[Prev][Next][Index][Thread]
b-greek-digest V1 #614
b-greek-digest Wednesday, 15 March 1995 Volume 01 : Number 614
In this issue:
Thayer
Re: For information
Re: Thayer
Re: Thayer
Re: Lexical Contacts and Cluster Analysis
Lexical Contacts and Cluster Analysis
----------------------------------------------------------------------
From: RoyRM@aol.com
Date: Tue, 14 Mar 1995 20:35:03 -0500
Subject: Thayer
I noticed during the discussion on software last month that it seemed like
almost all of the software listed Thayer's as a lexicon. I was under the
impression that at best, Thayer's was dated. Is this a bad assumption? If
it isn't, why is it making such an appearance? Is it length? popularity?
And why is BAGD not used? or even Louw & Nida?
Thanks,
Roy R. Millhouse
RoyRM@aol.com
------------------------------
From: Timothy Gaden <tjg@hermes.apana.org.au>
Date: Wed, 15 Mar 1995 08:20:13
Subject: Re: For information
>-- [ From: Alan R. Craig * EMC.Ver #2.10P ] --
>
>Some years back, in some book on Bible translations, I came across a
>comment that I have yet to find again nor identify who made it.
>Perhaps some of you might know who and where this was said:
>
> "Translation is interpretation."
I've not seen that exact quotation, but it seems to express the
same idea as Roland Barthes' famous statement, "Every decoding is
another encoding".
All German nouns are capitalised, even verbal nouns like "das
Glauben".
Tim.
- --------------------------------------------------------------
Timothy Gaden - tjg@hermes.apana.org.au - Melbourne, Australia
- --------------------------------------------------------------
------------------------------
From: John Baima <jbaima@onramp.net>
Date: Tue, 14 Mar 1995 20:03:55 -0600
Subject: Re: Thayer
>I noticed during the discussion on software last month that it seemed like
>almost all of the software listed Thayer's as a lexicon. I was under the
>impression that at best, Thayer's was dated. Is this a bad assumption? If
>it isn't, why is it making such an appearance? Is it length? popularity?
> And why is BAGD not used? or even Louw & Nida?
Thayer's is free. UChicago will not license BAGD now. Maybe someday they
will again. We and others use Louw & Nida.
- -John Baima
Silver Mountain Software
------------------------------
From: "James D. Ernest" <ernest@mv.mv.com>
Date: Tue, 14 Mar 1995 21:13:39 -0500 (EST)
Subject: Re: Thayer
On Tue, 14 Mar 1995 RoyRM@aol.com wrote:
> I noticed during the discussion on software last month that it seemed like
> almost all of the software listed Thayer's as a lexicon. I was under the
> impression that at best, Thayer's was dated. Is this a bad assumption? If
> it isn't, why is it making such an appearance? Is it length? popularity?
> And why is BAGD not used? or even Louw & Nida?
Think $$$$.
- -----------------------------------------------------------------
James D. Ernest Department of Theology
Manchester, New Hampshire, USA Boston College
Internet: ernest@mv.mv.com Chestnut Hill, Massachusetts
------------------------------
From: Kenneth Litwak <kenneth@sybase.com>
Date: Tue, 14 Mar 95 18:59:57 PST
Subject: Re: Lexical Contacts and Cluster Analysis
I stayed out of the earlier discussion of statistics, but now I wish to
make a few comments, since they relate to recent issues discussed here about
the basis and value of particular methods and hypotheses. I would like to
point out some concerns I have with these statistics.
1. No evidence has been presented to show that differences in vocabularly
or other stylometric data actually proves anything. You would
have to be able to say "Here's a document by author X and here's a
document that claims to be by author X and looks like a document from
author X but we know beyond question it is not, and this same
statistical analysis provides completely consistent results similar to the
case with the Pastoral epistles".
2. You have to be able to show that it is the case that Paul wrote
the "control group", e.g., Romans, Galatians, etc., and not the
Pastorals instead. The only reason Paul is assumed to have written
Romans as a starting point and not 1 Timothy is because F.C. Baur
said it, scholars believed it and that settled it for them.
3. These statistics take no account of the reasons why vocabularly
or style would be different between documents. A short, persoanl
letter of Philemon is not going to be the same as Romans. A letter
regarding pastoral/elder issues to Titus in Crete is quite obviously going
to be different than a more formal seeming document that is much longer,
like Romans, or a document that treats a totally different subject, like
Galatians.
4. I doubt that these statistical methods would even hold upto the
scrutiny of a modern author. Compare the vocabularly and style of a paper
in 1980 on Gregory of Nazianzus wrtings on the Holy Spirit with the
document I'm working on this week at work which is a test plan for a
new sfotware feature. Totally dissimilar. Totally different
vocabulary. Even in the same area, I have yet to write a paper that
mentions the Q hypothesis, while I'm sure that my forthcoming work for
a PhD, which begins next Fall, will have to refer to Q, even if only to
say that I am not using that hypothesis.
5. Remember what Mark Twain said about statistics :-).
6. I'm still looking for some kind of validation that the sample
size is large enough to be meaningful.
I know we have talked about hypotheses, but it seems to me that a
hypothesis should be able to answer the concerns I've expressed above,
particularly in accounting for different literary purposes and genres.
In the same vein, I recently finished browsing through an article I
was abstracting on statistics and Q, which argued that a proper statistical
analysis shows that Q is not a unity, challenging statistical assumptions
of others who argue for an alternative.Seems to back up what I learned
in psychology class in college. You can prove anything you want with
statistics. That is NOT to make a negative comment about Stephen
Carlson's work, but to recognize the limits to the validity of
statistical analyses in general.
Kenneth Litwak
Emeryville, CA
------------------------------
From: Greg Doudna <gdoudna@ednet1.osl.or.gov>
Date: Tue, 14 Mar 1995 21:34:51 -0800
Subject: Lexical Contacts and Cluster Analysis
To Stephen Carlson:
As one open to the potential of these methods I appreciate very
much your reporting of your data. Is the cluster analysis method
innovative with you, innovative with someone else in the NT
field, or is it tried and true with other corpuses of literature?
The claim for this method, which I agree does seem intuitively
sound, is that "[the method] can show that two corpora are
sufficiently distinct to cast doubt upon a thesis that they
have a common author."
And your notes do show, for the most part, substantial agreement
between your results and mainstream views and intuitions
regarding NT authorship--with one giant exception.
Your data appears to demonstrate that Luke and Acts were written
by different authors! Clustering between Luke and Acts happens
extremely "late" in all of your eight cases except your two
"pathological" ones. In your charts this is visually striking.
Is it not more likely that there is a flaw in the cluster
analysis method than that this particular conclusion is
correct?
To Ken Litwak:
> No evidence has been presented to show that differences in
> vocabulary or other stylometric data actually prove anything.
Would you be open to evidence of this nature, in the case
of New Testament writings, if it was presented?
Greg Doudna
West Linn, Oregon
- --
------------------------------
End of b-greek-digest V1 #614
*****************************
** FOR YOUR REFERENCE **
To unsubscribe from this list write
majordomo@virginia.edu
with "unsubscribe b-greek-digest" as your message content. For other
automated services write to the above address with the message content
"help".
For further information, you can write the owner of the list at
owner-b-greek@virginia.edu
You can send mail to the entire list via the address:
b-greek@virginia.edu