CHAPTER IV.

ANCIENT VERSIONS BASED UPON THE SEPTUAGINT.

THE Christian Churches of Greek-speaking countries throughout the Empire read the Old Testament in the Alexandrian Version. Few of the provinces were wholly non-Hellenic; Greek was spoken not only in Egypt and Cyrenaica, in Western Syria, Asia Minor, Macedonia, and Achaia, but to a great extent in the West, in Italy and at Rome. Roman satirists of the first century complained that the capital had become a Greek city; the upper classes acquired Greek; the freedmen and slaves in many cases spoke it as their mother tongue²³²²³²The evidence is collected by Caspari Quellen zur Gesch. d. Taufsymbols, iii. 267 f., and summarised by Sanday and Headlam, Romans, p. lii. ff.. Official letters addressed to the Roman Church or proceeding from her during the first two centuries were written in Greek; only three or at the most four of the Bishops of Rome during the same period bear Latin names²³³²³³The evidence is collected by Caspari Quellen zur Gesch. d. Taufsymbols, iii. 267 f., and summarised by Sanday and Headlam, Romans, p. lii. ff.. In Gaul the Greek tongue had spread up the valley of the Rhone from Marseilles to Vienne and Lyons; the Viennese confessors of A.D. 177 used it in their correspondence both with the Roman Bishops and with their brethren in Asia Minor; the Bishop of Lyons wrote in the same language his great work against the false gnosis of the age. The Old Testament as known to Clement of Rome and Irenaeus of Lyons is substantially the Greek version of 88the Seventy. To the Church of North Africa, on the other hand, the Greek Bible was a sealed book; for Carthage, colonised from Rome before the capital had been flooded by Greek residents, retained the Latin tongue as the language of common life. It was at Carthage, probably, that the earliest daughter-version of the Septuagint, the Old Latin Bible, first saw the light²³⁴²³⁴On the other hand reasons have been produced for suspecting that the Latin version had its origin at Antioch; see Guardian, May 25, 1892, p. 786 ff., and Dr H. A. A. Kennedy in Hastings' D. B. iii p. 54 ff. [This chapter was already in type when Dr Kennedy's article came into my hands. I regret that for this reason I have been unable to make full use of his exhaustive treatment of the Latin versions.]; certainly it is there that the oldest form of the Old Latin Bible first meets us in the writings of Cyprian. Other versions followed as the result of missionary enterprise; and to this latter source we owe the translations of the Old Testament which were made between the second century and the ninth into Egyptian, Ethiopic, Arabic, Gothic, Armenian, Georgian, and Slavonic. All these versions rest either wholly or in part upon the Septuagint, and therefore possess a special interest for the student of the Greek Bible. One other group has a claim upon his consideration. The earliest of the Syriac versions of the Old Testament is on the whole a translation from the Hebrew, but it shews the influence of the Septuagint in certain books. The rest, which belong to post-Nicene times, are based directly upon the Alexandrian Greek, and one of them forms the most important of extant witnesses to the text of the Hexaplaric recension.

1. LATIN VERSIONS FROM THE SEPTUAGINT.

(1) The Latin Bible before Jerome.

With the exception of Jerome himself, our earliest authority upon the origin of the Old Latin Bible is Augustine of Hippo, and it may be well to begin by collecting his statements upon the subject.

Aug. de civ. Dei xviii. 43 ex hac LXX. interpretatione etiam in Latinam linguam interpretatum est quod ecclesiae Latinae tenent. De doctr. Christ. ii. 16 [after a reference to the "Latinorum interpretum infinita varietas"] "qui enim scripturas ex Hebraea lingua in Graecam verterunt, numerari possunt, Latini interpretes nullo modo; ut enim cuique primis fidei temporibus in manus venit codex Graecus et aliquantulum facultatis sibi utriusque linguae habere videbatur ausus est interpretari." Ib. 22: "in ipsis autem interpretationibus Itala ceteris praeferatur." Ep. ii. 82 (ad Hieronymum): "ideo autem desidero interpretationem tuam de LXX. ut . . . tanta Latinorum interpretum qui qualescunque hoc ausi sunt quantum possumus imperitia careamus."

This is African testimony, but it belongs to the end of the fourth century, and needs to be verified before it can be unhesitatingly received. Many of the discrepancies to which Augustine refers may be due to the carelessness or officiousness of correctors or transcribers; if, as Jerome tells us, there were towards the end of the fourth century as many types of text as there were MSS. of the Latin Bible ("tot exemplaria quot codices"), it is clearly out of the question to ascribe each of these to a separate translator. A few specimens, taken from Cyprian and extant MSS. of the O. L., will enable the student to form some idea of the extent to which these differences are found in extant texts²³⁵²³⁵To facilitate comparison obvious errors of the MSS. and orthographical peculiarities have been removed..

Genesis xlviii. 17 f.

Exod. xxxii. 21— 24.

Leviticus iv. 27— 29.

Micah v. 2.

Isaiah xxix. 11, 18.

It is clearly unsafe to generalise from a few specimens, but the student will not fail to observe that the variations in these extracts may, perhaps without exception, be attributed either to the ordinary accidents of transcription or to the recensions of the original text. In the case of the New Testament Dr Hort²⁴²²⁴²Introduction, p. 78 ff. Cf. Westcott, Canon, p. 252 ff.; Wordsworth, O. L. Biblical Texts, i., p. xxx. ff. held that there was "some justification for the alternative view that Italy had an indigenous version of her own, not less original than the African," and where both types of text existed, he distinguished them by the designations 'African Latin' and 'European Latin,' applying the term 'Italian'²⁴³²⁴³On Augustine's use of this term see F. C. Burkitt, O. L. and Ita1a, p. 55 ff. to later revisions of the European text. The classification of the Old Latin authorities for the O. T. is less advanced, and owing to the fragmentary character of most of 92the MSS. it is more difficult; but we may assume that it will proceed on the same general lines, and that the pre-Hieronymian types of text in the Old Testament as in the New will be found to be mainly two, i.e. the African, and the European, with a possible sub-division of the latter class²⁴⁴²⁴⁴Cf. Berger, Histoire de la Vulgate, p. 6; Kennedy, in Hastings' D. B. p. 58 ff.. In pursuing this enquiry use must be made not only of the surviving fragments of O. L. MSS., but of the numerous quotations of the Latin versions which occur in writings anterior to the final triumph of the Vulgate. As Dr Hort has pointed out²⁴⁵²⁴⁵Introduction, p. 83., certain of the Latin fathers "constitute a not less important province of Old Latin evidence than the extant MSS., not only furnishing landmarks for the investigation of the history of the version, but preserving numerous verses and passages in texts belonging to various ages and in various stages of modification." These patristic materials were collected with great care and fulness by Sabatier (Bibliorum sacrorum Latinae versiones antiquae . . . opera et studio D. Petri Sabatier O. S. B., Reims, 1743, '49, Paris, 1751; vols. i. ii. contain the O. T.) ; but after the lapse of a century and a half his quotations can no longer be accepted without being compared with more recent editions of the Latin fathers²⁴⁶²⁴⁶For this purpose the Vienna Corpus Scriptorum Ecclesiasticorum Latinorum is the best collection available; but it is still far from complete., and they often need to be supplemented from sources which were not at his command²⁴⁷²⁴⁷A revised Sabatier is promised by the Munich Academy (Archiv, viii. 2, p. 311 ff.)..

These researches are important to the student of the Septuagint in so far as they throw light on the condition of the Greek text in the second and third centuries after Christ. The Latin translation of the Old Testament which is largely quoted by Cyprian was probably made in the second century, and certainly represents the text of MSS. earlier than 93the time of Origen. What Mr Burkitt has pointed out²⁴⁸²⁴⁸Rules of Tyconius, p. cxvi. f. in reference to the prophetic books is doubtless true in general; "no . . . passage [to which the asterisk is prefixed in Hexaplaric MSS. is found in any form of the African Latin." Thus, as he remarks, "the Old Latin brings us the best independent proof we have that the Hexaplar signs introduced by Origen can be relied on for the reconstruction of the LXX." Again, M. Berger²⁴⁹²⁴⁹Histoire de la Vulgate, p. 6. Cf. Driver, Samuel, p. lxxvii. f. has called attention to the prominence of Lucianic readings in certain Old Latin texts; and the fact that a Lucianic element is widely distributed in Old Latin MSS. and quotations has also been recognised by Vercellone²⁵⁰²⁵⁰Variae lectiones, ii., p. 426. and Ceriani²⁵¹²⁵¹Monumenta sacra et profana, 1. i., p. xvi.; Le recensioni dei LXX e la versione latina detta Itala (Rendiconti, Feb. 18, 1886). See also Driver, Notes on Samuel, p. lxxviii. f.; Kennedy, in Hastings' D.B., l.c.; Nestle, Einführung 2 , pp. 148 note, 280 [E. Tr., p. 182 f.]; Wordsworth-White. p.654.. This element is found even in the African text²⁵²²⁵²Burkitt, Rules of Tyconius, p. cxvii., and its occurrence there suggests that the Antiochian recension, though it was made at the beginning of the fourth century, has preserved ancient readings which existed also in the African copies of the LXX., though they found no place in our oldest codices.

We proceed to give a list of the extant remains of the Old Latin Version of the LXX., and the editions in which they are accessible.

OLD LATIN FRAGMENTS OF THE OLD TESTAMENT.

i. PENTATEUCH.

