CHAPTER II.

LATER GREEK VERSIONS.

1. At Alexandria and in Egypt generally the Alexandrian version was regarded, as Philo plainly says, with a reverence scarcely less than that which belonged to the original. It was the Bible of the Egyptian Jews, even of those who belonged to the educated and literary class. This feeling was shared by the rest of the Hellenistic world. In Palestine indeed the version seems to have been received with less enthusiasm, and whether it was used in the synagogues is still uncertain. But elsewhere its acceptance by Greek-speaking Jews was universal during the Apostolic age and in the next generation.

2. When the LXX. passed into the hands of the Church and was used in controversy with Jewish antagonists, the Jews not unnaturally began to doubt the accuracy of the Alexandrian version (Justin, dial. 68 τολμῶσι λέγειν τὴν ἐξήγησιν ἣν ἐξηγήσαντο οἱ ἑβδομήκοντα ὑμῶν πρεσβύτεροι παρὰ Πτολεμαίῳ τῷ τῶν Αἰγυπτίων βασιλεῖ γενόμενοι μὴ εἶναι ἔν τισιν ἀληθῆ). The crucial instance was the rendering of צַלְטָה by παρθένος in Isa. vii. 14, where νεᾶνις, it was contended, would have given the true meaning of the Hebrew word (ib. 71, 84; Iren. iii. 21. 1). But the dissatisfaction with which the LXX. was regarded by the Jewish leaders of the second century was perhaps not altogether due to polemical causes. The LXX. "did not suit the newer school of [Jewish] interpretation, it did not correspond with the received text⁸¹⁸¹Robertson Smith, The O. T. in the J. Ch., p. 64; cf. ib. p. 87 f.; Kirkpatrick, Divine Library, p. 63 ff.; cf. Buhl, p. 118 f.." An official text differing considerably from the text accepted in earlier times had received the approval of the Rabbis, and the Alexandrian version, which represented the older text, began to be suspected and to pass into disuse. Attempts were made to provide something better for Greek-speaking Israelites (Justin, dial. 71 αὐτοὶ ἐξηγεῖσθαι πειρῶνται). Of two such fresh translations Irenaeus speaks in terms of reprehension (l.c. οὐχ ὡς ἔνιοί φασιν τῶν νῦν μεθερμηνεύειν τολμώντων τὴν γραφήν . . . ὡς Θεοδοτίων . . . ὁ Ἐφέσιος καὶ Ἀκύλας ὁ Ποντικός, ἀμφότεροι Ἰουδαῖοι προσήλυτοι). Origen, who realised the importance of these translations, was able to add to those of Aquila and Theodotion the version of Symmachus and three others which were anonymous⁸²⁸²Eus. H. E. vi. 16.. Of the anonymous versions little remains, but Aquila, Theodotion, and Symmachus are represented by numerous and in some cases important fragments.

3. Aquila. The name had been borne in the Apostolic age by a native of Pontus who was of Jewish birth (Acts xviii. 2 Ἰουδαῖον ὀνόματι Ἀκύλαν, Ποντικὸν τῷ γένει). Aquila the translator was also of Pontus, from the famous sea-port⁸³⁸³Ramsay, Hist. Geogr. of Asia Minor, p. 27 f.; cf. Hort, Commentary on Peter, p. 172 ff. Sinope, which had been constituted by Julius Caesar a Roman colony; but he was of Gentile origin. He lived in the reign of Hadrian (A.D. 117—138), and was a connexion of the Emperor (πενθερίδης, Epiph., Dial. of Timothy and Aquila; πενθερός, Ps.-Ath., Chron. Pasch.). Hadrian employed his relative to superintend the building of Aelia Capitolina on the site of Jerusalem, and while there Aquila was converted to Christianity by Christians who had returned from Pella. Refusing, however, to abandon the pagan practice of astrology, he was excommunicated; upon which he shewed his resentment by submitting to circumcision and attaching himself to the teaching of the Jewish Rabbis. The purpose of his translation was to set aside the interpretation of the LXX., in so far as it appeared to support the views of the Christian Church.

That Aquila was a proselyte to Judaism is attested by the Jewish tradition (Jer. Talm. Meg. 1. 11, Kidush. 1. 1), in which he appears as הַגֵּר, ὁ προσήλυτος⁸⁴⁸⁴The name is written קילם ,אקילם ,עקילם or עקיעלם, and in the Bab. Talmud, אגקלום. On the identity of Aquila with Onkelos see Anger de Onkelo Chaldaico (before 1845), Friedmann Onkelos u. Akylas (Wien, 1896); or the brief statement in Buhl, p. 173.. After his conversion to Judaism, Aquila became a pupil of R. Eliezer and R. Joshua (Meg. f. 71 c) or, according to another authority, of R. Akiba (Kiddush. f. 59 a). The latter statement seems to have been current among the Jews of Palestine in Jerome's time (Hieron. in Isa. viii, 14 "scribae et Pharisaei quorum suscepit scholam Akybas, quem magistrum Aquilae proselyti autumant"), and it derives some confirmation from the character of the version.

According to Epiphanius the floruit of Aquila is to be placed in the 12th year of Hadrian (Epiph. de mens. et pond. 13 Ἁδριανὸς ἔτη κά, οὗτινος τῷ δωδεκάτῳ ἔτει Ἀκύλας ἐγνωρίζετο . . . ὡς εἶναι ἀπὸ τοῦ χρόνου τῆς ἑρμηνείας τῶν οβʹ ἑρμηνευτῶν ἕως Ἀκύλα τοῦ ἑρμηνευτοῦ, ἤγουν ἕως δωδεκάτου ἔτους Ἁδριανοῦ, ἔτη υλʹ καὶ μῆνας δʹ. The 12th year of Hadrian was A.D. 128—9, the year in which the Emperor began to rebuild Aelia. This date is doubtless approximately correct, if Aquila was a pupil of R. Akiba, who taught from A.D. 95 to A.D. 135⁸⁵⁸⁵Field, Hexapla, prolegg. p. xviii., or even of R. Eliezer and R. Joshua, who immediately preceded Akiba. It must have taken the Greek proselyte many years to acquire an adequate knowledge of Hebrew and of the Rabbinical methods of interpretation, and under these circumstances his great work could hardly have been completed before the fourth decade of the second century. When Irenaeus wrote his third book, in 33the ninth decade, Aquila's translation might still be regarded as comparatively recent (τῶν νῦν μεθερμηνεύειν τολμώντων τὴν γραφήν . . . ὡς . . . Ἀκύλας).

4. It was natural that the version of Aquila should be received with acclamation by his co-religionists. His teachers congratulated him in the words of Ps. xlv. 3, יָפְיָפִיתָ מִבְּנֵי אָדָם ‎⁸⁶⁸⁶Megilla 1. 9: in יפיפית there is a play upon יפת (cf. Gen. ix. 27).. The Talmud quotes or refers to his translation of not a few passages (Gen. xvii. 1; Lev. xix. 20, 23, 40; Esth. i. 6; Prov. xviii. 21, xxv. 11; Isa. iii. 20; Ezek. xvi. 10, xxiii. 43; Dan. v. 5, viii. 13). In Origen's time he was trusted implicitly in Jewish circles, and used by all Jews who did not understand Hebrew (ep. ad African. 2 φιλοτιμότερον πεπιστευμένος παρὰ Ἰουδαίοις . . . ᾦ μάλιστα εἰώθασιν οἱ ἀγνοοῦντες τὴν Ἐβραίων διάλεκτον χρῆσθαι, ὡς πάντων μᾶλλον ἐπιτετευγμένῳ); and the same preference for Aquila seems to have been characteristic of the Jews in the fourth and fifth centuries (cf. Jerome on Ezek. iii. 5, and Augustine de civ. Dei xv. z3), and at a still later period, for even Justinian, when regulating the public reading of the Scriptures in the synagogues, thought it expedient to permit the use of Aquila (novell. 146: "at vero ii qui Graeca lingua legunt LXX. interpretum utentur translatione . . . verum . . . licentiam concedimus etiam Aquilae versione utendi"). It was equally natural that the proselyte's version should be regarded with distrust by Christians, who saw in it the work of a champion of Rabbinism as well as a bold attempt to displace the Septuagint⁸⁷⁸⁷See Dr C. Taylor in the preface to Prof. Burkitt's Fragments of Aquila, p. vi.: "Aquila in a sense was not the sole or independent author of the version, its uncompromising literalism being the necessary outcome of his Jewish teachers' system of exegesis.". Yet the few Christian writers who were students of the Hebrew Bible learnt to recognise the fidelity of Aquila's work. He was 'a slave to the letter' (δουλεύων τῇ Ἐβραικῇ λέξει; whatever was wanting in the Hebrew text was not to be 34found in Aquila οὐ κεῖται παρὰ τοῖς Ἐβραίοις, διόπερ οὐδὲ παρὰ τῷ Ἀκύλᾳ). So Origen confesses⁸⁸⁸⁸Ep. ad Afric. 3. Cf. Aug. l.c.; and Jerome, though when in a censorious mood he does not spare the proselyte (e.g. praef. in Job, ep. ad Pammach.), elsewhere admits his honesty and diligence (ep. ad Damas. 12 "non contentiosius, ut quidam putant, sed studiosius verbum interpretatur ad verbum"; ep. ad Marcell. "iamdudum cum voluminibus Hebraeorum editionem Aquilae confero, ne quid forsitan propter odium Christi synagoga mutaverit, et—ut amicae menti fatear—quae ad nostram fidem pertineant roborandam plura reperio"). After these testimonies from the two most competent witnesses in the ancient Church, we need not stop to consider the invective of Epiphanius⁸⁹⁸⁹See p. 31..

