This post looks at whether Quran manuscripts perfectly preserve what Muhammad recited.
My conclusion is that today’s Quran seems to faithfully preserve what Uthman collected, but significant textual variants exist between the time of Muhammad to the time of Uthman.
Appreciating the textual tradition of the Quran should open doors for Muslims to understand that the textual tradition of the Christian and Jewish Scriptures does not invalidate them from being the Word of God.
What does the Quran teach about its preservation?
Devout Muslims believe the Quran is the Word of God, unchanged and unalterably transmitted from the time of Muhammad.1
It is “flawless” and fixed by an “imperishable tablet” and “not to be doubted.” To question otherwise is like “telling a Christian that Jesus was gay” as Abdou Filali-Ansary, a Moroccan scholar, put it.2
Muslims base their belief in the perfect preservation of the Quran from passages in the Quran in which Allah promises to perfectly preserve it:
- Verily We: It is We Who have sent down the Dhikr (i.e. the Quran) and surely, We will guard it (from corruption). (Quran 15:9, Muhsin Khan)
- Nay! This is a Glorious Quran, (Inscribed) in Al-Lauh Al-Mahfuz (The Preserved Tablet)! (Quran 85:21-22, Muhsin Khan)
Are there any textual variants in Quran manuscripts?
Some twenty years after Muhammad’s death, these readings were collected and standardized by Uthman.4
Here is an example of a difference:6
Since different readings are part of Uthmanic text tradition, it is counterproductive to insist that these invalidate what the Quran says about its preservation (Quran 15:9; 85:21-22).
However, Christians should use the qira’at to explain that differences between the Gospels do not invalidate them from being the Word of God.7
What can also be productive, as we will see below, is comparing copyist mistakes in Quran manuscripts and pre-Uthmanic Quran variants with similar kinds of variants in biblical manuscripts to show that variants do not invalidate a text from being the Word of God.
Are there copyist errors in Quran manuscripts?
Muslims acknowledge that there are unintentional copyist errors in Quran manuscripts. Muhammad Mustafa Al-Azami, Professor Emeritus at King Saud University, wrote:
We must nevertheless take into consideration that there are over 250,000 manuscripts of the Qur’an scattered all over the globe. When comparing them it is always possible to find copying mistakes here and there; this is an example of human fallibility, and has been recognized as such by authors who have written extensively on the subject of “unintentional errors.” Such occurrences cannot be used to prove any corruption within the Quran (The History of The Qur’anic Text: From Revelation to Compilation: A Comparative Study with the Old and New Testaments [Leicester: UK Islamic Academy, 2003], 13).8
This is helpful to understand because Muslims often claim that the Bible has been corrupted (tahrif) and point out that there are hundreds of thousands of variants in biblical manuscripts. But the vast majority of these are unintentional copyist errors, the kind found in Quran manuscripts (e.g. spelling variations).9
Why are there few meaningful and viable variants in Quran manuscripts?
A meaningful variant is something that changes the meaning of the text. A viable variant is one that is plausibly original. For biblical studies, probably less than 1% of New Testament variants are both meaningful and variable.10
In comparison with the Bible, there are fewer viable and meaningful variants in Quranic manuscripts after the time of Uthman.11 Different answers are given as to why there are few meaningful and viable variants in Quran manuscripts:
The typical answer Muslims give as to why there are few (or no) meaningful and viable Quran manuscript variants is belief that Allah will preserve the Quran in its original form until the Last Day (Quran 15:9; 85:21-22).
Francois De Blois, Teaching Fellow at the Department of the Study of Religions at the School of Oriental and African Studies in London, gives another reason for the lack of variants in Quran manuscripts:
…that the virtual absence of real textual variants in the Qur’an is the result of the fact that the transmission of the Qur’an has always been primarily through oral rather than through written tradition. The situation is similar with that of the Vedas, which were composed much earlier than the New Testament or the Qur’an and transmitted for many centuries exclusively orally. In the Vedas there are actually no real textual variants. But this means that the methodology of textual criticism and source-criticism, as applied with such success to the New Testament, cannot be transferred automatically to the Qur’an. A different kind of source requires a different kind of methodology (Francois De Blois, “Islam in Its Arabian Context.” In The Quran in Context, edited by Neuwirth, Sinai, and Marx [Brill 2010], pp.618, 619).
