Introduction

The term majmūʿa (pl. majmūʿāt), occasionally majmūʿ (pl. majāmīʿ), is the terminus technicus and past participle of jamaʿa, “to assemble, to collect,” and accordingly means “brought together,” in the sense of an assemblage, collection, or compilation.¹For a short discussion of the Arabic terminology, see ‎Gacek, The Arabic Manuscript Tradition, 26‎. ‎Friedrich and Schwarke, One-Volume Libraries, 12‎. It refers to manuscript containing multiple texts of any kind, either in full or excerpted form. Majmūʿāt are open to the entire world of Arabic literature: they can include any form of prose, poetry, or both. They may contain text in one language or in two or more languages. ²For the “conversation” of various linguistic and intellectual traditions in South Asian Islamic scholarship, see ‎Bahl, Mobile Manuscripts, 262–66, and passim‎.

Manuscripts containing multiple texts have long been intensively studied in several Western disciplines, such as classical, Byzantine, medieval Latin, German, and Romance philologies. Only recently has research in Arabic studies began to flourish within the framework of comparative codicology, which is dedicated to manuscript cultures worldwide.³For the beginnings and evolution of the research group Comparative Oriental Manuscript Studies (COMST, 2009–2014), at Hamburg University see the general introduction in the comprehensive manual ‎Bausi et al., Comparative Oriental Manuscript Studies, 9‎. The successor institution, the Centre for the Study of Manuscript Cultures (CSMC), is dedicated to manuscript cultures worldwide: https://www.csmc.uni-hamburg.de/ First and foremost, the focus here lies on a refined analysis of the material aspects of the codex, such as formats, quires, stratigraphy, layout, organization, and other elements that can provide information on the history and evolution of individual manuscripts in a given culture and during its production and circulation phases.⁴See the pioneering volumes at CSMC devoted exclusively to this manuscript types ‎Friedrich and Schwarke, One-Volume Libraries‎ and its follow-up, ‎Bausi, Friedrich and Maniaci, The Emergence of Multiple-Text Manuscripts‎ Both volumes also contain studies on Arabic manuscripts. Comparative manuscript studies draw on concepts and research approaches from cultural studies and material philology, see Friedrich and Schwarke, in their introduction to ‎One-Volume Libraries‎. This section outlines key contributions to the history of research: These include the conference volume ‎Nichols and Wenzel, The Whole Book‎ , ‎Gumbert, “Codicological Units”‎ as well as, based on Greek and Latin philology, ‎Andrist, Canart and Maniaci, La syntaxe du codex‎. Gumbert and Andrist et. al. focus strictly on the analysis of the codicological structure of manuscripts, whereas Nichols and Wenzel open various avenues of research into the medieval miscellany. See also the interdisciplinary publication series: https://www.materiale-textkulturen.org/ As a result of the mentioned emergence of new concepts and perspectives, majmūʿāt have increasingly become the focus of manuscript research.⁵This mainly applies to Arabic majmūʿāt and Ottoman-Turkish mejmuas. For insight into the Ottoman manuscript tradition as well as recent research activities, see https://mecmuaconference.univie.ac.at/ (in particular the conference´s booklet).

The majmūʿa manuscript type consists of two basis units that exist in all manuscript cultures. To highlight their substantial material difference, the two basic types “multiple-text manuscript” (MTM) and “composite manuscript” (CM) have been defined in comparative manuscript studies. Here, “multiple-text manuscript” is referred to as “a codicological unit with two or more texts resulting from one production process delimited in time and space.” In contrast, the less common type of “composite manuscript” designates “a codicological unit which is made up of formerly independent units.”⁶On the definition and discussion of the terminology of MTM and CM, see ‎Friedrich and Schwarke, One-Volume Libraries, 9‎, 15–16 (according to the definition of “unité du poroduction” and “unité du circulation” in ‎Andrist, Canart and Maniaci, La syntaxe du codex, 59–61‎). This differentiation “is closely tied with the production context” of the manuscript and was irrelevant for medieval Arabic scholars using and studying the texts.⁷See ‎Hirschler, “The Development of Arabic Multiple-Text and Composite Manuscripts,” 278–79‎.