5. Until the summer of 1897 Aquila's version was known to students only from the description of ancient writers, chiefly Christian, and the fragments of the Hexapla (c. iii.), which when complete contained the entire work. These sources were used with admirable skill by Dr Field (prolegomena in Hexapla, p. xix, ff.) and Dr C. Taylor (D. C. B. art. Hexapla) to illustrate the purpose and style of Aquila's work. But an unexpected discovery has since placed at our disposal several larger fragments of the version, emanating from a Jewish source. Among the débris of the Genizah of the Cairo synagogue brought to Cambridge in 1897 through the efforts of Dr Taylor and Dr Schechter, Professor Burkitt was so fortunate as to discover some palimpsest scraps which under later Hebrew writing contain in a good uncial hand of the sixth century Aquila's translation of 1 Kings xx. 9—17 and 2 Kings xxiii. 12—27⁹⁰⁹⁰Fragments of the Books of Kings according to the translation of Aquila (Cambridge, 1897).. From the same treasure Dr Taylor recovered portions of Pss. xc.-ciii., and a Hexaplar fragment of Ps. xxii.⁹¹⁹¹Hebrew-Greek Cairo Genizah Palimpsests (Camb. 1900). See also Amherst Papyri, i. p. 30 f. (London, 1900). 35The student will find below specimens of these discoveries, placed for the purpose of comparison in parallel columns with the version of the LXX.

3 Regn. xxi. (1 Kings xx.) 10—13.

4 Regn. (2 Kings) xxiii. 21—24.

Ps. xc. (xci.) 6b—13.

Ps. xci. (xcii.) 5—10.

6. If the student examines these specimens of Aquila's work and compares them with the Hebrew and LXX., the greater literalness of the later version and several of its most 39striking peculiarities will at once be apparent. He will notice especially the following. (1) There are frequent instances of an absolutely literal rendering of the original, e.g. 1 Kings xx. 10 ὃσ ἐν ποσίν μου = אֲשֶׁר בְּרַגְלָי (LXX. τοῖς πεζοῖς μου); 12 θέτε·καὶ ἔθηκαν = שִׂימוּ וַיָשִׂימוּ (LXX. οἰκοδομήσατε χάρακα, καὶ ἔθεντο χάρακα); 2 Kings xxiii. 21 τῷ λέγειν = לֵאמׂר (LXX. λέγων); 24 ἃ ὡράθησαν =
אֲשֶׁר נִרְאוּ (LXX. τὰ γεγονότα). (2) Under certain circumstances⁹⁹⁹⁹For these see Burkitt, Aquila, p. 12. σύν is employed to represent the Hebrew אֵת, when it is the sign of the accusative¹⁰⁰¹⁰⁰This singular use of σύν appears also in the LXX., but only in Ecclesiastes and the Song of Songs, which Freudenthal is disposed to assign to Aquila (p. 65); cf. König, Einleitung, p. 108 n., and McNeile, Introd. to Ecclesiastes.; e.g. 1 Kings xx. 12 σὺν τὸ ῥῆμα = אֶת־הַדָּבָר‎, 13 σὺν πάντα τὸν ὄχλον = אֶת־כָּל־הֶהָמוֹן‎, 2 Kings xxiii. 21 σὺν παντὶ τῷ λαῷ (where the dat. is governed by the preceding verb), 24 σὺν τοὺς μάγους κτλ. (3) The same Hebrew words are scrupulously rendered by the same Greek, e.g. καὶ καίγε = וְגַם occurs thrice in one context (2 Kings xxiii. 15, 19, 24); and in Ps. xcii. 8, 10 κατεργαζόμενοι ἀνωφελές twice represents פּעֲלֵי אָוֶן‎ (4) The transliterations adhere with greater closeness to the Hebrew than in the LXX.¹⁰¹¹⁰¹Aq. does not transliterate עחהא (see Burkitt, p. 14).; thus פֶּםַח becomes φέσα, יֹאׁשִׁיָּהוּ Ἰωσιαού, חִלְקִיָּהוּ Ἑλκιαού. (5) The Tetragrammaton is not transliterated, but written in Hebrew letters, and the characters are of the archaic type (, not יהוה); cf. Orig. in Ps. ii., καὶ ἐν τοῖς ἀκριβεστάτοις δὲ τῶν ἀντιγράφων Ἐβραίοις χαρακτῆρσιν κεῖται τὸ ὄνομα, Ἐβραικοῖς δὲ οὐ τοῖς νῦν ἀλλὰ τοῖς ἀρχαιοτάτοις—where the 'most exact copies' are doubtless those of Aquila's version, for there is no reason to suppose that any copyists of the Alexandrian version hesitated to write ο κς or κε for יהוה‎¹⁰²¹⁰²In a few Hexaplaric MSS. (e.g. Q, 86, 88, 243mg, 264) the Greek letters ΠΙΠΙ are written for יהוה, but (with the exception of the Genizah Palimpsest, Taylor, p. 27) the Greek MSS. use it solely in their excerpts from the non-Septuagintal columns of the Hexapla, and only the Hexaplaric Syriac admits ΠΙΠΙ into the text of the LXX., using it freely for κύριος, even with a preposition (as ). Oxyrh. Pap. 1007 (vol. VII.), late 3rd cent., has ZZ, representing doubled yod, in Gen. ii., iii. Ceriani expresses the opinion that the use of ΠΙΠΙ is due either to Origen or Eusebius, i.e. one of those fathers substituted ΠΙΠΙ for for in the non-Septuagintal columns, using the letters to represent the Hebrew characters which were familiar to them. On the whole subject the student may consult Ceriani, Monumenta sacra et profane, ii. p. 106 ff.; Schleusner, s.v. πίπι, Field, Hexapla ad Esa. i. 2; Hatch and Redpath, Concordance, p. 1135; Driver in Studia Biblia, i. p. 12, n. 3; Z. D. M. G. (1878), 465 ff., 501, 506. Prof. Burkitt acutely points out (p. 16) that (and doubtless also ΠΙΠΙ was read as Κύριος, since in one place in the Aquila fragments where there was no room to write the Hebrew characters "instead of οἴκῳ we find οἴκῳ κυ." On the orthography see Burkitt, p. 15, par. 4.. (6) That the crudities of Aquila's 40 style are not due to an insufficient vocabulary¹⁰³¹⁰³Even Jerome speaks of Aquila as "eruditissimus linguae Graecae" (in Isa. xlix. 5). is clear from his ready use of words belonging to the classical or the literary type when they appear to him to correspond to the Hebrew more closely than the colloquialisms of the LXX. The following are specimens; 1 Kings xx. 10 LXX. ἐκποιήσει, Aq. ἐξαρκέσει; LXX. ἀλώπεξιν, Aq. λιχάσιν¹⁰⁴¹⁰⁴See Prof. Burkitt's note (p. 26).; 12 LXX. σκηναῖς, Aq. συσκιασμοῖς; 2 Kings xxiii. 21 LXX. διαθήκης, Aq. συνθήκης; 24 LXX. θεραφείν, Aq. μορφώματα; LXX. εἴδωλα, Aq. καθάρματα; Ps. xc. 8 LXX. ἀνταπόδοσιν, Aq. ἀπότισιν; ib. 10 LXX. προσελεύσεται, Aq. μεταχθήσεται; LXX. μάστιξ, Aq. ἁφή; xci. 5 LXX. ποιήματι, Aq. κατέργῳ.

