- One of the earliest Quranic manuscripts that has survived is the Codex Parisino Petropolitanus. This manuscript attests roughly 45% of the Qur’an. François Déroche, who is a leading scholar on early manuscripts and palaeography, has dated this manuscript to between 671 and 695 CE. He states: When and where was the Codex Parisino-petropolitanus transcribed? As we have seen, the text of the Fihrist suggests that the ḥijāzī style was among the earliest varieties of Arabic script. On the other hand, being a copy of an earlier codex, it can hardly be dated before the middle of the first/seventh century, the more so because the copyists are adjusting the orthography in order to note more precisely the Qurʾanic text. The highly idiosyncratic way in which the five copyists worked suggests on the other hand that the reforms of ʿAbd al-Malik (who reigned from 65/685 to 86/705) had not yet been enforced. It should however be noted that Hand D writes in a very regular way, almost professional. If we accept the information provided by the Muslim tradition about the intervention of the Umayyad governor of Kufa, the famous al-Ḥajjāj b. Yūsu fand what Omar Hamdan called the “Maṣāḥif project” ,we have to note that some of its features lack entirely in the manuscript. The manuscript could therefore be attributed to the third quarter of the first century ah, between 671 and 695ad. It would thus be one of the earliest witnesses of book production in Umayyad times. (Qurans of the Umayyads, p. 34)
- What exactly were the reforms which were made by Abd al-Malik?
- The development of Qur’anic scripts
- Both manuscripts are particularly important as they can be dated with some precision. Their script has some connections with the ḥijāzī style, not so much the version exemplified in the portions of the Codex Parisino-petropolitanus transcribed by Hands A or B (fig.1–2),but more with the work of Hand D (fig.4) or the copyists of London, BL Or.2165 (fig.8). The latter on the one hand as well as both the Damascus and Fustat codices on the other hand witness a completely new feature in the—young—history of the Arabic script: the deliberate iteration of a style of writing. The two hands cooperating in the transcription of Or. 2165 or the two (or more) copyists of the other two copies of the Qurʾan were able to transcribe the text in such a way that the difference between the hands was not immediately detectable. In other words, they belonged to a world where scribes had a professional approach to their trade, learning a specific style and using it. We may go a step further: we have before our eyes the beginnings of a new concept, that of Qurʾanic script. Such a specialization is perhaps not completely new in the area, but is assuredly new in the Arabic manuscript tradition, a style becoming specific to a certain use. These elements point into the same direction: at the end of the first/seventh century, under the reign of caliph ʿAbd al-Malik, a fateful change occurred in the chancery of the empire. Arabic, both language and script, became the official medium of the administration. From some sources, we know that the caliphs had specific scripts devised for their own correspondence. The relationship between the script of ʿAbdal-Malik’s milestones and that of contemporaneous muṣḥafs suggests that the latter were involved in this transformation which concurred with reforms involving the Qurʾanic text. (p. 94-95)
- Illumination
- The illuminations tell a similar story: the repertory of the illuminators of both manuscripts belongs to the Umayyad period and is found on many contemporary works of art. They provide us with an important chronological argument, but they also show that a major change was occurring, at least in some circles. The conception of the Qurʾan as a book was modified. The austere presentation of the Qurʾanic text which was the hallmark of the earliest copies in ḥijāzī style which have come down to us and was probably a faithful replication of older codices, lost its attraction and its meaning for some patrons who felt the necessity to have a book which would have a nicer appearance and use a visual repertory in accordance with the taste of Near Eastern elites of that time. It is no surprise that, in the same way as the first generation of Muslims adopted the techniques they needed for the writing down of the Qurʾan— like the codex or the scriptio continua—from the dominant manuscript tradition, in the last quarter of the first/seventh century, the illuminators and their patrons, with the aim of beautifying the muṣḥaf, appropriated from the Late Antique tradition a suitable decorative repertor (p. 95-96)
- How does Shoemaker respond to the arguments stated above? He says:
- Déroche, through careful paleographic and codicological study, has confirmed that that the earliest extant Qur’ans were in fact produced in the imperial chancery during the reign of ʿAbd al-Malik.101 Only the so-called Codex Parisino-petropolitanus poses a possible exception to this courtly context. Déroche dates this manuscript, which was written by five different hands, to sometime during the last thirty years of the seventh century, on the basis of its orthographic differences from the textus receptus. Nevertheless, I am not entirely persuaded that these variations alone can securely date the manuscript before ʿAbd al-Malik’s standardization campaign. It is true that this codex does not conform perfectly to the new imperial standard, but it certainly is not out of the question that this manuscript could be a work from the early eighth century that had not yet been impacted by the new reform. (Creating the Quran, p. 68)
- First, he states that Déroche has “confirmed” that the earliest Qurans (excluding the Paris codex) were actually produced during the reign of Abd al-Malik. That certainly is not the case. Déroche identifies at least five other hijazi manuscripts which share features with the Paris codex. Secondly, he states that Déroche has argued for an early dating of the Paris codex on the basis of “orthographic differences from the textus receptus”. As seen above, there are several other reasons to date the Paris codex and other hijazi manuscripts earlier than the reign of Abd al-Malik: they still have their own unique, personal, writing styles and they generally lack margins and illumination. Shoemaker goes on to say:
- Nevertheless, whether this manuscript was written in 685 or 705 or 715, any of which years could be possible judging from the paleography alone, there is no question that it does not fit with the canonical narrative of an ʿUthmānic standardization, which it also belies. As Déroche rightly maintains, “When looking at the transcription of the Codex Parisino-petropolitanus, one sees that this copy as well as all those which belong to these chronological strata of the transmission are unable to prevent what the ʿUthmānic edition was supposed to achieve”—that is, variation in the Qur’anic text.
- If the manuscript was written in 685 (i.e. before Abd al-Malik’s project), then it is actually Shoemaker’s narrative that it would not fit. Deroche repeatedly states that with a few exceptions, the Paris codex agrees with the text that is used today. He states:
- On a more important level, that of canonicity, the text found in the Codex Parisino-petropolitanus, in spite of some peculiarities in the division into verses or in the text itself, is consistent with the ʿUthmanic rasm since we can surmise for the moment that the differences in orthography and the lack of diacritical marks do not impair the possibility to read it according to the canon. (p. 37)
- and:
- However, we can reconstruct the appearance of the first manuscripts of the Qurʾan on the basis of the earliest evidence known to us, like the Codex Parisino-petropolitanus, which was probably transcribed within the first two decades of Umayyad rule. The evidence they provide, when confronted with the accounts transmitted by the Islamic tradition about the writing down of the Qurʾan, confirms that these reports contain without doubt a historical core and, notably in the case of the Codex Parisino-petropolitanus, that a text compatible with the canonical version was transmitted (p. 136)
- If the Qur’an was only canonised in the early 8th century, as Shoemaker maintains, it would obviously be problematic if a manuscript from the 680s exists which is compatible with that canonical text. If we maintain the early dating, we only have 2 options:
- The text was compiled before Abd al-Malik.
- This manuscript represents one of the “proto-Quranic” sources and Abd al-Malik decided to copy it word for word. Amazingly enough, this proto-Quranic source has survived until today.