Cod. Lugdunensis, vi. (Ulysse Robert, Pentateuchi e Codice Lugdunensi versio Latina antiquissima, Paris, 1881; Librorurn Levitici et Numerorum versio antiqua Itala e cod. perantiquo in bibliotheca Ashburnhamiensi conservato, London, 1868; Delisle, Découverte d’une très ancienne version latine de deux livres de la Bible in the Journal des Savants, Nov. 1895, p. 702 ff.; U. Robert, Heptateuchi partis post. versio Lat. antiquissima e cod. Lugd., Lyons, 1900²⁵³²⁵³Cf. N. McLean in J. Th. St. ii. 305 ff..

ii. HISTORICAL BOOKS.

Joshua, Judges i. 1— x. 31.

Cod. Lugdunensis (in the portion published by Robert in 1900).

Ruth.

Cod. Complutensis, ix., Madrid, Univ. Libr. (S. Berger in Notices et Extraits, xxxiv. 2, p. 119 ff.).

1— 4 Regn.

Fragments of Corbie and St Germain MSS. (Sabatier); fragments from a Verona MS. and a Vatican MS. in Bianchini (Vindiciae, p. cccxli. ff.), from a Vienna MS. in Haupt's vet. antehieron. vers. fragmenta Vindobonensia, 1877, from an Einsiedeln MS. in Notices at Extraits xxxiv. 2, p. 127 ff., and from leaves found at Magdeburg and Quedlinburg²⁵⁸²⁵⁸See V. Schultze, die Quedlinburger Itala-Miniaturen der k. Bibliothek in Berlin (Munich, 1898). printed by W. Schum, 1876, Weissbrodt, 1887, and A. Düning, 1888. Fragments of 2 Regn. at Vienna published by J. Haupt, 1877. A Vienna palimpsest containing considerable fragments of 1— 2 Regn. (J. Belsheim, Palimpsestus Vind., 1885). Readings from the margin of Cod. Goth. Legionensis²⁵⁹²⁵⁹On these see Bergen Hist. de la Vulgate, p. 18 f., and the caution in O. L. and Itala, p. 9 f. printed by C. Vercellone, ii. p. 179 ff.; cf. Archiv, viii. 2. (The Verona and Vatican fragments should perhaps be classed as Vulgate.)

1 Esdras.

An O. L. text is to be found in the Paris MS. Bibl. Nat. lat. 111, the Madrid MS. E. R. 8, and another in a Lucca MS. ap. Lagarde, Septuagintastudien, 1892.

Judith, Tobit.

Cod. Complutensis.

Cod. Goth. Legionensis.

Cod. Vatic. regin. (Bianchini, Vindiciae, p. cccl. f.; Tobit only).

O. L. texts are also to be found in the Paris MSS. Bibl. Nat. lat. 6, 93, 161 (Tobit), 11505, 11549 (Judith), 11553, in the Munich MS. 6239, the Milan MS. Amb. E 26 infr. (Tobit), and the Oxford MS. Bodl. auct. E. infr. 2 (Judith). See Notices et Extraits xxxiv. 2, p. 142 ff. Of these texts some were printed by Sabatier, and Munich 6239 is in Belsheim's Libr. Tobiae, &c. (1893).

Esther.

Cod. Pechianus (Sabatier).

Cod. Vallicellanus (Bianchini, Vindiciae, p. ccxciv. ff.).

Cod. Complutensis (see above under Ruth).

An O. L. text of Esther is found also in the Paris MS. Bibl. Nat. lat. 11549 (= Corb. 7), the Lyons MS. 356, the Munich MSS. 6225, 6239, the Monte Casino MS. 35 (Biblioth. Casin. i., 1873), the Milan MS. Amb. E. 26 infr. (see S. Berger op. cit.).

1, 2 Maccabees.

O. L. texts are to be found in the Paris MS. Bibl. Nat. lat. 11553 (Sabatier) and the Milan MS. Amb. E. 26 inf. (A. Peyron, Cic. fragmm. i. 70 ff: (1824).

(See Berger, op. cit.)

iii. POETICAL BOOKS.

Psalms.

Cod. Veronensis (in Bianchini).

Cod. Sangermanensis (in Sabatier).

A Reichenau palimpsest described by Mone, l. u. gr. Messen, p. 40.

Fragments of the ῷδαί edited by F. F. Fleck (Leipzig, 1837), and L. F. Hamann (Jena, 1874).

Job.

Fragment. Floriacense (Sabatier). Containing c. xl. 3— 9.

Readings from the margin of Cod. Goth. Legionensis (Notices et Extraits, p. 111 ff.).

Proverbs, Ecclesiastes, Canticles.

Readings in a St Gallen MS., see Notices et Extraits, p. 137 ff. Fragments published by Sabatier, Vogel, Mone, Berger (Hastings' D. B. iii. p. 50).

Wisdom, Sirach.

See Lagarde, Mittheilungen i. (Göttingen, 1884). C. Donais, Une ancienne Version latine de l’Ecclésiastique (Paris, 1895).

iv. PROPHETS.

Fragmenta Wirceburgensia, vi. (?) (E. Ranke, Par palimp. Wirceb. p. 49 sqq.).

Containing Hos. i. 1— ii. 13, iv. 13— vii. 1; Jon. iii. 10— iv. 11; Isa. xxix. 1— xxx. 6, xlv. 20— xlvi. 11; Jer. xii. 12— xiii. 12, xiv. 15— xvii. 1O, xviii. 16— xxiii. 39, xxxv. 15— 19, xxxvi. 2— xxxvii. 11, xxxviii. 23— xl. 5, xli. 1— 17; Lam. ii. 16— iii. 40; Ezek. xxiv. 4— 21, xxvi. 10— xxvii. 4, xxxiv. 16— xxxv. 5, xxxvii. 19— 28, xxxviii. 8— 20, xl. 3— xlii. 18, xlv. 1— xlvi. 9, xlviii. 28— 35; Dan. i. 2— ii. 9, iii. 15— (26), viii. 5— ix. 10, x. 3— xi. 4, 20— 42, and Bel.

Fragmenta Fuldensia, v. (E. Ranke, Fragm. versionis ante-Hieronymianae, Marburg, 1868).

Containing Hos. vii. 6— ix. 1, Amos viii. 1— ix. 1, ix., 5— 9, Mic. ii. 3— iii. 3.

Fragmenta Weingartensia, v. (E. Ranke, Fragm. v. ante-H., Vienna, 1868; P. Corssen, Zwei neue Fragmente d. Weingartener Prophetenhandschrift, Berlin, 1899).

Containing Hos. iv. 13 f., v. 5, 7, vii. 16, viii. 1— 6, 13 f., ix. 1— 17, xii. 3, 7, 9, 12, xiii. 1, 3— xiv. 2; Amos v. 24— vi. 8; Mic. i. 5— iii. 3, iv. 3— vii. 20; Joel i. 1— 14, ii. 3— 5, iv. 2— 4, 15— 17; Jon. i. 14— iv. 8; Ezek. xvi. 52— xvii. 6, 19— xviii. 9, xxiv. 25— xxv. 14, xxvi. 10— xxvii. 7, 17— 19, xxviii. 1— 17, xxxiii. 7— 11, xlii. 5, 6, 14, xliii. 22— xliv. 5, 19— xlv. 2, xlvi. 9— 23, xlvii. 2— 15, xlviii. 22— 30; Dan. ii. 18— 33, ix. 25— x. 11, xi. 18— 23.

Fragmenta Stutgardiana (E. Ranke, Antiquissima V. T. versionis Latinae fragmenta, Marburg, 1888).

Containing Amos vii. 13— viii. 10; Ezek. xviii. 9— 17, xx. 18— 21, xxvii. 7— 17, xxxiii. 26— 30, xxxiv. 6— 12; Dan. xi. 35— 39.

Fragmenta monast. S. Pauli Carinthiaci (A. Vogel, Beiträge zur Herstellung der A. L. Bibelübersetzung, Vienna, 1868).

Containing Ezek. xlii. 5, 6, 14, xliv. 19— xlv. 2, xlvi. 9— 23, xlvii. 2— 15.

Fragmenta palimpsesta Vaticana (F. Gustafsson, Fragmenta V. T. in Latinum conversi a palimpsesto Vaticano eruta, Helsingfors, 1881)²⁶⁰²⁶⁰These fragments, as I am informed by Dr W. O. E. Oesterley, contain an almost purely Vulgate text, and should perhaps disappear from this list..

Containing Hosea iv. 6, 7; Joel ii. 5— 7; Amos v. 16— 18, vii. 2— 7, ix. 5— 8; Jon. iii. 7— iv. 2; Hab. i. 16— ii. 3; Zeph. iii. 13— 20; Zech. vii. 11— 14, viii. 16— 21.

Fragmenta palimpsesta Sangallensia (F. C. Burkitt, O. L. and Itala, Camb. 1896).

Containing Jer. xvii. 1— 17, xxix. 13— 19.

Codex Vallicellanus B. vii. (Bianchini, Vindiciae, p. ccxiii.).

Containing Baruch.

O. L. texts of Baruch are also to be found in the Paris MSS. Bibl. Nat. lat. 11, 161, 11951, and Arsenal 65, 70; and in the Monte Casino MS. 35, and the Reims MS. 1.

Copious extracts from most of the books of the O. L. Bible are given in the anonymous Liber de divinis scripturis sive Speculum, wrongly attributed to St Augustine (ed. F. Weihrich in the Vienna Corpus, vol. xii.). Two other patristic collections of O. L. excerpts may also be mentioned here—the Testimonia of St Cyprian (ed. Hartel, Corpus, vol. iii. 1), and the liber regularum Tyconii (ed. F. C. Burkitt, in Texts and Studies, iii. 1). See also the Collatio Carthaginiensis printed in Dupin's Optatus (Paris, 1700), p. 379 ff.

