Manuscripts and Textual Criticism

What you'll learn

Describe the number, age and key examples of New Testament manuscripts, including P52, P66, P75, Codex Sinaiticus and Codex Vaticanus.
Classify textual variants by type and significance.
Answer the “telephone game” objection and accurately state what Bart Ehrman concedes about the text.

What textual criticism is

None of the original documents (the autographs) written by Paul, Luke or John survive. Like every ancient text, the New Testament was copied by hand for centuries before printing. Copyists sometimes made mistakes or changes, so manuscripts differ in places. A textual variant is any place where manuscripts differ in wording, spelling or word order.

A common mix-up concerns the Dead Sea Scrolls, found from 1947 onward in caves near Qumran by the Dead Sea. They are not originals, and they are not New Testament manuscripts. They are copies of Old Testament books (about 230 biblical manuscripts, covering every book of the Hebrew Bible except Esther) together with many other Jewish writings, dating from roughly the third century BC to the first century AD. They contain no New Testament text; the claim that one tiny Greek scrap (7Q5) comes from Mark has been rejected by most scholars. Their value is for the Old Testament: the Great Isaiah Scroll, from about the second century BC, is roughly a thousand years older than the medieval Hebrew manuscripts previously known and agrees with them closely in substance. For both Testaments, then, we have copies, not autographs; the question is how good the copies are.

Textual criticism is the discipline of comparing manuscripts to reconstruct the original wording as closely as possible. It is used for every ancient text, from Homer to Julius Caesar. Christian scholars such as Bruce Metzger, Daniel Wallace, Peter Gurry and Wes Huff work in this field, as do sceptics such as Bart Ehrman. The tools and data are shared; what differs is the interpretation.

Explain in your own words what textual criticism is and why it is needed for the New Testament.Show answer

None of the original documents (the autographs) survive, and the New Testament was copied by hand for centuries, so copyists sometimes made mistakes or changes and manuscripts differ in places. Textual criticism compares the manuscripts to reconstruct the original wording as closely as possible. It is used for every ancient text, and both Christian and sceptical scholars work with the same tools and data; what differs is the interpretation.

autographs lost
hand-copying produced variants
compare manuscripts to reconstruct
used for all ancient texts

What is the common mix-up about the Dead Sea Scrolls, and what are they actually valuable for?Show answer

People sometimes think the Dead Sea Scrolls are original documents or New Testament manuscripts, but they are neither. They are copies of Old Testament books and other Jewish writings from roughly the third century BC to the first century AD, and the claim that the scrap 7Q5 comes from Mark has been rejected by most scholars. Their value is for the Old Testament: the Great Isaiah Scroll is about a thousand years older than the medieval manuscripts previously known and agrees with them closely in substance.

not originals, not NT
copies of OT and Jewish writings
7Q5 claim rejected
Isaiah Scroll agrees closely

An embarrassment of riches

The New Testament is by far the best-attested work of ancient literature. Scholars have catalogued more than 5,800 Greek manuscripts, ranging from small fragments to complete Bibles. In addition there are over ten thousand Latin manuscripts, thousands in Syriac, Coptic and other early translations, and so many quotations in the writings of early church leaders (the church fathers) that much of the New Testament could be reconstructed from them alone.

Compare other ancient works. For most classical authors we have a handful to a few hundred manuscripts, often copied many centuries after the original. Homer's Iliad, the runner-up, has around two thousand. Yet historians do not doubt that we know substantially what Tacitus or Caesar wrote. By the same standards the New Testament text is extraordinarily well preserved.

Age matters as well as number. Several manuscripts are usually dated to the second and third centuries, within roughly a century or two of the originals, and on its common dating P52 is closer still. Dates based on handwriting (palaeography) are approximate, and some are debated.

Why is the New Testament described as "an embarrassment of riches" compared with other ancient works?Show answer

Scholars have catalogued more than 5,800 Greek manuscripts, plus over ten thousand Latin manuscripts, thousands in other early translations, and so many quotations in the church fathers that much of the New Testament could be reconstructed from them alone. Most classical authors survive in a handful to a few hundred manuscripts, often copied centuries later, and even Homer's Iliad has only around two thousand. Yet historians do not doubt we substantially know what Tacitus or Caesar wrote, so by the same standards the New Testament is extraordinarily well preserved.

