The Origins of the Koran

This is a summary of The Origins of The Koran: Classic Essays on Islam's Holy Book, edited by Ibn Warraq (Prometheus Books: Amherst, New York. 1998). Ibn Warraq has provided a valuable collection of some of the most important critical studies of the Koran over the past century. Most of the essays are now a bit dated, and those familiar with the modern revisionist approach to Islamic history will recognise the areas where further study has proposed conclusions very different to some of the authors included here. These essays are foundational reading for all students of the Koran. They reveal many areas where new study is needed as well as providing a good grounding in the materials available to us both within the Islamic tradition and from non-Muslim source. Ibn Warraq himself provides a helpful discussion of the state of contemporary research, and the sections on the collation, variants, and sources of the Koran contains essays by such scholars as Arthur Jeffery and St. Clair-Tisdall. It is to be expected that this type of criticism will be summarily dismissed by most Muslim readers, but it should be very informative for students of religious history. This summary is not authorised by the editor, though it attempts to be a faithful representation of the ideas in this book and does not necessarily reflect my own views.

Summary by Sharon Morad, Leeds

The Origins of the Koran:
Classic Essays on Islam's Holy Book

Edited by Ibn Warraq; Prometheus Books, 1998
Summarised by Sharon Morad, Leeds

Part One: INTRODUCTION

Chapter One: Introduction (pp. 9-35)
-Ibn Warraq

There is a notable lack of critical scholarship on the Koran.

Major questions still needing answers include:

1. How did the Koran come to us? [issues of compilation and transmission]
2. When was it written and who wrote it?
3. What are the sources of the Koran? [the origin of stories, legends, and principles]
4. What is the Koran? [How do we determine authenticity?]

The traditional account claims that the Koran was revealed to Muhammad, written down in bits, and not collated before Muhammad's death.

The Collection Under Abu Bakr (p. 11)

Abu Bakr was caliph from 632-634. There are several incompatible traditions describing a collation during his reign.

1. 'Umar was worried that bits of the Koran would be lost after many Muslims were killed at the Battle of Yamama. Therefore he commissioned Zaid ibn Thabit to collect the Koran and write it down?
2. Or was it Abu Bakr's idea? Or maybe 'Ali's?
3. There are several other difficulties: Could this have been accomplished in only two years? The Muslims were fighting the Battle of Yamama (in Central Asia), why had these new converts memorised the Koran but the Arab converts had not? Why was this collation not an official codex but rather the private property of Hafsa?

It sounds like these traditions were invented to credit the popular Abu Bakr and (more significantly) to debit the much maligned 'Uthman.

The Collection of the Koran (pp. 12-13)

'Uthman was caliph from 644-656. He was asked for an official codex by one of his generals because the troops were fighting over which reading of the Koran was correct. Zaid was once again commissioned, with the help of three others. But…

1. The Arabic of the Koran was not a dialect.
2. There are variations between the number and names of the people working with Zaid. (One version lists somebody already dead at that time!)
3. In these stories there is no mention of Zaid's involvement in an earlier rescension.

Most scholars assume that the 'Uthmanic rescension is correct and the Abu Bakr rescension is fictitious, but they have no valid reasons for preferring it over the latter, as the same reasons for dismissing the Abu Bakr story (biased, unreliable, late sources, attempts to credit the collector etc…) can be applied to the 'Uthman story as well.

One major (and often un-addressed) question is – how much can we rely upon the memories of the early Muslims? Can we assume that they not only remembered everything perfectly, but that they heard and understood Muhammad perfectly in the first place?

Variant Versions, Verses Missing, Verses Added (pp. 13-18)

Modern Muslims assert that the current Koran is identical to that recited by Muhammad. But earlier Muslims were more flexible. 'Uthman, A'isha, and Ibn Ka'b (among others) all insisted that much of the Koran had been lost.

Codices were made by different scholars (e.g. Ibn Mas'ud, Ubai ibn Ka'b, 'Ali, Abu Bakr, al-Aswad). 'Uthman's codex supposedly standardised the consonantal text, yet consonantal variations persisted into the 4th century AH. An unpointed and unvowelled script contributed to the problem. Also, although 'Uthman tried to destroy rival codices variant readings survived. Standardisation was not actually achieved until the 10th century under the influence of Ibn Mujahid. Even he admitted 14 versions of the Koran. These are not merely differences in recitation; they are actual written variations.

Also, if some verses were omitted, why couldn't some have been added? For example, the Kharajites considered the Joseph story to be an interpolation, and most scholars suggest the addition of scribal glosses designed to explain the text or smooth out rhyme.

Scepticism of the Sources (pp. 18-34)

Muhammad died in 632. The earliest written material of his life is the sira of Ibn Ishaq (750), but Ibn Ishaq's work was lost. We only have parts of it available in quotation by Ibn Hisham (834). The hadith are even later. There are six authoritative collections of hadith: Bukhari, Muslim, Ibn Maja, Abu Dawud, al-Tirmidhi, and al-Nisai. All are dated between 200 and 300 years after Muhammad.

Scholars have attempted to distinguish which hadith contain real information from those containing legendary, theological or political embellishment. Wellhausen insists that the 8th century version (i.e. Ibn Ishaq) was accurate, and later versions were deliberate fictions designed to alter the 8th century story. Caetani and Cammens suggest that most sira were invented to construct an 'ideal' past and a justification for contemporary exaggerated exegesis of the Koran. Most scholars conclude that the stories about Muhammad prior to becoming a prophet are fictitious. In his important critique of the hadith Goldhizer argues that many hadith accepted even by the most rigorous collectors were 8th and 9th century forgeries with fictitious isnads. These hadith arose out of quarrels between the 'Umayyads and their opponents – both sides freely inventing hadith to support their respective positions. The manufacture of hadith speeded up under the 'Abbasids who were vying with the 'Alids for primacy. Even Muslims acknowledged a vast number of forgeries [~90% of hadith were discarded], but even so the collectors were not as rigorous as could be hoped. Even in the 10th century over 200 forgeries were identified in Bukhari. At one point 12 different versions of his work existed.

In his study of the hadith Schacht concludes:

1. Isnads only began to be widely used after the 'Abbasid revolution, and then they were formulated carelessly.
2. The better an isnad looks the more likely it was to be spurious
3. No existing hadith can reliably be ascribed to Muhammad
4. Most of the classical corpus was widely disseminated after Shafi'i (820) and most of he legal tradition was formulated in the 9th century.

His methodology includes looking at legal decisions – if they didn't refer to a crucial tradition it's because the tradition wasn't there. He argues that traditions were created in response to 9th century conditions and then redacted back several centuries. Islam cannot be traced accurately back before the 8th century.

Wansbrough argues that the Koran and the hadith developed out of sectarian controversies and were projected back to the time of Muhammad. Islamic law developed after contact with Rabbinic Judaism outside the Hijaz. Muhammad is portrayed as a Mosaic-type prophet, but the religion was Arabised – Arabic prophet, Arabic Holy language, Arabic scripture. At the same time as the formation of this Arabic religion we see the beginning of interest in pre-Islamic Arabic poetry, further suggestive of a rise in Arab nationalism. Negative evidence further supports a late date for the creation of the Koran. There is no record of the Koran being used in legal decisions before the 9th century, and the Fiqh Akbar I (a sort of Muslim creed drafted in the mid-8th century to represent orthodox views) contains no reference to the Koran.

Cook, Crone, and Hinds argue that Islam developed as an attempt to find a common identity among peoples united in conquests that began when the Arabs joined Messianic Judaism in an attempt to retake the Promised Land. Looking at non-Muslim all we can say is that Muhammad lived, was a merchant and taught about Abraham. But other than that non-Muslim sources do not confirm the traditional Islamic account. We have no reason to think that he lived in central Arabia (much less Mecca), or that he taught about the Koran. The Koran first appears late in the 7th century, and the first inscriptions with Koranic material (e.g. on coins and the Dome of the Rock) show trivial divergence from the canonical text. The earliest Greek sources say that Muhammad was alive in 634 (Muslim sources say he died in 632). In the 660's the Armenian chronicler describes the community of Jews and Arabs, but Muslims say that the Arabs split with the Jews during Muhammad's lifetime. The Armenian also describes Palestine as the focal point of the Ishmaelite (i.e. Arab) activity, though Muslims say this focus switched to Mecca in AH 2.

The result of their research is described in Hagarism: The Making of the Islamic World (1977). The major thesis of this work is that Muhammad preached a message of Jewish Messianism and became involved in a joint attempt by Jews and Arabs, citing common Abrahamic decent, to reconquer Palestine. Therefore the earliest non-Muslim sources report strong anti-Christian sentiment. But, eventually the Arabs quarrelled with the Jews in Palestine and needed to establish a separate religious identity. They were inhibited by lack of an indigenous religious structure, so they borrowed heavily from the Samaritans. For example, note the similar emphasis on the unity of God, the fatiha resembles a Samaritan prayer, the Koran only seems to know of the Torah or the Psalms (the Samaritans do not recognise the rest of the Hebrew scriptures), the importance of Moses, and the similarities between the Samaritan view of the Messiah and the Muslim concept of the Mahdi.