Destruction of meaningful and viable variants
Another plausible reason for a lack of meaningful and viable variants in Quran manuscripts is that they were destroyed Uthman.12
Muhammad Mustafa Al-Azami says:
But if any scrap of parchment falls into our inquisitive hands and, despite our best allowance for orthographic differences, fails to slip comfortably into the Uthmani skeleton, then we must cast it out as distorted and void (The History of The Qur’anic Text: From Revelation to Compilation: A Comparative Study with the Old and New Testaments [Leicester: UK Islamic Academy, 2003], 205).13
Destruction of potentially meaningful and viable variants doesn’t mean variants never existed. It means that they no longer exist and therefore cannot be studied.14
There isn’t a critical edition of the Quran
Adding to the confusion is that a critical edition of the Quran has never been published.
The common belief that the Qur’an has a single, unambiguous reading is due in part to the bravado of translators, who rarely express doubt about their choices. Yet it is above all due to the terrific success of the standard Egyptian edition of the Qur’an, first published on July 10, 1924 (Dhu l-Hijja 7, 1342) in Cairo, an edition now widely seen as the official text of the Qur’an… Minor adjustments were subsequently made to this text in following editions, one published later in 1924 and another in 1936. The text released in 1936 became known as the Faruq edition in honor of the Egyptian king, Faruq (r. 1936–52). Yet the influence of the Cairo text soon spread well beyond Egypt. It has been adopted almost universally by both Sunni and Shi‘i Muslims, and by critical scholars as well, who have long since given up Gustav Flugel’s 1834 edition. Writing in 1938, Otto Pretzl noted with amazement that in his day for the first time a de facto canonical text had emerged.
Yet the Egyptian project was never intended to be text-critical, at least as this term is commonly understood. The scholars who worked on that project did not seek to reconstruct the ancient form of the Qur’an, but rather to preserve one of the canonical qira’at “readings” (here meant in the specialized sense it has in Islamic tradition), that of Hafs (d. 180/796) ‘an ‘Asim (d. 127/745). But these qira’at are part of the history of the text, not its starting point…
When the scholars in Cairo decided to fix a standard text according to Hafs ‘an ‘Asim, they still had to decide which reports of it to trust. Their project, then, involved comprehensive research of the classical qira’at works. In fact, they conducted this research with great thoroughness and attention to detail, according to the observations of several western scholars. Gotthelf Bergstrasser, for example, noted that in only a small number of cases is their reading contradicted by earlier sources on Hafs ‘an ‘Asim. However, the Cairo text is often at odds with manuscript evidence (Gabriel Said Reynolds, “Introduction,” in The Qurʾān in its Historical Context, ed. Gabriel S. Reynolds [London, Routledge, 2008], pp.2-3).
This is important to understand because critical editions of the Bible (e.g. the Nestle-Aland edition of the New Testament) are viewed by many Muslims as evidence that the Bible is “corrupt;” when in fact, a critical edition of the Quran has not yet been accomplished.15
Are there any meaningful and viable variants in the Yemeni Quran manuscripts?
In 1972, some 12,000 Quranic parchment fragments were discovered in the Great Mosque in the Yemeni capital Sana’a. These fragments consist of tiny snippets of the Quran to whole folios belonging to some 926 copies of the Quran. To date, only a small portion of the fragments have been published, and it is unknown when the rest will be published.
According to Stefan Wild, a leading scholar in the Quran and Arabic lexicography wrote that the find at Sana had not yet produced any meaningful variants:
The earliest written Qur’anic fragments we know have a number of important orthographic and ornamental peculiarities like those found in the recently discovered Yemeni manuscripts from Sanaa, some of which may go back to the seventh century. But so far, no substantial textual difference has been found between these early documents and the Qur’anic text as we know it.16
Likewise, Muhammad Mustafa Al-Azami wrote:
Though certainly a great treasure containing a wealth of orthographic oddities, the Mushafs in Sana do not add anything new or substantial to the body of proof which already demonstrates the Quran’s completion within the first decades of Islam.17
More recently however, Behnam Sadeghi, Mohsen Goudarzi, and Uwe Bergmann have written about a palimpsest found at Sana’a.18 It dates from the the first half of the seventh century AD; it does not belong to the Uthmanic textual tradition;19 and it contains textual variants from the Uthmanic tradition.
