Septuagint

Septuagint is the common title of the earliest and most important version of the Old Testament, namely, into Greek, and is generally held to have derived its title (seventy) from the traditionary number of its translators (see below), rather than (as Eichhorn thought) from the authority of the Alexandrian Sanhedrim as consisting of seventy members. In the following account we shall endeavor to sift the truth out of the traditions on this subject. SEE GREEK VERSIONS.

I. Origin of the Version. — This is as great a riddle as the sources of the Nile. The causes which produced the translation, the number and names of the translators, the times at which different portions were translated, are all uncertain.

1. Ancient Testimony on the Subject. —

(1.) The oldest writer who makes mention of the Septuagint is Aristobulus, an author referred to by Eusebius (Proepar. Evangel. 13, 12) and Clement of Alexandria (Stromata, 5, 595). According to Eusebius, he was a Jew, who united the Aristotelian with the Jewish philosophy, and composed a commentary on the law of Moses, dedicated to Ptolemy Philometor. He is also mentioned in 2 Macc. 1, 10. Both Clement and Eusebius make him contemporary with Philometor (2d century B.C.). for the passages in their writings, in which they speak of him under Philadelphus must either have been corrupted by ignorant transcribers or have been so written by mistake (Valckenaer, § 10, 11; Dähne, p. 81 sq.). His words relative to the Septuagint are these:

"It is manifest that Plato has followed our law, and studied diligently all its particulars; for before Demetrius Phalereus a translation had been made by others of the history of the Hebrews' going forth out of Egypt, and of all that happened to them, and of the conquest of the land, and of the exposition of the whole law. Hence it is manifest that the aforesaid philosopher borrowed many things, for he was very learned, as was Pythagoras, who also transferred many of our doctrines into his system. But the entire translation of our whole law (ἡ δὲ ὅλη ἑρμήνεια τῶν διὰ τοῦ νόμου πάντων) was made in the time of the king named Philadelphus, a man of greater zeal, under the direction of Demetrius Phalereus." The entire passage has occasioned much conjecture and discussion. It is given by Valckenaer (Diatribe, etc.), Thiersch (De Versione Alexandrina), and Frankel (Vorstudien, etc.). It appears that the words of Aristobulus do not speak of any prior Greek translation, as Hody supposes, or indeed of any translation whatever. They rather refer to some brief extracts relative to Jewish history, which had been made from the Pentateuch into a language commonly understood by the Jews in Egypt, before the time of Demetrius. The entire law was first rendered into Greek under Philadelphus. Hody, and after him Eichhorn, conjectured that the fragments of Aristobulus preserved by Eusebius and Clement were written in the 2d century by another Aristobulus, a Christian, and that Aristobulus, the professed Peripatetic, was a heathen. But the quotation of Cyril of Alexandria (Contra Julianum, lib. 6), to which they appeal, was erroneously made by that father, as may be seen by comparing it with Clement. Richard Simon also denied the authenticity of Aristobulus's remains (Histoire Critique du V.T. p. 189). But Valckenaer has sufficiently established their authenticity. The testimony of Aristobulus is corroborated by a Latin scholion recently found in a MS. of Plautus at Rome, which has been described and illustrated by Ritschl in a little book entitled Die alexandrinischen Bibliotheken, etc. (Berlin, 1838). From the passage of Aristobulus already quoted, it appears that in the time of Aristobulus, i.e. the beginning of the 2d century B.C., this version was considered to have been made when Demetrius Phalereus lived, or in the reign of Ptolemy Soter. Hody, indeed, has endeavored to show that this account contradicts the voice of certain history, because it places Demetrius in the reign of Philadelphus. But the version may have been begun under Soter and completed under Philadelphus, his successor. In this way may be reconciled the discordant notices of the time when it originated; for it is well known that the Palestinian account, followed by various fathers of the Church, asserts that Ptolemy Soter carried the work into execution, while according to Aristeas, Philo, Josephus, etc., his son Philadelphus was the person. Hody harmonizes the discrepancy by placing the translation of the Pentateuch in the two years during which father and son reigned conjointly (B.C. 286 and 285). The object of Demetrius in advising Soter to have in his library a copy of the Jewish laws in Greek is not stated by Aristobulus, but Aristeas relates that the librarian represented it to the king as a desirable thing that such a book should be deposited in the Alexandrian library. Some think that a literary rather than a religious motive led to the version. So Hävernick. This, however, may be reasonably doubted. Hody, Sturz, Frankel, and others conjecture that the object was religious or ecclesiastical. Eichhorn refers it to private impulse; while Hug takes the object to have been political. It is not probable, however, that the version was intended for the king's use, or that he wished to obtain from it information respecting the best mode of governing a nation and enacting laws for its economic well-being. The character and language of the version unite to show that an Egyptian king, probably ignorant of Greek, could not have understood the work. Perhaps an ecclesiastical motive prompted the Jews who were originally interested in it, while Demetrius Phalereus and the king may have been actuated by some other design.

It is difficult, if not impossible, to ascertain whether Aristobulus' words imply that all the books of the Old Test. were translated into Greek under Philadelphus, or simply the Pentateuch. Hody contends that νόμος, the term used by Aristobulus, meant at that time the Mosaic books alone, although it was afterwards taken in a wider sense so as to embrace all the Old Test. Valckenaer thinks that all the books were comprehended under it, It is certainly more natural to restrict it to the Pentateuch. The Pentateuch, therefore, was completed under Philadelphus.

(2.) The next historical testimony regarding the Septuagint is the prologue of Jesus the son of Sirach, a document containing the judgment of a Palestinian Jew concerning the version before us. His words are these:

"And not only these things, but the law itself, and the prophets, and the rest of the books, have no small difference when they are spoken in their own language." Frankel has endeavored to throw suspicion on this passage, as if it were unauthentic, but his reasons are extremely slender (p. 21, note w). It appears from it that the law, the prophets, and the other books had been translated into Greek in the time of the son of Sirach, i.e. that of Ptolemy Physcon, B.C. 130.

(3.) The account given by Aristeas comes next before us (see Rosenmüller, Handb. d. Lit. d. bibl. Kritik u. Exeg. 2, 413 sq.). This writer pretends to be a Gentile, and a favorite at the court of Ptolemy Philadelphus, king of Egypt. In a letter addressed to his brother Philocrates, he relates that Philadelphus, when forming a library at great expense, was advised by Demetrius Phalereus to apply to the Jewish high priest Eleazer for a copy of the book containing the Jewish laws. Having previously purchased the freedom of more than a hundred thousand captive Jews in Egypt, the king sent Aristeas and Aidreas to Jerusalem with a letter requesting of Eleazer seventy-two persons as interpreters, six out of each tribe. They were dispatched accordingly with a magnificent copy of the law, and were received and entertained by the king for several days with great respect and liberality. Demetrius led them to an island, probably Pharos, where they lodged together. The translation was finished in seventy-two days, having been written down by Demetrius piece by piece, as agreed upon after mutual consultation. It was then publicly read by Demetrius to a number of Jews whom he had summoned together. They approved of it, and imprecations were uttered against any one who should presume to alter it. The Jews requested permission to take copies of it for their use, and it was carefully preserved by command of the king. The interpreters were sent home loaded with presents.

The work of Aristeas, which was first published in the original Greek by Simon Schard (Basel, 1561, 8vo), and several times reprinted, was also given by Hody in Greek and Latin, in his book entitled De Bibliorum Textibus Originalibus, Versionibus Groecis, et Latina Vulgata (Oxon. 1705, fol.). The most accurate edition, however, is that by Galland, in the Bibliotheca Vet. Patrum, vol. 2. It was translated into English by Whiston, and published at London in 1727, 8vo. See also Aristeas, Hist. 72 Int. ex Rec. Eld. de Parchum (Francf. 1610; Oxon. 1692).

(4.) In all discussions relative to the name of Septuagint, so universally appropriated to the Greek version of Alexandria, the scholion discovered by Osann and published by Ritschl ought to be considered. The origin of this Latin scholion is curious. The substance of it is stated to have been extracted from Callimachus and Eratosthenes, the Alexandrian librarians, by Tzetzes, and from his Greek note an Italian of the 15th century has formed the Latin scholion in question. The writer has been speaking of the collecting of ancient Greek poems carried on at Alexandria under Ptolemy Philadelphus, and then he thus continues: "Nam rex ille philosophis affertissimus [corr. "differtissimus," Ritschl; "affectissimus," Thiersch] et caeteris omnibus auctoribus claris, disquisitis impensa regiae munificentiae ubique terrarum quantum valuit voluminibus opera Demetrii Phalerei phzxa senum duas bibliothecas fecit, alteram extra regiam alteram autem in regia." The scholion then goes on to speak of books in many languages: "Quae summa diligentia rex ille in suam linguam fecit ab optimis interpretibus converti" (see Thiersch, De Pentateuchi Versione Alexandrina [Erlang. 1841], p. 8, 9). Bernhardy reads instead of "phzxa senum," "et lxx senum," and this correction is agreed to by Thiersch, as it well may be: some correction is manifestly needed, and this appears to be right. This gives us seventy elders associated in the formation of the library. The testimony comes to us from Alexandrian authority; and this, if true (or even if believed to be true), would connect the Septuagint with the library — a designation which might most easily be applied to a version of the Scriptures there deposited; and, let the translation be once known by such a name, then nothing would be more probable than that the designation should be applied to the translators. This may be regarded as the first step in the formation of the fables. Let the Septuagint be first known as applying to the associates in the collection of the library, then to the library itself, and then to that particular book in the library which to so many had a far greater value than all its other contents. Whether more than the Pentateuch was thus translated and then deposited in the royal library is a separate question.

2. Confirmation by Later Authorities. —

(1.) Of Jewish writers, Josephus (Ant. 12, 2) agrees in the main with Aristeas; but Philo's account (De Vita Mosis, lib. 2) differs in a number of circumstances.

