Fri May 31 12:37:07 1996

From owner-tc-list  Fri May 31 12:37:07 1996
Return-Path: 
Received: by scholar.cc.emory.edu (SMI-8.6/SMI-SVR4)
	id MAA09787; Fri, 31 May 1996 12:35:56 -0400
Date: Fri, 31 May 96 11:32 CDT
X-Sender: waltzmn@popmail.skypoint.com
Message-Id: 
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
To: tc-list@scholar.cc.emory.edu
From: waltzmn@skypoint.com (Robert B. Waltz)
Subject: Re: "Alexandrian" Text
Sender: owner-tc-list@scholar.cc.emory.edu
Precedence: bulk
Reply-To: tc-list@scholar.cc.emory.edu
content-length: 4703

On or about Wed, 29 May 1996, schmiul@uni-muenster.de wrote:

>On Tue, 28 May 1996, Larry W. Hurtado wrote:
>
>> Second, discussion of "text-types" raises the question of what we >mean.
>>Here
>I recommend *strongly* Epp's essay on "The Significance >of the Papyri...A
>Dynamic View of Textual Transmission," in the >Epp/Fee volume, _Studies in the
>Theory & Method of NT Textual >Criticism_ (pp. 274ff.).
>
>Epp identifies in this essay three text-types being in existence already
>in the
>second century: a "B" text-type (P75 + B as nucleus), a "C" text-type (P45 + W
>only in Mark), and a "D" text-type (P29, P48, P38, 071, and perhaps P69 + D).
>"(T)hese concentrations or lines of trajectory identify clusters that in turn
>differentiate themselves sufficiently from other clusters" (p. 297).
>In defining a cluster or text-type Epp refers to Colwells famous quantitative
>definition ("A text-type is a group of manuscripts that agree more than 70
>percent of the time and is seperated by a gap od about 10 percent from its
>neighbors."), and in the case of highly fragmentary MSS to the Alands'
>principle: "If a fragment preserves a passage where there is any variation in
>the tradition, it is quite sufficient to signal the textual character of the
>whole manuscript."
>
>With these essentials in mind I may shortly assess Epp's text-types "C"
>and "D".
>
>a.) Text-type "C" (P45 + W): "Yet the affinity of variations-units between P45
>and W in Mark virtually reaches the seventy percent mark (68,9%)" (Epp, p.
>292).
>In my words: ...shortly MISSES the seventy percent mark. What do definitions
>serve for, if they are not taken serious?

I must insert here a strong argument *against* the Colwell definition of a
text-type. Two, in fact.

The first is the use of exact numbers in deciding text-types. Unless we are
prepared to examine *all* variations in the manuscripts involved (as Hurtado,
to his credit, did), the numbers mean nothing. If we use a comparison based
on sampling, the rate of agreement between manuscripts will depend on the
sample being taken.

To show what I mean, let's take a sample. I'm going to use a biased one to get
my point across.

Take as our sample the variations shown in UBS3 for Luke 24. There are
nineteen of them. But *eight* of them consist of D and the latins against
*every other manuscript.* So in Luke 24, based on this sample, it turns
out that *every manuscript*, except D, agrees with *every other manuscript*
at least 70% of the time. Does that mean that all of them belong to the
same text-type? Hardly!

The cut-off line for defining a text-type *must* depend on the sample.

But even for a particular sample, I don't think we can fix a cut-off and
say, "All manuscripts that meet this criterion belong to this text-type."
The great problem with the Colwell definition is that it ignores mixture.

Take a manuscript like 424. Even after diligent correction, it is still
more Byzantine than anything else. Depending on the sample, it might well
be rated as 70% Byzantine. But is it a Byzantine manuscript? Well, maybe...
but what makes it interesting is not the 75% of the text that is Byzantine,
but the 25% that agrees very closely with 1739 6 etc.

424, as corrected, is not a Byzantine text. It is not a family 1739 text.
It is a text *with family 1739 elements*. And *that* is what makes it
important. Whereas, by Colwell's definition, it's just another Byzantine
manuscript.

Which leads to another problem: Can a text-type be defined without a "core
text"? This is precisely the problem of the "Caesarean text." If one omits
p45 and W (as Hurtado has shown we should), there is no "core" Caesarean
witness. Does this mean that the Caesarean text *cannot* exist? I think
that too strong an argument.

Note that I am not saying that the Caesarean text *does* exist; that remains
to be settled. I'm just saying that we cannot use Colwell's definition to
settle the matter.

>b.) Text-type "D" (P29, P48, P38, 0171, and perhaps P69 + D): These are all,
>save for D, fragments, and certainly 0171 and P69 desert from the D-text
>of Luke
>approximately as often as they testify to it. The handeling of fragments when
>establishing text-type relations is an open question to my mind.

This I heartily agree with. We should not draw conclusions about text-types
based on fragments. And 0171 and D are distinctly different, so they cannot
*both* be assumed to be "core" "Western" witnesses.

But I'll ask another question: Why should D be assumed to be a core witness
to the "Western" text? At least in Luke, D shows clear signs of editorial
handling!

Just another insertion of unorthodoxy into this debate. :-)

Bob Waltz
waltzmn@skypoint.com

Back