Slaves on Horses: The Evolution of the Islamic Polity argues that the traditions about the caliphate are fictitious, and Meccan Trade and the Rise of Islam claims that the existence of the Koran required the invention of stories to explain it. These stories became more detailed and elaborate over time and the further from Arabia that they were collected.

Chapter Two: The Koran (pp. 36-63)
-Theodor Nöldeke

The present Koran is identical with the original. Muhammad probably could read and write, but he tended to use a scribe. There is some suggestion that part of the Koran was written down during Muhammad's lifetime, since he had its inserted and deleted in large suras which he probably could not have remembered unless they were written down. The Koran itself admits that Muslims accused Muhammad of changing verses (S. 16:103). Variations are explained by the abrogation of verses and laws.

The Quraishites preferred the stories by Nadr son of Harith, who told Persian myths – so Muhammad had him executed.

The Koran contains many Biblical characters, but the stories are mixed up. The variations came from either the Jewish Haggada or the New Testament apocrypha or they are simply mistakes made by a listener (e.g. Haman is believed to be the minister of Pharaoh, and Mary is believed to be the sister of Aaron).

The style is semi-poetical. Rhyme is maintained throughout, but rhythm is rarely used. There are many reasons to criticise the style – arbitrary leaps between subjects, annoying word repetitions, and poor grammar. The challenge to 'produce a sura like it' is completely subjective. Muhammad repeatedly emphasised that the Koran is in Arabic, but he borrowed many foreign terms to express ideas that had no Arabic expression. Sometimes he misused these terms (e.g. the Aramaic 'furquan' meaning 'redemption' is used to mean 'revelation').

Differences between the Meccan and Medinan suras are due to a change in circumstances as Muhammad moved from being the preacher of a small, despised sect to becoming an autocratic ruler. However, establishing the chronology of revelation is almost impossible. The traditions that attempt to do so disagree with each other and are not reliable. In fact, there is very little reliable information at all about Muhammad before the Hijra. We are not even sure when to date the beginning of his prophethood (probably ~610). The Meccan suras tend to be short and are reminiscent of the oracles of pagan soothsayers, even beginning with the same oaths involving heavenly objects like stars. The greatest passage in the Koran is S. 1 – al-fatiha. This shows the influence of the Jews, especially in the reference to God as 'Rahman.' The Medinan suras are longer and contain sketches of the histories of previous prophets, laws, and diatribes against Jews and Christians. The beginning of each sura has a cryptic series of letters – for which no meaning is known.

After the death of Muhammad no one knew the entire Koran by heart. Many Arabs revolted against Abu Bakr and had to be forcibly put down. The greatest opposition came from Maslama (a.k.a. Musailima) who claimed to be a prophet but was executed by Abu Bakr. Then 'Umar asked Zaid ibn Thabit to collate the Koran. The suras were arranged from longest to shortest, as even then the chronological order was imperfectly known. That codex was given to Hafsa. Other scholars also compiled their own codices. These became sources of contention because they different from one another. So, 'Uthman asked Zaid to write another codex and all the others were destroyed despite a fair amount of grumbling by their compilers. The variations between the codices could not be variations of dialect, as at this point the Arabic script could not express such variations, being both unvowelled and unpointed. The distinctives of the destroyed codices have survived somewhat in oral tradition. Ibn K'ab's codex contains two extra suras (similar to al-fatiha) and Ibn Masu'd has a different order and omits suras 1, 113, and 114. Ibn Mas'ud seriously opposed the use of Zaid's codex over his own, arguing that he [ibn Mas'ud] had been a disciple of Muhammad for longer and knew the Koran better than Zaid. Even after the production of Zaid's codex a great variety of different readings (extending to meaning and not just pronunciation) were possible through different means of pointing and vowelling. Eventually seven systems of pointing [each with two systems of vowelling] were considered valid.

Part Two: THE COLLECTION AND THE VARIANTS OF THE KORAN

Chapter Three: Uthman and the Recension of the Koran (pp. 67-75)
-Leone Caetani

1. The Koran today is not the same as that given by Muhammad
   During the lifetime of the prophet and immediately afterwards verses were circulating that were either apocryphal or mistakenly attributed to the prophet. The 'Uthmanic recension was necessary to deal with the uncertainty regarding the canonical text. "It is clear that in the year 30 AH no official redaction existed. Tradition itself admits that there were various 'schools,' one in Iraq, one in Syria, one in al-Basrah, besides others in smaller places, and then, exaggerating in an orthodox sense this scandal, tries to make out that the divergences were wholly immaterial; but such affirmations accord ill with the opposition excited by the caliph's [i.e. 'Uthman's] act in al-Kufah. The official version must have contained somewhat serious modifications." (pg. 69)

2. The first recension under Abu Bakr and 'Umar is a myth

   1. Why did Abu Bakr practically conceal his copy, especially if the death of so many Muslims at the battle of Yamamah really did endanger the existence of the Koran?
   2. How was it that there was still no consensus regarding the Koran in AH 30 if this official codex had been made?

3. The 'Uthmanic recension was undertaken for political rather than religious motives
   Muhammad made no provision for continuing political and religious leadership after his death. Without his guidance, the knowledge of men who remembered his teaching (reciters or 'Qurra') became valuable. The Qurra spread with the empire establishing schools and teaching the lay populace and other Qurra. Rival groups developed, and many Qurra also began to voice strong disapproval of the caliph and of the military and political leaders who were profoundly ignorant of the Koran. The Qurra encouraged a general revolt against 'Uthman in AH 25. 'Uthman reacted quickly, ordered an official text to be complied and branded anyone who recited the Koran differently as a heretic. This effectively broke the power of the Qurra by taking the monopoly of knowledge about the Koran out of their hands.

4. We must revise our opinion of 'Uthman's character and not be mislead by later Muslim bad press.
   Tradition has many evil things to say about 'Uthman, but they dare not criticise his recension, because the Koran resulting from it is the foundation of Islam. Many of the complaints about 'Uthman are anti-'Ummayyad polemics and unjustly blame him for the financial blunders of his predecessor, 'Umar. The invention of the Abu Bakr recension effectively reduces 'Uthman's role to nothing more than copier of a previously compiled text. This accomplished the dual goal of preserving the authority of the existing text, while failing to give any credit to 'Uthman for preserving the Koran.

Chapter Four: Three Ancient Korans (pp. 76-96)
-Alphonse Mingana

1. The sources of the Koran - Muhammad was illiterate. He depended on oral information from Christians and especially from Jews. The corruption of oral transmission explains the inaccuracies of the stories. Historical errors include: Mary being the sister of Aaron(S. 3:31ff), Haman being Pharaoh's minister (S.28:38), and the conflation of Gideon and Saul (S. 2:250). There are contradictory attitudes toward non-Muslims. S. 2:189 says to fight against unbelievers and Suratut-Taubah says to make war on those who disagree, but S. 2:579 says there is no compulsion in religion and S. 24:45 says to dispute only kindly with Jews and Christians.
2. If we strip away the commentary, the Koran is inexplicable. Muslim theologians explain the contradictions by trying to put ayat (verses) in a historical context and by appealing to the doctrine of abrogated and abrogating verses. Without the commentary the Koran is completely garbled and meaningless.
3. Transmission from 612-632? – Muhammad never ordered the Koran to be written down, and when first asked to do so by Abu Bakr, Zaid ibn Thabit refused, arguing that he had no right to do so if Muhammad hadn't thought it necessary. (The wonderful memory of the Arabs has been overstated. For example, if we compare versions of the elegy 'Itabah' in different tribes we see significant variations.) Some verse were apparently written down, but we're not told which ones and we have no idea how they were preserved. What happened to the scraps after codification? They couldn't have been just chucked away – what sacrilege!
4. Who is the compiler of our standard text and is it authentic? Zaid ibn Thabit supposedly wrote the whole text of the Koran at least twice (under Abu Bakr and then under 'Uthman). The first copy was given to Hafsa, but 15 years later the believers were still arguing about what the Koran was, so 'Uthman had Zaid write up a second copy and destroyed all the others. Zaid probably tried to reproduce faithfully the words of Muhammad, otherwise surely he would have improved the style and grammar and amended the historical and typographical errors!) Indeed, the Koran today is substantially identical with this second recension, though not necessarily with the words of Muhammad. The claim that the Koran is perfect Arabic is absurd – there are many examples of repetition, weak rhyme, changing letters to force a rhyme, foreign words, bizarre usage or change of names (e.g. Terah to Azar, Saul to Talut (S. 2:248250), Enoch to Idris (S. 19:57)

II. The text of the Koran has traditionally been studied through (1) commentaries, (2) grammarians studying Arabic vowels and diacritical points, and (3) types of script used.

