Re: [textualcriticism] TR vs CT variants - variant count methodologies examined
- Sorry I didn't notice this previously. I'm on the first photographic expedition of the year, spending a month in Albania and Greece. Right now, I'm on the island of Lesbos. I'll be back for four days at the end of the month, before I hit the road again for 3 weeks.
When I wrote the piece about 1838 differences between the MT and TR (along with 6577 between the MT and Critical text as represented by the UBS 3/4 & NA 26/27), I looked strictly at the Hodges-Farstad text, second edition. Their apparatus has all places (allegedly) where the MT and Cr differ, as well as all places where the TR and MT differ. The MT, however, is not as pure an MT text as what Robinson and Pierpont have produced. I believe that they did not record movable nus as differences among the texts.
The statistics were not merely single words. The pericope adulterae and long ending of Mark, for example, constituted one difference between the critical text and the MT. But between the MT and TR, even though both have the PA, minor differences would each count as one variant.
As for Hodges and Farstad's 500 differences between the MT and TR, I think that depends on when they were asked about these. Before they completed the MT, their belief was that there would be between 500 and 1000 differences between the the TR and MT. I believe they both were surprised to learn that there were almost 2000 differences.
One other statistic should be noted: 657 MT shorter readings (i.e., shorter than the Cr text readings). Initially, I thought these were very promising as potentially authentic Byzantine readings--or, as Karavidopoulos put it, "Eastern Non-Interpolations" (picking up on WH's infamous phrase about potentially authentic Western readings). A doctoral student at Dallas Seminary (John Wu) examined all the shorter Byzantine readings in the Gospels for his dissertation and concluded that only one of them bore the marks of authenticity. The text in question is Matt 24.36. This came as no surprise since I had already held the Byzantine reading to be authentic there. Now, however, after another student of mine (Adam Messer) did work in the patristic commentaries on this passage, I think I am changing my mind about Byzantine authenticity here.
I wish I could engage in this discussion more, but I've got very limited access to the Internet, so that's unlikely for some time.
Daniel B. Wallace
Center for the Study of New Testament Manuscripts
----- Start Original Message -----
Sent: Sat, 14 Jun 2008 17:16:36 +0300
From: "ron minton" <ronminton@...>
Subject: Re: [textualcriticism] TR vs CT variants - variant count methodologies examined
>found is the six NTs: USB 3&4, NA26&27, H&F 1&2.
> Our method: We counted and cataloged every TR/CT variant that can be
On Wed, Mar 26, 2008 at 9:06 AM, Schmuel <schmuel@...> wrote:
> TR agrees with Maj.T 66.8%.
> Eddie Mishoe wrote:
> Dr. Daniel Wallace found 1,838 variants between HF (a MajT) and TR
(I gave wrong
> information last post and was properly corrected). I don't know
what % that represents,
> but your % here has 33.2% variants between the two. Are these
> the same? Sorry for my ignorance in this matter.
ron minton wrote:
We examined all the variants in USB 3&4, NA26&27, and HF1&2.
That is likely why we had more than the others.
Hi, Ron. It would be helpful if we knew the number of total
potential variants that were viewed by Daniel Wallace in reaching his
1,838 figure. He could be using the same 8,000 variants you use,
less as you conjecture, or he could be going down to the level of
letters and spelling and have 100,000 potential variants, or
something entirely different.
In fact, Thomas Strouse quotes the Daniel Wallace article "The
Majority Text Theory: History, Methods and Critique" in such a way
that the sense is that the number is over 100,000 ! And the Daniel
Wallace % results grossly different than yours (a point that can be
discussed more later, the question of alternate statistical
methodologies producing widely diverse results, and being quoted as
authoritative, with nary a notice in the textcrit world).
Fundamentalism and the Authorized Version - Thomas Strouse
The TR and the textus majoritas are found to be identical with
respect to 98.7% of their words and differ in 1,838 words, or about
1.3% (Daniel Wallace, "The Majority Text Theory: History, Methods and
Critique," Journal of the Evangelical Theological Society, June 1994,
One article that touches on the methodological issue is by Thomas
1,838 ... Others, such as Drs. Hodges and Farstad list only 500
(Ibid.). Both could be true if Wallace is counting difference such as
spelling (minor differences), while Hodges and Farstad count
differences in words and omissions (major differences)
Two text comparison numbers of absolute variants without methodology
specified are virtually meaningless. And even with methodology
specified, it would be proper to offer control numbers from known
alternate-related texts for the numbers to have meaning and not be
grossly misused. e.g. using the same methodology, the number of
differences between Aleph and B is x, or between the TR and NA27 is
And I may like to comment on these numbers more, but first I would
like to ask a question to yourself and Daniel Wallace.
What were your methodologies in a straight two-way comparison of MSS
of what is a variant vs. a match ? There are all sorts of potential
gray areas with spelling, word order, prepositions and conjunctions,
minor word differences that the methodology of determination is
primary, and the numbers have little meaning otherwise.
Please note that this is not the case in 3-way comparisons where the
totals add up to 100%. (e.g. in what % of places does the Vulgate or
Peshitta or Robinson-Pierpont MT or HF or Codex Alexandrinus agree
with the TR versus the NA-27 or vica versa on their variants.. in
such cases methodology is minor and results should always be similar,
the only differences will be a small number of cases where there is a
gray area between the three categories of agreement with one, the
other, or neither.).
However in two-way results where the numbers do not add to 100% ---
Methodology is king.
And without it, numbers have very little value.
Even with known methodology, unhinged numbers should be used with
as we learned back in the 1950's in the famous book written by
Darrell Huff on
----- End Original Message -----