Wed May 1 16:18:08 1996
From majordom Wed May 1 16:18:08 1996
Return-Path:
Received: by scholar.cc.emory.edu (5.0/SMI-SVR4)
id AA18167; Wed, 1 May 1996 16:18:08 +0500
Date: Wed, 1 May 96 15:14 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: Textual Theory
Content-Length: 13865
Sender: owner-tc-list@scholar.cc.emory.edu
Precedence: bulk
Reply-To: tc-list@scholar.cc.emory.edu
On Wed, 01 May 96, schmiul@uni-muenster.de (Ulrich Schmid)
>On Sun, 28 Apr 1996, Robert B. Waltz wrote:
>
>>Please note that this theory is ONLY for Paul; my work on other
>> areas is incomplete.
>
>I appreciate this statement very much, since, looking on so much statements in
>the discipline of tc, one is often confronted with results, though initially
>stemming only from one part of the NT (for example from the Gospels, or even
>just from Luke), but boldly extrapolated to the whole of NT textual
>transmission.
As a matter of fact, these results DO NOT apply to other areas in the
Mew Testament. Even in the Catholics, which superficially bear the
greatest similarity to the Paulines, the results are very different.
I found three major texts there, all with subdivisions. (In this
I independently verified the results of Duplacy.)
>>The first step is to identify text-types. This is a complex
>>matter which cannot be covered briefly; suffice it to say that
>>it is based on a statistical technique based on 550 readings
>>(for statistical purposes I would like to have more readings,
>>but I just haven't been able to come up with them) and about 50
>>manuscripts...
>
>>Based on this, we find a large number of manuscript groupings,
>>some of which (e.g. 330-451-2492) seem not to have been noticed
>>before, and others (e.g. Soden's family 1319, which ought to
>>be called family 2127) which have been inadequately classified.
>
>The _Text und Textwert_ volumes from 1991 for the Pauline Epistles give the
>collation results of 798 MSS at 231 variation units. Do the results drawn
>therefrom with respect to MSS grouping differ significantly from the results
>drawn from the 50 MSS at 550 variation units (supposing that "550 readings"
>means 550 variation units)? The examples mentioned above could be easily found
>out, when consulting _Text und Textwert_.
True to some extent. The examples I cited (330-451-2492, etc.)
are sister manuscripts.
What cannot be found from _Text und Textwert_ is the mixed manuscripts.
Take, for example, 6. Its closest relative, in my sample, is L, with
69% agreement. The figures for other Byzantine manuscripts are similar.
By contrast, 6 agrees with 1739 only 64% of the time.
I don't have _Text und Textwert_ in front of me, but if those figures
are representative, then T&T would show literally HUNDREDS of manuscripts
as being closer to 6 than is 1739.
Which is all well and good, except that 6's important kinship is with
1739. Where 6 is non-Byzantine, 6 and 1739 agree 85% of the time. Where
BOTH are non-Byzantine, they agree some 95% of the time. Other than the
members of family 1739, 6's closest relative in non-Byzantine readings
is B, at 72%.
So 6 is a weak member of family 1739 -- and this is not evident from
_Text und Textwert_.
>>There are, in Paul, FOUR manuscript families which give good
>>evidence of being independent of, and in my view predating,
>>the Byzantine text.
>
>a.)The independence of the Byz. text of these four manuscript families is one
>thing. But, on what reasons do they _predate_ the Byz. text?
Whether I can prove they predate the Byzantine text, of course, depends on
what date you assign the Byzantine text.
If you accept the Byzantine text as being fourth century, then the
four groups I mentioned can be shown to be earlier. The early date
of the p46-B group follows from the fact that p46 is third century
or earlier.
Aleph-A-C-I-33 was in existence by the fourth century, and the group
had already begun to fracture by that date; therefore it must predate
the fourth century.
In the case of D-F-G, the proof comes from the Old Latin.
In the case of family 1739, the evidence is that this text is
largely similar to that of Origen.
Of course, if you accept the Byzantine text as original, I can't
prove these text-types to precede it. But they are certainly all early.
>b.) What about the above mentioned MSS group (330-451-2492), and what about
>other MSS groups (for example 1505-2495-1611)? Are they not independent enough
>from the Byz. text? What makes the independent status of manuscript families?
I cannot trace the history of 330-451-2492 before the eleventh century.
