Article No. 69


This book is written with the intention of disproving the arguments which have been put forward in favour of the King James Version (KJV) and its underlying Greek text. It is designed for those who "find themselves influenced by the writings of the Trinitarian Bible Society and parallel groups, and do not know where to turn to find a popular rebuttal" (Introduction, p.10).

The Debate

The publishing of this book is an interesting development in the long controversy over the KJV, which began in earnest with the production of the Revised Version (RV) in the nineteenth century. As is well known, most scholars today adopt the form of Greek text underlying the RV and its successors in preference to the text underlying the KJV, from which it differs in more than five thousand places. In opposition to this 'scholarly' preference, however, there has arisen a sizable body of opinion favouring the traditional Greek text. Similarly while many people may prefer to read the Bible in 'twentieth-century English' there are others who still have a high regard for the quality of the KJV as a translation. Those who support the KJV and its underlying text have in recent years mounted a vigorous protest, in books and pamphlets. Dr. D. A. Carson is the first to write a book in direct reply, and is for that reason worthy of attention.


Most of the book (pp. 15-78, 105-123) is taken up with the question of the text of the New Testament, to which the present review is largely confined. The first two chapters (pp. 15-24) briefly explain how scribal errors arose during the manuscript transmission of the text, causing divergence among the manuscripts which survive. This introduces the central problem of textual criticism: how to determine the wording of the inspired original text when the surviving manuscripts disagree. The third chapter (pp. 25-27) explains the conventional classification of manuscripts into four 'text-types'. In the fourth chapter (pp. 29-31) are sketched the commonly accepted criteria for deciding between rival variant readings. Several popular fallacies occur here.

'Genealogical relationship'

Among external criteria, 'genealogical relationship' is said to be more important than the number of manuscripts (pp. 29-30; cf. also pp. 109-110, 120). In this way it is usually implied that the vast majority of manuscripts (the traditional text underlying the KJV) are all descended from a single ancestor, the deceased parent of the 'Byzantine text-type'. The author accordingly states "If the other three types agree on another reading, even though they collectively embrace a smaller number of actual manuscripts, then it is in principle more likely that they preserve the original reading, other things being equal". This statement depends on the assumption that the majority of manuscripts belong to a single text-type, called 'Byzantine', while the small minority of manuscripts between them form at least three text-types.

Under the current system of classification a separate text-type is assigned to any group of manuscripts which agree on a certain number of readings, no matter how corrupt they happen to be, and no matter how few the manuscripts. On the same principle a large group of manuscripts which are broadly similar to one another because they all faithfully preserve the original text would have to be assigned to just one text-type, even though they heavily outnumber the manuscripts from all other text-types put together. If the author's expedient of counting up the text-types is used the result is obviously heavily biassed in favour of the combined testimony of small groups of corrupt manuscripts and against a single large group of relatively pure manuscripts.

The current text-type classification stands in need of drastic revision to take full account of the numerous independent lines of transmission among the Byzantine manuscripts, stretching back to the earliest period. The best method is to assume that each manuscript is relatively independent and innocent of collusion unless there is definite proof of family relationship: innocent until proven guilty. It is likely that there are hundreds of independent lines of transmission within the Byzantine tradition, in which no two manuscripts are identical. If the author's principle of counting up the text-types were applied to these numerous relatively independent witnesses, the Byzantine manuscripts would invariably be found to preserve the original reading. Unfortunately scholars have been so prejudiced against the Byzantine manuscripts that they have not troubled to devote to them the painstaking research which these manuscripts deserve.

It is futile to object that since all manuscripts are descendants from the inspired original it therefore follows that genealogical relationships must exist (cf. p. 120), for in this sense not merely the Byzantine manuscripts but also all text-types are genealogically related to one another; this weak form of relationship is useless as a tool of textual criticism. Furthermore the reason for the broad similarity among the Byzantine manuscripts could well be that they are all relatively pure descendants of the original text rather than being the offspring of some complex process of corruption and collusion. It would not be at all surprising to find that the original text was faithfully perpetuated among the majority of manuscripts. It would on the other hand be very surprising to find that only a small minority of manuscripts tend to preserve the original, since this would imply that scribes were more likely to use a corrupt exemplar than a trustworthy exemplar. That such a surprising process occurred must first be satisfactorily proved by those scholars who advocate the numerous minority readings introduced into the modern critical text. And it will not be sufficient for them to show that this process could have occurred; they must show that it did occur or was very likely to occur. Until such proof is supplied these minority readings must be firmly rejected by all who desire the text of the Bible to rest on a sure foundation.