1. The first commentator was Ibn Abbas. He is the main source of traditional exegesis, though many of his opinions are considered heretical. Other important commentators include Tabari (839-923), az-Zamakhshari (1075-1144), and al-Baidhawi (d. 1286)
2. Diacritical marks did not exist before the 'Umayyad caliphate. They were borrowed from Hebrew and Aramaic. Important grammarians include Khalil ibn Ahmad (718-791) who invented the 'hamza', and Sibawaihi (Khalil). Vowels were not discovered until the end of the 8th at a study centre in Baghdad century under the influence of Aramaic.
3. Three major scripts are used – Kufic, Naskhi, and Kufo-Naskhi. The type of script gives the first rough division of age of manuscripts. More precise age determination is arrived at by considering other features, like the use of diacritical points.

Chapter Five: The Transmission of the Koran (pp. 97-113)
-Alphonse Mingana

According to Muslim writers (pp. 98-104)

• There is not much consensus among the traditions about the collection of the Koran. The earliest records about compilation are from Ibn Said (844), Bukhari (870) and Muslim (874).
• Ibn' Sa'd lists 10 different people who are supposed to have collected the Koran in the time of Muhammad (with a number of different hadith supporting each contender). Then he also gives hadith attributing collation to 'Uthman during 'Umar's caliphate, and in another place attributes collation to 'Umar himself.
• Bukhari's stories are different. He gives credit to the collection of the Koran during Muhammad's lifetime to a variety of people, but not the same list as Ibn Sa'd gives). Then he has the story of Abu Bakr's recension carried out exclusively by Zaid ibn Thabit. This is immediately followed by hadith about the 'Uthmanic recension work done by Zaid and three others.
• The last two traditions (the Abu Bakr and 'Uthmanic recensions) have been accepted above all the others – why? Also, if they had already assembled the whole Koran, why was it so hard to produce a codex? These two recensions are likely as fictitious as the others.
• Other Muslim historians confuse the picture farther:
  • The author of the Fihrist lists all the stories given by both Ibn Sa'd and Bukhari, then adds in two more.
  • Tabari tells us that Ali B. Abi Talib and 'Uthman wrote the Koran, but when they were absent ibn Ka'b and Zaid ibn Thabit did so. The people at that time accused 'Uthman of reducing the Koran from many books to one.
  • Wakidi writes that a Christian slave, ibn Qumta, taught Muhammad and that ibn Abi Sarh claimed that he could change what he wanted in the Koran just by writing to ibn Qumta.
  • Another source of traditions attributes the collection of the Koran to the caliph 'Abdul-Malik b. Marwan (684-704) and to his lieutenant Hajjaj b. Yusuf. Barhebraeus and Jaluld-Din as-Sayuti attribute it to the former, Ibn Dumak and Makrizi to the latter. Ibnul-Athir says that al-Hajjaj proscribed the reading of al-Masu'd's version, and Ibn Khallikan says that al-Hajjaj tried to get writers to agree on a text but was unsuccessful. Indeed variant readings continued and were recorded by Zamakhsharia and Baidhawi, though anyone who followed the variants was severely punished.

Transmission of the Koran according to Christian writers (pp. 104-111)

1. 639 CE - discussion between a Christian patriarch and 'Amr b. al-'Asd (summary of conversation recorded in a manuscript dated 874 CE). We learn:
   1. The Bible had not been translated into Arabic
   2. Teaching regarding the Torah, inheritance, and denial of the divinity and death of Christ existed in the Arab community.
   3. No reference was made to any Arab holy book.
   4. Some of the Arab conquerors were literate.
2. 647 CE – a letter from the patriarch of Seleucia, Isho'yabb III, refers to the beliefs of the Arabs without any reference to the Koran.
3. 680 CE – the anonymous writer at Guidi knows nothing about the Koran, thinks that the Arabs are simply professing the Abrahamic faith, and doesn't realise that Muhammad is a religious character.
4. 690 CE – John Bar Penkaye, writing under the reign of 'Abdul-Malik, has no idea that the Koran existed.

Only in the 8th century does the Koran become an item of debate between Muslims and Christians. Early Christian critics of the Koran include: Abu Nosh (secretary to the governor of Mosul), Timothy (the Nestorian patriarch of Seleucia), and, most importantly, al-Kindi (830 CE – i.e. 40 years before Bukhari!).

Kindi's major argument: 'Ali and Abu Bakr had been squabbling over the succession to Muhammad. 'Ali began collecting the Koran, and others demanded that their bits be included. A variety of codices were written. 'Ali pointed out the divergences to 'Uthman, hoping to undermine them, so 'Uthman had all but one copy destroyed. Four copies of 'Uthman's codex were made, but all the originals were destroyed. When Hajjaj b. Yusuf became powerful ('Abdul-Malik was caliph – 684-704) he gathered together all the copies of the Koran, changed passages as he wished, destroyed the others and made six copies of the new version. So, how can we possibly distinguish the original from the counterfeit?

A sort of Muslim response to Kindi is found in an apology for Islam written 20 years later in 835 CE by the physician 'Ali b. Rabbanat-Tabari at the request of the caliph Mutaw'akkil. In it Tabari ignores Kindi's historical point and merely asserts that the Sahaba (i.e. companions of the prophet) were good men. Then he lays out an apology for Islam that is significant because it pre-dates the hadith.

In summary – the Christians don't seem to know of the official Koran until the end of the 8th century and they seem to see Islam as a political venture with a bit of religious dressing.

Conclusion (pp. 111-113)

1. Almost nothing of the Koran was written at the death of Muhammad. It's uncertain as to how well known writing was in Mecca and Medina at that time.
2. Some years after Muhammad's death his companions began writing down oracles of Muhammad. This gave them prestige. 'Uthman's version was given royal sanction and the others were destroyed. Certainly dialectical differences were not the problem, as Arabic script at that time could not differentiate between dialectical variations anyway.
3. 'Uthman's Koran was probably written on scrolls of parchment (suhufs) and then, under 'Abdul Malk and Hajjaj b. Yusuf these were placed in book form with a fair amount of redaction, some parts deleted and others added.

Chapter Six: Materials for the History of the Text of the Koran (pp. 114-134)
-Arthur Jeffrey

Muslim writers have not seemed interested in textual criticism of the Koran since 322 AH when the text was fixed by Wazirs Ibn Muqla and Ibn 'Isa (helped by Ibn Ibn Mujahid). After that point those who used old or variant readings were punished (Ibn Miqsam and Ibn Shanabudh are good examples of what happened to those who made the attempt). Though the actual manuscripts have perished, these variations are somewhat preserved in the commentators of az-Zamakhshari (d. 538), Abu Hayyan of Spain (d. 745) and ash-Shawkani (d. 1250), and in the philology works of al-'Ukbari (d. 616), Ibn Khalawaih (d. 370), and Ibn Jinni (d. 392). None of this information has been used to produce a critical text of the Koran.

Muslim tradition (i.e. that before his death the prophet had the Koran ordered and written out though not in book form) is largely fictitious. After all, this same tradition says that very little had been recorded and that large amounts of the Koran were in danger of being lost when Muslims were killed at Yamama.

Abu Bakr probably did collect something, as did a variety of others (whose names are not agreed on in any two lists preserved in the tradition); but his collection was not an official recension, rather a private matter. Some orthodox Muslims say the word 'jama'a' ("to collect") only means "to memorise" in the traditions referring to the metropolitan codices, but as these collections were carried on camels and eventually burnt it is more likely that they were written codices. Different metropolitan areas followed different codices: Homs and Damascus followed al-Aswad, Kufa – Ibn Mas'ud, Basra – as-Ash'ari, and Syria – ibn Ka'b. Major divergences between these texts mandated 'Uthman's radical recension. The Qurra violently opposed him in this, and ibn Masu'd stubbornly refused to give his codex up until he was forced to do so.

Variants were preserved by commentators and philologists only when they were close enough to orthodoxy to help with tafsir. The ones they do preserve they insist were merely explanatory glosses on 'Uthman's text.

"The amount of material preserved in this way is, of course, relatively small, but it is remarkable that any at all has been preserved. With the general acceptance of a standard text other types of text, even when they escaped the flames, would gradually cease being transmitted from sheer lack of interest in them. Such readings from them as would be remembered and quoted among the learned would be only the relatively few readings that had some theological or philological interest, so that the great mass of variants would early disappear. Moreover, even with regard to such variants as did survive there were definite efforts at suppression in the interests of orthodoxy. On may refer, for instance, to the case of the great Baghdad scholar Ibn Shanabudh (245-328) who was admitted to be an eminent Koranic authority, but who was forced to make public recantation of his use of readings from the old codices." (pg. 119)

Any of the more striking variants were not recorded because of fear of reprisal.