(2) Latin versions of the LXX. revised or taken over by Jerome.

The great Pannonian scholar, Eusebius Hieronymus (A.D. 329— 420), began his "useful labours²⁶¹²⁶¹Aug. ep. 82 (ad Hieronymum): "hi qui me invidere putant utilibus laboribus tuis."" upon the Old Testament at Rome about the year 383, probably (as in the case of his revision of the Gospels) at the suggestion of the Roman Bishop Damasus († 384). His first attempt was limited to a revision of the Latin Psalter and conducted on lines which afterwards seemed to him inadequate. A few years later—but before 390— 1, when he began to translate from the Hebrew—a fresh revision of the Psalter from the LXX. was undertaken at the desire of Paula and Eustochium; its immediate purpose was to remove errors which had already found their way into the copies of the earlier work, but the opportunity was seized of remodelling the Latin Psalter after the example of the Hexapla.

Praef. in libr. Psalmorum: "psalterium Romae dudum positum emendaram et iuxta LXX. interpretes, licet cursim, magna illud ex parte correxeram. quod quia rursum videtis, o Paula et Eustochium, scriptorum vitio depravatum, plusque antiquum errorem quam novam emendationem valere, cogitis ut . . . renascentes spinas eradicem. . . . . notet sibi unusquisque vel iacentem lineam vel signa radiantia, id est vel obelos () vel asteriscos (); et ubicunque viderit virgulam praecedentem (), ab ea usque ad duo puncta (:) quae impressimus, sciat in LXX. translatoribus plus haberi; ubi autem stellae () similitudinem perspexerit, de Hebraeis voluminibus additum noverit aeque usque ad duo puncta, iuxta Theodotionis dumtaxat editionem qui simplicitate sermonis a LXX. interpretibus non discordat."

These two revised Latin Psalters were afterwards known as Psalterium Romanum and Psalterium Gallicanum respectively. Both recensions established themselves in the use of the Latin Church²⁶²²⁶²Cf. adv. Rufin. ii. 30 "psalterium . . . certe emendatissimum iuxta LXX. interpretes nostro labore dudum Roma suscepit"; where, as Westcott says (Smith's D. B. iii. 1698 n.), he seems to include both revisions., the former in the cursus psallendi, the latter in the bibliotheca or Church Bible. At length Pius V. († 1572) 99ordered the Gallican Psalter to be sung in the daily offices, an exception being made in favour of St Peter's at Rome, St Mark's at Venice, and the churches of the Archdiocese of Milan, which retained the 'Roman' Psalter²⁶³²⁶³Martène, de ant. rit. i. p. i8 f.. In MSS. of the Vulgate a triple Psalter not infrequently appears, shewing Jerome's two Septuagintal revisions side by side with the Psalterium Hebraicum, his later translation from the Hebrew; but the 'Hebrew' Psalter never succeeded in displacing the Hieronymian revisions of the Old Latin, and the Latin Church still sings and reads a version of the Psalms which is based on the Septuagint. The liturgical Psalter of the Anglican Church "followeth . . . the Translation of the Great English Bible, set forth and used in the time of King Henry the Eighth, and Edward the Sixth"; i.e. it is based on Coverdale's version, which was "translated out of Douche and Latyn into Englishe"; and many of its peculiarities may be traced to the LXX. through the Gallican Psalter incorporated in the Vulgate²⁶⁴²⁶⁴Cf. Bp Westcott, History of the English Bible, pp. 206 ff., 351 ff.; Kirkpatrick, Psalms, Intr. p. lxxiii f..

The following specimen (Ps. lxvii=lxviii. 12-14, 18-22) will enable the reader to form an idea of the relation between Jerome's two revisions of the Old Latin and his 'Hebrew'²⁶⁵²⁶⁵Editions published in 1874 by Baer and Tischendorf (Lib. Psalm. Heb. atque Lat.) and by Lagarde (Psalt. iuxta Hebraeos). Psalter.

ROMAN.	GALLICAN.	HEBREW.
12Dominus dabit verbum evangelizantibus virtute multa; 13rex virtutum dilecti, et speciei domus dividere spolia. 14si dormiatis in medios cleros, pennae columbae deargentatae, et posteriora dorsi eius in specie auri. [diapsalma]......18currus Dei decem milium multiplex, milia laetantium. Dominus	12Dominus dabit verbum evangelizantibus virtute multa; 13rex virtutum dilecti: et speciei domus dividere spolia. 14si dormiatis inter medios cleros pennae columbae deargentatae et posteriora dorsi eius in pallore auri, diapsalma.........18currus Dei decem milibus multiplex, milia laetantium:	12Domine, dabis sermonem adnuntiatricibus fortitudinis plurimae, 13reges exercituum foederabuntur, foederabuntur et pulcritudo domus dividet spolia. 14si dormieritis inter medios terminos, pennae columbae deargentatae et posteriora eius in virore auri.......18currus Dei innumerabiles, milia

in illis in Sina in sancto. 19ascendensin altum captivam duxit captivitatem, dedit dons hominibus. etenim non credunt inhabitare. 20Dominus Deus benedictus; benedictus Dominus de die in diem. , prosperum iter faclet nobis Deus salutaris poster. diapsalma. 21Deus noster deus salvos faciendi, et Domini exitus mortis. 22verumtamen Deus conquassabit capita inimicoruril suorum, verticem capilli perambulantium in delictis suis.

Dominus in eis in: Sina in sancto. 19ascendisti in altum: cepisti captivitatem, accepisti dona in hominibus. etenim non credentes inhabitare Dominum Deum. 20benedictus Dominus die quotidie; prosperum iter faciet nobis Deus salutarium nostrorum. disapsalma. 21Deus noster, Deus salvos faciendi: et Domini Domini: exitus mortis. 22verumtamen Deus confringet capita inimicorum suorum, verticem capilli perambulantium in delictis suis.

abundantium; Dominus in eis in Sina, in sancto. 19ascendisti in excelsum, captivam duxisti captivitatem, accepisti dona in hominibus; insuper et non credentes habitare Dominum Deum. 20benedictus Dominus per singulos dies; portabit nos Deus salutis nostrae. semper. 21Deus noster deus salutis, et Domini Dei mortis egressus. 22verumtamen Deus confringet capita inimicorum suorum, verticem crinis ambulantis in delictis suis.

The book of Job offered a still more promising field for the labours of the Hexaplarising reviser, for the Greek text as known to Origen fell greatly short of the current Hebrew, and it was this defective text which formed the basis of the Latin versions used by Cyprian and Lucifer and in the Speculum²⁶⁶²⁶⁶Burkitt, O. L. and Itala, pp. 8, 32 f.. Jerome, who had access to the Hexapla at Caesarea, took advantage of Origen's revision, in which the lacunae of the Greek job were filled up from Theodotion, and sent his friends, Paula and Eustochium, a Latin version of Job at once corrected and supplemented from the Hexaplaric LXX. The result gave him for the time profound satisfaction; he had lifted up job from the dunghill²⁶⁷²⁶⁷Praef. in libr. Job: "qui adhuc apud Latinos iacebat in stercore et vermibus scatebat errorum.", and restored him to his pristine state²⁶⁸²⁶⁸ibid. "integrum immaculatumque gaudete."; 101the difference between the Old Latin version and the new seemed to him to be nothing short of that which separate falsehood from truth²⁶⁹²⁶⁹Ad Pammach.: "veterem editionem nostrae translationi compara, et liquido providebitis quantum distet inter veritatem et mendacium." Jerome's satisfaction with his original revision of Job was continued even after he had produced a new version from the Hebrew; in the preface to the latter he leaves the student free to choose between the two ("eligat unusquisque quod vult"). . The asterisks shewed that from 700 to 800 lines had been restored to this long mutilated book²⁷⁰²⁷⁰Praef. in Job ed. Heb. See below, pt II., c. ii..

A few brief specimens from Lagarde's text²⁷¹²⁷¹In Mittheilungen, ii. will suffice to shew the character of the work.

x. 4. aut sicut homo perspicit, perspicis? aut sicut videt homo, videbis? aut humana est vita tua? aut anni tui sunt tanquam dies hominis?

xix. 17 et rogabam uxorem meam invocabam blandiens filios uteri mei ; at illi in perpetuum despexerunt me; cum surrexero, locuntur ad me.

xlii. 7 et defunctus est job senex plenus dierum. scriptum est autem resurrecturum cum his quos Dominus suscitabit.

Jerome also revised from the Hexaplaric Septuagint, for the benefit of Paula and Eustochium, the 'books of Solomon' (Proverbs, Ecclesiastes, Canticles), treating the Greek text after the manner of Origen; but his work has perished, the preface alone surviving. A like fate has overtaken a translation of Chronicles, undertaken at the desire of Domnio and Rogatianus. This version of Chronicles appears from the preface to have been influenced by Jerome's Hebrew studies, which were now sufficiently matured to enable him to form an independent judgement in reference to the merits of his Greek text, though he still clung to his old belief in the inspiration of the original Septuagint.

