CHAPTER 3:

THE CODICES OF IBN MAS'UD AND UBAYY IBN KA'B

1. ABDULLAH IBN MAS'UD: AN AUTHORITY ON THE QUR'AN TEXT.

No study of the early transmission of the Qur'an would be complete without an analysis of the contribution of Abdullah ibn Mas'ud, one of the most prominent of Muhammad's companions. He was one of his earliest disciples and we are told that he was "the first man to speak the Qur'an loudly in Mecca after the apostle" (Ibn Ishaq, Sirat Rasulullah, p.141). Throughout Muhammad's twelve years of mission at Mecca and until his death at Medina some ten years later Ibn Mas'ud applied himself very diligently to learning the Qur'an by heart. There is much evidence to show that he was regarded by Muhammad himself as one of the foremost authorities on the Qur'an, if not the foremost, as appears from the following hadith:

The same tradition in the other great work of hadith also specifically mentions that Muhammad "started from him" (Sahih Muslim, Vol. 4, p.1312), showing that he was deliberately mentioned first, indicating that Muhammad regarded him as the foremost authority on the Qur'an. Among others mentioned is Ubayy ibn Ka'b who, as we have already seen, also compiled a separate codex of the Qur'an before it was destroyed by Uthman.

It is significant to find no mention of Zaid ibn Thabit in this list which shows quite conclusively that Muhammad regarded Ibn Mas'ud and Ubayy ibn Ka'b as far better read in the Qur'an than him. In another hadith we find further evidence of Ibn Mas'ud's prominence in respect of his knowledge of the Qur'an:

In a similar tradition we read that he added to this that he had recited more than seventy surahs of the Qur'an in Muhammad's presence, alleging that all Muhammad's companions were aware that no one knew the Qur'an better than he did, to which Shaqiq, sitting by, added "I sat in the company of the Companions of Muhammad (may peace be upon him) but I did not hear anyone having rejected that (that is, his recitation) or finding fault with it" (Sahih Muslim, Vol. 4, p.1312).

Abdullah ibn Mas'ud obviously had an exceptional knowledge of the Qur'an and, as Muhammad himself singled him out as the first person to whom anyone should go who wished to learn the Qur'an, we must accept that any codex compiled by him would have as much claim to accuracy and completeness as any other. That he was one of the companions who did in fact collect the Qur'an apart from Zaid ibn Thabit cannot be disputed. Ibn Abi Dawud devotes no less than nineteen pages of his work on the compilation of the Qur'an manuscripts to the variant readings found between his text and that of Zaid which was ultimately the one standardised by Uthman (Kitab al-Masahif, pp. 54-73).

Having become a Muslim before even Umar, the second Caliph of Islam, Ibn Mas'ud had been on the hijrahs to both Abyssinia and Medina and was one of the highly regarded muhajirun who had followed Muhammad from Mecca. He participated in both the Battles of Badr and Uhud and his close association with the Prophet of Islam and prestige in the knowledge of the Qur'an resulted in his codex of the Qur'an being accepted as the standard text of the Muslims at Kufa before the recension done by Uthman. His reaction to Uthman's order that all codices of the Qur'an other than Zaid's should be burnt is most informative.

2. IBN MAS'UD'S REACTION TO UTHMAN'S DECREE.

When Uthman sent out the order that all codices of the Qur'an other than the codex of Zaid ibn Thabit should be destroyed, Abdullah ibn Mas'ud refused to hand over his copy. Desai openly speaks of "Hadhrat Ibn Mas'ud's initial refusal to hand over the compilation" (The Quraan Unimpeachable, p.44), but Siddique, in his article, prefers to leave the impression that no such objection from the distinguished companion of Muhammad ever took place, saying instead, "There is no indication that he ever objected to the 'text of Hafsah' during the entire Caliphate of Umar" (Al-Balaagh, op.cit., p.1). But why should he have raised any objection to Zaid's codex at that time? His own codex had become well-established at Kufa while Zaid's had receded into relative obscurity, simply being retained by the Caliph without any attempt whatsoever to establish it as the standard text for the Muslim community.

