(2) Founded mainly on the basis of their work, two theories were suggested to supply this deficiency.
(a) Rendel Harris (' A Study of Codex Bezte in TS ii. 1 [1891], and Foiir Lectures on the Western Text, Cambridge, 1894) and F. H. Chase (The Old Syriac Element in the Text of Codex Bezce, London,
1893) thought that retranslation from Latin and Syriac would solve the problem ; but no amount of retranslation will account for the relatively long Bezan additions.
(6) F. ^Mdiss, [Acta Apostolorum secundum formam quce videtur JRomanam, Leipzig, 1897, and also in his commentary, Acta Apostolorum, Gottingen, 1895) thought that Luke issued the Acts in two forms : one to Theophilus (the Alexandrian text), and the other for Rome (the Western text); but his reconstruction of the Roman text is scarcely satisfactory, and the style of the additions is not sufficiently Lucan.
(3) More recently von Soden [Die Schriften des Neuen Testaments, 1902-1910, p. 1834 fi'.), using the new facts as to the MSS summarized above, has revived Blass's theory in so far that he thinks that the interpolated text witnessed to by 55 and the Latin Versions and Fathers really goes back to a single original ; but, instead of assigning this original to Luke, he attributes it to Tatian, who, he thinks, added a new recension of Acts to his Diatessaron. The weak point in this theory is that the only evidence that Tatian edited the Acts is a passage in Eusebius * which states that he emended 'the Apostle.' This may refer to Acts, but more probably refers to the Ej)istles. Accord- ing to von Soden, the / text did not contain all the interpolations, K contained still fewer, and H contained none. He thinks that in the 2nd cent, there existed side by side the Tatianic text and a non-interpolated text which he calls I-H-K. From these two texts there arose the Latin Version — predominantly Tatianic — and most of the early Fathers were influenced by Tatian. Later on, in the 4th cent., three revisions were made : (a) H, by Hesychius in Alexandria, wiiich preserved in the main the text of I-H-K without the Tatianic ad- ditions, but with a few other corruptions ; (b) K, by Lucian, in Antioch, which had many Tatianic corruptions, as well as some of its own ; (c) /, in Palestine, possibly in Jerusalem, which preserved many Tatianic additions, though in a few cases keeping the I-H-K text against H. 55 (D) is the best example of this text, but has suffered from the addition of a much greater degree of Tatianic corruption than really belongs to the / text, owing to Latin influence.
The general relations of the various forms of the text, according to von Soden, can be shown roughly in the following diagram :
lU-K
H
I
/b
/c KT
Obviously this complicated theory cannot be dismissed without much more attention than it has yet received. It may prove that the ' text with additions ' is not Tatianic but is nevertheless a single text in origin. It is also very desirable to investigate how far it is possible to prove that there was an / text, derived from I-H-K, which
• TOv S' anoa-ToKov (fmcri To\iJifj<TaC Tivas avrhv fiLera4>pa.(rai tjytovat (US tniSiopdovixevov ai/riav rriv Trjs 4>pd(Teois trvvra^iv (Eus. HE iv. 29. 6). This scarcely sounds as though a series of interpolations was intended.
AUTtt OF THE APUiSTLES
ACTS or THE APOSTLES
17
nevertheless did not possess, in its oi-iginal state, all the ' Bezan ' interpolations.* If it were possible to say that the interpolations were a connected series (whether Tatianic or not is of minor im- portance), the text in which they are imbedded would become extremely valuable, and we should have no right to argue, as is now often done, that, because the interpolations are clearly wrong, there- fore the text in which they are found is to be condemned. For instance, in Ac 15^^ the Latin text interpolates the Golden Eule into the Apos- tolic decrees. That is no doubt wrong. Bat it does not follow that the text omitting ttviktov, in which this interpolation is placed, is not original.
Literature. — The general textual question can be studied in H. von Soden, Die Schriften des NT, Berlin, 1902-1910, esp. pp. 1649-1840 ; F. G. Kenyon, Handbook to the Textital Criti- cism, of the XT"^, London, 1912 ; E. Nestle, Einfilhrung in das griech. NT'-i, Gottingen, 1909 (the Eng. tr. is' from an older edition of the period before von Soden) ; K. Lake, The Text oj the AT6, London, 1911. Important for the study of the Latin are von Soden, ' Das lat. NT in Afrika zur Zeit Cyprians,' TU xxxiii. [Leipzig, 1909]; and Wordsworth-White, A'ow. Test. Dom. nost. les. Christi secundum edit. S. Hieronymi, vol. ii. pt. i. [Oxford, 1905] which also gives a clear statement of the best editions of the separate MSS of the Old Latin and the Vulgate (pp. v-xv).
II. Tradition as to Authorship.— So far back as tradition goes, the Acts is ascribed to St. Luke, the author of the Third Gospel, and com- panion of St. Paul (see, further, Luke). This tradition can be traced back to the end of the 2nd cent. (Clem. Alex. Strom, v. 12; Tertull. de Jejuniis, 10; Iren. adv. Hcer. I. xxiii. 1, in. xii. 12 fl"., IV. XV. 1 ; and the Canon of Muratori). If the connexion with the Third Gospel be accepted, as it certainly ouglit to be, the fact that Marcion used the Gosjiel is evidence for the existence of Acts, unless it be thought that the Gospel was written by a contemporary of Marcion who had not yet written Acts. Farther back tradition does not take us : there are no clear proofs of the use of Acts in the Apostolic Fathers (see The New Testa- ynent in the Apostolic Fathers, Oxford, 1905) or in the early Apologists. (For the later traditions concerning Luke and his writings see Luke.)