The Recovery of the New Testament

Was the New Testament corrupted? Or has it been restored?

Thomas Alderman 2024

The Text of the New Testament: Its Transmission, Corruption, and Restoration, by Bruce M. Metzger (Oxford University Press, 3d ed. 1992; 4th ed. with Bart D. Ehrman 2005)¹.

I. Introduction.

The Apostle Peter exhorts believers to “be ready always to give an answer to every man who asks of you a reason for the hope that is in you.”  1 Peter 3:15.  We certainly ought, therefore, to be prepared in advance to answer those who insist that the text of the New Testament has been so corrupted that we cannot know what its authors were trying to tell us.

It’s a claim that cannot be lightly dismissed. 

Moveable type was not invented until 1440, and the papyrus and parchment on which the original New Testament documents were written were perishable.  The only way either to transmit the books geographically or to preserve their substance was to copy them manually.  Manual copying entailed the possibility of error.  We now have over 5,000 Greek manuscripts or fragments, to say nothing of over 10,000 Latin manuscripts; and according to New Testament scholar Bart D. Ehrman, there are over 200,000 variant readings in the Greek manuscripts alone – and they are not all mistakes: there are many changes that were made intentionally.

In 2010, in a class in Biola University’s Master’s Decree program in Science and Religion, we were provided a list of works which disputed the reliability of the New Testament documents and were required to write a review of one of them.  I chose Ehrman’s Misquoting Jesus – The Story Behind Who Changed the Bible and Why (Harper One, 2005).  I now find myself wondering whether Misquoting Jesus belonged on the list.  Despite his use of a lot of worrying rhetoric, Ehrman acknowledges that the text of the New Testament has been recovered to a condition “closely (very closely)” related to what the original documents must have said.²

Needless to say I was very happy to find that out; but Ehrman did very little to explain how, if the NT documents had been so thoroughly compromised, they ever could have been restored.  It seemed like an impossible task.  But recently a friend suggested I read Bruce M. Metzger’s The Text of the New Testament – Its Transmission, Corruption, and Restoration (Oxford University Press, 1964, 1968, 1992, 2005³; hereinafter, “The Text of the NT”), which has answered my question.  The matter is of such importance, and it has become the source of such encouragement to me, that I am compelled to share what I have learned.

There are other important topics concerning the text of the New Testament besides the question of the integrity of the text itself, and it deserves to be mentioned what Metzger and Ehrman’s essay is not about.  It is not about canonicity – the question of which writings deserve the status of scripture.  Inspiration and inerrancy of the original autographs and the veracity of the recovered text are also not in view.  The question at hand was simply, What were the words written by the original authors of the books of the New Testament?

As I said, I got my answer, but it is not a simple one, and it is not one which will necessarily satisfy everyone.  Emphatically, none of the central teachings of the Church – as expressed, say, in the major creeds and confessions – are placed in question.  But some uncertainties in the text of the NT do persist, and not all are trivial.  Some may find it difficult to avoid reinterpretation of a few favorite passages.  What is both more important and more encouraging is that our awareness of another layer of complexity in understanding the Word may help us in at least two very important ways.  For one thing, we will become better Bible readers.  For another, we must, if possible, avoid showing ourselves ignorant of current conversations concerning our own sacred texts, and be ready to respond intelligently to those who say they have heard that the Bible has been so monkeyed with as to be unreliable.

II. Many variants.

Given the fact that errors in the transmission of the New Testament occurred early in the history of the Church, and given the fact that the errors resulted from human weakness both physical and moral, it must be understood that the corruption of the textual tradition of the Church presumably began with the very earliest copies.  I draw that inference from a famous complaint by the early church Father Origen Adamantius of Alexandria (185-253), writing perhaps 150 years after the publication of Mark’s Gospel:

[T]he differences among the manuscripts [of the Gospels] have become great, either through the negligence of some copyists or through the perverse audacity of others; they either neglect to check over what they have transcribed, or, in the process of checking, they lengthen or shorten, as they please.

A. Innocent mistakes.

Mistakes were made – many, many mistakes.  There were changes due to faulty eyesight – some Greek letters are not easily distinguishable from one another, especially if they are not carefully drawn.  Thus, in 2 Pet. 2:13, some manuscripts read ARAFIAII] (“love feasts”) and others read AIIATAIS (“deceptions”).  And in manuscripts of 2 Pet. 2:18, “scarcely” alternates with “really” where the tau and the gamma are confused.  Or, an entire sentence or even a verse might be dropped where its ending is identical to the ending of the preceding line or sentence or verse.

There were errors arising from faulty hearing, as where two letters or syllables sound alike.  There were errors of memory, substitutions of synonyms, and transpositions of letters.  There were errors of judgment, as when a scribe would incorporate into the text a note that an earlier scribe had made in the margin.

B. Intentional Changes.

i. “Corrections”.

And there were intentional changes.  Some were done in good faith.  Scribes would think they were correcting errors of spelling, grammar, or historical fact.

ii. Harmonizations.

Where the same event or teaching is recorded in two or more Gospels, scribes would sometimes change one or the other in order to harmonize them.  For instance, the words that belong in John 19:20, “It was written in Hebrew, in Latin, and in Greek,” have been introduced into the text of many manuscripts at Luke 23:38. Similarly, the shorter form of the Lord’s Prayer in Luke 11:2-4 was replaced in many copies with the more familiar, longer form in Matt. 6:9-13.  And Metzger describes the efforts of a few scribes “to harmonize the Johannine and Markan accounts of the Passion by changing ‘sixth hour’ in John 19:14 to ‘third hour’ as in Mark 15:25.  The statement in Mark 8:31 that “the Son of man must . . . be killed and after three days rise again,” was changed by some copyists to the more familiar expression “on the third day.”

iii. Clearing up Historical and Geographical Difficulties.

Matthew (27:9) attributes to the prophet Jeremiah what actually comes from Zechariah 11:12-13:

Then what was spoken by Jeremiah the prophet was fulfilled: “They took the thirty silver coins, the price set on him by the people of Israel, and they used them to buy the potter’s field, as the Lord commanded.”

Some scribes either substituted the correct name or omitted it altogether.

It is clearly Zechariah, despite some weak parallels between Matthew and Jeremiah 32:6-15, where Jeremiah purchases land at the price of 17 shekels of silver.  One possible conclusion is that Matthew made a mistake.  Another possibility, albeit a remote one, is that Matthew got it right and a scribe, thinking of Jeremiah 32, changed it from Zechariah to Jeremiah.  But parenthetically, the way the prophecy is fulfilled is truly amazing.  Here are the parallels: 

Zechariah Matthew

The shepherd receives 30 pieces of Judas receives 30 pieces of silver for his

silver for his services as shepherd. services as traitor.