It was only when this codex suddenly came into prominence and was decreed to be the official text during Uthman's reign that Ibn Mas'ud found his codex being threatened. He immediately refused to hand it over for destruction and we are told by Ibn al-Athir in his Kamil (III, 86-87) that when the copy of Zaid's text arrived for promulgation at Kufa as the standard text, the majority of Muslims there still adhered to Ibn Mas'ud's text. It must be quite obvious to any objective scholar that, just as Zaid had copied out a codex for Abu Bakr, so Ibn Mas'ud simultaneously compiled a similar codex and, given the latter's exceptional knowledge of the Qur'an, his text must be considered to be as accurate and reliable as that of Zaid. The two codices were of probable equal authority and reliability.

Because there are a wealth of evidences of differences between the two, however, and as it was Zaid's text that became the standardised text after Uthman's recension and the only one used to this day in the Muslim world, it is intriguing to find Muslim writers trying to play down and minimise the importance of Ibn Mas'ud's codex.

Desai claims that "his copy contained notes explanations as well. His copy was for his personal use, not for the use of the Ummah at large" (op.cit., p.45). No evidence is given for this claim. One of the great deficiencies in Desai's booklet is the almost total lack of documentation in respect of the factual allegations the author makes. Virtually nowhere do we find a reference to the traditional chapter and verse. The reader is expected to presume that the facts he alleges are well-founded. Desai leaves no room in his booklet for references by which a student can check whether the contents are factually reliable.

In fact it is well known that Ibn Mas'ud's codex, far from being for his personal use only, was widely used in the region where he was based and, just as Ubayy ibn Ka'b's codex became the standard text Syria before Uthman's recension, so Ibn Mas'ud's likewise became the standard text for the Muslim ummah in and around Kufa in Iraq (Ibn Abi Dawud, Kitab, p. 13).

Ahmad Von Denffer likewise attempts to minimise the importance of the other codices, saying of Ubayy ibn Ka'b's codex that "it was a mushaf for his own personal use, in other words, his private notebook" and goes on to say of all the other codices that these "personal notebooks became obsolete and were destroyed" (Ulum al-Qur'an, p.49). It is virtually impossible to understand how whole manuscripts of the Qur'an, carefully transcribed and widely used in the various provinces, can be reduced to the status of "personal notebooks", least of all how such codices could have become "obsolete" at any time.

Muslim writers resort to such strange reasonings solely because they are determined to maintain the declared textual perfection of the Qur'an as it stands today to the last dot and letter. As this text is only a revision and reproduction of the codex of just one man, Zaid ibn Thabit, they have to circumvent the fact that other equally authoritative codices of single companions existed and that all of them, Zaid's included, differed in many key respects. Thus the text of Zaid has become elevated to "official" status right from the time of its compilation, the other texts have been downgraded to the status of "personal notebooks", and the argument runs that they were destroyed because they differed from one another without any consideration for the fact that Zaid's own codex likewise differed from each of them in turn.

There are solid evidences to show why Abdullah ibn Mas'ud at first refused to hand over his codex for destruction. While Desai claims that it was only because he attached sentimental value to his compilation (p.45) and Siddique states that there was no difference between his text and Zaid's, we find, in fact, that it was precisely because the great companion of Muhammad considered his own text to be superior to and more authentic than Zaid's that he was angered at Uthman's decree. Before Hudhayfah had ever gone to Uthman to call upon him to standardise a single text of the Qur'an, Abdullah ibn Mas'ud had some sharp words with him and reacted to his proposal that the different readings in the various provinces should be suppressed.

Modern writers such as Siddique and others maintain that the only differences between the recitations of the text and the reading of each companion (qira'at) were in pronunciations and dialectal expressions, yet it is once again obvious that what Hudhayfah had in mind was the elimination of the actual written codices being used by Abdullah ibn Mas'ud and the others - you cannot drown a verbal recitation - and it was this proposal which so angered Ibn Mas'ud and which proves that the differences in reading were in the texts themselves. In other traditions we find clear evidences that he regarded Zaid's knowledge of the Qur'an, and therefore his written codex of the text, as inferior to his. After all, Abdullah ibn Mas'ud had become a Muslim at Mecca before Zaid was even born and he had enjoyed years of direct acquaintance with Muhammad while the early portions of the Qur'an were being delivered before Zaid ever accepted Islam.

