06.01. Appendix. THE TEXTUAL EVIDENCE OF THE GROUP DEFG

APPENDIX I THE TEXTUAL EVIDENCE OF THE GROUP DEFG THE evidence of the group DEFG as to the short recension is important, and complicated. The MSS. of the group, and their relations to each other, are as follows:-

D is Codex Claromontanus, of the sixth century, in the Bibliothèque Nationale in Paris. It is a Graeco-Latin MS., written colometrically, not in stichoi, that is to say, the size of the lines is regulated by the sense, not by the number of the letters. It represents two MSS., not merely a text with a translation, but the Greek text belongs to the same type, on the whole, as the original of the Old Latin version, though it has almost certainly been accommodated in many places to a more usual type. It is also famous for possessing the so-called Claromontane stichometry, one of the oldest lists of canonical books, representing, according to Harnack, an Alexandrian document of the fourth century. It was in modern times first used by Beza, who says that it came from the monastery of Clermont Beauvais.

E is Codex Petropolitanus Muralti xx., formerly belonging to the convent of St. Germains in Paris. When the convent was burnt the MS. was bought by Dobrowski at the end of the eighteenth century, and taken to St. Petersburg. It is a copy of D made in the ninth century, and is only valuable in places where D is no longer extant.

F is Codex Augiensis, of the ninth century, a Graeco-Latin MS. formerly the property of the monastery of Augia Dives or Reichenau, and now in the library of Trinity College, Cambridge. The Greek text greatly resembles that of G, but the Latin is that of the Vulgate written in a separate column and not between the Greek lines.

G is Codex Boernerianus, of the ninth century, a Greek MS. with an interlinear Latin translation of an Old Latin type. It was probably written by an Irishman in the Monastery of St. Gall, and is now in Dresden. The text belongs to the same type as D, but is inferior in value and has been much more contaminated with the usual type of late text. The most important points in connection with these MSS. are concerned with the relations subsisting between F and G, and those between D and the archetype of F and G. The relationship between F and G. There has always been a dispute among critics whether F is a copy of G or of the archetype of G, and it is not even now possible to say that any general agreement has been reached. The only way in which such a point can be settled is by a comparison of the places where there are differences of reading. If two MSS. make the same mistakes it is certain that they are closely connected, but it is not necessary to conclude that one is a copy—indeed, absolute proof of this point is almost impossible. If, however, the mistakes can be arranged in two classes, (α) those common to both MSS., and (β) those found in one alone, and the second can all be explained most naturally as mistakes made by the scribe of the second MS. in copying the first, the case for a direct derivation of one from the other is very strong. If, on the other hand, it appears that the MS. which is suspected of being a copy has nevertheless the right reading in some places where the supposed original has a mistake, then the theory of a direct derivation must be abandoned in favour of a common ancestry, unless it can be shown that these right readings are natural corrections made by the scribe. For instance, if it be found that a MS. which is supposed to be the original of another reads ἀνθώπῳ instead of ἀνθρώπῳ, nothing is proved by the fact that the supposed copy has correctly ἀνθρώπῳ, because the correction is obvious. It will, however, be seen that the application of this canon of criticism is very much more difficult than its statement, for who is to decide as to the limits of “natural corrections of obvious mistakes”? This is just the point on which everything turns with regard to G and F. There are a number of places in which G and F have mistakes in common, and a much smaller number where F has a right reading against a mistake in G. Zimmer believes that all of this latter class are “natural corrections of obvious mistakes” in G, while Corssen thinks that this explanation fails, and that F and G are two copies of the same original, G being the more accurate.

It is impossible to reproduce the arguments of these two scholars, for they turn on the nice consideration of a number of small points. Personally, I think that Corssen is right, and that F and G are independent witnesses to a common archetype, Y. The relationship between Y and D. On this point there is less theoretical difficulty. It is generally recognized that D is a better example of its type than Y, but there are sufficient places in which Y seems to have the family reading as against D to show that D is not the archetype of Y, but that D and Y are the representatives of a common ancestor, Z. The reconstruction of Z is not yet complete, and is one of the most obvious needs of textual criticism. But Dr. Corssen’s researches have gone some way to establishing various interesting points. He thinks that Z represents a Graeco-Latin edition of the fifth century, written in cola (i.e. in lines arranged according to the sense, rather than merely according to a fixed number of letters), and that it represents largely the European or Italian type of Latin found in Ambrosiaster and Victorinus. This result is supported by Dr. Souter’s investigations (Ambrosiaster, in Texts and Studies, p. 214), which show that the text of D is especially close to that used by Lucifer of Cagliari in Sardinia. In working out the problem of the text of Z the ideal would be to publish an edition of the three MSS. D, F, G with a reasoned critical commentary establishing the text of Y and of Z. Until the time when this edition appears it is necessary to attempt to anticipate its results for individual passages. For the present purpose two such passages are necessary:(1) The words ἐν Ρώμῃ in Romans 1:7 and Romans 1:15; (2) the Doxology.

