No man hath seen God at any time: The only begotten Son, which is in the bosom of the Father, he hath declared him.
(John 1.18)

Many modern versions of the Holy Scriptures are offered to the reader with the assurances that they embody the best fruits of the best scholarly research, that they are based on the best manuscripts, and that they present in the best contemporary English the precise meaning of the underlying Hebrew or Greek text. The treatment of ‘the only begotten Son’ by the translators of these versions shows very clearly that there is a great amount of disagreement among modern scholars regarding the authenticity and the precise meaning of the words. The reader is confronted with the problem of selecting from no less than forty variants the one which be must adopt in preference to the Authorised Version. Faced with such an array of conflicting opinions, all claiming to be scholarly and precise, he may feel that the wisest and most prudent course would be to retain the reading which has been most widely known among Christian people during the last nineteen hundred years, which is accurately and intelligibly rendered in the version quoted above, and is supported by the majority of the manuscripts.

Variations in the English Versions

In his article on Textual Criticism in the June 1972, issue of the Banner of Truth, the Rev. Donald MacLeod wrote, ‘The text which underlies many modern versions, however, is ‘The only-begotten God ...’. This is quite incorrect, as the Greek variant to which he refers does not contain the article ‘the’, and among the 40 variants offered in about 100 English versions very few correspond exactly with Mr MacLeod’s quotation, including Jehovah’s Witnesses’ New World Translation, the Concordant Version of 1931, and the margins of Alford and Weymouth. F. A. Spencer’s American R. C. version of 1951 has ‘the Only Begotten God’, and the New American Standard Version ‘the only begotten God’.

Variations in the manuscripts

Among the thousands of documents containing this verse there are five variants:

1) HO MONOGENES HUIOS — The only-begotten Son.

2) MONOGENES THEOS — God only-begotten.

3) HO MONOGENES THEOS — The only-begotten God.

4) HO MONOGENES — The only-begotten.

5) MONOGENES HUIOS THEOU — God’s only-begotten Son.

For the first two variants there is a great amount of very ancient documentary evidence, and the reading followed by the Authorised Version is supported by most of the Greek copies, ancient translations, and quotations in the works of early Christian writers-the ‘fathers’.

Misapprehension of the Facts

The third variant ‘the only-begotten God’ quoted by Mr MacLeod has negligible manuscript evidence to support it-papyrus 75 of the 3rd century, the third corrector of Codex Sinaiticus, and a single cursive, No. 33 of the 11th century. Westcott described the first two variants as ‘of equal antiquity, as far as our present authorities go’, and referring to the third variant he remarked in his commentary on John’s Gospel, ‘The strangely inaccurate statement of many commentaries that HO MONOGENES THEOS (the only-be-gotten God) is the reading of Sinaiticus, Vaticanus, Ephraemi, L, etc., shows a complete misapprehension of the facts and of the significance of the readings …’

The English and American Revisions

Westcott himself preferred the second variant MONOGENES THEOS, God only-begotten-and his preference very nearly found its way into the text of the English Revised Version of 1881. When the draft was submitted to the American Revision Committee under the chairmanship of Dr Phillip Schaff they strongly recommended that ‘Son’ should be retained in the text and that ‘God’ should be in the margin, and the English Committee adopted this course. Subsequently the American Revised Standard Version followed suit. It is regrettable that the New American Standard Version reversed this decision and introduced into the text ‘the only-begotten God’, thus perpetuating ‘a complete misapprehension of the facts and of the significance of the readings’.

The New English Bible

In the treatment of this particular text the edition of the Greek text underlying the New English Bible is more reliable than the English. A note on page 424 of the N.E.B. Greek Text edited by Professor Tasker explains why the translators here returned to the text followed by the A.V. In their opinion the second variant must be rejected because ‘it does not yield a tolerable sense’, and the first variant must be upheld ‘as intrinsically more probable’. In view of this admission it is strange that the N.E.B. English does not translate ‘the only-begotten Son’ as in the A.V., but actually follows the fifth variant-MONOGENES HUIOS THEOU-God's only-begotten Son-expressed in the N.E.B. as ‘God's only Son’. For this rendering including both "God" and ‘Son’ the documentary evidence is no more than the old Latin q of the 7th century, a Latin translation from the Greek of Irenaeus and a single quotation by Origen, AD. 254. The list of documents opposed to this reading would extend to several pages.

‘One Latin copy’

Referring to the 4th variant, Westcott commented, ‘The tempting reading of one Latin copy ‘the only-begotten’ has still less claim to be taken into account in the face of the facts of the case’.

The N.E.B. translators examined this ‘tempting reading’, and having dismissed it from their text, added the surprising statement ‘It was thought possible that the second alternative reading quoted, HO MONOGENES, though it is represented only in a single manuscript of the Latin Vulgate, might have given rise to the other two.’ This note suggests that the original text was ‘The only one’ or ‘The only-begotten’, and that this was expanded in one group of manuscripts by the addition of ‘Son’, and in another by the addition of ‘God’. This theory is advanced on the evidence of one solitary manuscript of the Latin Vulgate. One is entitled to ask how would the same panel of scholars react to the suggestion that the disputed words in 1 John 5.7, 8, which are found in a very large number of manuscripts of the Latin Vulgate, actually formed a part of the original text and gave rise to the later variants now reflected in the modem versions?

