Fri Nov 15 21:25:39 1996

From owner-tc-list  Fri Nov 15 21:25:39 1996
Return-Path: 
Received: by scholar.cc.emory.edu (SMI-8.6/SMI-SVR4)
	id VAA23902; Fri, 15 Nov 1996 21:24:36 -0500
Message-Id: <9611160320.AA17475@iris.arcadis.be>
Subject: long post about arabic Gospel mss
Date: Sat, 16 Nov 96 03:22:57 +0100
X-Sender: vale5655@194.111.178.251
X-Mailer: Claris Emailer 1.1
From: Jean VALENTIN 
To: "TC-LIST" 
Mime-Version: 1.0
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Sender: owner-tc-list@scholar.cc.emory.edu
Precedence: bulk
Reply-To: tc-list@scholar.cc.emory.edu
content-length: 17555

Hello Jimmy Adair and all tc-ers,

I have visited your web page on the versions of the NT. I have =
several remarks and suggestions, I hope they can be helpful. As for =
today, I will limit myself to the Arabic language.

SEVERAL PUBLICATIONS OF ARABIC VERSIONS
- the "Alexandrian Vulgate" of the XIIIth century has been published =
according to... one manuscript (!) by Paul de Lagarde. But as its =
transmission is rather stable, this edition must be considered as =
quite reliable.
P.  de  Lagarde  Die  vier  Evangelien  arabisch  aus  der  Wiener  =
handschrift herausgegeben (Leipzig, 1864; repr. Osnabr=FCck, 1972)

- The oldest Arabic version, originating from Southern Palestine in =
the IXth century, has partial publications. It was made from Greek =
manuscripts, and seems to be another witness of the "cesarean" or =
palestinian text (that is, it has many common variants with =
Koridethi, f1, f13, 565 etc... and the armenian, syropalestinian and =
georgian versions).
(1) S. Arbache Le t=E9tra=E9vangile Sina=EF arabe 72 (th=E8se, =
Louvain-la-Neuve, 1975)
Not a publication, but a study of the oldest dated ms (but in my =
opinion, this ms has a revised form of text compared to Sin. arb. 74 =
and others). The same author made a doctoral thesis in Bordeaux with =
a publication of Mk - arabic text and french translation - with a =
long study of the vocabulary and grammar. Arbache works at the =
university of Lille (north of France) and teaches textual criticism, =
but I don't know if he has plans to edit his manuscript.
(2) A. Galli Garland An Arabic Translation of the Gospel According to =
Mark thesis, Catholic University of America (Washington 1979) sees ms =
72 as a revision. I agree with this.
(3) B. Levin Die griechisch-arabische Evangelien-Ubersetzung, Vat. =
Borg. ar. 95 und Ber. orient. ar. 1108 (thesis, Uppsala, 1938) is an =
edition of two other mss of the same version. It has only Mt and Mk.

We should not forget, of course, the Arabic diatessaron. The latest =
edition is by Marmardji (full refercences in Metzger EVNT) but is to =
be used with caution. I have found several mistakes and omissions in =
the french translation he gives. I would say the french translation =
in unreliable, it is better to look at the Arabic text, and even =
then, knowing that it is a reconstructed text about which very few =
scholars were enthousiastic: always look at the critical apparatus. =
And, the variants are translated.
About this edition, I don't understand Marmardji's principles: he =
makes an "eclectic" text, but often he rejects in the apparatus the =
genuine diatessaric variant and prints the one that agrees with the =
peshitto, giving a wrong impression as to the degree of =
"vulgatization" of that text. At least, this is what I've found when =
I had to use it - would you agree or disagree, Bill Petersen?

