Chapter Five

The Compilation of
the Text of the Qur'an

1. THE INITIAL COLLECTION OF THE QUR'AN

THE FIRST COLLECTION UNDER ABU BAKR

Most books are written out as a complete text from cover to cover with the outline from the introduction to the conclusion planned well in advance before a word is written. The Qur'an, on the other hand, was never compiled into book form during the time of Muhammad and it was only his death which actually completed its text. It came to him during his lifetime in staggered portions and although its final form had been settled in principle prior to his death there was no single collection of its surahs and passages in a written form in anyone's possession.

While he lived there was always a possibility that fresh revelations could be added to the text. Indeed it would have seemed inappropriate to any of his companions to attempt to codify it in written form, especially as the main means of retaining its contents at the time was in the memory of those who had consciously endeavoured to learn the Qur'an by heart. Some of it had been written out on different materials such as pieces of wood, palm-leaves and the like. It also appears that new passages were coming to Muhammad with increasing frequency shortly before his demise, making an attempt at a single collection even more improbable:

It is expressly stated by one of the major Muslim scholars of the Qur'an in Islamic history that the text had been completely written down and carefully preserved but that it had not been assembled into a single location during the lifetime of the Prophet (As-Suyuti, Al-Itqan fii `Ulum al-Qur'an, p.96). Once the primary recipient of the Qur'an had passed away, however, it was only logical that a collection should be made of the whole Qur'an into a single text. The traditions of Islam state that four men knew the Qur'an during Muhammad's lifetime in its entirety, one of whom was Zaid ibn Thabit (Sahih al-Bukhari, Vol.5, p.97). He was soon called upon to compile a written codex of the text.

Shortly after the Prophet's death a number of tribes recently converted to Islam in the Arabian Peninsula reverted to Arabian paganism and revolted against Muslim rule. Muhammad's successor Abu Bakr sent an army to subdue them and in the subsequent Battle of Yamama a number of the companions who knew the Qur'an directly from their Prophet were killed. Others with a similar knowledge also passed away and with them their own readings of the text:

Abu Bakr realised that there was a danger that the Qur'an might be lost if any more of its best-known reciters passed away. He told Zaid that he was a young man above suspicion who had been known to write down portions of the Qur'an and he accordingly commissioned him to search for its portions and collect it into a single codex. Zaid was initially taken back at the idea and later recorded what followed:

Zaid eventually approved after Abu Bakr and `Umar had pressed the urgency of the task upon him and set about collecting the Qur'an. It was to be a unique undertaking as the contents of the book were spread widely among the companions and were recorded on various materials. His hesitancy at first shows that the project would not be easy. He did not believe that either he or any of the other companions who knew the text well could be relied on simply to write it out from memory. Instead he proceeded to make a thorough search for the text from a variety of sources and he recorded his investigation in these words:

The two primary sources, amongst the others mentioned, were later defined as ar-riqa`a ("the parchments") and sudur ar-rijjal ("the breasts of men"), namely not only texts from those who had memorised the Qur'an but also whatever written materials he could find (As-Suyuti, Al-Itqan, p.137). Nonetheless he was not the only companion of the Prophet to begin to codify the Qur'an into a single written text (a mushaf) and may not even have been the first to succeed in doing so. The following tradition states that another of the early reciters was the first to write it down and collect it:

This Salim is one of only four men who was recommended by Muhammad as the best reciters of the Qur'an from whom its contents should be learnt (Sahih al-Bukhari, Vol.5, p.96) and he was one of the qurra ("reciters") killed at the Battle of Yamama. As it was only after this battle that Zaid began to collect his material Salim's codex must indeed have preceded his as the first written copy of the Qur'an. Nonetheless Islamic tradition pays primary attention to Zaid's codex not only because it was called for by the first Caliph himself but also for other reasons which will shortly become apparent.

PERSPECTIVES ON ZAID'S COLLECTION OF THE QUR'AN

Muslims claim that the Qur'an as it stands today is an exact record of the original without so much as a dot or stroke ever having been lost, changed, or substituted in any way. This is a strange claim to make for a book which had to be compiled piecemeal from various sources scattered among the companions of Muhammad, particularly in the light of further evidences that some passages have been lost, that others have been abrogated, and that other codices compiled about the same time as Zaid's had numerous readings that differed from his and from each other's. These evidences will shortly be considered. At this point, however, it must be said that Zaid's final compilation was the result of an honest human attempt to collect the Qur'an as far as he was able to and there is no reason to suspect that it does not generally project the text as it stood by the time of the Prophet's death.

There are evidences even at this early stage, however, that portions of the Qur'an were irretrievably lost at the Battle of Yamama when many of the qurra who had memorised whole portions of it had perished:

The negative impact of this passage can hardly be missed: lam ya`alam – "not known", lam yuktab – "not written down", lam yuwjad – "not found", a threefold emphasis on the fact that these portions of the Qur'an which had gone down with the qurra who had died at Yamama were lost forever and could not be recovered.

There are evidences in the tradition literature to show that even Muhammad himself was occasionally inclined to forget portions of the Qur'an. One of these taken from a major Hadith work reads as follows:

There is no evidence to suggest that Zaid had compiled an official or standard codex of the Qur'an even though Abu Bakr was the immediate successor of Muhammad as head of the Muslim community. The object was apparently to ensure that he was in possession of a complete written text to ensure its preservation. For the next twenty years virtually nothing is said of this codex other than that, by the time of `Uthman's caliphate, it was in the private possession of Hafsah, `Umar's daughter and one of Muhammad's wives, and was kept under her bed.

It is important in concluding a study of Zaid's text to analyse the comment he made about two verses of the Qur'an which he had searched for and had found only with Abu Khuzaimah. The full text of his eventual discovery is recorded in these words:

It is quite clear from this passage that Zaid was dependent on one source alone for the last two verses of Surat at-Tauba. In fact there is another tradition which shows that it was not Zaid who sought earnestly for the exact text of a pair of verses which he recalled but could not trace. In this record it is stated that it was Abu Khuzaimah himself who drew the attention of the compilers to a text they were overlooking:

The significant feature of this passage is the implication that Zaid and his redactors would have missed these verses completely had Abu Khuzaimah not mentioned them. He makes a point of the fact that he received them "directly" (tilqiyya) from the Prophet meaning obviously that he had heard them firsthand and had not obtained them from secondary sources. The passage goes on to say that Abu Khuzaimah was subsequently asked where they should be inserted in the Qur'an and he suggested they be added to the last part of the text to be revealed, namely the close of Surat at-Tauba (Bara`a in the text).

