The past two centuries have also witnessed intense research activity by western scholars on the origins, provenance and collection of the Qur'ān. Their views have been subject to various approaches depending upon the Zeitgeist concurrent with their period. However, the Western accounts of the collection of the Qur'ān can be primarily divided into two categories. To the first category belong scholars who have formed their views by taking the traditional Muslim accounts of collection as a starting point in some form or the other, while to the second category belong scholars who have completely rejected the traditional Muslim accounts and have in fact come up with alternative accounts on the formation and collection of the Qur'ān. For the sake of simplicity, I will term the former category as traditional Western scholars and the latter as radical Western scholars.
Traditional Western Scholars
As far as traditional Western scholars are concerned, with the phenomenal German scholar Theodor Noldeke (d. 1930) began a new era in the study of the history of the Qur'ān. His seminal work Geschichte des Qorans (Gottingen 1860) revised first by Friedrich Schwally (d. 1919) (Leipzig 1909-1919) and later by Gottehelf Bergstrasser (d. 1933) and Otto Pretzl (d. 1944) (Leipzig 1909-1919) is an outstanding piece of research. This magisterial work on the history of the Qur'ān bears the stamp of vintage thoroughness of German scholarship.
Noldeke's student and disciple Friedrich Schwally in the revised edition of Geschichte des Qorans expresses the view that parts of the Qur'ān had been written in the time of the Prophet (sws) and existed in some rudimentary form; however, it is difficult to ascertain how much was written by him and how much was added by later collectors. He is sceptical of the collection under Abū Bakr (rta) but, with some criticism, accepts the account of the collection made under 'Uthmān (rta). He surmises that the reports of the collection attributed to Abū Bakr (rta) were later fabrications meant to bestow the honour of a Qur'ān collection to him and to 'Umar (rta) and to reduce the stature of the 'Uthmānic collection because people had many complaints against 'Uthmān (rta). According to Schwally, it cannot be denied that Hafsah (rta) had a Qur'ān: it was either 'Umar's Qur'ān which she received from him as an inheritance or it was one she personally compiled. In both cases, it was a personal collection, and had nothing to do with any state collection by Abū Bakr (rta). According to Schwally, the Qur'ān of 'Uthmān (rta) was prepared by copying out the text found in the main codex which was the most important of the texts available in Madīnah. What thus becomes evident, he concludes, is that the word jam' (collection) is not appropriate at all for what 'Uthmān (rta) actually did and neither does this word appear in the main narrative reported in this regard. It is certain secondary narratives as well as some works on the sciences of the Qur'ān which use this word. However, his conclusion is that interpolation in the final compilation cannot be ruled out. His words are:
Ich stimme aber mit Fischer darin Uberein, dass die Moglichkeit von Interpolationen in Qoran unbedingt zegegeben warden muss.
Regis Blachere (d. 1973), more or less follows the same line as Schwally in his views on the collection of the Qur'ān in the times of Abū Bakr (rta) and 'Uthmān (rta). He is of the opinion that a personal collection was made by Abū Bakr (rta) in his time. Later, in the times of 'Uthmān (rta), he says, that this personal collection and other material was used to form an official collection. This official collection was enforced in the empire, and all other masāhif of the Companions (rta) which contained revelations they had directly recorded from Muhammad (sws) in his times were destroyed.
Montgomery Watt (d. 2006), like Noldeke is also of the opinion that at the death of the Prophet (sws) parts of the Qur'ān had been written yet no one completely memorized it because it had not been collected. Muhammad (sws) had "brought together many revealed passages and given them a definite order, and that this order was known to and adhered to by his Companions." After raising various questions on the collection attributed to Abū Bakr (rta), he concludes that no "complete collection of the Qur'ān was officially made during the caliphate of Abū Bakr (rta)." Watt says that the leaves of the Qur'ān in possession of Hafsah (rta) can hardly be regarded as the main or sole basis of the 'Uthmānic text, which was prepared by available pieces of revelation at that time. He finally concludes that what we have today is "essentially the 'Uthmānic Qur'ān; 'Uthmān's commission decided what was to be included and what was excluded; it fixed the number and order of the sūrahs and the outline of the consonantal text."