5,800+ Greek manuscripts
translations and church fathers
far more than other ancient works
same standards applied

Key manuscripts to know

You do not need to memorise hundreds of names, but these five come up constantly. Key to the names: a papyrus is written on reed paper and numbered with a P (P52 is papyrus 52); a codex is a manuscript in book form rather than a scroll.

P52 (Rylands Library, Manchester): a credit-card-sized papyrus fragment of John 18:31-33 and 37-38. Its first editor dated it to the first half of the second century; more recent studies widen the range (Orsini and Clarysse suggest about AD 125-175, and some scholars allow the later second or even early third century). Found in Egypt, it shows John's Gospel was copied and circulating far from where it was written.
P66 (Bodmer Papyrus II): a substantial copy of John's Gospel, usually dated around AD 200, though some scholars argue for a later date. It shows corrections made by the scribe, which give a window into how copying actually worked.
P75 (Bodmer XIV-XV, now in the Vatican Library): large portions of Luke and John, traditionally dated about AD 175-225 (some recent scholars, such as Brent Nongbri, argue it could be as late as the fourth century). Its text is remarkably close to Codex Vaticanus, usually reckoned over a century later, showing careful transmission.
Codex Sinaiticus (mostly in the British Library): a fourth-century Bible, from roughly AD 330-360, containing the complete New Testament. It was found at St Catherine's Monastery at Mount Sinai by Constantin von Tischendorf in the nineteenth century.
Codex Vaticanus (Vatican Library): a fourth-century Bible from roughly AD 300-350, one of the most important witnesses, although its New Testament breaks off at Hebrews 9:14 and lacks the rest of Hebrews, 1 and 2 Timothy, Titus, Philemon and Revelation.

Which manuscript would you mention to show that John's Gospel was circulating early and far from where it was written, and why should you be careful about its date?Show answer

P52, a credit-card-sized fragment of John 18 found in Egypt, shows John's Gospel was copied and circulating far from where it was written. Its first editor dated it to the first half of the second century, but more recent studies widen the range, with Orsini and Clarysse suggesting about AD 125-175 and some allowing later. Dates based on handwriting are approximate, so it is honest to give a range rather than an exact date.

P52, John 18
found in Egypt
date debated, widened range
be honest about uncertainty

What does the comparison between P75 and Codex Vaticanus show, and what is distinctive about Codex Sinaiticus?Show answer

P75, containing large parts of Luke and John, has a text remarkably close to Codex Vaticanus, which is usually reckoned over a century later, showing careful transmission. Codex Sinaiticus is a fourth-century Bible from roughly AD 330-360 that contains the complete New Testament. It was found at St Catherine's Monastery at Mount Sinai by Tischendorf in the nineteenth century.

P75 close to Vaticanus
careful transmission
Sinaiticus: complete NT
fourth century

Types of variants

Estimates of the total number of variants range from a few hundred thousand to about 500,000 or more (Peter Gurry's careful 2016 estimate is around 500,000, not counting spelling differences), which sounds alarming. Ehrman often notes there are more variants than words in the New Testament. But the number is large precisely because we have so many manuscripts; a variant in one manuscript counted across thousands adds up fast. A text with only two copies would have very few variants and much less certainty.

Daniel Wallace helpfully sorts variants into four categories:

Spelling and nonsense errors: the largest group by far. For example, the Greek name John spelt with one or two *n*s, or a scribe's slip of the pen.
Minor changes that do not affect meaning: word order (Greek allows flexible order), such as “Christ Jesus” versus “Jesus Christ”; synonyms; or the presence or absence of the article (Greek can say “the Jesus”).
Meaningful but not viable: variants that would change meaning but appear in only one or two late manuscripts, so they have no real claim to be original. Example: the Trinitarian wording in the King James Version of 1 John 5:7-8 ("the Father, the Word, and the Holy Ghost") is in no early Greek manuscript and only a few late ones; Erasmus added it in 1522. The Trinity rests on many secure texts (Matt 28:19; John 1:1-18).
Meaningful and viable: a small fraction, often estimated at less than one per cent. Even these do not overturn any core doctrine. The best-known are the ending of Mark (16:9-20) and the woman caught in adultery (John 7:53-8:11), which most modern Bibles already mark in brackets or footnotes.