The Lord said to me, “throw it to the Judas throws the money into the Temple.

potter” – the handsome price at which 

they priced me!  So I took the thirty pieces The chief priests use the money to buy the

of silver and threw them into the house of potter’s field.

the Lord to the potter.

John 7:8: “I am not going to the festival yet.”  The NIV offers a marginal note: “Some early manuscripts do not have “yet.”  Did a copyist add it in order to avoid the implication that Christ fibbed?

At Mark 14:62, the statement of Jesus to the high priest that he, the high priest, would see the son of man “coming with the clouds of heaven” was deleted, the authors say, in order to prevent it from appearing that the Lord had made a factual error, since, in fact, the high priest died long before this event to end all events⁴ (although the high priest may yet see Him coming on the clouds!).

In Luke 23:32 the text of the manuscript denominated p⁷⁵, a codex⁵ of Luke and John, reads, “And also other criminals, two, were led away with him to be crucified.”  Says Metzger, “To avoid the implication that Jesus was also a criminal, most Greek witnesses have changed the text to “And also two others, criminals, were led away with him to be crucified.”⁶  (The NIV has it: “Two other men, both criminals. . . .”)

It must be noted that despite the negligence and hidden agendas of some scribes, most copyists were highly skilled and conscientious and devoted their lives to their painstaking work.  As Metzger writes, “There are . . . instances of difficult readings that have been transmitted with scrupulous fidelity.”⁷  The scribes, for all their faults, preserved the Word of God through the centuries, using the tools they had.  

C. Alterations Made Because of Doctrinal Considerations.

There were even deliberate alterations done for doctrinal reasons, whether to avoid the doctrinally unacceptable or inconvenient, or to introduce spurious evidence for a favored theological tenet.  “[The church Fathers] Irenaeus, Clement of Alexandria, Tertullian, Eusebius, and many others,” writes Metzger, “accused the heretics of corrupting the Scriptures in order to have support for their special views.  In the mid-second century Marcion expunged his copies of the Gospel according to Luke of all references to the Jewish background of Jesus.”⁸

There are several instances of changes made to avoid the implication that Joseph was Jesus’ biological father.  Metzger:

In Luke 2, there are several references to Joseph and Mary which, in the ordinary text, doubtless appeared to some persons in the early Church to require rephrasing in order to safeguard the virgin birth of Jesus.  In 2:41 and 43, instead of the words “his parents,” some manuscripts read “Joseph and Mary.”⁹

(The NIV refers to Joseph and Mary as “his parents,” without comment.)

In Luke 2:33 and 2:48, the oldest form of the text refers to Joseph as Jesus’ father.  In some later witnesses, however, it is “your relatives and I,” or “we.”)  (NIV: “the child’s father and mother” and “your father and I,” without comment.)

Metzger comments:

. . . This kind of change makes sense in the context of early Christological controversies, where some Jewish-Christian groups and other Christians holding to “adoptionist” Christological views were claiming that Jesus was a full flesh-and-blood human, the son of Joseph and Mary.¹⁰

Conversely, some passages were modified to emphasize Jesus’ divinity.  At the end of the prologue to John’s Gospel, most witnesses describe Christ as  “the only Son who is in the bosom of the Father” (1:18), whereas some other witnesses, principally Alexandrian, describe Him as “the only God who is in the bosom of the Father.”¹¹ (NIV: “No one has ever seen God, but God the one and only, who is at the Father’s side, has made him known;” and in a footnote: “Or the only begotten.”  And in another footnote: “Some manuscripts but the only (or only begotten) son.][Compare John 3:16: “one and only son,” or “his only begotten son.”)

The passage at Luke 22:43-44 (the bloody sweat) is missing from the oldest manuscripts, including most Alexandrian.  Metzger and Ehrman suggest they were added by those wishing to show that in addition to being wholly divine, Jesus was also wholly human, over against those who, like the Marcionites, maintained that Jesus was wholly and only divine.¹²

D. Jewish-Christian Relations.

In a chapter brand new to the Fourth Edition of The Text of the NT, and also bearing the unmistakable marks of Ehrman’s influence, the authors discuss several “Use[s] of Textual Data for the Social History of Early Christianity.”  Of particular interest are the sections on Jewish-Christian relations and the status of women.

It does seem that conflicts with Judaism, evident throughout the first three Christian centuries, affected some scribes.  Metzger and Ehrman note that Eldon Jay Epp “concluded that some 40% of Codex Bezae’s [aka the Western text] variant readings in Acts point toward an anti-Judaic bias.  The sensible inference is that the scribe himself, or his tradition, was anti-Jewish (in some sense) and that this prejudice came to be embodied in the transcription of the text.”¹³  (It should not be overlooked that Codex Bezae was the earliest and least reliable of the three major types of text: Bezae, Syrian (aka Byzantine), and Alexandrian.)

Many such instances are provided.  At Matt. 1:21, instead of Christ saving “his people,” it is said that he will save the world.  (NIV has “his people.”)  In John 4:22, salvation is “from Judea” instead of “of the Jews.”  (NIV: “from the Jews.”)  

Some manuscripts omit Jesus’ prayer from the Cross for the forgiveness of His enemies (“Father, forgive them, for they do not know what they are doing,” Luke 23:34) – “an omission,” say Metzger and Ehrman, “that makes particular sense if Jesus is understood to be asking God to forgive the Jews responsible for his crucifixion.”¹⁴  (NIV: “Some early manuscripts do not have this sentence.”)  Note also Acts 3:17, where Peter says to the crowd at the Temple: “I know that you acted in ignorance, as did your leaders.”  Could Peter have taken this directly from the Lord Himself?

Elsewhere, however, Metzger had written in 1994 (without Ehrman), “The logion, though probably not a part of the original Gospel of Luke, bears self-evident tokens of its dominical origin, and was retained¹⁵, within double square brackets, in its traditional place where it had been incorporated by unknown copyists relatively early in the transmission of the Third Gospel.”¹⁶  So was it removed from the original, or added later?  Even Bruce Metzger seems unable to decide.  Or was this a point of contention between him and Ehrman?