In another source we find that, when Uthman's order came for the destruction of the other codices and the uniform reading of the Qur'an according to Zaid's codex alone, Ibn Mas'ud gave a khutba (sermon) in Kufa and declared:

In the light of all these traditions, which can hardly be discounted, the evasive explanations of modern Muslim writers cannot be accepted. Abdullah ibn Mas'ud clearly resisted Uthman's order, not because of sentiment as Desai suggests, but clearly because he sincerely believed that his text of the Qur'an, gained firsthand from Muhammad himself, was more authentic than the text of Zaid. This conclusion cannot seriously be resisted by a sincere student of the history of the Qur'an text and its initial compilation.

It is also quite clear that the differences in reading were not confined to forms of dialect in pronunciation but in the actual contents of the text itself. An examination of some of these textual differences will show just how extensive those variant readings really were.

3. THE VARIANT READINGS IN IBN MAS'UD'S CODEX.

One of the anomalies recorded in respect of Ibn Mas'ud's text is that it is said to have omitted the Suratul-Fatihah, the opening surah, and the mu'awwithatayni, the two short surahs with which the Qur'an ends (Surahs 113 and 114). The form of these surahs has some significance - the first is purely in the form of a prayer to Allah and the last two are "charm" surahs, being recommended incantations of refuge with Allah which Muslims should recite as protection against sinister forces and practices. One tradition states that Ubayy ibn Ka'b was at one time challenged with the suggestion that Ibn Mas'ud had made certain negative statements about these surahs and he replied that he had asked Muhammad about them and was informed that they were a part of the revelation of the Qur'an and should be recited as such (Sahih al-Bukhari, Vol. 6, p.472).

The possibility that Ibn Mas'ud may have denied that these three surahs were a part of the Qur'an vexed early Muslim historians. The well-known Iranian philosopher and historian Fakhruddin ar-Razi, who wrote a commentary on the Qur'an titled Mafatih al-Ghayb ("The Keys of the Unseen") and who lived in the sixth century of Islam (1149-1209 AD) gave some attention to this problem and sought to prove that the allegations were unfounded.

Another Muslim historian, an-Nawawi, in his commentary on the Muhaththab said that the Fatihah and the two "charm" surahs were unanimously regarded by the Muslims as part of the Qur'an and that what had been said about Ibn Mas'ud was false and unjustified (as-Suyuti, Al-Itqan, p.187). The famous dogmatic Muslim scholar Ibn Hazm likewise rejected the suggestion that Ibn Mas'ud had omitted these surahs from his codex:

The record goes on to say that Ibn Hajar al-Asqalani however, in his commentary on the Sahih of al-Bukhari (his famous Fath al-Baari), accepted these reports as sound, quoting authorities who stated that Ibn Mas'ud would not include the two "charm" surahs in his manuscript as Muhammad had, to his knowledge, only commanded that they be used as incantations against evil forces. He regarded the isnad (the chain of transmitters) for this record as totally sound and attempted to harmonise the conflicting records instead, suggesting that Ibn Mas'ud accepted the Fatiha and "charm" surahs as genuinely revealed but was reluctant to inscribe them in his written text.

As Uthman ordered all the codices of the Qur'an other than Zaid's to be destroyed and as Ibn Mas'ud was eventually compelled to hand his over for elimination, it cannot be determined whether the three relevant surahs were actually included in his codex or not. If they were omitted, the reason is either that he was unaware that Muhammad had expressly stated that they were part of the Qur'an text (as alleged by Ubayy) or, less probably, that Ibn Mas'ud had actually determined that they were not part of the actual kitabullah, the Book of Allah, and that the other companions had assumed they were because they had come to Muhammad in the same form as the other surahs of the Qur'an.

