Sun Jun 2 12:03:48 1996
From owner-tc-list Sun Jun 2 12:03:48 1996
Return-Path:
Received: by scholar.cc.emory.edu (SMI-8.6/SMI-SVR4)
id MAA16740; Sun, 2 Jun 1996 12:02:26 -0400
Date: Sun, 2 Jun 96 10:59 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: 10564
On Sat, 01 Jun 96, schmiul@uni-muenster.de wrote, in part:
>So here I end up where I started: "Discussion of 'text-types' raises the
>question of what we mean" (Larry W. Hurtado). Any proposals what YOU mean,
>Larry, Bob,...?
Thanks to Ulrich Schmid for reminding me of the *essential* point of this
discussion. As I thought about that question, I realized that I was not
prepared to answer it.
The starting point for defining a text-type is "The largest group of texts
that can be recognized as displaying kinship." This is a correct definition,
but unfortunately not very useful; it's too vague. We need to look for
something better.
It seems to me that there are in fact *five* levels of relationship that we
can recognize. I don't claim to have proper names for them, but let me outline
them so we have some terminology to work with.
1) The "tight" family. These manuscripts are so closely associated that we can
prepare an actual stemma. Examples would be 209-205-205abs or certain of the
subgroups of family 13. I suspect that there are other examples among the
Byzantine text.
2) The "loose" family. This consists of manuscripts that cannot actually be
placed in an exact stemma, but among which we can find relationships that can
be clearly defined. Families 1 and 13 fall under this category. I would also
include family 2138 in the Catholics. For the most part we cannot tell which
members of this family are sisters or descendents of one another, but we can
group it. For instance, 614 and 2412 go together. So do 1505 and 2495. So do
630 and 1799. For this family we can create a sort of quasi-stemma, knowing
that each of the lines of the diagram may represent one manuscript or several.
Family 2138
Archetype
|
---------------------------------------------------------
| | | |
------ ------- ------- ----------------
| | | | | | | | | |
2138 | 1505 | 2412 | 630 1799 | |
1611 2495 614 206 429
3) The "tight" tribe. This consist of manuscripts, all descended from a common
archetype, which are too distant from each other to form even the sort of
stemma shown above, but which are close enough that we can at least assess
relationships to the archetypes. An example of this is family 1739 in Paul.
Here we can say, for instance, that 1739 and 0243 are closely related at a
level only slightly removed from the archetype (I believe they are first
cousins, with their common ancestor being about three removes from the
archetype). 6 and 424** are related to each other within the family, but split
off at a slightly earlier point and have both suffered extensive Byzantine
influence (with 6 receiving more corrections of the late Byzantine type). 1881
split off at about the same point as 1739, but comes from a different family
and has suffered more corruptions. And so forth.
4) The "loose" tribe. Consisting of manuscripts descended from a common
archetype but which cannot be placed within the structure of the tribe. An
example of this would be the "B group" that I described earlier (B, p75, T,
Sahidic, L). The five go together, but their exact relationships generally
cannot be determined, except to say something like, "B is a good text; L has
been corrupted."
5) The "text-type." These are related manuscripts that *do not* derive
directly from a single common ancestor (other than the autographs).
It is evident, at least to me, that the final definition is the poorest of the
three. All I've really said is that a text-type is something looser than a
loose tribe. That's not much help.
Another thing that needs to be noted in using these definitions is the role of
the Byzantine text. No matter what one thinks of the origin of that text,
Byzantine mixture is the single most important element in the history of the
text.
In "tight" families, there is no Byzantine influence at all. Or more
correctly, what separates the members of the family is not Byzantine
influence; the family text may be heavily Byzantine. But all that separates
the family members is the errors and peculiarities of scribes.
In "loose" families, Byzantine influence begins to appear. In family 2138, for
instance, all members of the family seem to disagree with the (hypothetical)
archetype by about 10-20%. But in every case we find that the divergence is
Byzantine. Either the manuscript displays the family text, or it displays a
Byzantine reading. There are *no* instance of other readings (at least, none
that I've noticed).
In "tight" tribes, Byzantine influence becomes important, as do the side
effects of many generations of copying. Still, there is no sign of influence
other than the Byzantine.
In "loose" tribes, we see evidence of mixture from sources other than the
Byzantine text. Thus L, in the gospels, is mostly a mixture of p75/B readings
and Byzantine readings, but has a few readings that seem to belong to a
different non-Byzantine strain (perhaps from the Aleph group, or possibly from
some later phase of the Alexandrian text -- what the Alands apparently call
the "Egyptian" text, though I can't see how this differs from Hort's
"Alexandrian" text).