E. The Oppression of Women in Early Christianity.  

The status of women in the church is, if anything, more controversial now than ever before.  This is a topic into which I have not myself delved significantly until now.  If Metzger and Ehrman’s treatment of the situation is any indication, the lack of resolution in regard to this matter probably stems at least in part from the surprisingly numerous points of ambiguity in the original source documents, leaving room for debate.

Here is the NIV:

33a For God is not a God of disorder but of peace — (b) as in all the congregations of the Lord’s people.

34 Women should remain silent in the churches. They are not allowed to speak, but must be in submission, as the law says. 

35 If they want to inquire about something, they should ask their own husbands at home; for it is disgraceful for a woman to speak in the church.

The Revised Standard Version, however, includes 33b not as part of the preceding sentence as in the NIV, but as part of the next sentence:

33a For God is not a God of confusion but of peace.

b As in all the churches of the saints, 

34 the women should keep silence in the churches (etc.).

The latter reading, it could be argued, suggests that here Paul is appealing to an existing, prevailing practice: a tradition.  Surely this had been the practice in the synagogues since time immemorial.  Also the sentence, “They are not allowed to speak, but must be in submission, as the Law says,” seems slightly curious.  The Law? “It is for freedom that Christ has set us free.”  Galations 5:1.  

The editors of the 1984 edition of the NIV observe that it is clear from 11:5 that women prayed and prophesied in church.  Paul’s purpose here, they say, was not to limit women’s role in the church but to promote unity.  They observe that in Corinthian culture, women were not permitted to confront men in public, and that women were not afforded the opportunity of formal religious education; and the editors suggest that in their newfound Christian freedom the women were asking questions that could more easily be addressed at home.

Metzger and Ehrman describe still another theory:

Many scholars have claimed that the passage is not Pauline but represents an interpolation, made perhaps by the author of (the pseudepigraphic?) 1 Timothy (cf 2.1-10).  While one common objection to the interpolation theory has been the lack of manuscript attestation – the passage is present in all of our witnesses — Gordon Fee has stressed the text-critical evidence in its support, observing that the verses in question occur in a different location in Western witnesses [which are the earliest among the major traditions] (giving the passage the appearance of a marginal note incorporated at more or less appropriate junctures).  If Fee is correct concerning their secondary character, the interpolation may show that women came to be oppressed  more severely in a later period of Pauline Christianity (perhaps around the end of the first century) than at the outset.

. . . Ben Witherington has summarized some of the evidence that suggests that the scribe of Codex Bezae was intent on de-emphasizing the prominent role that women played in the early church, as recorded in the narrative of Acts. . . .  Witherington observes that in Bezae’s text of Acts 17.4, Paul’s Thessalonian converts are unambiguously “wives of prominent men” rather than “women of prominence” [NIV: “prominent women”], that the high profile of women is occasionally compromised by the insertion of references to their children (Acts 1.14) or to men of high profile (Acts 17.12) [NIV: “prominent Greek women and many Greek men”, without comment], and that the regular transposition of “Aquila” to precede “Priscilla” may intimate the scribe’s uneasiness with the woman’s implicit priority.¹⁷

But it appears that as of 2005 at least, the matter still lacked consensus:

While other scholars have also discussed, in brief order, the significance of textual problems for assessing the oppression of women in early Christianity, we are still awaiting an extensive and rigorous analysis.¹⁸

III. A Short History of Recovery.

We have seen that textual scholars have been highly successful in identifying variant readings in the hand-copied New Testament manuscripts.  Does that mean they have been equally successful in determining which of those readings are faithful to the original autographs?  Surprisingly, and thankfully, the answer is strongly affirmative: scholars agree that the text of the New Testament has been substantially recovered, with a limited number of remaining problematic texts which do not affect any central doctrine of the faith.  How was this done?

Through the centuries efforts were made to refine the text; but such efforts were frustrated by the need to preserve and transmit the new versions by the same means – manual copying – that had beset the New Testament from the beginning.  For instance, in about 382 A.D. the Pope invited the great scholar Jerome to prepare a revision of the Latin Bible, which he did. Sadly, that version itself became corrupted in the same way earlier versions had done.

Even after the invention of the printing press gave scholars a means of preserving renewed text, the project developed slowly at first and became a multi-generational undertaking.  Indeed, it was not until the posthumous publication of John Mill’s Greek New Testament in 1710 that the movement to recover the text started getting traction.  Mill collected “all the evidence from Greek manuscripts and . . . Fathers that lay within his power to procure.”  In the process he identified more than 30,000 variant readings in the Greek New Testament alone.¹⁹

Many scholars were disturbed by this, but at least it helped to define the problem.  Lutheran Greek scholar Johann Albrecht Bengel (1687-1752) resolved to devote himself to the study of the transmission of the text.  He became the first to identify particular lineages of manuscripts, which he designated the Asiatic, the Latin, and the Alexandrian.  He also articulated an axiom that has governed textual criticism to this day, namely, that the difficult reading is to be preferred to the easy, based on the insight that a scribe is more likely to simplify a difficult passage than to complicate an easy one.

The project was further advanced by Johannes Martin Augustinus Scholz (1794-1852), who traveled the world in order to compile “the first comprehensive listing of Greek manuscripts of the New Testament, adding 616 new manuscripts to those previously known.”²⁰

In 1881 Westcott and Hort distinguished four principal types of text, which they called the Syrian (aka the Byzantine), the Western (aka Bezae), the Alexandrian, and the Neutral.  The Syrian they characterized as being farthest from the original.  Unfortunately, it was the Syrian text which formed the primary basis for the Textus Receptus (also known as the Vulgate).  This became the principal Latin version of the Bible, prepared, as mentioned above, mainly by Jerome in the late 4th century, and (as revised in 1592) adopted as the official text of the Roman Catholic Church.

The Western text is the most ancient, dating to “perhaps before the middle of the second century,” but is characterized by a love of paraphrase.  “Words, clauses, and even whole sentences were changed, omitted and inserted with astonishing freedom. . . .  But its most dangerous work is “harmonistic” corruption, that is, the partial or total obliteration of differences in passages otherwise more or less resembling each other.”²¹

Scholars generally agree that a major contribution of Westcott and Hort was to establish that the Syrian text is most recent.  Three types of evidence support this conclusion:

The Syrian text contains conflate readings clearly composed of earlier forms.