'Internal criteria'

1. "In general, the shorter reading is to be preferred, since it is demonstrable that later scribes, at least, tended to add bits rather than remove them" (p. 30). This principle favours the shorter form of text found in the Alexandrian text-type, as opposed to the longer form in the Byzantine text-type. Whatever the faults of later scribes, it is certain that the earlier scribes of Egypt, where the Alexandrian text-type grew up, had a tendency towards omission. This is evident from a study of the papyri by Colwell (undiscussed by Dr. Carson). In his chapter on 'scribal habits' in his "Studies in Methodology in Textual Criticism" (1969) Dr. E. C. Colwell shows that the scribes of Papyri 45, 66, 75 are all prone to losing their place in the text and consequently skip over letters, syllables, words or phrases. This proven tendency of Egyptian scribes throws grave suspicion upon the quality of the shorter Alexandrian text favoured by most critics.

2. "In general the most difficult reading ... is to be preferred, since a scribe is more likely to emend a difficult reading than an easy one" (p. 30). This criterion falls prey to a simple reductio ad absurdum: whatever obscure or nonsensical wording a solitary scribe happens to invent, we would be obliged to accept it because it would be a more difficult reading than another reading which makes sense and is found in all the other manuscripts. Further, it ignores the probability that the writers of the inspired originals would have wished to make their meaning clear to the reader, and would often have deliberately avoided using a 'more difficult reading'.

3. "One must also take into account what the original human author was more likely to have written ... sometimes it is possible to make an educated guess from among three or four variants on the basis of what the author has done elsewhere" (p. 30). This is an extremely dangerous procedure, which is again capable of a reductio ad absurdum: wherever a New Testament writer happens to have used words and stylistic constructions uniquely, we would be obliged to reject the testimony of nearly all the manuscripts and accept the reading of a solitary scribe who has altered the passage to conform with the author's usual practice. This kind of textual criticism is a compound of guesswork and subjectivity, leading to a frivolous treatment of the words of Scripture.

The fourth chapter thus only serves to show how inferior is the internal evidence compared with the external. It is the neglect of the external evidence of manuscripts which has led to the modem form of the Greek text.

The Textus Receptus

The fifth chapter (pp. 33-37) handles the origins of the Received Text ('textus receptus' or TR) of the Greek New Testament. There is the usual criticism of Erasmus' first edition on the grounds of its "countless hundreds of printing errors", its narrow manuscript base, and its additional readings derived from the Vulgate. The implication is that the KJV of 1611 (which was based on the TR) was guilty of all the blemishes found in Erasmus' edition of 1516. It should be stressed that while numerous printing errors were made in his first edition, Erasmus himself subsequently corrected many of these, and in the important edition of Stephanus in 1550 there are hardly any printing errors to be found. None of these errors has any effect upon the KJV. As for its manuscript basis, it is evident that the TR stands close to the text found in the majority of all manuscripts, and the number of passages derived from the Vulgate are very few. The TR was certainly not perfect but it by no means deserves the scorn and obloquy that have been heaped upon it during the past century.

Fourteen Theses

In chapter six (pp. 39-42) miscellaneous arguments in favour of the 'Byzantine text-type' (the traditional text) are listed. From the manner in which the arguments are there presented, the reader may well receive the mistaken impression that they do not represent a coherent case. In chapter seven (pp. 43-78), the main chapter of the book, the author offers fourteen theses in which he deals with such arguments. Before we examine the theses in detail, it should be noted that only the first four are of central importance. The positions criticised in theses five to fourteen are not crucial to the debate over the text, but Dr. Carson has identified some fallacies which should certainly be abandoned by their advocates. It is undesirable for supporters of the traditional text to spoil an excellent case by putting forward supplementary arguments deficient in logic and evidence.

Byzantine versus Alexandrian

Thesis 1: "There is no unambiguous evidence that the Byzantine text-type was known before the middle of the fourth century" (p. 44).

It is true that there is no surviving pre-4th century manuscript which contains a high proportion of the readings peculiar to the Byzantine 'text-type'. But there is no reason to expect such manuscripts to be forthcoming. It should be remembered that the survival of any manuscript from that early period is a freak of history and climate. The pre-350 AD manuscripts which do survive are relatively few in number and extremely fragmentary; it would not be possible to construct a complete New Testament from among them. The origin of these manuscripts is almost exclusively Egyptian, and it would be unsafe to rely upon evidence confined within such narrow geographical limits. It is likely that such manuscripts were the basis of the Alexandrian text-type, which is associated with the same geographical region. As has been mentioned already, when the early papyri are closely studied proved to be prone to omitting portions of the text through scribal carelessness (or editorial activity). Hence it is quite unreasonable to expect to find unambiguous pre4th century manuscript evidence for the Byzantine 'text-type'.