(1) ἐν Ρώμῃ inRomans 1:7andRomans 1:15. The facts are these: G reads—

ΠΑCΙ ΤΟΙC ΟΥCΙΝ ΕΝ ΑΓΑΠΗ ΘΕΟU ΚΛΗΤΟΙC ΑΓΙΟΙC omnibus qui sunt in caritate dei vocatis sanctis. As F does not exist at this point, we must assume that this was the reading of Y.

D is not fully extant—the MS. begins with the words κλητοῖς ἁγίοις, but d (the Latin version of D) reads qui sunt Romae in caritate dei vocatis sanctis, with a sign against in referring to a marginal note which has perished. Tischendorf thought that this note was probably a reference to the Vulgate reading dilectis instead of in caritate, but E, the copy of D which is later than D, and has adopted all the corrections in it, reads πᾶσιν τοῖς οὖσιν ἐν Ῥώμῃ, κλητοῖς ἁγίοις. It is therefore more probable, as Zahn suggests (Commentary, p. 617), that the note stated that ἐν ἀγάπῃ was an alternative to ἐν Ῥώμῃ. and E has acted on this and chosen ἐν Ῥώμῃ. In this case the evidence of (D)E goes to support the omission of ἐν Ῥώμῃ. which must be credited to Z as well as to Y; and it remains an open point whether Z may not even have omitted ἐν ἀγάπῃ Θεοῦ as well as ἐν Ῥώμῃ. But this last point, on which no final decision is possible, is not nearly so important as the establishment of the fact that ἐν Ῥώμῃ was not in Z. In Romans 1:15, where the words ἐν Ῥώμῃ recur, G omits them, and is probably to be regarded as the representative of Z, though D F have been accommodated to the usual text. The Doxology (Romans 16:25-27). Either Z omitted the doxology altogether, or it placed it after Romans 14:23. This result is reached by the following considerations. D has the doxology at the end of the Epistle, but F omits it altogether, and G leaves a blank space. It is clear that Y either omitted it or placed it after Romans 14:23. The blank space in G may point to the scribe’s objection to the position of the doxology in Romans 14:23, in spite of the fact that he found it there in his exemplar, or to his knowledge of the fact that Romans 14:23 was the usual place in which to insert it, in spite of the fact that it was not there in his exemplar, and therefore he did not feel justified in inserting it. In any case, Y did not insert the doxology after Romans 16:23. The question therefore only remains, whether D or Y best represents Z. While admitting that there must always be an element of doubt on the subject, I think that Y must be regarded as transcriptionally more probable; the doxology is obviously in a more natural place at the end of the Epistle, and the tendency must have been to move it from Romans 14:23, Romans 15:1-33, Romans 16:1-23, rather than the reverse. The fact that the Antiochene text as a whole kept to Romans 14:23 is no answer to this fact, but merely shows that the Antiochene text preserved, on this point, an early text.

Whether the text of Y really had the doxology at Romans 14:23, or omitted it altogether, is more doubtful. I am inclined to think that there is a slight balance of probability in favour of omission: the tendency of scribes was to invent and insert doxologies and other liturgical additions, not to omit them, and therefore the omission is transcriptionally slightly the more probable reading. The question remains whether, supposing that Y omitted the doxology, it did so because it disturbed the sense, or because it was already omitted by Z. Here unfortunately the evidence will not take us, and it is useless to indulge in guesses.

Dr. Corssen, however, argues that the text of Z in Romans 15:1-33 and Romans 16:1-23 belongs to a different archetype from the rest of the Epistle. His argument is that in these two chapters there are almost as many singular readings which may be attributed to Z as in all the other chapters put together. On this point he seems to be right, and though of course his explanation of the fact is not the only one possible, it is a plausible theory that behind Z was a copy of the Epistle which omitted Romans 15:1-33 and Romans 16:1-23 and ended with Romans 14:23, with or without the doxology, and had no reference to Rome in the opening verses of the Epistle. But the scribe of Z was acquainted with the tradition which had the concluding chapters, and he added them from another MS.

If this be so, the archetype of Z must have been a pure copy of the short recension. This result, though, of course, it cannot be regarded as certain, is important as evidence for the short recension, and also is textually important as tending to show that in the text of the group representing Z the considerable differences from the other early uncials are really due to its representing a different collection of Epistles.