From the fragments which survive in the margins of hexaplaric MSS. it is possible to illustrate certain other characteristic features of Aquila which arise out of his extreme loyalty to the letter of his Hebrew text. (1) Jerome remarks upon his endeavour to represent even the etymological meaning of the Hebrew words (ad Pammach. 11 "non solum verba sed etymologias quoque verborum transferre conatus est)," and by way of example he cites the rendering of Deut. vii. 13, where Aquila substituted χεῦμα, ὀπωρισμόν, στιλπνότητα for σῖτον, οἶνον, ἔλαιον in order to reflect more exactly the Hebrew יִצְהָר ,תִּירשׁ ,דָּגָן—as though, adds Jerome humorously, we were to use in Latin fusio, pomatio, splendentia. Similarly, 41Aquila represented עִצֵּם by ὀστεοῦν, and הִשְׂכִּיל by ἐπιστημονίζειν or ἐπιστημονοῦν, and even coined the impossible form ἁφημένος to correspond with נָגוּעַ‎. (2) An attempt is made to represent Hebrew particles, even such as defy translation; thus ה local becomes the enclitic δε (e. g. νότονδε = הַנָּגְבָּה, Gen. xii. 9, Κυρήνηνδε = קִירָה‎, 2 Kings xvi. 9); and similarly prepositions are accumulated in a manner quite alien from Greek usage (e.g. εἰς ἀπὸ μακρόθεν = לְמֵרָחוֹק‎, 2 Kings xix. 25). (3) Other devices are adopted for the purpose of bringing the version into close conformity with the original; a word of complex meaning or form is represented by two Greek words (e.g. עֲזָאזֵל is converted into τράγος ἀπολυόμενος and צִלְצַלּ into σκιὰ σκιά; a Hebrew word is replaced by a Greek word somewhat similar in sound, e.g. for אֵלוֹן (Deut. xi. 30) Aquila gives αὐλών, and for תְּרָפִים ‎ (1 Sam. xv. 23) θεραπεία¹⁰⁵¹⁰⁵The student who wishes to pursue the subject may refer to Field, Prolegg. p. xxi. sqq., and Dr Taylor's article Hexapla in Smith and Wace's Dict. Chr. Biog. iii. p. 17 ff. Jerome speaks more than once of a second edition of Aquila "quam Hebraei κατ᾿ ἀκρίβειαν nominant." The question is discussed by Field (prolegg. xxiv. ff.)..

Enough has been said to shew the absurdity of Aquila's method when it is regarded from the standpoint of the modern translator. Even in ancient times such a translation could never have attained to the popularity which belonged to the LXX.; that it was widely accepted by the Greek synagogues of the Empire can only have been due to the prejudice created in its favour by its known adherence to the standard text and the traditional exegesis¹⁰⁶¹⁰⁶See Prof. Burkitt's article Aquila in the Jewish Quarterly Review, Jan. 1898, p. 211 ff.. The version of Aquila emanated from a famous school of Jewish teachers; it was issued with the full approval of the Synagogue, and its affectation of preserving at all costs the idiom of the original recommended it to orthodox Jews whose loyalty to their faith was stronger than their sense of the niceties of the Greek tongue. For ourselves the work of 42Aquila possesses a value which arises from another consideration. His "high standard of exactitude and rigid consistency give his translation, with all its imperfections, unique worth for the critic¹⁰⁷¹⁰⁷Dr Taylor, pref. to Fragments of Aquila, p. vii.." Its importance for the criticism of the Old Testament was fully recognised by the two greatest scholars of ancient Christendom, and there are few things more to be desired by the modern student of Scripture than the complete recovery of this monument of the text and methods of interpretation approved by the chief Jewish teachers of the generation which followed the close of the Apostolic age.

7. Theodotion. With Aquila Irenaeus couples Theodotion of Ephesus, as another Jewish proselyte who translated the Old Testament into Greek (Θεοδοτίων ἡρμήχευσεν ὁ Ἐφέσιος καὶ Ἀκύλας . . . ἀμφότεροι Ἰουδαῖοι προσήλυτοι). Himself of Asiatic origin, and probably a junior contemporary of Theodotion, Irenaeus may be trusted when he assigns this translator to Ephesus, and describes him as a convert to Judaism. Later writers, however, depart more or less widely from this statement. According to Epiphanius, Theodotion was a native of Pontus, who had been a disciple of Marcion of Sinope before he espoused Judaism. According to Jerome, he was an Ebionite, probably a Jew who had embraced Ebionitic Christianity. His floruit is fixed by Epiphanius in the reign of the second Commodus, i.e. of the Emperor Commodes, so called to distinguish him from L. Ceionius Commodus, better known as L. Aurelius Verus.

Epiph. de mens. et pond. 17 περὶ τὴν τοῦ δευτέρου Κομόδου βασιλείαν τοῦ βασιλεύσαντος μετὰ τὸν προειρημένον Κὸμοδον Λούκιον Αὐρήλιον ἔτη ιγʹ, Θεοδοτίων τις Ποντικὸς ἀπὸ τῆς διαδοχῆς Μαρκίωνος τοῦ αἱρεσιάρχου τοῦ Σινωπίτου, μηνίων καὶ αὐτὸς τῇ αὐτοῦ αἱρέσει καὶ εἰς Ἰουδαισμὸν ἀποκλίνας καὶ περιτμηθεὶς καὶ τὴν τῶν Ἐβραίων φωνὴν καὶ τὰ αὐτῶν στοιχεῖα παιδευθείς, ἰδίως καί αὐτὸς ἐξέδωκε. Hieron. ep. ad Augustin.: "hominis Judaei atque blasphemi"; 43praef. in Job: "Iudaeus Aquila, et Symmachus et Theodotio Judaizantes haeretici"; de virr. ill. 54 "editiones . . . Aquilae . . . Pontici proselyti et Theodotionis Hebionaei"; praef. ad Daniel.: "Theodotionem, qui utique post adventum Christi incredulus fuit, licet eum quidam dicant Hebionitam qui altero genere Iudaeus est¹⁰⁸¹⁰⁸Marcion flourished c. A.D. 150; Commodus was Emperor from 180—192. The Paschal Chronicle, following Epiphanius, dates the work of Theodotion A.D. 184.."

The date assigned to Theodotion by Epiphanius is obviously too late, in view of the statement of Irenaeus, and the whole account suspiciously resembles the story of Aquila. That within the same century two natives of Pontus learnt Hebrew as adults, and used their knowledge to produce independent translations of the Hebrew Bible, is scarcely credible. But it is not unlikely that Theodotion was an Ephesian Jew or Jewish Ebionite. The attitude of a Hellenist towards the Alexandrian version would naturally be one of respectful consideration, and his view of the office of a translator widely different from that of Aquila, who had been trained by the strictest Rabbis of the Palestinian school. And these expectations are justified by what we know of Theodotion's work. "Inter veteres medius incedit" (Hieron. praef. ad evang.); "simplicitate sermonis a LXX. interpretibus non discordat" (praef. in Pss.); "Septuaginta et Theodotio . . . in plurimis locis concordant" (in Eccl. ii.)—such is Jerome's judgement; and Epiphanius agrees with this estimate (de mens. et pond. 17: τὰ πλεῖστα τοῖς οβʹ συνᾳδόντως ἐξέδωκεν). Theodotion seems to have produced a free revision of the LXX. rather than an independent version. The revision was made on the whole upon the basis of the standard Hebrew text; thus the Job of Theodotion was longer than the Job of the LXX. by a sixth part of the whole (Orig. ep. ad Afric. 3 sqq., Hieron. praef. ad Job)¹⁰⁹¹⁰⁹See Field, Hexapla, p. xxxix.; Hatch, Essays, p. 215; Margoliouth, art. 'Job' in Smith's Bible Dict. (ed. 2)., and in Daniel, on 44the other hand, the Midrashic expansions which characterise the LXX. version disappear in Theodotion. His practice with regard to apocryphal books or additional matter appears not to have been uniform; he followed the LXX. in accepting the additions to Daniel and the supplementary verses in Job¹¹⁰¹¹⁰Orig. ep. ad Afric. 3., but there is no evidence that he admitted the non-canonical books in general¹¹¹¹¹¹On Baruch see Nestle's remarks in Hastings' D. B. iv. (art. Septuagint). .

