XIII.

ON THE READING "AN ONLY-BEGOTTEN GOD," OR "GOD ONLY-BEGOTTEN," JOHN I. 18.

[From the Unitarian Review and Religious Magazine for June, 1875.]

Θεὸν οὐδεὶς ἑώρακεν πώποτε· ὁ μονογενής υἱός [var. reading μονογενής θεός], ὁ ὢν εἰς τὸν κόλπον τοῦ πατρός, ἐκεῖνος ἐξηγήσατο.

As the writer of the present article has already twice discussed the reading of this passage,— first in the Appendix to Norton's Statement of Reasons, etc. (2d ed., 1856), pp. 448469, and afterwards in the Bib. Sacr. for Oct., 1861, p. 840 sqq. [see Essay XII.],—an apology may be needed for returning to the subject. The question, however, has acquired a new interest in connection with the revision of the common English version of the Bible which is now in progress in England and this country. It is well known that two of the most eminent among the scholars of the British Committee engaged in this work, Dr. Westcott and Mr. Hort, have adopted the reading "God" in the text of their (as yet unpublished) critical edition of the Greek Testament. Another of the British revisers, Professor Milligan, has accepted it as the true reading, in Milligan and Roberts's The Words of the New Testament, etc. (1873), p. 162 ff.; and Professor Lightfoot, in his valuable work On a Fresh Revision of the English New Testament (2d ed., 1872), p. 27, remarks that "the "Onlybegotten God' would seem to have equal or superior claims to 'the Only-begotten Son' in John i. 18, and must either supersede it or claim a place side by side with it." Dr. Tregelles receives it into the text of his important edition of the Greek Testament (Part II., 1861), and had previously defended its genuineness in his Account of the Printed Text of the Greek New Testament (1854), p. 234 f.; Lachmann placed it in the margin of his critical editions (1831 and 1842) as an

alternate reading, and would undoubtedly have taken it into the text, had he known all the authorities by which it is supported.

It must not be supposed, however, that there is a general agreement of scholars in favor of this reading. Tischendorf, though he had adopted it in the second edition of his Synopsis Evangelica (1864), has restored the reading "Son" to the text in his eighth critical edition of the Greek Testament (1869); Alford retains viós (6th ed., 1868), though giving ɛóc a place in his margin; Dr. Scrivener, also a member of the British Biblical Revision Committee, defends the reading "Son" in his Introduction to the Criticism of the New Testament (2d ed., 1874), p. 525 f.;* and Bishop Wordsworth does. not even notice the reading eóc in his edition of the Greek Testament (5th ed., 1866). The reading "Son" is also defended by Rev. T. S. Green, M.A., in his Course of Developed Criticism, etc. (1856), p. 73, and Critical Appendix to the Twofold New Testament (187-), p. 33; by Dr. Samuel Davidson, art. "Manuscripts, Biblical," in Kitto's Cycl. of Bibl. Lit. (3d ed., 1870), iii. 60; by Professor James Drummond, of Manchester New College, in an able article in the Theological Review for October, 1871, pp. 468–495; and by Rev. J. B. McClellan, M.A., in his recent learned and elaborate work, The New Testament... a New Translation... from a critically revised Greek Text, etc. (Lond. 1875), vol. i. p. 707 f. Among scholars of the present century on the continent of Europe, I know of none who have adopted the reading tóc.† It is emphatically rejected as a dogmatic gloss by Godet (1864) and Meyer (1869) in their recent Commentaries on the Gospel of John; and it is also rejected or ignored entirely by Olshausen (1838), Lücke (1840), Tholuck (1857), Ewald (1861), Brückner and De Wette (1863), Bäumlein (1863), Hengstenberg (1867), Lange (1868), and so by Dr. Schaff in his American translation; see his note. It is also ignored in the most important recent translations which professedly represent a critically revised Greek text; as that of Holtz

*[So, too, in the 3d ed. (1883); see especially p. 606, note.]

[Now adopted by Harnack, Weiss, al.]

mann in Bunsen's Bibelwerk, vol. iv. (1864), the American Bible Union (2d revis., 1867), the new authorized Dutch translation by Van Hengel and others (Amst. 1868), the French version of Oltramare (Genève, 1872), and the German translation by Weizsäcker (1875). The French translation of Rilliet (Genève, 1860) is not an exception, as that only represents the Vatican MS.

The question, then, is evidently an open one; and the object of the present article is to state and weigh, as fairly as possible, the evidence for the rival readings. It may be proper to mention that the substance of the paper was prepared at the request of the New Testament Company of the American Biblical Revision Committee, though no one but the writer is responsible for any statement or argument which it may contain. It is hoped that the account of the evidence will be found somewhat fuller and more accurate than has elsewhere been given; but, to avoid unnecessary repetition of what has already been published, I shall often refer, for details, to the articles in the Bibliotheca Sacra and the Theological Review, which are mentioned above. Alford's note also gives very fully the context of some of the passages cited from the Christian Fathers. In adducing authorities not noticed in Tischendorf's last critical edition (1869), references are given; but it is assumed that one who is specially interested in the investigation of the question will have that edition at hand.

