When Was the Quran Collected and Standardized and Who Did It?
Source: https://www.reddit.com/r/AcademicQuran/comments/1bdh9v1/when_was_the_quran_collected_and_standardized_and/?share_id=CVgDdR5_XLEdXGcPf79E6&utm_content=2&utm_medium=android_app&utm_name=androidcss&utm_source=share&utm_term=1
Although the standard story of the collection and standardization of the text of the Koran maintains it occurred primarily under the Caliph Uthman (644-656 A.D.), here the contrarian view is asserted that this process actually occurred almost entirely under Caliph al-Malik (685-705) of the Uymayyad dynasty, with the crucial assistance of his right-hand man, al-Hajjajj. The main (secondary) source for this thesis, and this entire post, is Stephen J. Shoemaker’s “Creating the Qur’an: A Historical-Critical Study,” which cites the necessary and relevant primary sources to support it. The intention of the first part of this text will deal with why the standard Sunni story of the Quran’s compilation, as told by Bukhari that focuses on the role of Uthman in standardizing its text, simply isn’t true. Then I’ll give some evidence that al-Malik, with the able if ruthless assistance of al-Hajjajj, did this instead.
A standard common claim of Muslims is that the text of the Quran has absolutely no errors or variations in it. However, when the actual history of the Quran’s transmission, collection, and standardization is examined in reasonably contemporaneous primary sources, it’s obvious that it had many, many variations and different regional text types before al-Malik (r. 685-705) and al-Hajjajj used their imperial authority to forcibly standardize the “received text” of the Quran out of these sources. The standard story of the standardization of the Quran’s text appears in Bukhari’s important collection of hadith (sayings/teachings attributed to Muhammad), which the Sunni sect of Islam upholds and many Western historians uncritically have signed off on (i.e., the “Noldekean-Schwallian” paradigm, as Shoemaker labels it). In this telling, Abu Bakr (632-34), the first caliph and Muhammad’s right-hand man, was asked by the man who would succeed him, Umar (634-644), to collect together the recitations of the Quran together because many of those with knowledge of what Muhammad had said recently had died in battle, thus taking their memories to the grave. Abu Bakr initially objected, by saying if Muhammad hadn’t told his fellow Muslims to do this during his lifetime, why do it now? However, Umar persisted, and Abu Bakr said it was fine to do so and that the services of the scribe Zayd b. Thabit should be used for assembling the text of the Quran together. Oddly enough, Zayd made the same objection that Abu Bakr did, but Umar prevailed with him as well. So then, Zayd went to work looking for various fragments of Muhammad’s revelations as they were preserved in various ways, including stones, palm branches, camel bones, and “in the hearts of men.” Then Zayd gave the sheets of paper resulting from this project to Abu Bakr, who later at his death passed them along to Umar. After Umar died, he left these sheets with his daughter Hafsa, who had been a wife of Muhammad. Then roughly 20 years later, the Caliph Uthman (644-656) towards the end of his reign became concerned about the various differing renditions of the Quran in circulation among Muslims. One of his top generals, Hudhayfa ibn al-Yaman, told Uthman that in Iraq and Syria significantly different versions of the Quran were in circulation. He was troubled because Muslims would eventually become divided over which version was the best. Uthman acted on Hudhayfa’s concerns by getting the sheets that Hafsa had kept, which became the basis for an official version of the Quran that a group of scribes under the direction of Zayd produced. Then Uthman sent out copies of this official text to the major cities of his realm (Kufa, Mecca, Basra, and Damascus). He also directed that all the other defective versions of Quran should be gathered up and destroyed. So then, since these final events took place around 650 A.D, Muslims will claim that the Quran has no textual variations.