For many years there was no further development in reducing the text of the Qur'an to a standard form for the whole Muslim community. Events in the time of `Uthman's caliphate, however, led to the next stage.

2. `UTHMAN'S RECENSION OF ZAID'S COMPILATION

THE ORDER TO DESTROY ALL THE OTHER CODICES

The codex of Zaid ibn Thabit was clearly one of great importance and its retention in official custody during the caliphates respectively of Abu Bakr and `Umar testify to its key significance during the time of the Qur'an's initial codification. There can be little doubt, however, that this codex was at no time publicised during this period or declared to be the official text for the whole Muslim world.

There were a number of other masters among the qurra who had gone to great lengths to memorise the Qur'an. Islamic tradition states that by the time `Uthman became caliph twelve years after the death of the Prophet there were written codices in use in different provinces compiled by other well-known companions, in particular `Abdullah ibn Mas`ud and Ubayy ibn Ka`b. There was no official reaction at first to this development as Zaid's text had never been intended as an official copy and the credibility of these men in their knowledge of the Qur'an had never been doubted. They are mentioned along with two others as having been acknowledged by Muhammad himself during his lifetime as the foremost authorities on the Qur'an:

The special mention of the fact that Muhammad started with `Abdullah ibn Mas`ud indicates that the Prophet regarded him as the most knowledgeable Qur'an reciter among his companions. In fact, while the codices of this man and other prominent reciters became prominent in the developing Muslim world the codex of Zaid faded into virtual obscurity. It had simply receded into the private custody of Hafsah, one of the widows of the Prophet (Sahih al-Bukhari, Vol.6, p.478).

Seven years after his accession to leadership of the Muslim world, however, `Uthman was faced with a crisis which threatened to break up the Muslim world and undermine his unchallenged leadership over it. It came from the very areas where the other companions were so highly respected because of their unique knowledge of the Qur'an and the fame their codices enjoyed. Circumstances gave him an opportunity to severely subvert their authority by ordering that their codices be destroyed in the interests of standardising one text for the whole Muslim community. His opportunity came when the Muslim general Hudhayfah ibn al-Yaman, leading an expedition of Muslim forces from what is today Syria and Iraq, discovered that the people there were disputing with each about the reading of the Qur'an. The codex of `Abdullah ibn Mas`ud was the standard text of the Muslims at Kufa while that of Ubayy ibn Ka`b held sway in Damascus. Hudhayfah immediately reported the matter to `Uthman. What followed is described in the following tradition:

There is no suggestion that he considered the other codices to be unreliable. It was the divisions between the Muslims in the reciting of the text that made him realise the need to act as he foresaw the possibility that the Muslim world would break up into sects and divisions. By unifying the people on a single text of the Qur'an he saw an occasion to prevent such a partition occurring. The following tradition gives a balanced picture of the circumstances and explains why he chose Zaid's codex as the basis on which the Qur'an text was to be standardised for the Muslim community. `Ali is reported to have said of `Uthman:

The motive is twice stated in this extract to simply be the desire to bring consensus among the Muslims on the basis of a single text. If any of the leaders involved in the process had believed that the other codices were unreliable or that Zaid's was a perfect compilation of the Qur'an to the last dot and letter they would simply have ordered their scribes to transcribe it. Their decision to choose Zaid and Sa`id because of their proficiency in the reading and Arabic knowledge of the Qur'an respectively shows that, as at the time when Zaid's text was first commissioned, the aim was to get as close to the original as possible.

The question that might well be asked, however, is why Zaid's text was called for and why copies were made to be sent as the official copies of the Qur'an in each province while the others then in use had to be burnt and destroyed. One reason has already been given, namely to reimpose `Uthman's authority over the Muslims scattered throughout the Muslim provinces. Zaid's text, being kept in official custody at Medina, was ideal for this purpose. Also, it had not been in general public use so there had been no division about its contents. The standardising of a Medinan text at the seat of the Caliph's government enabled him to suppress the popularity of other reciters in areas where he was becoming unpopular. He was placing members of his own family, the descendants of Umayya who had opposed Muhammad until the conquest of Mecca, in positions of authority over them. Zaid's text was thus chosen not because it was believed to be superior to the others but because it suited `Uthman's purposes in standardising the text of the Qur'an.

The fact that none of the other texts was spared shows that not one of them, Zaid's included, was in complete agreement with any of the others. There must have been serious textual variants between the codices to warrant such drastic action. The order to consign all but one of the written texts (masahif) to the flames indicates that serious divisions existed between them. This was perhaps a circumstance to be expected when it is remembered that the Qur'an had not been reduced to a single text at Muhammad's death. At the time it was widely scattered piecemeal among a number of his companions and that mainly in their memories, the most fallible of sources.

THE REVISION OF ZAID'S CODEX OF THE QUR'AN

Muslims often claim that all that `Uthman sought to achieve was to cancel out the different readings of the Qur'an in its various dialects. The issue was, they say, purely one of eliminating different pronunciations. This line of reasoning is subjectively advanced to maintain the hypothesis that the Qur'an, in its written form, is a divinely preserved and therefore perfect text. There were no vowel points, however, in those early codices and any differences in pronunciation would not have appeared in the texts. He could only have ordered the burning of all other codices if there were serious differences in the text itself. Evidences will be given in the next section to show that this was indeed the case.

In fact, shortly after his decree had been put into effect, `Uthman enquired what the grievances were of the Muslims whose opposition to him was intensifying. One of their complaints was that he had "obliterated the Book of Allah" (Ibn Abi Dawud, Kitab al-Masahif, p.36). They did not accuse him simply of destroying their masahif (codices) but of burning the kitabullah, the Qur'an, itself. Although his action contributed towards the standardising of an official text it also left a keen antagonism as they believed he had ruined authentic manuscripts of the Qur'an compiled by some of Muhammad's closest companions.

