Wed Nov 13 14:43:57 1996

From owner-tc-list  Wed Nov 13 14:43:57 1996
Return-Path: 
Received: by scholar.cc.emory.edu (SMI-8.6/SMI-SVR4)
	id OAA02152; Wed, 13 Nov 1996 14:42:42 -0500
X-Sender: waltzmn@popmail.skypoint.com
Message-Id: 
In-Reply-To: <19B82342DDD@div.ed.ac.uk>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Date: Wed, 13 Nov 1996 14:39:07 -0700
To: tc-list@scholar.cc.emory.edu
From: "Robert B. Waltz" 
Subject: Re: Patristic statistics
Sender: owner-tc-list@scholar.cc.emory.edu
Precedence: bulk
Reply-To: tc-list@scholar.cc.emory.edu
content-length: 3006

On Wed, 13 Nov 1996, "Professor L.W. Hurtado"  wrote:

>Small note in response to Robt. Waltz:
>The logic behind using "best" reps of text-types in attempting to see
>the relationships of another given ms is as follows:  (1) We can't
>realistically consider all available NT mss--too many to deal with in
>this lifetime;

No argument on that one; observe what I said.

(2) We don't need to anyway if the question is to what
>main group ("text-type") a given ms might belong--all we need are mss
>sufficiently representative of the text-types to serve the purpose.

The defect here is that this assumes the solution. How do we *know*
what are the best mss until we have examined enough of them to have
a feeling for the text-type?

Case in point (in Paul, as usual from me...). B was long considered
a member of the Alexandrian group. Then came p46, and Zuntz, and a
clear demonstration that B is *not* Alexandrian, but part of the
p46 text-type.

>Finally, as Fee and I have demonstrated, a firm determination of the
>textual relationships of mss requires (1) a good quantitative
>method--we both opt for Colwell;

I agree with the need for a good quantitative method. But your method
should not be called "Colwell"; it is "Colwell-Tune 1963" (for those
of you who have Studies in Methodology, the decisive quote is on page
59.

I would opt for a different "Colwell": "Colwell 1959."

Otherwise the method loses its point. Yes, using Colwell-Tune 1963,
or the Claremont Profile Method, you can find the *closest* relative
of any manuscript. But so what? The closest relative usually adds
relatively little to our knowledge. We want the more distant relatives.

Taking another example from Paul, the manuscript 104 is listed in most
tables of witnesses as Alexandrian. And yet, it agrees with the major
Byzantine witnesses much more often than with any Alexandrian text.
We call it Alexandrian not because that is its dominant text-type,
but because its Alexandrian readings -- even though they are in the
minority -- are those that interest us.

To take an analogy from genetics: If you want to learn something about
the human genome, and how it evolved, you don't take DNA from me and
from my parents; you take DNA from me, and from an ape, and from
a monkey.

>(2) plus assessment of the
>*significance* of readings shared by mss, which I tried to do in some
>detail for
>Codex W and representative witnesses.

This analysis takes place *after* the main comparison. I maintain they
should be done simultaneously. But this is a minor point.

Now let me make an observation: I have Hurtado's dissertation, and use it,
and think it one of the more methodologically sound works in my library
(as opposed to, say, Richards, which has 'way too many errors in it).
I am simply pointing out that we need to do more.

P.S. Apologies to the list if I'm grumpy today. I've had a computer go down
and several miscellaneous sorts of idiocy to deal with today.

Bob Waltz
waltzmn@skypoint.com

Back