The evidence, then, for the different readings is as follows:

I. MANUSCRIPTS.

ὁ μ. θ.). *

For μovoyeviç bɛós, §, B, C*, L, 33 (xo, 33,

For pov. viós, A, X, г, A, A, II, C3, E, F, G, H, K, M, S, U, V,

* In the Bibliotheca Sacra for October, 1861 (p. 850), Ɖɛóç was given as the reading of a prima manu, as if it had been afterwards corrected. The text of the MS. was not then published, and I was misled by Tischendorf, who, in his Notitia Cod. Sinaitici (1860), p. 18, gave the reading as follows: "Joh. 1, 18 a prima, novo, svo (absque ó cum BC*L) fɛoơ (cum_BC*L etc.) ɛra (om o wr)." I naturally supposed the "a prima" to refer to all the variations from the Received Text, not merely to the first and the last. Dr. Tregelles before me had fallen into the same error (Text. Crit., 2d ed., p. 780).

the cursives 1, 22, 28, 118, 157, 209, all of which are of exceptional importance and value, also Professor Ferrar's group, 13, 69, 124, 346, which he regards as representing an early uncial akin to D, but with a purer text;* and all other known cursives, several hundred in number, but the majority of them not carefully examined, or of little worth. D, unfortunately, is mutilated here; but the versions and Fathers with which it usually agrees support viós.

Of these MSS., and B are assigned to the middle of the fourth century, A and C to the middle of the fifth, E and L are of the eighth century, the other uncials of the ninth or tenth. Among the later uncials, L, X, г. A, A, II, are distinguished from the rest, L pre-eminently, by their more frequent agreement with the oldest authorities. The cursive MS. 33 is of remarkable excellence.

The MS. authority for eos is weighty, though confined to the representatives of an Alexandrian or Egyptian text. In a large majority of cases, the reading supported by these MSS. against the rest is confirmed by other ancient evidence and by intrinsic probability, and has a good claim to be adopted. On the other hand, they all, or the most of them, sometimes concur in readings which are clearly false, or exceedingly improbable, or very doubtful. See, for example, *, B, C, L, U, г, Matt. xxvii. 49; *, B, C*, 4, Cop., Hcl. Syr.marg. Aeth., Mark iii. 14; §, B, D, L, 4, Mark vi. 22; B*, C, L, V, Luke i. 17; 8, B, D, U, X, 33, Luke xv. 21 ; Noa, B*, C*, John i. 15%; §, B, L, T, 33, John iii. 13; B, L, T, X, г, 4, etc., John vii. 8; *, B, H, L, P, 61, Acts xii. 25 (impossible). See also for &, B, in particular, Matt. vi. 8; xvi. 21; Mark iv. 21 (inó for iñi, impossible); Luke xxiii. 32; Acts xvi. 32; James i. 17; 2 Peter ii. 13. As to the cursives, those first named above are nearly, some of them perhaps quite, equal in value to 33 in the Gospels; 1, 22, and 209, especially, are often right where it is wrong. See, e.g., Matt. v. 44; vi. 1, 6, 13; xix. 16, 17; Luke xi. 2, 4, and numberless other passages.

* See Scrivener's Introduction to the Criticism of the New Testament, p. 167, and Addenda to do., p. ix. (2d ed., 1874); [see 3d ed. (1883), p. 181].

Such being the state of the case, the MS. authority for . Bóc, though important, cannot be regarded as in itself decisive.

II. ANCIENT VERSIONS.- For fós, the Coptic or Memphitic (3d cent., or perhaps even the 2d), Peshito Syriac (in its present form, 4th cent.; so Tregelles, Westcott and Hort, Crowfoot, Payne Smith, Lightfoot), Harclean Syriac in the margin (A.D. 616), Aethiopic (4th or 5th cent.) in the Roman edition.

For viós, Old Latin (2d cent.), a, b, c, e, f, ff2, 1, filius, q fil. dei; Vulgate (A.D. 384), Curetonian Syriac (2d cent.), Jerusalem Syriac (5th cent. ?), Harclean Syriac in the text (here probably Philoxenian, A.D. 508), Aethiopic in Platt's edition, which is the best; Armenian (cir. A.D. 431).

Though Wilkins and Malan (Gospel of St. John) in their translations of the Coptic give its reading as "the onlybegotten of God," and are followed by Scrivener and McClellan, this is doubtless an error; see Schwartze's note in loc.

III. FATHERS. (In citing their names, the year when they flourished is noted, generally as assigned by Cave.)

For fós. Clem. Alex.192, once, but once in reference viòç fɛóç ; Excerpta ex Theodoto (Valentinians), 2d cent. (?); Epiphan.38 three times, and one ref.; Didym. Alex.370 twice, and one ref. (?); Cyr. Alex.412 four times, and one ref., but viós three times (Opp. iv. 103o, v. i. 365o, vi. i. 90o), also (allus.) ¿ pov, rov Bɛoč žájos (VI. i. 187). Perhaps 2d Synod of Ancyra358 one ref., and Greg. Nyss.379 one ref. and eight allusions (Opp. iii. 291a, and in addition to what Tisch. cites, ii. 432", 478, 506, 595 [605], 691), but viós twice in similar allusions (Opp. ii. 466, iii. 648"), and iv pirog róg once (i. 697"). The inconclusiveness of such references and allusions is illustrated in Bib. Sac. (as above), pp. 855-857; see also below, p. 280.

No quotation of the passage with the reading deus has been produced from any Latin Father.*

*The apparent exception in the case of Hilary (De Trin. xii. 24, Migne x. 448a) is merely