8. Specimens of Theodotion's style and manner may be obtained from the large and important fragments of his work which were used by Origen to fill up the lacunae in Jeremiah (LXX.). The following passage, preserved in the margin of Codex Marchalianus, will serve as an example¹¹²¹¹²O. T. in Greek, iii. pp. vii. ff., 320 f..

Jeremiah xl. (xxxiii.) 14—26.

¹⁴ Ἰδοὺ ἡμέραι ἔρχονται, φησὶ Κύριος, καὶ ἀναστήσω τὸν λόγον μου τὸν ἀγαθὸν ὃν ἐλάλησα ἐπὶ τὸν οἶκον Ἰσραὴλ καὶ ἐπὶ τὸν οἶκον Ἰούδα. ¹⁵ ἐν ταῖς ἡμέραις ἐκείναις καὶ ἐν τῷ καιρῷ ἐκείνῳ ἀνατελῶ τῷ Δαυὶδ ἀνατολὴν δικαίαν, ποιῶν κρίμα καί δικαιοσύνην ἐν τῇ γῇ. ¹⁶ ἐν ταῖς ἡμέραις ἐκείναις σωθήσεται ἡ Ἰουδαία καὶ Ἰερουσαλὴμ κατασκηνώσει πεποιθυῖα· καὶ τοῦτο τὸ ὄνομα ὃ καλέσει αὐτὴν ¹⁷ ὅτι τάδε λέγει Κύριος, Οὐκ ἐξολοθρευθήσεται τῷ Δαυὶδ ἀνὴρ καθήμενος ἐπὶ θρόνον οἴκου Ἰσραήλ· ¹⁸ καὶ τοῖς ἱερεῦσι τοῖς Λευίταις οὐκ ἐξολοθρευθήσεται ἀνὴρ ἐκ προσώπου μου, ἀναφέρων ὁλοκαυτώματα καὶ θύων θυσίαν. ¹⁹ καὶ ἐγένετο λόγος Κυρίου πρὸς Ἰερεμίαν λέγων ²⁰ Τάδε λέγει Κύριος Εἰ διασκεδάσετε τὴν διαθήκην μου τὴν ἡμέραν καὶ τὴν διαθήκην μου τὴν νύκτα, τοῦ μὴ εἶναι ἡμέραν καὶ νύκτα ἐν καιρῷ αὐτῶν· ²¹ καίγε ἡ διαθήκη μου διασκεδασθήσεται μετὰ Δαυὶδ τοῦ δούλου μου, τοῦ μὴ 45εἶναι αὐτῷ υἱὸν βασιλεύοντα ἐπὶ τὸν θρόνον αὐτοῦ, καὶ ἠ πρὸς τοὺς Λευίτας τοὺς ἰερεῖς τοὺς λειτουργοῦντάς μοι. ²² ὡς οὐκ ἐξαριθμηθήσεται ἡ δύναμις τοῦ οὐρανοῦ, οὐδὲ ἐκμετρηθήσεται ἡ ἄμμος τῆς θαλάσσης, οὕτως πληθυνῶ τὸ σπέρμα Δαυὶδ τοῦ δούλου μου καὶ τοὺς Λευίτας τοὺς λειτουργοῦντάς μοι. ²³ καὶ ἐγένετο λόγος Κυρίου πρὸς Ἰερεμίαν λέγων ²⁴ Ἆρά γε οὐκ ἴδες τί ὁ λαὸς ἐλάλησαν λέγοντες Αἱ δύο πατριαὶ ἃς ἐξελέξατο Κύριος ἐν αὐταῖς, καὶ ἰδοὺ ἀπώσατο αὐτούς·; καὶ τὸν λαόν μου παρώξυναν τοῦ μὴ εἶναι ἔτι ἔθνος ἐνώπιόν μου. ²⁵ τάδε λέγει Κύριος Εἰ μὴ τὴν διαθήκην μου ἡμέρας καὶ νυκτός, ἀκριβάσματα οὐρανοῦ καὶ γῆς, οὐκ ἔταξα, ²⁶ καίγε τὸ σπέρμα Ἰακὼβ καὶ Δαυὶδ τοῦ δούλου μου ἀποδοκιμῶ, τοῦ μὴ λαβεῖν ἐκ τοῦ σπέρματος αὐτοῦ ἄρχοντα πρὸς τὸ σπέρμα Ἀβραὰμ καὶ Ἰσαὰκ καὶ Ἰακώβ· ὅτι ἐπιστρέψω τὴν ἐπιστροφὴν αὐτῶν, καὶ οἰκτειρήσω αὐτούς¹¹³¹¹³Another considerable fragment of Theodotion may be found in Jer. xlvi. (xxxix.) 4—13, see O. T. in Greek, p. 534 f..

Unfortunately there is no other Greek version which can be compared with Theodotion in this passage, for the LXX. is wanting, and only a few shreds of Aquila and Symmachus have reached us. But the student will probably agree with Field that the style is on the whole not wanting in simple dignity, and that it is scarcely to be distinguished from the best manner of the LXX.¹¹⁴¹¹⁴Hexapla, prolegg. p. xxxix. "Theodotionis stylus simplex et gravis est." LXX. of Jer. xxiii. 5, 6 may be set beside Θ of xl. 14, 15. With his Hebrew Bible open at the place, he will observe that the rendering is faithful to the original, while it escapes the crudities and absurdities which beset the excessive fidelity of Aquila. Now and again we meet with a word unknown to the LXX. (e.g. ἀκριβάσματα = חֻקּוֹת‎)¹¹⁵¹¹⁵Cod. A employs ἀκριβασμός in this sense (Jud. v. 15, 3 Regn. xi. 34, 4 Regn. xvii. 15), but under the influence of Theodotion, at least in the last two passages; see Field ad loc., or a reminiscence of Aquila; on the other hand Theodotion agrees with the LXX. against Aquila in translating בְּרִית by διαθήκη. If in one place 46Theodotion is more obscure than Aquila τὴν διαθήκην τὴν ἡμέραν . . . τὴν νύκτα, Aq. τῆς ἡμέρας . . . τῆς νυκτός), yet the passage as a whole is a singularly clear and unaffected rendering. His chief defect does not reveal itself in this context; it is a habit of transliterating Hebrew words which could have presented no difficulty to a person moderately acquainted with both languages. Field gives a list of 90 words which are treated by Theodotion in this way without any apparent cause¹¹⁶¹¹⁶0p. cit. p. xl. sq.. When among these we find such a word as אֵל (which is represented by ἤλ in Mal. ii. 11), we are compelled to absolve him from the charge of incompetence, for, as has been pertinently asked, how could a man who was unacquainted with so ordinary a word or with its Greek equivalent have produced a version at all? Probably an explanation should be sought in the cautious and conservative temperament of this translator¹¹⁷¹¹⁷D. C. B. art. Hexapla (iii. p. 22). Cf. ib. iv. p. 978.. Field's judgement is here sounder than Montfaucon's; Theodotion is not to be pronounced indoctior, or indiligentior, but only "scrupulosior quam operis sui instituto fortasse conveniret¹¹⁸¹¹⁸Thus in Mal. l. c. he was perhaps unwilling to use θεός in connexion with the phrase אִל גֵכָר.."