When we come to the rest of the Qur'an, however, we find that there were numerous differences of reading between the texts of Zaid and Ibn Mas'ud. As mentioned already the records in Ibn Abi Dawud's Kitab al-Masahif fill up no less than nineteen pages and, from all the sources available, one can trace no less than 101 variants in the Suratul-Baqarah alone. We shall mention just a few of the differences here in illustration of the nature of the variations between the texts.

1. 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 after the last word there was added the expression yawmal qiyaamati, that is, they would not be able to stand on the "Day of Resurrection". The variant is mentioned in Abu Ubaid's Kitab Fadhail al-Qur'an (cf. Nöldeke, Geschichte, 3.63; Jeffery, Materials, p.31). The variant was also recorded in the codex of Talha ibn Musarrif, a secondary codex dependent on Ibn Mas'ud's text, Taiha likewise being based at Kufa in Iraq where Ibn Mas'ud was based as governor and where his codex was widely followed (Jeffery, p.343).

2. Surah 5.91, in the standard text, contains the exhortation fasiyaamu thalaathati ayyaamin' - "fast for three days". Ibn Mas'ud's text had, after the last word, the adjective mutataabi'aatin, meaning three "successive" days. The variant derives from at-Tabari (7.19.11 - cf. Nöldeke, 3.66; Jeffery, p.40) and was also mentioned by Abu Ubaid. This variant reading was, significantly, found in Ubayy ibn Ka'b's text as well (Jeffery, p.129) and in the texts of Ibn Abbas (p.199) and Ibn Mas'ud's pupil Ar-Rabi ibn Khuthaim (p.289).

3. Surah 6.153 begins Wa anna haathaa siraatii - "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 - cf. Nöldeke 3.66; Jeffery, p.42). Ubayy ibn Ka'b had the same reading, except that for rabbakum his text read rabbika (Jeffery, p.131). 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 Mds'ud and Ubayy ibn Ka'b (Jeffery, p.318). 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).

4. Surah 33.6 contains the following statement about the relationship between Muhammad's wives and the 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". The variant was also recorded by at-Tabari (21.70.8 - cf. Nöldeke 3.71; Jeffery p.75). This variant was likewise recorded in the codices of Ubayy ibn Ka'b (Jeffery, p.156) as well as those of Ibn Abbas (p.204), Ikrima (p.273) and Mujahid ibn Jabr (p.282), except that in these three cases the statement that Muhammad is the father of the believers precedes that which makes his wives their mothers. In the codex of Ar-Rabi ibn Khuthaim, however, where the variant also occurs, it is placed in the same position in the text as in the codices of Ibn Mas'ud and Ubayy (p.298). 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.

These four examples are of texts where the variant consisted of the inclusion of extra words or clauses not found in Zaid's codex and, in each case, the variant is supported by inclusion in other codices, notably those included in Ubayy's text. The majority of variants, however, relate to consonantal variants in individual words or different forms of these words. In some cases whole words were omitted, such as in Surah 112.1 where Ibn Mas'ud omitted the word qul - "say" as did Ubayy ibn Ka'b (Fihrist S.26 Z.26 - cf. Nöldeke 3.77; Jeffery, pp. 113 and 180).

In other cases the variant related to the form of a word which also slightly altered its meaning, as in Surah 3.127 where Ibn Mas'ud and Ubayy both read wa saabiquu ("be ahead") for wa saari'uu ("be quick") in the standard text (cf. Nöldeke, 3.64; Jeffery, pp. 34 and 125).

In yet other cases one single word might be added not affecting the sense of the text, as in Surah 6.16 where once again both Ibn Mas'ud and Ubayy recorded the same variant, namely yusrifillaahu - "averted by Allah" - for the standard yusraf - "averted" (recorded from Maki's Kitab al-Kasf, cf. Nöldeke, 3.66; Jeffery, pp. 40 and 129).

