Textual Evidence

D.A.Carson &

Excerpt from: D. A. Carson, The King James Version Debate, (Baker House, Mich., 1979)

Page Index

    Most Changes are Errors - Carson's Thesis 9
    A Non-Credible Theory - what is required to classify omissions as additions
    Reality Check - what needs to be admitted and corrected
    Carson's Example: Tischendorf's Apparatus and missing footnotes

Return to Index

D.A.Carson &


The following is taken from:
Nazaroo, Stupid Gaffs in the UBS Greek Text - Haplography
Forum thead, Theology Online Religion section, 2009

Headings have been added for clarity and navigation purposes.


Evidence for the Byzantine Text

Even a more liberal scholar such as D.A. Carson, (who is no friend of the KJV or TR), confesses the force of Haplographic errors.

In fact, he depends upon them to support his Thesis 9, in his book,
The King James Version Debate:

Thesis 9: 'The charge that non-Byzantine text-types are theologically aberrant is fallacious.'

... "The second thing to note is the fact that the omission of an individual title or phrase or verse does not constitute evidence for theological heresy. Perhaps the omission was part of the original, and the MSS that include the title or phrase or verse are guilty of additions.

Beyond that, however, even if the omission were not part of the autograph, the omission per se would not prove heterodoxy. One would have to ask why the omission had taken place.

The reason might prove to be nothing more than homoeoteleuton. 23 ...

What we discover in practice is that most changes or omissions (or additions, depending on the point of view) are "quite trivial and wholly devoid of theological significance." 24

Where they do affect the meaning of a passage in such a way that the passage can no longer be called upon to support a particular doctrine, nevertheless the doctrine itself remains unchanged because it is still supported by many other passages in the same textual tradition [text-type].

One must conclude, therefore, that if a heretic changed the particular passage in order to suppress some specific doctrine, then he was singularly incompetent if he did not systematically change all the other placed in the document that supported the same doctrine.'

Carson's Original Footnotes:

23. For example, because the last 5 words of Luke 14:26/27 are exactly the same in Greek, it is easy to understand the omission of the latter verse in about a dozen MSS.

24. Donald MacLeod, "The Bible and Textual Criticism" BanT no. 111 (1972) 12-26.

- D.A. Carson, , p. 62-63 (Baker, 1979)

Here Carson admits that most variants are devoid of theological influence, and have all the appearance of random, undirected, unmotivated accidents.

He even gives a clear example of homoeoteleuton: Luke 14:26/27

Its obvious that he is willing to concede that in spite of multiple manuscript support, this is an error of omission by homoeoteleuton.

But once accident is conceded, we are not free to chose between omission and addition in most of these cases. For most variant units with haplographic features do NOT have the features of repetition by dittography.

And it is statistically absurd to suggest that most cases of addition would have perfect features of homoeoteleuton/arcton (a false appearance of omission) when they should have the ordinary appearance of dittographic duplication instead.

A Fantastic New Theory...

The only possible way that this large collection of apparent homoeoteleuton/arcton errors (omissions) could really be dittographic (additions) would be through a second deliberate editing process:

This process would have to have:

(a) found simple dittographic errors, then changed their appearance to make them read better in the editors' minds, and

(b) coincidentally in every case introduced new homoeoteleuton/arcton features,

(c) while obliterating the dittographic features. At the same time,

(d) this deliberate editing process would have to use this consistent policy of handling the majority of dittographic errors in an absurd manner,

(e) without any other identifiable, unified theological or doctrinal direction.

Historically however, there is no such process or practice known to have been practiced. All ancient editing had a rationale, usually to correct errors or previous edits; editors did not get creative. And such activity was directed, through simple and explicable guidelines; it wasn't free-form literary art.

But once again, if you need to posit a deliberate editing process to explain things, you don't need dittographic errors as a 'starter' for every variant, nor is there any reason to think these would be involved.

You might as well just say an editor deliberately changed the text without any prompting by what should have been obvious dittography errors. Why would such free-flying editors need an excuse to emend the text anywhere he chose? At the same time, such editors would have continued to correct dittography errors in the normal way (deletion).