9. The relation of the two extant Greek versions of Daniel is a perplexing problem which calls for further consideration. In his lost Stromata Origen, it appears¹¹⁹¹¹⁹Jerome on Dan. iv.: "Origenes in nono Stromatum volumine asserit se quae sequuntur ab hoc loco in propheta Daniele non iuxta LXX. interpretes . . . sed iuxta Theodotionis editionem disserere.", announced his intention of using Theodotion's version of Daniel; and an examination of Origen's extant works shews that his citations of Daniel "agree almost verbatim with the text of Theodotion now current¹²⁰¹²⁰Dr Gwynn in D. C. B. (iv. p. 974).." The action of Origen in this matter was generally endorsed by the Church, as we learn from Jerome (praef. in Dan.: "Danielem prophetam iuxta LXX. interpretes ecclesiae 47non legunt, utentes Theodotionis editione"; cf. c. Rufin. ii. 33). Jerome did not know how this happened, but his own words supply a sufficient explanation: "hoc unum affirmare possum quod multum a veritate discordet et recto iudicio repudiata sit." So universal was the rejection of the LXX. version of Daniel that, though Origen loyally gave it a place in his Hexapla, only one Greek copy has survived¹²¹¹²¹The Chigi MS. known as Cod. 87 (H. P. 88); see O. T. in Greek, iii. pp. vi., xii., and cf. the subscription printed ib. p. 574., Theodotion's version having been substituted in all other extant Greek MSS. of Daniel.

But the use of Theodotion's Daniel in preference to the version which was attributed to the LXX. did not begin with Origen. Clement of Alexandria (as edited) uses Theodotion, with a sprinkling of LXX. readings, in the few places where he quotes Daniel (paed. ii. 8, iii. 3, strom. i. 4, 21). In North Africa both versions seem to have influenced the Latin text of Daniel. The subject has been carefully investigated by Prof. F. C. Burkitt¹²²¹²²Old Latin and Itala, p. 18 ff., who shews that Tertullian used "a form of the LXX. differing slightly from Origen's edition," whilst Cyprian quotes from a mixed text, in which Theodotion sometimes predominates. Irenaeus, notwithstanding his reverence for the LXX. and distrust of the later versions, cites Daniel after Theodotion's version¹²³¹²³An exception in i. 19. 2 (Dan. xii. 9 f.) is due to a Marcosian source.. Further, Theodotion's Daniel appears to be used by writers anterior to the date usually assigned to this translator. Thus Hermas (vis. iv. 2, 4) has a clear reference to Theodotion's rendering of Dan. vi. 22¹²⁴¹²⁴See Salmon, Intr. to the N. T.7 p. 639.. Justin (dial. 31) gives a long extract from Dan. vii. in which characteristic readings from the two versions occur in almost equal proportions¹²⁵¹²⁵On the trustworthiness of Justin's text here see Burkitt, op. cit. p. 25 n. (against Hatch, Essays, p. 190).. Clement of Rome (1 Cor. 34) cites a part of the same context, 48with a Theodotionic reading (ἐλειτούργουν, LXX. ἐθερά͜ευον). Barnabas (ep. iv. 5) also refers to Dan. vii., and, though his citation is too loose to be pressed, the words ἐξαναστήσονται ὄπισθεν αὐτῶν are more likely to be a reminiscence of ὀπίσω αὐτῶν ἀναστήσεται (Th.) than of μετὰ τούτους στήσεται (LXX.). The Greek version of Baruch (i. 15—18, ii. 11—19) undoubtedly supports Theodotion against the LXX. Still more remarkable is the appearance of Theodotionic renderings in the New Testament. A writer so faithful to the LXX. as the author of the Epistle to the Hebrews, in his only reference to Daniel Heb. xi. 33 = Dan. vi. 23) agrees with Theodotion against the Chigi version¹²⁶¹²⁶Heb. l. c. ἔφραξαν στόματα λεόντων (Dan. Th., ἐνέφραξεν τὰ στόματα τῶν λεόντων; LXX., σέσωκέ με ἀπὸ τῶν λεόντων). . The Apocalypse, which makes frequent use of Daniel, supports Theodotion on the whole; cf. Apoc. ix. 20 (Dan. v. 23), x. 6 (Dan. xii. 7), xii. 7 (Dan. x. 20), xiii. 7 (Dan. vii. 21), xix. 6 (Dan. x. 6), xx. 4 (Dan. vii. 9), xx. 11 (Dan. ii. 35)¹²⁷¹²⁷The references are from Dr Salmon's Intr. p. 548 f. He adds: "I actually find in the Apocalypse no clear evidence that St John had ever seen the so-called LXX. version." See Bludau in Th. Q. 1897 (p. 1 ff.).. Even in the Synoptic Gospels Theodotion's rendering in Dan. vii. 13 (μετὰ τῶν νεφελῶν) occurs as well as the LXX. ἐπὶ τῶν ν. comp. Mc. xiv. 62 with Mt. xxiv. 30, xxvi. 64¹²⁸¹²⁸The N. T. occasionally inclines to Theodotion in citations which are not from Daniel; cf. Jo. xix. 37 (Zech. xii. 10), 1 Cor. xv. 54 (Is. xxv. 8); see Schürer³, iii. p. 324 "entweder Th. selbst ist älter als die Apostel, oder es hat einen 'Th.' vor Th. gegeben.".

From these premisses the inference has been drawn that there were two pre-Christian versions of Daniel, both passing as 'LXX.', one of which is preserved in the Chigi MS., whilst the other formed the basis of Theodotion's revision¹²⁹¹²⁹D. C. B. art. Theodotion iv. p. 970 ff. Dr Salmon (Intr. p. 547) is disposed to accept this view.. It has been urged by Dr Gwynn with much acuteness that the two Septuagintal Books of Esdras offer an analogy to the two versions of Daniel, and the appearance of the phrase ἀπηρείσατο αὐτὰ ἐν τῷ εἰδωλίῳ αὐτοῦ in 1 Esdr. ii. 9 and Dan. i. 2 (LXX.) 49has been regarded as an indication that the Greek Esdras and the Chigi Daniel were the work of the same translator¹³⁰¹³⁰D. C. B. iv. p. 977 n.; cf. Hastings' D. B., i. p. 761.. An obvious objection to the hypothesis of two Septuagintal or Alexandrian versions is the entire disappearance of the version which was used ex hypothesi not only by the authors of the Epistle to the Hebrews and the Apocalypse, but by Theodotion and other writers of the second century. But Theodotion's revision of Daniel may have differed so little from the stricter Alexandrian version as to have taken its place without remark¹³¹¹³¹On the whole question of the date of Theodotion, see Schürer, G. J. V.³ iii. 323 f., where the literature of the subject is given..

10. Symmachus. Of this translator Irenaeus says nothing, and it has been inferred, perhaps too hastily, that he was unknown to the Bishop of Lyons, and of later date. Origen knew and used Symmachus, and had received a copy of his commentary on St Matthew from a wealthy Christian woman named Juliana, to whom it had been given by the author. According to Eusebius, Symmachus was an Ebionite, and this is confirmed by Jerome; a less probable tradition in Epiphanius represents him as a Samaritan who had become a convert to Judaism¹³²¹³²The name סומכום occurs in the Talmud as that of a disciple of R. Meir, who flourished towards the end of the second or beginning of the third century. Geiger desires to identify our translator with this Symmachus; see Field, prolegg. ad Hex. p. xxix..

Eus. H. E. vi. 17 τῶν γε μὴν ἑρμηνευυῶν αὐτῶν δὴ τούτων ἰστέον Ἐβιωναῖον τὸν Σύμμαχον γεγονέναι . . . καὶ ὑπομνήματα δὲ τοῦ Συμμάχου εἰσέτι νῦν φέρεται ἐν οἷς δοκεῖ πρὸς τὸ κατὰ Ματθαῖον ἀποτεινόμενος εὐαγγέλιον τὴν δεδηλωμένην αἵρεσιν κρατύνειν. ταῦτα δὲ ὁ Ὠριγένης μετὰ καὶ ἄλλων εἰς τὰς γραφὰς ἑρμηνειῶν τοῦ Συμμάχου σημαίνει παρὰ Ἰουλιανῆς τινος εἰληφέναι, ἣν καὶ φησι παῤ αὐτοῦ Συμμάχου τὰς βίβλους διαδέξασθαι. Hieron. de virr. ill. 54 "Theodotionis Hebionaei et Symmachi eiusdem dogmatis" (cf. in Hab. iii. 13); praef. in Job: "Symmachus et Theodotion Iudaizantes haeretici." Epiph. de mens. et pond. 15 ἐν τοῖς τοῦ Σευήρου χρόνοις Σύμμαχός τις Σαμαρείτης τῶν παῤ αὐτοῖς σοφῶν μὴ τιμηθεὶς ὑπὸ τοῦ οἰκείου ἔθνους . . . προσηλυτεύει καὶ περιτέμνεται δευτέραν περιτομήν . . . οὗτος τοίνυν ὁ Σύμμαχος πρὸς διαστροφὴν τῶν 50παρὰ Σαμαρείταις ἑρμηνειῶν ἑρμηνεύσας τὴν τρίτην ἐξέδωκεν ἑρμηνείαν.