Incidentally, I have found that one *cannot* define the exact level of a
relationship simply by looking at statistics. One must compare the actual
documents over at least a few chapters of text. In cases where I have not been
able to do so (e.g. family 2127, where I am largely dependent on Merk's
inaccurate collations), I cannot define the precise nature of the family.
Family 2127 could, from what I know, be a "loose" family or a "tight" tribe.
This system comes out of my own method, which works from the bottom up. Taking
Paul as an example, I start by looking for tight families. There aren't many
of those -- 1739/0243, 1505/2495, and F/G are about all that spring to mind;
for our purposes they can largely be ignored. Then I find the "loose"
families. There are many more of those -- e.g. 330/451, 1505/1611/2495,
perhaps family 2127. Then come the "tight" tribes -- e.g. family 1739,
330/451/2492. Then "loose" tribes, such as certain phases of the Alexandrian
text (Aleph/33, 436/1962).
Now comes the hard part. We now have to group these tribes, and all the
remaining manuscripts, into types. Some types are easy -- e.g. everyone agrees
that D F G OL go together, even though they are not close enough to form
tribes. But that classifying types is not always easy is shown by the fact
that what Hort called one text type (the Alexandrian), Zuntz calls two (the
"Alexandrian" and "proto-Alexandrian") and I call three (Alexandrian, p46/B,
and family 1739).
I think my results are solid and verifiable -- for Paul. I agree with Zuntz
that D/F/G/, Aleph/A/C/I/33, and p46/B are distinct. (Though the Bohairic goes
with Aleph, not p46/B!) And since family 1739 is distinct from all three, and
closer to the other three than they are to each other, it stands alone and
deserves to be considered its own text-type (indeed, it looks on examination
to be the closest type to the original. But that is irrelevant to the argument
here).
This is the part of Colwell's definition that works best: the *gap*. There is
a clear gap between *all four* of my proposed text-types and the Byzantine
text (if one sets aside mixed manuscripts). There is a slightly smaller gap
between D/F/G and the other three groups. And a still smaller, but still very
distinct, gap between p46/B, family 1739, and Aleph/A/C/I/33.
Of course, this gap needs to be measured only among the non-Byzantine readings
of the witnesses. If we allow in Byzantine readings, the gaps become much
harder to see.
But this sort of technique does not work for the gospels. There the lines are
*not* clear. The D type, since it has only one Greek witness, is vague. The
"Caesarean" witnesses have a tendency to smear out between B and D. (For
instance, family 1 is notably more Alexandrian than Theta or family 13.) Even
the later Alexandrian witnesses don't fall into neat categories. The two
best-known "Alexandrian" minuscules, 33 and 892, fall generally between B and
Aleph (with, of course, a lot of Byzantine influence also).
Having come to this non-conclusion, I went to the textbooks -- and found very
little. Hort really doesn't say what a text-type is. Turning to the
introductory manuals in English, I found that:
* Greenlee does not define text-types, just gives a (not very accurate)
list of which manuscripts belong to which types
* Aland/Aland use the term but never define it (they rarely define anything!)
* Metzger devotes a few pages to determining textual relationships, but
only at a low level
* Vaganay/Amphoux devotes much more attention to text-types -- but
in the end defines the text-types *historically* rather than in terms
of manuscripts
I obviously couldn't check over every book on the subject, or even every book
in my library, in two days, but looking through specialized studies was hardly
more helpful. We've already noted the classical work by Colwell. It does a
good job of describing the problem and what needs to be done -- but then gives
us the useless 70% definition. Epp and Fee is devoted mostly to individual
readings, not text-types. Epp does devote some attention to the subject -- the
well-known "trajectory" theory -- but I could not find any usable definition.
Wisse's work on the Claremont Profile Method deals only with low-level
relationships. I can't really tell how Richards defines text-types -- but the
system is clearly wrong, because he lumps family 2138 with the Alexandrian
text! The most recent work I could find -- by Thomas C. Geer in Studies &
Documents 46 -- is little more than a plea for complete collations of all
manuscripts.
To sum up: If there is *any* reliable modern definition of a text-type, I
can't find it. Basically people are still saying, "I know one when I see one."
So now you can see why I brought up the question of the relationship between B
and Aleph. I could, with enough work, state a definition of a text-type that
works for Paul or the Catholics, and presumably for Acts or the Apocalypse. I
can't yet do so for the Gospels. And, obviously, until I/we do I/we cannot
assess the relationship between B and Aleph or among the "Caesarean"
witnesses.
I welcome anyone's thoughts.
Bob Waltz
waltzmn@skypoint.com
Back