Praef. in libros Salomonis: "tres libros Salomonis, id est, Proverbia, Ecclesiasten, Canticum canticorum, veteri LXX. auctoritati reddidi, vel antepositis lineis () superflua quaeque 102designans, vel stellis () titulo (?) praenotatis ea quae minus habebantur interserens . . . et ubi praepostero ordine atque perverso sententiarum fuerat lumen ereptum suis locis restituens feci intellegi quod latebat" Praef. in libr. Paralipomenon: "cum a me nuper litteris flagitassetis ut vobis librum Paralipomenon Latino sermone transferrem, de Tiberiade legis quondam doctorem qui apud Hebraeos admirationi habebatur assumpsi . . . et sic confirmatus ausus sum facere quod iubebatis. libere enim vobis loquor, ita et in Graecis et Latinis codicibus hic nominum liber vitiosus est ut non tam Hebraea quam barbara quaedam . . . arbitrandum sit. nec hoc LXX. interpretibus qui Spiritu sancto pleni ea quae vera fuerant transtulerunt, sed scriptorum culpae adscribendum. . . . ubicunque ergo asteriscos . . . videritis ibi sciatis de Hebraeo additum . . . ubi vero obelus, transversa scilicet virga, praeposita est, illic signatur quid LXx. interpretes addiderint."

Whether Jerome dealt with the rest of the canonical books of the Old Latin in the same manner must remain an open question. No trace remains either of such revised versions or of prefaces which once belonged to them, nor does he refer to them in the prefaces of his translations from the Hebrew. On the other hand his letters occasionally speak of his revision of the Old Latin in terms which seem to imply that it was complete, and in one of them there is a passage which suggests that the disappearance of the other books was due to the dishonesty of some person whose name is not given.

Adv. Rufin. ii. 24: "egone contra LXX. interpretes aliquid sum locutus quos ante annos plurimos diligentissime emendatos meae linguae studiosis dedi?" Ep. 71 (ad Lucinium): "LXX. editionem et te habere non dubito." Ep. 106 (ad Sunn. et Fret.): "editionem LXX. interpretum quae et in ἑξαπλοῖς codicibus reperitur et a nobis in Latinum sermonem fideliter versa est." Cf. Ep. Augustini ad Hieron. (116), (c. 405): "mittas obsecro interpretationem tuam de LXX. quam te edidisse nesciebam." At a later time (c. 416) Jerome excuses himself from doing as Augustine had desired, since "pleraque prioris laboris fraude cuiusdam amisimus" (Ep. 134).

In any case Jerome's Hexaplarised version had little or no influence on the text of the Latin Bible, except in the Psalter. Even his translations from the Hebrew did not easily supersede the Old Latin. The familiar version died hard and, 103as the list of MSS. will have shewn, parts of it were copied as late as the seventh century. Even at Rome the old version long held its ground by the side of the new; in the last years of the sixth century, Gregory the Great, while basing his great commentary on Job upon the Vulgate, claimed a right to cite the Old Latin when it served his purpose, "quia sedes apostolica utrique nititur²⁷²²⁷² Praef. ad Moralia in Job. ."

The coexistence of the two versions naturally produced mixture in the MSS.²⁷³²⁷³Cf. e.g. Berger, op. cit. p. xi.: "les textes des anciennes versions et de la nouvelle sont constamment mêlés et enchevêtrés dans les manuscrits.", which was not altogether removed by the revisions of the sixth and ninth centuries. Moreover, the Old Latin version continued to hold its place in those books of the Church Bible which had no Semitic original, or of which the Semitic original was no longer current. In the preface to the Salomonic Books Jerome says explicitly: "porro in eo libro qui a plerisque Sapientia Salomonis inscribitur et in Ecclesiastico . . . calamo temperavi, tantummodo canonicas scripturas vobis emendare desiderans." The books of Tobit and Judith²⁷⁴²⁷⁴On the relation of Jerome's Latin Judith to the Septuagint see C. J. Ball in Speaker's Commentary, Apocrypha, p. 257 ff. were afterwards translated by him from the Aramaic (praeff. in librum Tobiae, in librum Judith), and these versions have been incorporated in the Vulgate, but the Vulgate Wisdom, Ecclesiasticus, Baruch, 1, 2 Maccabees are supplied from ante-Hieronymian sources. Thus to this day a considerable part of the Latin Bible is in greater or less degree an echo of the Septuagint.

LITERATURE. Besides the editions already mentioned the student may consult with advantage Eichhorn, Einleitung, i. 321; N. Wiseman, Essays, i. (London, 1853)—a reprint of his Two letters on some parts of the controversy concerning 1 Joh. v. 7; B. F. Westcott, art. Vulgate in Smith's D. B. iii.; H. Rönsch, Itala u. Vulgata (Marburg, 1869); F. Kaulen, Handbuch zur Vulgata (Mainz, 1870); Ziegler, Die lat. Bibelübersetzungen vor 104Hieronymus (Munich, 1879); Lagarde, Probe einer neuen Ausgabe der lat. Übersetzungen des A. T. (1870); A. Ceriani, Le recensioni dei LXX e la versione latina detta Itala, 1886; L. Salembier, Une page inédite de l’histoire de la Vulgate, Amiens, 1890; Bleek-Wellhausen (1893), p. 553 ff.; Scrivener-Miller, ii, p. 191 ff.; Gregory, p. 949 ff.; F. C Burkitt, The Old Latin and the Itala, in Texts and Studies (Cambridge, 1896); E. Nestle, Urtext, pp. 84 ff. [especially valuable for the bibliography of the Latin versions]; H. A. A. Kennedy, The Old Latin Versions, in Hastings' D. B. iii. pp. 47— 62; Corssen in Jahresb. f. d. class. Altertumswissensch (1899); Latin Versions of the O. T., art. in Ch. Q. R. (Apr. 1901); W. O. Oesterley in J. Th. Stud. v. vi. (text of Min. Proph.).

2. THE EGYPTIAN VERSIONS.

The tradition of St Mark's episcopate at Alexandria²⁷⁵²⁷⁵See Gospel acc. to St Mark, p. xiv. f. The Clementine Homilies (i. 8 ff.) attribute the foundation of the Alexandrian Church to Barnabas. But a yet earlier beginning is possible. In Acts xviii. 24 cod. D reads Ἀλεξανδρεὺς . . . ὃς ἦν κατηχημένος ἐν τῇ πατρίδι τὸν λόγον τοῦ κυρίου, on which Blass (Acta app. p. 201) remarks: "itaque iam tum (id quod sine testimonio suspicandum erat) in Aegyptum quoque nova religio permanaverat." may be taken as evidence, so far as it goes, of the early planting of the Church in that city. The first converts were doubtless, as at Rome, Greek-speaking Jews, descendants of the old Jewish settlers²⁷⁶²⁷⁶Acts ii. 9 f. οἱ κατοικοῦντες . . . Αἴγυπτον. Ib. vi. 9 τινὲς ἐκ τῆς συναγωγῆς τῆς λεγομένης . . . Ἀλεξανδρέων. Cf. Report of the Egypt Exploration Fund, 1899— 1900, p. 54., and their Greek proselytes; and the first extension of the movement was probably amongst the Greek population of the towns on the sea-coast of the Mediterranean. As it spread to the interior, to the villages of the Delta, to Memphis, Oxyrhynchus, Panopolis, and eventually to Thebes, it encountered native Egyptians who spoke dialects of the Egyptian tongue²⁷⁷²⁷⁷Cf. what is said of St Anthony in the Vita Antonii (Migne, P. G. xxvi. 944 sq.).. How soon they were evangelised there is no direct evidence to shew, but the process may have begun shortly after the Gospel reached Alexandria. The native Church retained its own tongue, and in the fourth and fifth centuries Greek was still unknown to many of the monks and ecclesiastics of Egypt. Christianity however is probably responsible for either introducing or spreading the use of a new system of 105writing with characters which are chiefly of Greek origin²⁷⁸²⁷⁸Of the 31 letters of the Coptic alphabet 7 only () are not from the Greek. On the pre-Christian systems see Clem. strom. v. 4 οἱ παῤ Αἰγυπτίοις παιδευόμενοι πρῶτον μὲν πάντων . . . ἐκμανθάνουσι τὴν ἐπιστολογραφικὴν καλουμένην (the Demotic), δευτέραν δὲ τὴν ἱερατικὴν . . . ὑστάτην δὲ καὶ τελευταίαν τὴν ἱερογλυφικήν. . This writing, known as Coptic—a corruption of Αἰγύπτιος—is found with some variations in all MS. fragments of the Egyptian versions of the Old and New Testaments.

The analogy of the Old Latin would lead us to suppose (as Bp Lightfoot remarks²⁷⁹²⁷⁹Scrivener-Miller, ii. p. 97.) that no long interval passed between the acceptance of Christianity by any large number of native Egyptians, and the first attempts to translate the Scriptures into the Egyptian tongue. "We should probably not be exaggerating if we placed one or both of the principal Egyptian versions, the Bohairic and the Sahidic, or at least parts of them, before the close of the second century." The Bishop is writing with only the New Testament in view, but his argument applies equally to the Old. His view is on the whole supported by Dr Hort²⁸⁰²⁸⁰Intr. to N. T. in Greek, p. 85., Ciasca²⁸¹²⁸¹Sacr. bibl. fragmenta Copto-Sahidica, i. p. viii. , and Mr A. C. Headlam²⁸²²⁸²Scrivener-Miller, ii. p. 105 f.: but Mr Forbes Robinson, following Guidi, produces reasons for regarding it as 'not proven,' and prefers to say that "historical evidence . . . on the whole, points to the third century as the period when the first Coptic translation was made." "But this view," he adds, "can only be regarded as tentative. In the light of future discoveries it may have to be modified²⁸³²⁸³Hastings' D. B. i. p. 672. Cf. F. E. Brightman in J. Th. St. i. 254.."