"For example, Abu Hayyan, Bahr VII 268, referring to a notorious textual variant, expressly says that in his work, though it is perhaps the richest in uncanonical variants that we have, he does not mention those variants where there is too wide a divergence from the standard text of 'Uthman."

The Masahif Books (pp. 120-126)

During the fourth Islamic century three books were written by Ibn al-Anbari, Ibn Ashta, and Ibn Abi Dawud, each entitled Kitab al-Masahif, and each discussing what was known of the lost codices. The former two are lost to us and known only in quotation; the third has survived. Ibn Abi Dawud is the third most important Hadith collector. He refers to fifteen primary codices and thirteen secondary codices (the later were mostly based on Mas'uds primary codex).

One major drawback to tracing variants through the Hadith is that there was not the same meticulous care taken over the transmission of the variants as over the canonical version, so authenticity is difficult to ascertain. However, despite the limitations, significant information is available to contribute toward the formation of a critical text. Thirty-two different books contain the main sources of variants.

Codex of Ibn Mas'ud (d. 33) (pp. 126-129)

Ibn Masu'd was an early convert. He participated in the Jijra's to Abyssinia and Medina, was present at the battles of Badr and Uhud, was a personal servant of Muhammad, and learned seventy suras from the prophet. He was one of the earliest teachers of Islam, and was commended by the prophet himself for his knowledge of the Koran.

He produced a codex that was used in Kufa, and many copies were made of it. He indignantly refused to give his codex up because he argued it was more accurate than Zaid ibn Thabit's. His codex did not include Suras 1, 113, and 114. He did not consider them a part of the Koran though he knew of them and offered variant readings of them. The order of his suras is also different from that 'Uthman's official codex.

Codex of Ubai B. Ka'b (d. 29 or 34) (pp. 129-131)

Ibn Ka'b was one of the Ansar. He was a secretary to Muhammad in Medina and is said to have written the treaty with the people of Jerusalem and to have been one of the four instructors commended by Muhammad. His personal codex was dominant in Syria even after standardisation. He appears to have been involved with the creation of 'Uthman's text, but tradition is garbled as to exactly how. He seems to have known the same number of suras as the authorised version, though the order is different. His personal codex never attained the popularity of Ibn Mas'ud's codex, and it was destroyed early by 'Uthman.

Codex of 'Ali (d. 40) (pp. 132-134)

'Ali was Muhammad's son-in-law and supposedly began compiling a codex immediately upon the death of Muhammad. He was so engrossed in the task that he neglected to swear fealty to Abu Bakr. Some say he had access to a hidden store of Koranic materials. 'Ali's sura divisions were very different from 'Uthman's so it is difficult to tell if material was missing or added. 'Ali supported 'Uthman's recension and burnt his own codex. It is hard to know if the variants ascribed to 'Ali were in fact due to the original codex or to his interpretations of 'Uthman's codex.

Chapter 7: Progress in the Study of the Koran Text (pp. 135-144)
-Arthur Jeffrey

A quick look at Muslim commentaries reveals many difficulties with the vocabulary of the Koran. The commentators tended to assume that Muhammad meant the same things as they would mean by certain words, and they interpreted the Koran in light of the theological and judicial controversies of their time.

Jeffrey has already produced a lexicon of the non-Arab words in the Koran, but the Arabic words cannot properly be investigated until a critical text exists. The closest thing to a textus recepticus is the text tradition of Hafs from 'Asim (the best of the three traditions of the Kufan school). A standard issue of this text tradition was officially produced by the Egyptian government in 1923.

Following the Muslim traditions, the text resulting from the 'Uthmanic recension was unpointed and unvoweled. When diacritical marks were invented different traditions of pointing developed in the major metropolitan centers. Even when the consonants (huruf) were agreed different ways of voweling could be devised. So a large number of ikhtiyar fi'l huruf (i.e. traditions as to the consonants, as variations in pointing resulting in a varying consonantal text) developed. These systems not only differed regarding pointing and voweling, but occasionally used different consonants altogether, as if attempting to improve the 'Uthmanic text. [NB: There are seven systems of pointing (i.e. ikhtiyar f'il huruf), each with two traditions of voweling, providing a total of fourteen canonical variations in reading. When citing a system both the source of the huruf and the source of the voweling are mentioned.)

In AH 322 Ibn Mujahid of Baghdad (a great Koranic authority) pronounced a fixed huruf (supposedly 'Uthmanic) and forbade any other ikhtiyar and limited the variations in voweling to seven different systems. Later, three other systems were considered equally valid by some.

So, the text of the Koran has two major categories of variants, the canonical variants, restricted to patterns of voweling (of which the system of 'Asim of Kufa according to Hafs is most popular for some reason), and the uncanonical consonantal variations.

Chapter 8: A Variant Text of the Fatiha (pp. 145-149)
-Arthur Jeffrey

The Fatiha (Sura 1) is generally not considered to be an original part of the Koran. Even the earliest Muslim commentators (e.g. Abu Bakr al Asamm d. 313) did not consider it canonical.

One variant form of the Fatiha is given in the Tadhkirat al-A'imma of Muhammad Baquir Majlisi (Tehran, 1331), another is given in a little book of fikh written about 150 years ago. These two vary from one another and from the textus recepticus though the sense of all three remains the same. Variations include: replacing synonyms, changes in verb form, and one or two changes of words that are not synonyms by have generally related meanings (e.g. 'r-rahmana (merciful) to 'r-razzaqui (bountiful).) These variants to not improve grammar or clarity and seem to have no doctrinal significance; they are the sort that would exist in an oral prayer that was later fixed.

Khalil b. Ahmad, a Reader of the Basran school, offers yet another variant. He is a known to have transmitted from 'Isa b. 'Umar (d. 149) and was a pupil of Ayyub as-Sakhtiyani, (d. 131), both of whom are famous for their transmission of uncanonical variants.

Chapter 9: Abu 'Ubaid on the Verses Missing from the Koran (pp. 150-153)
-Arthur Jeffrey

There are perhaps a few invalid proclamations that have been interpolated into the Koran, but what is far more certain is that many authentic proclamations have been lost. Jeffrey gives the complete text of a chapter in Abu Ubaid's Kitab Fada'il-al-Qru'an, folios 43 and 44, concerning chapters that have been lost from the Koran.

Abu 'Ubaid al-Qasim . Sallam (154-244 AH) studied under renown scholars and himself became well known as a philologist, jurist and Koranic expert. His chapter contains a list of Hadith on the missing verses of the Koran. According to these Hadith:

• 'Umar is recorded as saying that much of the Koran has disappeared.
• Ai'sha ways that sura 33 used to have 200 verses, but much of it has been lost.
• Ibn Ka'b says that Sura 33 had as many verses as sura 2 (i.e. at least 200 verses), and included the verses on stoning [NB: as the Sura 33 has 73 verses today.]
• 'Uthman also refers to the missing verses on the stoning of adulterers (several different Hadith all report this).
• Ibn Ka'b and al-Khattab differed over whether S. xxxlii:6 (sic) was part of the Koran or not.
• Several people (Abu Waqid al-Laithi, Abu Musa al-Ash'ari, Zaid b. Arqam, and Jabir b. 'Abdallah) remember an aya about humans being greedy which is not now in the Koran.
• Ibn Abbas confesses to hearing things and not knowing if they were part of the Koran or not.
• Abi Ayyub b. Yunus reports a verse that he read in A'isha's codex that is not now in the Koran, and adds that A'isha accused 'Uthman of having altered the Koran.
• ' Adi b. 'Adi comment on the existence an other missing verses, the previous existence of which was confirmed by Zaid ibn Thabit.
• 'Umar questioned the loss of another verse, and was informed by 'Abd ar-Rahman b. 'Auf that "It dropped out among what dropped from the Koran."

'Ubaid concludes the chapter by asserting that these verses were all genuine and used to be recited during prayers, but they were not passed down by the savants because they were considered extra, similar to verses contained elsewhere in the Koran.

Chapter Ten: Textual Variations of the Koran (pp. 154-162)
-David Margoliouth

Orthodox Islam does not demand uniformity of the Koran. It permits 7-10 variant readings differing usually (but not always) in minutia.

Other (non-orthodox) variations can be attributed to the fact that Muhammad frequently changed his revelation and some of his followers might not have known what the abrogating version was. After his death it was a political necessity for 'Uthman to standardise the text, and al-Hajjaj produced yet another recension at the end of the7th century.

For a long time there was confusion about what was Koran and what was not. Sometimes verses of poets were cited as words of Allah. Even the religious leaders weren't always sure what the correct text was. For example, in one of his letters the Caliph Mansur grossly misquotes S. 12:38, relying on the word 'Ishmael' to prove his point, when the word is not even in the text. Significantly, neither Mubarrad nor Ibn Khaldun, who both reproduce this letter, notice the mistake. Even Bukhari, at the beginning of his Kitab al-Manaqib cites something as 'revealed' that was not in the Koran. These mistakes were made after a written existed; it's scarcely credible that mistakes would not have crept in while the text was still transmitted orally.