However, as is often the case, the term “multiple-text manuscript” can be misleading when used in a general and overarching sense that silently absorbs the second basis type without differentiation. The term “miscellany,” well-established in Western research and bibliographical documentation, has also been subject to criticism because of its ambiguity, as the term suggests a conceptless mixture of disparate texts and ignores “the dynamics of individual examples of the genre.”⁸‎Friedrich and Schwarke, One-Volume Libraries, 15‎. For “power and limits” of the notion of miscellaneity, see ‎O’Donnell, “Retractions.”‎ As a result, there is as yet no single English term that encompasses the two basic types.

The terminological shortcomings may also explain why multiple-text and composite manuscripts are repeatedly referred to using metaphorical terms—for example “one-volume library,”⁹In his classic essay ‎“From Arabic Books and Manuscripts”,‎ Rosenthal recognizes a close relationship between the content of a majmūʿa—which he defines as a comprehensive compilation of philosophical and scientific works—and the intellectual outlook of its owner and compiler, the Yemeni bookseller Muḥammad b. Ḥasan al-Nihmī (ninth/fifteenth century). , “house,”¹⁰‎Nichols and Wenzel, The Whole Book, 127‎. “a unique textual monument […] within the scholarly landscape”¹¹‎Hirschler, “The Development of Arabic Multiple-Text and Composite Manuscripts,” 290‎. or “treasure trove of intellectual […] history.”¹²‎Schmidt, “From One-Volume-Libraries to Scrapbooks,” 218‎. Some designations use rather technical terms, such as “hard disk,” alluding to the ability to store a variety of signs and symbols,¹³ See back cover on ‎Nichols and Wenzel, The Whole Book‎. or “corpus organizer,”¹⁴Term by Alessandro Bausi, see ‎Friedrich and Schwarke, One-Volume Libraries, 17‎. a term which focuses on the universal intellectual act of depositing and ordering knowledge. All these terms imply a human being who strategically reads, copies, selects, plans, and organizes in order to preserve texts or excerpts for a particular purpose. Therefore, disclosing hidden correlations between compiler, collection and the manuscript´s material structure is one of the most promising research dimensions with regards to majmūʿāt.

Majmūʿāt were produced in many subtypes and forms, ranging from personal notebooks filled with excerpts, quotations, text snippets, lists, and sketches; to collections of various works from a single genre or several genres; to canonical compendia in any discipline. Most multiple-text manuscripts and composite manuscripts do not have a specific name or title. Some selections, such as florilegia and epistles, can be designated as muntakhab (or muntakhaba), muntajab (or muntajaba), or zubda … (butter, cream of….), all of which are prestigious names meaning “exquisite selection,” “choice,” or “the best of…” Manuscripts of “anthologies,” which are either defined by a specific subject matter or a literary theme,¹⁵On these broad categories and their respective differentiation, subtypes and forms see ‎Bauer, “Literarische Anthologien der Mamlūkenzeit.”‎ however, stand at the verge of being classified as “multiple-text manuscripts.” This is because they constitute independent and self-contained works, with an entertaining, didactic, philosophical, or religious purpose and a correspondingly ornate title.¹⁶In many cases metaphorical titles refer to precious objects, such as al-Kashkūl (the title of an anthology by Bahāʾ al-Dīn al-ʿĀmilī, who died in 1030 AH/1621 CE), which alludes to an artistically decorated and richly filled beggar´s bowl, or al-Iqd al-farīd, “The unique necklace,’´ the title of an anthology by Ibn ʿAbd al-Rabbīh al-Andalusī (d. 328 AH/920 CE). Widespread during Mamluk and Ottoman times, anthologies take many forms in terms of theme and structure, and thus cannot be clearly delimited.¹⁷See ‎Bauer, “Anthologies, Arabic Literature (Post-Mongol Period)”‎; ‎Schmidt, “Anthologies, Ottoman.” However, they differ from the majority of multiple-text materials, which usually contain collections with no fixed order¹⁸Several contributions in ‎Bausi, Friedrich and Maniaci, The Emergence of Multiple-Text Manuscripts‎, reflect on subvarieties, as well on common and differentiating aspects of the production, transmission, and reception of MTM and CM. and, unlike anthologies, are not considered independent, coherent works whose meaning is created through specific internal arrangement and semantic cohesion.¹⁹See ‎Bauer, “Anthologies, Arabic Literature (Post-Mongol Period)”‎; ‎Schmidt, “Anthologies, Ottoman”‎. Nevertheless, anthologies share essential characteristics and functions with most multiple-text manuscripts. Like any other majmūʿa, guided by the principles of collection and selection.²⁰See ‎Bauer, “Literarische Anthologien der Mamlūkenzeit,” 107‎. Their composition, and thus their inherent structure and meaning, can also undergo significant transformation during the process of transmission.