The plurality of the Egyptian versions is well ascertained. Perhaps the geographical form of Egypt gave special opportunities for the growth of popular dialects; certain it is that increased knowledge of the language has added to the dialectic complications with which the Coptic scholar has to struggle²⁸⁴²⁸⁴The Demotic, as it is known to us, appears to present no dialectic variation, perhaps because the specimens which have reached us were all the work of the single class—the scribes: see Hyvernat, Étude sur les versions Coptes in Revue Biblique, v. 3, p. 429; A. C. Headlam in Scrivener-Miller, p. 105.. 106It was in these popular dialects that the translations of the Bible were made. "Christianity . . . was in Egypt a great popular movement . . . the Scriptures were translated, not into the literary language, but into that of the people; and the copies of these translations in each locality reflected the local peculiarities of speech." Fragments of Biblical versions have been found in the Bohairic²⁸⁵²⁸⁵Formerly known as the Memphitic, a name which might be more appropriately applied to the form of Middle Egyptian current at Memphis. 'Bohairic' is derived from el-Bohairah, a district S. of Alexandria. 'Sahidic,' also called Thebaic, is from es-sa‘îd = Upper Egypt. On some characteristics of the several dialects see Hyvernat, p. 431., Sahidic, and Middle Egyptian dialects. The Bohairic dialect was spoken in Lower, the Sahidic in Upper, Egypt, and the Middle Egyptian in the intermediate province of Memphis. Some authorities speak of two other dialects, the Fayumic and Akhmimic, assigning to them certain Biblical fragments which are regarded by others as belonging to the Middle Egyptian.

Translations of books of the Old Testament into these Egyptian dialects were naturally made from the Alexandrian Greek version, and, if we may judge from the extensive use of the Old Testament in early Christian teaching, there is no reason to doubt that they were translated at as early a date as the Gospels and Epistles, if not indeed before them. Portions of the Old Testament exist in each of the Egyptian dialects. Hyvernat mentions fragments of Isaiah, Lamentations and Ep. of Jeremiah in Fayumic and Middle Egyptian, and of Exodus, Sirach, 2 Macc., and each of the Minor Prophets in Akhmimic²⁸⁶²⁸⁶Cf. Steindorff, Die Apokalypse des Elias, p. 2.; in Bohairic he enumerates 6 MSS. of the Pentateuch, 14 of the Psalms, 5 of Proverbs, 3 of Job, 4 of the Minor Prophets, 5 of Isaiah, 3 of Jeremiah, 4 of Daniel, and 107one MS. of Ezekiel; in Sahidic, though few complete MSS. of any Biblical book have survived, there is a large number of extant fragments representing most of the canonical books and certain of the non-canonical (the two Wisdoms, the Ep. of Jeremiah, and the Greek additions to Daniel).

The following list gives the more important publications which contain portions of the Old Testament in the Egyptian versions.

BOHAIRIC. D. Wilkins, Quinque libri Moysis, 1731; Fallet, La version Cophte du pentateuque, 1854; Lagarde, Der Pentateuch koptisch, 1867; Bruchstücke der kopt. Übersetzungen des A. T. in Orientalia i. 1879. The Psalter has been edited by R. Tuki, 1744, J. L. Ideler, 1837, Schwartze, 1848, Lagarde, Psalterii versio Memphitica, Göttingen, 1875, F. Rossi, Cinque manoscritti &c., 1894; Job by H. Tattam, 1846; the Prophets by Tattam (Prophetae minores, 1836, Proph. maiores, 1852).

SAHIDIC. Lagarde, Aegyptiaca, 1883; Ciasca, Sacr. bibl. fragm. Coptosahidica Musei Borgiani, 1885— 9; Amélineau, Fragments coptes in Recueil v. (1884), and Fragments de la version thébaine, ib. vii.— x. (1886— 9); the same scholar has edited Job in Proceedings of the Soc. of Bibl. Arch., 1887; O. v. Lemm, Bruchstücke, 1885, Sahidische Bibelfragmente, 1890; Krall, Mittheilungen, 1887; F. Rossi, Papiri Copti, 1889, Un nuovo codice, 1893; Maspéro, Fragments de l᾿Ancien Testament in Mémoires publiés par les membres de la mission arch. française au Caire, vi., 1892; E. A. T. W. Budge, The earliest known Coptic Psalter, 1898²⁸⁷²⁸⁷On the correspondence of this Psalter with cod. U see below, p, 143.; Coptic Biblical Texts in the Dialect of Upper Egypt, 1912; N. Peters, Die sahidisch-koptische Übersetzung d. Buches Ecclesiasticus . . . untersucht, 1898; P. Lacau, Textes de l᾿A. T. en copte sahidique, 1901; Sir H. Thompson, The Coptic Version of certain books of the O. T., 1908; A Coptic Palimpsest, 1911.

MIDDLE EGYPTIAN, &c. Tuki, Rudimenta linguae Coptae, 1778; Quatremère, Recherches sur la langue et la littérature de l’Egypte, 1808; Zoega, Catal. codd. Copt., 1810; Engelbreth, Fragmenta Basmurico-Coptica V. et N. T., 1811; Von Lemm, Mittelägyptische Fragmente, 1885; Krall, Mittheilungen, 1887; Bouriant in Mémoires de l’Institut égyptien ii., 1889, and in Mémoires publiés par &c. vi. 1; Steindorff, die Apokalypse des Elias, p. 2 ff. (Leipzig, 1899).

It may reasonably be expected that the Egyptian versions of the Old Testament; when they have been more fully recovered and submitted to examination by experts, will prove 108to be of much importance for the criticism of the text of the LXX. Ceriani²⁸⁸²⁸⁸See O. T. in Greek, iii. p. ix. has shewn that the Greek text of Cod. Marchalianus agrees generally with that which underlies the Bohairic version of the Prophets, whilst both are in harmony with the text which is quoted by Cyril of Alexandria. A German scholar²⁸⁹²⁸⁹A. Schulte in Theol. Quartalschrift, 1894— 5; see Hyvernat, p. 69., starting with the Bohairic Prophets, finds that their text is similar to that of the Codex Alexandrinus, the Codex Marchalianus, a series of cursive Greek MSS., some of which had been recognised by Cornill²⁹⁰²⁹⁰Ezechiel, p. 66 ff. as Hesychian (22, 23, 26, 36, 40, 42, 49, 51, 62, 86, 91, 95, 97, 106, 114, 130, 147, 153, 185, 228, 233, 238, 240, 310, 311), and the Greek columns of the Complutensian Polyglott. Of the Sahidic fragments, Job is perhaps "a translation of Origen's revised text, with the passages under asterisk omitted²⁹¹²⁹¹Burkitt in Encycl. Brit. iv. 5027; cf. Hatch, Essays, p. 215 ff.; Dillmann, Textkritiches zum Buche Ijob, p. 4; Burkitt, O. L. and Itala, p. 8; Kenyon, Our Bible and the ancient MSS., p. 751.," whilst Isaiah is distinctly Hexaplaric, and traces of the influence of the Hexapla are also to be found in Proverbs, Ecclesiastes and Ezekiel, although in varying degrees. On the whole it is natural to expect the Hesychian recension to be specially reflected in Egyptian versions. But other influences may have been at work²⁹²²⁹²Hyvernat, p. 71., and much remains to be done before these versions can be securely used in the work of reconstructing the text of the Greek Old Testament²⁹³²⁹³See the remarks of F. Robinson in Hastings' Dict. of the Bible i. 673a..

LITERATURE. Quatremère, Recherches; Zoega, Catalogus; L. Stern, Koptische Grammatik, 1880; Kopten, Koptische Sprache u. Litteratur, 1886; Scrivener-Miller, ii. p. 91 ff. (J. B. Lightfoot and A. C. Headlam); Gregory, prolegg., p. 859 ff.; J. P. P. Martin, Intr., partie théor., p. 310 ff.; H. Hyvernat, Étude sur les versions coptes de la Bible in Revue biblique, v. 3, 4, vi. 1; E. Nestle, Urtext, p. 144 ff.; W. E. Crum, Coptic Studies, 1897— 8; Catalogue of Coptic MSS. in Brit. Museum, 1905; A. E. Brooke in J. Th. St. iii.

3. THE ETHIOPIC VERSION.

Ethiopia is said to have been evangelised in the fourth century from Tyre. The Tyrian missionaries were probably of Greek speech²⁹⁴²⁹⁴Charles (art. Ethiopic Version, in Hastings' D. B. i. p. 792) states that "the Abyssinians first received Christianity through Aramaean missionaries." But Tyre in the fourth century was as Greek as Alexandria and Antioch., and brought with them the Greek Bible. But apart from this, the contiguity of Ethiopia to Egypt, and the circumstance that the first Bishop of Auxume received consecration at Alexandria, create an a priori probability that any early translations from the Old Testament into Ethiopic were based upon the Septuagint, whether immediately or through the Coptic versions.

