Poetical Books.—The extant remains of Job are so scanty that it is difficult to come to any con- clusion regarding the text. Apparently, the Frag. of Fleury, which is found both in the ‘Speculum’ and Priscillian, belongs to the earliest form of the Lat. VS, following the same type of Gr. text as Cyp. and Lucif., and therefore, perhaps, being entitled to the designation ‘ African.’
According to Burkitt, a second ὙΡΘ of OL is found in the quotations of Ambrose, based on the leading uncials of the LXX and in intimate con- nexion with the Greek. The Fragg. which Berger has pub. from the margin of the Leon Cod. also reveal a close attachment to the Greek (esp. Cod. A), and coincide most frequently with the quota- tions of Ambr. and Aug. Perhaps the two last types of text ought to be called ‘Italian.’
For a genuinely ‘ African’ text of Ps our most trustworthy authority is MS L. of Cyprian’s Testi- monia. The Verona and St. Germain Psalters both exhibit a later type of text, although the former has suffered less revision. It would be rash to specify either text definitely as ‘ European’
* Rendel Harris points out some very curious :¢semblances in spelling between Cod. d and Cod. Lugd., which go to suggest, he thinks, that both were Rhéne-valley MSS (Study of Cod. Bez. pp. 29, 30).
t See Berger, Notice, pp. 12, 13.
tSee Lagarde, Septuaginta-Studien, 1892, i. pp. 71, 72; Driver, Notes on Heb. Text of Samuel, pp. lxxvii-lxxxii ; Bur- kitt, Old-Latin and Itala, p. 9.
§ See Berger, op. cit. pp. 14, 15.
ἢ Still more defective appears to be the Cod. Pechianus whicb Sab. gives for the latter part of the book. OL of Est is to be pub. by Thielmann.
{| Old-Latin, etc. pp. 8, 32-34.
LATIN VERSIONS, THE OLD
LATIN VERSIONS, THE OLD 59
or ‘Italian.’ A noteworthy feature is that the
Verona MS shows a striking agreement with Ang.'s text of Ps, while decidedly marked is the
affinity between the St. Germ. Psalter and the
quotations of Cassiodorus the Calabrian. The ortions of the OL Psalter found in the Mozarabic iturgy belong to this latter type of text.*
Proverbs.—Here we can distinguish two recen- sions. The one is represented by Vogel’s Fragg., which agree with the quotations of Cyp. and Vigilius of Thapsus, having also a close resem- blance to the ‘Speculum.’ It may be designated ‘African.’ The other is seen in the Fragg. of the St. Gall MS, No. 11. These have their chief
arallels in Jerome, Ambrose, and Augustine. They therefore belong to the ‘Italian’ family.
The Fragg. of Ecclesiastes and Canticles named in our list are of precisely the same character as the second recension in Proverbs. We may here note that for OT it seems even more difficult than in the NT to draw a line between ‘ European’ and ‘Italian’ texts. Often, indeed, there appears to be none.
Prophetical Books.—In attempting to classify the extant OL texts of the Prophets, we are met, as in the Hexateuch, by the difliculty of conflicting evidence. Fortunately, part of the ground has been cleared by Burkitt in his Rules of Tyconius. Much of what follows depends on his important investigations. The extant Fragg. of the Prophets plainly do not belong to the oldest stratum of the OL tr®. It is needful, therefore, to begin as usual with Cyprian; as we have seen, Tertullian’s quota- tions are of doubtful value. Now, Burkitt has clearly proved} that Tyconius the Donatist (c. A.D. 400) used an OL version of the Prophets (in Isaiah at least +) almost identical with that of Cyp., the only difference being a slight revision of ane Latin. How does this writer stand towards our two chief MS authorities, the Weingarten (w) and Wiirzburg (A) Fragg.? At many points he is in close agreement with both, but especially, perhaps (e.g. in Ezk), with w. Now, a comparison of wand A shows so many agreements in proportion to differences that we cannot help believing that they are copies of the same tr" of the Prophets, whose variation is accounted for by varying elements in the Gr. texts by which they were revised.§ In ἢ, e.g. in the midst of a great mixture of types, the Lue. strain is considerably more prominent than in w.|| Accordingly, we may perhaps call A an ‘Italian’ text; its coincidences with brose, and to a less extent with Augustine, are noteworthy. w is possibly an earlier revision of the same tr®.**
* There seems to be a close resemblance in Ps between Pris- cillian and the ‘Speculum.’ The Latin Psalter with Anglo- Saxon paraphr. pub. by Thorpe, Oxf. 1835, is almost identical with 6 so-called ‘Roman’ Psalter of Jerome, although occasionally it diverges. The text of Cassiod. has also a most intimate connexion with Roman Psalter.
+ Rules of Tyconius Pp. lii-evii. φ
tIt must be noted that there is 8 considerable difference between Tyconius’ text of Is and of Ezk. See important table on p. cvi of op. cit. Burkitt suggests that perhaps there was ‘a partial revision of the African Bible anterior to Cyprian,’ the result of which is seen in the text of Ezk in Tyc. This point has Lope already brought forward in connexion with the quotations ο rian.
# See also Cornill, Das Buch ἃ. Proph. Ezech. p. 31ff. But see Corssen’s most important discussion of the two MSS in Zwet neue I’'ra ὁ ἃ. Weing. Propheten-MS, Berlin, 1899, in which he shows that the variation is largely due to the insertion of glosses in the texts.