No ante-Nicene Father quotes a distinctively Syrian reading.  And

Careful comparison with other readings decisively undermine its claim to be original.²²

In contrast, Westcott and Hort found the Alexandrian text to be superior, as what “might be expected from the influence of a Greek literary centre,” says Metzger, “a delicate philological tact in correcting forms, syntax, and in subtle changes made in the interest of attaining a greater degree of polish in language and style. . . .”²³

Westcott and Hort considered the “Neutral” text to be the most free from corruption.  It is best represented by codex Sinaiticus and codex Vaticanus.  Codex Sinaiticas is the oldest extant complete copy of the Christian Bible and dates from the middle of the Fourth Century.  All but a few of its leaves are housed in the British Museum in London, where my eyes have seen it! Codex Vaticanus is a nearly complete copy of the Christian Bible.  It also dates from the Fourth Century and is housed in the Vatican.

Westcott and Hort’s Greek New Testament was adopted as the underlying text for the 1881 revision of the King James Version of the Bible.

Metzger offers this evaluation of Westcott and Hort’s work:

. . . [T]he overwhelming consensus of scholarly opinion recognizes that their critical edition was truly epoch-making.  They presented what is doubtless the oldest and purest text that could be attained with the means of information available in their day.  Though the discovery of additional manuscripts has required the realignment of certain groups of witnesses, the general validity of their critical principles and procedures is widely acknowledged by textual scholars today.”²⁴

Not everyone was happy.  John W. Burgon (1813-88), Dean of Chichester, was indignant.

. . . [H]e could not imagine that, if the words of Scripture had been dictated by the inspiration of the Holy Spirit, God would not have providentially prevented them from being seriously corrupted during the course of their transmission.  Consequently it was inconceivable to Burgon that the Textus Receptus, which had been used by the Church for centuries, could be in need of the drastic revision which Westcott and Hort had administered to it.²⁵

This is very reminiscent of the problem of Bart Ehrman himself as stated in his work, Misquoting Jesus, published in the same year as the Fourth Edition, of which he was a co-author with Bruce Metzger.  For Ehrman, it is not enough that the text of the New Testament has been very substantially recovered.  It is not enough that, as he acknowledges, “the essential Christian beliefs are not affected by the textual variants in the manuscript tradition of the New Testament.”²⁶  For Ehrman, confidence in the Scriptures is precluded by the mere fact that there were ever any variant manuscripts; that there are any remaining uncertainties; and that the books of the New Testament were written by human authors.  Note how similar the above description of Burgon’s problem, by the authors of the Fourth Edition, is to Ehrman’s description of his own problem in Misquoting Jesus: 

This became a problem for my view of inspiration, for I came to realize that it would have been no more difficult for God to preserve the words of scripture than it would have been for him to inspire them in the first place. . . .  The fact that we don’t have the words surely must show, I reasoned, that he did not preserve them for us.  And if he didn’t perform that miracle, there seemed to be no reason to think that he performed the earlier miracle of inspiring those words.²⁷

This is not an account of a past stage in Ehrman’s intellectual development: it is a statement of his belief as of 2005.  He says, “This is the shift in my own thinking that I ended up making, and to which I am now fully committed.”²⁸

Ehrman and Burgon both make the same mistaken assumption about what God would or would not do: that if God inspired the text, he would also prevent it from being corrupted.  The only difference is that Ehrman recognizes that the text was corrupted and Burgon does not.  So for Burgon, inspiration means that God did prevent the corruption of the New Testament, which therefore must be and is pristine and in no need of any “recovery”; whereas, for Ehrman, God’s failure to prevent the corruption of the text means it was never inspired in the first place.

Both lines of reasoning are fallacious.  Ehrman is mistaken in assuming that if God didn’t preserve the text once it had been written, there is therefore “no reason” to suppose He inspired the autographs themselves.  There is a very clear distinction to be made, however, between what the original authors wrote on the one hand, and everything that was done to the text after them, on the other hand.  Matthew writes at 14:26 that at the Last Supper Jesus promised the disciples that “the counselor, the Holy Spirit, whom the Father will send in my name, will teach you all things and will remind you of everything I have said to you.”  Jesus did not make a similar promise to any subsequent generation.

Clearly, neither Ehrman nor Burgon knows what God would do.  God may see a greater benefit in a recovered text than in one never wanting recovery.  And God has protected His Word – perhaps not in the manner or at the time Ehrman expected, but through recovery itself.  Is God not permitted to do that?  Let’s remember that even Ehrman acknowledges that the variants do not shake any central doctrine of the faith, so it’s not as if there had ever been a period in church history during which seekers were adrift without a clear scriptural path to salvation.

It is interesting that Ehrman describes Misquoting Jesus and The Text of the New Testament, both published in 2005, as written for separate audiences – the former for a lay audience, the latter for a professional audience.  And in the same paragraph he states that the former “is written for anyone who might be interested in . . . seeing how we might, through the application of some rather rigorous methods of analysis, reconstruct what those original words actually were.”²⁹  We can do that?  Then where is the problem?

IV. It Could Have Been Worse.

With thousands of manuscripts and hundreds of thousands of variant readings, the very notion that the original text might be uncovered must have seemed preposterous.  Yet there were a number of factors which have facilitated the restoration of the text:

The classification of the manuscripts into distinct families and hence, the ability to trace the changes in the text through space and time.

The identification of the textual tradition having the greatest degree of integrity (i.e., the Alexandrian tradition).

The sheer number of manuscripts.

The availability of the writings of the early church Fathers.

The relative purity of the Pauline corpus.

The great learning of textual critics.

The vigilance of concerned parties.

The advent of digital computing.

A. Families of witnesses

As mentioned earlier, Johann Albrecht Bengel (1687-1752) was the first to recognize that most of the thousands of NT manuscripts could be associated with forms of the text which had originated at the great centers of ancient Christianity: the “Western” text, the Alexandrian text, and the Syrian text.

1. The Western text

Of these, the Western form of the text is deemed the earliest because it was used by Marcion, Justin, Heracleon, Irenaeus, Tertullian, and other church leaders of the second century.  Unfortunately, the Western text was changed by its copyists “with astonishing freedom,” and it is characterized by a  fondness of paraphrase.  Words, clauses, and even whole sentences are freely changed, omitted, or inserted.  Sometimes the motive appears to have been harmonization, while at other times it was the enrichment of the narrative by inclusion of traditional or apocryphal material.³⁰

In the Fourth Edition, and in what seems a distinctively Ehrmanian gloss, it is hypothesized that the earliest copyists would not have been trained professionals, but simply literate members of a congregation who had the time to do the work:

. . . Since most, if not all, of them would have been amateurs in the art of copying, a relatively large number of mistakes no doubt crept into their texts as they reproduced them.  It is possible that after the original was placed in circulation it soon became lost or was destroyed, so all surviving copies may conceivably have derived from some single, error-prone copy made in the early stages of the book’s circulation.³¹

Yes, one may suppose that is possible.  Yet it is far from self-evident that professional copyists would not have been available in first- and second-century Israel, as the scribal profession had long been well-established in a society governed by ancient writings – which themselves, of course, could only be preserved or disseminated by manual copying.  There is even reason to suppose that any number of the Jewish scribes became Christians, as did many Jewish priests.  Acts 6:7.  Nor is there any suggestion that undisciplined copying practices affected the Alexandrian form of the text.  Quite the contrary is the case, as we shall soon see.