Although several scholars have appreciated, researched and described the distinctive, highly developed culture of Muslim scholarship,²¹In his comprehensive and classical work, ‎“The Technique and Approach of Muslim Scholarship”‎ Rosenthal draws on the work of the Damascene scholar ʿAbd al-Bāsiṭ al-ʿAlmawī (d. 981/1573) to describe the practices and norms in the Arab-Islamic scholarly culture. many others have overlooked the intrinsic value and high research potential of majmūʿāt.²²In his classic essay ‎“From Arabic Books and Manuscripts”‎ Rosenthal recognizes a close relationship between the content of a majmūʿa—which he defines as a comprehensive compilation of philosophical and scientific works—and the intellectual outlook of its owner and compiler, the Yemeni bookseller Muḥammad b. Ḥasan al-Nihmī (ninth/fifteenth century). For more on the general lag in Arabic studies in this field, see ‎Friedrich and Schwarke, One-Volume Libraries, 11–12‎. Valuable standard works like Déroche, Islamic Codicology and Gacek, The Arabic Manuscript Tradition do not yet take MTM and CM into account. The nearly complete lag of knowledge and research on majmūʿāt is also obvious in ‎Sagaria Rossi, “Manuscripts in Arabic Script”,‎ where this manuscript form is not mentioned. In most modern library catalogues produced in the Arab world or in the West, majmūʿāt were not treated as a distinct codicological type.²³See ‎Hirschler, A Monument of Medieval Syrian Book Culture, 119‎; ‎Hirschler, “The Development of Arabic Multiple-Text and Composite Manuscripts,” 275–76‎. Apart from notable exceptions,²⁴See ‎Sawwas, Fihris makhṭūṭāt Dār al-Kutub al-Ẓāhirīya‎ ; ‎Schoeler, Arabische Handschriften‎; ‎Wiesmüller, Die Rifāʿīya (Refaiya)‎. this manuscript type is often inadequately catalogued and described. As a result, the diverse works compiled within a multiple-text or composite manuscript are often either ignored or distributed across the catalogue under their respective thematic categories. The lack of consistency in cataloguing majmūʿāt is not a modern phenomenon. It is, for example, similar to the majāmīʿ in two otherwise carefully and well thought-out arranged Mamluk library documents analyzed by Konrad Hirschler.²⁵See ‎Hirschler, Medieval Damascus, 67–70‎; 86; ‎Hirschler, A Monument of Medieval Syrian Book Culture, 172‎.

