third from late in the first century A.D., the fourth also from the first century A.D.).

(c) The Pseudepigraphs are of a very large number: The Psalter of Solomon was originally written in Hebrew in the latter part of the first century B.C., but is preserved in Greek. The book of Enoch, originally written in Hebrew, is preserved entire only in Æthiopic. The Assumption of Moses is from the first Christian century. Fourth Ezra is from early in the second century A.D. The Apocalypse of Baruch, recently found in the Ambrosian Library at Milan by Ceriani, is from the early second century A.D. The Ascension of Isaiah is from the second half of the second century B.C. The Testament of the Twelve Patriarchs is from the early part of the second century. The book of Jubilees, or Little Genesis, is from the first century B.C. The Sibylline Oracles are in fourteen books, from the second century B. C. to the close of the first century A.D.1

4. The next step in Textual Criticism is to ascertain the original autographs of the Canon of the Law and the Prophets, when they were first collected and fixed. The Septuagint version of the Law and the Prophets, and possibly also of some of the Writings, takes us back of the Maccabean text. The Samaritan codex of the Law gives us on the whole the earliest independent witness to the original text of the Canon of the Law.

5. We have as a final step to ascertain the original text, the autographs of the authors of the Sacred Writings. This we can ascertain on the basis of the texts thus far established, by bringing into consideration parallel passages, such as those of Samuel and Kings on the one side and Chronicles on the other; parallel versions of the same poem, as Ps. 14=53; Ps. 18 = 2 Sam. 22; citations of earlier writings in later ones; and the rules of internal evidence.

The following examples of the application of the genealogical principles to particular passages will suffice:

The English Authorized Version reads in Gen. 4910 "until Shiloh come." The Revised Version retains this in the text,

1 See Briggs, Messiah of the Gospels, pp. 9 seq.; and Messiah of the Apostles, pp. 2 seq.

but puts on the margin other renderings. The Massoretic text,

.may be translated in this way ,עד כי יבא שילה

(a) But the first appearance of this translation known to us is by Sebastian Münster in 1534. Through his influence it passed over into the Great Bible in 1539, and has been retained in all subsequent English versions. Münster seems to have been misled to this interpretation by the use of as a name of the Messiah in the Talmud.1 But that does not justify the translation "until Shiloh come "" any more than the use of Yinnon, Ps. 72, Chaninah, Jer. 1613, Menachem, Lam. 116, and the leprous one, Is. 53*, as names of the Messiah, would justify a translation of all these passages in accordance therewith. In fact there is no such translation of Gen. 4910 known to Jewish tradition. is found in the Old Testament as the name of a place, but nowhere as the name of a person.

(b) The Massoretic pointing really represents the traditional opinion that was a noun with the archaic suffix, meaning his son. This is the interpretation of the Targum Yerushalmi and many Jewish scholars of the tenth century. It is true that there is no such word in Biblical Hebrew. But the Mishna uses the form with the meaning embryo, and it would seem that the ancient Jews interpreted as a cognate stem with . Calvin followed this opinion, but few others have adopted it since the Reformation.

(c) The is of the nature of a Massoretic interpretation, as is so frequently the case with the quiescent letters in the Hebrew text. The original consonantal text read. This is evident from the Arabic of Saadia of the tenth century, who did not follow the Massoretic pointing, but translated it as if it were pointed; that is, the relative, the preposition, and the suffix . Saadia is sustained by Aquila, who testifies to the official interpretation of the rabbins of the second Christian century. Symmachus and Theodotion give the same witness. Jerome read = one sent, qui mitten

שָׁלְח but he interpreted it as ,שלח or שלה

dus est.

(d) We may now go back of the official text of the second Christian century to the Maccabean text. The Targum of Onkelos and the Syriac version testify to, and translate: the Targum, "whose is the kingdom," the Syriac, "whose it is," which is explained by Aphraates and Ephraem as "whose is the kingdom."

(e) We may now go back to the text of Ezra. The ancient 1 Sanh., 98 b. See Driver, Journal of Philology, 1885, in an article on

.שילה

Greek version and the Samaritan codex both confirm w, and the former renders ἕως ἂν ἔλθῃ τὰ ἀποκείμενα αὐτῷ.

(f) We may also go a step still further backward under the guidance of an apparent citation in Ezek. 212, where the phrase

seems to be not only a reminiscence עד בא אשר לו המשפט

אשר לו interpretation

but an interpretation of Gen. 4910, and confirms

with the

Thus the genealogical principle establishes, beyond the shadow of a doubt, that the original reading of the passage was, and that the interpretation was either "that which belongs to him," or "whose it is."

For another example we may use Ps. 2210 (9), which is translated in our English Bible, "Thou didst make me trust (when I was) upon my mother's breasts." This is a correct translation of the Massoretic text (Hiphil participle). But in the time of Jerome the unpointed text was, for he takes it as the noun

, my trust. So do the Syriac and ancient Greek versions, leading us back to the Maccabean Psalter. But we may go further back still, for Ps. 22 is quoted in paraphrase in the later Ps. 715, where we have, the noun.