(2.) Among the Greek Church fathers Irenaeus (lib. 3, c. 24) relates that Ptolemy Lagi, wishing to adorn his Alexandrian library with the writings of all nations, requested from the Jews of Jerusalem a Greek version of their Scriptures; that they sent seventy elders well skilled in the Scriptures and in later languages; that the king separated them from one another and bade them all translate the several books. When they came together before Ptolemy and showed their versions, God was glorified, for they all agreed exactly, from beginning to end, in every phrase and word, so that all men may know that the Scriptures are translated by the inspiration of God.

Justin Martyr (Cohort. ad Groecos, p. 34) gives the same account, and adds that he was taken to see the cells in which the interpreters worked.

Epiphanius says that the translators were divided into pairs, in thirty-six cells, each pair being provided with two scribes; and that thirty-six versions agreeing in every point were produced, by the gift of the Holy Spirit (De Pond. et Mens. c. 3-6).

(3.) Among the Latin fathers Augustine adheres to the inspiration of the translators — "Non autem secundum LXX interpretes, qui etiam ipsi divino Spiritu interpretati, ob hoc aliter videntur nonnulla dixisse, ut ad spiritualem sensum scrutandum magis admoneretur lectoris intentio" (De Doctr. Christ. 4, 15).

But Jerome boldly throws aside the whole story of the cells and the inspiration — "Et nescio quis primus auctor Septuaginta cellulas Alexandriae mendacio suo extruxerit, quibus divisi eadem scriptitarent, cum Aristseus ejusdem Ptolemaei ὑπερασπιστής, et multo post tempore Josephus, nihil tale retulerint: sed in una basilica congregatos, contulisse scribant, non prophetasse. Aliud est enim vatem, aliud esse interpretem. Ibi Spiritus ventura praedicit; hic eruditio et verborum copia ea quae intelligit transfert" (Proef. ad Pent.).

3. Modern Opinions. —

(1.) Until the latter half of the 17th century the origin of the Sept. as given by Aristeas was firmly believed; while the numerous additions that had been made to the original story in the progress of centuries were unhesitatingly received as equally genuine. The story was first reckoned improbable by L. Vives (in a note to Augustine's De Civitate Dei); then Scaliger asserted that it was written by a Jew; and Richard Simon was too acute a critic not to perceive the truth of Scaliger's assertion. Hody was the first who demonstrated with great learning, skill, and discrimination that the narrative could not be authentic (De Bibl. Text. Orig. Vers. Groec. et Lat. Vulg. [Oxford, 1705] lib. 4). It is now universally pronounced fabulous.

(2.) But the Pseudo-Aristeas had a basis of fact for his fiction; on three points of his story there is no material difference of opinion and they are confirmed by the study of the version itself: (a.) The version was made at Alexandria. (b.) It was begun in the time of the earlier Ptolemies, about B.C. 280. (c.) The law (i.e. the Pentateuch) alone was translated at first. It is also very possible that there is some truth in the statement that a copy was placed in the royal library. (The emperor Akbar caused the New Test. to be translated into Persian.)

(3.) But by whom was the version made? As Hody justly remarks, "It is of little moment whether it was made at the command of the king or spontaneously by the Jews; but it is a question of great importance whether the Hebrew copy of the law and the interpreters (as Pseudo-Aristeas and his followers relate) were summoned from Jerusalem and sent by the high priest to Alexandria." On this question no testimony can be so conclusive as the evidence of the version itself, which bears upon its face the marks of imperfect knowledge of Hebrew, and exhibits the forms and phrases of the Macedonic Greek prevalent in Alexandria, with a plentiful sprinkling of Egyptian words. The forms ἤλθοσαν παρενεβάλοσαν, betray the fellow-citizens of Lycophron, the Alexandrian poet, who closes his iambic line with κἀπὸ γῆς ἐσχάζοσαν. Hotldy (2, 4) gives several examples of Egyptian renderings of names and coins and measures; among them the hippodrome of Alexandria for the Hebrew Cibrath (Ge 48:7), and the papyrus of the Nile for the rush of Job (Job 8:11). The reader of the Sept. will readily agree with his conclusion, "Sive regis jussu, sive sponte a Judaeis, a Judaeis Alexandrinus fuisse factam." The question as to the moving cause which gave birth to the version is one which cannot be so decisively answered either by internal evidence or by historical testimony. The balance of probability must be struck between the tradition, so widely and permanently prevalent, of the king's intervention, and the simpler account suggested by the facts of history and the phenomena of the version itself. It is well known that after the Jews returned from the captivity of Babylon, having lost in great measure the familiar knowledge of the ancient Hebrew, the readings from the books of Moses in the synagogues of Palestine were explained to them in the Chaldaic tongue in Targums or paraphrases; and the same was done with the books of the prophets when, at a later time, they also were read in the synagogues. The Jews of Alexandria had probably still less knowledge of Hebrew; their familiar language was Alexandrian Greek. They had settled in Alexandria in large numbers soon after the time of Alexander and under the earlier Ptolemies. They would naturally follow the same practice as their brethren in Palestine; the law first, and afterwards the prophets, would be explained in Greek, and from this practice would arise in time an entire Greek version. All the phenomena of the version seem to confirm this view; the Pentateuch is the best part of the version; the other books are more defective, betraying probably the increasing degeneracy of the Hebrew MSS. and the decay of Hebrew learning with the lapse of time.

(4.) Nevertheless, the opinion that the Pentateuch was translated a considerable time before the prophets is not warranted by the language of Justin, Clement of Alexandria, Tertullian, Epiphanius, and Hilary of Poitiers; although we are aware that Aristeas, Josephus, Philo, the Talmudists, and Jerome mention the law only as having been interpreted by the seventy-two. Hody thinks that the Jews first resorted to the reading of the prophets in their synagogues when Antiochus Epiphanes forbade the use of the law, and therefore that the prophetic portion was not translated till after the commencement of Philometor's reign. It is wholly improbable, however, that Antiochus interdicted the Jews merely from reading the Pentateuch (comp. 1 Macc. 1:41, etc.; and Josephus, Ant. 12, 5; Frankel, p. 48, 49). The interval between the translating of the law and the prophets, of which many speak, was probably very short. Hody's proof that the book of Joshua was not translated till upwards of twenty years after the death of Ptolemy Lagi, founded upon the word γαισός, is perfectly nugatory, although the time assigned cannot be far from the truth. The epilogue to the book of Esther does not state that this part of the Old Test. was translated under Ptolemy Philometor or that it was dedicated to him. On the contrary it refers to a certain epistle containing apocryphal additions to the canonical book of Esther (Valckenaer, p. 33, 63). It is a fruitless task to attempt to ascertain the precise times at which separate portions of the version were made. All that can be known with any degree of probability is that it was begun under Lagi and finished before the thirty-eighth year of Ptolemy Physcon.

It is obvious from internal evidence that there were several translators, but certainly not seventy-two. Hody has endeavored to parcel out their version into small portions, assigning each part to a separate person, and affirming that they were put together in one cento without revision; but his notions of rigid uniformity in the translators are such as exclude perspicuity, freedom, variety, and elegance. There is no ground for believing that the Pentateuch proceeded from more than one interpreter, who was unquestionably the most skilful of all. The entire work was made by five or six individuals at least, and must, consequently, be of unequal value. Comp. Amersfoordt, De Variis Lectio. Holmes. Loc. quorund. Pent. Mos. (Lugd. 1815); Thiersch, De Pent. Vers. Al. Libri III (Erlang. 1841); Frankel, Ueber d. Einfluss d. palest. Exeg. auf d. alex. Hermen. (Leips. 1851); Rosenmüller, op. cit. p. 435 sq.

(5.) In opposition to the Pseudo-Aristeas, we cannot but maintain that the translators were Alexandrian, not Palestinian, Jews. The internal character of the entire version, particularly of the Pentateuch, sufficiently attests the fact. We find, accordingly, that proper names and terms peculiar to Egypt are rendered in such a manner as must have been unintelligible to a Greek- speaking population other than the Egyptian Jews. That the translators were Egyptians has been proved, to the satisfaction of all, by Hody; although some of his examples are not appropriate or conclusive. Frankel supposes that the version was made not only at different times, but at different places. This is quite arbitrary. There is no reason for believing with him that different books originated after this fashion, the impulse having gone forth from Alexandria and spreading to localities where the Jews had settled, especially Cyrene, Leontopolis, and even Asia Minor.

(6.) The division into verses and chapters is much later than the age of the translators. Our present editions have been printed in conformity with the division into chapters made in the 12th century, though they are not uniform in this particular. Still, however, many MSS. have separations in the text. The Alexandrine Codex is said by Grabe to have 140 divisions, or, as they may be called, chapters, in the book of Numbers alone (Prolegomena, c. 1, § 7).

The titles given to the books, such as Γένεσις, etc., could hardly have been affixed by the translators, since often they do not harmonize with the version of the book itself to which they belong.

II. Textual Basis of the Version. —

1. It has been inquired whether the translator of the Pentateuch followed a Hebrew or a Samaritan codex. The Sept. and Samaritan harmonize in more than a thousand places, where they differ from the Hebrew. Hence it has been supposed that the Samaritan edition was the basis of the version. Various considerations have been adduced in favor of this opinion; and the names of De Dieu, Selden, Whiston, Hottinger, Hassencamp, and Eichhorn are enlisted on its behalf. But the irreconcilable enmity subsisting between the Jews and the Samaritans, both in Egypt and Palestine, effectually militates against it. Besides, in the prophets and Hagiographa, the number of variations from the Masoretic text is even greater and more remarkable than those in the Pentateuch; whereas the Samaritan extends no further than the Mosaic books. No solution, therefore, can be satisfactory which will not serve to explain at once the cause or causes both of the differences between the Seventy and the Hebrew in the Pentateuch, and those found in the remaining books. The problem can be fully solved only by such a hypothesis as will throw light on the remarkable form of the Sept. in Jeremiah and Esther, where it deviates most from the Masoretic MSS., presenting such transpositions and interpolations as excite the surprise of the most superficial reader. The above solution of the question must be rejected not only for the reasons assigned, but also for the following.

(1.) It must be taken into account that if the discrepancies of the Samaritan and Jewish copies be estimated numerically, the Sept. will be found to agree far more frequently with the latter than the former.

(2.) In the cases of considerable and marked passages occurring in the Samaritan which are not in the Jewish, the Sept. does not contain them.

