Variants

A textual variant is any place where surviving manuscripts of the New Testament differ. Estimates run from about 300,000 to 500,000 differences across roughly 5,800 Greek manuscripts.

That figure exceeds the New Testament's word count, which is about 138,000 words in Greek. The reason is arithmetic: a single word differing in a thousand copies counts as a thousand variants.

This page covers what the differences consist of and how they are assessed. Passages judged to be later additions are treated separately in Interpolations.

What the variants consist of#

Bart Ehrman and Bruce Metzger both estimate the last category — differences that both change the meaning and have a genuine claim to being original — at well under one percent.

How scribes introduced changes#

Unintentional.

• Haplography: writing once what should be written twice
• Dittography: writing twice what should be written once
• Homoeoteleuton: the eye skipping from one line ending to a similar ending further down, omitting the text between
• Confusion of similar letters, particularly in the older uncial script written without spaces between words
• Errors of hearing, where a text was dictated to several copyists at once

Intentional.

• Harmonising a Gospel to its parallel in another Gospel
• Smoothing awkward grammar
• Clarifying an ambiguous pronoun by naming the person
• Expanding titles — "Jesus" becoming "the Lord Jesus Christ"
• Adjusting a reading that appeared theologically problematic

The last category is the contested one. Ehrman argued in The Orthodox Corruption of Scripture (1993) that some changes were made to strengthen positions that later became orthodox.

Examples of meaningful variants#

Two of these are informative about direction. ΟΣ ("who") and ΘΣ (an abbreviation for "God") differ by two small strokes, and the change at 1 Timothy 3:16 runs toward the stronger christological reading. The omission of "nor the Son" at Matthew 24:36 removes a statement that Jesus did not know something.

The Mark 1:41 case is contested in the other direction. "Anger" is the harder reading and the one a scribe would more plausibly soften, which is why some editors prefer it despite weaker manuscript support.

How readings are assessed#

Textual criticism weighs external and internal evidence.

External. The age of the manuscripts, their geographical spread, and their groupings. A reading found in early witnesses from separate regions is stronger than one confined to a single late family.

Internal. Two principles, applied with judgement rather than mechanically:

• Lectio difficilior potior — the harder reading is preferred, since scribes tended to simplify rather than complicate.
• Lectio brevior potior — the shorter reading is often preferred, since scribes tended to add explanatory material. This principle is now applied more cautiously than in the past.

Modern critical editions — Nestle-Aland and the United Bible Societies text — print the editors' preferred reading with an apparatus recording the alternatives and the manuscripts supporting each. The reasoning is published rather than hidden.

Common objections#

"400,000 variants means the text is hopelessly corrupt"

The number is accurate and it is also misleading without context.

Variants are counted across all manuscripts, so a single spelling difference appearing in a thousand copies contributes a thousand to the total. The count rises with the number of surviving manuscripts, which means a better-attested text produces a larger number.

The overwhelming majority are spelling, word order, and obvious slips. For most of the New Testament the wording is well established, and modern editions differ from one another very little.

"No essential doctrine depends on a disputed reading"

Broadly defensible and stated by scholars across the spectrum, including both Metzger and Ehrman.

It holds because doctrines rest on multiple passages, so removing one leaves the others standing.

It does not show the disputed passages are unimportant. 1 John 5:7 is the only verse stating the Trinity explicitly and is a later addition; 1 Timothy 3:16 in its later form is a direct statement that God was manifested in the flesh.

The claim is also compatible with some changes being theologically motivated, which is a separate point about how the text was transmitted.

"The sheer number of manuscripts lets us recover the original"

Comparison does allow errors to be identified and usually corrected, and this is why the New Testament's wording is better established than that of most ancient works.

What the method reconstructs is the earliest recoverable text, which is not necessarily the author's original. For most books the evidence begins in the third or fourth century, and the copying before that cannot be inspected.

D. C. Parker and others have argued that for works which circulated and were revised in their earliest period, a single original text may not be a coherent target.

"Ehrman exaggerates — he agrees the text is 99% certain"

Accurate. Ehrman has said repeatedly that the vast majority of variants are insignificant, and he worked on the standard handbook with Metzger, who held conservative conclusions.

His argument is not that the text is unrecoverable but that some changes were theologically motivated, which is a claim about scribal behaviour rather than about the extent of corruption.

That specific claim is contested. Some scholars accept parts of it; others hold that the changes are better explained by harmonisation and clarification than by doctrinal intent.

What the evidence shows#

The New Testament survives in more manuscripts than any other ancient work, and the large variant count is partly a consequence of that abundance.

Nearly all differences are spelling, word order, and copying slips. A small number change the sense, and a few of those touch passages of doctrinal interest.

Comparison across manuscripts identifies most changes, and the critical editions publish the alternatives and the reasoning rather than presenting a single text as settled.

What the method reaches is the earliest recoverable wording, not a verified copy of an original — and the gap between composition and the earliest substantial witnesses is not directly observable.

Related material: New Testament Manuscripts and Interpolations.