Arthur Jeffrey (d. 1959) is of the opinion that at the death of the Prophet (sws) "there was no collected, arranged, collated body of revelations." Various Companions (rta) had personally collected the Qur'ān in a codex. Like others, the collection of Abū Bakr (rta) was "a private affair". All these collections differed from one another and some of them became popular in various territories. Disputes and controversies sprung forth in the time of 'Uthmān (rta) as people wanted to adhere to the codex of their own area. In order to resolve these disputes, 'Uthmān (rta) actually canonized the Madīnan codex, assuming that there was one. He also says that contrary to this fact, some accounts mention that Madīnah "depended largely on oral tradition" and that 'Uthmān (rta) actually embarked upon collecting the Qur'ān afresh.
Nabia Abbott (d. 1981) agrees with the views of Schwally regarding the canonization of the text in the times of 'Uthmān (rta) with one qualification which in her own words is: "However, we do not accept the completeness and authenticity of 'Uthmān's edition to the extent Noldeke and Schwally do, for we admit with Hirschfeld not only omissions but interpolations of textual matter and even go so far as to admit with 'Abd al-Masīh al-Kindī, Casanova and Mingana possible changes introduced by Hajjāj, though both the nature and extent of these are not to be readily determined."
The view of the traditional western scholars can perhaps be summed up by saying that it was 'Uthmān (rta) in whose time the consonantal text of the Qur'ān was finalized, either by merely copying out Hafsah's collection (Schwally), or by using Hafsah's collection and other sources as well (Watt and Blachere) or by collecting the Qur'ān totally afresh (as per one view mentioned by Jeffrey). This prima facie might seem similar to the traditional Muslim accounts of collection. However, there is a world of difference. According to most Western scholars, the text finalized by 'Uthmān (rta) was not a true copy of what was revealed to the Prophet (sws).
Radical Western Scholars
Prominent among these scholars include Alphonso Mingana (d. 1937), Gunther Luling (b. 1928), Christoph Luxenberg, John Wansbrough (d. 2002), Yehuda Nevo (d. 1992) and John Burton (d. 2001). A brief summary of their views follows.
Under the influence of the French scholar Paul Casanova (d. 1926), Alphonso Mingana concludes that the Qur'ān was given a final shape in the times of 'Abd al-Malik ibn Marwān. While regarding Muslim sources on the accounts of collection as unreliable and contradictory, he focuses on some non-Muslims sources of those times which do not mention the presence of the Qur'ān among Muslims. He primarily draws on the 9th century Apology of the Christian faith written by al-Kindy at Māmūn's court which describes an account of the collection of the Qur'ān. As per this account, the Qur'ān was finalized by al-Hajjāj ibn Yūsuf, the powerful lieutenant of 'Abd al-Malik ibn Marwān who "caused to be omitted from the text a great many passages. Amongst these, they say, were verses revealed concerning the House of Omeyya with the names of certain, and concerning the House of 'Abbās also with names." Besides this, he also refers to some other accounts which are also devoid of any mention of the Muslim scriptural book.
Gunther Luling (b. 1928), uses philology to re-discover the Ur-Qur'ān in his work Uber den Ur-Qur'ān: Ansatze zur Rekonstruktion vorislamischer christlicher Strophenlieder im Qur'ān, 1st ed. Erlangen: Luling, 1974 which has recently been translated and developed as A Challenge to Islam for Reformation. In his opinion, the Qur'ān has four textual strata. The first stratum which is the original text is a strophic hymnal composed by the Christians of Makkah which comprised both Trinitarians and non-Trinitarians and these were written at least a century before Muhammad (sws). The second stratum consists of passages from that hymnal which were edited and Islamized in Muhammad's time. The third stratum contains sections which were exclusively written in the time of Muhammad. The fourth stratum is sections altered by later scholars during the process of orthographic editing. In a nutshell, the Qur'ān, in the opinion of Luling, is the product of several textual revisions.
Christoph Luxenberg also uses philology to re-interpret the original text of the Qur'ān. His basic thesis is that Syro-Aramaic was the lingua franca of Arabia in the 7th century and was replaced by Arabic much later by Arabs bred in this Syro-Aramaic culture. The Qur'ān he concluded was a mixture of Arabic and Syro-Aramaic words (aramaisch-arabische Mischsprache). In his Die syro-aramäische Lesart des Koran: Ein Beitrag zur Entschlüsselung der Koransprache(2000) recently translated into English as The Syro-Aramaic Reading of the Koran: A Contribution to the Decoding of the Language of the Koran, 1st ed. Berlin: Verlag Hans Schiler, 2007, he tries to demonstrate that many of the words and passages of the Qur'ān if interpreted as Syro-Aramaic words give a much more appropriate and meaningful interpretation to the text.