Regarding the evidence of the few early versions, it is likewise true that none of them fully supports the Byzantine 'text-type'. The versions were made for the benefit of those who did not speak Greek and lived at a considerable distance from the areas where the original Greek autographs were probably kept. Translators who worked in these distant regions were less likely to be in possession of good exemplars than scribes who lived in the region close to the autograph. In fact while they do not consistently reflect any of the main 'text-types', the early versions which stand closest to the Alexandrian text-type (namely the Coptic and Sinaitic Syriac) turn out to have a strong geographical association with Egypt, and are subject to the same Greek manuscripts from this region. Hence we need not expect early versions to belong to the Byzantine group.

Like the early versions the early fathers do not consistently follow any of the main 'text-types', but they do sometimes cite exclusively Byzantine readings. This is eventually admitted by the author on p.111, contradicting the emphatic denial he makes on p. 47. From the studies of Burgon and Miller it appears that in the works of the early fathers the Byzantine text is not cited just occasionally but is predominant, compared with the text of Westcott and Hort (the text which has exercised enormous influence on twentieth century critical editions). The author wishes to avoid this conclusion by demanding that the citations should be shown 'distinctively' Byzantine: "the vast majority of so-called Byzantine readings in the ante-Nicene fathers are also Western or Alexandrian readings. They become distinctively Byzantine only by their conflation in individual manuscripts after the fourth century has got under way" (p. 110). Such a method is clearly biassed against manuscripts which preserve a longer text and in favour of manuscripts which present a shorter text, since the longer reading will be said to be 'conflation' of different kinds of shorter reading and hence not 'distinctive'. This is a built-in bias against the Byzantine manuscripts, which preserve a longer text than the Alexandrian manuscripts.

On p. 47 (and p. 108) the author pours scorn on the explanation that the early Byzantine manuscripts wore out or were destroyed. But he must make use of a similar explanation himself, if he is to explain how the Alexandrian text-type could have been the dominant text in the first three centuries and yet left so few fragments behind.

All these considerations show that the first thesis is making unreasonable demands.

Thesis 2: "The argument that defends the Byzantine tradition by appealing to the fact that most extant manuscripts of the Greek New Testament attest to this Byzantine text-type, is logically fallacious and historically naive" (p. 48).

The logic of that argument has already been demonstrated. What the author must show is that scribes were historically more likely to corrupt the text than to preserve it, or were more likely to use a corrupt exemplar than a trustworthy exemplar, at any given passage. The vast numerical superiority of the Byzantine manuscripts is being explained away as due to the influence of Chrysostom and 'monarchical bishops' (pp. 113-115). But there is no evidence that Chrysostom imposed a particular form of text; on the contrary (as the author concedes in a footnote on p. 114) Chrysostom himself often did not use the majority Byzantine text. The existence of significant variants within the Byzantine tradition shows that Byzantine scribes were not seeking to conform with an authorised standard. The Byzantine empire was large enough to allow for hundreds of independent lines of manuscript transmission. The concept of an imposed uniformity is utterly groundless.

In the Appendix(pp. 115-116) the author suggests that a "historical accident" analogous to the Genesis flood took place and overthrew the dominant Alexandrian text in favour of the Byzantine, just as all the descendants of Adam were drowned except the family of Noah. On the basis of this he accuses Pickering of historical naivety...To substantiate this charge, he must provide evidence that a "historical accident" occurred which was capable of destroying most of the good manuscripts in the civilised world and allowing a corrupt text to go forth and multiply. This evidence simply does not exist.

Thesis 3: "The Byzantine text-type is demonstrably a secondary text" (p. 51).

The justification offered for this assertion is that the Byzantine text contains frequent 'harmonization' (the removal of an apparently difficult reading by altering it to conform with a parallel passage). Like most internal criteria of textual criticism this is entirely subjective in application, and falls under the same objection as the principle of adopting the more difficult reading, as discussed under 'Internal criteria'.

Thesis 4: '"The Alexandrian text-type has better credentials than any other text-type now available" (p. 52).

In the light of all the foregoing observations, this claim should be considered already disproved.