Examples of Major Variants in the Sana’a palimpsest (a.k.a. C-1 or the lower writing of Saṇʿāʾ 1)
(Behnam and Goudarzi, 21)20
This is strong manuscript evidence that differences in the Quran are more than just dialect and pronunciation. The Sana Quran is evidence for non-Uthmanic versions of the Quran to which literary sources also give evidence.21
The Uthmanic Quran may not contain corruption, but it is guilty of corruption
Textual history, contrary to what many Muslims believe, is not reason for dismissing the Christian and Jewish Scriptures as corrupt. For Muslims to argue otherwise would be to argue against the Quran.22
The real corruption, comment below if you think otherwise, is that the Quran does not preserve what God revealed because the Quran omits the books of Moses, Prophets, the Psalms, and the Gospels. Furthermore, many Muslims have resorted to arguing that the Gospel of Barnabas was an authentic Gospel. Adding to and subtracting from Scripture is undoubtedly corruption (tahrif).
Talking points with Muslims
- Does your Muslim friend know that there are unintentional copyist errors in Quran manuscripts? The point of this question is not to denigrate the Quran, but to help Muslims understand that copyist errors in biblical manuscripts—the vast majority of which are similar to the unintentional copyist errors in the Quran—do not invalidate the Bible from being the Word of God. Invite your Muslim friend to read the Gospels.
- Does your Muslim friend know that there are textual variants in Quran palimpsest fragments found in Sana’a? The point of this question is that the existence of textual variants in the biblical tradition does not automatically disqualify the Bible from being the Word of God. Again, invite your Muslim friend to read the Gospels.
- Islamic tradition has recognized different readings of the Quran.23 This can be pointed out for the purpose of explaining that differences between the Gospels do not mean they are corrupt.
- Explain why Christians include the Jewish Scriptures in their Bible, and ask your Muslim friend why Uthman omitted both the Jewish and Christian Scriptures.
FOR SIMILAR KINDS OF TALKING POINTS BE SURE AND READ:
Watch, Is the Bible Reliable?
- “The text of the Qur’an is entirely reliable. It has been as it is, unaltered, unedited, not tampered with in any way, since the time of its revelation” (Fethullah Gülen, Questions this Modern Age Puts to Islam). [↩]
- Andrew Higgins, “The Lost Archive,” Wall Street Journal, http://online.wsj.com/article/SB120008793352784631.html; accessed August 11, 2012. [↩]
- Muhammad Al-Azami writes that over “twenty Companions have narrated hadiths confirming that the Quran was revealed in seven dialects” (The History of the Quranic Text from Revelation to Compilation, 154). Here is one such hadith:
Narrated ‘Umar bin Al-Khattab:
I heard Hisham bin Hakim reciting Surat-al-Furqan during the lifetime of Allah’s Apostle, I listened to his recitation and noticed that he was reciting in a way that Allah’s Apostle had not taught me. I was about to jump over him while He was still in prayer, but I waited patiently and when he finished his prayer, I put my sheet round his neck (and pulled him) and said, “Who has taught you this Sura which I have heard you reciting?” Hisham said, “Allah’s Apostle taught it to me.” I said, “You are telling a lie, for he taught it to me in a way different from the way you have recited it!” Then I started leading (dragged) him to Allah’s Apostle and said (to the Prophet), ” I have heard this man reciting Surat-al-Furqan in a way that you have not taught me.” The Prophet said: “(O ‘Umar) release him! Recite, O Hisham.” Hisham recited in the way I heard him reciting. Allah’s Apostle said, “It was revealed like this.” Then Allah’s Apostle said, “Recite, O ‘Umar!” I recited in the way he had taught me, whereupon he said, “It was revealed like this,” and added, “The Quran has been revealed to be recited in seven different ways, so recite of it whichever is easy for you .” (Bukhari, Book 9, Volume 93, Hadith 640; also Hadith No. 514, Vol. 6); see Frederik Leemhius, “Readings of the Qur’an.” Encyclopaedia of the Qur’ān; Michael Cook, “The Stemma of the Regional Codices of the Koran,” Graeco-Arabica, in Festschrift in Honour of Vasilios Christides, ed. by G.K. Livadas, vol.9-10 (Athends, 2004), p. 89-104. [↩]
- Although there are problems with it:
The most widely-accepted version of the traditional history of the Qurʾan places the collection of the final consonantal text in the caliphate of ʿUthman about twenty years after Muhammad’s death. The occasion for the final collection of the Qurʾan, according to this account, was a military expedition to Azerbayjan and Armenia under the leadership of the general Ḥudhayfa. Apparently his Muslim contingents from Syria and those from Iraq fell into dispute about the correct way of reciting the Qur’an during the communal prayers. Trying to establish order, ʿUthman appointed a commission of four respected Meccans, presided over by Zayd b. Thabit, to copy the “sheets” that were in Ḥafṣa’s personal possession. Where variant readings of words were encountered, they chose the one in the dialect of the Quraysh. When the scribes completed their assignment, ʿUthman kept one copy in Medina and sent other copies to al-Kūfa, al-Baṣra and Damascus. He then commanded that all other extant versions be destroyed. His order, however, was not heeded in al-Kūfa by the Companion Ibn Masʿūd (d. 32/653) and his followers. The difficulties of this version of the story center on essential points, namely the doubt that accuracy in the recitation of the Qur’an would have caused significant unrest in the military during the early conquests of Islam, the widely-accepted view that the Qur’an is not actually in the dialect of the Quraysh (q.v.) and the improbability that the caliph would have given an order to destroy the already existing copies of the Qur’an. Further, the appearance of Ḥafṣa in this narrative probably functions simply as a mechanism to link the Abū Bakr/ ʿUmar and ʿUthman versions together and to establish an unbroken chain of custody for an authoritative text that remained largely unnoticed in the community. Despite the difficulties in this version of the chronology of the collection of the Qur’an, scholars generally accept that the official consonantal text of the Qur’an was established in ʿUthman’s caliphate and that Zayd b. Thabit played a significant role in effecting it. (Gerhard Böwering, “Chronology and the Qur’an.” In Encyclopaedia of the Qurʾan. General Editor: Jane Dammen McAuliffe [Leiden and Boston: Brill, 2005], CD-ROM version.)
For a Muslim account of Uthman’s collection of the Quran see Muhammad Mustafa al-Azami, The History of the Quranic Text from Revelation to Compilation, 87-97. [↩]
- Other versions include “a 1964 mushaf from Algeria in the Warsh riwāya and another version from Morocco. A Tunisian edition of the Qālūn riwāya was published by al-Dār al-Tūnisiyya lil-Nashr. In the Sudan the Dūrī reading was printed in 1989 by the Department of Religious Affairs and Endowments (awqāf)” (Michael W. Albin, “Printing of the Qur’an.” Encyclopaedia of the Qurʾān. General Editor: Jane Dammen McAuliffe [Leiden and Boston: Brill, 2005], CD-ROM version).
- Taken from “Which Quran” by Layth Al-Shaiban [↩]
- Jesus would have preached similar, but not the exact same, sermons in different places which can explain many differences in the Gospels. [↩]
- “The Islamic scholarly tradition does not purport to have preserved the spelling of every word in the codices sent out by Uthman. Rather, Muslim tradition preserves the original Uthmanic codices at least at the skeletal-morphemic level, that is, with respect to features of the skeletal (unpointed) text that would necessarily change a word or part of word (morpheme) into something else if they were different. Some skeletal variations, such as different spellings of a word are not skeletal-morphemic because they do not necessarily change a word. Moreover, differences in the way consonants are pointed may change a word, but they are not skeletal-morphemic either since they do not change the skeleton. Normally, a reading is said to differ from the standard Uthmanic rasm only if it changes both the skeleton and the word, that is, if the change is skeletal and morphemic. All of this has been well-understood for many centuries and is simply taken for granted in the way most Muslim Quran specialists have written about the different readings (qira’at)” (Sadeghi, Behnam; Mohsen Goudarzi, “Ṣan’ā’ 1 and the Origins of the Qur’ān,” Der Islam [March 2012], 2-3]). [↩]
- Arguably, the Quran has a positive view of the Bible (cf. Quran 3:3; 5:44-48, 68-69; 6:115; 10:37,64; 12:111; 18:27; 34:31; 35:31), but many Muslims today say that these Quranic passages refer to Scriptures that no longer exist.