Yehuda De Nevo (d. 1992) relying on numismatics, epigraphic evidence and archaeological findings in the Negev Desert in Jordan reconstructs the early history of Islam. Disregarding traditional accounts of early Islamic history, he is of the opinion that the paganism alleged to be found in Hijaz in the age of jāhiliyyah a back-projection of the paganism found in Negev in the time of Abbasids. This is to couple the fact that archaeological researches show no trace of Jewish settlements in Khaybar and other parts of Arabia where they were thought to exist. He writes: "From the fact that the Qur'ān contains many phrases present in the Muslim inscriptions of the second century AH and later, but absent from inscriptions of Hishām's days or earlier, I would conclude that it was canonized quite late, ie. after these phrases had entered the religious vocabulary."
John Wansbrough (d. 2002) working axiomatically on the findings of Joseph Schacht (d. 1969) regarding Hadīth, dismisses the reports on the collection of the Qur'ān as historically un-reliable. He bases his studies as Motzki puts it on "a form-critical study of the Qur'ān and Muslim exegetical literature." In Wansbrough's view, "the structure itself of Muslim scripture lends little support to the theory of a deliberate edition. Particularly in the exempla of salvation history, characterized by variant traditions, but also in passages of exclusively paraenetic or eschatological content, ellipsis and repetition are such as to suggest not the carefully executed project of one or of many men, but rather the product of organic development from originally independent traditions during a long period of transmission." While commenting upon the schemata of revelation, he says that they "include a number of conventions typically employed to introduce the major themes of the Qur'ānic theodicy. Exhibiting a comparatively limited lexical range, those formulae serve to confirm the impression of a composition made of originally unrelated pericopes." He traces the chronological development of early (pre-Tabarī) Muslim exegetical literature to show that it reflected the needs of the emerging Muslim community. It produced in his words "the following exegetical typology: 1) Haggadic, 2) Halakhic, 3) Masoretic, 4) Rhetoric and 5) Allegoric." By employing an argumentum e silentio, he concludes that it was not before the period of the Masoretic exegesis that the scripture was canonized since the Haggadic and the Halakhic exegeses contain no reference to a stable standard text. This meant that the Muslim community, which in his opinion actually developed outside Arabia in Abbasid Iraq, remained without a ne varietur text of the Qur'ān until the end of 2nd century AH. The standard text emerged from an oral transmission of Prophetic logia and in his words the "establishment of a standard text such as is implied by the 'Uthmānic recension traditions can hardly have been earlier." In support of his thesis, he also tries to substantiate that the classical Arabic (poetic koinē) in which the Qur'ān is couched developed in the early 3rd century.
John Burton, like Wansbrough takes the conclusions of Goldziher-Schacht regarding Hadīth as the starting part of this inquiry and concludes that the narratives of the collection of the Qur'ān in the times of Abū Bakr and 'Uthmān were concocted by legal scholars in wordings which reflected an incomplete redaction of the Qur'ān. The motive behind giving an impression of an incomplete redaction was to justify certain legal views they held: they contended the source of their views were certain extraneous verses of the Qur'ān which were originally part of the Urtext. "This motive induced the Muslims to exclude their Prophet from the history of the collection of their Qur'ān texts. It was a compelling motive. It was their only motive."
A prominent example of such an extraneous verse being the stoning verse, which though not found in the Qur'ān is held to be still operational though its reading has been suppressed (mansūkh al-tilāwah dūn al-hukm). Central to Burton's thesis is the issue of abrogation which had occupied early usūl scholars. His conclusion in the wake of the fabricated reports on the collection of the Qur'ān is very simple but compelling: "What we have today in our hands is the mushaf of Muhammad."
In more recent times, Michael Cook (b. 1940) and Patricia Crone (b. 1945), both students of Wansbrough, have postulated an alternative account for the rise of Islam. Like their mentor, they also place the final canonization of the Qur'ān at a much later date than alleged by Muslim scholars. Basing their conclusion on archaeological findings and non-Islamic sources, they assert that except for one implicit piece of evidence, there is no proof for the existence of the Qur'ān before the end of the 7th century AH. In their words, the Qur'ān "is strikingly lacking in overall structure, frequently obscure and inconsequential in both language and content, perfunctory in its linking of disparate materials, and given to the repetition of whole passages in variant versions. On this basis it can plausibly be argued that the book is the product of belated and imperfect editing of materials from a plurality of traditions."