OLD ARABIC? RECENSIONS? REVISIONS?
Now here I answer also to some questions that have been sent to me in =
the "what do you expect" thread. Bob Waltz, for example, asked me if =
there is an "old arabic" version. Some of you also reacetd to my =
question about what to do with recensions.
In Arabic there is not one version, but many. Also, we cannot say =
they are dependent of each other (like for example in syriac or latin =
where the successive versions are revisions of the older ones). Each =
confession (copts, nestorians, melkites, maronites and jacobites) =
produced its own version, even several according to the region, the =
liturgical language prevalent, the use (liturgical or apologetical) =
etc... 
Of course, some of these versions have been revised in the course of =
time. There are several processes we find, sometimes gradually as =
each scribe introduced several corrections, sometimes with what i =
call a recension, when there's a deliberate project, recurrent =
modifications following the same pattern in a ms or a group of mss.
What I've observed is (1) correction of the language, from a =
colloquial form of Arabic to a more literary form; (2) modification =
of the text-type: there's a progress of the byzantine text-type, it's =
already at work in the first revision of the first version (IXth =
century) and it seems to be definitively victorious in the XIIIth =
century in Palestine; (3) progressive influence of the "Alexandrian =
vulgate" from the XIIIth century on.

THE SINAI COLLECTION
In my study of the Arabic gospel mss of Mount Sinai (to which I added =
several mss from Jerusalem), I see the following: there are 42 mss =
(both tetravangelion and lectionaries) and they represent about 13 =
different versions. The number may change as it is not always easy to =
see clearly in the several recensions of the melkite version of the =
XIth century. I hope to publish a survey in due time, but here =
already some informations.

The oldest versions come first. 

(1) The version of mss 72, 74 and 54 appears in the IXth century. It =
is made from greek. It is in a quite un-classical arabic, except that =
72 ameliorates the arabic and makes the text (a little) closer to the =
byzantine text-type. It is probably the earliest one - at least the =
earliest mss contain it. It is also helpful in that it is quite =
literal in its rendition of Greek, sometimes to the point of being =
un-semitic. Samir Arbache of the University of Lille, is working on =
ms 72, I don't know what his plans are.

(2) The version of ms 75, also from the IXth century, is in a very =
good classical arabic. According to some authors, it is a revision of =
(1), but some others deny this. Sinai Arabic 75 is the only ms known.

(3) The version of ms 71 (Xth century) is very special. It is very =
often the shortest text of the Gospels that I've ever seen. But is it =
because it follows a very old shorter text, or because it is a free =
translation? It is written in an arabic that is much colored with =
aramaisms and unknown words and roots. Nevertheless, it was =
translated from a Greek text. We can be very sure of this, as several =
mistakes and nonsenses can be explained only by Greek. When a =
distinct greek text-type can be detected, we are very close to codex =
Koridethi (even closer than the syropalestinian is). I'm preparing an =
edition and translation of this very special ms.
A revised form exists in ms 133, which is a lectionary from the XIIth =
century. This second ms has some syropalestinian annotations. Its =
arabic is much better, and many readings from the syriac peshitto =
have been introduced.

(4) Ms 70 from the IXth century is also very old. I haven't studied =
it much, but it seems to be also a mixed text based on greek cesarean =
texts and the syriac peshitto. 

These were the earliest tries, here follow an important milestone.

(5) Then there is "the melkite version of the XIth century" as I call =
it. It is by far the version that is represented in the biggest =
number of manuscripts. The oldest ones (69 and 106) represent a mixed =
text with greek and syriac readings. With the time, the syriac =
readings are eliminated. Two mss (82 and 89) seem to represent a =
revision where a conscious effort was produced to produce a text very =
close to the greek and in good arabic. In total there are 15 mss that =
represent, in all their variations, this version in its evolution. It =
is about this text that I was sending the "What do you expect" =
message.
But there are related forms about which I don't know yet what I =
should think. A ms of the XVth century, also in Sinai, but written by =
a scribe from nort-western Syria, has basically the same version, but =
has many... old syriac readings that are not present, even in mss 69 =
and 106. Other mss (6 in Sinai and 2 in Jerusalem) represent three =
distinct lectionaries, where this version was revised in order to =
produce the lectionary.
There are also mss of this version in Leiden and other places, and =
there have been some allusions to it in ealier scientific works, =
including an important article by Curt Peters who publishes a few =
sample texts - but doesn't use the Sinaitic mss . Peters, Voobus and others remarked that, according to the =
big number of manuscripts, this version must have enjoyed an official =
status.