2. The Byzantine Text.

If the Western form of the text was the earliest, scholars agree that the Byzantine (aka Syrian) text was the latest, and hence the furthest removed from the autographs.³²  Nevertheless the vast majority of extant manuscripts are of this type because the Byzantine became popular in Constantinople and throughout the Byzantine Empire, and eventually formed the basis for the Textus Receptus.³³

3. The Alexandrian text.

“It would be a mistake,” write Metzger and Ehrman, “to think that the uncontrolled copying practices that led to the formation of the Western textual tradition were followed everywhere. . . .”  They continue:

In particular, there is solid evidence that in at least one major see of early Christendom, the city of Alexandria, there was conscious and conscientious control exercised in the copying of the books of the New Testament. . . .  Alexandria, widely known throughout the ancient word as a major center for learning and culture, had a long history of classical scholarship, attached principally to its famous museum and library but influential on larger parts of its population.  It is no surprise, then, to find that textual witnesses connected to Alexandria attest a high equality of textual transmission from the earliest times.  It was there that a very ancient line of text was copied and preserved, as is evidenced in such Alexandrian church writers of the third and fourth centuries as Origen, Athanasius, and Didymus the Blind, in such notable manuscripts as p⁶⁶, p⁷⁵, Codex Vaticanus, and Codex Sinaiticus, and in copies of the Coptic versions.  In light of the striking similarities in text between the fourth-century [Codex Vaticanus] and the early third-century [Codex Sanaitcus], it is clear that the Christian scholars of Alexandria worked assiduously to preserve an accurate form of text.

With the discovery . . . of p⁶⁶ and p⁷⁵, both dating from about the end of the second or the beginning of the third century, proof became available that Hort’s Neutral text goes back to an archetype that must be put early in the second century. . . .  Most scholars are still inclined to regard the Alexandrian text as on the whole the best ancient recension and the one most nearly approximating the original.³⁴

B. Thousands of Manuscripts.

The proliferation of biblical manuscripts sets the Christian texts apart from other ancient documents, of which there were often very few extant exemplars – sometimes a handful, or even only one or two.  At the same time, and paradoxically, the multiplication of copies and of generations of copies meant that the “genealogy” of the text was preserved for those with the skill, the desire, and the time to scrape away the detritus of the centuries.

C. The Writings of the early Church Fathers.

In addition to the thousands of NT manuscripts, the early Church Fathers’ published letters, sermons, and commentaries contain extensive scriptural quotations.  For instance, Polycarp (69-155), a disciple of the Apostle John, in a letter to the church at Philippi, quoted from all twenty-seven books of what became the Canon, and no others, as authoritative.  “Indeed,” says Metzger, “so extensive are these [patristic] citations that if all other sources for our knowledge of the text of the New Testament were destroyed, they would be sufficient alone for the reconstruction of practically the entire New Testament.”³⁵

D. The Relative Purity of the Pauline Corpus.

“There are relatively few variant readings in the textual tradition of Paul’s letters,” according to F. F. Bruce.³⁶

E. The Great Learning of the NT Textual Critics.

In addition to the principals of textual criticism and a working knowledge of the English, Greek, Hebrew, and German languages, a textual scholar must be proficient in the following disciplines:

Knowledge of the development of Christian doctrine and cult, as well as all the heretical aberrations in the early Church.

Paleographical features of majuscule³⁷ and minuscule³⁸ hands, along with a knowledge of dialectical variations in Greek orthography and syntax.

The style, usage, and thought of the authors of the NT, in order to know whether any given anomaly should be judged to be foreign to the author’s intention.³⁹

F. Doctrinal Disputes of Early Christianity.

Metzger reports that Bludau “concluded that the manuscripts of the New Testament were not easily susceptible to deliberate falsification, given the vigilance exercised over their production by all concerned parties.”⁴⁰

G. Computers.

If the project of recovery was nearly complete in 2005, one may imagine that the digital revolution has almost tied it all up in a neat bow in the years since then.  It seems likely, too, that the development of artificial intelligence will find useful applications in the winnowing of the remaining variants.

Thus what must have seemed an overwhelming task became manageable for a brotherhood of dedicated scholars, stretching across centuries, and animated by the love of God and of His Word.

That doesn’t mean it was easy.

V. Methods.

“To teach another how to become a textual critic is like teaching another how to become a poet.”⁴¹

One of the purposes for this exercise has been to provide the reader with a means for forming his or her own conclusion as to whether the tools of the textual critics were sufficient to the task of recovery.  What were those tools?

A. Every Case is a Special Case.

First, it must be constantly borne in mind that not every textual problem will yield to the application of a rule.  Every case is special.  According to Kenney, this principle enshrines “what must be accounted the only completely and universally valid principle of textual criticism ever formulated.”⁴²

B. Prefer the Reading Which Best Explains the Origin of the Others.

Sometimes a change from one reading to another seems more likely than a change from the other to the one.

. . . We all follow this common-sense criterion when confronted with errors and “variant readings” in modern printed books.  For example, two editions of John Bunyan’s classic The Pilgrim’s Progress diverge in the story of Christian’s finding and using a key by which he is able to make his escape from Doubting Castle. One edition reads “The lock went desperately hard,” while the other reads “The lock went damnable hard.”  Which is the original reading and which has been altered?  Did Bunyan write “desperately” and a modern editor change it to “damnable” for some inexplicable reason?  Or did Bunyan write “damnable” (using the word in its nonprofane sense) and did someone subsequently alter it in order to remove what was deemed to be an offensive expression?  There can surely be no doubt what the answer is.⁴³

C. Prefer the Difficult to the Easy.

Prefer the reading which is more difficult for the scribe, unless it is so difficult as to be impossible.