(3.) In the passages in which slight variations are found, both in the Samaritan and Sept., from the Jewish text, they often differ among themselves, and the amplification of the Sept. is less than that of the Samaritan.

(4.) Some of the small amplifications in which the Samaritan seems to accord with the Sept. are in such incorrect and non-idiomatic Hebrew that it is suggested that these must be translations, and, if so, probably from the Sept.

(5.) The amplifications of the Sept. and Samaritan often resemble each other greatly in character, as if similar false criticism had been applied to the text in each case. But as, in spite of all similarities such as these, the Pentateuch of the Sept. is more Jewish than Samaritan, we need not adopt the notion of translation from a Samaritan codex, which would involve the subject in greater difficulties, and leave more points to be explained. (On some of the supposed agreements of the Sept. with the Samaritan, see bishop Fitzgerald in Kitto's Journal of Sacred Literature, Oct. 1848, p. 324-332.)

Some suppose that the one was interpolated from the other — a conjecture not at all probable. Jahn and Bauer imagine that the Hebrew MS. used by the Egyptian Jews agreed much more closely with the Samaritan in., the text and forms of its letters than the present Masoretic copies. This hypothesis, however, even if it were otherwise correct, would not account for the great harmony existing between the Samaritan and Sept.

Another hypothesis has been put forth by Gesenius (Commentatio de Pent. Samar. Orig. Indole, et Auctor.), viz. that both the Samaritan and Sept. flowed from a common recension (ἔκδοσις) of the Hebrew Scriptures, one older than either, and different in many places from the recension of the Masoretes now in common use. "This supposition," says Prof. Stuart, by whom it is adopted, "will account for the differences and for the agreements of the Sept. and Samaritan." The following objections have been made to this ingenious and plausible hypothesis.

(a.) It assumes that before the whole of the Old Test. was written there had been a recension or revision of several books. But there is no record or tradition in favor of the idea that inspired men applied a correcting hand in this manner till the close of the canon. To say that others did so is not in unison with right notions of the inspiration of Scripture, unless it be equally affirmed that they corrupted, under the idea of correcting, the holy books.

(b.) This hypothesis implies that a recension took place at a period comparatively early, before any books had been written except the Pentateuch, Joshua, Judges, and the writings of David and Solomon. If it be improbable that a revised edition was made before the completion of the canon, it is much more improbable that it was undertaken when few books were written.

(c.) It supposes that an older recension was still current after Ezra had revised the whole collection and closed the canon. In making the Sept. version, it is very improbable that the Jews, who were the translators, followed a recension far inferior in their estimation to the copy of the sacred books corrected by Ezra. This objection rests on the assumption that Ezra completed the canon of the Old Test., having been prompted, as well as inspired, to arrange and revise the books of Scripture. Such is the Jewish tradition; and although a majority of the German critics disallow its truth, yet it is held by very able and accomplished men.

Prof. Lee (Prolegomena to Bagster's Polyglot) accounts for the agreement between the Sept. and Samaritan in another way. He conjectures that the early Christians interspersed their copies with Samaritan glosses, which ignorant transcribers afterwards inserted in the text. But he has not shown that Christians in general were acquainted with the Samaritan Pentateuch and its additions to the Hebrew copy; neither has he taken into account the reverence entertained by the early Christians for the sacred books. We cannot, therefore, attribute the least probability to this hypothesis.

Another hypothesis has been mentioned by Frankel, viz. that the Sept. flowed from a Chaldee version, which was used before and after the time of Ezra — a version inexact and paraphrastic, which had undergone many alterations and corruptions. This was first proposed by R. Asaria di Rossi, in the midst of other conjectures. Frankel admits that the assumption of such a version is superfluous, except in relation to the Samaritan Pentateuch, where much is gained by it. This Chaldee version circulated in various transcripts here and there; and as the same care was not applied in preserving its integrity as was exercised with respect to the original Hebrew, the copies of it presented considerable differences among themselves. Both the Greek version and the Samaritan Pentateuch were taken from it. Frankel concedes that this hypothesis is not satisfactory with regard to the Sept., because the mistakes found in that version must have frequently originated in misunderstanding the Hebrew text. There is no evidence, however, that any Targum or Chaldee version had been made before Ezra's time, or soon after. Explanations of the lessons publicly read by the Jews were given in Chaldee, not regularly perhaps, or uniformly; but it can scarcely be assumed that a Chaldee version had been made out in writing, and circulated in different copies. Glosses, or short expositions of words and sentences, were furnished by the public readers for the benefit of the people; and it is by no means improbable that several of these traditional comments were incorporated with the version by the Jewish translators, to Whom they were familiar.

In short, no hypothesis yet proposed commends itself to general reception, although the Vorstudien of Frankel have probably opened up the way towards a correct solution. The great source from which the striking peculiarities in the Sept. and the Samaritan flowed appears to us to have been early traditional interpretations current among the Jews, targums, or paraphrases — not written, perhaps, but orally circulated. Such glossarial versions, which must have circulated chiefly in Palestine. require to be traced back to an early epoch — to the period of the second Temple. They existed, in substance at least, in ancient times, at once indicating and modifying the Jewish mode of interpretation. The Alexandrian mode of interpretation stood in close connection with the Palestinian; for the Jews of Egypt looked upon Jerusalem as their chief city, and the Sanhedrim of Jerusalem as their ecclesiastical rulers. If, therefore, we can ascertain the traditional paraphrases of the one. those of the other must have been substantially the same (see Gieseler's Eccles. Hist., transl. by Cunningham, 1, 30).

Tychsen (Tentamen de Variis Codd. Heb. V.T. MSS. Gener.) thought that the Sept. was made from the Hebrew transcribed into Hebrew-Greek characters. It is almost unnecessary to refer to such a notion. It never obtained general currency, having been examined and refuted by Dathe, Michaelis, and Hassencamp.

2. Evidence as to the Verbal Condition of the Original. Here we naturally inquire as, to two obvious points:

(1.) Was the version made from Hebrew MSS. with the vowel-points now used? A few examples will indicate the answer.

A. PROPER NAMES.

Hebrew. Septuagint. Ex 6:17, לַבנַי, Libni. Λοβενεί. 6:19, מִחלַי, Machli. Μοολεί. 13:20, אֵתָם, Etham. Ο᾿θώμ De 3:10, סִלכָה, Salchah. ῾Ελχᾶ. 4:43, בֶּצֶר, Bezer. Βοσόρ.

34:1, פַּסגָּה, Pisgah. Φασγά.

B. OTHER WORDS.

Hebrew. Septuagint. Ge 1:9, מָקום, place. συναγωγή (מַקוֶה). 15:11, וִיֵּשֶׁב אֹתָם καὶ συνεκὰθισεν αὐτοῖς and he drove them away. (וִיֵּשֶׁב אַתָּם). Ex 12:17, אֶתאּהִמִּצּוֹת, τὴν ἐντολὴν ταύτην unleavened bread. אֶתאּהִמַּצוָה). Nu 16:5, בֹּקֶר, in the ἐπέσκεπταί morning. (בָּקִר).

De 15:18, מַשׁנֵה, double. ἐπέτειον, (מַשָּׁנָה). Isa 9:8, דָּבָר, a word. θάνατον (דֶּבֶר ). Examples of these two kinds are innumerable. Plainly the Greek translators had not Hebrew MSS. pointed as at present. In many cases (e.g. Ex 2:25; Na 3:8) the Sept. has possibly preserved the true pronunciation and sense where the Masoretic pointing has gone wrong.

(2.) Were the Hebrew words divided from one another, and were the final letters וֹ, Š, ן , ֹם ,, in use when the Sept. was made? — Take a few out of many examples:

Hebrew. Septuagint.

(1) De 26:5, אֲרִמַּי אֹבֵד, Συρίαν ἀπέβαλεν a perishing Syrian. (ארם יאֹבד). (2) 2Ki 2:14, אִŠאּהוּאַ, ἀφφώ he also. [it joins the two words in one]. (3) 2Ki 22:20, לָכֵן οὐχ οὕτως therefore. (לאֹאּכֵן) (4) 1Ch 17:10, וָאִגַּד לךָ , καὶ αὐξήσω δε and I told thee. (וִאֲגִדֶּלךָ),

(5) Ho 6:5, יֵצֵא וּמַשׁפָּטֶיךָ אוֹר καὶ τὸ κρίμα μον ὡς φῶς ἐξελεύ σεταί and thy judgments The Sept. reads: [are as] the light [that] goeth forth וּמִשׁפָטִי כָאוֹר(6) Zec 11:7, לָכֵן עֲנַיֵּי הִצּאֹןεἰς τὴν Χανανῖτιν even you, O poor of the flock. [it joins the first two words].

Here we find three cases (2, 4, 6) where the Sept. reads as one word what makes two in the present Hebrew text; one case (3) where one Hebrew word is made into two by the Sept.; two cases (1, 5) where the Sept. transfers a letter from the end of one word to the beginning of the next. By inspection of the Hebrew in these cases it will be easily seen that the Hebrew MSS. must have been written without intervals between the words, and that the present final forms were not then in use. In three of the above examples (4, 5, 6), the Sept. has perhaps preserved the true division and sense. In the study of these minute particulars, which enable us to examine closely the work of the translators, great help is afforded by Cappelli Critica Sacra, and by the Vorstudien of Frankel, who has most diligently anatomized the text of the Sept. His projected work on the whole of the version has not been completed, but he has published a part of it in his treatise Ueber den Einfluss der palästinischen Exegese auf die alexandrinische Hermeneutik, in which he reviews minutely the Sept. version of the Pentateuch.