And if we postulate that early editors were really this crazy and stupid, why not then admit that no choice of reading can be more plausible than another. There can be no credible, unified explanatory account for uncontrolled caprice, so we might as well look upon such free-texting as 'random changes of the unknown kind'. In that case, tossing a coin to choose the correct reading will be as good a method as any.

Facing up to Reality...

The sensible (and only plausible) option is to accept the fact that variation units that look like homoeoteleuton/arcton ARE homoeoteleuton/arcton, and units that look like dittography ARE dittography, and that the statistics resulting from this classification reflect real scribal tendencies and errors.

Carson should just admit that the nearly 100 obvious homoeoteleuton/arcton errors mistakenly adopted by Westcott/Hort and NA27/UBS4 are really exactly what they look like: errors of OMISSION by homoeoteleuton/arcton.

What Modern Translators should do...

The translators of modern NT versions should also take serious heed that what is a proper procedure for reconstructing an early intermediate exemplar or lost ancestor is NOT an appropriate procedure for reconstructing the text of the NT.

The proper procedure is to mark and catalog early scribal errors and leave them in the margin of Greek reconstructed texts and technical papers, and not introduce them into the text of modern Bible translations.

Carson's Example:
Luke 14:26-27

By the way, searchers will look in vain for any footnote or apparatus anywhere that references and discusses Luke 14:27.

However, if we turn to Tischendorf's 1877 edition of the Greek NT (p 290-291), we will find a relatively lengthy footnote as follows:

Luke 14:26-27
(Tischendorf's main 1877 text: )

26....ου δυναται ειναι μου μαθητης.
27. οστις ου βασταζει τον σταυρον
εαυτου και ερχεται οπισω μου,
ου δυναται ειναι μου μαθητης.

Versum om M*(9th cent.) R(7th cent.) Γ(9th cent.) al mu (alii multi)

(om/add kai: )
| οστισ sine copula cum א* L cop (accedunt B* aeth)
...ς Ln Ti και οστισ cum אc A (D) X Δ Λ Π unc9 al fere omn it7 vg syrcu etutr
... οστ.γαρ 157. arm

(+ substitution: )
| εαυτου c. A B L2 M2 Δ al10
...ς αυτου c. א D L* Λ Π unc8 al pler B a sbapt628 : : ut Mt.

(+ word order reversal: )
| ειν. μου μαθ. cum א Β Ε F G H L S V X Λ Δ etc.
...ς Ln μου ειν. μαθ. cum A K M2 U Π al plu c ff2 vg (et. am*)

Ln: Lachmann 1842/1850 ed.
Ti: Tischendorf's previous 1859 ed.
ς: Elzev. 1624 ed., Steph. 1550 ed.: when differing, ς = Steph ς* = Elz.


This is what Carson is referring to, about a handful of MSS omitting this verse; and its also an excellent example of why the UBS and NA27 critical apparatus is near-useless in determining what the manuscripts really are doing.

W. Willker's online Textual Commentary also fails to even note this variant, which is nonetheless important because it gives witness both to the scribal habits of real copyists and the frequency and non-reliability of omissions in certain witnesses and groups.

Why was this variation unit left out, and no notice given in the NA/UBS apparatus?

Because it was recognised as an obvious homoeoteleuton error of omission.

Why then were 50 or so other similar variation units included, and the omissions actually followed, even though the exact same features of an obvious homoeoteleuton error of omission are found there?

Because these haplography errors occurred in the ancestor of א & B!

What a double-standard, and what a coverup.

Another reason that readings like these are left out of the critical apparatus is that they go against the current theory regarding the Byzantine text-type. Critics claim that this text tends to "conflate", "insert", "expand" the text.

But evidence like this (and there is plenty more) shows that contrary to claims, the Byzantine scribes had the same tendencies as all other scribes, namely a tendency to OMIT, not ADD.

Hiding this crucial evidence regarding Byzantine copyists again obscures the truth in favour of the Alexandrian text-type (Aleph/B).

Return to Index