But is the mainstream Sunni story of the Quran’s compilation historically true? Even in this account, the Quran’s assembly and production was haphazardly performed. Furthermore, Sunni coercive imperial authority was applied very early on to the promulgation of a standardized text. There was no “bottom up” consensus of believers involved in this process, nor did the Muslim scribes have available the knowledge of the techniques and processes of textual reconstruction (as part of “lower criticism”) that the Christian West’s scholars eventually developed. (By contrast, no Christians had such coercive authority over the New Testament’s text for its first 200 years because they were a persecuted religious minority under the pagan Roman government’s watchful eye). When Uthman ordered the destruction of the alternative regional variations of the Quran, how did he know that they were wrong in all cases and that his was right?
Furthermore, there’s no unanimity in the primary Islamic sources supporting the story of the Quran’s standardization by Uthman. There are at least three other accounts of Umar’s or Abu Bakr’s involvement that don’t agree with Buhkari’s version as retold above. One version says that Umar did the work of collecting the Quran from disparate media without the involvement of Abu Bakr at all. Another rendition says that Abu Bakr ordered Zayd to write Muhammad’s recitations on palm branches, shoulder bones, and leather before Umar later had Zayd write these down into one document. Another telling of the story has Abu Bakr fully refuse Umar’s request to have the Quran written down. So when Umar became caliph, only then he had the Quran written down on leaves. Then there’s in both Shiite and Sunni sources the claim that Ali, who was Muhammad’s son-in-law and cousin, was the first one to collect the Quran together. There’s one account that Salim b. Ma’qil supposedly assembled the text right after Muhammad died. Another report says that Aisha, Muhammad’s favorite wife, had a copy of the Quran in the form of a codex. The rival regional versions of the Quran before Uthman supposedly had its text standardized have been called “the companion codices.” Purportedly four early followers of Muhammad were respectively responsible for them: Abd Allah ibn Mas‘ud’s version (in Kufa), Miqdad b. al-Aswad (in Hims), and Ubayy b. K’ab (in Syria), and Abu Musa al-Ash’ari (Basra). There’s hardly any unanimity in the tradition about how the Quran’s text was collected in the primary sources of Islam when other primary sources outside of Bukhari’s own harmonized story are examined. (See generally Shoemaker, “Creating the Qur’an,” pp. 24-25).
In other early Islamic historical works, outside of the hadith, more inconsistencies about how the Quran was compiled arise. Ibn Shabba (d. 876) in his “History of Medina,” has a collection of accounts about how the Quran came together, but surprisingly none of them mention Abu Bakr’s role. One report here says that Umar had begun collecting the Quran’s text together, but was assassinated before the job was done. Another tradition, by contrast, says that Umar owned a codex of the Quran. Yet another story says that Umar had disagreements with the version of the text that Ubayy b. Ka’b had collected. One report says that Zayd and Umar proofed a version of the Quran of Ubayy and routinely edited it based on the authority of Zayd. As one reads over the stories of Umar’s involvement in the collecting of the Quran, he actually wasn’t trying to compile the Quran but was trying to support the authority of one version among several that had already been collected together. According to Ibn Shabba, by the time Umar had become the caliph, several versions of the Quran had already been independently compiled, with each having its supporters in different areas. Umar wanted to assert the authority of the version of the Quran found in Medina against the versions enjoying favor in Iraq and Syria. Ibn Shabba dedicates an entire long chapter to the traditions about the efforts of Uthman’s compilation of the Quran. Besides the version of the story that Bukhari preserved, he gives a number of other accounts about Uthman’s participation in standardizing the text of the Quran. But much like the stories about Umar, Uthman wasn’t collecting the text from scratch, but rather was trying to correct versions of the Quran that were already in circulation to fit in with his caliphate’s preferred rendition. (See generally Shoemaker, “Creating the Quran,” pp. 25-26).