. . . The characteristic of most scribal emendations is their superficiality, often combining “the appearance of improvement with the absence of its reality.” Obviously, the category “more difficult reading” is relative, and a point is sometimes reached when a reading must be judged to be so difficult that it can have arisen only by accident in transcription.)⁴⁴

D. Prefer the Shorter Reading.

In general, prefer the shorter reading, “except where . . . the scribe may have omitted material that he deemed to be superfluous, harsh, or contrary to pious belief, liturgical usage, or ascetic practice.”⁴⁵

E. Prefer the Dissident.

In parallel passages, the reading which stands in verbal dissidence with the others, is to be preferred, since scribes would frequently bring divergent passages into harmony with one another.

F. Reconstruct Histories.

Can a reading be traced to earlier manuscripts (the genealogical relationship or the “family tree”)?⁴⁶

. . . By way of example, suppose that there are seven manuscripts of an ancient book and that in a certain paragraph three of them agree in lacking a sentence that is present in the other four manuscripts. From this circumstance we would deduce either that a common ancestor of the three had omitted the sentence or that an ancestor of the four had added it.  Suppose, moreover, that we find that the seven manuscripts frequently range themselves so that one of them (which we may designate A) stands apart, showing no great similarity to any of the other six, while B, C, and D, on the one hand, and H, F, and G, on the other hand, greatly resemble each other, though differing somewhat from the rest. We can express this by saying that B, C, and D form a family, descended from a hypothetical common ancestor that we may call X, and that E, F, and G form another family, descended from a hypothetical ancestor that we may call Y. The readings of X that can be deduced by comparing those of B, C, and D will be of a higher antiquity and of greater authority than any of the readings in B, C, or D taken singly; and the same may be said for the readings of Y when compared with those in E, F, and G. Indeed, it is possible to go further: we may compare the readings of X and Y with each other and with those of A and thus deduce the readings of a still more remote ancestor that we may call Z, the hypothetical archetype of all the manuscripts. Thus, the pedigree of all ten manuscripts (the seven extant and the three hypothetical) would be as shown in Appendix 1.

It follows that because B, C, and D may agree in a given reading against A, such a reading is not three times more likely to be correct than the reading in A.  In fact, it is obvious that, other things being equal, there is a fifty-fifty chance that either of the two readings may be correct, for where B, C, and D agree they represent a manuscript (X) which is as far removed from the archetype (Z) as is A.  Thus, instead of merely counting the number of manuscripts supporting a given reading, the editor must weigh their significance in accordance with their mutual relations.  Often, however, difficulties hinder the construction of a stemma of manuscripts.  The simple example given above assumes that the different lines of descent have remained independent of one another.  But a disturbing element enters when mixture has occurred, that is, when a copyist has had two or more manuscripts before him and has followed sometimes one and sometimes the other or, as sometimes happened, when a scribe copied a manuscript from one exemplar and corrected it against another.  To the extent that manuscripts have a “mixed” ancestry, the genealogical relations among them become progressively more complex and obscure to the investigator.⁴⁷

G. Consider the geographical distribution of the witnesses that support a given variant.

H. Prefer that reading which is what the author is more likely to have written. 

There are a number of touchstones in this assessment:

1. The style, vocabulary, and theology of the author throughout the book.

2. The immediate context.

3. Harmony with the usage of the author elsewhere.

4. The Aramaic background of the teaching of Jesus.

5. The priority of the Gospel according to Mark.  And

6. The influence of the Christian community upon the formulation and transmission of the passage in question.⁴⁸

The appropriate question to ask is whether intrinsic evidence opposes the conclusion commended by genealogical considerations, the geographical distribution of witnesses, and transcriptional probabilities.⁴⁹

I. The Critic’s Intuition.

“Not all of these criteria are applicable in every case,” Metzger cautions.  “The critic must know when it is appropriate to give primary consideration to one type of evidence and not to another.  Since textual criticism is an art as well as a science, it is understandable that in some cases different scholars will come to different evaluations of the significance of the evidence. “⁵⁰

J. Conjecture.

Occasionally the manuscripts will refuse to yield their secrets.

. . . If the only reading, or each of several variant readings, that the documents supply is impossible or incomprehensible, the editor’s only remaining resource is to conjecture what the original reading must have been.⁵¹

One must admit the theoretical legitimacy of applying to the New Testament a process which has so often been found essential in the restoration of the right text in classical authors.  On the other hand, the amount of evidence for the text of the New Testament, whether derived from manuscripts, early versions, or patristic quotations, is so much greater than that available for any ancient classical author that the necessity of resorting to emendation is reduced to the smallest dimensions.⁵²

Before a conjecture can be regarded as even probable, it must satisfy the two primary tests that are customarily applied in evaluating variant readings in manuscripts: (1) it must be intrinsically suitable and (2) it must account for the corrupt reading or readings in the transmitted text. . . .  We require of a successful conjecture that it shall satisfy them absolutely well. . . .  The only criterion of a successful conjecture is that it shall approve itself as inevitable.  Lacking inevitability, it remains doubtful.⁵³

VI. Illustrations.

Metzger further elucidates the application of the principles of textual criticism by providing a number of examples of the kinds of problems which textual critics encounter, together with descriptions of how they have been resolved.

Certainly among the best known variants are the woman taken in adultery, the missing conclusion to Mark’s Gospel, and the Johannine Comma of 1 John 5.

A. The Woman Taken in Adultery.

. . . The story of the woman taken in adultery . . . has all the earmarks of historical veracity; no ascetically minded monk would have invented a narrative which closes with what seems to be only a mild rebuke on Jesus’ part: “Neither do I condemn you; go, and do not sin again.”  At the same time the pericope, which is usually printed as John 7:53-8:11, must be judged to be an intrusion into the Fourth Gospel. . . .⁵⁴

The authors base this conclusion on the absence of the pericope from all the oldest manuscripts and from the writings of the church Fathers for a thousand years, and on the marked differences in style and vocabulary.  “The pericope is obviously a piece of floating tradition,” they say, “which circulated in certain parts of the Western Church. . . .  The best disposition to make of the pericope as a whole is doubtless to print it at the close of the fourth Gospel, with a footnote advising the reader that the text has no fixed place in the ancient witnesses,” which is exactly what the NIV does.⁵⁵

B. The Missing Conclusion.

How did Mark end his Gospel?  Unfortunately, we do not know, Metzger says.  There are four endings that are current among the manuscripts, but after recounting the evidence for each one, he concludes that “probably none of them represents what Mark originally intended to stand as the close of his Gospel.”⁵⁶  (The NIV notes that “The earliest manuscripts and some other ancient witnesses do not have Mark 16:9-20.”)