There are further evidences that Zaid's codex was not at this time considered an infallible copy of the Qur'an. `Uthman not only ordered his text to be copied but also called for it to be revised at the same time. When he appointed the four redactors mentioned he chose the other three because they were from the Quraish tribe at Mecca while Zaid came from among the ansar of Medina. He said that, if they should differ at any point in respect of the language of the Qur'an, they were to overrule Zaid and write it in the Quraish dialect as it had been originally revealed in it (Sahih al-Bukhari, Vol.4, p.466).

At the same time Zaid, after the manuscripts had been copied out, suddenly remembered another text that was missing from the Qur'an:

A similar record of this omission is recorded in Sahih al-Bukhari (Vol.6, p.479). It shows that even Zaid's original attempt to produce a complete codex was not entirely successful. It is remarkable in the light of these evidences to hear Muslims not only claiming that the Qur'an in their hands today is an exact, perfect redaction of the original but also alleging that this proves the divine origin of the book. The facts show otherwise. It was not Allah who arranged the text in its present form but rather the young man Zaid and that only according to the best of his ability. Nor was it Muhammad who codified or standardised it for the Muslim ummah but `Uthman and that only after a complete revision of one codex at the expense of all the others. The Qur'an in the possession of Muslims today is simply a revised edition of Zaid's initial compilation.

Even after this time disputes still arose regarding the authenticity of the text. A good example concerns a variant reading of Surah 2:238 which, in the Qur'an standardised by `Uthman, reads "Maintain your prayers, particularly the middle prayer (as-salaatil wustaa), and stand before Allah in devoutness". The variant reading is given in this hadith:

`Aishah was a very prominent woman in Islam being one of the widows of the Prophet, and she would not have recommended such a change lightly. She ordered the scribe to add the words wa salaatil `asr meaning "and the afternoon prayer", giving Muhammad himself as the direct source of her authority for this reading. On the same page there is a similar tradition where Hafsah, another of his widows, ordered her scribe `Amr ibn Rafi to make the same amendment to her codex. It is known that Hafsah had a codex of her own in addition to the codex of Zaid in her possession. Ibn Abi Dawud refers to it as a separate manuscript under the heading Mushaf Hafsah Zauj an-Nabi (saw) ("The Codex of Hafsah, the Widow of the Prophet"). He specifically records this same incident as a variant reading in her codex:

Ibn Abi Dawud also states on the same page that this variant was found in the codices of Ubayy ibn Ka`b, Umm Salama and Ibn Abbas. Some commentators accepted that it contained an injunction to specially observe the afternoon prayer in addition to the midday prayer while others said it was merely an elaboration of the text and that the salatil-wusta was the same as the salatil-`asr as in this tradition:

It was variants such as this that led to Hafsah's codex being destroyed when Marwan ibn al-Hakam was governor of Medina some time after the death of `Uthman. While Hafsah was still alive she refused to give it up though he anxiously sought to destroy it (Ibn Abi Dawud, Kitab al-Masahif, p.24) and it was only upon her death that he got hold of it and ordered its destruction fearing, he said, that if it became well-known the same variant readings `Uthman sought to suppress would occur again.

The Muslim world today boldly professes a single text of the Qur'an yet those of `Uthman's time accused him, saying that the Qur'an had been in many books and that he had discredited them all except one. A high price had been paid to obtain one standardised text for all time.

3. VARIANT READINGS IN THE OTHER CODICES

`ABDULLAH IBN MAS`UD: AN AUTHORITY ON THE QUR'AN

Although `Uthman succeeded in destroying the other codices he was unable to suppress the fact that they had been compiled. Because the preferred method of learning the Qur'an was still by memorisation he could not entirely eliminate the variant readings known to exist between them and Zaid's codex. He also had to contend with the fact that many of their compilers were renowned Qur'an reciters. One of the best known was `Abdullah ibn Mas`ud who is recorded as being "the first man to speak the Qur'an loudly in Mecca after the apostle" (Ibn Ishaq, Sirat Rasulullah, p.141). The hadith record which records that Muhammad specifically started with him as a leading authority on the Qur'an is supported by the following tradition where he expresses his own knowledge of the book:

In a similar tradition he added to this that he had once recited more than seventy surahs in Muhammad's presence and claimed that all of the Prophet's companions were aware that no one knew the Qur'an better than he. Shaqiq, one of the companions sitting there, stated that no one argued with him or found any fault in his recitation (Sahih Muslim, Vol.4, p.1312). It also cannot be doubted that he was one of those who collected the Qur'an into written form shortly after Muhammad's death. Ibn Abi Dawud devotes no less than nineteen pages to the variant readings between his text and that of Zaid ibn Thabit (Kitab al-Masahif, pp. 54-73). It is also well known that Ibn Mas`ud initially refused to hand his codex over for destruction and for a while after one of the copies of Zaid's manuscript arrived at Kufa the majority of the Muslims there still adhered to Ibn Mas`ud's text.

There are solid evidences that his reason for resisting `Uthman's order was that he considered his own codex to be far superior to Zaid's and before Hudhayfah ever reported the existence of variant readings to the Caliph he had some sharp words with him.

When Hudhayfah also challenged him that he had been sent to the people of Kufa as their teacher and there had made them submit to his reading of the Qur'an, Ibn Mas`ud replied that he had not led the people astray, again claiming that no one knew the Qur'an better that himself (Ibn Abi Dawud, p.14). On another occasion he had this to say about his knowledge of the Qur'an in contrast with Zaid's proficiency:

It is also recorded that when news of `Uthman's order to destroy the other codices and to use Zaid's alone to obtain uniformity in reading reached Kufa Ibn Mas`ud gave a khutba, a sermon on the subject and declared to the Muslims of the city:

One thing is obvious from these statements – Ibn Mas`ud regarded his codex as a more authentic record of the original Qur'an text than the one compiled by Zaid and standardised by `Uthman as the sole text to be used throughout the Muslim world thereafter.