Dillmann, who at one time had explained the numerous transliterations and other approaches to the Hebrew in the existing Ethiopic version by assuming that the translators worked upon a Hexaplaric text, ultimately found cause to classify the MSS. under three heads, (1) those which on the whole represent the text of the LXX. on which he supposed the version to have been based; (2) those of a later recension—the most numerous class—corrected by other MSS. of the LXX.; (3) those in which the original version has been revised from the Hebrew²⁹⁵²⁹⁵Nestle, Urtext, p. 148. Loisy, Histoire critique, I. ii. p. 231.. Lagarde, on the other hand, suggested that the version was translated from the Arabic, as late as the fourteenth century, and maintained that in any case the printed texts of the Ethiopic Old Testament depend upon MSS. which are too late and too bad to furnish a secure basis for the employment of this version in the reconstruction of the Septuagint²⁹⁶²⁹⁶Ankündigung einer neuen Ausgabe der gr. Übersetzung d. A.T., p. 28; cf. Materialen, i. p. iii.. "These suggestions are not however supported by a closer examination of the Ethiopic version of the Octateuch. The text as printed by Dillmann, and especially the readings of the oldest MS. he used, which is supported by a dated thirteenth century MS. brought from Abyssinia to Paris since 110his edition was published, betray direct descent from a Septuagint text of a somewhat interesting type, which had apparently undergone less Hebrew or hexaplar revision than the Greek ancestors of the Armenian and Syro-hexaplar versions. We are safe in concluding with Charles, 'It is unquestionable that our version was made in the main from the Greek²⁹⁷²⁹⁷This criticism of Lagarde's view is due to Mr N. McLean, who has recently examined the Ethiopic Genesis for the larger Cambridge Septuagint..'"

The Ethiopic version of the Old Testament contains all the books of the Alexandrian canon except 1— 4. Maccabees, together with certain apocrypha which are not found in MSS. of the LXX. (Enoch, the Book of Jubilees, 4 Esdras, &c.). A considerable part of it has appeared in print. Dillmann edited the Octateuch and the four books of Kingdoms (1853— 71), and the deuterocanonical books (1894); the book of Joel appeared in Merx, Die Prophetie des Joels, the book of Jonah in W. Wright's Jonah in four Semitic versions (London, 1857). The Psalms were printed by Ludolf (1701), Rödiger (1815), Dorn (1825), and Jeremiah, Lamentations and Malachi by Bachmann (1893); Bachmann also edited the Dodecapropheton, and part of Isaiah.

Lists of the MSS, may be seen in Wright, Ethiopic MSS. of the British Museum (London, 1878); Zotenberg, Catalogue des MSS. éthiopiens de la Bibliothèque Nationale (Paris, 1877); D’Abbadie, Catalogue raisonné de MSS. éthiopiens (Paris, 1859); Dillmann, Catalogus MSS. Aethiop. in Bibliotheca Bodleiana (Oxford, 1848), and Abessinische Handschr. d. k. Biblioth. zu Berlin; Müller, Aethiofi. Handschr. der k. Hofbiblioth. in Wien (ZDMG. xvi. p. 554). For fuller information as to this Version see F. Prätorius, Urtext, p. 147 ff.

4. THE ARABIC VERSION.

The Arabic Old Testament printed in the Paris and London Polyglotts is a composite work, the Hexateuch being a translation from the Hebrew, and the books of Judges, Ruth, 1 Regn. i.—2 Regn. xii. 17, Nehemiah i.—ix. 27, and Job from the Peshitta; the Septuagint has supplied the basis for 111the other poetical books and for the Prophets²⁹⁸²⁹⁸Loisy, Hist. crit., I. ii. p. 239. Mr Burkitt in Hastings' D. B. (i. p. 137) writes "J(udges), S(amuel), K(ings), and Ch(ronicles), are all from the Peshitta.". Some of the MSS. exhibit in certain books a translation which has come from the LXX. through the Coptic; the book of Job in this version has been published by Lagarde (Psalterium Job Proverbia arabice, Göttingen, 1876)²⁹⁹²⁹⁹Lagarde gives for the Psalter four texts, viz. those published at Rome (1614), Paris (1645), Quzhayya (1612), Aleppo (1706); for Job, besides the versions mentioned in the text, that of the Paris Polyglott..

The Arabic version directly derived from the LXX. is said to exhibit in the Prophets a text akin to that of Cod. A (Ryssel, in ZA W. 1885, p. 102 ff., 158). It shews traces of Hexaplaric influence (H. Hyvernat, in Vigouroux, D. B. i. p. 846).

EDITIONS of Arabic versions of the Septuagint. Besides the Polyglotts (Paris, 1645; London, 1652), mention may be made of the Psalters published at Genoa, 1516; Rome, 1614 and 1619; Aleppo, 1706; London (S.P.C.K.), 1725. In W. Wright's Book of Jonah the Arabic is from a MS. in the Bodleian (see p. vii.). Cf. H. Hyvernat, op. cit.

MSS. Lists of MSS. of the Arabic versions of the Old Testament will be found in the Preface to Holmes and Parsons, vol. i.; Slane's Catalogue des MSS. Arabes de la Bibl. nat.; Mrs M. D. Gibson's Studia Sinaitica, iii. (London, 1894), Catalogue of Arabic MSS. at Sinai (codd. 1— 67). Cf. Hyvernat, op. cit.

LITERATURE. Schnurrer, Bibliotheca Arabica, 1780; H. E. G. Paulus, Bodleiana specimina versionum Pent. Arab., 1789; Eichhorn, Einleitung, § 275 ff.; R. Holmes, Praef. ad Pent.; Rödiger, De origine et indole Arab. libr. V. T. interpretationis (Halle, 1829). Among more recent works reference may be made to Cornill, Ezechiel, p. 49 f.; Loisy, Hist. crit. I. ii. p. 238; Nestle in Urtext, p. 150 ff.; F. C. Burkitt, art. Arabic Versions, in Hastings' D. B. i. p. 136 ff.; H. Hyvernat, op. cit.

5. THE SYRIAC VERSIONS.

According to Moses bar-Cephas († 913), there are two Syriac versions of the Old Testament—the Peshitta, translated 112from the Hebrew in the time of King Abgar, and the version made from the Septuagint by Paul, Bishop of Tella. This statement is neither complete nor altogether to be trusted, but it may serve as a convenient point of departure for a summary of the subject.

(1) The origin of the Peshitta is still as obscure as when Theodore of Mopsuestia wrote: ἡρμήνευται δὲ ταῦτα εἰς μὲν τὴν τῶν Σύρων παῤ ὅτου δήποτε, οὐδὲ γὰρ ἔγνωσται μέχρι τῆς τήμερον ὅστις ποτὲ οὗτός ἐστιν³⁰⁰³⁰⁰Migne, P. G., lxvi. 241; cf. ib. 252 f., 263, 466 ff., 492 ff.. That the translation on the whole was made from the Hebrew is the verdict of modern scholars as it was that of Moses bar-Cephas. Yet certain books display the influence of the LXX. While "the Pentateuch follows the Hebrew text and the Jewish exegesis, Isaiah and the twelve Minor Prophets contain much which is from the LXX., and the influence of the Greek version appears to have been felt also in the Psalter³⁰¹³⁰¹Nestle in Urtext, p. 230; cf. Bleek-Wellhausen, pp. 558-560; W. E. Barnes in J. Th. St. ii. 186 ff.." From the first the Peshitta seems to have included the non-canonical books of the Alexandrian Bible except 1 Esdras and Tobit, "and their diction agrees with that of the canonical books among which they are inserted³⁰²³⁰²Gwynn, D. C. B., iv. p. 434.."

(2) The Syriac version ascribed to Paul, Bishop of Telladhe-Mauzelath (Constantine) in Mesopotamia, was a literal translation of the LXX. of the Hexapla, in which the Origenic signs were scrupulously retained. A note in one of the rolls of this version assigns it to the year 616— 7; the work is said to have been produced at Alexandria under the auspices of Athanasius, Monophysite Patriarch of Antioch, who with five of his suffragans had gone thither to visit the Alexandrian Patriarch. Paul of Tella and Thomas of Harkel appear to have been of the party, and their visit in Alexandria led to 113the translation of the entire Greek Bible into Syriac, the New Testament having been undertaken by Thomas, while Paul worked upon the Old³⁰³³⁰³Gwynn, Paulus Tellensis and Thomas Harklensis, in D. C. B. iv. pp. 266 ff., 1014 ff..

The version of Paul of Tella, usually called the Syro-Hexaplar, was first made known to Europe by Andreas Masius (Andrew Du Maes, † 1573). In editing the Greek text of Joshua he used a Syriac MS. which contained part of Deuteronomy, Joshua, Judges, Kings, Chronicles, Ezra, Esther, Judith, and part of Tobit, in this translation. The codex which he employed has disappeared, but the Ambrosian library at Milan possesses another, possibly a second volume of the lost MS., which contains the poetical and prophetic books, in the order Psalms, Job, Proverbs, Ecclesiastes, Song of Solomon, the two Wisdoms, the twelve Prophets, Jeremiah (with Baruch, Lamentations, and the Epistle), Daniel (with Susanna and Bel), Ezekiel, Isaiah. Portions of the historical books of the Syro-Hexaplar³⁰⁴³⁰⁴Viz., parts of Genesis and Joshua; half of Numbers, nearly the whole of Judges, Ruth, and 3 Kingdoms, and Exodus complete. have been discovered among the Nitrian MSS. of the British Museum, and a catena, also at the Museum, contains fragments of Chronicles and the books of Esdras, while the Paris Library contributes 4 Kingdoms. Norberg edited Jeremiah and Ezekiel in 1787; Daniel was published by Bugati in 1788 and the Psalms in 1820; Middeldorpf completed the prophetical and poetical books in his edition of 1835, and in 1861 Ceriani added Baruch, Lamentations, and the Ep. of Jeremiah. Of the historical books Judges and Ruth were published by Skat Rördam in 1861, and Genesis and Exodus (i.—xxxiii. 2) by Ceriani (Mon. sacr. et prof. ii.), who has also given to the world the Milan fragments in Mon. vol. vii.