Miscellaneous secondary arguments


Thesis 5: "The argument to the effect that what the majority of believers in the history of the church have believed is true, is ambiguous at best and theologically dangerous at worst; and as applied to textual criticism, the argument proves nothing very helpful anyway" (PĚ 54).

The author does not say who has applied such an argument to textual criticism. The argument would certainly be incapable of proving that the majority Byzantine text should be adopted, since it cannot be shown that the majority of past believers ever held specific views about text-types. If anyone really holds this view, Dr. Carson is quite right to object.

Thesis 6: '"The argument that defends the Byzantine text by appealing to the providence of God is logically and theologically fallacious" (p.55). It is convenient to take this together with Thesis 12:'The argument that ties the adoption of the TR to verbal inspiration is logically and theologically fallacious" (p. 68).

The author is right to urge caution in the interpretation of the doctrines of divine providence and verbal inspiration. When Christ promised that His words shall not pass away and that not one jot or tittle shall pass from the law, He did not promise that no scribes and editors would ever corrupt the true text of the Scriptures, or that the true text would be restored to perfection in the Textus Receptus, or that the true text would always be preserved in the majority of manuscripts. His promise would not be inconsistent with the original text being preserved in just one manuscript, or even being temporarily lost like the book of the law in the days of Josiah. The doctrine of the verbal inspiration of the original autographs of Scripture does not lay God under an obligation to make available perfect copies of the original text for the use of every generation. It is theologically justifiable to ascribe to the providence of God the preservation of all New Testament manuscripts, but the fact of providential preservation does not prove anything about the quality of one manuscript or another. In the absence of a new revelation as to which are the best manuscripts, the text we now adopt should be soundly based upon sciencific textual criticism. There is good reason to believe that the Textus Receptus and the versions based upon it are on the whole a faithful representation of

the inspired words of God. But we should not neglect the duty of supporting this belief by the application of scientific principles to the available evidence.

Thesis 7: "The argument that appeals to fourth-century writing practices to deny the possibility that the Byzantine text is a conflation, is fallacious" (p. 57).

The author is here concerned solely with a secondary argument put forward by Dr. E. F. Hills. Even if the argument is inconclusive, this does not mean that there is any justification for regarding the Byzantine text as a 'conflation'. No convincing examples of such conflation have ever been produced.

Thesis 8: "Textual arguments that depend on adopting the TR and comparing other text-types with it are guilty, methodologically speaking of begging the issue; and in any case they present less than the whole truth (pp.57-J8).

There is nothing wrong in principle with using the Textus Receptus (or the majority text) as a standard of comparison. When it has been established that the TR is generally based on the most reliable manuscript tradition, then it is justifiable to compare modern versions with the TR and conclude that these versions are relatively of the TR with the manuscripts is understood. (However, the author's criticisms of Gill's commentary on 1 John 5.7 are entirely correct, as this verse of the TR has practically no support from the Greek manuscripts.)

Thesis 9: "The charge that the non-Byzantine text-types are theologically aberrant is fallacious" (p. 62).

The author is probably right in thinking that most various readings are not the result of deliberate changes, but he goes too far in implying that the modern form of text does not significantly weaken any of the doctrines established in the KJV and the majority of manuscripts. He is particularly concerned with the treatment of the deity of Christ. In a table, eight passages are listed where various versions call Jesus "God". The KJV, RSV and NEB are each shown as ascribing deity to Christ in just four out of the eight passages, while the NIV does so in seven. In fact the RSV, NEB and NIV each weaken their testimony in several of the passages through the use of footnotes. It is interesting that the table does not include 1 Timothy 3.16 which in the Byzantine text and the KJV contains one of the most explicit references to the deity of Christ: "God was manifest in the flesh". Modem versions, including the NIV, follow the small minority of manuscripts which have the equivocal reading "He was manifest in the flesh". The existence of a few such readings is insufficient to prove any systematic and deliberate corruption of the text, but when these readings are found in manuscripts produced during a period of Arian influence, or in versions whose translators are known to hold unitarian views, there are good grounds for suspecting that the adoption of these readings is due to heretical bias. In this sense the charge of theological aberration is correct and not at all fallacious. Further, the deity of Christ is not the only doctrine weakened by the modern form of text, and it is useless to pretend that the alteration of any doctrinal passage does not to some extent weaken the testimony to that doctrine. Nor is it right to confine our attention solely to doctrinal passages. There is no word of Scripture which is not exceedingly precious.