Further confusion resulted from the lack of diacritical marks. For example, Hamza, who later helped invent point notation, confesses to having confused 'la zaita fihil' (no oil in it) with 'la raiba (no doubt) because of the lack of points. (So the lack of pointing could quite dramatically alter meaning!) Eventually a system of pointing based on Aramaic was adopted, though the caliph Ma'mun (198-218 AH) is said to have forbidden the use of both diacritical and vowel marks. Variant traditions of pointing developed over time, usually with little difference to sense, but in some places the differences in pointing resulted in greatly different meanings.

Sometimes the textual variants look like deliberate attempts to amend the text (e.g. 24:16- did the pre-Islamic Arabs only worship inathan (females) or authanan (idols)? ). Sometimes the Readers used historical research to supplement grammatical studies in determining the authentic text. For example Ibraham was chosen over Ibrahim (which seems to be necessary for the rhyme.) Also, three different ways of vowelling sura 30:1 result in three different meanings. One awkward rendition was chosen because it fits history.

Part Three: THE SOURCES OF THE KORAN

Chapter Eleven: What Did Muhammad Borrow From Judaism? (pp. 165-226)
-Abraham Geiger

THOUGHTS BELONGING TO JUDAISM WHICH HAVE PASSED OVER INTO THE KORAN?

Conceptions Borrowed from Judaism (pp. 166-172)

That these 14 words of Hebrew origin are used in the Koran suggests that ideas about divine guidance, revelation, and judgement after death were all borrowed from Judaism by Islam. Otherwise why wouldn't Arabic words have been used?

Views borrowed from Judaism (pp. 172-185)

1. Doctrinal views
   1. Unity of God
   2. Creation - 6 days, 7 heavens (asserted in Chagiga, also the '7 paths' is used in the Talmud), 7 hells – including 7 gates and trees at the gates
   3. Mode of Revelation
   4. Retribution, including the last judgement and Resurrection – e.g. linkage of resurrection and judgement, evil state of the world before the Messiah/Mahdi, the war between Gog and Magog, a person's body will testify against them (e.g. S. 24:24), idols will be cast into hellfire, the wicked will be allowed to prosper so as to increase their iniquity. 1000 years is like a day to the Lord, the resurrected person will appear in the clothes in which he is buried
   5. Doctrine of spirits - similar beliefs regarding angels and demons (djinn). Though Islam has a much more earthy idea of paradise, some similarities remain.
2. Moral and Legal Rules
   1. Prayer
      • Matches the rabbis' positions for prayer (standing, sitting, reclining) see Sura 10:13
      • shorten prayer in war
      • prayer forbidden to the drunken
      • prayer must be vocalised by not said loudly
      • Daybreak discerned by the ability to distinguish a blue (black) from a white thread
   2. Woman
      • divorced woman waits three months before remarriage
      • suckling time is two years
      • same limits on intermarriage
3. Views of Life
   Death with the righteous is to be prized – S. 3:191 and Num. 23:10
   Full understanding at 40 years – S. 46:14 and Aboth 5:21
   Interceding effectively leads to reward – S. 4:87 and Baba Kamma 92
   At death family and goods don't follow a person, only works do – Sunna 689 and Pirke Rabbi Eliezer 34

Stories Borrowed from Judaism (pp. 185-223)

We can assume that Muhammad acquired the Old Testament narratives from the Jews, because nothing is included that would be of particular interest to Christians.

Patriarchs (pp. 187-204)

1. From Adam to Noah
   • Creation – Adam is wiser than the angels are because he could name the animals (S. 2:28-32) c.f. Midrash Rabbah on Numbers para. 19, Midrash Rabbah on Genesis para. 8 and 17, and Sanhedrin 38
   • The story of Satan refusing to worship Adam (S. 7:10-18; 17:63-68, 18:48, 20:115, 38:71-86) was explicitly rejected by the Jews. c.f. Midrash Rabbah on Genesis para. 8
   • Cain and Abel – sacrifice and murder.
     Koran – raven tells Cain how to bury the body (S. 5:31)
     Jews – raven tells parents how to bury body (Pirke Rabbi Eliezer Ch. 21)
     Koran – slaying a soul is like slaying all mankind (S. 5:35) this is taken out of context from Mishna Sanhedrin 4:5
   • Idris (Enoch) – taken to Paradise after death and raised to life again. c.f. S. 19:58 with Gen. 5:24 and Tract Dereen Erez (cited in Midrash Yalku Ch. 42)
2. From Noah to Abraham
   • Angels living on earth, lusting after women and dividing marriages. S. 2:96 – alludes to Mdr. Abhkhir (quoted in Midr. Yalkut Ch. 44)
   • Noah – role as teacher and seer and the flood of hot water both match rabbinical ideas. [Compare S. 7:57-63, 10:72-75, 11:27-50, 22:43, 23:23-32, 25:39, 26:105-121, 29:13-14, 37:73-81, 54:9-18, 71:1ff with Sanhedrin 108, and S. 11:40 with Midrash Tanchuma, Section Noah, S. 11:42, 23:27 with Rosh Hashanan 162.] Noah's words are indistinguishable from Muhammad's (or Gabriel/Allah).
3. Abraham to Moses
   • Abraham – Archetypal prophet, friend of God, lived in temple, wrote books. Conflict over idols lead to danger of being burned alive but he was rescued by God. (Compare S. 2:60, 21:69-74, 29:23-27; 37:95-99 with Midrash Rabba on Genesis para. 38). So strong is Muhammad's identification with Abraham that he places words in Abraham's mouth that are not suitable to anyone outside Muhammad's context (e.g. S. 24:88, 29:17-23)
   • Joseph is the subject of almost all of the 12th sura. Additions to the Biblical story are derived from Jewish legends. (e.g. Joseph is warned away from Potiphar's wife in a dream (s. 12:24, Sotah 6:2), Egyptian women cut their hands because of Joseph's beauty (S. 12:31, compare with references in Midrash Yalkut to 'The Great Chronicle'.)

Moses and His Time (pp. 201-216)

This is very similar to the Biblical account, but with some additions from Jewish fables and some errors.

• The infant Moses refused the breast of Egyptian women (S. 28:11, Sotah 12,2)
• Pharaoh claims divinity (S. 26:28, 28:38, Midrash Rabba on Exodus para. 5)
• Pharaoh eventually repents (S. 10:90ff, Pirke Rabbi Eliezar section 43)
• God threatens to overturn the mountain onto the Israelites (S. 2:60, 87; 7:170, Abodah Zerah 2:2)
• There is a confusion as to the exact number of plagues – is it 5 (S. 7:130) or 9 (S. 17:103; 27:12)
• Haman (S. 28:5,7,38; 29:38; 28:38) and Korah (S. 29:38; 40:25) are thought to be advisors to Pharaoh.
• Miriam the sister of Aaron is also thought to be the mother of Jesus (S. 3:30ff, 29:29, 46:12)

The Kings Who Ruled Over Undivided Israel (pp. 216-220)

Very few particulars are given about Saul or David. Solomon is discussed in much more detail. The story about the Queen of Sheba (S. 27:20-46) is virtually identical to the 2nd Targum on the Book of Esther.

Holy Men After the Time of Solomon (pp. 220-223)

Elijah, Jonah, Job, Shadrach, Mishach, Abednego (not by name), Ezra, Elisha

Conclusion: Muhammad borrowed a great deal from Judaism – both scripture and legend. He freely altered what he heard. 'Conceptions, matters of creed, views of morality, and of life in general, and more especially matters of history and traditions, have actually passed over from Judaism into the Koran.' (p. 222)

Appendix: Statements in the Koran Hostile to Judaism (pp. 223-226)

Muhammad's aim was to bring about the union between all religions, but Judaism, with its host of laws, stood in his way. So he made a break with the Jews, declaring them enemies (S. 5:85) who killed the prophets (S. 2:58, 5:74), thought themselves favoured by God (S. 5:21) believed they alone would enter paradise (S. 2:88, 62:6), held Ezra to be the son of God (S. 9:30), trusted in the intercession of their predecessors (2:128, 135), and perverted the Bible (S. 2:73). To emphasise this break he changed some of the Jewish traditions. For example: (1) Supper precedes prayer (sunna 97ff) in opposition to the Talmud's adamant stance that prayer has priority, (2) Sex is permitted during Ramadan. The Talmud forbids it on the evening of fasts. Also, men may only remarry the wives they have divorced if the woman has first married and divorced someone else (S. 2:230). This is in direct opposition to the Bible, (3) Most of the Jewish dietary regulations are removed, (4) Muhammad cites 'eye for eye' and rebukes the Jews for replacing it with the payment of money (S. 5:49).