The Hexapla, Tetrapla, and occasionally the Heptapla, are 114mentioned as the sources of the text in the subscriptions to the books of the Syro-Hexaplar. These subscriptions were doubtless translated with the rest of the Greek archetypes, but they shew the character of the copies employed by the translators. The version is servile to such an extent as sometimes to violate the Syriac idiom³⁰⁵³⁰⁵Field, Prolegg. in Hex., p. lxix., where many instances are produced.. It is obvious that this extreme fidelity to the Greek, while it must have hindered the use of the version in the Monophysite churches of Syria, is of vast advantage to the Biblical critic. It places in his hands an exact reflexion of the Hexaplaric LXX. as it was read at Alexandria at the beginning of the 7th century, derived ultimately from the Hexapla and Tetrapla through the recession of Eusebius. Thus it supplements our scanty stock of Greek Hexaplaric MSS., and indeed forms our chief authority for the text of Origen's revision. In the case of one of the canonical books the version of Paul of Tella renders even greater service. One of the Greek texts of Daniel—that which Origen regarded as the true Septuagintal text—has survived only in a single and relatively late MS. The Syro-Hexaplar here supplies another and earlier authority, which enables us to check the testimony of the Chigi Greek.

(3) Other Syriac versions made from the Greek.

(a) Fragments of a Syriac version in the Palestinian dialect have been printed by Land, Anecdota Syriaca, iv. (Leyden, 1875), J. R. Harris, Biblical Fragments from Mt Sinai (London, 1890), G. H. Gwilliam, Anecdota Oxoniensia, Semitic Series, I. v., ix. (Oxford, 1893— 6), G. Margoliouth, Liturgy of the Nile (London, 1897), and Mrs Lewis, Studia Sinaitica, vi. (London, 1897)³⁰⁶³⁰⁶The fragments in Studia Sinaitica are accompanied by critical notes, the work of Dr Nestle, in which they are carefully compared with the Greek text (pp. xl.— lxxiv.).. This version has been made from the LXX.; in the Books of Kings the text is now known not to be Lucianic, as it was at first supposed to be (Anecd. 115Oxon. ix. p. 32); in the Greater Prophets, it is in part at least Origenic (Studia Sinaitica, pp. xvi., lxiii.); Job seems to have contained the interpolations from Theodotion which are found in the extant Greek texts of that book³⁰⁷³⁰⁷Burkitt in Anecd. Oxon., Semitic ser., 1. ix. p. 44, and cf. Nestle's notes to Studia Sinaitica, vi..

The following is a complete list of the Palestinian fragments included in the publications mentioned above: Gen. i. 1— iii.— 24, vi. 9— ix. 19, xviii. 1— 5, 18— xix. 30, xxii. 1— 19; Ex. viii. 22^b— xi. 10, xxviii. 1— 12^a; Num. iv. 46 f., 49— v. 2 f., 4, 6, 8; Deut. vi. 4— 16, vii. 25— 26^a, x. 12— xi. 28, xii. 28— xiv. 3; 2 Regn. ii. 19— 22; 3 Regn. ii. 10^b— 15^a, ix. 4— 5^a; Pss. viii. 2 f., xxi. 2, 19, xxii. 1, 5, xxiv. 1 f., xxix. 2, 4, xxx. 2, 6, xxxiv. 1, 11, xxxvii. 2, 18, xl. 2, 5, 7, xliii. 12— 27, xliv.— xlvi., xlviii. 15 ff., xlix. 1— 9, liv. 2, 22, lv. 7 ff., lvi. 1— 7, lxiv. 2, 6, lxviii. 2, 3, 22, lxxvi. 2, 21, lxxvii. 52— 65, lxxxi., lxxxii. 1— 10, lxxxiv. 2, 8, lxxxv. 1, 15 f., lxxxvii. 2, 5— 7, 18, lxxxix. 1— xc. 12, xcvii. 1, 8 f., ci. 2 f.; Prov. i. 1— 19, ix. 1— 11; Job xvi. 1— xvii. 16, xxi. 1— 34, xxii. 3— 12; Sap. ix. 8— 11, 14— x. 2; Amos ix. 5— 14^a, viii. 9— 12; Mic. v. 2— 5; Joel i. 14— ii. 27, iii. 9— 21; Jonah; Zech. ix. 9— 15, xi. 11^b— 14; Isa. iii. 9^b— 15, vii. 10— 16, viii. 8— xi. 16, xii. 1— 6, xiv. 28— 32, xv. 1— 5, xxv. 1— 3^a, xxxv. 1— 10, xl. 1— 17, xlii. 5— 10, 17— xliii. 21, xliv. 2— 7, l. 4— 9, lii. 13— liii. 12, lx. 1— 22, lxi. 1— 11, lxiii. 1— 7; Jer. xi. 18— 20³⁰⁸³⁰⁸See Studia Sin., vi. p. xiv. f. For recent additions see Nestle in Hastings' D. B. iv. 447..

(b) Mention is made³⁰⁹³⁰⁹Bickell, Conspectus rei Syr. lit., p. 9; cf. Ebedjesu in Assemani, iii. 71. of a version of the Greek Old Testament attempted by the Nestorian Patriarch Mar Abbas (A.D. 552). But notwithstanding the declared preference of Theodore for the LXX., the Nestorians have always used the Peshitta, and there is no extant Nestorian version from the Greek.

(c) Of Jacobite versions from the LXX. there were several. (1) Polycarp the chorepiscopus, who in the fifth century laboured upon a translation of the New Testament under the auspices of Philoxenus, the Monophysite Bishop of Mabug, is known to have rendered the Greek Psalter into Syriac. The margin of the Syro-Hexaplar³¹⁰³¹⁰Field, Hexapla, ii. p. 448. mentions a Philoxenian 'edition' of Isaiah, 116to which two fragments printed by Ceriani³¹¹³¹¹Mon. sacr. et prof. v.; cf. Gwynn in D. C. B. iv. p. 433. from the British Museum MS. Add. 17106 are believed to belong. The text of these fragments agrees on the whole with that of the Lucianic MSS. of the Prophets. (2) Another Monophysite, Jacob of Edessa, applied himself in 704— 5 to the revision of the Syriac Old Testament, using for the purpose the Hexaplaric LXX.³¹²³¹²Gwynn, D. C. B. iii., and the fragments of the other Greek translations. Some books of this revised version exist in MS. at London and Paris³¹³³¹³1 Regn. i. 1— 3 Regn. ii. 11, and Isaiah are in the London MSS. lx., lxi. (Wright, Catalogue, p. 37 ff.), and the Pentateuch and Daniel are preserved at Paris., and a few specimens have been printed³¹⁴³¹⁴See Ladvocat, Journal des savants, for 1765; Eichhorn, Bibliothek, ii. p. 270; De Sacy, Notices et extraits, iv. p. 648 ff.; Ceriani, Mon. sacr. et prof. v. i. 1. .

(d) From Melito downwards the Greek fathers refer occasionally to the Greek renderings of an interpreter who is called ὁ Σύρος The student will find in Field's prolegomena a full and learned discussion of the question who this Syrian interpreter was. Field inclines to the opinion that he was a bilingual Syrian, of Greek origin, who translated into Greek from the Peshitta³¹⁵³¹⁵On the other hand see Scrivener-Miller, ii. p. 7, note; and Bleek Wellhausen (1893), p. 560..

EDITIONS. PESHITTA. Lee, V. T. Syriace (London, 1823); O. and N. T., 1826. A complete Syriac Bible has recently been published by the Dominicans of Mosul (⁽¹⁾1887— 91, ⁽²⁾1888— 92).

SYRO-HEXAPLAR. A. Masius, Josuae-historia illustrata (1574); M. Norberg, Codex Syriaco-Hexaplaris (1787); C. Bugati, Daniel (1788), Psalmi (1820); H. Middledorpf, cod. Syrohexapl., lib. IV. Reg. e cod. Paris. Iesaias &c. e cod. Mediol. (1835): Skat Rördam, libri Iudicum et Ruth sec. Syro-hexapl. (1861); P. de Lagarde, V. T. ab Origene recensiti fragmenta ap. Syros servata v. (1880), and V. T. Graeci in sermonem Syrorum versi fragm. viii. (in his last work Bibliothecae Syriacae . . . quae ad philologiam sacram pertinent, 1892); G. Kerber, Syro-hexaplarische Fragmente (ZATW., 1896). Ceriani has published 117the contents of the London MS. in Monumenta sacra et profana, ii., and those of the Milan MS. in vol. vii. (1874) of the same series³¹⁶³¹⁶For the Apocryphal books see Lagarde, Libri V T. apocr. Syriace, and Bensly-Barnes, The fourth book of Maccabees in Syriac (Camb. 1895). .

LITERATURE. G. Bickell, Conspectus rei Syrorum literariae (1871); Field, Hexapla, I. p. lxvii. sqq. (1875); W. Wright, Syriac literature in Encycl. Britannica, xxii. (1887); E. Nestle, Litteratura Syriaca (1888), and Urtext (1897), p. 227 ff.; Scrivener-Miller, ii. p. 6 ff.; Gregory, p. 807 ff.; J. P. P. Martin, Introduction (p. théor.), p. 97 ff.; Loisy, Histoire critique I. ii. p. 234 f.; E. Nestle, Syriac Versions (in Hastings' D. B. iv.

