Monday, November 13, 2006

An Arabic Perspective on Orthodox Corruption

After I worked for several years on Arabic manuscripts - not of the New Testament, but of Arabic horsemanship (Edition and Translation of ibn ahi Hizam al-Khuttali’s works) - I thought it might be helpful to draw some parallels to NT Textual Criticism. I employed about ten mss of the middle ages, two later recensions, and two (not very careful prepared) printed editions.

1. The quality of the transmission differs widely from ms to ms. Some have only half of the stuff, some have a hiatus of a large number of paragraphs, some totally mixed up the paragraph-order (there are 170 paragraphs - aetiologies and treatments, each of them subdivided, sometimes up to 15 times).

2. There are manuscript families. Families have a certain amount of common readings and of common misreadings, but that does not mean that they always behave like a family: Sometimes you find in a ms which belongs definitely to a certain family - let’s say family I - a reading from family II. That suggests that a lot of these variants may either a) be influenced by a text-critical endeavour of the scribe, or b) have come in by chance - the same situation (e.g. a seemingly missing subject / object / personal pronoun / preposition) brought about the same changes.

3. The best mss are not reliable in every reading. Better trained scribes seem to be more prone to homoioteleuta than their less trained colleagues. Even the worst ms helps sometimes to restore a difficult reading.

4. When each paragraph is considered separately, changes are on a moderate scale. Usually, you have what you always had in textual criticism: substitution of pronouns for nouns or proper nouns, changes of the 2nd person to 3rd person and vice versa, addition of clarifying attributes or short sentences, homoioteleuta/arcta, dittographies, abridgements of long lists, smoothing out of grammar and meaning, etc.

5. Nonsense readings emerge when a scribe did not understand what he had before him (there are often pharmacological expressions of Greek, Latin or Persian origin in these texts), and the scribe created either (in an effort to keep the original) the best he could make out of the reading, or (when he did not much care) a new reading adapted to the meaning of the context, but totally different from what was in the original. On rare occasions only are these corruptions irreparable.

6. A large part of the variants that came into being look similar to the original, but mean something quite different, even if they seldom damage the context. This is especially true of Arabic, where only small diacritical dots differentiate between letters. To speak in terms of NT textual criticism: If you have the famous reading 1Tim 3:16, every one muses on the possibility that ΟΣ was changed to ΘΣ by only (accidentally) dropping the horizontal nomen-sacrum-stroke, and perhaps adding the middle bar in the ΟΣ. Others say that the change in the one or the other direction was made by intention.
Coming from my own experience with these Arabic mss, I would now pose a different or third solution: Scribes of antiquity obviously introduced changes into the text when a) the context and the outward appearance of the letters etc. granted permission and/or suggested/triggered a change of the text, b) the text - in their estimation - needed an improvement to keep its meaning from becoming misused. When these two conditions meet each other, you are on the best hunting ground for variants. We have the same in Mt 1:7-8; 1:10; John 1:18; 1 Thess 2:7; we have it in the numerous changes brought about through itacism; etc. That does not mean that there are not also a lot of accidents (misreadings, examples of carelessness etc.). This is not orthodox corruption - it is something different: The concept of orthodox corruption does not consider the outward appearance of the text, it focusses on the meaning, and suggests a corruption of the meaning. What I think is worth investigating are those variants where the similarity of the letters / of the word(s) “allowed” to a certain extent to bring in a similarity of the meaning. These variants did not change or even damage the meaning of the text; usually, they only underlined it; they “improved” the meaning.
As far as I know, we can observe similar phenomena in the LXX translation of the OT. The LXX reflects quite often an only slightly different reading of the Hebrew original (letters dalet-resh, sin/shin, similarity of sound in ‘ayin / ghayn / aleph etc.), or knowledge of a different Semitic reading/root of the very same letters (James Barr has a lot of examples in his “Comparative Philology and the Text of the OT”). Here also, some cases may be accidental, but I wonder whether it might not have happened that a lot of cases came into being by the process mentioned above: outward similarity (letters, sound) suggested / allowed / triggered / provoked a change of the text to improve its meaning. These ancient scribes were not dumb fools who carelessly brought about changes of the text whenever it suited their taste ...

7. In general, the text of the mss, even of mutilated mss (a lot of pages missing / mixed up) has not lost its original meaning. Even those scribes who were careless did not dare to change the text ad libitum. The original meaning was either transmitted, or, if lost via transmission, can be restored via textual criticism.

8. There is one modern Arabic edition which used only one manuscript. That edition has, of course, certain shortcomings, and a lot of questionable readings (and this reminds us of Richard Bentley’s truism: If there had been but one manuscript of the Greek Testament, at the restoration of learning about two centuries ago; then we had had no various readings at all. And would the text be in a better condition then, than now we have 30 000? So far from that; that in the best single copy extant we should have had hundreds of faults, and some omissions irreparable. Besides that the suspicions of fraud and foul play would have been increas’d immensely).

With five mss, I reached already a point were it was clear that new mss would not change the text very much. Ibn Khuttali’s text must have been transmitted in hundreds of mss, and goes back to a Greek tradition, which was later compiled into the Corpus hippiatricorum graecorum. There is even an Arabic translation made from an Armenian Vorlage. With more mss at hand, and some compilations which were made by different authors, the text now as a solid basis - but can, of course, be slightly improved with every new ms which may be unearthed.

“Not only do variants not damage the study of Holy Writ, they are rather useful, according to Augustine.” (Erasmus of Rotterdam, Letter EE 860: 61-64).

1 comment

  1. Thanks for sharing this.
    This reminds me of a professor I met once who insisted that Paul Maas's book on textual criticism is a better resource for NT TC than anything specifically on that topic. This professor's opinion was that TC is the same thing no matter where it's being applied, and issues like the number and age of NT mss should not change the nature of the discipline in any way.

    I still haven't gotten around to reading Maas, partly because I'm skeptical of the advice I was given. Does anyone here have an opinion on that basic issue?