Chapter Twelve: The Sources of Islam (pp. 227-292)
-W. St. Clair-Tisdall

Ch. I – Views of Muslim Divines as to the sources from which Islam sprang (232)

The Koran is direct from heaven from God via Gabriel to Muhammad. God is the only 'source' of Islam.

Ch. II – Certain Doctrines and Practices of the Arabs in the "Days of Ignorance" Maintained in Islam (pp. 232-236)

Islam retains much from pre-Islamic Arabia including Allah, the name for God. The concept of monotheism did exist in the jahiliyya – even the pagans conceived of a supreme God that ruled over all the others. There are hints that some idolatry would remain (e.g. the Satanic verses). The Ka'ba was the masjid of many tribes as early as 60 BC, and the pagans first had the tradition of kissing the black stone. Two passages from the Sabaa Mu'allaqat of Imra'ul Qays are quoted in the Koran (S. 54:1, 29:31&46, 37:69, 21:96, 93:1). There is also a hadith where Imra'ul mocks Fatima because her father is plagiarising him and claiming to be quoting revelation.

Ch. III – How Far Some of the Doctrines and Histories in the Koran and Tradition were taken from Jewish Commentators, and Some Religious customs from the Sabaeans (pp. 236-257)

Sabeans – a religious group now disappeared. Among the little known about them we see the following customs:

• 7 daily prayers, 5 of them at the same times as those chosen by Muhammad
• prayed for the dead
• fasted 30 days from night to sunrise
• observed Eed from the setting of 5 starts
• venerated the Ka'ba

Jews – Three important tribes lived in the vicinity of Medina: Bani Quraiza, Qainuqa'a, and Nadhir.

1. Cain and Abel – S. 5:30-35, compare with the Targum of Jonathan ben Uzziah, the Targum of Jerusalem. Specifically there are parallels with Pirke Rabbi Eleazer(the story of the raven teaching people how to bury), and with Mishnah Sanhedrin (the commentary about the shedding of blood).
2. Abraham saved from Nimrood's fire – (S. 2:260, 6:74-84, 21:52-72, 19:42-50, 26:69-79, 29:15,16; 37:81-95, 43:25-27, 60:4) taken from Midrash Rabbah (Gen. 15:7). The parallels are especially clear when the relevant hadith are consulted. The only significant difference is that in the Koran Abraham's father is called Azar, not Terah, but Eusebius tells us that this is similar to the name used in Syria. This Jewish commentary was the result of a mistaken translation of 'Ur', which means 'city' in Babylonian, but was apparently mistaken for the word 'Or' meaning fire, so the commentator (Jonathan ben Uzziah) thought Abraham had been delivered out the 'fiery oven' of the Chaldeans.
3. Visit of the Queen of Saba (Sheba) to Solomon (S. 21:17ff) is taken from the 2nd Targum of the Book of Esther
4. Harut and Marut (S. 2:96, especially Araish al-Majalis – the commentary on that ayat) similar to several accounts in the Talmud, especially Midrash Yalkut. The stories are the same except for the manes of the angels. The manes in the Koran are the same as those of two goddesses worshipped in Armenia.
5. A few other things taken by Islam from the Jews
   • 'Sinai overhead' – S. 2:172 and Abodah Sarah
   • The golden calf lowing – s. 2:90 and Pirke Rabbi Eleazer
   • Also, the Koran uses the word 'Sameri' for the man who built the golden calf – but Samaritans didn't exist until 400 years after Moses.
6. A few other Jewish Matters
   • Many words in the Koran are not Arabic but Hebrew, Chaldaean, Syriac, etc…
   • The concept of 7heavens and 7 hells are in the Jewish books Hagigah and Zohar (S. 15:44, 17:46)
   • God's throne is above the waters (S. 11:9) from the Jewish Rashi
   • The angel Malik rules over Jehennam – the names is taken from Molech, the ruler of fire in pagan Palestine.
   • There is a wall or partition separating heaven and hell (S. 7:44) – a variety of places in the Jewish Midrash.
7. Religious usages of Islam taken from the Jews
   • Daybreak begins when you can distinguish a white from a black(Islam)/blue(Jewish) thread (S. 2:83, Mishnah Berakhoth)
   • S. 21:105 is a quotation of Psalm 37:11. How could the Koran quote the Psalms unless it came after them, therefore either the Psalms must be eternal as well, or the Koran is not.
   • The Koran is preserved on heavenly tablets (S. 85:21-22) – similar to the stone tablets of the decalogue (Deut. 10:1-5) which Jewish legend had embellished to include the entire Torah, Writings, Prophets, Mishnah, and the Gemara (Rabbi Simeon).

Ch. IV – On the Belief that Much of the Koran is Derived from the Tales of Heretical Christian Sects

Many heretics were expelled from the Roman Empire and migrated to Arabic before the time of Muhammad.

1. The Seven Sleepers, or Companions of the Cave (S. 18:8-26) is a story of Greek origin found in a Latin work of Gregory of Tours ('Story of Martyrs' 1:95) and was recognised by Christians as pious fiction.
2. The History of Mary (S. 19:16-31, 66:12, 3:31-32&37-42, 25:37). Mary is said to be the sister of Aaron, the daughter of Imran (Hebrew Amran the father of Moses), and the mother of Jesus. The hadith tell us that Mary's mother was an aged, barren woman who promised to give her child to the temple if God gave it to her (from the Protevangelium of James the Less). The hadith also explain that the casting of rods mentioned in the Koran refers to when 6 priests were vying for who would raise Mary. They threw their rods into the river, only Zaccharias' rod floated (from the History of our Holy Father the Aged, the Carpenter (Joseph), and Arabic apocryphal book). Mary was denounced as an adulteress but pleaded her innocence (from Protevangelium a Coptic book on the Virgin Mary), and gave birth under a palm tree that aided her (from History of the Nativity of Mary and the Saviour's Infancy)
3. The Childhood of Jesus – Jesus spoke from the cradle and created birds of clay which he then turned to life (S. 3:41-43, 5:119), from The Gospel of Thomas the Israelite and The Gospel of the Infancy Ch. 1, 36, 46. Jesus was not really crucified (s. 4:156) in accordance with the heretic Basilides (quoted by Iraneus). The Koran erroneously thinks that the Trinity consists of father, mother, and son (s. 4:169, 5:77).
4. Some other stories from Christian or heretical writers: In the hadith (Quissas al-Anbial) God sends angels together dust to create Adam and Azrael brings it from every quarter (Ibn Athir via Abdul Feda). This is from the heretic Marconion who argued that it was an angel (the 'God of the law') who created people, not the true God. The balance of good and bad deeds (S. 42:16, 101:5-6) is from the 'Testament of Abraham' and from the Egyptian 'Book of the dead.' Two New Testament verses are alluded to: (a) camel through the eye of a needle (S. 7:38, Mt. 19:24), God has prepared for the righteous things that eyes have not seen nor ears heard (Abu Hureira quoting the prophet in Mishkat of the Prophet, 1 Cor. 2:9).

Ch. V – Some Things in the Koran and Tradition Derived from Ancient Zoroastrian and Hindu Beliefs (pp. 275-286)

Arabian and Greek historians tell us that much of the Arabian peninsula was under Persian rule before and during Muhammad's life. Ibn Hisham tells us that the stories of Rustem, Isfandiyar and ancient Persia were told in Medina and the Quraish used to compare them with tales in the Koran (e.g. the tales told by Nadhr, son of al-Harith).

1. Ascent (Miraj) of the prophet (S. 17:1) – There is a great variation in interpretation. Ibn Ishaq quotes A'isha and the prophet as saying this was an out of body journey. Muhyiad-Din [ibn al-'Arabi] agrees. But Ibn Ishaq also quotes the prophet saying that it was a literal journey. Cotada relates the prophet saying that it was a literal journey into the 7th heaven. In a Zorastrian story the Magi send one of their number into heaven to get a message from God (Ormazd) (from a Pahlavi book Arta Viraf Namak – 400 B.H.) Also, the 'Testament of Abraham' tells of Abraham being taken up to heaven in a chariot.
2. Paradise – filled with houris (S. 55:72, 56:22) – like the 'paries' in Zorastrianism. The words 'houry', 'djinn', and 'bihist' (Paradise) are derived from Avesta or Pahlavi sources. The 'youths of pleasure' (ghilunan) are also in Hindu tales. The name of the Angel of death is taken from the Jews (in Hebrew two names are given, Sammael and Azrael, the latter was borrowed by Islam), but the concept of the angel killing those in hell was taken from Zoroastrianism.
3. Azazil coming from hell – in the Muslim traditions he worshiped God 1000 years in each of the 7 heavens before reaching earth. Then he sat 3000 years by the gates of paradise trying to tempt Adam and Eve and destroy creation. This is very similar to the Zoroastrian tale regarding their devil (Ahriman) in the book Victory of God. The peacock agreed to let Iblis into Paradise in exchange for a prayer with magical qualities (the Bundahishnih) - an association also noted by the Zoroastrians (Eznik in his book Against Heresies).
4. The light of Muhammad was the first created thing (Qissas al-Anbia, Rauzat al Ahbab). The light was divided into 4, then each into 4. Muhammad was the first of the first divisions of light. This light was then placed on Adam and descended to the best descendent. This is virtually identical to the Zoroastrian view which described 4 divisions of light (the Minukhirad, Desatir-i Asmani, Yesht 19:31-37); the light was placed on the first man (Jamshid) and passed to his greatest descendent.
5. The Bridge Sirat is a concept from Dinkart, but it is named Chinavad by the Zoroastrians.
6. The concept that each prophet predicts the next prophet is from Desatir-i Asmani where each Zoroastrian prophet predicts the next one. Also, the openings of these books (i.e. the Desatir-i Asmani) is "In the name of God, the Giver of gifts, the Beneficent' which is similar to the opening of all the Suras 'In the name of God the Merciful, the Gracious.'
7. How could Muhammad have learned these stories? Rauzat al-Ahbab tells us that the prophet used to talk to people from all over the place. Al-Kindi accuses the Koran of including foolish old-wives tales. Also, in Sirat-Rasul we learn of the Persian, Salman, who advised Muhammad regarding the battle of the trench and was accused of helping compose the Koran. (The Koran mentions him, though not by name, in S. 16:105).