A somewhat earlier primary source than Ibn Sa’d’s is “Kitab al-tabaqat al-kahib” of Ibn Sa’d (d. 845), which is made up of biographies of the early caliphs and of Muhammad himself. He provides a wealth of reports about how the Quran was gathered together, which are hardly unanimous about how the process occurred. As de Premere writes about Ibn Sa’d’s perspective in the early ninth century, “the real history of the Qur’anic corpus seemed blurry and the identity of its architects uncertain.” Like Ibn Shabba, he says nothing about Abu Bakr’s supposed role in assembling the Quran together. Most interestingly, when focusing on the rule of Uthman himself, Ibn Sa’d says nothing about Uthman’s supposed role in compiling the Quran, which makes for a major inconsistency with Bukhari’s standard story. Even more surprisingly, in Zayd’s biography, Ibn Sa’d’s omits any mention of Zayd’s efforts to collect the Quran. Ibn Sa’d doesn’t make any mention of the sheets that Hafsa supposedly had, which were supposedly used to create the canonical version of the Quran that Uthman commanded to have made. In yet another account, Uthman indeed did command the Quran to be compiled, but his order went to Ubayy instead of to Zayd.
One problem in examining the accounts of the Quran’s collection concerns the ambiguity of the Arabic word “jama’a,” which can mean both “to memorize” and “to collect.” This makes the accounts of whether anyone wrote down anything Muhammad said during his lifetime unclear, since it could have meant the “memorization” of what he said, not its “collection.” In these reports, two men stand out, who were already mentioned above repeatedly, Zayd b. Thabit and Ubayy b. Ka’b, which later traditions say they were Muhammad’s scribes. Ibn Sa’d has contradictory reports about Umar’s role in compiling the Quran: One report says that Umar was the first to collect the Quran on sheets, but another says Umar was assassinated before he could compile the Quran together. Sa’d clearly didn’t know anything about the standard canonical story of Bukhari’s about Uthman, Zayd, and Hafsa’s sheets at the beginning of the ninth century. As de Premare observes, the silences and inconsistencies of Sa’d are disturbing about the real support that Bukhari’s story actually has in the primary sources. There’s no uniformity or unanimity in the relevant sources about how the Quran was compiled. (See Shoemaker, pp. 26-28).
A somewhat earlier version about the collection of the Quran appears in “Book of the Conquests,” by Sayf ibn ‘Umar (d. 796-797). In one key regard, his report agrees with Bukhari’s version in describing the general Hudhayfa’s sense of consternation about the different renditions of the Quran in use by Muslims in different areas of Uthman’s domain. In one regard, the reported conflicts were worse, however, since rival groups of believers were proclaiming the cases for their preferred versions of the Quran while condemning those found elsewhere. Since Hudhayfa was greatly distressed about these sharp disputes and major variations in the text of the Quran, he told Uthman in Mecca about this serious problem. To summarize the situation regionally, the Kufans favored the codex of Abd Allah ibn Mas’ud, the Syrians preferred that of Miqdad b. Al-Aswad (and seemingly Salim), and the Basran’s liked the rendition of Abu Musa al-Ash’ari. Oddly, the version of Ubayy b. Ka’b receives no mention in this source. Uthman commanded the partisans of each of these versions of the Quran to appear before him in order to make the case for their respective summary of the words of Muhammad. These are clearly discordant books in dispute, since Sayf ibn ‘Umar’s account identifies these productions as “codexes.” So clearly, from the bottom up, rival groups of Muslims in different geographical areas had written down what they believed were Muhammad’s words. Confronted with this mess, Uthman’s solution wasn’t to create a new collection of the Quran, but to take the version available in Medina, of which he had copies made and then he had them sent out to these other areas of his realm. He ordered that all the other versions should be destroyed. It’s not clear that his commands were followed or that he had the effective political/police power to enforce his decisions on this matter on believers who lived far from the Hejaz. So in the earliest account that we have of the Quran’s compilation in Islamic primary sources, Uthman made no effort to textually reconstruct the “best” version of the Quran out of various regional versions. There is no primary source before the ninth century that confirms that Uthman and the scribes he directed engaged in any kind of careful systematic process of textual reconstruction. (See Shoemaker, pp. 28-30).
