TEXTUAL CRITICISM OF THE KORAN.

• Author: BELLAMY, JAMES A.

The presidential address of 2000, in which the author suggests emendations and/or explanations of seven proper names in the Koran, which have not been satisfactorily identified. These are: Tuwan, al-Jibt, Azar, Idris, [Uzayr.[subset]], al-Rass, and [Isa.[subset]]

F. W. HALL, A WELL-KNOWN CLASSICAL SCHOLAR in the early part of this century, once said that "many people tend to regard textual criticism as a disease." [1] Hall disagreed with this opinion, and so do I. Textual criticism is not really a disease; I would prefer to call it an addiction. Some people like myself feel a compulsion to emend texts that do not make sense, and if they do this in a careful, common-sense way, they may restore the original and so deserve the gratitude of all those who use it. Others are bound by hoops of steel to the printed text they have before them, and cannot conceive that it might contain some corruptions.

However, all lengthy texts contain some corruptions, and it is the goal of textual criticism to remove them and produce a text as near as possible to the original. Another goal is to provide some insight into the production, transmission, and editing of the work in question.

Koranic studies have not benefited much from textual criticism. The editors who around A.D. 650 produced the Uthmanic recension of the Koran, which is the standard version, were much better textual critics than the generations that followed, both East and West. Later Muslim scholars felt obliged to accept the received text as correct, and most Western scholars have followed suit, which is the main reason why so little progress has been made in establishing a better text.

The Uthmanic recension has certain characteristics that make it unusually subject to misreading. First of all, it contains no short vowels at all, and even long a is often not indicated. Only long u and i, which sometimes can be pronounced aw or ay, are usually, but not always, noted. An even more serious problem is presented by the fact that there were no diacritics in the earliest Koranic text, so differences between sh and s, and between b, t, th, n, and y, cannot be distinguished. On the other hand, the consonantal outlines of the words, with a few minor exceptions, are firmly established, so the range of possible errors is quite limited. The vast majority of the words in the Koran can be read in one way, only, dictated by the requirements of syntax and sense. However, the fact that the writing system is imperfect has to be kept constantly in mind by anyone working on the text of the Koran.

The recitation of the Koran soon became professionalized, and many of the best-known readers made selections of variants for their own use. In the early tenth century, about three-hundred years after the promulgation of the Uthmanic recension, a well-known scholar named Ibn Mujahid decreed that seven readings were canonical and all the others shadhdh, or "deviant," hence not to be used in ritual recitation. Not everyone agreed with his classification, and some later scholars approved of ten, or even fourteen, recitations. But in the course of time the availability of readings has declined gradually to the point that today only two are in common use. The others, both canonical and deviant, are known only to advanced students in Koranic recitation. All Muslims recite the Koran today in the Uthmanic recension, and most of them use the reading of [Asim.[subset]] b. Abi Najud, a Kufan scholar, as transmitted by his student Hafs b. Sulayman. This has the happy result that we have a good recension against which we can record our emendations and not have to make one of our own.

These variants, however--I have counted more than two-hundred that make a difference in the meaning--are important in that they tell us that there was no solid oral tradition stemming directly from the prophet to prove which variant was correct. For example, in Surah 6:63, of the seven readers, the two from Kufah recite [[subset].sup.njyn.[subset]] (anjana) "he saves us." and the other five [[blank].sup.[contains]][njytn.sup.[contains]] (anjaytana) "you (sg.) save us." [2] These two words sound so different that no one, unless he were deaf, could mistake one for the other, and the words on both sides of the word in question are unambiguous. One cannot argue that the prophet used one variant one day and the other the next. Nor can one maintain that there is a firm oral tradition that guarantees the reading of the unambiguous words but breaks down when more than one reading is possible. It seems clear that the earliest readers got their readings from the written text of the Uthmanic recension, and since Arabic was their native language, they read the unambiguous parts correctly, and where the text was ambiguous, they exercised their knowledge of the language and came up with what pleased each of them the most. [3]

There is, however, another set of mistakes in the Koranic text which are serious copyists' or readers' errors and do not result simply in a choice of variants. These have no recorded variants in the consonantal outline (rasm) in the canonical readings, which means, of course, that all the readers accepted them as correct. This tells us immediately that the mistakes were already in the tradition before the new recension was undertaken.

I have written about some of these before, so I will not repeat myself here, but will propose emendations and/or explanations for a few proper names that have so far not been identified. I have chosen proper names deliberately for two reasons. First, names point to a specific place or a definite person, so if they make no sense, this is a clear indication that we are faced with a mistake. Second, one can sometimes find supporting evidence, which goes beyond palaeography, for emending them, either in the Bible or elsewhere in the Koran. In the case of common nouns and other ordinary words, we are in the same position as the early readers; we can base our readings only on sense and syntax. I would further point out that all the mistakes I shall correct occur very commonly in later manuscripts, so one cannot take exception to them palaeographically. Most important, the majority of them (four out of seven) are not really copyists' errors, but are errors made by the early readers, who misread the consonantal outl ines, which are in fact correct. In other words, the Koranic text is better than the recitations of it by the early readers.

TUWAN

One of the most curious textual problems in the Koran occurs in the retelling of the story of Moses and the burning...