Ch. VI – The Hanefites: Their Influence on Muhammad and On His Teaching (pp. 286-292)

The influence of the Hanefites (Arab monotheists) on Muhammad is most reliably described by Ibn Hisham quoting Ibn Ishaq'a Sirat. Six Hanefites are mentioned by name – Abu Amir (Medina), 'Ummeya (Tayif), Waraqa (became a Christian), 'Ubaidallah (became a Muslim, moved to Abyssiniya and gave up Islam for Christianity), 'Uthman, Zaid (banished from Mecca, lived on Mt. Hira where Muhammad went to meditate) (the latter four were from Mecca).

Conclusion – All this said, the variety of sources does not mean than Muhammad had no role in creating Islam. But we see that as circumstances in his life changed, so too did his revelation. For example, s. 22:44 (pre-Hegira) permission is given to fight when persecuted, but in s. 2:212-214) war is commanded even during the sacred months (post-Hegira). Then again after the Banu Quraiza are conquered comes s. 5:37 commanding dire punishments for anyone who opposes Muhammad. Towards the end of Muhammad's life the sacred months come back into favour (s. 9:2,29), but Muslims are also commanded to kill idolaters wherever they may find them, (even if they are not fighting against Islam!), because they do not profess the true religion.

Chapter Thirteen: The Jewish Foundation of Islam (pp. 293-348)
-Charles Cutler Torrey

Allah and Islam (pp. 293-330)

Muhammad was trying to create a religious history for the Arabs, but Arabian religious history did not provide many sources for him. What references there are occur mainly in the Meccan period. He refers to Hud, the prophet of the people 'Ad; Salih, the prophet of the Thamud; and Shu'aib, prophet of Midian. All pagan customs not directly involving idolatry were preserved in Islam, e.g. the rituals of the Haj.

After exhausting the Arabian possibilities Muhammad began to rely on Jewish material because it was well-known and would give the new religion greater credibility in the wider world. In addition to apocryphal works, Muhammad must have been familiar with the canonical Bible, especially the Torah. He only knows the prophets with interesting stories and is therefore ignorant of Isaiah, Jeremiah, Ezekiel and all the minor prophets except Jonah. From popular tales the Arabs knew that the Jews felt that they had descended from a common ancestor, Abraham, via Isma'il and Isaac respectively. Hagar is not mentioned in the Koran. The Koran says that they built the Ka'ba (though later Muslim doctrine says that Adam built it and Abraham cleansed it of idols). It is possible that the 'hanifs' (Arab monotheists following the religion of Abraham) are an invention of later Islam. The story of Iblis (or Shaitan) prostrating himself before Adam (38:73-77) may not refer to worship as there is a possible Jewish source for this story in Sanhedrin 596 and Mir. Rabba 8. Shu'aib is probably the Biblical Jethro. 'Uzair is Ezra, and the Jews are accused of declaring him to be the son of God. Idris is also Ezra (the Greek name). Hebrew chronology is very week in the Koran, e.g. Muhammad seems to associate Moses near to Jesus (as Moses' sister is also Jesus' mother).

'Isa ibn Maryam is Jesus. Very little is known about him by Muhammad and there are no uniquely Christian doctrines in the Koran. The little that was known about Jesus came from (1) the facts and fancies that were spread throughout all Arabia, and (2) a little via the Jews. The name 'Isa is itself inappropriate, it should be Yeshu in Arabic. Either it was given by the Jews (associating Jesus with their ancient enemy Esau) or it is a corruption of the Syriac name (Isho). In the Koran itself Jesus doesn't have a position higher than Abraham, Moses, or David. This elevation occurred later in the caliphate when the Arabs had closer contacts with Christians. A few Christian terms (e.g. Messiah, Spirit) work their way into the Koran without any real understanding of what they mean. It was probably the migration to Abyssinia that increased Muhammad's interest in the Christian stories. Rudolph and Ahrens argue that if Muhammad had learned about Jesus from the Jews then he would have ignored or insulted him. But many Jews appreciated Jesus as a teacher while rejecting Christian dogmas. Also, Muhammad was aware of the large Christian empire, so he would have distrusted anyone who insulted Jesus. The only information about Christ in the Koran is the kind of stuff that wouldn't bother the Jews. The Koran's view of Jesus' mission is: (1) confirm the true doctrine of the Torah, (2) preach monotheism, (3) warn against new sects. S. 15:1-15 is a literary connection with the New Testament (Lk. 1:5-25, 57-66). This is the story of Zechariah and John was probably related by a learned man but not a Christian as it was isolated from any association with Jesus' birth. In summary, there is nothing particularly Christian about Jesus in the Koran.

Torrey now digresses to a discussion of the composite Meccan suras, following the traditional Muslim accounts closely. He points out the implausibility of Meccan and Medinan verses being intermingled if in fact the prophet was publicly reciting his revelations and having them memorised by his followers as they were revealed. Would it not cause confusion (or scepticism) to be continually inserting new material into previously revealed suras? The traditional commentators frequently neglect the Jewish population in Mecca that may have been the target of some ayat in the Meccan suras. In fact, Muhammad's personal contact with Jews was longer and closer pre-Hijra than post-Hijra. Why would we assume that there was no hostility to Muhammad from the Meccan Jews? And, after the eviction or butchery of the yews in Yathrib, it's scarcely surprising that the Jews quickly left Mecca. Torrey recommends considering the Meccan suras to be complete without interpolations unless there is unmistakable proof to the contrary. Doing this decreases the variation in style and vocabulary assumed to exist between the two periods. [NB: Basically he is arguing for literary criticism instead of form criticism.]

'The origin of the term Islam' (pp. 327-330)

Traditionally 'Islam' is said to mean 'submission', especially to Allah. But, this is not the normal meaning one would expect of the 4th stem of the verb 'salima'. It is especially strange since 'submission' is not a prominent feature of Muhammad or his religion nor especially emphasised in the Koran. It is, however, an important attribute of Abraham, especially in his potential sacrifice of Ishmael.

The Narratives of the Koran (pp. 330-348)

Muhammad's use of stories about prophets served two functions: (1) it provided a clear connection with the previous 'religions of the book', and (2) it showed his countrymen that his religion had been preached before and those who rejected it were punished. But, Muhammad's storytelling was boring and he was mocked by an-Nadr ibn al-Harith who insisted that his own tales of Persian kings were far more interesting. (After the battle of Badr the prophet had his revenge and slew an-Nadr.) Muhammad himself appreciated a good story and incorporated pretty bits of folk tale into the Koran where he could. However, this provided a dilemma for Muhammad. If he merely reproduced tales he would be accused of plagiarism, but if he changed them he would be accused of falsifying. He couldn't just invent new stories, for his imagination was vivid but not creative. All of his characters talk the same way and he has very little sense of action. His solution was to repeat the stories he had learned, but in fragments, using introductory words which imply that he could tell more if he chose (e.g. 'and when…', 'and then there was that time…')

The story of Joseph is the most complete narrative in the Koran, but it is still annoyingly short in detail. Why were the women given knives? What does the banquet have to do with anything? Why was Joseph put in prison after Potipher's wife confessed? Solomon and the Queen of Sheba (27:16-45) is taken directly from the Haggada (see above pp. 181-186). Jonah (37:139-148) is a condensation of the Biblical account, but the name given is based on the Greek rather than Hebrew form. Saul and Goliath ('Talut' and 'Jalut') is a confusion of the story of Gideon (Jdg. 7:47) with that of David and Goliath. The story of Moses (s. 28:2-46) is a summary of most of Ex. 1-4, though Muhammad does not associate Moses with the Israelites. Haman is believed to be Pharaoh's vizier (also in s. 29 and 40). As in the Talmud (Sotah 126) the baby Moses refuses to suckle at an Egyptian breast. The marriage of Moses in Midian is loosely patterned after Jacob and Rachael; and a tower (virtually identical to the tower of Babel) is built by Pharaoh to reach Allah. This narrative illustrates the freedom which Muhammad felt as a prophet to alter the Biblical tradition.

