Fri Nov 29 10:14:34 1996

From owner-tc-list  Fri Nov 29 10:14:34 1996
Return-Path: 
Received: by scholar.cc.emory.edu (SMI-8.6/SMI-SVR4)
	id KAA13119; Fri, 29 Nov 1996 10:13:35 -0500
X-Sender: waltzmn@popmail.skypoint.com
Message-Id: 
In-Reply-To: 
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Date: Fri, 29 Nov 1996 09:10:08 -0700
To: tc-list@scholar.cc.emory.edu
From: "Robert B. Waltz" 
Subject: Re: Cesarean Text
Sender: owner-tc-list@scholar.cc.emory.edu
Precedence: bulk
Reply-To: tc-list@scholar.cc.emory.edu
content-length: 4918

On 29 Nov 96, "Jean Valentin"  wrote:

>Hello all!
>
>Am I having problems with my connections, or is it really so quiet on
>the list these last days? :-)

Remember that November 28 was a holiday in the United States. Also,
it seems like disputes on this list always die off after a few
messages... whether they've been settled or not.

So let's start a new arguement!

>So tell me: what do you have to say about the so-called "cesarean
>text"? Some scholars deny it even existed (I think Kurt Aland was of
>this opinion), others seem to make it very important - specially
>people who publish oriental versions, so I might be concerned with my
>Arabic.
>
>So what is this all about?

The basic problem with the "Caesarean" text, as defined by Lake, Blake,
New, and Streeter, was that it was defined in terms of divergences
from the TR. This is always a bad procedure, but it was *particularly*
bad in the case of the "Caesarean" text, because family 1 is considered
a member of that text -- and 1 was also used to compile the TR. The
net result was that Streeter, in particular, labelled *everything*
"Caesarean." (Streeter, for instance, considers effectively all of
Soden's I group, except D, to be weak "Caesarean.")

The finding of the "Caesarean" text led everyone to try to find more
witnesses. So, for instance, Kenyon classified p45 (which Colwell
has shown to be a wild text) as "Caesarean."

Once this mad rush was over, people sat back and looked to see what
they had found. And what they found was problems.

One of the most important steps was Hurtado's study (which we have
argued about on this list). This work, if nothing else, proved that
there is no link between p45/W and the rest of the "Caesarean" text.

This left a problem, because it meant that there is no pure "Caesarean"
witness. If one applies the Colwell 70% criterion, the "Caesarean"
text ceases to exist.

I think people will agree up to this point. It's what follows that
gets complicated. The question is, "Can a text-type exist without
a pure representative?"

The answer, if one follows the Colwell definition, is NO. (This is
one of the reasons I don't like the Colwell definition. It simply
does not allow for mixture. Whereas we know that mixture is almost
universal among manuscripts that are not purely Byzantine.)

But the "Caesarean" text was not defined in terms of Colwell; it was
defined in terms of places where *certain manuscripts (Theta, fam 1,
fam 13, 28, 565, 700, arm, geo) diverged from the Byzantine text.*
(The fact that Lake et al used the TR to represent the Byzantine
text made their results invalid, but it does not influence their
definition.) This is a much harder definition to apply, but also
could bear more fruit.

Let me demonstrate the difficulty. Take two manuscripts, A and X.
Let's assume that they are both Caesarean. Let's even assume that
they are sisters, and agree absolutely.

Now let's subject them to heavy correction against a Byzantine text. Assume
that half the "Caesarean" readings of each are conformed to the Byzantine
text. However, by the nature of things, each will receive a *different*
set of corrections.

Now let's copy A, as corrected, to produce B. Let's copy X, and corrected,
to produce Y.

Both B and Y are 50% Byzantine and 50% "Caesarean." But -- and this is
important -- they do not agree very often. In fact, they will agree, on
average, *only 50% of the time.* To be precise, in 25% of the cases,
both will retain the original "Caesarean" reading. In 25% of the cases
both will have the same Byzantine correction. In the remaining 50% of
variants, one will be "Caesarean" and the other Byzantine.

I would maintain, therefore, that the only was to truly examine the
nature of the "Caesarean" text is to look at the cases where *both*
manuscripts are *non-Byzantine.* Only there can we safely assume that
they are not mixed. This, obviously, means that we don't have as
much data available. Even the best "Caesarean" texts (Theta, family 1)
appear to have lost half or more of their "Caesarean" readings, and
most of the others are worse (28, e.g., is "Caesarean" only in Mark,
and 565, while it has some early readings outside Mark, is really
strong only in that book).

Still, it would appear that, if we follow this procedure, the "Caesarean"
text *does* exist. (I think; recall that I've spent most of my time on
Paul and the Catholics, and my results for the Gospels are incomplete.)

But note that this all depends on definitions. Anyone who accepts the
Colwell definition will say that the "Caesarean" text does not exist.
By the Lake definition, it does. Until we can agree on a definition
of a text-type (or find a pure witness), the "Caesarean" text will
remain in limbo.

If anybody wants to see more of my opinions on the matter, check my
article on text-types at

http://www.skypoint.com/~waltzmn

Bob Waltz
waltzmn@skypoint.com

Back