That Symmachus, even if of Jewish or Samaritan birth, became an Ebionite leader is scarcely doubtful, since an Ebionitic commentary on St Matthew bearing his name was still extant in the fourth century¹³³¹³³Euseb. l. c.; the Symmachians, an Ebionite sect probably named after him, are mentioned by Ambrosiaster (comm. in Gal., prolegg.) and Augustine (c. Faust. xix. 4, c. Crescon. i. 36)¹³⁴¹³⁴Philastrius, who represents the Symmachiani as holding other views, says (c. 145): "sunt haeretici alii qui Theodotionis et Symmachi itidem interpretationem diverso modo expositam sequuntur." See Harnack, Gesch. d. altchr. Litt., 1. i. p. 212.. His floruit is open to some question. Dr Gwynn has shewn¹³⁵¹³⁵D. C. B. iv. p. 971 ff. Σευήρου in de pond. et mens. 16 is on this hypothesis a corruption of Οὐήρου. Cf. Lagarde's Symmicta, ii. p. 168. that Epiphanius, who makes Theodotion follow Symmachus, probably placed Symmachus in the reign of Verus, i.e. Marcus Aurelius. Now in the Historia Lausiaca, c. 147, Palladius says that Juliana sheltered Origen during a persecution, i.e. probably during the persecution of the Emperor Maximius (A.D. 238—241). If this was so, the literary activity of Symmachus must have belonged, at the earliest, to the last years of M. Aurelius, and it may be questioned whether Epiphanius has not inverted the order of the two translators, i.e. whether Theodotion ought not to be placed under M. Aurelius and Symmachus under Commodus (A.D. 180—192)¹³⁶¹³⁶The Gospel of Peter, which cannot he much later than A.D. 170, and may be fifteen or twenty years earlier, shews some verbal coincidences with Symmachus (Akhmîm fragment, pp. xxxiv. 18, 20), but they are not decisive.. The version of Symmachus was in the hands of Origen when he wrote his earliest commentaries, i.e. about A.D. 228¹³⁷¹³⁷Cf. D. C. B. iv. p. 103.; but the interval is long enough to admit of its having reached Alexandria.

11. The aim of Symmachus, as Jerome perceived, was to express the sense of his Hebrew text rather than to attempt 51a verbal rendering: "non solet verborum κακοζηλίαν sed intellegentiae ordinem sequi" (in Am. iii. 11). While Aquila endeavoured "verbum de verbo exprimere," Symmachus made it his business "sensum potius sequi" (praef. in Chron. Eus., cf. praef. in Job). Epiphanius, who believed Symmachus to have been a Samaritan proselyte to Judaism, jumped to the conclusion that his purpose was polemical (πρὸς διαστροφὴν τῶν παρὰ Σαμαρείταις ἑρμηνειῶν ἑρμηνεύσας). But if Symmachus had any antagonist in view, it was probably the literalism and violation of the Greek idiom which made the work of Aquila unacceptable to non-Jewish readers. So far as we can judge from the fragments of his version which survive in Hexaplaric MSS., he wrote with Aquila's version before him, and in his efforts to recast it made free use of both the LXX. and Theodotion. The following extracts will serve to illustrate this view of his relation to his predecessors.

MALACHI II. 13¹³⁸¹³⁸The Hexaplaric renderings are from Cod. 86 (Cod. Barberinus): Field, Hexapla, ii. p. 1033.

But it must not be supposed that Symmachus is a mere reviser of earlier versions, or that he follows the lead of Aquila as Theodotion follows the LXX. Again and again he goes his own way in absolute independence of earlier versions, and sometimes at least, it must be confessed, of the original. This is due partly to his desire to produce a good Greek rendering, more or less after the current literary style; partly, as it seems, to dogmatic reasons. The following may serve as specimens of the Greek style of Symmachus when he breaks loose from the influence of his predecessors: Gen. xviii. 25 ὁ πάντα ἄνθρωπον ἀπαιτῶν δικαιοπραγεῖν, ἀλρίτῶς μὴ ποιήσῃς τοῦτο; Job xxvi. 14 τί δὲ ψιθύρισμα τῶν λόγων αὐτοῦ ἀκούσομεν, ὅπου βροντὴν δυναστείας αὐτοῦ οὐδεὶς ἐννοήσει; Ps. xliii. 16 δἰ ὅλης ἡμέρας ἡ ἀσχημόνησίς μου ἄντικρύς μου, καὶ ὁ καταισχυμμὸς τοῦ προσώπου μου καλύπτει με. Ps. lxviii. 3 ἐβαπτίσθην εἰς ἀπεράντους καταδύσεις, καὶ οὐκ ἔστιν στάσις· εἰσῆλθον εἰς τὰ βάθη τῶν ὑδάτων, καὶ ῥεῖθρον ἐπέκλυσέν με. Eccl. iv. 9 εἰσὶν ἀμείνους δύο ἑνός· ἔχουσιν γὰρ κέρδος ἀγαθόν. Isa. xxix. 4 ὐπὸ γῆν ἐδαφισθήσεται ἡ λαλιά σου, καὶ ἔσται ὡς ἐγγαστρίμυθος ἡ φωνή που καὶ ἀπὸ τῆς γῆς ἡ λαλιά σου ῥοίσεται.

It cannot be said that these renderings approach to excellence, but a comparison with the corresponding LXX. will shew that Symmachus has at least attempted to set himself free from the trammels of the Hebrew idiom and to clothe the thoughts of the Old Testament in the richer drapery of the Greek tongue. It is his custom to use compounds to represent ideas which in Hebrew can be expressed only by two or more words (e.g. בְּלי־פֶשַׁע, Symm. ἀναιτίως, עַיִן ∙בְּעַיִן, Symm. ὁφθαλμοφανῶς, לְראׁשׁ פִּנָּה Symm. ἀκρογωνιαῖος); he converts into a participle the first of two finite verbs connected by a copula (Exod. v. 7 Symm. ἀπερχόμενοι καλαμάσθωσαν, 4 Regn. i. 2 σφαλέντες ἔπεσον); he has at his command a large supply of Greek particles (e.g. he renders אַךְ by ἄρα, ὄντως, ἴσως, δἰ ὅλου, μόνον, οὕτως, ἀλλ᾿ 53ὅμως)¹³⁹¹³⁹For other examples see Field, prolegg. p. xxx. ff.; D. C. B. iv. p. 19 f.. More interesting and important is the tendency which Symmachus manifests to soften the anthropomorphic expressions of the Old Testament; e.g. Gen. i. 27, ἔκτισεν ὁ θεὸς τὸν ἄνθρωπον ἐν εἰκόνι διαφὸρῳ¹⁴⁰¹⁴⁰Reading, perhaps, בעלם ובעלם אלהים; cf. Nestle, Marginalien, pp. 3, 15. · ὄρθιον ὁ θεὸς ἔκτισεν αὐτόν. Exod. xxiv. 10, εἶδον ὁράματι τὸν θεὸν Ἰσραήλ. Jud. ix. 13 τὸν οἶνον . . . τὴν εὐφροσύνην τῶν ἀνθρώπων. Ps. xliii. 24 ἵνα τί ὡς ὑπνῶν εἶ, Δέσποτα; In these and other instances Symmachus seems to shew a knowledge of current Jewish exegesis¹⁴¹¹⁴¹See D. C. B. iii. p. 20. which agrees with the story of his Jewish origin or training.