Muslim writers try to get around the implications of these evidences by suggesting that it was only a sentimental attachment to his codex that made Ibn Mas`ud react so strongly against the Caliph's order or, once again, that the variant readings were confined solely to differences in  pronunciation. It is quite clear, however, that it was his conviction that his codex was superior to Zaid's that made him angry and, as shall be seen, the variant readings related to real differences in the text itself.

THE VARIANT READINGS IN THE OTHER CODICES

One of the interesting facets of Ibn Mas`ud's codex was the total omission of the opening chapter, the Suratul-Fatihah, from his text as well as the mu`awwithatayni, the last two surahs of the Qur'an. The form of these chapters has some significance – the first is purely a prayer to Allah and the last two are "charm" surahs against evil forces. In all three the words are the expression of the believer as speaker rather than Allah himself. The possibility that Ibn Mas`ud had denied the validity of these surahs troubled early Muslim historians. Fakhruddin ar-Razi, the author of a commentary on the Qur'an titled Mafatih al-Ghayb ("The Keys of the Unseen") who lived in the sixth century of Islam admitted that this had "embarrassing implications" and used the strange reasoning that Ibn Mas'ud had probably not heard himself from the Prophet that they were to be included in the Qur'an. Ibn Hazm, another scholar, simply charged without giving any reasons that this was "a lie attributed to Ibn Mas`ud". Ibn Hajar al-Asqalani, however, in his famous Fath al-Baari (a review of the Sahih al-Bukhari) accepted these reports as sound, stating that Ibn Mas`ud had omitted them because the Prophet, to his knowledge, had only commanded that the surahs be used as incantations against evil forces and that, while he accepted them as sound, he had been reluctant to include them in his text (As-Suyuti, Al-Itqan, pp.186-187).

There were numerous differences between Ibn Mas`ud's codex and Zaid's in respect of the rest of the text and no less than one-hundred-and-one occur in Suratul-Baqarah alone. A review of some these will indicate the nature of these variant readings.

Surah 2:275 begins with the words Allathiina yaakuluunar-ribaa laa yaquumuuna – "those who devour usury will not stand". Ibn Mas`ud's text had the same introduction but added the words yawmal qiyaamati, namely "on the Day of Resurrection". The variant is mentioned in Abu Ubaid's Kitab Fadhail al-Qur'an and was also recorded in the codex of Talha ibn Musarrif, a secondary codex said to have been dependent on Ibn Mas`ud's text, Talha likewise being based at Kufa.

Surah 5:91, in the standard text, contains the exhortation fasiyaamu thalaathati ayyaamin – "fast for three days". Ibn Mas`ud's text added the adjective mutataabi`aatin meaning three "successive" days. This variant is derived from at-Tabari's famous commentary titled Jami` al-Bayan `an Ta`wil ay al-Qur'an (7.19.11) and was also mentioned by Abu Ubaid. This variant wasfound in Ubayy ibn Ka`b's text as well as in the codices of Ibn `Abbas and Ibn Mas`ud's pupil Ar-Rabi ibn Khuthaim.

Surah 6:153 begins Wa anna haathaa siraati – "Verily this is my path". Ibn Mas`ud's text read Wa haathaa siraatu rabbakum – "This is the path of your Lord". The variant derives again from at-Tabari (8.60.16). Ubayy ibn Ka`b had the same reading, except that for rabbakum his text read rabbika. The secondary codex of Al-A`mash, mentioned by Ibn Abi Dawud in his Kitab al-Masahif (p.91), also began with the variant wa haathaa as in the texts of Ibn Mas`ud and Ubayy ibn Ka`b. Ibn Abi Dawud also adds a further variant, suggesting that Ibn Mas`ud read the word siraat with the Arabic letter sin rather than the standard sad (Kitab al-Masahif, p.61).

Surah 33:6 contains the following statement about the relationship between Muhammad's wives and the community of Muslim believers: wa azwaajuhuu ummahaatuhuu – "and his wives are their mothers". Ibn Mas`ud's text added the words wa huwa abuu laahum – "and he is their father". This variant is also recorded by at-Tabari (21.70.8) and was also recorded in the codices of Ubayy ibn Ka`b, Ibn `Abbas, Ikrima and Mujahid ibn Jabr except that in the last three texts mentioned the statement that Muhammad is their father precedes the one which makes his wives their mothers. The codex of Ar-Rabi ibn Khuthaim, however, follows Ibn Mas`ud's in placing it at the end of the clause. The considerable number of references for this variant reading argue strongly for its possible authenticity over and against its omission in the codex of Zaid ibn Thabit.

In many other examples the variant relates to the form of a word which has slightly altered its meaning, as in Surah 3:127 where Ibn Mas`ud and Ubayy ibn Ka`b both read wa saabiquu ("be ahead") for wa saari`uu ("be quick") in the standard text. The variant again derives from at-Tabari (4.109.15). In other instances a single word has been added not affecting the sense of the text as in Surah 6:16 where once again Ibn Mas`ud and Ubayy ibn Ka`b recorded the same variant, namely yusrifillaahu – "averted by Allah" – for the standard yusraf – "averted". This variant is recorded in Maki's Kitab al-Kasf.

It is important to remember that these are not variants which reflect adversely on the codices which were destroyed as though the text standardised by `Uthman was above reproach while all these were full of aberrant readings. Zaid's codex was just one of many which had been compiled shortly after Muhammad's death and it was purely as a matter of convenience that it was preferred above the others. The prominence which Ibn Mas`ud enjoyed as a reciter, and his claim that he knew the Qur'an better than Zaid, should also be remembered. It is also most significant to find that Ubayy ibn Ka`b also was regarded as one of the best readers of the Qur'an by the Prophet himself:

As a result he became known as Sayyidul-Qurra, the "Master of the Readers". Another tradition states that `Umar himself confirmed that he was the best of the Muslims in the recitation of the Qur'an (Sahih al-Bukhari, Vol.6, p.489). It is therefore significant to find that numerous variant readings existed between his and Zaid's text.