Thesis 10: "The KJV was not accepted without a struggle, and some outstanding believers soon wanted to replace it" (p. 66).

This statement is correct as far as it goes but, as the author acknowledges, it does not prove anything about the quality of the KJV: "I make these points not because they prove the imperfection of the KJV, but because they inject a little historical perspective into the question at hand" (p. 67). It should be remembered that as far as the underlying Greek text is concerned, the KJV made no radical changes from the text underlying the Great Bible and other sixteenth century English versions, which were all based upon the Received Text (except the Rheims version, which was translated from the Vulgate). As the author concedes, "One reason for the eventual acceptance of the KJV was its excellence." The same may not be said of modern versions, which have deserted the traditional Greek text in favour of a text that is demonstrably corrupt.

Thesis 11: "The Byzantine text-type must not be thought to be the precise equivalent of the TR" (p. 67).

This too is true, and should not be forgotten by those who support the TR. There are blemishes in the TR which should eventually be removed by reference to the majority Byzantine text.

Thesis 12: see under Thesis 6.

Thesis 13: "Arguments that attempt to draw textual conclusions from a prejudicial selection of not immediately relevant data, or from a slanted use of terms, or by a slurring appeal to guilt by association, or by repeated appeal to false evidence, are not only misleading, but ought to be categorically rejected by Christians who, above all others, profess both to love truth and to love their brothers in Christ" (p. 74).

These remarks are to be warmly endorsed. Neither side helps its case by the excesses which are here complained of. The debate is one which should always be conducted in a spirit of Christian charity.

(In the course of his remarks, the author says 'What shall we say too about the vast majority of evangelical scholars, including men in whom were found the utmost piety and fidelity to the Word along with scholarship second to none? These men hold that in the basic textual theory Westcott and Hort were right..." It is interesting that those who scorn the appeal to the numerical weight of the Byzantine manuscripts, are perfectly content to make an appeal to the numerical weight of the scholars who reject these manuscripts. And yet the debate must be decided according to the vote of the manuscripts, and not according to the vote of the scholars.)

Thesis 14: "Adoption of the TR should not be made a criterion of orthodoxy" (P. 78).

This thesis is correct, insofar as the field of textual criticism is one in which there can be honest differences of opinion without the need to make accusations of heresy against an opponent.

W. N. Pickering's "The Identity of the New Testament Text" (1977)

Dr. Carson's Appendix (pp. 105-123) is devoted to a critique of Pickering's book, and overlaps with arguments which have been dealt with above. We shall examine here just two additional points.

On p. 116 Dr. Carson suggests that it is contradictory to hold that early Christians were on principle careful copyists, at the same time as holding that most corruptions to the text took place before the end of the second century. There is no real contradiction here: it is quite likely that most of the early copyists performed their task with relative accuracy, giving rise to the majority Byzantine text, while a minority of copyists negligently or deliberately corrupted the text on a large scale and were between them responsible for most of the variants now known, the worst of which are found in the non-Byzantine manuscripts.

On p. 117 it is suggested that neither Papyrus 75 nor Codex Vaticanus is 'recensional' (i.e. the immediate product of editorial activity), that P75 is not the parent of Cod. Vaticanus, that they are independent early witnesses to the Alexandrian text-type, which must therefore stand "closest to the autograph". But if, as is generally admitted, most corruptions took place in the first two centuries A.D., it is quite possible that both these manuscripts are descended from a single corrupt copy made earlier in the second century. The early date of P75 and Cod. Vaticanus is far from being a guarantee of excellence.


The second part of the book (pp. 81-102) handles non-textual questions, with a discussion of seven common arguments in favour of retaining the KJV. Most of this section is taken up with the question of the KJV's quality as a translation compared with the NIV and other modern versions. For reviews of the NIV and other modern versions, reference should be made to separate articles obtainable from the Trinitarian Bible Society.


Dr. Carson has put his finger upon logical and theological deficiencies in some of the supplementary arguments of those who support the traditional New Testament text, and he has rightly criticised the spirit in which these arguments are sometimes presented. For this reason his contribution to the Debate is to be welcomed. Concerning the more important arguments in favour of the traditional text, however, no one need be discouraged by Dr. Carson's objections, which have been shown to be entirely lacking in validity.

A. J. Brown 13.11.1979


For more information write to:

Trinitarian Bible Society
Tyndale House, Dorset Road
London, SW19 3NN, England
Telephone: 0 181 543 7857
Fax: 0 181 543 6370