Literature. On Aquila the student may consult R. Anger de Onkelo Chaldaico, 1845; art. in D. C. B. (W. J. Dickson); M. Friedmann, Onkelos u. Akylas, 1896; Lagarde, Clementina, p. 12 ff.; Krauss, Akylas der Proselyt (Festschrift), 1896; F. C. Burkitt, Fragments of Aquila, 1897; C. Taylor, Sayings of the Jewish Fathers², 1897 (p. viii.); Schürer³, iii. p. 317 ff. On Symmachus, C. H. Thieme, pro puritate Symmachi dissert., 1755; art. in D. C. B. (J. Gwynn); Giov. Mercati, l’età di Simmaco interprete, 1892. On Theodotion, Credner, Beiträge, ii. p. 253 ff.; art. in D. C. B. (J. Gwynn); G. Salmon, Intr. to the N. T.⁷, p. 538 ff.; Schürer³, iii. p. 323 ff. Works which deal with the ancient non-Septuagintal versions in general will be mentioned in c. iii., under Literature of the Hexapla.

12. Other ancient Greek versions. The researches of Origen (A.D. 185—253) brought to light three anonymous versions besides those of Aquila, Theodotion and Symmachus; from their relative position in the columns of his great collection (see c. iii.) they are known as the Quinta (εʹ), Sexta (ςʹ), and Septima (ζʹ) respectively. The following are the chief authorities:

Eus. H. E. vi. 16 τοσαύτη δὲ εἰσήγετο τῷ Ὠριγένει τῶν θείων λόγων ἀπηκριβωμένη ἐξέτασις ὡς . . . καί τινας ἑτέρας παρὰ τὰς· καθημαξευμένας ἑρμηνείας ἐναλλαττούσας . . ., ἐφευρεῖν, ἃς οὐκ οἶδ᾿ ὅθεν ἔκ τινων μυχῶν τὸν πάλαι λανθανούσας χρόνον εἰς φῶς ἀνιχνεύσας προήγαγεν . . . τίνος ἄῤ εἶεν οὐκ εἰδὼς αὐτὸ τοῦτο μόνον ἐπεσημήνατο 54ὡς ἄρα τὴν μὲν εὕροι ἐν τῇ πρὸς Ἀκτίῳ Νικοπόλει . . .ἐπὶ μιᾶς αὖθις σεσημείωται ὡς ἐν Ἰεριχοῖ εὑρημένης ἐν πίθῳ κατὰ τοὺς χρόνους Ἀντωνίνου τοῦ υἰοῦ Σεβήρου. Epiph. de mens. et pond. 18 μετὰ τὸν διωγμὸν τοῦ βασιλέως Σευήρου ηὑρέθη ἡ πέμπτη ἐν πίθοις ἐν Ἰεριχῷ κεκρυμμένη ἐν χρόνοις τοῦ υἱοῦ Σευήρου τοῦ ἐπικληθέντος Καρακάλλου τε καὶ Γέτα . . . ἐν δὲ τῷ ἑβδόμῳ αὐτοῦ ἔτει ηὑρέθησαν καὶ βίβλοι τῆς πέμπτης ἐκδόσεως ἐν πίθοις ἐν Ἰεριχῷ κεκρυμμένης μετὰ ἄλλων βιβλίων Ἐβραικῶν καὶ Ἑλληνικῶν. τὸν δὲ Καράκαλλον διαδέχεται Ἀντωνῖνος ἕτερος . . . μετὰ τοῦτον ἐβασίλευσεν Ἀλέξανδρος . . . ἔτη ιγʹ· ἐν μέσῳ τῶν χρόνων τούτων ηὑρέθη ἕκτη ἔκδοσις, καὶ αὐτὴ ἐν πίθοις κεκρυμμένη, ἐν Νικοπόλει τῇ πρὸς Ἀκτίῳ. Pseudo-Ath. syn. scr. sacr. 77 πέμπτη ἑρμηνεία ἐστὶν ἡ ἐν πίθοις εὑρεθεῖσα κεκρυμμένη ἐπὶ Ἀντωνίνου βασιλέως τοῦ Καρακάλλα ἐν Ἰεριχῷ παρά τινος τῶν ἐν Ἰεροσολύμοις σπουδαίων. ἕκτη ἐρμηνεία ἐστὶν ἡ ἐν πίθοις εὑρεθεῖσα, καὶ αὕτη κεκρυμμένη, ἐπὶ Ἀλεξάνδρου τοῦ Μαμαίας παιδὸς ἐν Νικοπόλει τῇ πρὸς Ἄκτιον ὑπὸ Ὠριγένους γνωρίμων. Hieron. de virr. ill. 54 "quintam et sextam et septimam editionem, quas etiam nos de eius bibliotheca habemus, miro labore repperit et cum ceteris editionibus conparavit": in ep. ad Tit. "nonnulli vero libri, et maxime hi qui apud Hebraeos versu compositi sunt, tres alias editiones additas habent quam 'quintam' et 'sextam' et 'septimam' translationem vocant, auctoritatem sine nominibus interpretum consecutas." Cf. in Hab. ii. 11, iii. 13.

It appears from the statement of Eusebius¹⁴²¹⁴²Jerome (prol. in Orig. exp. Cant.) confirms Eusebius, on whose words see Dr Mercati, Studi a Testi 5, v. p. 47 (1901). that Origen found the Quinta at Nicopolis near Actium, and that either the Sexta or the Septima was discovered in the reign of Caracalla (A.D. 211—217) at Jericho; while Epiphanius, reversing this order, says that the Quinta was found at Jericho c. A.D. 217, and the Sexta at Nicopolis under Severus Alexander (A.D. 222—235)¹⁴³¹⁴³The Dialogue of Timothy and Aquila identifies Nicopolis with Emmaus Nicopolis in Palestine.. According to Epiphanius both the Quinta and the Sexta, according to Eusebius the Sexta only, lay buried in a πίθος (dolium), one of the earthenware jars, pitched internally, and partly sunk in the ground, in which the mustum was usually stored while it underwent the process of fermentation¹⁴⁴¹⁴⁴D. of Gk and Lat. Ant. p. 1202. These πίθοι are said to have been sometimes used instead of cistae or capsae for preserving books. In 1906 five Greek documents were found in an earthenware jar at Elephantine; see Dr F. G. Kenyon in Egypt Exploration Fund Archaeological Report for 1907—8, p. 50.. Since 55Origen was in Palestine A.D. 217, and in Greece A.D. 231, it is natural to connect his discoveries with those years. How long the versions had been buried cannot be determined, for it is impossible to attach any importance to the vague statements of Eusebius (τὸν πάλαι λανθανούσας χρόνον). The version found at or near Nicopolis may have been a relic of the early Christianity of Epirus, to which there is an indirect allusion in the Pastoral Epistles¹⁴⁵¹⁴⁵Lightfoot, Biblical Essays, p. 432.. The Jericho find, on the other hand, was very possibly a Palestinian work, deposited in the wine jar for the sake of safety during the persecution of Septimius Severus, who was in Palestine A.D. 202, and issued edicts against both the Synagogue and the Church¹⁴⁶¹⁴⁶Cf. Eus. H. E. vi. 7; Spartian. in Sev. 17.. Of Septima nothing is known, beyond what Eusebius tells us, and the very sparing use of it in the Psalter of some Hexaplaric MSS.; the few instances are so dubious that Field was disposed to conclude either that this version never existed, or that all traces of it have been lost¹⁴⁷¹⁴⁷Prolegg. ad Hexapla, p. xlvi; see however R. Sinker, Psalm of Habakkuk (Camb. 1890), p. 42. Ps.-Athanasius calls Lucian the seventh version: ἐβδόμη πάλιν καὶ τελευταία ἑρμηνεία ἡ τοῦ ἁγίου Λουκιανοῦ..

There is no conclusive evidence to shew that any of these versions covered the whole of the Old Testament¹⁴⁸¹⁴⁸According to Harnack-Preuschen (i. p. 340) the opposite is implied by Eusebius' use of ἐναλλαττούσας in reference to these versions: "d. h. die eine war nur für diese, die andere nur für jene Bücher vorhanden.". Renderings from Quinta¹⁴⁹¹⁴⁹On Quinta see Mercati, Studi e Testi 5, IV. p. 28; and Burkitt in Proc. Soc. Bibl. Archaeology, June 1902. are more or less abundant in 2 Kings, Job, Psalms, Canticles, and the Minor Prophets, and a few traces have been observed in the Pentateuch. Sexta is well represented in the Psalms and in Canticles, and has left indications of its existence in Exodus, 1 Kings, and the Minor Prophets.