One problem that also arises in the efforts to trace back the earliest version of the Quran as a text concerns the original ambiguities between what are now called “hadith” or the sayings/teachings attributes to Muhammad, and the Quran’s text, which purports to only be the words of God Himself. There clearly was confusion about how to make a distinction between these two kinds of records. For example, in an early letter said to have been written by Zayd ibn Ali (695-740), two of the hadith quoted are almost identical to what’s in the Quran (5:56; 21:24). Ibn Sa’d relays Salima b. Jarmi’s assertion that he has gathered “many qurans” from Muhammad together, which presumably were his teachings or hadith, not reports of direct revelations from God. So then, given these stories about how the Quran was collected by Abu Bakr, Umar, Uthman, Zayd, Ubayy, etc. they may not have been able in their time to make the clear distinctions that later Sunni scholarship made between the Quran (i.e., divine revelation) and the hadith (i.e., teachings/sayings of Muhammad). These regional versions of the Quran may well have been hadith to one degree or another. (See Shoemaker, p.30).
So when all these conflicting stories examined in the early primary Islamic historical sources about the Quran’s original compilation, it becomes obvious that that the standard Bukhari story, which the mainstream Sunni tradition has endorsed, is much too simplistic. The key error of many Western historians, such as those who endorse the generally reigning “Noldekean-Schwallian” paradigm, has been to uncritically endorse and support the mainstream Sunni viewpoint when it simply isn’t well supported in the primary historical sources. As Shoemaker quotes Burton as summarizing the relevant primary sources about the Quran’s compilation: “The reports are a mass of confusions, contradictions, and inconsistencies. By their nature, they represent the product of a lengthy process of evolution, accretion and ‘improvement.’ They were framed in response to a wide variety of progressing needs. . . . The existence of such reports makes it clear that the Muslims were confused. The earliest stage of the traditions on the collection of the Qur’an did consist in incompatible attributions of the first collection to Abu Bakr, to ‘Umar, to ‘Uthman.” De Premare somewhat cynically observes: “such variation among the reports [indicates] that each one seems to reflect later circumstances rather than the fact that it is alleged to relate.” (See Shoemaker, pp. 30-31).
The Muslims’ standard claims that there are no variations in the Koran’s text are simply not true. Most significantly, the variations that still are known to exist are those that survived the ruthless standardization process of the Quran during the reign of Abd al-Malik (685-705). Abu Hayyan al-Gharnait, who has been an important collector of the Quran’s textual variants, has explicitly noted that he has deliberately not gathered “those variants where there is too wide a divergence from the standard text of ‘Uthman.’” (See Shoemaker, p. 33). The Quranic inscriptions found in the Dome on the Rock on Jerusalem’s Temple Mount area are among the oldest in existence. (Since Jerusalem was mainly a Christian city at the time, these inscriptions often bore witness against Christian teachings and beliefs). However, as Shoemaker notes, these inscriptions, placed by the caliph Abd al-Malik (685-705 A.D.) “are our earliest surviving evidence for the text of the Qur’an, and yet they different from the now canonical version of the Qur’an.” He asked how this could be possible, if the text of the Quran had been standardized some 40 years earlier in the time of Uthman. (Shoemaker, p. 64).
One of the oldest Qurans, the Sanaa manuscript of the eighth century, has actually two differing texts. The newer one, dating to the middle eighth century, was written over an erased version that dates to the early eighth century. So why would the same folio pages have two different Qurans laboriously handwritten on them? Well, the older erased “palimpsest” version varies regularly from the newer “Uthmanic” rendition. In this case, it’s obvious that that when the newer standardized text of the Quran was promulgated throughout the caliphate of Abd al-Malik, the older version was erased from this particular manuscript’s pages. What was erased, however, is still recoverable and legible. It indicates that at least until 700 A.D. or later, non-canonical versions of the Quran were still being copied, which is long past the dates of Uthman’s reign (644-656 A.D.) (See Shoemaker, p. 77). Most likely the great majority of the variants that existed in the regional codexes of Ubayy b. Ka’b, Abd Allah ibn Mas’ud, Abu Musa al-Ash’ ari, and Miqdad b. Al-Aswad were totally destroyed; what has been preserved is a feeble remnant. So then, how do we know what was preserved is really what Muhammad allegedly heard from God as opposed to what was destroyed? Ha
Another interesting set of witnesses about the Quran’s formation comes from early Shiite witnesses from the first three centuries of Islamic history. The partisans of Ali as the legitimate caliph opposed the rendition of the Quran that mainstream Sunni tradition attributed especially to Uthman’s efforts. According to the Shiites, it was Ali, not Abu Bakr, Umar, or Uthman, who first gathered together the Quran’s text shortly after Muhammad’s death. However, the Shiites maintain that Ali’s version of the Quran was much longer than Uthman’s rendition. The first three caliphs that the Sunnis recognize (Abu Bakr, Umar, and Uthman) twisted and falsified what Ali had compiled. For obvious political reasons, the Shiites asserted, these caliphs deleted the section that clearly named Ali as Muhammad’s legitimate political heir. So the Quranic text traditionally attributed to Uthman is a distorted version designed to promote the religious and political program of the first three Sunni caliphs, according to the early Shiite writers.