C. The Johannine Comma.

Metzger avers that what has become known as the Johannine Comma was probably inserted into 1 John 5 in order to foil the Arians, who denied the deity of Christ, and he pronounces it “certainly spurious.”⁵⁷

1 John 5:7-8 reads in the NIV:

7 For there are three that testify: 

8 The Spirit, the water, and the blood; and the three are in  agreement.

But the King James Version has this reading:

7 For there are three that bear record in heaven, the Father, the Word, and the Holy Ghost: and these three are one.

8 And there are three that bear witness in earth, the Spirit, and the water, and the blood: and these three agree in one.

The editors of the NIV write that the extra text is “not found in any Greek manuscript before the sixteenth century.”  It now appears, says Metzger, that the disputed text was taken from the Latin Vulgate by a Franciscan friar who prepared it specifically for insertion into the Greek New Testament which Erasmus was preparing at the time – that is, in the early 16^th Century.  How did it get into the Vulgate?  “The Comma probably originated,” Metzger says, “as a piece of allegorical exegesis of the three witnesses and may have been written as a marginal gloss in a Latin manuscript of 1 John, whence it was taken into the text of the Old Latin Bible during the fifth century.”⁵⁸

That makes sense, in that no final resolution of the seeming paradox of the Trinity was reached in the church until the Fourth Century, and it was not formalized until Nicea in 325.

Of course that does not mean that the doctrine itself is spurious, as it has clear scriptural basis.  Matt. 28:19:

Therefore go and make disciples of all nations, baptizing them in the name of the Father and of the Son and of the Holy Spirit

and 2 Cor. 13:14:

May the grace of the Lord Jesus Christ, and the love of God, and the fellowship of the holy Spirit be with you all.

And let’s not forget “I and the Father are one.”  John 10:30.

D. More Examples.

Metzger provides several illustrations involving variant readings which turn on the presence or absence of a single letter.  One of these is the angelic chorus of Luke 2:14.  Did the angelic host at Jesus’ birth sing, as in the KJV, “Glory to God in the highest and on earth peace, good will toward men,” or did they sing, as in the RSV, “Glory to God in the highest, and on earth peace among men with whom he is well pleased”?  They differ only by one Greek letter.  Each is widely supported in the manuscripts and in the early church fathers.  Metzger favors the RSV, for reasons which do not greatly concern us here, but which reflect a playful erudition.

In 1 Thess. 2:7, did Paul say “we were gentle among you, like a nurse taking care of her children,” or did he say, “We were babes among you like a nurse taking care of her children”?  Metzger concludes it was the former.⁵⁹

There are instances where almost all of the “good” manuscripts are in error and the correct reading is preserved in “inferior” witnesses. “Thus at Matt. 12:47 the best Alexandrian witnesses, joining with representatives of the Western group, are convicted of having accidentally omitted a whole verse.  Verse 47 closes with the same word as verse 46, and the eyes of copyists have chanced to wander from the end of one to the end of the other.”⁶⁰  (NIV: “Some manuscripts do not have verse 47.”)

In Matt. 22:34-35, did the author identify Jesus’ inquisitor as a lawyer, or not?  Probably not.  This is another instance of  harmonization, where the copyist inserts material from another Gospel in order to harmonize the two.

In 1 Thess. 3:2, did Paul refer to Timothy as “our brother and God’s fellow worker,” or as “our fellow worker,” or as “God’s servant,” or as “servant and God’s fellow worker,” or as “God’s servant and our fellow worker”?⁶¹

In Acts 6:8, was Stephen:

Full of grace and power;

Full of faith and power;

Full of grace and faith; or

Full of faith and grace of the Spirit?

Each has significant support in the manuscripts.

Metzger proceeds to analyze these variant readings, and in so doing demonstrates the sophistication of some of the critical tools available to the textual critic.  First, he recounts the manuscript evidence in favor of each reading.  In view of its length, I set forth the rest of his analysis in full in Appendix 2 for the reader’s edification.

Did the angels sing, “Glory to God in the highest, and on

earth peace, good will toward men,” Luke 2.14 (King James Version), or did they sing, “Glory to God in the highest, and on earth peace among those whom he favors”?  (New Revised Standard Version)?  The answer (it is the latter reading) turns on one letter of the Greek text.

The authors provide many other examples, for which I heartily recommend the book itself, available on Amazon.

Metzger summarizes the accomplishments and the limitations of New Testament textual criticism:

By way of conclusion, let it be emphasized again that there is no single manuscript and no one group of manuscripts that the textual critic may follow mechanically. All known witnesses of the New Testament are to a greater or lesser extent mixed texts, and even several of the earliest manuscripts are not free from egregious errors.  Although in very many cases the textual critic is able to ascertain without residual doubt which reading must have stood in the original, there are not a few other cases where only a tentative decision can be reached, based on an equivocal balancing of probabilities.  Occasionally, none of the variant readings will commend itself as original, and one will be compelled either to choose the reading that is judged to be the least unsatisfactory or to indulge in conjectural emendation.  In textual criticism, as in other areas of historical research, one must seek not only to learn what can be known but also to become aware of what, because of conflicting witnesses, cannot be known.⁶²

VII. Conclusion.

In 1978 the International Council on Biblical Inerrancy convened more than 200 evangelical leaders, who adopted their Chicago Statement on Biblical Inerrancy.  Signatories to the statement came from a variety of evangelical Christian denominations, and included Robert Preus, James Montgomery Boice, Kenneth Kantzer, J. I. Packer, Francis Schaeffer, R. C. Sproul, and John F. MacArthur.  Article X of the Chicago Statement sets forth the Council’s views concerning our topic of textual criticism:

We affirm that inspiration, strictly speaking, applies only to the autographic text of Scripture, which in the providence of God can be ascertained from available manuscripts with great accuracy. . . .

We deny that any essential element of the Christian faith is affected by the absence of the autographs.   We further deny that this absence renders the assertion of Biblical inerrancy invalid or irrelevant.

Additionally, the Council’s “Exposition” of the Statement affirmed:

Since God has nowhere promised an inerrant transmission of Scripture, it is necessary to affirm that only the autographic text of the original documents was inspired and to maintain the need of textual criticism as a means of detecting any slips that may have crept into the text in the course of its transmission.  The verdict of this science, however, is that the Hebrew and Greek text appear to be amazingly well preserved, so that we are amply justified in affirming, with the Westminister Confession, a singular providence of God in this matter and in declaring that the authority of Scripture is in no way jeopardized by the fact that the copies we possess are not entirely error-free.