With regard to the literary character of Quinta and Sexta, the style of Quinta is characterised by Field as "omnium elegantissimus . . . cum optimis Graecis suae aetatis scriptoribus comparandus." Sexta also shews some command of Greek, 56but is said to be disposed to paraphrase; Field, while he regards that charge as on the whole 'not proven,' cites a remarkable example of the tendency from Ps. xxxvi. 35, which ςʹ renders, Εἶδον ἀσεβῆ καὶ ἀναιδῆ ἀντιποιούμενον ἐν σκληρότητι καὶ λέγοντα Εἰμὶ ὡς αὐτόχθων περιπατῶν ἐν δικαιοσύνῃ. Jerome¹⁵⁰¹⁵⁰adv. Rufin. attributes both versions to 'Jewish translators,' but the Christian origin of Sexta betrays itself¹⁵¹¹⁵¹"Prodens manifestissime sacramentum," as Jerome himself remarks. No doubt the primary reference is to Joshua (Field), but the purport of the gloss is unmistakable. at Hab. iii. 13 ἐξῆλθες τοῦ σῶσαι τὸν λαόν σου διὰ Ἰησοῦν τὸν χριστόν σου¹⁵²¹⁵²leg. fors. Ἰησοῦ τοῦ χριστοῦ σου..

The Greek fathers of the fourth and fifth centuries quotes non-Septuagintal renderings from an interpreter who is styled ὁ Ἐβραῖος. Ὁ Σύρος is also cited, frequently as agreeing with ὁ Ἐβραῖος. Nothing is known of these translators (if such they were), but an elaborate discussion of all the facts may be seen in Field¹⁵³¹⁵³Prolegg. pp. lxxv.—lxxxii. See also Lagarde, Ueber den Hebräer Ephraims von Edessa. On τὸ Σαμαρειτικόν see Field, p. lxxii. ff., and Nestle, Urtext, p. 206. For some ambiguous references to other(?) versions see Philostr. haer. cc. 143, 144..

13. The 'GRAECUS VENETUS.' This is a version of the Pentateuch, together with the books of Ruth, Proverbs, Canticles, Ecclesiastes, Lamentations, and Daniel, preserved in St Mark's Library at Venice in a single MS. of cent. xiv.—xv. (cod. Gr. vii.)¹⁵⁴¹⁵⁴See Eichhorn, p. 421 ff.; De Wette-Schrader, p. 122 f.. It was first given to the world by de Villoison (Strassburg, 1784) and C. F. Ammon (Erlangen, 1790—1); a new edition with valuable prolegomena by O. von Gebhardt appeared at Leipzig in 1875¹⁵⁵¹⁵⁵Graecus Venetus Pentateuchi &c. versio Graeca. Ex unico biblioth. S. Marci Venetae codice nunc primum uno volumine comprehensam atque apparatu critico et philologico instructam edidit O. G. Praefatus est Fr. Delitzsch.. This translation has been made directly from the M. T., but the author appears to have occasionally availed himself of earlier Greek versions (LXX., 57Aq., Symm., Theod.)¹⁵⁶¹⁵⁶Gebhardt, p. lvii. ff.. His chief guide however appears to have been David Kimchi, whose interpretations are closely followed¹⁵⁷¹⁵⁷Ib. p. lxii.. That he was a Jew is clear from incidental renderings (e.g. in Exod. xxiii. 20 he translates הַמָּקוֹם τὸν ὀντωτήν¹⁵⁸¹⁵⁸Ὀντωτής, ὁντουργός, οὐσιωτής are his usual renderings of יהוה., sc. יְהוָֹה). From the fact of his having undertaken a Greek version Gebhardt infers that he was a proselyte to Christianity, but the argument may be used to support an opposite conclusion; as a Jew he may have been moved by a desire to place before the dominant Orthodox Church a better rendering of the Old Testament than the LXX. Delitzsch wishes to identify him with Elissaeus, a Jewish scholar at the court of Murad I., who flourished in the second half of the 14th century.

The style of this remarkable version will be best illustrated by a few specimens:

Gen. vi. 2 f.

² τεθέανται γοῦν οἱ υἱεῖς τοῦ θεοῦ τὰς θυγατέρας τοῦ ἀνθρὡπου ὅτι καλαὶ ἐτέλουν, καὶ ἔλαρον ἑαυτοῖς γυναῖκας ἀπὸ πασῶν ὦν εἵλοντο. ³ ἔφη τοίνυν ὁ ὀντωτής Οὐ κρινεῖ πνεῦμα τοὐμὸν ἐν τῷ ἀνθρώπῳ ἐς αἰῶνα, ἐφ᾿ οἷς ἔτι πέρ ἐστι σάρξ· τελέσουσι δ᾿ αἱ ἡμέραι αὐτοῦ ἑκατὸν καὶ εἴκοσιν ἔτη.

Prov. viii. 22 ff.

²² ὁ ὀντωτὴς ἐκτήσατό με ἀρχὴν ὁδοῦ οἱ, πρὸ τῶν ἔργων αὐτοῦ ἐκ τότε. ²³ ἀπ᾿ αἰῶνος κέχυμαι, ἀπὸ κρατὸς, ἀπὸ προλήμματος γῆς. ²⁴ ἐν οὐκ ἀβύσσοις πέπλασμαι, ἐν οὐ πηγαῖς δεδοξασμένων ὑδάτων· ²⁵ πρὶν ὄρη ἐμπαγῆναι, πρὸ τῶν βουνῶν ὡδίνημαι· ²⁶ ἄχρις οὐκ ἐποίησε γῆν, διόδους καὶ κεφαλὴν κόνεων τῆς οἰκουμένης.

Daniel vii. 13.

¹³ ὁράων ἐκύρησα ἐν ὁράσεσιν εὐφρόνας, αὐτίκα τε ξὺν ταῖς 58νεφηλαις τῶν πόλων ὡς υἱεὺς ἀνθρώπω ἀφικνούμενος ἔην, μέχρι τε τῶ παλαιῶ ταῖς ἁμέραις ἔφθασε κἀνώπιον τήνω προσήγαγόν ἑ. ¹⁴ τήνῳ τ᾿ ἐδόθη ἀρχὰ τιμά τε καὶ βασιλεία, πάντες τε λαοὶ ἔθνεα καὶ γλῶτται τήνῳ λατρευσείοντι· ἁ ἀρχύ εὑ ἀρχὰ αἰῶνος ὃς οὐ παρελευσείεται, ἅ τε βασιλεία εὗ ἅπερ οὐκ οἰχησείεται.

The student will not fail to notice the translator's desire to render his text faithfully, and, on the other hand, his curiously infelicitous attempt to reproduce it in Attic Greek; and lastly his use of the Doric dialect in Daniel to distinguish the Aramaic passages from the rest of the book. The result reminds us of a schoolboy's exercise, and the reader turns from it with pleasure to the less ambitious diction of the LXX., which, with its many imperfections, is at least the natural outgrowth of historical surroundings.

Klostermann (Analecta p. 30) mentions a MS. Psalter (Vat. Gr. 343), bearing the date 22 April, 1450, which professes to be a translation into the Greek of the fifteenth century (κατὰ τὴν νῦν κοινὴν τῶν Γραικῶν φωνήν). A version of the Pentateuch into modern Greek in Hebrew characters was printed at Constantinople in 1547, forming the left-hand column of a Polyglott (Hebrew, Chaldee, Spanish, Greek). It is described in Wolf, Bibliotheca Hebraea, ii. p. 355, and more fully in La version Neo-grecque du Pentateuche Polyglotte . . . remarques du Dr Lasare Belléli (Paris, 1897). This Greek version has recently been transliterated and published in a separate form with an introduction and glossary by D. C. Hesseling (Leide, 1897). A Greek version of job (1576) is mentioned by Neubauer in J. Q. R. iv. p. 18 f.