It’s true, however, that starting in the tenth century Shiite scholars of the “Twelver” tradition began to repudiate their sect’s earlier witness and began to embrace the Uthman text such as it is. Because it had become very dangerous to cast doubt on the Quran by this time, they had obvious motives of self-preservation to change their opinions. Although mainstream Western scholarship has generally arbitrarily rejected the oldest Shiite viewpoint on the development of the Quran’s text, this decision to embrace the Sunni viewpoint has to be regarded as prejudiced. As already related above, the Sunni primary sources are full of inconsistencies and contradictions, so why should they deemed to be automatically reliable compared to the Shiite viewpoint by impartial historians? The Shiite viewpoint also has the advantage of being the “minority” viewpoint that lost, which may be all the more valuable since it reports details and has a general perspective that the (Sunni) victors have ignored, twisted, or censored. Interestingly enough, two of the discordant regional texts of the Quran were developed in southern Iraq, where Ali’s support was at its strongest, such as in Kufa and Basra. Indeed, Kufa was briefly the capital of Ali’s short-lived caliphate. (See Shoemaker, p. 40).
In this context, it’s important to realize how violently the Ummayyads persecuted and quelled the partisans of Ali’s cause in their realm. Unlike the case for the orthodox Catholics who counter-attacked the Gnostic Christians in the second and third centuries A.D., who were armed only with the power of the pen, the Sunni caliphs had authority over the sword and willingly wielded it to favor their cause. Amir-Moezzi explains the power of the Umayyad caliphs to impose their political and religious will on their opponents: “In an attempt to justify these measures [that distorted records of the past], caliphal power set up a complex system of propaganda, censorship, and historical falsification. First it altered the text of the Qur’an and forged an entire body of traditions falsely ascribed to the Prophet, drawing great scholars, judges, jurists, preachers, and historians into its service—all this within a policy of repression that was as savage as it was methodical, aimed at its opponents at large, but at Alids in particular.” (As quoted by Shoemaker, p. 37). The standardized text of the Quran is actually, according to Michael Cook, “a remarkable testimony to the authority of the early Islamic state.” The imperial efforts to find and destroy dissident Qurans were especially aimed at the proto-Shiites of southern Iraq. They were so successful, according to Omar Hamdan, “that one could only wonder in disbelief . . . if any remnant of a differing recension [of the Qur’an] were to come to light. Therefore, given the power of the Sunni caliphs by the eighth century, they easily could have thoroughly censored the viewpoint of Ali’s partisans from the historical sources that they controlled, in a manner bordering upon the fictional Ingsoc’s in Orwell’s novel “ 1984.” (See generally Shoemaker, pp. 35-38).