For example, in place of wa yush-hidullaaha in Surah 2:204 he read wa yastash-hidullaaha. He also omitted the words in khiftum from Surah 4:101. Then again, in Surah 5:48 where the standard reading is wa katabnaa `alayhim fiiha – "and we inscribed therein for them (the Jews)" – the reading of Ubayy was wa anzalallaahu alaa banii Isra`iila fiiha – "and Allah sent down therein to the Children of Israel". The variant was also recorded by at-Tabari (6.153.24).

The evidences all show that, prior to the endeavour by `Uthman to standardise one codex for the purposes of obtaining uniformity of reading, there were numerous different readings of the Qur'an among the best known of the reciters. It took time for the Qur'an to become a single text and, as shall be seen, a second redaction was necessary some centuries later to standardise the vocalised text as well. One thing is quite obvious from all these readings, however – there is no foundation for the Muslim claim that the Qur'an presently read in the Muslim world is an exact copy of the original text at the time of Muhammad.

4. MISSING PASSAGES OF THE QUR'AN TEXT

THE MUSHAF: AN INCOMPLETE RECORD OF THE QUR'AN TEXT

During the Battle of Yamama shortly after Muhammad's death a number of the qurra, reciters of the Qur'an, perished and, as has been seen already, some passages of the text are said to have disappeared with them. No one else is said to have known these texts and it must be assumed that they passed away with them. There are many other records to show that individual verses and, at times, whole passages are missing from the Qur'an in its standardised form. These all serve to indicate that the mushaf of the Qur'an, as Muslims read it today, is in fact an incomplete record of the original handed down to them. `Abdullah ibn `Umar, in the earliest days of Islam, had this to say on the subject:

There are many examples that could be quoted but a selection of these should suffice to prove the point. A typical case relates to a verse which is said to have read:

It is said that this verse at one time formed part of Suratul-Bayyinah (Surah 98). This is quite possible as it fits well into the context of the short surah which contains, in other verses, some of the words appearing in the missing text, such as din (religion, v.5), `aml (to do, v.7) and hunafa (upright, v.4). It also contrasts the way of Allah with the beliefs of the Jews and Christians.

It is also interesting to note that, whereas the standard text of Surah 3:19 today reads innadiina `indallaahil-Islaam – "the religion before Allah is al-Islam (i.e. the Submission)", Ibn Mas`ud read in place of al-Islam the title al-Haniffiyah, i.e. "the Upright Way". At the beginning of Muhammad's prophetic mission there were a number of people in Arabia who disclaimed the worship of idols and called themselves hunafa, specifically meaning those who follow the upright way and who scorn the false creeds surrounding them.

It is possible that this was the initial name of of the specific faith which Muhammad later called Islam as his religion took on its own special identity and as his followers specifically came to be called Muslims, those who submit to Allah. This would account for the subsequent lapse of the title in the Qur'an and the omission of the text which is said to have been part of the text.

There are further evidences of whole surahs said to be missing from the Qur'an in its present form. Abu Musa Al-`Ashari, one of the earliest authorities on the Qur'an text and a companion of Muhammad, is reported to have said to the qurra in Basra:

The one verse he said he could recall is one of the well-known texts said to be missing from the Qur'an. Abu Musa went on to say:

The tradition as here quoted follows the record of it in the Sahih Muslim where it is set out after the statement about the surah resembling the ninth surah and containing the verse about the son of Adam (Vol.2, p.501). The Musabbihaat are those surahs of the Qur'an (numbers 57, 59, 61, 62 and 64) which begin with the words Sabbahu (or yusabbihu) lillaahi maa fiis-samaawaati wal-ardth – "Let everything praise Allah that is in the heavens and the earth". These are records from the most authoritative of Islamic sources and they indicate very clearly that the Qur'an in its present form is somewhat incomplete.

SPECIFIC VERSES SAID TO BE MISSING FROM THE QUR'AN

Much is said in the Hadith literature about the missing verse about the "son of Adam". The tradition is so widely reported that it must be authentic in its basic details. As-Suyuti records a number of these to show how well-known it was, one of which reads:

This record is followed by a similar tradition recorded by Ubayy ibn Ka`b which gives the verse in much the same words, except that in this case the companion expressly stated that Muhammad had quoted this verse as part of the Qur'an text which he had been commanded to recite. Following this is the tradition of Abu Musa, similar to the record in the Sahih Muslim, which states that the verse was from a surah resembling Suratul-Bara`ah in length. In this case, however, Abu Musa is not said to have forgotten it but rather that it had subsequently been withdrawn (thumma rafa`at – "then it was taken away"), the verse on the greed of Adam alone being preserved (As-Suyuti, Al-Itqan, p.525).

Abu Ubaid in his work Fadhail al-Qur'an and Muhammad ibn Hazm in his Kitab al-Nasikh wa`l Mansukh both recorded this verse as well but alleged that it was part of a surah that was later abrogated and duly withdrawn. Nonetheless it remained in the memory of many reciters as a portion of the original Qur'an text.

Another very well-known passage said to be missing from the Qur'an relates to the "stoning verses" initially brought to the attention of the growing Muslim community by `Umar, the second Caliph of Islam. They state that Muhammad once ordered all adulterers to be stoned to death in contrast with the statement in Surah 24:2 that they should be lashed with a hundred strokes. `Umar is said to have drawn the attention of the Muslim community to this passage from the mimbar (pulpit) in the mosque of Medina towards the end of his life. He is reported to set the matter before those gathered before him as follows:

`Umar was clearly persuaded that this verse was originally a part of the Qur'an as revealed to Muhammad and was concerned that over a period of time it would be forgotten by the next generation of Muslims. In another record of this incident it is said that `Umar added: "Verily stoning in the Book of God is a penalty laid on married men and women who commit adultery if proof stands, or pregnancy is clear, or confession is made" (Ibn Ishaq, Sirat Rasulullah, p.684). Both of these records add that `Umar mentioned another missing verse which was once part of the kitabullah (viz. the Qur'an) which the earliest of Muhammad's companions used to recite, namely "O people! Do not claim to be the offspring of other than your fathers, as it is disbelief on your part to be the offspring of other than your real father" (Sahih al-Bukhari, Vol.8, p.540). There are indeed many Hadith records which record that Muhammad during his lifetime duly ordered the stoning of adulterers:

The difference between this tradition and the Qur'anic text quoted on giving adulterers a hundred stripes has led to much discussion among Muslim commentators. They generally concluded that, as `Umar had so much to say about the missing verse, it must have been part of the original text but had possibly been withdrawn. Nevertheless it was presumed that the teaching and prescription found in the verse remained binding as part of the sunnah, the "example" of the Prophet. They decided that stoning of adulterers was the penalty for married men and women who commit adultery but that a hundred lashes was the punishment for a single person who cohabited with a married person. In the early days Muslim scholars struggled with the implications of the many traditions which stated very clearly that certain passages were missing from the Qur'an.