(6) There are still other mss which I haven't studied much, so I =
won't say much about them. (a) The "Alexandrian vulgate" is an =
eclectic recension. It was produced in Egypt in the XIIIth century, =
but was based on an earlier version supposedly made from Coptic. It =
is strange that I find this version in a XIIth century in Mount =
Sinai, but its text-type is definitely not Alexandrian - once again =
mixed cesarean-syriac. A second, later Sinaitic ms represents the =
usual type. Ignazio Guidi in his pionner article , gives a lengthy discussion of the elaboration and =
characteristics of the "Alexandrian vulgate", which quickly evinced =
all the other versions. It also progressively influenced the =
transmission of, then eliminated the Melkite version on its own =
ground.
(b) Ms 112 is a strange patchwork. I have done some surveys in Mt, =
and therefore this qualification: chapters 1-5 are purely peshitto, =
chap 23 is "cesarean" and ch. 28 has several typically old syriac =
variants!
(c) In Jerusalem, ms Staurou 26 is very peculiar. It is a bilingual =
lectionary (greek-arabic) from the XIth century. Its greek text is =
byzantine, but the Arabic follows a very different text-type, having =
agreements again with the palestinian text-type (Koridethi and its =
allies).
(d) There are other versions, mostly with a mixed text. A version =
tries; like centuries before ms 75 did, to have an elevated style in =
a very good literary Arabic. Others were made for the purpose of a =
lectionary.
Etc, etc... There are 13 main versions in Mt Sinai and Jerusalem (at =
least for the mss to which I had an access). "Main" versions because =
as I say, we could split some of them in two, three, four recensions. =
The melkite version of the XIth century has 24 of the 42 mss =
surveyed. Some are represented in one ms only. And of course, this =
represents only the situation in the Jerusalem melkite patriarchate, =
which according to the places and the periods used four languages: =
greek, syriac, syropalestinian and arabic. The situation is surely =
very different in Egypt or in Iraq, to mention only those places.

A QUESTION ABOUT METHOD
Now some words about the very interesting question by Henry =
Carmichael: how do I determine a version was made from Greek, Syriac, =
Syropalestinian, or Coptic? These are the four main candidates in the =
east.
The question of the Vorlage is a very important one, and also a =
difficult one. I must say I'm still thinking about the method in =
finding the language of the Vorlage. Also, for those wishing to have =
a greek retroversion, such a thing can only be done when we are sure =
that only the greek text was used by the translator. 
For Michael Holmes: Yes, I've read the article by Baarda. In fact I =
was thinking about his argumentation when I wrote about his point in =
my recent post. Also, Blake in the introduction to his edition of the =
Gospel of mark (p. 444-445) argues for latin, also from the fact that =
it has more similarities in its functionment with georgian that greek =
has, and it allows him to be very litteral. But as to Arabic, this =
language is much too different both from greek or latin or even =
french or english to allow a literal translation. I think I still =
need some more thought about it.

Now, back to the determination of the Vorlage. Here follows the =
actual state of my thinking about it, but maybe I will change my =
mind. I would really appreciate your comments about what follows.