III. Ecclesiastical Authority and Influence. — The Sept. does not appear to have obtained general authority among the Jews so long as Hebrew was understood at Alexandria. It is remarkable that Aristobulus quotes the original, even where it departs from the text of the Sept. The version was indeed spread abroad in Egypt, Northern. Africa, and Asia Minor. It seems to have been so highly esteemed by the Jews as to be publicly read in some of their synagogues. From the 146th Novella of Justinian, it would seem that some Jews wished the public interpreter, who read the lessons out of the law and the prophets in Hebrew, to give his explanations of them in Greek, while others desired to have them in Chaldee. The reader, therefore, employed this translation as, explanatory of the sections recited in the original, yet, although they highly esteemed the Greek, they did not regard it as equal to the Hebrew. Even the Talmudists make honorable mention of its origin. It is true that the Talmud also speaks of it as an abomination to the Jews in Palestine; but this refers to the 2d century and the time following, not to the period immediately after the appearance of Christ. When controversies arose between Christians and Jews, and the former appealed with irresistible force of argument to this version, the latter denied that it agreed with the Hebrew original. Thus by degrees it became odious to the Jews — as much execrated as it had before been commended. They had recourse to the translation of Aquila, who is supposed to have undertaken a new work from the Hebrew, with the express object of supplanting the Sept. and favoring the sentiments of his brethren, Among the Christians the ancient text, called κοινή, was current before the time of Origen. We find it quoted by the early Christian fathers — in Greek by Clemens Romanus, Justin Martyr, Irenaeus; in Latin versions by Tertullian and Cyprian. We find it questioned as inaccurate by the Jews (Just. Martyr, Apol.), and provoking them to obtain a better version (hence the versions of Aquila, etc.). We find it quoted by Josephus and Philo; and thus we are brought to the time of the apostles and evangelists, whose writings are full of citations and references, and imbued with the phraseology of the Sept. From all this we are justified in the following conclusions on this head:

1. This version was highly esteemed by the Hellenistic Jews before the coming of Christ. An annual festival was held at Alexandria in remembrance of the completion of the work (Philo, De Vita Mosis, lib. 2). The manner in which it is quoted by the writers of the New Test. proves that it had long been in general use. Wherever, by the conquests of Alexander or by colonization, the Greek language prevailed; wherever Jews were settled, and the attention of the neighboring Gentiles was drawn to their wondrous history and law, there was found the Sept., which thus became, by Divine Providence, the means of spreading widely the knowledge of the One True God and his promises of a Savior to come throughout the nations; it was indeed ostium gentibus ad Christum. To the wide dispersion of this version we may ascribe, in great measure, that general persuasion which prevailed over the whole East (percrebuerat Oriente toto) of the near approach of the Redeemer, and which led the magi to recognize the star that proclaimed the birth of the King of the Jews.

2. Not less wide was the influence of the Sept. in the spread of the Gospel. Many of those Jews who were assembled at Jerusalem on the day of Pentecost, from Asia Minor, from Africa, from Crete and Rome, used the Greek language; the testimonies to Christ from the law and the prophets came to them in the words of the Sept.; St. Stephen probably quoted from it in his address to the Jews; the Ethiopian eunuch was reading the Sept. version of Isaiah in his chariot (ὡς πρόβατον ἐπὶ σφαγὴν ἤχθη); they who were scattered abroad went forth into many lands, speaking of Christ in Greek, and pointing to the things written of him in the Greek version of Moses and the prophets; from Antioch and Alexandria in the East to Rome and Massilia in the West, the voice of the Gospel sounded forth in Greek; Clemens of Rome, Ignatius at Antioch, Justin Martyr in Palestine, Irenaeus at Lyons, and many more, taught and wrote in the words of the Greek Scriptures; and a still wider range was given to the Sept. by the Latin version (or versions) made from it for the use of the Latin churches in Italy and Africa; and in later times by the numerous other versions into the tongues of Egypt, Ethiopia, Armenia, Arabia, and Georgia. For a long period the Sept. was the Old Test. of the far larger part of the Christian Church (see the Hulsean Prize Essay, by W.R. Churton, On the Influence of the Sept. on the Progress of Christianity [Camb. 1861J; and an art. in the Zeitschr. f. wissensch. Theol. 1862, vol. 3).

A number of other versions have been founded on the Sept.

1. Various early Latin translations, the chief of which was the Vetus Itala; 2. The Coptic and Sahidic, belonging to the 1st and 2d centuries; 3. The Ethiopic, belonging to the 4th century; 4. The Armenian, of the 5th century; 5. The Georgian, of the 6th century; 6. Various Syriac versions, of the 6th and 8th centuries; 7. Some Arabic versions, SEE ARABIC VERSIONS; 8. The Slavonic, belonging to the 9th century.

IV. Liturgical Origin of Portions of the Version. — This is a subject for inquiry which has received but little attention; not so much, probably, as its importance deserves. It was noticed by Tregelles many years ago that the headings of certain psalms in the Sept. coincide with the liturgical directions in the Jewish Prayer book. The results were at a later period communicated in Kitto's Journal of Sacred Literature, April 1852, p. 207-

209. The results may be briefly stated: The 23d Psalm, Sept. (Heb. 24th), is headed in the Sept. τῆς μιᾶς σαββάτου; so, too, in Heb. in De Sola's Prayers of the Sephardim, ביום הראשׁון: Psalm 47, Sept. (Hebrews 48), δευτέρᾷ σαββάτου ליום שׁני: Psalm 93, Sept. (Hebrews 94), τετράδι σαββάτου, ליום רביעי: Psalm 92, Sept. (Hebrews 93), εἰς τὴν ἡμέραν τοῦ προσαββάτου,ליום שׁשׁי; There appear to be no Greek copies extant which contain similar headings for Ps 81; Ps 30 (Hebrews 82 and 81), which the Jewish Prayer book appropriates to the third and fifth days; but that such once existed in the case of the latter psalm seems to be shown from the Latin Psalterium Vetus having the prefixed quinta sabbati, ליום חמישׁי. Delitzsch, in his Commentary on the Psalms, has recently pointed out that the notation of these psalms in the Sept. is in accordance with certain passages in the Talmud.

It is worthy of inquiry whether variations in other passages of the Sept. from the Hebrew text cannot at times be connected with liturgical use, and whether they do not originate in part from rubrical directions. It seems to be at least plain that the Psalms were translated from a copy prepared for synagogue worship.

V. Character of the Version. — Under this head we have to consider several special questions relating to its internal character as a translation:

1. Is the Sept. Faithful in Substance? — Here we cannot answer by citing a few examples; the question refers to the general texture, and any opinion we express must be verified by continuous reading. For a purely philological examination, SEE SEPTUAGINT, LINGUISTIC CHARACTER OF.

(1.) It has been clearly shown by Hody, Frankel, and others that the several. books were translated by different persons, without any comprehensive revision to harmonize the several parts. Names and words are rendered differently in different books; e.g. פֶּסִח, the Passover, in the Pentateuch is rendered πάσχα; in 2Ch 35:6,. φασέκ.

אוּרַי, Urim, Ex 28:26, δήλωσις; De 33:8, δῆλοι; Ezr 2:63, (φωτίζοντες; Ne 7:65, φωτίσων.

תֻמַּ, Thummim, Ex 28:26, ἀλήθεια; Ezr 2:63, τέλειον.

The Philistines in the Pentateuch and Joshua are φυλιστείμ; in the other books ἀλλόφυλοι.

The books of Judges, Ruth, Samuel, and Kings are distinguished by the use of ἐγώ είμι instead of ἐγώ.

These are a few put of many like variations.

(2.) Thus the character of the version varies much in the several books; those of the Pentateuch are the best, as Jerome says ("Confitemur plus quam caeteris cum Hebraicis consonare"), and this agrees well with the external evidence that the law was translated first, when Hebrew MSS. were more correct and Hebrew better known. Perhaps the simplicity of the style in these early books facilitated the fidelity of the version.

(3.) The poetical parts are, generally speaking, inferior to the historical, the original abounding with rarer words and expressions. In these parts the reader of the Sept. must be continually on the watch lest an imperfect rendering of a difficult word mar the whole sentence. The Psalms and Proverbs are perhaps the best.

(4.) In the major prophets some of the most important prophecies are sadly obscured — e.g. Isa 9:1, τοῦτο πρῶτον πίε ταχὺ ποίει, χώρα Ζαβουλών, κ. τ. λ.; and in 9:6, καὶ τοῦτο τὸ ὄνομα αὐτοῦ ὅ καλέσει αὐτὸν Κύριος Ι᾿ωσεδὲκ ἐν τοῖς προφήταις. Ezekiel and the minor prophets (speaking generally) seem to be better rendered. The Sept. version of Daniel was not used, that of Theodotion being substituted for it.

(5.) Supposing the numerous glosses and duplicate renderings which have evidently crept from the margin into the text to be removed (e.g. Isa 7:16; Hab 3:2; Joe 1:8) — for these are blemishes not of the version itself, but of the copies — and forming a rough estimate of what the Sept. was in its earliest state, we may perhaps say of it, in the words of the well known simile, that it was, in many parts, "the wrong side of the Hebrew tapestry," exhibiting the general outlines of the pattern, but confused in the more delicate lines, and with many ends of threads visible; or, to use a more dignified illustration, the Sept. is the image of the original seen through a glass not adjusted to the proper focus — the larger features are shown, but the sharpness of definition is lost. On Judges, see Grabii Ep. ad J. Millium qua Ostend. L. Judd. Gen. LXX Ves,. eam esse quam MS.

Alex. Exhibet, etc. (Oxf. 1705); Ziegler, Theol. Abhandl. (Gött. 1791), vol. 1. On Samuel and Kings, Thenius, Kurzgef. exeg. Hdb. z. A.T. 4, 24 sq.; 9, 13 sq. On Chronicles, Movers, Krit. Unters. (Bonn, 1834). On Esther, Fritsche's ed. (Zür. 1848). See Jeremiah s.v. Jud. Alex. ac Relig. init. Groec., em. Notisque Crit. ill. G.L. Spohn (Lips. 1794; 2d ed. 1824, by F.A.G. Spohn).

2. Is the Version Minutely Accurate in Details? — We have anticipated the answer to this question, but will give a few examples:

(1.) The same word in the same chapter is often rendered by differing words — Ex 12:13, פָּסִחתַּי, "'I will pass over," Sept. σκεπάσω, but 23, פָּסִח, "will pass over," Sept. παρελεύσεται.