The practice of collecting various texts into one manuscript is well documented from the third/ninth century.²⁶An example of a very old multiple-text manuscript dating from the fourth/tenth century ‎Leipzig, Universitätsbibliothek, MS Vollers 505 (Majmūʿa)‎ which once belonged to the al-Rifāʿīya private library in Damascus. This manuscript, copied in the year 389/990, contains the divans of the pre-Islamic and early Islamic poets Abū Ṭālib, Abū-l-Aswad al-Duʾalī, and Suḥaym ʿAbd-Banī l-Ḥasḥās. Production of multiple-text and composite manuscripts spread considerably during the institutionalization of Islamic scholarship in Mamluk and Ottoman times. Recent, pathbreaking investigations into the categorization and physical arrangement of multiple-text and composite manuscripts in Mamluk and Ottoman libraries offer rich new insights into the development of scholarship and library culture in specific collections and locations, for example in the Ṣāliḥiyya quarter of Damascus.²⁷‎Hirschler, Medieval Damascus, 69‎, takes as a starting point of his research a unique document, namely the catalogue of the Ashrafiyya Library housed in the mausoleum of the Ayyubid al-Malik al-Ashraf in the center of medieval Damascus. The catalogue, compiled in the 670s/1270s, provides insights into the classification system and the holdings of a large library: 8% of the approximately 2,000 manuscripts held there were majmūʿāt. In ‎A Monument of Medieval Syrian Book Culture, 121‎, Hirschler examines the late Mamluk private library predominantly devoted to Hadith, owned by the Damascene scholar Ibn ʿAbd al-Hādī (d. 909/1503), on the basis of an inventory (Fihrist). There, 48% of 578 catalogue entries are majmūʿāt, most of them of the CM type. The private Rifāʿīya library, researched by Liebrenz and catalogued by Wiesmüller, was like the above located in the Ṣāliḥīya quarter of Damascus (eighteenth–nineteenth centuries CE). This collection was sold by its owner to Leipzig University Library in 1853. It contained 471 manuscripts of which just under 20% were majmūʿāt; for a detailed characterization see ‎Liebrenz, Die Rifa’iya aus Damaskus, 121‎. In her descriptions, Wiesmüller ‎Die Rifāʿīya (Refaiya)‎ documents the individual composition of each majmūʿa, 92 in total.

As a result of the mentioned emergence of cross-disciplinary and comparative approaches majmūʿāt have increasingly become the focus of manuscript research.²⁸Two (and more) chapters in ‎Bausi, Friedrich and Maniaci, The Emergence of Multiple-Text Manuscripts, 215–46; 245–304‎ document the meanwhile advanced research paradigms in Arabic Studies are based on key insights from cross-disciplinary codicology: ‎Ragetti, “Rolling Stones Do Gather”,‎ and ‎Hirschler, “The Development of Arabic Multiple-Text and Composite Manuscripts.”‎ In the following, three case studies on the vast corpus of Ismaili text production are presented, providing insights into the different features of majmūʿāt which are of relevance for manuscript studies.

Corpus

The three case studies presented in the following section are based on multiple-text and composite manuscripts which originate from the field of Ismaili Studies, the subject of my research.²⁹My research on the literary culture of an Islamic minority is part of the project “Bibliotheca Arabica – Towards a New History of Arabic Literature” (https://www.saw-leipzig.de/en/research/projects/bibliotheca-arabica). The manuscripts presented were created and used by Tayyibi-Ismailis in India and Nizari Ismailis in north-west Syria.³⁰The Ismailis are a religious community within Shiʿi Islam going back to the 3rd/9th century. On Ismaili literature and intellectual history, see ‎Madelung, “Ismāʿīliyya”‎ The two main branches of the Ismāʿīliyya are the Nizari Ismailis (today a cosmopolitan global community with historical roots in Syria, Persia, Central and South Asia) and the Tayyibi Ismailis (today known as Bohras with their main base in India but with historical roots in Yemen and South Asia). For the Fatimids, see ‎Klemm, “Fatimiden”.‎ Ismaili literature dates back to the period of immediate pre-Fatimid and early Fatimid dynasty (297–567/ 909–1171). For literary historical aspects of Fatimid and Ismaili Literature see: ‎Walker, Exploring an Islamic Empire‎ ; ‎Daftary, “Intellectual Life among the Ismailis”‎; ‎Klemm, “A Library in One Volume”‎ (with further references). As is the case with most religious minorities around the world, manuscripts—as part of global religious and cultural heritage—are strictly protected and safeguarded from threats and loss. Furthermore, Ismaili religious and philosophical works are regarded as strictly secret and accessible only to the religious elite. However, in specific cases and under certain conditions, some owners of manuscripts have opened their collections for researchers.³¹See ‎Akkerman, A Neo-Fatimid Treasury of Books‎. The author has conducted her pioneering study on the continuous manuscript production and the socio-religious significance of manuscripts in the Khizānat al-daʿwa, the library of the leading Dāʿī of the tiny congregation of Bohra-Alawis in the city of Vadodara (Baroda) in Gujarat. The collection built up by the Institute of Ismaili Studies in London (IIS) since the 1980s comprises around 3,000 manuscripts in Arabic and Persian, as well as Central and South Asian regional languages, and consists of acquisitions and bequests from owners keen to ensure the safe preservation, restoration, and accessibility of the manuscripts.³²However, the collection of the Institute of Ismaili Studies in London is still not accessible on a digital platform, except for the presentation of a few selected highlights: https://www.iis.ac.uk/research/ismaili-special-collections/access/ As for the Hamdani Collection see https://www.iis.ac.uk/multimedia/manuscripts-in-hamdani-collection/ as well as the extensive, research-based catalogue of ‎De Blois, Arabic, Persian and Gujarati Manuscripts‎. For this reason, the renowned Hamdani Collection as well as a group of Ismaili majmūʿāt from Syria were entrusted to the institute as estate. The following case studies draw on these two corpora within the holdings of the IIS.