With today’s technological advances – computers, word-processing software, the internet, search engines, etc. – the age of New Testament recovery is coming to a conclusion.  It’s not perfect, but even the imperfections are, in general, known, and the editors of modern versions of the New Testament have flagged them for us.  For all essential purposes, we can be confident that we have the writings of the apostolic generation.

ENDNOTES

¹All references are to the Fourth Edition unless otherwise noted.

²Bart D. Ehrman, Misquoting Jesus – The Story Behind Who Changed the Bible and Why (Harper One, 2005), p. 62.

³Ehrman, who studied under Metzger at Princeton Theological Seminary, is co-author with Metzger of the 2005 Fourth Edition.  Unless otherwise indicated, all references are to the Fourth Edition.

⁴Metzger 3d ed., 202.

⁵A codex is a manuscript in book form, as distinguished from the earlier usage of scrolls.

⁶Metzger 3d ed., 202.

⁷Metzger 271.

⁸Metzger 265-266.

⁹Metzger, 267.

¹⁰Metzger, 285.

¹¹Metzger, 286.

¹²Metzger, 286.

¹³Metzger, 284.

¹⁴Metzger, 287.

¹⁵Here Metger is referring to his translation of the Greek New Testament.

¹⁶Bruce Manning Metzger, United Bible Societies, A Textual Commentary on the Greek New Testament, Second Edition a Companion Volume to the United Bible Societies’ Greek New Testament (4th Rev. Ed.) (London; New York: United Bible Societies, 1994), 154.  My emphasis.

¹⁷Metzger, 289-290.

¹⁸Metzger, 289-290.

¹⁹Metzger, 154-155.

²⁰Metzger, 169.

²¹Metzger, 178, quoting Hort.

²²Metzger, 180.

²³Metzger, 179.

²⁴Metzger 183.

²⁵Metzger, 181.

²⁶Bart Ehrman, Misquoting Jesus, 252.

²⁷Ehrman, 11.

²⁸Ehrman, 13.

²⁹Ehrman, 15.

³⁰Metzger, 3d ed., 132.

³¹Metzger, 275.

³²Metzger, 280.

³³Metzger, 3d ed., 170.

³⁴Metzger, 277-278, 312.

³⁵Metzger, 3d ed., 86.

³⁶F.F. Bruce, The Canon of Scripture (IVP Academic 1988), 130.

³⁷Larger, uppercase lettering, found uniformly in earlier manuscripts.

³⁸Lowercase lettering typifying later manuscripts.

³⁹Metzger, 313-314.

⁴⁰Metzger, 283.

⁴¹Metzger, 211.

⁴²Metzger, 209.

⁴³Metzger, 300.

⁴⁴Metzger, 303.

⁴⁵Metzger, 303.

⁴⁶The following extended example is set forth at length due to the importance of the critical tool in view and the difficulty in preserving clarity in a briefer presentation.

⁴⁷Metzger, 207-208.

⁴⁸Metzger, 303-304.

⁴⁹Metzger, 3d ed., 218.

⁵⁰Metzger, 304.

⁵¹Metzger, 226-227.

⁵²Metzger, 230.

⁵³Metzger, 227.

⁵⁴Metzger, 3d ed., 223.

⁵⁵Metzger, 3d ed., 223-224; Metzger, 320-321.

⁵⁶Metzger, 3d e., 226.

⁵⁷Metzger, 3d ed., 62.

⁵⁸Metzger, 3d ed., 102

⁵⁹Metzger, 3d ed., 230.

⁶⁰Metzger, 3d ed., 239.

⁶¹Metzger, 3d ed., 240.

⁶²Metzger, 343.

APPENDIX 1

APPENDIX 2

Stephen, Full of Grace

Of these four variant readings, it is obvious that either the first two are independent abridgements of the longer readings or the third and fourth readings arose from combining the elements of the first two.  Considerations of both external evidence and internal probability unite to demonstrate that readings 3 and 4 are secondary, being alternative conflations of the other two.  Reading 3 is supported by the majuscule manuscript E^a, which dates from the sixth century and is one of the earliest representatives of the Koine or Byzantine type of text in Acts.  Reading 4 is supported by the majuscule manuscript Ψ, which dates from the eighth or ninth century and has a mixed type of text in Acts.  Transcriptional considerations lead one to conclude that both readings 3 and 4 presuppose the priority of the other two readings, for it is easier to believe that a scribe, knowing the existence of readings 1 and 2, decided to join them, lest the copy that he was writing lose one or the other, than to believe that two scribes independently took offense at the longer reading and that each chose to perpetuate half of it in his copy.  Thus, external evidence, which is meager in extent and relatively late in date, and transcriptional probabilities unite against the originality of readings 3 and 4.  

Variant reading 2 is supported by two majuscule manuscripts, H^a of the ninth century and P^a of the tenth century, both representative of the Koine or Byzantine type of text.  The majority of the minuscule manuscripts join these two majuscule witnesses. . . .

Variant reading 1 is supported by a wide variety of witnesses, including representatives of the major pre-Koine types of text.  Codices Sinaiticus and Vaticanus, both of the fourth century, are the earliest and best majuscule representatives in Acts of the Alexandrian type of text.  Codex Bezae, of the fifth century, is the chief Greek representative of the Western group of witnesses.  Codex Alexandrinus, of the fifth century, and p⁷⁴, dating from about the seventh century, have a mixed type of text. The evidence of the early versions, including the Latin, Syriac, Coptic, and Armenian, reflects the wide geographical area over which the reading was accepted.  The external evidence in support of reading 1 is, therefore, far superior in point of age and diversity of text type to that supporting reading 2.

Internal probabilities likewise favor reading 1.  If the account originally stated that Stephen was “full of faith,” there is no discernible reason why a scribe should alter it to “full of grace.”  On the other hand, in view of the statement made three verses earlier that Stephen was a man “full of faith and the Holy Spirit” (verse 5), it is easy to understand that in transcribing the later statement in verse 8 copyists would be likely, either consciously or unconsciously, to substitute “faith”, which they recalled from the earlier passage, for the correct reading “grace”.  The presence of “of the Spirit” in reading 4 is to be explained in the same way.

Thus, the converging of several strands of evidence, both external and internal, leads one to the firm conclusion that the author of Acts 6:8 wrote “full of grace and power.”¹

[The NIV has it so.]