(2.) Differing words by the same word — Ex 12:23, עָבִר, "pass through," and פָּסִח , "pass over," both by πρελεύσεται; Nu 15:4-5, מַנחָה, "offering," and זֶבִח, "sacrifice," both by θυσία.

(3.) The divine names are frequently interchanged; Κύριος is put for אֵֹלהַים , God, and Θεός for יהוֹה, Jehovah; and the two are often wrongly combined or wrongly separated.

(4.) Proper names are sometimes translated, sometimes not. In Genesis 23: — by translating the name Machpelah (τὸ διπλοῦν), the version is made to speak first of the cave being in the field (ver.9), and then of the field being in the cave (ver. 17), ὁ ἀγρὸς Ε᾿φρών, ὅς ην ἐν τῷ διπλῷ σπηλαίῳ, the last word not warranted by the Hebrew. Zec 6:14 is a curious example of four names of persons being translated — e.g. לַטוֹבַיָּה, "to Tobijah," Sept. τοῖς χρησίμοις αὐτῆς; Pisgah in De 34:1, is φασγά, but in De 3:27, τοῦ λελαξευμένου. (5.) The translators are often misled by the similarity of Hebrew words — e.g. Nu 3:26, מֵיתָרָיו, "the cords of it," Sept. τὰ κατάλοιπα. and 4:26, τὰ περισσά. In other places οἱ κάλοι, and Isa 54:2, τὰ σχοινίσματα, both rightly. Ex 4:31, יַשׁמַעוּ, "they heard," Sept. ἐχάρη (יַשׂמהוּ); Nu 16:15, "I have not taken one ass" (חֲמוֹר), Sept. οὐκ ἐπιθύμημα,( חמר) εἴληφα; De 32:10, יַמצָאֵהוּ, "he found him," Sept. αὐτάρκησεν αὐτόν; 1Sa 12:2, .

שִׂבתַּי, "I am gray headed," Sept. καθήσομαι (שִׁבתַּי);Ge 3:17, בִּעֲבוּרֶךָ" for thy sake," Sept. ἐν τοῖς ἔργοις σου (ד forר).

In very similar cases the error may be thus traced to the similarity of some of the Hebrew letters, ד and ר, ה and ת, י and 5, etc.; in some it is difficult to see any connection between the original and the version — e.g. De 32:8, בּנֵי יַשׂרָאֵל, "the sons of Israel," Sept. ἀγγέλων Θεοῦ. Aquila and Symmachus, υἱῶν Ι᾿σραήλ.

Isaiah 21:11, 12. Septuagint.

Watchman, what of the night? Watchman, what of the night? The watchman said, The morning cometh, and also the night: Φυλάσσετε ἐπάλξεις; Θυλάσσω τοπρωϊv καὶ τήν νύκτα. If ye will inquire, inquire ye. ἐὰν ζητῇς ζήτει·, Return, come. καὶ παῤ ἐμοὶ οἴκει. (6.) Besides the above deviations and many like them, which are probably due to accidental causes — the change of a letter, or doubtful writing in the Hebrew — there are some passages which seem to exhibit a studied variation in the Sept. from the Hebrew, e.g. Ge 2:2, on the seventh (השביעי) day God ended his work; Sept. συνετέλεσεν ὁ Θεὸς ἐν τῇ ἡμέρᾷ τῇ ἕκτῃ τὰ ἔργα αὐτοῦ.. The addition in Ex 12:40, καὶ ἐν τῇ γῇ Χαναάν, appears to be of this kind, inserted to solve a difficulty.

Frequently the strong expressions of the Hebrew are softened down; where human parts are ascribed to God for hand the Sept. substitutes power; for mouth, word, etc. Ex 4:16, "Thou shalt be to him instead of God" (לֵאלֹהַום), Sept. σὺ δὲ αὐτῷ ἔσῃ τὰ πρὸς τὸν Θεόν (see Ex 4:15). These and many more savor of design rather than of accident or error.

The version is, therefore, not minutely accurate in details; and it may be laid down as a principle, never to build any argument on words or phrases of the Sept. without comparing them with the Hebrew. The Greek may be right; but very often its variations are wrong.

3. We shall now be prepared to weigh the tradition of the fathers, that the version was made by inspiration (κατ᾿ ἐπίπνοιαν τοῦ Θεοῦ, Irenaeus; "Divino Spiritu interpretati," Augustine). Even Jerome himself seems to think that the Sept. may have sometimes added words to the original "ob Spiritus Sancti auctoritatem, licet in Hebraeis voluminibus non legatur" (Proefat. in Paralip. tom. 1, col. 1419).

Let us try to form some conception of what is meant by the inspiration of translators. It cannot mean what Jerome here seems to allow, that the translators were divinely moved to add to the original, for this would be the inspiration of prophets, as he himself says in another passage (Prolog. in Genesin), "Aliud est enim vertere, aliud esse interpretem." Every such addition would be, in fact, a new revelation. Nor can it be, as some have thought, that the deviations of the Sept. from the original were divinely directed, whether in order to adapt the Scriptures to the mind of the heathen or for other purposes. This would be, pro tanto, a new revelation, and it is difficult to conceive of such a revelation; for, be it observed, the discrepancy between the Hebrew and Greek Scriptures would tend to separate the Jews of Palestine from those of Alexandria, and of other places where the Greek Scriptures were used; there would be two different copies of the same books dispersed throughout the world, each claiming divine authority; the appeal to Moses and the prophets would lose much of its force; the standard of divine truth would be rendered doubtful; the trumpet would give an uncertain sound. No! If there be such a thing as an inspiration of translators, it must be an effect of the Holy Spirit on their minds, enabling them to do their work of translation more perfectly than by their own abilities and acquirements; to overcome the difficulties arising from defective knowledge, from imperfect MSS., from similarity of letters, from human infirmity and weariness; and so to produce a copy of the Scriptures, setting forth the Word of God and the history of his people, in its original truth and purity. This is the kind of inspiration claimed for the translators by Philo (Vit. Mosis, lib. 2): "We look upon the persons who made this version not merely as translators, but as persons chosen and set apart by divine appointment, to whom it was given to comprehend and express the sense and meaning of Moses in the fullest and clearest manner." The reader will be able to judge from the foregoing examples whether the Sept. version satisfies this test. If it does, it will be found not only substantially faithful, but minutely accurate in details: it will enable us to correct the Hebrew in every place where an error has crept in; it will give evidence of that faculty of intuition in its highest form which enables our great critics to divine from the faulty text the true reading; it will be, in short, a republication of the original text, purified from the errors of human hands and eyes stamped with fresh authority from heaven. This is a question to be decided by facts, by the phenomena of the version itself. We will simply declare our own conviction that, instead of such a divine republication of the original, we find a marked distinction between the original and the Sept. — a distinction which is well expressed in the words of Jerome (Prolog. in Genesin): "Ibi Spiritus ventura praedicit; hic eruditio et verborum copia ea quae intelligit transfert." It will be remembered that this agrees with the ancient narrative of the version, known by the name of Aristeas, which represents the interpreters as meeting in one house, forming One council, conferring together, and agreeing on the sense (see Hody, lib. 2, c. 6).

There are some, perhaps, who will deem this estimate of the Sept. too low; who think that the use of this version in the New Test. stamps it with an authority above that of a mere translation. But as the apostles and evangelists do not invariably cite the Old Test. according to this version, we are left to judge by the light of facts and evidence. Students of Holy Scripture, as well as students of the natural world, should bear in mind the maxim of Bacon, "Sola spes est in vera inductione."

VI. Benefits to be Derived from the Study of the Septuagint. — After all the notices of imperfection above given, it may seem strange to say, but we believe it to be the truth, that the student of Scripture can scarcely read a chapter without some benefit, especially if he be a student of Hebrew, and able, even in a very humble way, to compare the version with the original.

1. We have seen above that the Sept. gives evidence of the character and condition of the Hebrew MSS. from which it was made with respect to vowel points and the mode of writing. This evidence often renders very material help in the correction and establishment of the Hebrew text. Being made from MSS. far older than the Masoretic recension, the Sept. often indicates readings more ancient and more correct than those of our present Hebrew MSS. and editions, and often speaks decisively between the conflicting readings of the present MSS. The following are instances: Ps 22:17 (in the Sept. 21:16). The printed Hebrew text is כארי; but several MSS. have a verb in the third person plural, כארַו: the Sept. steps in to decide the doubt, ὤρυξαν χεῖράς μου καὶ πόδας μου, confirmed by Aquila, Vσχυναν.

Ps 16:10. The printed text is חסידי , in the plural; but near two hundred MSS. have the singular, חסיד, which is clearly confirmed by the evidence of the Sept., οὐδὲ δώσεις τὸν ὅσιόν σου ἰδεῖν διαφθοράν.

In passages like these, which touch on the cardinal truths of the Gospel, it is of great importance to have the testimony of an unsuspected witness in the Sept. long before the controversy between Christians and Jews.

In Ho 6:5, the context clearly requires that the first person should be maintained throughout the verse; the Sept. corrects the present Hebrew text, without a change except in the position of one letter, τὸ κρίμα μου ὡς φῶς ἐξελεύσεται, rendering unnecessary the addition of words in italics in our English version.

Other examples might be given, but we must content ourselves with one signal instance of a clause omitted in the Hebrew (probably by what is called ὁμοιοτέλευτον) and preserved in the Sept. In Ge 4:8 is a passage which in the Hebrew and in our English version is evidently incomplete: "And Cain talked (וִיּאֹמֶר) with Abel his brother; and it came to pass when they were in the field," etc. Here the Hebrew word וִיּאֹמֶרis the word constantly used as the introduction to words spoken, "Cain said unto Abel;" but, as the text stands, there are no words spoken, and the following words "... when they were in the field" come in abruptly. The Sept. fills up the lacuna Hebroeorum codicum (Pearson), καὶ ειπε Κάϊν πρὸς Α᾿βὲλ τὸν ἀδελφὸν αὐτοῦ, διέλθωμεν εἰς τὸ πεδίον (נֵלכָה הִשָּׂדֶה). The Samaritan Pentateuch and the Syriac version agree with the Sept., and the passage is thus cited by Clemens Romanus (Ep. 1, 4). The Hebrew transcriber's eye was probably misled by the word שָׂדֶה terminating both the clauses.