While the alleged corruption of the Christian Scriptures is a central part of Muslim discourse against Christianity, it was not always this way.
In the first five centuries AH (seventh to eleventh centuries ad), Muslim commentators, theologians, and jurists usually dealt with the “previous Scriptures” assuming, on the basis of qur’anic references to these Scriptures, that the Law, Psalms, and New Testament were given by God but misinterpreted by Jews and Christians. Muslim-Christian dialogue during these centuries frequently involved reference to and debate about the proper interpretation of specific passages from the Bible. After the eleventh century, there is a noticeable shift in the character of the dialogue. After this point Muslims began arguing that the actual text of the Bible has been intentionally corrupted and is therefore not trustworthy and cannot provide any basis for dialogue. Martin Accad has argued, ‘it can be demonstrated that until the time of Ibn Hazm in the eleventh century, the accusation of tahrif in the sense of ‘intentional corruption of the Holy Scriptures’ was virtually non-existent. Even where some grave and serious suspicions were raised against the integrity of the text, the accusation can certainly not be considered to have been a central or foundational element of the Muslim discourse against Christianity. If it has become the starting point of that discourse today, it is certainly worth knowing that it has not always been the case, and that it is therefore possible to think otherwise. Even after Ibn Hazm, as late as the fourteenth century, Ibn Taymiyya recognized that the Islamic position towards tahrif as textual corruption was still diverse and ambiguous’. (Martin Accad, “The Gospels in the Muslim Discourse of the Ninth to the Fourteenth Centuries.” In A Common Word: Muslims and Christians on Loving God and Neighbor. Edited by Miroslav Volf, Ghazi bin Muhammad, and Melissa Yarrington [Grand Rapids: Eerdmans, 2010], 59). [↩]
- For a helpful ten minute overview of New Testament variants watch Dan Wallace Explains the New Testament Variants in Ten Minutes. Dan Wallace is founder and Executive Director of the Center for the Study of New Testament Manuscripts. [↩]
- “[T]he ʿUthmānic text itself still left room for different readings. The codices of Medina, Mecca, Damascus, Kūfa and Baṣra are said to have presented some slight differences in a number of places, mainly concerning an extra wāw or alif, or a dhī instead of dhū or dhā. The chapter about the differences among these codices in Ibn Abī Dāwūd al-Sijistānī’s (d. 316/929) book on the ancient codices (Jeffery,Materials, 39-49 of the Arabic text) sums them up in lists that appear to have been well established by then” (Frederik Leemhius, “Readings of the Qur’an.” Encyclopaedia of the Qur’ān. General Editor: Jane Dammen McAuliffe [Leiden and Boston: Brill, 2005], CD-ROM version). [↩]
- According to one report it was said to Uthman:
“You have burned the book of God.” Uthman replied, “People read (the Qur’an) in different ways. One would say, ‘My Qur’an is better than yours.’ The other would say, ‘No, mine is better.’”
It was then objected to Uthman, “But why did you burn the (other) collections?”
Uthman replied, “I wanted nothing else to exist except what had been written in front of the Messenger of God, and was contained in the pages (suhuf) of Hafsa” (Al-Baladhuri, Ansab, vol.4.1, 550ff. quoted by Gregor Schoeler, “The Codification of the Qur’an.” In The Quran in Context. Edited by Neuwirth, Sinai, Marx [Brill, 2010], p.787. [↩]
- Interestingly, “the Egyptian government was motivated to begin the project that would lead to the Cairo Qur’an edition due to the variations (or “errors,” as an appendix to the Cairo edition describes them) found in the Qur’anic texts that they had been importing for state schools. In response, the government destroyed a large number of such texts by sinking them in the Nile River and issued its own text. The Cairo project thus followed in the spirit of the caliph ‘Uthman, and the governor al-Hajjaj b. Yusuf (d. 95/714), who are reported to have destroyed competing versions and distributed their own text of the Qur’an in the first Islamic century” (Gabriel Said Reynolds, “Introduction,” in The Qurʾān in its Historical Context, ed. Gabriel S. Reynolds [London, Routledge, 2008], p.3). [↩]