Nonetheless `Umar was quite statisfied that the order to stone those guilty of adultery was indeed a part of the original text  as appears from this particular tradition of the same incident:

These traditions, among many others of a similar nature, all give the impression that the Qur'an, once it was compiled into a single text at the end of Muhammad's life, was incomplete. Numerous passages, although not entirely forgotten by the companions of the Prophet, had nevertheless fallen out of the text of the book as it was generally recited by the Muslims and are no longer a part of it. While they do not appear to affect the teaching of the Qur'an as it stands today, they nevertheless do witness against its complete authenticity.

5. SAB`AT-I-AHRUF: THE SEVEN READINGS

THE SAB`AT-I-AHRUF IN THE HADITH LITERATURE

`Uthman succeeded in standardising a single written text of the Qur'an but, as the pronunciation of words and clauses was not reflected in the earliest manuscripts, the Qur'an was still read in different ways. No vocalisation of the written text existed at that time and so the script (as much of written Arabic does today) was transcribed with consonants only. Vowel points were only added much later. At the same time a tradition had been recorded that the Prophet himself had stated that the Qur'an was in fact sent down with more than one form of recitation:

This statement concludes an incident where `Umar one day heard Hisham ibn Hakim reciting Suratul-Furqan in a way very different to that which he had learnt it. In his typical impulsiveness he intended to spring upon him but controlled himself until Hisham had finished his reading. `Umar immediately confronted him with being a liar when he claimed that he had learned his recitation directly from Muhammad himself. When they approached the Prophet for a decision he confirmed both their readings, adding the above statement that the Qur'an had been revealed alaa sab`ati ahruf, "in seven readings". A similar tradition stating that the Qur'an was originally sent down in seven different forms reads as follows:

Another tradition states that Ubayy ibn Ka`b recalled an occasion where Muhammad reported that Jibril (Gabriel) had one day informed him that Allah had commanded that the Qur'an be recited in only one dialect, to which Muhammad replied that his people were not capable of this. After some negotiation the angel informed him that Allah had allowed the Muslims to recite the Qur'an in seven different ways and that each one would be correct (Sahih Muslim, Vol.2, p.391).

Apart from these traditions there are no records to define exactly what these seven different forms of reading were. As a result numerous different explanations have been given, some saying that this was to accommodate the different dialects of the Arab tribes and others that they were seven distinct forms conveyed to the centres of Islam by approved readers in the second century of Islam. Abu `Amr is said to have taken one of these to Basrah, Ibn Amir took another to Damascus, and so on (Sunan Abu Dawud, note 3365, Vol.3, p.1113). No one can possibly say what they were, however, as nothing more is said in the Hadith literature than that they were confined to differences in dialect and pronunciation.

It is important to note that these are a different type of variant reading to those which `Uthman suppressed. The records which have been kept of the latter were, as has been seen already, of words, clauses and other real differences in the text itself. In the case of the sab`at-i-ahruf, however, the distinction was confined to finer points of pronunciation and expression of the text. `Uthman was well aware of the different types and he obviously wanted to eliminate both of them. To erase the textual differences he simply chose Zaid's codex in preference to the others and ordered that they be burnt. To deal with the dialectal variants, on the other hand, he ordered Sa`id ibn al-As and two others from the Quraysh to amend Zaid's text where necessary to confine the text to their dialect. The following impression of his action is very informative:

He succeeded eminently in eliminating the real differences in the text between the different codices by destroying all but one, but he was unable to eradicate the differences in dialect as these could not be defined in a written text that had no vowel points. It is these alone that the sab`at-i-ahruf are said to have affected. Muslim scholars and writers in modern times often attempt to blur the distinction by suggesting that the only variant readings that existed were in the pronunciation of different dialects (lugaat) and that, although `Uthman sought to suppress them, the Prophet of Islam himself had originally authorised them. It is obvious, however, that the prime purpose of the Caliph's action was to eliminate serious differences in the actual text of the Qur'an and that he could not, in fact, have succeeded in deleting the dialectal variations (which would have been negligible in comparison with the textual variants).

Abu Dawud records a selection of the latter type in his Kitab al-Huruf wa al-Qira`at ("The Book of Readings and Recitation"). These three examples show how the differences in pronunciation affected the text:

In each case the variant is found solely in the vowelling of the text and would not have been reflected in the consonantal text transcribed by `Uthman. It can clearly be seen that this type of variant reading has virtually no effect on the text or its meaning, unlike the other type which covers whole words and clauses found in some codices and not in the others. It was to be some centuries before serious attention was given to actually defining the sab`at-i-ahruf, the "seven different readings".

THE PERIOD OF IKHTIYAR UNTIL IBN MUJAHID

For the next three centuries after `Uthman there were considerable differences in the recitation of the Qur'an as a result of his inability to eliminate the dialectal variants, but the differences were confined to these alone. This was a time of ikhtiyar, a period of "choice" when Muslims were free to read the Qur'an in whichever dialect they chose on the strength of the tradition that there were seven legitimate ways in which the Qur'an could be recited. During this period until the year 322 AH (934 AD), all scholars of the Qur'an agreed that such recitations were valid although no one could define exactly what the seven readings were. They would be at the discretion of anyone who attempted to specify them.

In that year, however, the well-known authority on the Qur'an at Baghdad, Ibn Mujahid, took it upon himself to resolve the issue. He had considerable influence with Ibn `Isa and Ibn Muqlah, two of the wazirs (ministers) in the Abassid government of the day. Through them he managed to establish an official limitation on the permissible readings of the Qur'an. He wrote a book titled Al-Qira`at as-Sab`ah ("The Seven Recitations") and in it he established seven of the readings current in the Muslim world as canonical and declared the others shadhdh ("isolated") and no longer legitimate. He gave no source of authority for his decision and it appears it was entirely his own discretion which guided him.