Thus, despite their limited but increasingly improving accessibility, research into the multifarious Ismaili manuscript cultures constitutes a relatively new and flourishing field of research within the scope of Ismaili studies and beyond. Accordingly, the following examination will focus on recent research into the historical production, transmission, and reception of Ismaili majmūʿāt. Although Ismaili manuscript traditions represent distinct cultural spheres, they form part of the highly diverse Arabo-Islamic manuscript traditions. Thus, among Ismailis communities outside the Arab world, Arabic is revered as the language of religious tradition and continues to be cultivated in their religious and scholarly writings, often coexisting with multilingualism.

The following examples cannot therefore be regarded as representative of a cohesive and overarching Islamic-Arabic manuscript culture. Instead, they are part of the great diversity of its historical and cultural traditions.

Historical Semantics and Contextual Analysis

Majmūʿāt as Production Units: Social and Religious Milieus

Multiple-text and composite manuscripts are valuable sources for exploring various aspects of ‘codicology in context,’ particularly the social environments of manuscript production and use. This will be illustrated using the example of a group of ten majmūʿāt belonging to the Nizari Ismailis, who lived in seclusion until the late nineteenth century in the coastal mountain range of Jabal al-ʿAlawiyya (today’s Jabal al-Anṣāriyya) in northwestern Syria. Only little is known about the literary history of this community in Syria, which retreated into the mountains after the end of Fatimid rule (297–567/909–1171). There, their manuscripts often fell prey to their local enemies. As a result, many manuscripts of the Syrian Ismailis have been destroyed, lost, or remained undocumented and hidden until the present day.

An opportunity to gain deeper insight into the literary heritage of the community was provided thanks to the estate of a private collector from Salamiyya.³³In 1265/1848, the Ottoman sultan ʿAbd al-Majīd I (1255–77/1839–61) granted the Nizaris the right to settle in the Syrian town Salamiyya, the long-abandoned place of origin of the Fatimids, see ‎Klemm, “Hidden Literary History,” 195‎. Housed at the IIS, this private collection represents a small but valuable subcorpus of its rich collection. It contains a small group of ten majmūʿāt, most of them copied in the nineteenth and early twentieth century.³⁴See ‎Klemm, Regional and Transregional Ismaʿili Tradition in Syria‎ (forthcoming): The study is dedicated to a corpus of ten Ismaʿili majmuʿat from Syria acquired by the Institute of Ismaili Studies from the estate of the literary historian Mustafa Ghalib from Salamiyya. Some of these manuscripts have been accessible to me in physical or at least in digital form. A further fundament of research was Cortese’s catalogue ‎Cortese, Ismaili and Other Arabic Manuscripts‎ that offers descriptions and literary-historical explorations into the works collected in the majmūʿāt. An initial examination of their predominately religious content revealed historical transmission networks stretching between Tartus in the Syrian coastal region to Aleppo and the Ismaili state of Alamut in northern Persia.³⁵Centre of the Nizari Ismaili State in Northern Iran (483/1050 until the Mongol conquest 654/1256): ‎Daftary, “Alamūt.”‎ One manuscript contained a rare textual witness from the region, an almost complete copy of the popular hagiography (Manāqib al-Mawla Rāshid al-Dīn) of Rāshd al-Dīn Sinān (d. ca. 589/1193), called the “Old Man of the Mountain” by the Crusaders, who was the charismatic founder and leader of the Ismaili mini-state in Syria. This collection of popular stories about his heroic deeds and miracles dates back to the Mamluk period. This textual witness highlights the function of majmūʿāt as “treasure troves” for works that would otherwise have been lost.