Example: Weighing a real variant: Mark 1:41

The variant: When a leper begs Jesus for healing, most manuscripts say Jesus was "moved with pity" (Greek splanchnistheis). Codex Bezae and a few Old Latin manuscripts say he was "angry" (orgistheis).

External evidence (the manuscripts): "Pity" has far wider and earlier support, including Codex Sinaiticus and Codex Vaticanus.

Internal evidence (scribal habits): "Angry" is the harder reading. Scribes were more likely to soften a startling word than to introduce one, so some scholars think "angry" is original. The NIV puts "indignant" in its text; most translations keep "pity" or "compassion".

What is at stake: Very little. Mark elsewhere shows Jesus angry at hardness of heart (Mark 3:5), and in every manuscript Jesus touches the leper and heals him.

What to say: "That's a real variant, and scholars weigh it openly. Either way, the passage shows the same Jesus who heals the unclean. No doctrine depends on it."

How would you answer someone who says, "There are more variants than words in the New Testament"?Show answer

The number is large precisely because we have so many manuscripts; one variant counted across thousands of copies adds up fast, and a text with only two copies would have few variants but much less certainty. Most variants are spelling and nonsense errors or minor changes like word order that do not affect meaning. Only a small fraction, often estimated at under one per cent, are both meaningful and viable, and even these do not overturn any core doctrine.

many manuscripts = many variants
most are spelling/minor
under 1% meaningful and viable
no core doctrine overturned

What is the difference between a "meaningful but not viable" variant and a "meaningful and viable" one? Give an example of the second.Show answer

A meaningful but not viable variant would change the meaning but appears in only one or two late manuscripts, so it has no real claim to be original. A meaningful and viable variant both affects meaning and has a real claim to be original. The best-known examples are the ending of Mark (16:9-20) and the woman caught in adultery (John 7:53-8:11), which most modern Bibles already mark with brackets or footnotes.

not viable: one or two late manuscripts
viable: real claim to originality
Mark 16:9-20, John 7:53-8:11
already marked in Bibles

Using Mark 1:41 or 1 John 5:7-8, explain how textual critics weigh a variant and why no doctrine is put at risk.Show answer

Critics weigh external evidence, which manuscripts support each reading and how early and widespread they are, and internal evidence, such as which reading better explains the other. In Mark 1:41 "pity" has far wider and earlier support, but "angry" is the harder reading that scribes might have softened, so scholars debate it; either way Jesus heals the leper. The Trinitarian words in 1 John 5:7-8 are in no early Greek manuscript and only a few late ones, so they are not viable, yet the Trinity is taught throughout the New Testament.

external evidence: which manuscripts
internal evidence: harder reading
Mark 1:41: Jesus still heals
1 John 5:7-8 not viable; Trinity taught elsewhere

The telephone game objection

Many people imagine transmission like the children's game of telephone (Chinese whispers): one person whispers to the next in a single line, and the message is garbled by the end. But manuscript transmission was nothing like that.

First, it was not a single line. Copies spread rapidly across the Roman world in many branches, so an error in one branch can be detected by comparison with others. Second, it was written, not whispered; we can examine early copies directly rather than relying on the last person in the line. Third, we have early copies close to the source, like being able to ask the second and third people in the line what they heard. Wes Huff often makes the point that we can actually see the process and correct for it, which is the opposite of the telephone game.

Variants are also usually easy to explain. When we can see that a scribe's eye skipped from one line ending to a similar ending (homoeoteleuton) or that he harmonised one Gospel to another, we can work back to the earlier reading.

The same goes for translation. Modern Bibles are not translations of translations of translations. Translators work directly from the Hebrew, Aramaic and Greek, using the best manuscripts. The ESV, NASB, CSB, NKJV, NIV and KJV are all faithful translations; their differences mostly reflect translation style (more word-for-word or more thought-for-thought) and, in a few places, textual decisions that footnotes usually point out.

