The Manuscript Treasure

The NT is far better attested than any other ancient document. The numbers are extraordinary.

5,800+
Greek MSS
10,000+
Latin MSS
9,300+
Other Ancient Versions
1M+
Patristic Citations

Compare this with major works of classical antiquity. Caesar's Gallic Wars: about 10 manuscripts, the earliest 900 years after Caesar. Tacitus's Annals: about 35 manuscripts. Plato: roughly 200 manuscripts, the earliest about 1,200 years after Plato. Homer's Iliad — the best-attested classical work — about 1,800 manuscripts.

The NT, with over 5,800 Greek manuscripts plus thousands of ancient translations, is in a different league. We could lose every single Greek manuscript and reconstruct nearly the entire NT from the patristic citations alone — over a million quotations of the NT from early church writers.

This is not because of any unusual circumstance. It is because the NT was treated from the beginning as Scripture worth preserving, copying, and translating into every language the gospel reached. The result is a manuscript tradition unmatched in the ancient world.

Major Manuscripts

The most important witnesses to the NT text. Every modern Greek New Testament rests on these.

𝔓⁵²Rylands Papyrus
c. AD 125 · Manchester · A fragment of John 18 — the earliest known NT manuscript

A small papyrus fragment containing portions of John 18:31–33, 37–38. Discovered in Egypt and now held at the John Rylands Library in Manchester. Its dating to roughly AD 125 — within a generation of John's likely composition c. AD 90 — sets a powerful constraint on any theory of late Gospel composition. If John was being copied and circulated in Egypt by AD 125, it was clearly written and accepted earlier.

𝔓⁴⁶Chester Beatty II
c. AD 175–225 · Dublin/Ann Arbor · Most of the Pauline epistles plus Hebrews

One of the earliest substantial NT manuscripts — a near-complete codex of the Pauline letters dating to the late second or early third century. Contains Romans, 1–2 Corinthians, Galatians, Ephesians, Philippians, Colossians, 1 Thessalonians, and Hebrews. The witness it provides to the Pauline corpus is extraordinarily early.

𝔓⁶⁶Papyrus Bodmer II
c. AD 200 · Cologny, Switzerland · Most of John's Gospel

An early codex containing most of John 1–14 and substantial portions of chapters 14–21. Around AD 200 — extraordinary closeness to the original. The careful production of this manuscript suggests John's Gospel was already an established and prized text.

𝔓⁷⁵Papyrus Bodmer XIV–XV
c. AD 175–225 · Vatican Library · Most of Luke and John

Contains substantial portions of Luke and John. Its text is remarkably similar to that of Codex Vaticanus (a 4th-century codex), demonstrating that the Vaticanus text is not a 4th-century revision but reflects a much earlier text-type.

אCodex Sinaiticus (aleph)
c. AD 350 · British Library, St Catherine's, Leipzig, St Petersburg · The complete Bible

Discovered by Constantin von Tischendorf at St Catherine's Monastery on Mount Sinai in the 1840s and 1850s. Contains the entire NT plus most of the OT (LXX) plus the Epistle of Barnabas and the Shepherd of Hermas (which were among the disputed books eventually not received into the canon). One of the two foundational pillars of modern NT textual criticism.

BCodex Vaticanus
c. AD 325–350 · Vatican Library · The complete Bible

Held in the Vatican Library since at least the 15th century, but only made widely available to scholars in the 19th. Contains nearly the entire OT and NT in Greek (the end of Hebrews and Revelation are missing). With Sinaiticus, the joint backbone of modern critical NT editions. Extremely high text-critical value.

ACodex Alexandrinus
c. AD 400–450 · British Library · The complete Bible (lightly damaged)

Slightly later than Sinaiticus and Vaticanus. Brought from Alexandria to England in the 17th century as a gift to Charles I. Contains nearly the complete Greek Bible. Its text in the Gospels reflects a slightly different (Byzantine) tradition; in Acts and the Epistles, it is closer to Sinaiticus and Vaticanus.

CCodex Ephraemi Rescriptus
5th century · Paris · A palimpsest — Bible text overwritten by sermons of Ephrem

A palimpsest: the original 5th-century Bible text was scraped off in the 12th century and overwritten with the works of Ephrem the Syrian. Modern technology has made the underlying biblical text largely recoverable. A reminder that some manuscripts have been preserved by accident as much as by design.

DCodex Bezae Cantabrigiensis
c. AD 400 · Cambridge · Greek-Latin diglot of Gospels and Acts

A bilingual manuscript with Greek on the left page and Latin on the right. Famous for its distinctive 'Western' text, particularly in Acts, where it is roughly 10% longer than the standard text and contains fascinating additional details. Whether these represent original readings or later expansions is one of the open questions in textual criticism.

Text Types

Manuscripts cluster into 'families' based on shared readings. Recognizing the family helps weigh testimony.

NT manuscripts can be grouped into broad text-types based on their readings. Three are classically recognized:

Alexandrian — manuscripts originating from or copied in Alexandria, Egypt. Generally regarded as the earliest and most carefully preserved text-type. Sinaiticus, Vaticanus, P75, P66 are key witnesses. Modern critical editions lean heavily on the Alexandrian text.

Byzantine — the dominant text-type of the medieval Greek-speaking church. Codex Alexandrinus (in the Gospels) is an early witness; the great majority of Greek manuscripts are Byzantine. Smoother and more harmonized in some places, it tends to expand and clarify rather than abbreviate. Underlies the Textus Receptus from which the King James Version was translated.

Western — represented chiefly by Codex Bezae and the Old Latin translations. Distinctive for paraphrastic expansion and unique readings. Acts in the Western text is roughly 10% longer than in other text-types.