The 40 English Variants

Authorities Supporting the A.V. Rendering

Critics of the Authorised Version often give the impression that its underlying text depends only upon the few documents available to Erasmus, Stunica, Stephens and Beza in the 16th century, but this is a serious misrepresentation of the facts. The majority of the documents recovered since that time preserve the same form of the text as that which underlies the Authorised Version. The evidence may be divided into six groups.

I. The Uncials

Greek manuscripts in large characters. The majority of the uncials have HUIOS-Son in John 1. 18, including those known by the symbols W A C³ Θ K X Δ Π Ψ. These documents span the period from the 4th century to the 10th.

ll. The Cursives

Greek manuscripts in small characters written with a ‘running’ hand. Many of these are later than the uncials, but were copied from uncials written earlier than any of the surviving uncials. Scrivener referred to such cursives as representing ‘respectable ancestors known only through their descendants’. Most of the cursives preserve the Greek text in the form represented by the documents available to the A.V. translators. They number many hundreds and with only one exception they support the A.V. rendering of John 1. 18 ‘Son’. The list begins with ‘family 1’ (1, 118, 131, 209 etc.) and continues with ‘family 13’ (13, 69, 124, 174, 230, 346, 543, 788 etc.).

lll. Cursives

Cursives sometimes stand apart from the mainstream. Most of the cursives preserve the text sometimes called ‘Byzantine’ because of its prevalence in the Byzantine period, but there is a group of cursives which sometimes present readings apparently independent of the mass of Byzantine documents. With regard to John 1. 18 this ‘independent’ group stands with the main body of Byzantine manuscripts in reading ‘Son’ as in the A.V. The list includes the manuscripts numbered 28, 565, 700, 892, 1009, 1010, 1071, 1079, 1195, 1216, 1230, 1241, 1242, 1253, 1344, 1365, 1546, 1646, 2148.

IV. The Ancient Versions

As early as the 2nd century the Greek New Testament was translated into Syriac and Latin and in the next three or four hundred years into Coptic, Armenian, Georgian and Ethiopic. Surviving copies of these ancient versions represent the Greek text as it stood in the still more ancient copies used by their translators. For instance, the old Latin copies known by the symbols, a, b, c, e, f, ff², l, aur were made between the 4th and 8th centuries, but represent the translation work undertaken in the 2nd century based on Greek copies no longer in existence. All of these Old Latin copies support the reading ‘Son’ in John 1.18 in the A.V., and the other ancient versions referred to above preserve the same reading.

V. The ‘Fathers’

It has been said that the entire text of the New Testament could be recovered from quotations in the works of early Christian writers. If a writer in the second century comments upon a verse which later writers were inclined to omit or radically alter, it may be assumed that the ancient commentator had in his possession a Greek copy containing the disputed verse. The second century copy has not survived, but the commentary demonstrates that it was familiar to the writer. In John I. 18 ‘Son’ is attested by the following ancient authorities: Irenaeus, Hippolytus, Clement of Alexandria, Tertullian, Synodical Epistle of the Council of Antioch, Archelaus, Eustathius, Alexander of Alexandria, Eusebius of Caesarea, Phoebadius, Victorious of Rome, Hilary, Ambrosiaster, Athanasius, Ambrose of Milan, Titus Bostra. Basil, Gregory Nazianzen, Augustine of Hippo, Chrysostom, Synesius, Theodoret, Theodore of Mopsuestia, Nonnus, Proclus, Vigilius. These writings span the period from the second century to the end of the fifth century.

Less evidence for the alternative reading ‘God’ The variant MONOGENES THEOS, God adopted by some of the modern versions is not so well attested, the authorities being papyrus 66, the uncials Aleph, B, C*, L, some of the Coptic, Syriac and Ethiopic versions, and some quotations by Theodotus, Va1en-tinians, Ptolemy, Diatessaron, Heracleon, Irenaeus, Clement of Alexandria, Origen, Hilary, Basil, Apostolic Constitutions, Didymus, Gregory Nyssa, Jerome, Cyril of Alexandria, Epiphanius, Arius, and Synesius. Papyrus 66 introduces a sixth sub-division of the ancient evidence, the papyrus fragments surviving from the 2nd-7th centuries.

The antiquity of some of these fragments is no guarantee of their reliability. Most of them were made and preserved in Egypt and exhibit the same deficiencies as those which characterise later copies originating and circulating in that part of the world. In many instances the variants of Papyrus 66 are rejected by modern editors and translators, but are regarded as respectable if they agree with Codex Vaticanus. It is a mistake to suggest that ‘only-begotten God’ renders more honour to Christ than ‘only-begotten Son’, for the innovation can be used to support the dangerous error that Christ’s Godhead was inferior to that of the Father, that His Godhead was not eternal and essential, but subordinate and derived. It was no doubt for this reason that Arius adopted the incorrect reading because it was more favourable to his heretical view of the Person of the Son of God. For the same reason Jehovah’s Witnesses approve this reading. The true reading relates ‘only-begotten’ to His Sonship, not to His Deity, and is truly and accurately rendered in the Authorised Version, with the support of the overwhelming mass of the ancient documentary evidence.

This article was first published in 1972 in Quarterly Record 441. Edited for online publication in January 2023.