The genealogy of the Greek Bible is traced back in a similar way. Lagarde represented that in the case of the Septuagint it was necessary to ascertain the three great official texts of the third century, Lucian, Hesychius, and Eusebius. All the manuscripts should be classified so far as possible to show their descent from these. On the basis of these three one may work back to the common parent. Westcott and Hort have shown that we have two groups of texts that are older than these recensions; namely (1) the Western text, represented by D, the old Latin, the old Syrian, and sundry citations; and (2) the neutral text of B, §, going back to a common parent in the second century. The translations all come into evidence in showing the texts from which they were translated, and the Christian Fathers of the different centuries in the use of the versions and manuscripts from which they cited.1

An interesting example of the use of the genealogical principle in the New Testament is in 1 Peter 315. The Authorized Version

1 I think it unnecessary to give a classification of the Fathers for the purpose of showing the descent of citations. These are accessible easily to all students. I have given the Jewish Literature because it is not so accessible.

reads: "But sanctify the Lord God in your hearts." But this reading is found only in the uncials of the ninth century, K, L, P, and in no earlier writers than Theophylact and Ecumenius. The great uncials, B, and A, C, the Syriac, Sahidic, Coptic, and Armenian versions, all give Xporróv, Christ, in place of eor, God. The genealogical principle therefore determines, without doubt, the original reading, and so the Revised Version renders, "But sanctify in your hearts Christ as Lord." This evidence might be fortified by the usage of the New Testament. But no further evidence is needed.

The genealogical method does not always determine the origi nal reading; then we have to fall back on the internal evidence. As an example of the failure of the genealogical method I may cite the case of Acts 2028. I shall quote from myself:

"There is a great difference of opinion as to the reading here. The external authority of MSS., versions, and citations is not decisive. Tischendorf, De Wette, Meyer, and the mass of German critics read Church of the Lord'; Scrivener, Westcott, and Hort, and the leading British scholars read Church of God.' If any unprejudiced man will compare the great mass of authorities cited on both sides, he will be convinced that there is ample room for difference of opinion. The context favors Church of the Lord.' This reading is also favored by the fact that it is a unique reading, and therefore difficult. Nowhere else in the New Testament do we find the phrase 'Church of the Lord.' The scribe in doubt would follow the usual phrase. That the more difficult reading has survived is a proof of its originality. The reading 'Church of God' gives by implication blood of God.' This is found in Ignatius and other early writers, possibly on the basis of this passage, but it involves a conception which is alien to the New Testament. It is extremely improbable that Luke would put into the mouth of Paul such an unexampled and extraordinary expression under the circumstances. It involved a doctrine of startling consequences. Such a doctrine would not come into the language of Holy Scripture in such an incidental way. The American Revision, therefore, is to be followed in its reading 'Church of the Lord' rather than the A. V. or the British Revision 'Church of God.'"1

1 Briggs, The Messiah of the Apostles, 1895, p. 81. See Ezra Abbot, Critical Essays, pp. 294 seq.

R

VI. CONFLATION AND OTHER CORRUPTIONS

It is characteristic of the late Syrian texts, and in a large measure also of Lucian's text of the Old Testament, that they indulge in a considerable amount of conflation. Underlying conflation is the feeling that, as far as possible, all of the original text should be preserved; and that, in cases of doubt, it is better to preserve all than to run the risk of losing anything. Conflation is indeed found in the earliest texts both of the Old Testament and the New Testament, and must have taken place to a considerable extent back of any versions known to us. Conflation arises partly from the comparison of earlier authorities, and partly from the insertion of ancient marginal explanations, or glosses. A very good example of conflation is given. in Westcott and Hort.

"Mk. 949.

“ (α) πᾶς γὰρ πυρὶ ἁλισθήσεται (*) Β Δ 1-118–209 61 81 435 al me. codd. the arm. codd.

"(B) Tâσa yàp Ovσía áλì ádobýσeraι D cu2 (a) b cff-i (k) tol holm gig (a c tol holm gig omit al: a omits yáp: k has words apparently implying the Greek original πᾶσα δὲ (or γὰρ) οὐσία ἀναλωθήσεται, ο being read for θ, and a Na for A/C).

“ (δ) πᾶς γὰρ πυρὶ ἁλισθήσεται, καὶ πᾶσα θυσία ἁλὶ ἁλισθήσεται, ACNXEFGHKMSUVгπ cu. omn. exc. 15 fq vg syr. vg hl me. codd æth arm. codd go Vict (culo vg. codd. opt omit dλ; X adds it after πυρί).

"A reminiscence of Lev. vii. 13 (κaì πâν Swрov Ovσías vμôv åλì ἁλισθήσεται) has created β out of a, ΠΥΡΙ ICO being read as OYCCO with a natural reduplication, lost again in some Latin copies. The change would be aided by the words that follow here, kaλòv Tò åλas K.T.λA. In & the two incongruous alternatives are simply added together, yáp being replaced by Kaí. Besides. ACNX, 8 has at least the Vulgate Syriac, and the Italian and Vulgate Latin, as well as later versions."

Here we see the original in the neutral text, a variation by a mistake in the Western text, and then a full conflation in the Syrian texts.

An interesting example of corruption of an original text is presented in Ps. 25. This Psalm is an alphabetical hexameter. All

1 Westcott and Hort's New Testament in Greek, 1882, pp. 101, 102.