Give three reasons why manuscript copying was not like the telephone game.Show answer

First, transmission was not a single line; copies spread rapidly across the Roman world in many branches, so an error in one branch can be found by comparing others. Second, it was written, not whispered, so we can examine early copies directly. Third, we have early copies close to the source, like being able to ask the second and third people in the line what they heard. As Wes Huff points out, we can see the process and correct for it.

many branches, not one line
written, not whispered
early copies near the source
we can correct for errors

How would you answer someone who says the Bible has been translated so many times that the meaning must have been lost?Show answer

Modern Bibles are not translations of translations; translators work directly from the Hebrew, Aramaic and Greek, using the best manuscripts. Faithful translations such as the ESV, NASB, CSB, NKJV, NIV and KJV differ mainly in style, more word-for-word or more thought-for-thought, and in a few places in textual decisions that footnotes usually point out.

translated directly from original languages
not a chain of translations
many faithful translations
differences: style and noted textual choices

What Bart Ehrman concedes

Bart Ehrman, a leading textual critic at the University of North Carolina, popularised textual scepticism in Misquoting Jesus (2005). It is important to represent him fairly. He argues that scribes changed the text, sometimes for theological reasons, and that we cannot be certain of the original wording in some places. He has also said the autographs are lost and that we have copies of copies.

Yet Ehrman acknowledges in Misquoting Jesus that most variants are immaterial and insignificant, and in an interview printed at the back of the paperback edition he says his position does not actually conflict with Bruce Metzger's view that essential Christian beliefs are not affected by textual variants; for the most part, he agrees. He adds that variants still affect the meaning of some theologically important passages and that some original wording may never be recovered. Metzger, his teacher and later co-author, held that view, and their textbook The Text of the New Testament treats the original text as largely recoverable. So the disagreement is less about the data than about how much uncertainty to emphasise.

As a Reformed believer, you can say with the Westminster Confession (1.8) that God has kept his word “pure in all ages” by his singular care and providence, and that this providence worked through many manuscripts, not by miraculous identical copies. The abundance of witnesses is itself the means of preservation. Evangelical inerrantists say the same. The Chicago Statement on Biblical Inerrancy (1978) affirms that inspiration strictly applies to the original text, which "in the providence of God can be ascertained from available manuscripts with great accuracy" (Article X). You met the Chicago Statement in Know Your Faith, Lesson 1.

Represent Bart Ehrman fairly: what does he argue, and what does he concede?Show answer

Ehrman argues that scribes changed the text, sometimes for theological reasons, that the autographs are lost, and that we cannot be certain of the original wording in some places. Yet he acknowledges that most variants are immaterial, and in the interview at the back of Misquoting Jesus he says that for the most part he agrees with Metzger that essential Christian beliefs are not affected by textual variants. So the disagreement is less about the data than about how much uncertainty to emphasise.

scribes changed text
some wording uncertain
most variants immaterial
agrees essential beliefs unaffected

How does the Westminster Confession (1.8) help a Reformed believer think about preservation?Show answer

The Confession says God has kept his word pure in all ages by his singular care and providence. This providence worked through many manuscripts rather than by miraculous identical copies. So the abundance of manuscript witnesses is itself the means God used to preserve his word.

kept pure in all ages
by providence
through many manuscripts
abundance is the means

Key texts

Isaiah 40:8: “The word of our God will stand forever”; the theological basis for confidence in preservation.
Matthew 24:35: Jesus promises his words will not pass away, a promise God fulfils through providence.
Jeremiah 36:27-32: When King Jehoiakim burned Jeremiah's scroll, God had it rewritten; a vivid picture of God preserving his word through copying.
2 Peter 3:15-16: Peter already treats Paul's letters as Scripture circulating among churches, showing early collection and copying.
Colossians 4:16: Paul tells churches to exchange and read his letters, explaining the rapid, multi-branch spread of copies.

Objections and answers

Try answering each one yourself before you open the response.

“There are more variants than words in the New Testament. How can you say we know what it originally said?”Show a response

That statistic is true but misleading. Variants are counted per manuscript, so if one spelling difference appears in 2,000 manuscripts, it can count as 2,000 variants. The large number is a sign of abundant evidence, not corruption; an ancient book with three copies would have few variants and far less certainty. Most variants are spelling slips, word order or synonyms that make no difference in translation. Only a small fraction, often estimated at less than one per cent, are both meaningful and plausibly original, and none of them overturns a central Christian belief. Even Bart Ehrman has said that, for the most part, he agrees essential Christian beliefs are not affected by textual variants. More data means more confidence, not less.