There are several indications, none of which is 100 percent sure.
(1) Geographical area. In Egypt, Coptic is possible. In Syria, Iraq =
and Persia, Syriac is dominant. In Palestine, the melkite church is =
by far dominant, and so the greek language is favored. But as you =
know, the curetonian syriac version was discovered in Egypt. =
Alexandrian scholarship did much work directly from the greek. Syriac =
and syropalestinian had an important role in big sectors of the =
melkite church. Etc, etc...
(2) Liturgical language of the Church that produces the version. But =
here again, we should remember that the greek language, as reputed =
original, must always be considered. And, from one period to another, =
from a geographical area to another, it can change.
(3) Variants. I don't think they prove much. Such variants that we =
find only in syriac or coptic mss may have existed in greek. Except =
for a word-for-word rendering of a well known version (e.g. the =
peshitto) with all its peculiarities, there's not much help from the =
variants. Also, many versions mix several text-types. The alexandrian =
vulgate, according to Guidi, was made with the intention to produce =
an eclectic text, a fusioned text that would incorporate the readings =
of the three great vulgates of the East: the Byzantine greek text, =
the syriac peshitto and the bohairic versions. Using variants to =
determine the vorlage is just confusing - I've tried, and except in =
very few cases like Mt 1-5 in Sinai Arabic 112 leads to nothing.
(4) Linguistic peculiarities. You can find aramaisms in Arabic and =
think they are induced by the use of an aramaic (syriac or =
syropalestinian) text. But (a) maybe the translator had aramaic as =
his mothertongue and it influenced his rendition into Arabic (just =
like some of my constructions in English must be influenced by =
French), or (b) maybe we are in an area where spoken Arabic is still =
under the influence of an important aramaic substratum.
(4b) Now this also: Arabic has a very large vocabulary. There is =
always, when translating, the choice between several roots. The =
choice of the same root or word than in syriac proves nothing, except =
when it is _recurrent_ - I mean by this: the syriac or =
syropalestinian has probably guided the translator in the choice of =
his vocabulary when there are too many agreements between the syriac =
and arabic vocabularies. But this is quite difficult to determine.

There's only one criterium that works very well, but the problem is =
that you have to be lucky to find several examples. It is if the =
translator can be caught "red-handed" (en fran=E7ais: la main dans le =
sac) making a mistake or a nonsense that can be explained only by ONE =
language. For this, you need a translator that makes real blunders, =
and most of the Arabic translators were strong, multilingual, =
encyclopedic sholars.
But there are sometimes examples. Here are TWO that allow me to =
determine that ms Sinai Arabic 71 was translated from Greek.
(1) In Mt 23.5 we find in Arabic something that must read like =
"AHFAZ". It must be a plural, but it can't be found in any Arabic =
dictionary. It just doesn't mean anything in Arabic. We can see that =
it must be derived from the root H-F-Z, that means "to keep". Now, =
here it is: in greek, we have "phylakteria", which comes from the =
same root as the verb "phylasso", meaning also "to keep". Now you all =
get it: the translator came upon this technical word that he didn't =
understand, but he could see its etymology. So he tried to reproduced =
in Arabic the etymological derivation of the greek word, and left it =
to the reader to understand... And, we can be sure it is from greek =
and not from an aramaic dialect: all the syriac and syropalestinian =
version use a word akin to the hebrew tefillin, which has no =
connection whatsoever with the meaning "to keep". So the =
mistake/nonsense could not come from syriac or syropalestinian.
(2) In Mt 26.18, we find another nonsensical word in Arabic: Dayna. =
It also means nothing in Arabic, and no etymological derivation from =
Aramaic is possible. But in the greek text we have "deina", a hapax =
in the NT according to my concordance! (Go in town to such_and_such =
and tell him...). So here, the scribe came upon a rare word that he =
didn't understand and chose simply to transliterate it. And, if he =
had consulted the syriac or syropalestinian texts, he would have read =
"felan", which corresponds exactly to the Arabic "fulan". He would =
have been guided in his translation by the resemblance between his =
own language and the one of his Vorlage. This example indicates, not =
only that he translated from greek, but also that he used _only_ the =
greek text. Somebody working with both a greek and a syriac exemplar, =
if he didn't understand one, would have looked at the other... Do you =
never-never-never have an english bible opened in front of you =
besides your greek text ? :-)
As you see, it is not easy to find such cases as these two ones, and =
a good translator doesn't leave you many chances to catch him in this =
way. I consider myself very lucky to have found two such examples, =
and maybe there will be more in the future as this translator seems =
rather careless.

That's all for Arabic today.

Sorry for this long post.



Jean Valentin - Brussels - Belgium

Ce qui est trop simple est faux, ce qui est trop complexe est =
inutilisable.
What's too simple is wrong, what's too complex is unusable.

Back