A fourth, the Caesarean, is sometimes proposed but is now contested. Modern textual criticism increasingly treats text-types as overlapping rather than discrete categories.

The eclectic method dominant in modern scholarship does not simply prefer one text-type but evaluates each variant on its own merits, considering external evidence (which manuscripts have which reading) and internal evidence (which reading best explains the rise of the others).

The Famous Variants

A handful of textual variants are theologically substantial and well known. Modern translations flag them all.

Mark 16:9–20 — The Longer Ending of Mark
The two oldest and best manuscripts (Sinaiticus and Vaticanus) end Mark at 16:8 — 'they were afraid.' The longer ending (16:9–20) appears in most later manuscripts but is regarded by virtually all textual scholars as a later addition by a scribe troubled by Mark's abrupt ending. Modern translations include it but flag it as not in the earliest manuscripts. The longer ending contains the snake-handling passage often misused in fringe Pentecostal practice; this is yet another reason its dubious textual status matters.
John 7:53–8:11 — The Pericope Adulterae
The famous story of the woman caught in adultery is missing from the earliest manuscripts of John (P66, P75, Sinaiticus, Vaticanus, etc.). It floats in the manuscript tradition — sometimes appearing after John 7:52, sometimes after John 21, sometimes after Luke 21:38, sometimes omitted entirely. Most scholars conclude it is genuine apostolic tradition (it has the ring of Jesus' teaching) but was not part of John's original Gospel. Modern translations typically include it with a note.
1 John 5:7–8 — The Comma Johanneum
'For there are three that bear record in heaven, the Father, the Word, and the Holy Ghost: and these three are one.' The classical Trinitarian proof-text in 1 John — but it appears in only a tiny handful of late Greek manuscripts and is absent from all the early Greek witnesses. It entered the Latin Vulgate tradition late and from there into the Textus Receptus and the King James Version. Modern translations omit or footnote it. The Trinity is well-established without this verse; the doctrine does not depend on a textually disputed reading.
1 Timothy 3:16 — 'God was manifest in the flesh' vs. 'who was manifest'
A small variant with christological weight. The Textus Receptus (and KJV) reads 'God was manifested in the flesh.' The earliest manuscripts read 'who was manifested in the flesh' (referring to Christ). The difference in Greek is between theos (God) and hos (who) — a single letter in cursive script. The 'who' reading is older and almost certainly original; it still affirms Christ's incarnation but does not include the bare word theos. The doctrine of Christ's deity rests on many other texts.
Acts 8:37 — The Ethiopian's Confession
'And Philip said, If thou believest with all thine heart, thou mayest. And he answered and said, I believe that Jesus Christ is the Son of God.' Present in the KJV; absent from the earliest manuscripts. Likely a later liturgical addition — the kind of confessional formula a baptizer might have added by the second century. Modern translations omit or footnote it.

The pattern is consistent. The substantive textual variants in the NT — the ones that change meaning rather than just spelling — are few, well-known, and clearly flagged in any modern translation. None affects a doctrine of Christianity that is not also taught elsewhere. The doctrine of the Trinity, the deity of Christ, the resurrection, justification by faith — all stand firm whether or not the disputed verses are original.

Why So Many 'Variants'?

Bart Ehrman's Misquoting Jesus popularized the figure of '400,000 variants' in the NT manuscripts. The number is technically correct but profoundly misleading.

The number is enormous because the manuscript base is enormous. Every single spelling difference between every single manuscript counts as a 'variant.' If a single manuscript spells 'John' differently in three places, that contributes three variants to the count. Multiply across 5,800 Greek manuscripts and the figure climbs quickly.

Of these 400,000+ variants:

• The vast majority are spelling differences in proper names (Iōannēs vs Iōanēs, etc.) — meaningless for translation.

• A great many are word-order variations — meaningless in inflected Greek where word order does not change meaning.

• Many are missing or added articles — meaningless in translation.

• Many are obvious copyist errors — duplicated words, dropped lines, easy to detect.

The variants that are both meaningful (would change meaning if adopted) and viable (have substantial manuscript support) are roughly 1% of the total. That is to say, in the entire NT, perhaps a few thousand variants are genuinely in question — and as we have seen, of those, only a handful affect anything theologically substantial. The rest is what Daniel Wallace has called 'noise without signal.'

The honest summary: the NT text is in remarkably good shape. We can be confident that what we read in any modern critical translation closely reflects what the apostolic authors wrote — closer than for any other ancient document.

Modern Critical Editions

The standard scholarly Greek New Testaments today are:

Nestle-Aland (NA28) — the 28th edition (2012) of the critical text begun by Eberhard Nestle in 1898. The most widely used scholarly edition. Includes a comprehensive critical apparatus showing variants and their manuscript support.

UBS5 — the United Bible Societies' Greek New Testament, 5th edition. Same base text as NA28 but with a smaller apparatus designed for translators rather than textual critics.

Tyndale House Greek New Testament (THGNT) — 2017. A more conservative edition that gives greater weight to the earliest manuscripts and aims to recover the format of early Christian writing.

The Robinson-Pierpont Byzantine Text — represents the majority Byzantine text-type. Used by those who prefer the dominant medieval text.

The Textus Receptus — the printed Greek text underlying the King James Version, compiled by Erasmus in 1516 and refined by later editors. Now superseded by modern critical editions but historically significant.

Most modern English translations (NIV, ESV, NASB, NRSV, NKJV with footnotes, CSB) translate from one of these critical editions. The KJV translates from the Textus Receptus. The differences between KJV and modern translations on textual matters are real but small — and clearly flagged in any responsible translation.

Test Your Understanding

Two quizzes covering the manuscript witness and the textual variants. Test your grasp of why we can trust that we have the New Testament the apostles wrote.

§1 — The Manuscript Witness
§2 — Variants and Reliability