It may be post-Byzantine. Also, it is more Byzantine than not. So
this group, while a family with independent text, does not qualify as
an early text-TYPE.
In the case of 1505-1611-2495-syh (to which 1022 can be added in the
Pastorals), the family is obviously old, since it is found in the
Harklean version. Beyond that it cannot be dated. It MAY be the
vague remnants of an independent (Syriac?) text-type.
But in any case, it is more Byzantine than anything else. So it
offers little help in group analysis. One can never tell whether a
group reading is an early survival or just Byzantine mixture.
I noted that I had to resort to other methods when I had a two-versus-two
split in text-types. One of my "other methods" was consulting the other
manuscript groups.
>>Since we now have FOUR groups, rather than the two that come up
>>so often in textual studies, we are usually able to proceed on
>>a Majority basis. If a reading is found in all four groups, it
>>is certain. If a reading is found in three, I will accept it
>>unless there is VERY strong internal evidence. (Usually the only
>>evidence I accept in this case is assimilation of parallels.)
>>If the groups split evenly, then some other criterion must be
>>resorted to. This is where internal evidence usually applies.
>
>What about cases in which _the_ group readings may be uncertain, for the
>intra-group testimony is split? I assume internal evidence is applied
>here, too.
Sometimes, but not often. For example, if family 1739 is split, it
will usually be found that several members of the family are
Byzantine, while the others are non-Byzantine. Since all the members
of the family suffer Byzantine influence, and all but 1730 have suffered
heavy Byzantine influence, I take the non-Byzantine reading.
Technically, I suppose this is "internal evidence." But the process
is hardly subjective.
The only place where I really have trouble is where p46 and B split.
There I do have to guess at readings sometimes -- but this rarely
makes a difference in the ultimare result.
>>However, I take a very limited view of internal canons of
>>criticism. I maintain that there is ONLY ONE canon for internal
>>evidence: "That reading is best which best explains the others."
>>All other canons ("Prefer the harder reading," "prefer the shorter
>>reading," etc.) are simply corollaries of this rule, and should
>>be applied in this light. (This simple rule has other corollaries
>>which are often forgotten, e.g., "Prefer the MIDDLE reading.")
>
>This I appreciate very much: "That reading is best which best explains the
>others". But, why excluding the Byzantine text from this procedure?
Who said I omitted the Byzantine text from this procedure?
I will admit that I may miss something in analysing it. But in
analysing which reading best explains the others, I try to take
ALL text-types into account.
>Sometimes
>the Byz. text gives readings which may help to explain other readings, and
>there
>is at least one instance within the Corpus Paulinum where the Byz. text
>gives a
>reading that best explains all readings and settles the case of the somehow
>scattered evidence of the four manuscript families mentioned above: The
>position
>of the doxology of Romans 16,25-27.
>
>a.) Parts of the "Western" family omit the doxology: F (G) 629.
>b.) Family 1739 is split: 1739 and 630 give the doxology after 16,23; 6, 424,
>and 1881 give the doxology after 14,23.
>c) The (other) Alexandrian family is divided, too: B and Sahidic give the
>doxology after 16,23, whereas P 46 gives the doxology after 15,33.
>d) The traditional Alexandrian family is split: aleph, C, Bohairic, 81,
>and 436
>give the doxology after 16,23; 1175 gives the doxology after 14,23; so
>does 1506
>but omitting chapter 16; A and 33 give the doxology after 14,23 _and_ after
>16,23.
>
>Arguing from internal evidence the position of the doxology after 14,23
>clearly
>best explains all other readings, since it is most unlikely that it should
>have
>been inserted at that position, as long as it is found in other positions
>(i.e.
>after 15,33 or 16,23). On the other hand, the removal of the doxology away
>from
>14,23 seems most logical as long as chapters 15-16 follow.
>Puzzling enough, no homogeneous family gives this reading that best
>explains the
>others (at best three members out of five, cf. family 1739). But, on the other
>hand, the Byz. text most homogeneously gives this reading.
>
>To conclude therefrom: In this peculiar instance, the Byzantine text as a
>distinct entity displays a state of the text which is earlier than most of the
>scattered evidence of the four manuscript families mentioned above.
In assessing this evidence, let me point out that one of the purposes of
my method is to eliminate the effects of Byzantine mixture. That's one
reason why we need family groupings.