These are but a small selection of the hundreds of variant readings between the texts of Ibn Mas'ud and Zaid giving a rough idea of the kind of differences that existed between their codices. They do serve, however, to show that these differences in their readings were not purely dialectal or confined to the pronunciation of the text as is conveniently suggested by writers like Siddique who are bound to the popular dogma "one text, no variants", but rather radically affected the contents of the text itself. The extent of the variant readings between all the codices in existence at the time of Uthman before he singled out that of Zaid to be the preferred text at the expense of the others is so great - they fill up no less than three hundred and fifty pages of Jeffery's Materials for the History of the Text of the Qur'an - that one can understand why the others were ordered to be destroyed.

Far from the Qur'an being universally accepted in a standard form there were, on the contrary, vast differences in the texts distributed in the various provinces. Uthman's action brought about the standardisation of a single text for the whole Muslim world - it was not a perpetuation of an already existing unity - and Zaid's codex, which from the evidences we have considered had no greater claim to authenticity than Ibn Mas'ud's, was simply arbitrarily chosen as the standard text because it was close at hand in Medina, had been compiled under official supervision, and had not become the accepted or rival text of any one province like some of the others before Uthman's decree. Before closing this chapter let us give some attention to the other great compiler of the Qur'an, Ubayy ibn Ka'b.

4. UBAYY IBN KA'B - MASTER OF THE QUR'AN RECITERS.

Among the authorities on the Qur'an other than Abdullah ibn Mas'ud the most well known was Ubayy ibn Ka'b. There are two very interesting hadith relating to his prominence as an expert on the Qur'an text, the first reading as follows:

In consequence he became known as Sayyidul-Qurra - "the Master of the Readers". Umar himself, the second Caliph of Islam, confirmed that he was in fact the best of all the Muslims in the recitation of the Qur'an (Sahih al-Bukhari, Vol. 6, p.489). The second hadith in this respect reads as follows:

We are not informed as to why Muhammad considered himself especially obliged to commit parts of the Qur'an to Ubayy but these two traditions do serve to show how highly regarded he was as an authority on the Qur'an. Nonetheless his codex also contained a vast number of readings which varied from Zaid's text and, as we have already seen, these readings often agreed with Ibn Mas'ud's text instead. The addition of the word mutataabi'aatin in Surah 5.91, which we have already seen was recorded by at-Tabari as part of the codex of Ibn Mas'ud, was independently attributed to Ubayy as well (Ibn Abi Dawud, Kitab al-Masahif, p.53). His order of Surahs, in some ways similar to Zaid's, was nonetheless different at many points (as-Suyuti, Al-Itqan fii Ulum al-Qur'an, p.150).

Some examples of instances where he agreed with Ibn Mas'ud and differed in turn from Zaid (there were in fact a very large number which could be mentioned) are the following:

1. For the standard reading wa yush-hidullaaha in Surah 2.204 he read wa yastash-hidullaaha (cf. Nöldeke 3.83; Jeffery, p.120).

2. He omitted the words in khiftum from Surah 4.101 (cf. Nöldeke 3.85; Jeffery, p.127).

3. He read mutathab-thibiina for muthabthabiina in Surah 4.143 (cf. Jeffery, p.127).

There are a number of cases where whole clauses differed in his text. In Surah 5.48, where the standard text reads wa katabnaa 'alayhim fiiha - "and We inscribed therein for them (the Jews)" - the reading of Ubayy ibn Ka'b was wa anzalallaahu alaa banii Isra'iila fiiha - "and Allah sent down therein to the Children of Israel" (cf. Nöldeke 3.85; Jeffery, p.128).

From Abu Ubaid we find that, whereas Surah 17.16 in the standard text reads amarnaa mutrafiihaa fafasaquu, Ubayy read this clause ba'athnaa akaabira mujri-miihaa fdmakaruu (cf. Nöldeke 3.88; Jeffery, p.140).

One can go on and on to show how vastly Ubayy's text, like Ibn Mas'ud's and all the others, is said to have differed from Zaid's text which ultimately became standardised as the official reading of the Qur'an, but these examples serve once again to show that the variant readings were in the contents of the text itself and not just in niceties of pronunciation and recitation as many modern Muslim writers choose to assume.

