Wed Feb 5 23:54:13 1997

From owner-tc-list  Wed Feb  5 23:54:13 1997
Return-Path: 
Received: by shemesh.scholar.emory.edu (SMI-8.6/SMI-SVR4)
	id XAA06611; Wed, 5 Feb 1997 23:54:13 -0500
Date: Wed, 5 Feb 1997 23:53:50 -0500 (EST)
From: Maurice Robinson 
To: tc-list@shemesh.scholar.emory.edu
Subject: Re: Majority mss
In-Reply-To: 
Message-ID: 
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-tc-list@shemesh.scholar.emory.edu
Precedence: bulk
Reply-To: tc-list@scholar.cc.emory.edu
content-length: 2644



On Wed, 5 Feb 1997, Ronald L. Minton wrote:

> I was cruising the textual criticism web sites today, and 
> I came up with a question that I had been asked last year.  How many 
> manuscripts can we say for certain are represented by the 
> Robinson-Pierpont and the Hodges-Farstad texts?

Based upon the data cited by Von Soden regarding his "K" MSS (and
eliminating the MSS cited only cursorily) as well as comparing such with
the Nestle-Aland "M" data cited in the appendix to the 26th or 27th
edition, one could say that a solid "K" or "M" reading generally would
reflect a consensus of around 500 or so MSS -- quite sufficient
statistically for establishing the dominant Byzantine Textform in most
instances. 

I have an unpublished paper ("How Many MSS are Necessary to Establish the
'Majority Text'?") which was distributed by the Majority Text Society some
years back, in which I demonstrated that by using the consensus reading of
only 13 MSS randomly selected in a sample chapter one would approach
either the Von Soden "K" or the Nestle-Aland "M" with approximately 96%
certainty, and that every MS added to such a pool would increase the
percentage of agreement in moving towards the desired goal.

Statistically I suspect that any random sampling of 100 MSS or more would
suffice to produce a strongly Byzantine consensus text which would
probably approach "K" or "M" with 98% or more certainty. If approximately
500 MSS were used as the base (i.e., following the data of Von Soden and
Nestle-Aland), I suppose 99%+ certainty should be expected in establishing
beyond doubt nearly all of the primary readings of the Byzantine Textform
(math and statistics majors out there might help me out on that aspect;
someone did send me some statistical data on this point after my article
was distributed, but I confess an inability to comprehend chi-square and
such other niceties which lie outside of my own area of specialization).

I should also note that at least in the case of the Robinson-Pierpont
edition the _Text und Textwert_ series as well as IGNTP/Luke and the
earlier IGNTP/Legg were also consulted.  Of these, _Text und Textwert_ of
course provides *all* the continuous-text Greek MS data for the variant
cited, usually averaging around 500-600 MSS in the portions of Acts, Paul,
and the General Epistles so far published.

_________________________________________________________________________
Maurice A. Robinson, Ph.D.           Professor of Greek and New Testament
Southeastern Baptist Theological Seminary     Wake Forest, North Carolina
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Back