- Keith Small, visiting lecturer for at the Centre for Islamic Studies and Muslim-Christian Relations at the London School of Theology, believes that the editorial process is responsible for the lack of meaningful and viable variants in Quran manuscripts:
The great bulk of the Quranic manuscript tradition contains the same basic form of consonantal text, even in the great majority of the very earliest manuscripts. For instance, aside from early orthography issues, the consonantal text one finds in the oldest Quran in the British Library which dates to the early to mid 700’s is basically the same as the consonantal text in current printed Qurans. One does not find variant phrases, verses, or portions in the great majority of the earliest manuscripts. Only occasionally does one find a variant word. In fact, the degree of the lack of variant readings in these manuscripts is an evidence for a major editing/standardization project taking place within the first Islamic century, as one finds claimed in the Hadith (“A Quranic Window onto New Testament Textual History.” Journal of the Institute of Islamic Studies, Number 1, 2011, p.34). [↩]
- “While the classical literature records thousands of textual variants, the task of reconstructing the history of the Qur’an is complicated because the variants are not found in any extant manuscripts known to Western scholars. Several valuable works on the history of the Kur’an were written during the 4th/10th century, but later Muslim scholars, with just a few exceptions, have shown little interest in the problem of reconstructing the history of the canon” (Alford T. Welch, “Al-Kur’an.” In The Encyclopaedia of Islam, ed. P. Bearman et al. [Leiden: Brill, 1986-2004], CD-ROM version). [↩]
- Wild, “Why Self-referentiality?” In Self-referentiality in the Qur’an; Diskurse der Arabistik 11. Edited by Stefan Wild [Wiesbaden: Otto Harrassowitz, 2006], 16 [↩]
- The History of the Quranic Text from Revelation to Compilation, 318. [↩]
- Sadeghi, Behnam; Mohsen Goudarzi, “Ṣan’ā’ 1 and the Origins of the Qur’ān,” Der Islam (March 2012); Behnam Sadeghi; Uwe Bergmann, “The Codex of a Companion of the Prophet and the Qur’ān of the Prophet,” Arabica, Volume 57, Number 4, 2010 , pp. 343-436. [↩]
- All known Quran manuscripts—with the exception of one, the lower writing of Sanʿāʾ 1—“belong to the standard textual tradition, called ʿUtmānic in accordance with the traditional account of its origin, which traces it to the recension promulgated by ʿUtmān b. ʿAfān, the Companion of the Prophet Muhammad who ruled the Muslim empire as the third caliph during 23-35/644-56.” (Behnam Sadeghi; Uwe Bergmann, “The Codex of a Companion of the Prophet and the Qur’ān of the Prophet,” Arabica, Volume 57, Number 4, 2010, 364). [↩]
- Behnam Sadeghi plans a future article with a systematic textual analysis of all the variants. [↩]
- Literary sources also indicate different text traditions from the Companions of Muhammad: Ibn Masud and Ubay ibn Ka’b; however, these rescensions are only known from descriptions in literary sources (Behnam Sadeghi; Uwe Bergmann, “The Codex of a Companion of the Prophet and the Qur’ān of the Prophet,” Arabica, Volume 57, Number 4, 2010, 344).
Stefan Wild says there was a prototype and preliminary stage preceding the Uthmanic version:
Before the establishment of a canonically closed book, Muslim scripture was open to invasive procedures and strategic re-workings…the Qur’an, just like Jewish and Christian scripture, was a battleground for interpretive traditions. An important difference seems to be that the process leading up to a textus receptus was much shorter than the parallel processes for Jewish and Christian scriptures (Wild, “Why Self-referentiality?” In Self-referentiality in the Qur’an; Diskurse der Arabistik 11. Edited by Stefan Wild [Wiesbaden: Otto Harrassowitz, 2006], 17). [↩]
- I leave it to Muslims to decide whether the Sana Quran disproves Quran 15:9; 85:21-22. [↩]
- Ibn al-Jazari, a distinguished scholar in the field of the qira’at of the Qur’an, noted that in his time (1350 AD/751 AH – 1429 AD/833 AH) “important and unimportant scholars alike believed that seven modes (ahruf ) in which the Qurʾān was revealed went well beyond the ʿUtmānic text. He quotes one authority to the effect that other modes can include additions, omissions, substitutions, and transpositions of words” (Behnam Sadeghi; Uwe Bergmann, “The Codex of a Companion of the Prophet and the Qur’ān of the Prophet,” Arabica, Volume 57, Number 4, 2010, 346). [↩]