Wed Feb 12 09:21:09 1997

From owner-tc-list  Wed Feb 12 09:21:09 1997
Return-Path: 
Received: by shemesh.scholar.emory.edu (SMI-8.6/SMI-SVR4)
	id JAA18850; Wed, 12 Feb 1997 09:21:08 -0500
X-Sender: waltzmn@popmail.skypoint.com
Message-Id: 
In-Reply-To: <199702121332.FAA08487@m8.sprynet.com>
References: <9702120808.AB29497@iris.arcadis.be>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Date: Wed, 12 Feb 1997 08:25:02 -0700
To: tc-list@shemesh.scholar.emory.edu
From: "Robert B. Waltz" 
Subject: What is a theory (Was: Re: Koren vs BHS text)
Sender: owner-tc-list@shemesh.scholar.emory.edu
Precedence: bulk
Reply-To: tc-list@scholar.cc.emory.edu
content-length: 2685

On Wed, 12 Feb 1997, "Lewis Reich"  wrote:

>On 12 Feb 97 at 8:08, Jean VALENTIN wrote:
>
>> It shows that these three mss disagree in many instances, specially when 
>> it comes to matres lectionis. And, of course, this raises the inevitable 
>> remark that this theory of equidistant letters works only with one text. 
>
>I wonder if it is appropriate to describe the equidistant letters 
>matter as a theory, since what is being reported is a result, and as 
>far as I know, no particular theory was advanced in the statistical 
>articles to account for the result.  
>
>The fact that results of this kind are obtainable only with one text 
>is not an objection to the results, in effect it is a confirmation of 
>the statistical analysis that the effects are such that could not 
>have resulted from mere chance.

Let's get our terminology straight. (This is not a criticism of anyone;
just a matter of getting things straight.)

1. Facts. Facts can only be determined by direct and repeatable
   observation. So the readings of a particular manuscript (at least
   where it can be clearly read) are facts.

2. Statistical models. Statistical models are *not* facts, even if
   they embrace all the data. (I don't know if this "equidistant
   letters" item uses all the data -- but I doubt it.) The same
   data is subject to all sorts of statistical analysis. Statistical
   analysis is repeatable, but that does not make it fact in its
   own right. Moreover, statistical analyses can be flawed.

3. Theory. *Anything* not falling under the above heads is theory.
   All interpretation is theory. A theory may have strong backing,
   but it cannot be proved. You cannot prove an interpretation; you
   can only *dis*prove it. (You disprove it by providing a
   counterexample.)

So, in the case of the "equidistant letters" thing, the facts are the
readings of the manuscripts. Whatever examination the researchers did
is a statistical analysis. Their conclusions are a theory.

And *any* results they derive *could* be coincidence. We can *never*
say that coincidence is "impossible"; merely highly unlikely.

Be it noted that I still have not seen the paper, so I cannot make
any judgment on the value of its conclusions -- or even its underlying
facts. But please, let's use our scientific terminology correctly.

-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-

                            Robert B. Waltz
                         waltzmn@skypoint.com

Want more loudmouthed opinions about textual criticism?
Try my web page: http://www.skypoint.com/~waltzmn
(A very rough draft of part of the Encyclopedia of NT Textual Criticism)

Back