The seven readings now authorised were those of Nafi (Medina), Ibn Kathir (Mecca), Ibn `Amir (Damascus), Abu `Amr (Basrah), Asim, Hamzah and al-Kisai (Kufa). In each case there were certain recognised transmitters who had executed a recension (riwayah) of their own of each reading and two of these, those of Warsh (who revised Nafi's reading) and Hafs (who revised Asim's) eventually gained the ascendancy as the others generally fell into disuse. Warsh's riwayah has long been used in the Maghrib (the western part of Africa under Muslim rule, namely Morocco, Algeria, etc) mainly because it was closely associated with the Maliki school of law but it is the riwayah of Hafs that has gradually gained almost universal currency in the Muslim world. This has particularly been so since the printing of the Qur'an became commonplace.

Ibn Mujahid's determination to canonise only seven of the readings current in the Muslim world of his day was upheld by the Abbasid judiciary. Very soon after his decree a scholar named Ibn Miqsam was publicly forced to renounce the widely-held opinion that any reading of the basic consonantal outline was acceptable as long as it contained good Arabic grammar and made good sense. The period of ikhtiyar duly closed with Ibn Mujahid's action. He did to the vocalised reading of the Qur'an what `Uthman had done to the consonantal text many centuries earlier. Just as the Caliph had destroyed the different codices so this scholar outlawed all dialectal readings in use except seven. So likewise, just as the text standardised by `Uthman cannot be regarded as a perfect  copy of the Qur'an exactly as it was delivered by Muhammad because it only standardised the text of one redactor at the Caliph's personal discretion, so the seven readings canonised by Ibn Mujahid cannot be accepted as an exact reflection of the sab`at-i-ahruf as they were likewise arbitrarily chosen by the redactor according to his own preference and judgment.

It is obvious that no one with any real authority can say precisely what the seven different readings referred to in the Hadith actually were. A very good example of the confusion caused in subsequent generations about these readings is found in the following quote attributed to Abu al-Khair ibn al-Jazari prior to Ibn Mujahid's declarations:

This statement shows how impossible it was to define the seven different readings. Any good reading was automatically considered to be one of them, not because it could be proved to belong to them, but because of other factors – its isnad (chain of authorities), its consistency with the `Uthmanic consonantal text, and its compliance with proper Arabic grammar. The decision rested purely on matters of discretion.

Contrary to the oft-stated Muslim sentiment that the Qur'an as it stands today is an exact replica of the one said to be inscribed on the Preserved Tablet in heaven, it is obvious that the book went through a long period in which distinctions in both the actual text and thereafter in dialectal reading were eliminated in the interests of obtaining a single text. The Qur'an became standardised into the form in which it is found today, mainly through the actions of `Uthman and Ibn Mujahid respectively but also through other means such as the gradual lapse of most of the readings accepted by the latter scholar. The book only contains a uniform, defined text because certain Muslims of earlier times made it their express purpose to reduce it to a single text upon which the whole Muslim world could be united. The evidences show, however, that whole passages are now missing from certain surahs, that considerable numbers of variant readings existed in the original codices, and that a host of different dialectal readings survived until some three centuries later until these were reduced to seven. Only the printing of the Qur'an has finally given the Muslim world a single, unvaried text.

There is even evidence to show that some time after `Uthman's action to standardise the mushaf, the written text of the Qur'an, certain amendments were made to Zaid's text. Under the heading Bab: Ma Ghaira al-Hajjaj fii Mushaf `Uthman ("Chapter: What was Altered by al-Hajjaj in the `Uthmanic Text") Ibn Abi Dawud lists eleven changes made by the governor of Iraq during the caliphate of `Abd al-Malik many decades after the death of `Uthman. His narrative begins as follows:

The whole section continues to name each of the changes made by the governor in what appears to have been a further minor recension of the text. Interestingly each one of the readings amended was also originally the reading of Ubayy ibn Ka`b as well.

There can be little doubt that the Qur'an has generally survived intact and that its present text is a relatively authentic reproduction of the book as it was originally delivered. There is no justification, however, for the Muslim dogma that nothing in it, to the last dot or letter, has ever been changed or amended, or that any portion of it has since been lost or omitted.

6. THE EARLY SURVIVING QUR'AN MANUSCRIPTS

THE INITIAL DEVELOPMENT OF THE TEXT

Numerous early manuscripts of the Qur'an from approximately one- hundred-and-fifty years after Muhammad's death have survived though none is in complete form. Large portions have been preserved intact but on the whole only fragments exist. It was generally assumed, as it is today, that the Arabic language was so familiar to its speakers that vowelling of the text was not necessary. A number of consonants were not distinguished from one another either so that only seventeen were employed in the very early texts. As time passed, however, the similar consonants were separated by diacritical points above or below the letters and vowelling soon followed to clearly identify the reading of each text. Today almost without exception printed Qur'ans are fully vocalised.

No form of dating appears in the earliest manuscripts either so that the date and place of origin of these texts is generally a matter of conjecture. It was only in later centuries that the calligrapher's name was disclosed in a colophon (usually at the end of a text) together with the date and place where the codex was transcribed. Unfortunately some colophons in the early manuscripts are known to have been forged so that accurate identification often becomes almost impossible.

Even after vocalisation became common some Qur'an manuscripts were transcribed in the original form. A good example is the superb text written in gold script on blue vellum which survives almost intact from Kairouan in Tunisia where it was originally inscribed in the late ninth or early tenth century AD. It has been suggested that the scribe's intention was to produce a work of great beauty for commemorative purposes rather than for general reading. This Qur'an was intended to be presented to the Abbasid Caliph al-Ma`mun for the tomb of his father, Harun ar-Rashid, at Mashad in Persia. For some reason it never left Tunis and the bulk of it is preserved in the National Library of Tunisia in the city. A number of individual leaves are in private collections.