However, it’s unlikely that much of what Muhammad said was written down during his lifetime because the small, poor communities of Mecca and Medina were made up of people for whom the spoken word was primary and few were literate in a broader sense. Shoemaker spends a good amount of space making the case that those living in the Hejaz in Muhammad’s time in these communities wouldn’t have been able to write a complex text like the Quran; most of their writings are short personal messages placed on rocks that are the equivalent of “Kilroy was here.” He also makes a detailed case against any idea that Mecca was important in the spice trade, in the mining of minerals, or as a pilgrimage site. Patricia Crone’s work has been particularly devastating against any idea that Mecca was a thriving center of an international spice trade. Mecca, being a community incapable of growing crops, unlike Medina, was functionally the local version of an almost entirely non-literate, uncultured, impoverished “Gopher Prairie;” Medina wasn’t much better off despite it could irrigate some crops. Given this realistic portrayal of cultural and economic conditions in the Hejaz, it’s fully believable that Muhammad was indeed illiterate, much like many others in his community. (See Shoemaker generally, pp. 96-133).
Working from a skeptical, naturalistic perspective, Shoemaker and others who have examined the history of the development of religions and their texts find the standard Sunni Bukhari/Uthman story of the compilation and standardization of the Quran’s text to be exceedingly implausible. It would have happened way too quickly. Chase Robinson explains why this standard tradition of how the Quran’s text was collected is so unlikely (as quoted in Shoemaker, pp. 38-39): “The complicated and protracted processes that generated monotheist scripture in antiquity and late antiquity are generally measured in centuries or at least several decades; the [Sunni] tradition would have us believe that in the case of Islam they were telescoped into about twenty years. Are we really to think that within a single generation God’s word moved from individual lines and chapters scribbled on camel shoulder-blades and rocks to complete, single, fixed and authoritative text on papyrus or vellum? It would be virtually unprecedented. It is furthermore unlikely in the light of what we know of early Arabic: the nature of early Arabic scripture, which only imperfectly described vowels and consonants, and conventions of memorization and reading, which often privileged memory over written text, would militate against the very rapid production of the fixed and authoritative text that the tradition describes.” Here a strong contrast arises with the environment in which the New Testament was produced, which had much more widespread literacy, including the Founder’s own literacy (i.e., Luke 4:16), among Jewish people and also the educated gentiles with whom the likes of Paul rubbed shoulders, unlike the case for c. 700 A.D. Hejaz. The Old Testament was already a long standardized text upon which the earliest Jewish Christians would have found to serve as the obvious model to base their own faith upon when the Gospels were written down in Greek after a certain period of oral transmission in (mainly) Aramaic. The story of the transmission of Paul’s letters in this regard was simpler, however, since they started their lives as written text. As already surveyed above and as Shoemaker observes (p. 39), the primary sources that portray situation in which the Muslims received and produced the Quran have so many inconsistencies and contradictions demonstrates that historians shouldn’t mechanically place their faith in the standard Sunni Bukhari/Uthman story.
Another factor that likely retarded the collection and standardization of the Quran was the great authority given to the early caliphs over the community of Muslim believers. They were treated almost like vicars of God on earth functionally, because of the power they had to determine the beliefs and practices of the Muslim community, which was over and above their martial powers to wage war and to administer the law. As a result, the Quran itself gets very little attention from believers until the end of the eight century. Shoemaker explains the consequences of this dynamic (p. 41, italics removed): “This dynamic of a gradual shift from the caliphs’ direct authority as deputies of God to recognizing instead the authority of Muhammad’s teachings as remembers by the members of the ‘ulama also goes a long way toward explaining the Qur’an’s apparent absence from the Believers’ faith until the end of the seventh century, as evidenced by both the Islamic tradition itself and the various contemporary reports from writers outside of the community of the Believers.” Most strikingly, Muhammad isn’t mentioned at all by the early Muslim governmental authorities before the time of the Caliph Marwan I (684-85). According to Shoemaker (p. 41), the founder of Islam “is not named by any one of the papyri, inscriptions, or coins from this period.” However, by the time Abd al-Malik becomes caliph, Marwan I’s son, a pronounced shift occurs: Now the authority of Muhammad and the Quran are often publicly proclaimed to the Muslim community and to the wider world, when they had been neglected for 50 or 75 years by the Umayyad governmental authorities. A related reason why Uthman wouldn’t have been important in standardizing the Quran stems from his personal unpopularity and the weakness of the governmental apparatus at his command to coerce obedience in matters of faith at a distance from the Hejaz. He may have chosen a regional version of the Quran, such as that of Mecca or Medina, and then tried to impose its text on others, but lacked success in doing so. (See Shoemaker, pp. 40-41). Much like the standard weakness of the rabbinical sources making up the Mishna and the Talmud, who often projected earlier in time practices and institutions that actually came later, it’s overwhelmingly likely the same problem in reconstruction the past occurred here, in which what al-Malik actually did was projected onto Uthman and the earlier caliphs, who were seen as having more historical legitimacy since they lived and ruled in time closer to Muhammad.
