VARIANT READINGS · MANUSCRIPT HANDS · CRITICAL EDITIONS

A passage through the manuscripts

One passage. Several witnesses. Learn to see what changed.

A reader’s introduction to textual criticism. Three real cases from the Greek New Testament, from a spelling to a disputed ending: sit at the desk, follow each witness to the reading it supports, and keep the object, the transcription and the edition apart.

PASSAGES

WITNESSES COMPARED

11

OPEN RECORDS

1,138

OLDEST COPY

c. 175–225

WRITTEN TO COPIED

120–177 yrs

MARK’S OLDEST

c. 250

SOURCES

26

01 · PROLOGUE · BEFORE COMPARING

Keep the layers apart.

Textual criticism compares surviving witnesses to investigate a text’s transmission and earlier wording. These three Greek New Testament cases range from spelling to a substantial disputed passage. No Greek is required: each extract has an English gloss.

  • 01 The object Ink, a correction, an intact ending or a broken edge.
  • 02 The transcription A scholar’s record of the marks, with conventions for loss and uncertainty.
  • 03 The edition An editor’s chosen text, with an apparatus recording alternatives.

02 · CHAPTER I · GALATIANS 1:3

Spelling & order. Small differences, different kinds.

Paul’s opening greeting wishes grace and peace to the churches in Galatia. Start with the spelling of “to you”, then move “our” through the sentence.

291 dated open records hold Galatians. Point at a century.

  • 01 The readings
  • 02 Read the apparatus
  • 03 How an editor reasons

GALATIANS 1:3

Small differences. Different kinds.

Paul’s opening greeting wishes grace and peace to the churches in Galatia. Start with the spelling of “to you”, then move “our” through the sentence.

COMPARE A DIFFERENCE

  • P46 Papyrus 46 MANUSCRIPT χάρις ὑμεῖν καὶ εἰρήνη Grace to you and peace …
  • SBL SBLGNT CRITICAL EDITION χάρις ὑμῖν καὶ εἰρήνη Grace to you and peace …
  • NA28 Nestle–Aland 28 CRITICAL EDITION χάρις ὑμῖν καὶ εἰρήνη Grace to you and peace …
  • ὑμεῖν 1 manuscript
  • ὑμῖν 2 editions

Manuscript evidence Editor’s text

Michigan · diplomatic transcription↗ (opens in a new tab) SBLGNT · Galatians text↗ (opens in a new tab) SBLGNT · Galatians apparatus↗ (opens in a new tab)

03 · CHAPTER II · ROMANS 5:1

A changed vowel. One letter, a different force.

After speaking of justification by faith, does Paul state that “we have peace”, or urge “let us have peace”? Compare the first writing with the corrections.

298 dated open records hold Romans. Point at a century.

  • 01 The readings
  • 02 Three questions
  • 03 How an editor reasons

ROMANS 5:1

One letter. A different force.

After speaking of justification by faith, does Paul state that “we have peace”, or urge “let us have peace”? Compare the first writing with the corrections.

INSPECT THE MANUSCRIPT HANDS

  • א* Codex Sinaiticus MANUSCRIPT εἰρήνην ἔχωμεν Let us have peace.
  • B* Codex Vaticanus MANUSCRIPT εἰρήνην ἔχωμεν Let us have peace.
  • SBL SBLGNT CRITICAL EDITION εἰρήνην ἔχομεν We have peace.
  • ω 2 manuscripts
  • ο 1 edition

Manuscript evidence Editor’s text

NET · Romans 5:1, note 4 (PDF p. 9)↗ (opens in a new tab) SBLGNT · Romans text↗ (opens in a new tab)

04 · CHAPTER III · MARK 16:8–20

A disputed ending. Twelve verses, several histories.

The women flee the empty tomb in fear. Some witnesses end at 16:8; others continue with appearances, a commission and an ascension. First check what each object actually preserves.

932 dated open records hold Mark. Point at a century.

  • 01 The readings
  • 02 Read the join
  • 03 How an editor reasons

MARK 16:8–20

Twelve verses. Several histories.

The women flee the empty tomb in fear. Some witnesses end at 16:8; others continue with appearances, a commission and an ascension. First check what each object actually preserves.

Presence, absence and loss are different kinds of evidence.

  • א Codex Sinaiticus MANUSCRIPT ἐφοβοῦντο γάρ · ends at 16:8 “For they were afraid.” Then the end-title; Luke follows.
  • B Codex Vaticanus MANUSCRIPT ἐφοβοῦντο γάρ · ends at 16:8 The ending is followed by a blank column.
  • A Codex Alexandrinus MANUSCRIPT 16:8 → 16:9–20 present Continues with the Longer Ending.
  • P45 Papyrus 45 NO SURVIVING TEXT Mark 16 not extant No reading to compare at this location.
  • SBL SBLGNT CRITICAL EDITION 16:8 → ⟦16:9–20⟧ The Longer Ending is printed within double brackets.
  • ends at 16:8 2 manuscripts
  • 16:9–20 present 1 manuscript
  • no reading no leaf survives here
  • ⟦16:9–20⟧ 1 edition

Manuscript evidence Editor’s text Not extant ≠ omitted

Sinaiticus · BL folio 228↗ (opens in a new tab) CSNTM · P45 folio inventory (PDF)↗ (opens in a new tab) SBLGNT · Mark 16↗ (opens in a new tab)

05 · AT THE BROKEN EDGE · FOUR KINDS OF EVIDENCE

Not every gap is the same gap.