In all the foregoing cases we do not attribute any paramount authority to the Sept. on account of its superior antiquity to the extant Hebrew MSS., but we take it as an evidence of a more ancient Hebrew text, as an eyewitness of the texts, 280 or 180 years B.C. The decision as to any particular reading must be made by weighing this evidence, together with that of other ancient versions, with the arguments from the context, the rules of grammar, the genius of the language, and the comparison of parallel passages. Thus the Hebrew will sometimes correct the Greek, and sometimes the Greek the Hebrew; both liable to err through the infirmity of human eyes and hands, but each checking the other's errors.

2. The close connection between the Old and the New Test. makes the study of the Sept. extremely valuable, and almost indispensable to the theological student. Pearson quotes from Irenaeus and Jerome as to the citation of the words of prophecy from the Sept. The former, as Pearson observes, speaks too universally when he says that the apostles "prophetica omnia ita enunciaverunt quemadmodum Seniorum interpretatio continet." But it was manifestly the chief storehouse from which they drew their proofs and precepts. Grinfield says that "the number of direct quotations from the Old Test. in the Gospels, Acts, and Epistles may be estimated at 350, of which not more than fifty materially differ from the Sept. But the indirect verbal allusions would swell the number to a far greater amount" (Apol. for LXX, p. 37). The comparison of the citations with the Sept. is much facilitated by Grinfield's Editio Hellenistica of the New Test., and by Gough's New Test. Quotations, in which the Hebrew and Greek passages of the Old Test. are placed side by side with the citations in the New. (On this subject see Hody, p. 248, 281; Kennicott, Dissert. Genesis § 84; Cappelli Critica Sacra, vol. 2.)

3. Further, the language of the Sept. is the mold in which the thoughts and expressions of the apostles and evangelists are cast. In this version Divine Truth has taken the Greek language as its shrine, and adapted it to the things of God. Here the peculiar idioms of the Hebrew are grafted upon the stock of the Greek tongue; words and phrases take a new sense. The terms of the Mosaic ritual in the Greek version are employed by the apostles to express the great truths of the Gospel, e.g. ἀρχιερεύς, θυσία, ὀσμὴ εὐωδίας. Hence the Sept. is a treasury of illustration for the Greek Testament. Many examples are given by Pearson (Proef. ad LXX), e.g. σάρξ, πνεῦμα, δικαιόω, φρόνημα τῆς σαρκός,, "Frustra apud veteres Graecos quaeras quid sit πιστεύειν τῷ Θεῷ vel εἰς τὸν Θεόν quid sit εἰς τὸν Κύριον, vel πρὸς τὸν Θεὸν πίστις, quae toties in Novo Foedere inculcantur, et ex lectione Seniorum facile intelliguntur." Valckenaer also (on Luke 1:51) speaks strongly on this subject: "Graecum Novi Testamenti contextum rite intellecturo nihil est utilius, quam diligenter versasse Alexandrinam Antiqui Foederis interpretationem, e qua una plus peti poterit auxilii, quam ex veteribus scriptoribus Graecis simul sumtis. Centena reperientur in N.T. nusquam obvia in scriptis Graecorum veterum, sed frequentata in Alexa. versione." E.g. the sense of

τὸ πάσχα in De 16:2, including the sacrifices of the Paschal week, throws light on the question as to the day on which our Lord kept his last Passover, arising out of the words in Joh 18:28, ἀλλ᾿ ἵνα φάγωσι τὸ πάσχα.

4. The frequent citations of the Sept. by the Greek fathers, and of the Latin version of the Sept. by the fathers who wrote in Latin, form another strong reason for the study of the Sept. Pearson cites the appellation of Scaraboeus bonus applied to Christ by Ambrose and Augustine, as explained by reference to the Sept. in Hab 2:11, κάνθαρος ἐκ ξύλου.

5. On the value of the Sept. as a monument of the Greek language in one of its most curious phases, this is not the place to dwell. Our business is with the use of this version as it bears on the criticism and interpretation of the Bible; and we may safely urge the theological student who wishes to be "thoroughly furnished" to have always at his side the Sept. Let the Hebrew, if possible, be placed before him; and at his right, in the next place of honor, the Alexandrian version. The close and careful study of this version will be more profitable than the most learned inquiry into its origin; it will help him to a better knowledge both of the Old Test. and the New.

VII. Objects to be Attained by the Critical Scholar.

1. Among these a question of much interest, suggested above, still waits for a solution. In many of the passages which show a studied variation from the Hebrew (some of which are above noted), the Sept. and the Samaritan Pentateuch agree — e.g. Ge 2:2; Ex 12:40.

They also agree in many of the ages of the post-diluvian patriarchs, adding one hundred years to the age at which the first son of each was born, according to the Hebrew (see Cappelli Critica Sacaa, 3, 20, 7). SEE PATRIARCH.

They agree in the addition of the words διέλθωμεν εἰς τὸ πεδίον (Ge 4:8), which many have seen reason to think rightly added.

Various reasons have been conjectured for this agreement — translation into Greek from a Samaritan text, interpolation from the Samaritan into the Greek, or vice versa; but the question does not seem to have found a satisfactory answer (see § 2 above).

2. For the critical scholar it would be a worthy object of pursuit to ascertain as nearly as possible the original text of the Sept. as it stood in the time of the apostles and Philo. If this could be accomplished with any tolerable completeness, it would possess a strong interest, as being the first translation of any writing into another tongue, and the first repository of divine truth to the great colony of Hellenistic Jews at Alexandria.

The critic would probably take as his basis the Roman edition from the Codex Vaticanus as representing most nearly the ancient (κοινή) texts. The collection of fragments of Origen's Hexapla, by Montfaucon and others, would help him to eliminate the additions which have been made to the Sept. from other sources, and to purge out the glosses and double renderings; the citations in the New Test. and in Philo, in the early Christian fathers, both Greek and Latin, would render assistance of the same kind; and perhaps the most effective aid of all would be found in the fragments of the old Latin version collected by Sabbatier in 3 vols. fol. (Rheims, 1743).

3. Another work of more practical and general interest still remains to be done, viz. to provide a Greek version, accurate and faithful to the Hebrew original, for the use of the Greek Church, and of students reading the Scriptures in that language for purposes of devotion or mental improvement. Field's edition is as yet the best of this kind. It originated in the desire to supply the Greek Church with such a faithful copy of the Scriptures; but as the editor has followed the text of the Alexandrian MS., only correcting, by the help of other MSS., the evident errors of transcription (e.g. in Ge 15:15, correcting τραφείς in the Alexandrian MS. to ταφείς, the reading of the Complut. text), and as we have seen above that the Alexandrian text is far from being the nearest to the Hebrew, it is evident that a more faithful and complete copy of the Old Test. in Greek might yet be provided.

We may here remark, in conclusion, that such an edition might prepare the way for the correction of the blemishes which remain in our authorized English version. Embracing the results of the criticism of the last two hundred and fifty years, it might exhibit several passages in their original purity; and the corrections thus made, being approved by the judgment of the best scholars, would probably, after a time, find their way into the margin at least of our English Bibles.

One example only can be here given, in a passage which has caused no small perplexity and loads of commentary. Isa 9:3 is thus rendered in the Sept.: To πλεῖστον τοῦ λαοῦ, ὅ κατήγαγες ἐν εὐφροσύνῃ σου καὶ εὐφρανθήσονται ἐνώπιόν σου, ώς οἱ εὐφραινόμενοι ἐν ἀμήτῳ, καὶ ὃν τρόπον οἱ διαιρούμενοι σκῦλα It is easy to see how the faulty rendering of the first part of this has arisen from the similarity of the Hebrew letters ת and ה, ד, and ר, and from an ancient error in the Hebrew text. The following translation restores the whole passage to its original clearness and force:

Ε᾿πλήθυνας τὴν ἀγαλλίασιν (הִגַּיל) ἐμεγαλύνας τὴν εὐφροσύνην· εὐφραίνονται ἐνώπιόν σου ὡς οἱ εὐφραινόμενοι ἐν ἀμήτῳ, ὃυ τρόπον ἀγαλλιῶνται οἱ διαιρούμενοι σκῦλα.

"Thou hast multiplied the gladness, Thou hast increased the joy; They rejoice before thee as with the joy of harvest, As men are glad when they divide the spoil."

Here ἀγαλλίασις and ἀγαλλιῶνται, in the first and fourth lines, correspond to גַּיל and יָגַילוּ; εὐφροσύνη and εὐφραίνονται, in the second and third lines, to שַׂמחָה and שָׂמחוּ. The fourfold introverted parallelism is complete, and the connection with the context of the prophecy perfect.

It is scarcely necessary to remark that in such an edition the apocryphal additions to the book of Esther, and those to the book of Daniel, which are not recognized by the Hebrew canon, would be either omitted or (perhaps more properly, since they appear to have been incorporated with the Sept. at an early date) would be placed separately, as in Field's edition and our English version. SEE APOCRYPHA; SEE CANON; SEE DANIEL, BOOK OF; SEE ESTHER, BOOK OF; SEE SAMARITAN PENTATEUCH.

VIII. Manuscripts and Early Critical Labors. —

1. The various readings given by Holmes and Parsons enable us to judge, in some measure, of the character of the several MSS. and of the degree of their accordance with the Hebrew text. Many other MSS., chiefly fragments, have since been brought to light by Tischendorf, and most of them have been published in his Monum. Sacra Ined. They are distinguished thus by Holmes: the uncial by Roman numerals, the cursive by Arabic figures. Among them may be specially noted, with their probable dates and estimates of value as given by Holmes in his preface to the Pentateuch:

UNCIAL.— PROBABLE DATE CENTURY.

(Sinaiticus. Royal Library, St. Petersburg]—. 4

1. COTTONIANUS. Brit. Mus. (fragments) — 4 2. VATICANUS. Vat. Library, Rome — 4 3. ALEXANDRINUS. Brit. Mus — 5 7. AMBROSIANUS. Ambros. Lib., Milan — 7 10. COISLINIANUS. Bibl. Nat., Paris — 7

CURSIVE.

