Some ancient reports and modern scholars assign this event to the reign of the third caliph and link it with his standardizing the text of the Qur’ān around AD 650. However, the analysis shows that the sūras were formed earlier.

There is the traditional account that is associated with most premodern scholars. They held that the Prophet Muḥammad (d. AD 632) disseminated the Qur’ān gradually. Some of his Companions compiled copies of the scripture. These codices had differences. Motivated by the differences and seeking uniformity among Muslims, the Caliph ‘Uthmān (d. AD 656), himself a Companion, established a standard version. However, at the very least we know the text at the “skeletal-morphemic” level. Rather, Muslim tradition preserves the original ‘Uthmānic 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.

There are scholars in North America and Europe who support key features of the traditional narrative as recounted above. They do not take all the reports in the later sources at face value, but they believe that critical and detailed analysis of the literary evidence confirms elements of the traditional account. These scholars have their counterparts in the Muslim world. Notable members of this group include Michael Cook, Muḥammad Muḥaysin, and Harald Motzki, the first one being a defector from the revisionist camp. For their works on the Qur’ān, see the Bibliography. For a brief discussion of Muḥaysin’s work, see Behnam Sadeghi, “Criteria for Emending the Text of the Qur’ān,” in Law and Tradition in Classical Islamic Thought, ed. Michael Cook, et al. (New York: Palgrave Macmillan, forthcoming, 2012). For a summary and discussion of Cook’s work, see Sadeghi and Bergmann, “The Codex,” 364, 367–9.

1. regardless of the date of the lower codex, the textual tradition to which it belonged and the ‘Uthmānic tradition must have diverged sometime before the spread of the ‘Uthmānic tradition in the mid-seventh century AD. The lower writing, on paleographic and art-historical grounds, is almost certainly from the seventh century AD, and probably not from the latter part of that century. More precision may be obtained by radiocarbon dating, which assigns the parchment, and hence the lower codex, to the period before AD 671 with a probability of 99% (before 661 with the probability of 95.5%, and before 646 with a probability of 75%). Radiocarbon dating was performed on a sample from the “Stanford 2007” folio. For the details, see Sadeghi and Bergmann, “The Codex,” 352–4. On the assumption that the codex was not made a long time after the parchment was prepared, see “The Codex,” 354. Yasin Dutton, “An Umayyad Fragment of the Qur’an and its Dating,” Journal of Qur’anic Studies 9.2 (2007): 57–87; Efim Rezvan, “On the Dating of an ‘ʿUthmānic Qur’ān’ from St. Petersburg,” Manuscripta Orientalia 6.3 (2000): 19–22; Hans-Caspar Graf von Bothmer, “Die Anfänge der Koranschreibung: Kodikologische und kunsthistorische Beobachtungen an den Koranfragmenten in Sanaa,” Magazin Forschung (Universität des Saarlandes), 1 (1999): 45.(edited)
2. Among the Qur’ān manuscripts, twenty-two are in the Ḥijāzī script, and therefore are probably from the first century AH (7th century and early 8th century AD). Puin wrote that there are about 90 Ḥijāzī manuscripts (Gerd-Rüdiger Puin, “Observations on Early Qur’ān Manuscripts in Ṣan‘ā’,” in The Qur’ān as Text, ed. Stefan Wild (Leiden and New York: E.J. Brill, 1996), 108). This estimate is wrong by a factor of four. Bothmer cites Puin’s error and corrects it, mentioning that the correct number is twenty-two (Bothmer, “Die Anfänge der Koranschreibung,” 46, footnote 28). All but eight of these twenty-two Ḥijāzī manuscripts are in the “vertical format,” that is, are longer in height than width. There are also many manuscripts in the Kūfī script, some of which are probably from the first century AH.