Thus in two cases the family reading is obvious:
family 1739, as represented by 1739 630, places the doxology after 16:23
(the other manuscripts of the group have been influenced by the
Byzantine text)
(Which brings up a footnote: readings of 424* have NO SIGNIFICANCE.
The absence of a correction in 424** does not mean that the
exemplar used for correction agreed with 424*. The corrector could
have missed a reading, or been unsure of what the correction
meant, or could have been sure that the corrected text was wrong.
We cannot speculate on readings we cannot see -- e.g. the readings
of the manuscript used to correct 424 when 424 shows no correction.)
The Alexandrian text, as represented by aleph, C, bo, has the
doxology after 16:23. The reading of A 33 is the result of mixture.
1175, although late Alexandrian in most of Paul, is thoroughly
Byzantine in Romans (so is 33, for that matter. And 2464. I didn't
get into those details).
The readings of p46 are 1506 are peculiar, and deserve notice (I've been
fascinated for years by 1506's omission of Romans 16). But since they
are singular, no great case can be built upon them.
I concede the split in the "Western" text.
I can't agree with your assessment of the internal evidence. It omits at
least one point: Marcion's editorial work on the epistle. I think the
possibility must at least be considered that the variation in the placement
of the doxology is the result of Marcion's using a shorter form of Romans.
We must also consider the possibility that Marcion _wrote_ the doxology.
If it makes any difference, I concede this is a very difficult reading.
I could cite others (Col. 3:6 springs to mind). But one or two difficult
readings do not invalidate the method.
>>Technically this is an eclectic method. But it is a very limited
>>eclecticism; the only place where I REALLY have to be subjective
>>are when
>>1. The four text-types split into two groups, or
>>2. The text-types split into FOUR groups, or
>>3. The members of the various text-types are so divided that
>> the group reading cannot be determined.
>
>The above mentioned example from the Romans doxology might point to at least a
>fourth scenario concerning the subjective element (combining 2. and 3.):
>4. The text-types split into FOUR groups, _and_ the members of the various
>text-types are so divided that the group reading cannot be determined.
I will agree -- theoretically. But I know of only one reading where this
_both_ conditions apply to the extent that it affects things. The reading
is in 1 Cor. 6:5 (and the NA27 apparatus does not treat it). I'm sure there
are others -- but not many.
>To sum up: The limited eclectic method exposed looks quite impressive to my
>mind. Nevertheless, I suspect the subjective element involved is much greater
>than expected.
I don't deny subjectivity. But I consider it much less subjective than
the system followed by the UBS editors. And more solidly grounded than
the method followed by Westcott and Hort.
>I may also point to my above mentioned questions concerning the
>lateness of the Byzantine text and the singeling out of _only_ four manuscript
>families.
There could well be more manuscript families. There COULD even be some
early families. I don't claim this is the last word. If I could, I would
apply my techniques to all the manuscripts and be certain. But as a non-
professional, I haven't had time. I'm doing what I can.
The one thing I would observe is that the results in _Text und Textwert_
would seem to imply that there are no more "outstanding" manuscripts
out there. (By "outstanding" I mean entirely free of Byzantine influence.)
I have little use for the Alands' system of classifying minuscules, but
it does provide an indication of Byzantine influence. Of their Category
I and II manuscripts, I have classified almost all:
p46 -- with B
aleph -- Alexandrian
A -- Alexandrian, with some slight mixture
B -- with p46
C -- Alexandrian, with some mixture
D -- "Western"
F -- "Western" (and why is F category II and not G?)
I -- Alexandrian
33 -- Alexandrian (except in Romans)
81 -- Late Alexandrian
256 -- family 2127 (probably; based on von Soden's collations)
1175 -- Late Alexandrian (Mostly; not in Romans or Thessalonians)
1506 -- Alexandrian (although of a peculiar type; needs a good study)
1739 -- family 1739
1881 -- family 1739
1962 -- Late Alexandrian; with 436
2127 -- family 2127
2464 -- Late Alexandrian (and not really good enough to be category II)
This list includes everything but 442 (which MIGHT go with 1962, but
I haven't much to go on). So it seems unlikely that there are any
PURE representatives of PURE text-types omitted.
And I'm sure this is all anyone ever wanted to hear on the subject!
Bob Waltz
waltzmn@skypoint.com
Back