The best indication of an early manuscript's origin, however, is its script. A number of different styles were used in the early days and these went through various stages of development. These factors help to determine the likely origin of any particular text. Prior to the advent of Islam the only proper script known to exist was the Jazm script. It had a very formal and angular character and it was from this style that the other famous early scripts developed. No Qur'an fragment, however, is known to have been written out in this form. The earliest Qur'an script known was employed in Arabia and is called the al-Ma`il script. It was first utilised in Medina. It is unique in that it uses vertical letters which are written at a slight angle. The very name means "the slanting" script and its upright form resulted in the early manuscripts being produced in a vertical format similar to that used for most books today. Only a few pages and fragments and, in a few cases, whole portions of the Qur'an are known to have survived yet they are almost certainly the oldest in existence. They date not earlier than about one hundred and fifty years after the death of Muhammad. A sign of their early origin is the fact that no vowel strokes or diacritical points were used in the text and no verse counts or chapter headings were recorded.

The second early script originating from Medina was the Mashq, the "extended" style which was used for a few centuries. It was the first to use a horizontal form and had a cursive and somewhat leisurely style. The most common early script, however, was the Kufi, more properly known as al-Khatt al-Kufi. Its title does not indicate its form but rather its place of origin, namely Kufa in Iraq where Ibn Mas`ud's codex had been so highly prized until its destruction at `Uthman's direction. It took some time to become predominant but, when it did, it became pre-eminent for three centuries and many superb texts survive.

Like the Mashq script it employed a horizontal, extended style and as a result most of the codices compiled were oblong in format. In time it became supplemented with diacritical marks and vowel strokes. No Kufic Qur'ans are known to have been written in Mecca and Medina in the first two centuries when the al-Ma`il and Mashq scripts were most regularly used. Nonetheless most of the early surviving Qur'an texts are written in Kufi script.

Another script which derives from the Hijaz in Arabia is the Naskh, the "inspirational" script. It took some time before it became widely accepted but, when it did, it replaced virtually all the others including the Kufi as the standard form of transcribing the Qur'an. It remains so until this day and virtually all Qur'ans printed and written out by hand since the eleventh century are written in this form. It is easily readable and also yields readily to artistic calligraphy. One of the earliest Qur'ans to use this form which survives intact as a complete text is the famous manuscript written out by Ibn al-Bawwab at Baghdad in 1001 AD. It is now in the Chester Beatty Library in Dublin in Ireland.

One other script amongst a few which developed after the Naskhi is the Maghribi, the "Western" script which, as its name indicates, comes from the extreme western region of the traditional Islamic world. It was first employed in Morocco and Moorish Spain and is still used in the area to this day. It is a very cursive script, not easy to read for those unfamiliar with the Arabic language, but highly attractive when written artistically.

THE TOPKAPI AND SAMARQAND CODICES

Despite the evidences that no Qur'an manuscripts can be reliably dated till the late eighth century, it is a popular fiction in the Muslim world that one or more of the copies of Zaid's codex that `Uthman distributed to the Muslim provinces survives intact to the present day. The motive for this popular belief is the desire to prove from existing texts that the Qur'an is unchanged to its last letter from its first written codices down to its most recent copies.

It is known for certain that Zaid's original manuscript, which was originally in Abu Bakr's possession and thereafter under the control of `Umar and Hafsah, came into the hands of Marwan upon the latter's decease, having been sent to him by `Abdullah ibn `Umar. It is expressly stated that this manuscript was destroyed by him immediately thereafter (Ibn Abu Dawud, Kitab al-Masahif, p.21). Two of the early Kufic manuscripts surviving without vowel points, however, are especially presented as originals of the copies which `Uthman made from Zaid's codex, one being a codex said to be preserved in the State Library at Tashkent in Uzbekistan. It is popularly known as the "Samarqand" codex as it is said to have first come to this city about 1485 AD and to have remained there until 1868. Thereafter it was removed to St. Petersburg and in 1905 fifty facsimile editions were prepared by one Dr. Pisarref at the instigation of Czar Nicholas II under the title Coran Coufique de Samarqand, each copy being sent to a distinguished recipient. In 1917 it was taken to Tashkent where it now remains.

Not more than about a half of this manuscript survives. It only begins with the seventh verse of Suratul-Baqarah and many intervening pages are missing. The whole text from Surah 43:10 has been lost. What remains, however, indicates that it is obviously of great antiquity, being devoid of any kind of vocalisation although here and there a diacritical stroke has been added to a letter. Nonetheless it is clearly written in Kufi script which immediately places it beyond Arabia in origin and of a date not earlier than the late eighth century. No objective scholarship can trace such a text to Medina in the seventh century.

Its actual script is also very irregular. Some pages are neatly and uniformly copied out while others are distinctly untidy or imbalanced. On some pages the text is fairly smoothly spread out while on others it is severely cramped and condensed. At times the Arabic letter kaf is written uniformly with the rest of the text, at others it has been considerably extended and is the dominant letter. The manuscript may well be a composite text of portions from different original codices, alternatively different scribes were employed to transcribe it. It also has artistic illumination between some of the surahs with coloured medallions. The very appearance of the text compared with the known development of the early scripts prevents a date earlier than one hundred and fifty years after Muhammad's death or a place of origin anywhere in Arabia.

The other famous manuscript is known as the "Topkapi" codex as it is preserved in the Topkapi Museum in Istanbul in Turkey. Once again, however, it is written in Kufi script, giving its date away to not earlier than the late eighth century. Like the Samarqand codex it is written on parchment and is virtually devoid of vocalisation though it, too, has occasional ornamentation between the surahs. It also appears to be one of the earliest texts to have survived but it cannot sincerely be claimed that it is an `Uthmanic original.

A comparison between these two codices in any event shows that they were not transcribed in the same place at the same time. The Topkapi codex has eighteen lines to the page while the Samarqand codex has between eight and twelve. The whole text of the former is uniformly written and spaced while the latter, as mentioned already, is often haphazard and distorted. They may well both be two of the oldest sizeable manuscripts of the Qur'an surviving but their origin cannot be taken back earlier than the second century of Islam.

The oldest surviving texts of the Qur'an, whether in fragments or whole portions, date not earlier than about a hundred and fifty years after the Prophet's death.