There is a very interesting record of a whole verse which was found in Ubayy's text and which is not found today in Zaid's text which we shall consider in the next chapter. We cannot close on Ubayy, however, without giving some consideration to two extra surahs which we are told belonged to his codex. We are informed that, whereas Ibn Mas'ud omitted the two "charm" surahs from his codex, Ubayy included two extra surahs, al-Hafd (the Haste) and al-Khal' (the Separation) (as-Suyuti, Al-Itqan, p.152-153). The narrative continues by stating that Abu Ubaid said:

Suyuti goes on to give the full text of these two surahs, stating that they were also found in the codex of Ibn Abbas following the reading of both Ubayy and Abu Musa who also recorded them (Al-Itqan, p.154). Both surahs are similar to the Suratul-Fatihah, containing prayers to God for forgiveness and declarations of faith, praise, service and trust in his mercy. We are told that these are the supplications which Muhammad occasionally offered at his morning prayers after recitation of other surahs, being described as "the preserved suratal-quunut (chapters of humble obedience toward God) in the surahs respectively titled al-Khal' and al-Hafd" (as-Suyuti, Al-Itqan, p.527).

It is intriguing to consider that, in their likeness to the Suratul-Fatihah (which extends to their length also - the Fatihah has seven verses while the other two have been set out in three and six verses respectively - cf. Nöldeke, Geschichte 2.35), they were regarded as of equal authority from different stand-points by Ibn Mas'ud and Ubayy respectively. The former had none of them in his codex, the latter all three! It seems that Muhammad himself used them interchangeably and that some of his companions were uncertain whether they should be recorded as part of the written kitabullah, especially as each one constitutes a prayer of supplication in the words of the believers and worshippers in contrast to the rest of the Qur'an where Allah is always made to be the speaker.

We have, in this chapter, given some consideration to the codices of the two most prominent authorities on the Qur'an to show how considerably they differed from the codex of Zaid ibn Thabit and how uncertain much of the Qur'an text was when it was first compiled after the death of Muhammad. We could also go on to consider the numerous other codices that are recorded as having been transcribed before Uthman's decree that they should be burnt, but let it suffice to say that in each of these as well there were large numbers of variant readings which have been preserved. (Uthman was able to blot out the written codices in which they were recorded, he was unable to erase them from the memories of those who had recorded them).

In fact one should not speak so much of the readings in Zaid's text as the "standard" readings and of the others as "variant" readings as though the latter were the exception. The truth is that, between all the codices that existed in the early days of Islam ibn Mas'ud's, Zaid's, Ubayy's, Abu Musa's, etc. there were a wealth of differences and Zaid's readings qualify just as readily as the others do. In his case his qira'at became standardised as the only readings allowable in the Muslim world and copies of his codex were distributed to replace the others in popular use purely to establish a uniform reading of the Qur'an text.

The Qur'an as it has come down through the centuries is not the single text without any variants that has been divinely preserved without so much as a dispute regarding even one letter as Muslim writers conveniently choose to believe. Rather it is simply but one form of it as it existed during the first two decades after Muhammad's death, the compilation of but one man, Zaid ibn Thabit, and commissioned for the Muslim world as the only text to be accepted, not by divine decree, but by the arbitrary discretion of yet another single individual, Uthman ibn Affan.

The popular sentiment of the Muslims that the Qur'an has, right from the beginning, been preserved without the slightest variation in a single text would carry weight if it could be shown that this was the only text accepted by the whole Muslim community from the time of Muhammad himself.

The records of the Qur'an's compilation in the heritage of Islam, however, show convincingly that there were a whole number of different codices in vogue during the first generation after Muhammad's demise and that these all varied considerably from one another. The adoption of a single text came only twenty years after his death and only through the unilateral choice of one of the varying codices as the standard text at the expense of the others. The universally accepted text of the Qur'an in the Muslim world is not so much the mushaf of Muhammad but rather the mushaf of Zaid ibn Thabit, and its unchallenged authority today has come about, not through divine decree or preservation, but by the imposition of one man acting on his own initiative against the many other codices of equal authority which he summarily consigned to the flames.


Jam' Al-Qur'an: Table of contents
Answering Islam Home Page