Presence, absence, loss and restoration are different kinds of evidence. Start at the torn edge of P46, then test the other three on real cases.

P46 · GALATIANS 1:4 · LINE 8

A word can be restored. The ink is still missing.

In Galatians 1:4, Michigan’s diplomatic transcription marks lost letters in the word for “will”. Turn on the restoration and watch the boundary between evidence and reconstruction.

Three letters are recorded here. A gap is not evidence that the scribe meant to leave a shorter word.

Michigan · diplomatic transcription↗ (opens in a new tab)

READING THE MARKS

One word, three records.

  • THE INK θελ What the papyrus itself still shows: three letters, then the break.
  • THE DIPLOMATIC TEXT θελ[ημα A close transcription, as in papyrological publications: no accents, the bracket where the ink stops, and the letters after it supplied.
  • THE PRINTED TEXT θέλημα Accents and breathings, as in a modern edition. SBLGNT prints the whole word.

IN ITS VERSE · GALATIANS 1:4

κατα το θελ[ημα του θ(εο)υ και π(ατρο)ϲ ημων

κατὰ τὸ θέλημα τοῦ θεοῦ καὶ πατρὸς ἡμῶν

according to the will of God our Father

Michigan · the opening text of Galatians↗ (opens in a new tab) Michigan · nomina sacra in P46↗ (opens in a new tab) SBLGNT · Galatians text↗ (opens in a new tab)

06 · THE WORK BEHIND THE PRINTED LINE

An edition is an argument that can be revised.

The goal here is the earliest recoverable text of the Greek New Testament. It is not a survey of Hebrew Bible or Septuagint textual criticism, whose editorial histories and questions require their own close studies.

1857–1881

Tregelles’s edition (1857–1879) and Westcott–Hort (1881) remain visible in SBLGNT’s apparatus. They are historical editorial arguments, not additional manuscripts. Romans 5:1 shows why later editors sometimes depart from them.

Holmes · how SBLGNT was edited↗ (opens in a new tab) SBLGNT · Romans apparatus↗ (opens in a new tab)

WHY A COMPUTER DOES NOT SIMPLY COUNT THE COPIES

No autograph of any New Testament writing survives. The earliest papyrus fragments date from the second century, the first complete copies (Sinaiticus and Vaticanus) from the fourth, and most of the roughly 5,700 manuscripts from the second millennium. The text at the start of the tradition, the initial text, has to be reconstructed.

Gerd Mink’s Coherence-Based Genealogical Method (CBGM), developed at the INTF for the Editio Critica Maior, compares the states of text that manuscripts carry, not the books themselves. A late manuscript can preserve an early text; scribes can combine sources.

At each place where witnesses differ, editors judge which reading most likely gave rise to the others: a local stemma. The method then combines thousands of these local judgments into a picture of how whole texts relate.

Where text X more often has the earlier reading than text Y, X is a potential ancestor of Y. Coherence is how closely two texts agree, together with which way the flow mostly runs between them. A textual flow diagram joins each witness to its closest potential ancestor.

The method does not draw stemmata automatically or make textual decisions by itself. Coincidental agreement, mixed ancestry and lost witnesses complicate reconstruction, so the results are hypotheses about texts: inspectable, not infallible. No CBGM calculation is performed by this study.

Münster INTF · genealogical method↗ (opens in a new tab) Gerd Mink · texts, mixture and genealogy↗ (opens in a new tab)

Illustrative schematic of a textual flow diagram · not a real calculation. Drawn for this study after the kind of diagram the INTF publishes; the sigla are examples and every link is invented.

Each line joins a witness to its closest potential ancestor: the text that, where the two differ, more often has the reading judged earlier. The stronger the coherence (agreement, plus which way the flow runs), the closer the link. Choose a witness to light its path to A.

Closest potential ancestor Carries the new reading Second source (mixture)

07 · THE WITNESSES ON THE DESK

Six objects, one study.

Every manuscript named above, from the open records. Choose one to open its record: date, material, contents, where it is now and its sources.

08 · KEEP INVESTIGATING

Follow the evidence further.

READ FURTHER

ON THIS SITE

WHAT REMAINS TO BE CHECKED

Romans’s correction labels follow a published apparatus; the hands have not been freshly collated here. The GA 304 investigation follows Houghton’s published analysis. Full ECM volumes, the complete translated-witness evidence and Monier’s inaccessible 2019 paper remain research gaps.

Before choosing a reading, ask what kind of evidence you have.

Rooted in Scripture. Open for discovery.

A place to read slowly, ask deeply, and follow the connections. Scripture, study and the voices of Christian history, brought together for a lifetime of learning.

SPEND TIME IN THE WORD

  • Open the reader
  • Books & reading progress
  • Compare Scripture editions
  • Find a passage

FOLLOW YOUR QUESTIONS

  • Study collections
  • People & places
  • Faith & apologetics
  • Stories of faith

A library with a paper trail.

Meet the authors. Explore the collection. Trace each work to its source.

Download the code.

Open source on GitHub.