The innumerable ‘Uthmānic manuscripts and the different ‘Uthmānic readings preserved in the literary sources provide a very clear picture of the degree and types of change that could arise during the period in which the ‘Uthmānic tradition flourished. These changes are small enough in scope and few enough in number to be compatible with written transmission or with dictation in which the result is checked against the original. The standard tradition thus appears to have achieved a high level of transmission fidelity already around the mid-seventh century AD. The conclusion that C-1’s origin must have predated ca. AD 650 is largely independent of the date of Ṣan‘ā’ 1. For example, it would not be invalidated if it were found that the lower Ṣan‘ā’ 1 codex was produced, say, in the eighth century AD. This codex would still be only a representative of a C-1 text type, and the late date of the manuscript would still beg the question of when this textual tradition originated. Moreover, since the differences between the C-1 text type and the ‘Uthmānic text type outstrip in magnitude and number the range of differences expected to arise in the period after ca. AD 650, most of these differences must have originated before then.

1. A positive answer to this question is supported by textual criticism, as described above, which assigns the beginning of the C-1 text type to the period before the spread of the standard text type, that is, before ca. AD 650. In sum, the “Companion” codices indeed existed at the time of the Companions, as the literary sources maintain.
2. Analysis resolves a fundamental question about the early history of the Qur’ān: who joined the existing verses to form the sūras (chapters) and when? Many scholars and some early reports hold that this was accomplished after the death of the Prophet by the committee that ‘Uthmān charged with the task of standardizing the Qur’ān. Some other early reports however indicate that this was done already by the Prophet himself. This last view is now found to be better supported. It follows from the fact that the ‘Uthmānic Qur’ān, C-1, and the Companion codices generally have the same passages within the sūras, that the sūras were fixed before these various textual traditions branched off, in particular before the spread of the ‘Uthmānic version. With only a few exceptions, the differences among the codices are at the level of morphemes, words, and phrases – not at the level of sentences or verses. The exceptions in C-1 include the very short consecutive verses 31 and 32 in sūra 20, which are three words long apiece, and which appear in C-1 in reverse order. Literary sources record that these verses were also transposed in the Codex of Ubayy b. Ka‘b.47 Another exception concerns verse 85 of sūra 9, which is missing.

At sixteen words, this omission is found to be an outlier when compared to the sizes of other missing elements in C-1, which are much shorter. The anomaly may be explained by the common phenomenon of parablepsis, a form of scribal error in which the eye skips from one text to a similar text, in this case, from the instance of ūna followed by a verse separator and the morpheme wa at the end of verse 84 to the instance of ūna followed by a verse separator and the morpheme wa at the end of verse 85. The conclusion that the sūras were constituted prior to the ‘Uthmānic text helps one assess the accuracy of some early Muslim accounts. It disproves the reports that imply that it was under ‘Uthmān that the sūras were assembled from the preserved pieces of the revelation.48 There are some traditions about ‘Uthmān’s team finding the last two verses of sūra 9 with a man named Khuzayma, or Abū Khuzayma, or Ibn Khuzayma.49 C-1 has these verses in the expected place. Since they are also found in the ‘Uthmānic Qur’ān, and since it is not reported that any Companion codex was without them, these verses must have belonged to the prototype from which the C-1 and ‘Uthmānic text types emerged. Therefore, one should not read too much into the report.

1. The differences between this codex and the Uthmanic one can be summarized as follows:
   • Variants which arose due to the Prophet teaching the Qur’an differently. The variants in this category are as equally authoritative as Uthmanic readings.

1. As the text is non-Uthmanic, and is very old, it probably represents some sort of companion reading of the Qur’an that is different from the Uthmanic one. In a sense, this is probably the type of manuscript that Uthman would have taken to write his codex and then would have had burned. The fact that this Qur’an differs from the Uthmanic one does not mean that the Qur’an is somehow “changed”. We already have the original, as the Uthmanic copy along with its readings is ensured to be a perfect representation of the ‘Prophetic prototype’, this manuscript is simply an extra version that represents some other reading. As long as one of the divinely authorized readings reach us, we have the whole Qur’an (See: History of the Qur’anic Text. Mustafa Al-Azami).
2. Ibn Masud & Ubayy b.Ka’b:

Surviving fraction of the texts:

Symbols/Conventions:

Folia 2A

Folia 2B

1. Uthman did not change the Qur’an
2. This contention is quite a common one, even though it could be dispelled by simply reading any historical account of the Uthmanic compilation, it has been repeated very often among apologetics circles. If historical reports from the Islamic sources constitute good evidence to prove otherwise, that Uthman did not change anything from the Qur’an, this manuscript is even stronger evidence for this, since it is flesh-and-blood proof. There is nothing in the manuscript so far that suggests any meaningful difference from the standard Qur’an. There is no proof that Uthman somehow added any theological or legal idea into the Qur’an. The biggest differences are words being used in place of their synonyms. This manuscript, for the first time ever, gives us hard physical evidence for something Muslims have already known: Uthman was sincere in his efforts to compile the Qur’an.

1. The Qur’an really did exist during the time Muslims say it did
2. I should say this likely refutes the Wansbrough thesis of a very late Qur’an. Since we actually do have a Qur’an dating from during this time, we can safely put to rest any fanciful ideas on the origins of the Qur’an. The manuscript also confirms a part of the Islamic Hadith tradition. Though these sources are written much later they are remarkably accurate in telling us that there were early manuscripts owned by the companions of the Prophet that had variants. They also relate to us what types of variants are found in these manuscripts, and at times the variants mentioned in the Islamic literature is word for word the same as those that are found in this particular manuscript. Sadeghi and Bergmann conclude on this point that:
3. It is now equally clear that recent works in the genre of historical fiction are of no help. By “historical fiction” I am referring to the work of authors who, contentedly ensconced next to the mountain of material in the premodern Muslim primary and secondary literature bearing on Islamic origins, say that there are no heights to scale, nothing to learn from the literature, and who speak of the paucity of evidence. Liberated from the requirement to analyze the literature critically, they can dream up imaginative historical narratives rooted in meager cherry-picked or irrelevant evidence, or in some cases no evidence at all. They write off the mountain as the illusory product of religious dogma or of empire-wide conspiracies or mass amnesia or deception, not realizing that literary sources need not always be taken at face value to prove a point; or they simply pass over the mass of the evidence in silence.
4. A pioneering early example of such historical fiction was Hagarism, written by Patricia Crone and Michael Cook. While few specialists have accepted its narrative, the book has nevertheless profoundly shaped the outlook of scholars. It has given rise to a class of students and educators who will tell you not only that we do not know anything about Islamic origins, but also that we cannot learn anything about it from the literary sources. All this would be good and well if the mountain of evidence had been studied critically before being dismissed as a mole hill; but the modern critical reevaluation of the literary evidence has barely begun. And, significantly, any number of results have already demonstrated that if only one takes the trouble to do the work, positive results are forthcoming, and that the landscape of the literary evidence, far from being one of randomly-scattered debris, in fact often coheres in remarkable ways. A good example of such findings would be some of Michael Cook’s own fruitful recent studies in the literary sources in two essays of his already discussed here. It is not his confirmation of some elements of the traditional account of the standard Qurʾān that I wish to highlight here, noteworthy as it may be, but rather his demonstration that we can learn from the study of the literary sources.

1. The Uthmanic text and the current variant readings
2. When Uthman compiled the Qur’an, what he essentially did was to limit the number of the possible readings of the Qur’an. He was unable (or perhaps unwilling) to confine the Qur’an to just one qira’at (reading) because the arabic text at the time did not have dots. There were multiple ways to read the mushaf (written Qur’anic text) because of this. As such, some of the dialects survived: As long as any dialect fit the Uthmanic manuscripts, and actually did trace back to the Prophet himself, it was and still is considered a divinely inspired and authoritative reading. Take the example I have given above: The word underlined can be read “Ta’maloona” and “Ya’maloona”, while both of them have the same skeletal (ie. not dotted) text, they still sound and mean a different thing. This point cannot be stressed enough: The Qur’an was allowed to have certain selected variants in them by the permission of the author. Nobody was allowed to simply replace any word they liked because it was difficult to pronounce or they did not like it: The change in wording had to be approved by the Prophet himself.