Sura 18 is unusual because the stories in it are not from the Bible or Rabbinic literature, and Muhammad makes not mention to it elsewhere in the Koran.

1. The seven sleepers is from the legend of 7 Christian youths who fled from Ephasus to the mountains to escape the persecution of Decius (250 AD). Though a Christian tale it seems to have come to Muhammad via the Jews for several reasons (a) The hadith say that the Jews of Mecca were especially interested in this story (See Baidawi on vs. 23), (b) the rest of the stories in the chapter seem to have come via a Jewish rescension, and (c) internal evidence points to verse 18, which mentions the importance of 'clean' food, a concept important to Jews, not to Christians. There is nothing uniquely Christian about this tale. It could just as easily have been Israelite youths. Apparently the legend existed in different forms and Muhammad was challenged to know what was the correct number of youths. The Koran diffuses the challenge by insisting that only God knows the right answer.
2. The next story is a common parable of a god-fearing poor man vs. an arrogant, impious rich man. The latter is punished.
3. Then we have the story of Moses searching for the fountain of life which is the same as an episode from the legend of Alexander the Great with the name changed. This legend has roots in the Gilgamesh epic.
4. Finally, the narrative of the 'Two-horned' hero is again from Alexander the Great. He journey's to the place of the setting sun and to the place of its rising, as an emissary of God. He is protected against Gog and Magog (Yajuj and Majuj in the Koran) and Alexander builds a great wall. These fantasies echo those found in the Haggada, which reinforces the possibility of a Jewish source for the entire sura, likely a single document.

So, the sources of the Koran used by Muhammad include:

1. Biblical narrative with alteration
2. Jewish Haggada, well preserved
3. A small amount of ultimately Christian material from Aramaic.
4. Legends common to world literature introduced via the Jews at Mecca.

All of these were altered and rearranged for the purpose of providing his listeners with an Arabian revelation with enhanced credibility because it could be seen as part of a universal divine revelation.

Part Four: MODERN TEXTUAL CRITICISM OF THE KORAN

Chapter Fourteen: Literary Analysis of Koran, Tafsir, and Sira: The methodologies of John Wansbrough (pp. 351-363)
-Andrew Rippin

Christianity and Judaism are both seen as religions rooted in history. 'What really happened' is seen as an important criteria for determining the truth or falsehood of the religion. It assumes that the sources available to us contain discernible historical data which enable us to achieve positive historical results.

Modern scholarship of Islam has the same desire to achieve positive results, but the literary qualities of the available sources are often overlooked. Neutral testimony, archaeological data, datable documents, and evidence from external sources, are profoundly lacking. The few external sources that we have (as recounted by Crone and Cook in Hagarism) have questionable authenticity and are based on polemic. Internal sources are recorded two centuries after the event, influenced by the intervening years and intended to provide a 'salvation history' legitimising the faith and the scripture of Islam. For example, the stories known as asbab al-nazul (occasions of revelation) are significant not for their historical value but for their exegetical value – they provide a framework for interpretation of the Koran. Yet these basic literary facts are often ignored by historians.

The Nature of the Sources

John Wansbrough (SOAS) argues for a critical literary assessment of the sources so as to avoid the inherent theological view of history. His two major books are Qur'anic Studies: Sources and Methods of Historical Interpretation, dealing with "the formation of the Koran along with the witness of exegetical writings (tafsir) to that formation., and The Sectarian Milieu: Content and Composition of Islamic Salvation History, examining the traditional biographies of Muhammad to see "the theological elaboration of Islam as a religious community" especially "questions of authority, identity, and epistemology." [pg. 354] Wansbrough's basic methodology is to ask the question: What is the evidence that the stories are accurate regarding the formation of the scripture and the community? The earliest non-Islamic sources testifying to the Koran are the 2nd/8th century. Islamic sources (excluding those whose primary purpose was defending the canon) suggest that the Koran itself was not totally fixed until the 3rd/9th century. Manuscript evidence doesn't allow for much earlier dating.

Many scholars ask why they should not trust Islamic sources. In answer Wansbrough, rather than pointing to contradictions between and within them (like John Burton, The Collection of the Koran), argues that "the entire corpus of early Islamic documentation must be viewed as 'salvation history.' What the Koran is trying to evidence, what tafsir, sira, and theological writings are trying to explicate, is how the sequence of worldly events centred on the time of Muhammad was directed by God. All the components of Islamic salvation history are meant to witness the same point of faith, namely, an understanding of history that sees God's role in directing the affairs of humankind." [pp. 354-355] Salvation history is not attempting to describe what really happened, it is attempt to describe the relationship between God and men and vice versa. (Wansbrough does not use 'salvation' with it's Christian connotations, i.e. the saving of an individual soul from damnation, but in a more general literary sense that could just as easily be called 'sacred' history.)

This concept has been fully developed within biblical and Mishnaic studies by the likes of Bultmann and Neusner. "All such works start from the proposition that the literary records of salvation history, although presenting themselves as being contemporary with the events they describe, actually belong to a period well after such events, which suggests that they have been written according to later points of view in order to fit the purposes of that later time… The records we have are the existential records of the thought and faith of later generations." [pp. 355-356] Goldziher and Schact recognised that many of the sayings attributed to the prophet were invented to settle legal and doctrinal disputes in later generations. However most scholars since Schact have tended to ignore the implications of his work. Wansbrough argues that we do not (and probably cannot) know what 'really happened'. Literary analysis can only tell us about the disputes of later generations. The whole point of Islamic salvation history is to adapt Judeo-Christian religious themes for the formulation of an Arabian religious identity. The Koran itself demands that it be placed within a Judeo-Christian context (e.g. the line of prophets, sequence of scriptures, common narratives). "This notion of extrapolation is, in a sense, the methodological presupposition that Wansbrough sets out to prove within his books by posing the question: If we assume this, does the data fit? At the same time he poses the question: What additional evidence appears in the process of the analysis to corroborate the presupposition and to define it more clearly?" [pg. 357] Attacking the presupposition misses the entire point To evaluate his work one must first weigh the evidence and the conclusions proposed.

Wansbrough's Approach to the Sources (pp. 358-363)

Wansbrough argues that modern studies of the Koran, even those which purport to use modern biblical methodologies (e.g. Richard Bell), acquiesce to the traditional interpretation of the data. Major reasons for this include: (1) increasing specialisation means that there are fewer scholars capable of interacting with the wide variety of necessary languages and religions. Most think that a knowledge of Arabic and 7th century Arabia is sufficient. (2) The irenic approach (e.g. Charles Adams), aimed at appreciation of Islamic religiousness, avoids the basic question 'How do we know?'

In his analysis of the basic character of the Koran Wansbrough identifies four major motifs common to monotheistic imagery: divine retribution, sign, exile, and covenant. He points out that the Koran is written in a 'referential' style, presupposing detailed audience knowledge of the Judeo-Christian traditions which can be alluded to with only a few words without losing meaning (similar to Talmudic references to the Torah). Only as 'Islam' moved out of the Arabian peninsula and obtained a fixed identity (based on political structure) does the Koran become detached from its original intellectual environment and require explanation – i.e. the tafsir and sira. The similarities between the Koran and Qumram literature show a "similar process of biblical-textual elaboration and adaptation to sectarian purposes." [pp. 360] So the Koran is a composite of referential passages developed on the context of Judeo-Christian sectarian polemics joined together through a variety of literary and narrative conventions. Textual stability goes hand-in-hand with canonisation and was not really feasible until political power was well established; "thus the end of the 2nd/8th century becomes a likely historical moment for the gathering together of oral tradition and liturgical elements leading to the actual concept "Islam." [pg. 361] This coincides with the rise of literary Arabic. Wansbrough analyses Koranic tafsir into five genres – haggadic, halakhic, masoretic, rhetorical, and allegorical – and then shows a chronological development of increasing concern with the textual integrity of the Koran and then with its use as scripture. The sira have some exegetical function, but are more important in providing a narrative of the Islamic version of salvation history. Much of the contents of the sira fit nicely as elaborations of 23 well-known polemical motifs traditional to the Near Eastern sectarian milieu.

Critics have largely accused Wansbrough of creating a method that determines results rather than allowing material determining results. However, Rippin points out that the traditional theologico-historical methods are just as likely to condition results. What is needed is for scholars to become more aware of the limitations of their own methods and to be prepared to considered the validity of other methods. A closer examination of the basic data is necessary to determine the validity and implications of Wansbrough's method.