6. THE GOTHIC VERSION.

About the year 350 a translation of the Bible into the Gothic tongue was made by Ulfilas (Wulfila)³¹⁷³¹⁷Socr. ii. 11, iv. 33, Theodoret iv. 37, Philostorg. ii. 5., the descendant of a Cappadocian captive who had been brought up among the Goths in Dacia, and was in 341 consecrated Bishop of the Gothic nation, which was then beginning to embrace Arian Christianity. According to Philostorgius he translated the whole of the Old Testament except the books of Kingdoms, which he omitted as likely to inflame the military temper of the Gothic race by their records of wars and conquests (Philostorg. loc. cit.: μετέφρασεν εἰς τὴν αὐτῶν φωνὴν τὰς γραφὰς ἁπάσας πλήν γε δὴ τῶν Βασιλειῶν ἅτε τῶν μὲν πολέμων ἱστορίαν ἐχουσῶν, τοῦδε ἔθνους ὄντος φιλοπολέμου). Unfortunately only a few scanty fragments of the Gothic Old Testament have been preserved, i.e., some words from Gen. v. 3— 30, Ps. lii. 2— 3, 2 Esdr. xv. 13— 16, xvi. 14— xvii. 3, xvii. 13— 45. With the exception of the scrap from Genesis, they are derived from palimpsest fragments belonging to the Ambrosian Library which were discovered by Mai in 1817 and subsequently published at Milan by Mai and Castiglione; and they are printed in the great collection of Gabelentz and Loebe (Ulfilas: V. et N. Testamenti . . . fragmenta, Lipsiae, 1843) and in Migne P. L. xviii.; more recent editions are those of Uppstrom, Upsala, 1854— 7; Massmann, Stuttgart 1855— 7; Stamm, Paderborn, 1865; Bernhardt, Halle, 1875, 1884; G. H. Balg, The First Germanic Bible, Milwaukee, 1891; Stamm-Heyne, 1896.

Lagarde (Librorum V T. canonicorum pars i., p. xiv., 1883) shews by an examination of the Esdras fragments that Ulfilas probably used MSS. of the Lucianic recension, and the same view is held by A. Kisch, Der Septuaginta-Codex des Ulfilas (Monatschrift f. Gesch. u. W. des Judenthums, 1873), and F. Kauffmann, Beiträge zur Quellenkritik d. gothischen Bibelübersetzung (Z. f. d. Phil. 1896). Ulfilas was in Constantinople for some time about 340, and his MSS. of the LXX. were doubtless obtained in that city, which according to Jerome was one of the headquarters of the Lucianic LXX. ("Constantinopolis usque Antiochiam Luciani martyris exemplaria probat").

7. THE ARMENIAN VERSION.

Armenian writers of the fifth century ascribe the inception of the Armenian Bible to Mesrop (354— 441) and his associates. The book of Proverbs was the first translated, whether because it stood first in the volume³¹⁸³¹⁸So F. C. Conybeare (Hastings, i. p. 152). In Scrivener-Miller, ii. p. 151, he suggests that the earlier books had been rendered previously. on which the translators worked, or because its gnomic character gave it a special importance in their eyes. The work is said to have been begun at Edessa, but MSS. were afterwards obtained from Constantinople; and Moses of Khoren, a nephew and pupil of Mesrop, was despatched to Alexandria to study Greek in order to secure "a more accurate articulation and division"³¹⁹³¹⁹On this see Conybeare, Scrivener-Miller, ii. p. 153. of the text. Moses indeed affirms that the earliest translations of the O.T. into Armenian were from the Syriac, and his statement receives some confirmation from the mention of Edessa as the place of origin, and from the circumstance that Syriac was the Church-language of Armenia before the introduction of the Armenian alphabet³²⁰³²⁰See Dr Salmon in D. C. B., iii. p. 908.. On the other hand the existing Armenian version 119is clearly Septuagintal. It fits the Greek of the LXX. "as a glove the hand that wears it"; keeping so close to the Greek that it "has almost the same value for us as the Greek text itself from which (the translator) worked would possess³²¹³²¹Conybeare, op. cit., p. 151 f. He attributes the composite character of the Armenian text (of which he gives instances) to Hexaplaric influences.." But, as Lagarde has pointed out³²²³²²Genesis Gr., p. 18., the printed text is untrustworthy, and the collation made for Holmes and Parsons cannot be regarded as satisfactory. A fresh collation will be made for the larger edition of the Cambridge Septuagint³²³³²³Mr McLean, who has collated the greater part of the Octateuch, informs me that "the Armenian shews a typical hexaplar text in Genesis and Exodus, agreeing closely with the Syriaco-hexaplar version, and in varying degrees with the MSS. that compose the hexaplar group." "The hexaplar element (he adds) is much less in evidence in Leviticus, Numbers, and Deuteronomy, but again appears strongly in Joshua, Judges, and Ruth.".

The order of the books of the O.T. in Armenian MSS., as given by Conybeare³²⁴³²⁴Op. cit., p. 152 f. (Octateuch, 1—4 Regn., 1—2 Paralipp., 1 and 2 Esdr., Esther, Judith, Tobit, 1—3 Macc., Psalms, Proverbs, Ecclesiastes, Canticles, Wisdom, Job³²⁵³²⁵In some MSS. Job precedes the Psalter., Isaiah, the Minor Prophets, Jeremiah, with Baruch and Lamentations, Daniel, Ezekiel) is on the whole consistent with the grouping found in the oldest Greek authorities³²⁶³²⁶See Part II. c. i., and seems to point to the use by the translators of good early codices.

MSS. Few codices of the entire Bible are earlier than the 13th century; one at Edschmiatzin belongs to the year 1151. Holmes assigns his Arm. 3 to A.D. 1063, but according to Conybeare it is a MS. of the eighteenth century.

EDITIONS. Venice (Psalter), 1565; Amsterdam, 1666; Constantinople, 1705; Venice, 1805 (the first edition which is of any critical value, by J. Zohrab); Venice, 1859— 60 (by the Mechitarist fathers of San Lazzaro).

LITERATURE R. Holmes, Praef. ad Pent.; F. C. Conybeare in Scrivener-Miller, ii. 148 ff. and in Hastings' D. B., l.c.; 120H. Hyvernat, in Vigouroux' D. B.; C. R. Gregory, Prolegg. p. 912 ff.; J. P. P. Martin, Introd. (p. théor.), p 323 ff.; E. Nestle in Urtext, p. 155, where fuller bibliographical information will be found.

8. THE GEORGIAN VERSION.

The origin of this version is obscure. According to Moses of Khoren, the Georgian as well as the Armenian version was the work of Mesrop. Iberia seems to have received the Gospel early in the fourth century, if not before; but it may have possessed no translation of the Scriptures until the movement initiated in Armenia by Mesrop had communicated itself to the neighbouring region. That the Georgian Old Testament was based upon the Greek is said to be manifest from the transliteration of Greek words which it contains.

MSS. A Psalter of cent. vii.—viii. is preserved at the monastery of St Catherine's, Mt Sinai, and at Athos there is a MS., dated 978, which originally contained the whole Bible, but has lost Lev. xii.—Joshua. Both the Sinai library and the Patriarchal library at Jerusalem are rich in Georgian MSS.

EDITIONS. The Georgian Bible was printed at Moscow in 1743 and at St Petersburg in 1816 and 1818; the Moscow edition is said to have been adapted to the Russian Church Bible.

LITERATURE. F. C. Alter, über Georgianische Litteratur (Vienna, 1798); A. A. Tsagarelli, An account of the monuments of Georgian Literature (in Russian, St Petersburg, 1886— 94; A. Khakhanow, Les MSS. Georgiens de la Bibliothèque Nationale à Paris (without place or date, ? 1898).

9. THE SLAVONIC VERSION.

The Greek Bible was translated into Slavonic by the brothers Cyril and Methodius, from whom in the ninth century the Slavs received the faith. Of the Old Testament the Psalter alone was finished before the death of Cyril, but according to contemporary testimony Methodius brought the work to completion. As a whole this original version no 121longer exists, the codices having perished in the Tartar invasion of the thirteenth century; and the fragments of the Old Testament of Cyril and Methodius which are embedded in the present Slavonic Bible are "so mixed up with later versions as to be indistinguishable³²⁷³²⁷The Russian Bible, in Ch. Quart. Review, xli. 81 (Oct. 195), p. 219. ." The existing version has not been made uniformly from the Greek. Esther was translated from the Hebrew, while Chronicles, Ezra and Nehemiah, and certain other books, were rendered from the Latin Vulgate in the fifteenth century. On the other hand the Octateuch, the books of Kingdoms, and the poetical books are from the Greek, and some of them, especially the Octateuch, contain old materials probably due, at least in part, to the work of Cyril and Methodius.

A Psalter in the Glagolitic script, preserved at Sinai, has been edited by Geitler (Agram, 1883); and there is a critical edition of the Slavonic Psalter by Amphilochius (Moscow, 1874— 9).

So far as the Slavonic Old Testament is based on the LXX., its text is doubtless Lucianic; cf. Lagarde, Praef. in Libr. V. T. can. i. p. xv. "ni omnia fallunt Slavus nihil aliud vertit nisi Luciani recensionem," and Leskien in Urtext, p. 215, "dass im allgemeinen der Kirchenslavischen Übersetzung der griech. Text der Lucianischen (Antiochenisch-Konstantinopolitanischen) Rezension zu Grunde liegt ist sicher."

LITERATURE. The Russian authorities are given by Mr Bebb in Scrivener-Miller, ii. p. 158. See also Gregory, Prolegg. p. 1112 ff.; Professor Leskien of Leipzig in Urtext, p. 211 ff.; the article in Ch. Quarterly Review cited above; and Th. Literaturzeitung, 1901, col. 571.