16. Mediceus. Med. Laurentian Lib., Florence — 11 19. Chigianus. Similar to Complut. text and 108, 118 — 10 25. Monachiensis. Munich — 10 58. Vaticanus (No. x). Vat. Lib., similar to 72 — 13 59. Glasguensis — 12 61. Bodleianus. Laud. 86, notae optimae — 12 64. Parisiensis (11). National Library — 10 or 11 72. Venetus. Maximi faciendus — 13 75. Oxoniensis. Univ. Coll — 12 84. Vaticanus (1901), notae optimae — 11

106. 107.} Ferrarienses. These two agree — 14

108. {Vaticanus (330). Similar to Complut —14

118. {Parisiensis. Nat. Lib. text and (19) — 13

The texts of these MSS. differ considerably from each other, and consequently differ in various degrees from the Hebrew original (see Grabe, De Variis Vitiis LXX, etc. [Oxf. 1710]).

The following are the results of a comparison of the readings in the first eight chapters of Exodus:

(1.) Several of the MSS. agree well with the Hebrew; others differ very much.

(2.) The chief variance from the Hebrew is in the addition, or omission, of words and clauses.

(3.) Taking the Roman text as the basis, there are found eighty places (a) where some of the MSS. differ from the Roman text, either by addition or omission, in agreement with the Hebrew; twenty-six places (β) where differences of the same kind are not in agreement with the Hebrew. There is therefore a large balance against the Roman text in point of accordance with the Hebrew.

(4.) Those MSS. which have the largest number of differences of class (a) have the smallest number of class (β). There is evidently some strong reason for this close accordance with the Hebrew in these MSS.

(5.) The divergence between the extreme points of the series of MSS. may be estimated from the following statement:

72 differs from the Roman text — in 40 places, with Hebrew, in 4 places against Hebrew

59 differs from the Roman text — in 40 places with Hebrew, in 9 places against Hebrew Between these and the Roman text lie many shades of variety. The Alexandrine text falls about half-way between the two extremes:

Differing from Roman text { in 25 places, with Hebrew. in 16 places against Hebrew.

The diagram below, drawn on a scale representing the comparison thus instituted (by the test of agreement with the Hebrew in respect of additions or omissions), may help to bring these results more clearly into view. The baseline R.T. represents the Roman text.

T 72. Venetus. 59. Glasguensis. 58. Vaticanus (No. 10). 10. COISLINIANAS. 16. Mediceus. 7. AMBROSTANUS. Ed. Complutensis Codd. 19, 108, 118. 3. ALEXANDRINUS.

84. Vaticanus (1901). R Edit. Aldina.

The above can only be taken as an approximation, the range of comparison being limited. A more extended comparison might enable us to discriminate the several MSS. more accurately, but the result would perhaps hardly repay the labor.

2. But whence these varieties of text? Was the version at first more in accordance with the Hebrew, as in (72) and (59), and did it afterwards degenerate into the less accurate state of the Codex Vaticanus? Or was the version at first less accurate, like the Vatican text, and afterwards brought, by critical labors, into the more accurate form of the MSS. which stand highest in the scale?

History supplies the answer. Jerome (Ep. ad Suniam et Fretelam, 2, 627) speaks of two copies, one older and less accurate, κοινή, fragments of which are believed to be represented by the still extant remains of the old Latin version; the other more faithful to the Hebrew, which he took as the basis of his own new Latin version. In another place (Proefat. in Paralip. vol. 1, col. 1022) he speaks of the corruption of the ancient translation, and the great variety of copies used in different countries:

"Cum germana illa antiquaque translatio corrupta sit... Alexaudriat et AEgyptus in Sept. suis Hesychium laudant auctorem; Constantinopolis usque Antiochiam Luciani Martyris exemplaria probat; mediae inter has provinciae Palaestinos codices legunt: quos ab Origene elaboratos Eusebius et Pamphilus vulgaverunt: totusque orbis hac inter se contraria varietate compugnat." The labors of Origen, designed to remedy the conflict of discordant copies, are best described in his own words (Comment. in Matthew 1, 381, ed. Huet.):

"Now there is plainly a great difference in the copies, either from the carelessness of scribes, or the rash and mischievous correction of the text by others, or from the additions or omissions made by others at their own discretion. This discrepance in the copies of the Old Covenant we have found means to remedy, by the help of God, using as our criterion the other versions. In all passages of the Sept. rendered doubtful by the discordance of the copies, forming a judgment from the other versions, we have preserved what agreed with them; and some words we have marked with an obelos as not found in the Hebrew, not venturing to omit them entirely; and some we have added with asterisks affixed, to show that they are not found in the Sept., but added by us from the other versions, in accordance with the Hebrew." The other ἐκδόσεις, or versions, are those of Aquila, Theodotion, and Symmachus. Origen (Comm. in Joann. 2, 131, ed. Huet.) says, "The same errors in names may frequently be observed in the law and the prophets, as we have learned by diligent inquiry of the Hebrews, and by comparing our copies with their copies, as represent in the still uncorrupted versions of Aquila, Theodotji, and Symmachus." It appears from these and other passages that Origen, finding great discordance in the several copies of the Sept., laid this version side by side with the other three translations, and, taking their accordance with each other as the test of their agreement with the Hebrew, marked the copy of the Sept. with an obelos, — , where he found superfluous words, and supplied the deficiencies of the Sept. by words taken from the other versions with an asterisk, *, prefixed. The additions to the Sept. were chiefly made from Theodotion (Jerome, Prolog. in Genesin, vol. 1; see also Proef. in Job, p. 795). From Eusebius, as quoted below, we learn that this work of Origen was called Τετραπλᾶ, the fourfold Bible. The following specimen is given by Montfaucon:

Ge 1:1

ΑΚΥΛΑΣ ΣΥΜΜΑΧΟΣ ΟιΟ῾ ΘΕΟΔΟΤΙΩΝ ἐν κεφαλαίῳ ἔκτισεν ὁ Θεός σύν τὸν οὐρανὸν καὶ σὺν τὴν γῆν ἐν ἀρχῆ ἔκτισεν ὁ Θεὸς τὸν οὐρανὸν καὶ τὴν γῆν ἐν ἀρχῇ ἐποίησέν ὁ Θεὸς καὶ τὴν γῆν ἐν ἀρχῇ ἔκτισεν ὁ Θεός τὸν οὐρανόν καὶ τὴν γῆν But this was only the earlier and the smaller portion of Origen's labors: he rested not till he had acquired the knowledge of Hebrew, and compared the Sept. directly with the Hebrew copies. Eusebius (Hist. Eccl. 6, 16, p. 217, ed. Vales.) thus describes the labors which led to the greater work, the Hexapla; the last clause of the passage refers to the Tetrapla:

"So careful was Origen's investigation of the sacred oracles that he learned the Hebrew tongue, and made himself master of the original Scriptures received among the Jews in the Hebrew letters; and reviewed the versions of the other interpreters of the Sacred Scriptures, besides the Sept.; and discovered some translations varying from the well-known versions of Aquila, Symmachus, and Theodotion, which he searched out, and brought to light from their long concealment in neglected corners;... and in his Hexapla, after the four principal versions of the Psalms, added a fifth, yea, a sixth and seventh translation, stating that one of these was found in a cask at Jericho, in the time of Antoninus, son of Severus: and bringing these all into one view, and dividing them in columns over against one another, together with the Hebrew text, he left to us the work called Hexapla; having arranged separately, in the Tetrapla, the versions of Aquila, Symmachus, and Theodotion, together with the version of the Seventy." So Jerome (in Catal. Script. Eccl. vol. 4, pt. 2, p. 116):

"Quis ignorat, quod tantum in Scripturis divilis habuerit studii, ut etiam hebraeam linguam contra aetatis gentisque suae naturam edisceret; et acceptis LXX interpretibus, alias quoque editiones in unum volumen congregaret: Aquilae scilicet Pontici proselyti, et Theodotionis Ebionaei, et Symmachi ejusdem dogmatis....

Praeterea quintam et sextam et septimam editiouem, quas etiam nos de ejus bibliotheca habemus, miro labore reperit, et cum caeteris editionibus comparavit." From another passage of Jerome (in Epist. ad Titum, vol. 4, pt. 1, p. 437) we learn that in the Hexapla the Hebrew text was placed in one column in Hebrew letters, in the next column in Greek letters:

HEXAPLA Ho 11:1

Το ΕΒΡΑΙΚΟΝ Το ΕΒΡ ΕΛΛΗΝΙΚΟΙ Σ ΓΡ ΑΚΥΛΑΣ ΣΥΜΜΑΧΟΣ Οἰ ο ΘΕΟΔΟΤΙ ΩΝ קראתי לבני וממצוים ואהבהו כו נער ישׁראלχι νερ Ισραηλ ουεαβηου ουμεμεσραι μ καραθι λεβανι οτι παις Ισραηλ και ηγαπησα αυτον και απο Αιγυπτου εκαλεσα τον υιον μου οτι παις Ισραηλ και ηγαπημενος εξ Αιγυπτου κεκληται υιος μου οτι νηπιος Ισραηλ και εγω ηγαπησα αυτον και εξ Αιγυπτου κεκληται υιος μου οτι νηπιος Ισραηλ και ηγαπησα αυτον και εκαλεσα υιον μου εξ Αιγυπτου It should here be mentioned that some take the Tetrapla as denoting, not a separate work, but only that portion of the Hexapla which contains the four columns filled by the four principal Greek versions. Valesius (Notes on Eusebius, p. 106) thinks that the Tetrapla was formed by taking those four columns out of the Hexapla, and making them into a separate book. But. the testimony of Origen himself (2, 381; 2, 131), above cited, is clear that he formed one corrected text of the Sept. by comparison of the three other Greek versions (Α., Σ., Θ.), using them as his criterion. If he had known Hebrew at this time, would he have confined himself to the Greek versions? Would he have appealed to the Hebrew, as represented by Aquila, etc.? It seems very evident that he must have learned Hebrew at a later time, and therefore that the Hexapla, which rests on q comparison with the Hebrew, must have followed the Tetrapla, which was formed by the help of Greek versions only. The words of Eusebius also (Hist. Eccl. 6, 16) appear to distinguish very clearly between the Hexapla and Tetrapla as separate works, and to imply that the Tetrapla preceded the Hexapla. The order of precedence is not a mere literary question; the view above stated, which is supported by Montfaucon, Usher, etc., strengthens the force of Origen's example as a diligent student of Scripture, showing his increasing desire integros accedere fontes.

