Essay: What the British Library’s Hijazi Maʾil Quran (Or. 2165) reveals about early scribal practice and Quranic transmission

For many Muslims, Quran manuscripts are familiar objects woven into daily life. I started learning Quranic Arabic at age six, surrounded by mass-produced printed Qurans whose scripts I never questioned. Encountering the British Library’s Or. 2165 (fig. 1), an early Hijazi manuscript from the 8th century CE and often described as one of the oldest surviving Qurans (Dutton 2004), felt both extraordinary and deeply personal. Its handwritten slanted script and unadorned parchment revealed a formative moment in Islamic textual history very different from the polished copies I grew up with.

Fig. 1. Page from the Hijazi Maʾil Quran. British Library Or. 2165, fol. 76v (8th century), ink on vellum. London: British Library.

This essay argues that Or. 2165 illuminates a transitional moment in the transmission of the Quran, revealing how early scribes negotiated oral tradition, evolving written conventions, and collaborative manuscript production. Through close visual analysis and comparison with scholarship, I show how the manuscript embodies the earliest stages of the Islamic book tradition.

At first glance, Or. 2165 presents twenty-three lines of tightly compressed text in dark brown ink on a worn parchment ground. The letters lean noticeably to the right and are formed with thick strokes, creating a dense visual effect that suggests a manuscript intended for practical use rather than ceremonial display. This interpretation, however, invites further nuance. Illumination was far less common in this early period than in later centuries, and because the Quran survives only in fragmentary form, we cannot know whether it originally included an elaborate frontispiece. The question of how and by whom this manuscript was used, therefore, remains a complex one.

Amid the monochrome surface, a small red circle surrounded by dots stands out (fig. 3). There are no diacritical vowel marks, decorative headings, or illumination. Looking at Or. 2165, I found myself wondering about the scribe. Who they were, what environment they worked in, and how these fragile pages travelled from the early Islamic world to a British reading room. Its claim to antiquity invites both wonder and scepticism, prompting a close investigation of the script, materials, and scholarly debates surrounding its origins.

To understand the manuscript’s dating, it is necessary to identify why scholars classify Or. 2165 as written in Hijazi (Maʾil) script, and why this is important. Originating in the Hijaz region, Qurans of this type are generally dated to a few decades after the Prophet’s death, placing them in the late 7th to early 8th century (Deroche, 2014). Its rightward slant is consistent with descriptions attributed to prominent scholar and bibliographer Ibn al-Nadim (d. 380/990) (El Menshawy, Al-Maadeed & Adam 2018).

Letter forms such as the rounded mim with a fine tail and the nun ending in a semicircular stroke closely parallel the features identified by Francois Deroche (2014) in his palaeographic comparison with the related Codex Parisino‑petropolitanus (Paris Arabe 328a). His analysis situates this Hijazi hand within the earliest phase of Quran copying, approximately 695 CE, and more broadly within c. 650–704 CE. On palaeographic grounds, Paris 328a is usually placed toward the later part of this early range, in the late 7th rather than the 8th century.

Though visually modest, features such as the rightward slant, the small red circle encircled by dots, rounded mims with delicate tails, and nuns ending in semi-circular strokes reveal deliberate stylisation rather than a purely functional hand. These elements suggest that early scribes actively contributed to the emergence of a distinctive Quranic aesthetic in Islam’s first centuries.

The material features of Or. 2165, its parchment, ink, and codicological structure also help establish its antiquity, even while dating remains imprecise. Hijazi manuscripts were typically written on parchment during the 7th to 8th centuries (Deroche 2014). For comparable manuscripts, such as the Maʾil Quran in the Tareq Rajab Museum, radiocarbon dating has been used to establish the age of the parchment. This Quran leaf has been dated to the 7th century; however, radiocarbon analysis only reflects the preparation date of the skin, not the moment of writing (El Menshawy et al 2018).

Inks used for early Qurans were made from gallnuts, gum Arabic and vitriol, and were sometimes scented with camphor or musk (El Menshawy et al 2018). Whether Or. 2165 once carried such fragrance is unknown, but the knowledge that scribes crafted ink with such care underscores the reverence and craftsmanship involved in producing the Quran’s earliest written forms.

Early scribes worked within an oral culture, producing written copies that functioned as aids to accurate recitation (Rabb 2006). Scholarly consensus traditionally holds that the Quran was standardised under the caliph Uthman around 650 CE, producing the consonantal skeleton (rasm) that underpins all later readings (Bothmer 1986; Rabb 2006). Although this view was later challenged by Wansbrough, summarised in van Putten (2019), who argued that the canonical text could only have taken shape several centuries after the traditional date, more recent work supports the existence of an early written archetype.