At this point, let’s turn to presenting the evidence that Caliph al-Malik (685-705) and his right-hand man, al-Hajjajj compiled and edited the Quran. Although Muslims at times will admit that they had some influence on the text of the Quran, they attempt to limit those changes to minor amendments, such as the addition of diacritical marks and standardized spellings. However, many manuscripts copied after the early eighth century still lacked these features while others clearly did, which proves the falsity of this attempt to minimize al-Malik’s role in substantially producing the text of the Quran as we have it today. Francois Deroche perceives the problem with this kind of analysis (italics omitted): “If we turn to the reports stating that the diacritics were introduced in the course of al-Hajjajj’s ‘Masahif project’ and that ta and ya were selected in order to distinguish between the second and third person of some verbal forms, we have to admit that the manuscript evidence says otherwise.” However, these modest concessions to al-Malik’s role appear to be an attempt to arbitrarily harmonize the historical primary sources, which also mention the (supposed) roles of Abu Bakr, Umar, and Uthman as well.
Let’s examine some of the primary sources that attribute major roles to al-Malik and/or al-Hajjajj in editing and compiling the text of the Quran. One tradition attributes says that al-Malik said that he feared death in the month of Ramadan because (italics omitted), “That is the month in which I was born, it is the month in which I was weaned, it is the month in which I gathered together the Qur’an [jama’tu l-Qur’an], and it is the month in which I was sworn allegiance [as the caliph].” Another tradition maintain that al-Hajjajj sent codices with the new text of the Quran to all the major centers of the imperial realm, such as Medina, Kufa, Mecca, Basra, Damascus, and Egypt, with the goal of its replacing the local versions of the Quran then in use. In some cases, it was said that he was not only the first one to sent official codices to all the important cities of his master’s realm, but also he was the one who created the practice of having the Qur’an read aloud in mosques. He also instructed that all the older, local versions of the Quran should be collected and destroyed, much like it was said that Uthman had done in the official Sunni/Bukhari story. All the privately owned manuscripts of the Quran with the wrong text were to be seized and disposed of after paying the owners 60 dirham each. The Islamic governor of Egypt, confronted with al-Hajjajj’s order to accept the new text of the Quran, regarded his command as presumptuous, since he was of the same rank as al-Hajjajj. He replied that al-Hajjajj “permits himself to send a mushaf [codex] to the very military district [jund] where I am serving, me!” The Egyptian governor then responded by making his own edition of the Qur’an. This story completely undermines the standard Sunni narrative of Bukhari, which maintains Uthman’s efforts standardized the Quran’s text, since it indicates it didn’t exist in Egypt at the time al-Malik ruled. In Medina around this time, Uthman’s own family objected to al-Hajjajj’s edition of the Quran, according to Ibn Shabba. The people of Mecca were said to have asked Uthman’s family to produce a copy of the Quran of Uthman’s so they may read it. Uthman’s family responded to this request by saying that it had been destroyed on the same day when Uthman had been assassinated. (See Shoemaker, pp. 44-46).
Stephen J. Shoemaker, “Creating the Qur’an: A Historical-Critical Study,” is available for a free download at the University of California’s Luminos Web site, which provides Open Access to academic books. Click here for the details: https://luminosoa.org/site/books/m/10.1525/luminos.128/