The labors of Origen, pursued through a long course of years, first in procuring by personal travel the materials for his great work, and then in comparing and arranging them, made him worthy of the name Adamantius. But what was the result of all this toil? Where is now his great work, the Hexapla, prepared with so much care, and written by so many skilful hands? Too large for transcription, too early by centuries for printing (which alone could have saved it), it was destined to a short existence. It was brought from Tyre and laid up in the library at Caesarea, and there probably perished by the flames, A.D. 653. One copy, however, had been made, by Pamphilus and Eusebius, of the column containing the corrected text of the Sept., with Origen's asterisks and obeli, and the letters denoting from which of the other translators each addition was taken. This copy is probably the ancestor of those codices which now approach most nearly to the Hebrew, and are entitled Hexaplar; but in the course of transcription the distinguishing marks have disappeared or become confused; and we have thus a text composed partly of the old Sept. text, partly of insertions from the three other chief Greek versions, especially that of Theodotion.

The facts above related agree well with the phenomena of the MSS. before stated. As we have codices derived from the Hexaplar text (e.g. 72, 59, 58), and at the other extreme the Codex Vaticanus (II), probably representing nearly the ancient uncorrected text, κοινή; so between these we find texts of intermediate character in the Codex Alexandrinus (III), and others, which may perhaps be derived from the text of the Tetrapla.

To these main sources of our existing MSS. must be added the recensions of the Sept. mentioned by Jerome and others, viz. those of Lucian of Antioch and Hesychius of Egypt, not long after the time of Origen. We have seen above that each of these had a wide range that of Lucian (supposed to be corrected by the Hebrew) in the churches from Constantinople to Antioch; that of Hesychius in Alexandria and Egypt; while the churches lying between these two regions used the Hexaplar text copied by Eusebius and Pamphilus (Jerome, vol. 1, col. 1022). The great variety of text in the existing MSS. is thus accounted for by the variety of sources from which they have descended.

IX. Modern Editions. —

1. This version appears at the present day in five principal editions:

1. Biblia Polyglotta Complutensis (1514-17). 2. The Aldine edition (Venice, 1518). 3. The Roman edition, edited under pope Sixtus V (1587). 4. Facsimile edition of the Codex Alexandrinus, by Baber (1816). 5. Facsimile edition of the Codex Sinaiticus, by Tischendorf (St. Petersburg, 1862, 4 vols. fol.).

The texts of (1) and (2) were probably formed by collation of several MSS. The Roman edition (3) is printed from the venerable Codex Vaticanus, but not without many errors. This text has been followed in most of the modern editions. A transcript of the Codex Vaticanus, prepared by cardinal Mai, was lately published at Rome by Vercelloni It is much to be regretted that this edition is not so accurate as to preclude the necessity of consulting the MS. The text of the codex, and the parts added by a later hand, to complete the codex (among them nearly all of Genesis), are printed in the same Greek type, with distinguishing notes. The facsimile edition by Baber (4) is, printed with types made after the form of the letters in the Codex Alexandrinus (British Museum Library) for the facsimile edition of the New Test. by Woide in 1786. Great care was bestowed upon the sheets as they passed through the press. The Codex Sinaiticus (5) was published in facsimile type at the expense of the emperor of Russia, and a very limited edition was printed. SEE SINAITIC MS.

2. Other important editions are the following: The Septuagint in Walton's Polyglot (1657) is the Roman text, with the various readings of the Codex Alexandrinus. The Cambridge edition (1665) (Roman text) is only valuable for the preface by Pearson. An edition of the Codex Alexandrinus was published by Grabe (Oxford, 1707-20), but its critical value is far below that of Baber's. It is printed in common type, and the editor has exercised his judgment on the text, putting some words of the codex in the margin,: and replacing them by what he thought better readings, distinguished by a smaller type. This edition was reproduced by Breitinger (Zurich, 1730-32, 4 vols. 4to), with the various readings of the Vatican text. The edition of Bos (Franeq. 1709) follows the Roman texts with its scholia, and the various readings given in Walton's Polyglot, especially those of the Codex Alexandrinus. This has often been reprinted, and is now the commonest text. The valuable critical edition of Holmes, continued by Parsons, is similar in plan to the Hebrew Bible of Kennicott; it has the Roman text, with a large body of various readings from numerous MSS. and editions (Oxford, 1798-1827). The Oxford edition by Gaisford (1848) has the Roman text, with the various readings of the Codex Alexandrinus below. Tischendorf's editions (the 5th, 1875) are on the same plan; he has added readings from some other MSS. discovered by himself, with very useful Prolegomena. Some convenient editions have been published by Bagster, one in 8vo, others of smaller size, forming part of his Polyglot series of Bibles. His text is the Roman. The latest edition, by Field (1859), differs from any of the preceding. He takes as his basis the Codex Alexandrinus, but corrects all the manifest errors of transcription by the help of other MSS., and brings the dislocated portions of the Septuagint into agreement with the order of the Hebrew Bible. The text in Stier and Theile's Polyglotten Bibel (Bielefeld, 1854) is revised arbitrarily, and without the aid of the Codex Sinaiticus. Scrivener has promised a new critical edition.

3. Editions of particular books, more or less critically prepared, have occasionally been issued: Genesis, by Lagarde (Lips. 1868); Esther, by Fritzsche (Turici, 1848); Ruth, by the same (ibid. 1867); Jeremiah, by Spohn (Lips. 1794-1828); Ezekiel, by Vincent (Romans 1840); Jonah, by Hohner (Lips. 1787-88). The genuine text of Daniel (which was long supposed to be lost, the translation of Theodotion having been substituted for it in the common MSS.) was first published separately by Simon de Magistris in 1772, from the Codex Chigianus; and it was reprinted by J.D. Michaelis (1773-74), Segaar (1775), and more critically by Hahn (1845), from the Codex Ambrosianus.

The best Lexicon to the Septuagint is that of Schleusner, published at Leipsic (1820-21, 5 pts.), and reprinted at Glasgow (1822, 3 vols. 8vo).. An earlier one is that of Biel (Hag. 1779-80, 3 vols.). The best for the Apocrypha is Wahl's Clavis (Lips. 1863). The best Concordance is that of Trommius (Amst. 1718, 2 vols. fol.). An earlier one is that of Kircher (1607). Winer's V.T. Grammar serves an excellent purpose for philological comparison. The student may also consult Sturz, De Dialecto Macedonica (Lips. 1808); Maltby, Two Sermons before the University of Durham (1843). SEE GREEK LANGUAGE.

X. Literature. — In addition to the works named by Walch, Bibl. Theol. 4, 31 sq., 156 sq.; Rosenmüller, Handb. d. Literatur, 2, 279 sq.; and Danz; Wörterb. d. Theol. s.v. "Alex. Vers.," the following are important: Cappelli Critica Sacra (Par. 1650); Waltoni Proleg. ad Bibl. Polyglott. (Lond.

1657); Pearsoni [Bp.] Pref. Paroenetica ad LXX (ibid. 1655); Vossius, De LXX Interp. (Hag. 1661; app. 1663); Montfaucon, Hexaplorum Origenis quoe Supersunt (Par. 1710; Lips. 1740); Hody, De Bibl. Text. Original. Vers. Grecis, et Latina Vulgata (Oxf. 1704); Hottinger, Thesaurus (Zur. 1649); Owen, Inquiry into the Sept. (Lond. 1769); Brief Account, etc. (ibid. 1787); Kennicott. Dissertationes to his Vet. Test. (Oxon. 1776-80); Wornier, De LXX Interpretibus (Hamb. 1617, 8vo); Knapp, De Versione Alex. (Hal. 1775-76, 4to); Hasenkamp, De Pentat. LXX Interp. (Marb. 1765, 4to); Stroth, Symboloe Criticoe (Lips. 1778-83); Sulzner, De LXX Interp. (Hal. 1700, 4to ); Weyhenmeyer, De Fersione LXX (Ulm. 1719, 4to); Reineke, De Dissensu Vers. Alex. ab Archetypo (Magd. 1771, 4to); Holmes, Prolegg. ad LXX (Oxf. 1798-1827); Valckenaer, Diatribe de Aristobulo Judoeo (L.B. 1806); Schleusner, Opusc. Crit. ad Verss. Gr. V.T. (Lips. 1812); Dähne, Jüdisch-alexandrinische Philosophie (Hal. 183134); Töpler, De Pentat. Interp. Alex. Indole Crit. et Hermen. (Hal. Sax. 1830); Gfrörer, Urchristenthum (Stuttg. 1831, 8vo); Fabricii Bibliotheca Sacra, ed. Harless, vol. 3; Studer, De Versionis Alexandrinoe Origine, Historia, Usu, et Abusu Critico (Bernie, 1823, 8vo); Credner, Beiträge zur Einleitung, etc. (Halle, 1838, 2 vols. 8vo); Amersfoordt, Dissertatio de Variis Lectionibus Holmesianis (Lugd. Bat. 1815, 4to); Plüschke, Lectiones Alex. et Hebr. (Bonn, 1837); Thiersch, De Pent. Fers. Alex. (Erlang. 1841); Frankel, Vorstudien zu der Septuaginta (Leips. 1841); Ueber den Einfluss der palästinischen Exegese, auf die alex. Hermeneutik (ibid. 1851); Grinfield, N.T. Editio Hellenistica (ibid. 1848), and Apology for the Septuagint (ibid. 1850); Selwyn, Notoe Criticoe in Ex. 1-24, Numeros, Deuteronomium (ibid. 1856-58); also Hor. Hebr. on Isaiah 9 (ibid. 1848); Churton, Hulsean Essay (ibid. 1861); Pearson [G.], Papers, in the Journal of Sacred Lit. 1, 4, 7, 3d series.