Diem’s orthographic studies, discussed in van Putten (2019), show that certain features, such as the plene spelling of ā, represent Hijazi innovations beyond what was possible in the earlier Nabataean Aramaic script, indicating early scribal development rather than late editorial invention. Sadeghi and Bergmann (van Putten 2019) further define the ‘Uthmanic text type’ as a tradition in which surviving codices agree on surah order, verse order, and wording.

Dutton (2004) notes that Or. 2165 marks individual verses with clusters of six dashes (fig. 2) and groups of ten with red rosettes (fig. 3). Deroche (2014) adds that blank spaces separate suras (fig. 4), while the basmala appears to be a later addition (Al-Shareef et al 2020). These organisational features align closely with the Uthmanic tradition, supporting Or. 2165’s status as an early Quranic manuscript.

Scholars, however, differ on the manuscript’s precise provenance. Dutton (2004) argues that Or. 2165, along with the related MS Paris 328a, reflects the reading tradition of the Syrian Qari Ibn Amir (d. 118/736). He notes that their verse-ending patterns correspond to the Himsi system, leading him to date the manuscript to around 85 AH (704 CE).

Intisar Rabb (2006) broadly agrees with the Syrian attribution but refines it: she proposes that the manuscript likely originated in the Ḥimṣ region and only later moved toward Damascus, placing its production in the early Umayyad period (650–705 CE). Deroche (2014) remains more cautious, acknowledging Syrian features but refraining from assigning a precise geographical origin. Taken together, these views show how palaeography, orthography, and reading traditions intersect but do not yield certainty.

Producing a Quran of this scale often required multiple scribes. Deroche (2014) argues that at least two hands worked on Or. 2165: one writing the main skeletal text and another adding verse markers and later structural elements. Rabb (2006) identifies two early scribes plus several later hands who added corrections and headings, suggesting a long life of transmission and revision. The presence of multiple hands hints at a semi-organised scribal environment, perhaps a workshop or mosque context, where manuscripts travelled, were corrected, and augmented over time. It also reinforces the collaborative and devotional nature of early Quranic manuscript production.

Analysing the Hijazi Maʾil Quran has offered a window into the earliest stages of the Quran’s material formation. Although its exact dating and regional origin remain debated, Or. 2165’s script, verse markers, and multi-hand production strongly indicate its place among the earliest surviving Qurans. It embodies a transitional moment when oral transmission, emerging scribal practices, and evolving aesthetics met on the page.

Standing before Or. 2165 feels like witnessing the Quran as it first took material form, a profoundly moving thought.

Fig. 2. Individual verse marked with a cluster of six dashes. Hijazi Maʾil Quran (Or. 2165) British Library Or. 2165, fol. 76v (8th century), ink on vellum. London: British Library.

Fig. 3. Groups of ten verses marked with red rosettes. Hijazi Maʾil Quran (Or. 2165) , British Library Or. 2165, fol. 76v (8th century), ink on vellum. London: British Library.

Fig. 4. A blank space separating suras where Basmala appears to have been added later, Hijazi Maʾil Quran (Or. 2165), British Library Or. 2165, fol. 76v (8th century), ink on vellum. London: British Library.

References

Al-shareef, A.M. & Abdul Salam, Y.I. (2020) ‘Early manuscripts of Quran (through data of Hijazi calligraphy and archaeological evidence)’, Journal of the General Union of Arab Archaeologists, 5(1), Article 1.

Bothmer, H.C.G. von (1986) Masterworks of Islamic Book Art: Koranic Calligraphy and Illumination in the Manuscripts Found in the Great Mosque in Sanaa. In: Yemen – 3000 Years of Art and Civilisation in Arabia Felix. Munich: Priese Publishing; Innsbruck: Pinguin-Verlag; Frankfurt/Main: Umschau-Verlag; Amsterdam: Royal Tropical Institute.

Déroche, F. (2014) Qurʾans of the Umayyads: A First Overview. Leiden: Brill.

Dutton, Y. (2004) ‘Some Notes on the British Library’s “Oldest Qur’an Manuscript” (Or. 2165)’, Journal of Qur’anic Studies, 6(1), pp. 43–71.

El Menshawy, S., Al-Maadeed, S. & Adam, K. (2018) ‘An evaluation of Maʾil Qur’an manuscript’, Scientific Culture, 4(1), pp. 75–81.

Rabb, I.A. (2006) ‘Non-Canonical Readings of the Qurʾān: Recognition and Authenticity (The Ḥimṣī Reading)’, Journal of Qur’anic Studies, 8(2), pp. 84–127.

Van Putten, M. (2019) ‘‘The Grace of God’ as evidence for a written Uthmanic archetype: the importance of shared orthographic idiosyncrasies’, Bulletin of the School of Oriental and African Studies, 82(3), pp. 447–464.