“We don't have the originals, just copies of copies of copies. So we can't trust it.”Show a response

We do not have the original of any ancient book: not Plato, not Tacitus, not Caesar's Gallic Wars. If the absence of autographs made a text untrustworthy, all of ancient history would collapse. The question is how close and how many the copies are. For the New Testament we have manuscripts usually dated to the second and third centuries, such as P66 and P75, plus thousands of later copies in many independent branches, early translations and patristic quotations. By comparing them, textual critics can identify where errors entered and recover the earlier wording. P75 and Codex Vaticanus, usually dated over a century apart, agree to a remarkable degree, which shows careful copying. Copies of copies, when abundant and early, are a strength.

“Scribes deliberately changed the text to make Jesus look more divine.”Show a response

There are a few places where scribes seem to have made theologically motivated changes, and textual critics openly discuss them; Ehrman's earlier book The Orthodox Corruption of Scripture collects examples. But two points matter. First, we can detect those changes precisely because other manuscripts preserve the earlier reading, so they did not succeed in corrupting the text. Second, the deity of Christ does not depend on any disputed verse. It is taught in passages that are textually secure: John 1:1-18, John 20:28, Philippians 2:5-11, Colossians 1:15-20, Hebrews 1:1-8 and many more. If you removed every disputed reading, the full doctrine of Christ's deity would remain untouched.

“What about the ending of Mark and the story of the woman caught in adultery? Doesn't that prove the Bible was added to?”Show a response

These are the two largest variants, and Christians do not hide them; most modern Bibles mark them with brackets or notes, which is textual criticism at work in plain sight. Many scholars think Mark 16:9-20 and John 7:53-8:11 were not part of the original texts, though some defend the longer ending of Mark. Either way, we know about them precisely because earlier manuscripts like Sinaiticus and Vaticanus lack them. No doctrine depends on these passages; the resurrection appearances and Jesus' compassion to sinners are taught elsewhere. Identifying a later addition is not evidence of an uncontrolled text but of a text we can check.

“Even your earliest manuscripts are a century or more after the events. Anything could have been changed in that gap.”Show a response

It's a fair question, but the gap is far smaller than for almost any other ancient text, and it is not a blank. Several manuscripts are usually dated to the second and third centuries, and on its common dating P52 is closer still; for most classical authors the gap is many centuries, yet historians trust what they have. More importantly, copies spread quickly across the Roman world in many independent branches. To change the text wholesale, someone would have had to track down and alter copies in Egypt, Syria, Italy and elsewhere, and the early translations and church fathers' quotations would also have to match. Instead, we see early manuscripts like P75 closely agreeing with later ones like Codex Vaticanus, which points to careful transmission. Textual critics can see where copyists slipped and correct for it.

“Fine, maybe the text was copied accurately. That only proves we know what the authors wrote, not that what they wrote is true.”Show a response

You're exactly right, and I'm glad you made that distinction. Textual criticism answers one question: do we know what the New Testament originally said? The answer is yes, to a remarkable degree, and even sceptical scholars like Ehrman agree the essential teaching is not affected by the variants. Whether that teaching is true is a different question, and it needs other evidence. But the first question still matters, because it means we can examine what the apostles actually claimed rather than guessing. And what they claimed is public and testable, especially that Jesus rose from the dead and appeared to many witnesses (1 Cor 15:3-8). So reliable manuscripts don't end the conversation; they let it begin on solid ground. I'd be glad to look at the evidence for the resurrection with you next.

Written task

A friend shares a YouTube video claiming “the Bible has been changed so many times we have no idea what it originally said.” Write a 250-400 word reply that explains what textual criticism is, gives the key facts about New Testament manuscripts, answers the telephone game analogy, and fairly states what Bart Ehrman concedes.

Hand this in to Mr Scott. He will look for:

Gives accurate figures and at least three named manuscripts with approximate dates.
Explains the categories of variants and why the raw number is misleading.
Clearly shows why the telephone game analogy fails.
Represents Ehrman fairly, neither overstating nor understating his position.
Connects the evidence to God's providential preservation of his word.