The group of ten multiple-text and composite manuscripts contains about one hundred titled entries of prose or poetry, as well as additional extracts and text snippets in various forms and styles. Each manuscript is a unique composition with regard to the combination and arrangement of its content.

The heterogeneous literary materials contained in the majmūʿāt are represented either in full or as extracts and text fragments (mukhtaṣara, faṣl). As for the genres or forms, there are prose forms, mostly labeled as book (kitāb) or treatise (risāla). Prose forms include spoken speech, as in question-and-answer-form (masāʾil), sessions of wisdom (majālis), oral traditions (ḥadīth, riwāya) of the Prophet and the first six Shiʿite Imams, as well as praise to the Imam (madḥ al-Imām) and prayers (adʿiya). Some texts combine prose and poetry. Poetic forms are mostly qaṣīdas as well as the local religious hymns (anāshīd dīniyya), which were composed by Ismaili Shaykhs (shuyūkh)—religious authorities and instructors.

Many of the works and extracts in these majmūʿāt belong to the secret (bāṭin), esoteric, and philosophical spheres of the Ismaili faith and doctrine, and were intended to be accessible only to the Shaykhs and initiated members of the community. The bāṭin literary sphere includes allegoric interpretations (taʾwīl) of the religious tradition; words, citations, and invocations spoken by the dāʿī (the religious leader of the community) during the ceremonial taking of the oath of allegiance; and another rare textual witness, the Dustūr al-Mawla ʿAlāʾ al-Dīn containing the religious and social code for the adherents and higher ranks of the daʿwa, which was issued by the Imam ʿAlāʾ al-Dīn Muḥammad III (d. 653/1255) of the castle of Alamut.

Most of the works and extracts in the majmūʿāt belong to the sphere of secrecy, since they contain ḥaqāʾiq, “spiritual truth.” This content clearly points to the hierarchical status of their owners. They must have belonged to the religious elite who underwent the process of initiation into the concealed bāṭin (esoteric) spheres. This is also confirmed by an ownership statement found in two composite manuscripts. Accordingly, the owner of the two manuscripts was a high-ranking envoy to the newly established axis that, from the the end of the nineteenth century, linked the Syrian Ismailis with Imam Muḥammad Shāh Agha Khan III. in Karachi (r. 1885–1957).

In the case of other majmūʿāt constituting multiple-text manuscripts written by one hand, the owners may also have been the scribes of their manuscripts. Due to the bāṭin status of many of the represented works, they had access to the esoteric writings from which they selected their material. Therefore, it is reasonable to assume that they used their personal majmūʿāt as one-volume libraries for spiritual edification. According to oral information from the region, majmūʿāt were used in subterranean rooms in Salamiyya for teaching and communal reading purpose until the second half of the twentieth century.

In conclusion, the religious content of the majmūʿāt suggests the elevated religious status of their owners, as well as the function of these manuscripts as repositories and practical handbooks for rituals and recitations. The majmūʿāt therefore offer an authentic insight into Ismaili religious literature and culture in Syria.

Majmūʿāt as Circulation Units: Composition and Transformation

The Majmūʿ al-tarbiya, a self-contained compendium of fifty-one separate Ismaili religious texts, was examined by Delia Cortese on the basis of twelve manuscripts housed in the Institute of Ismaili Studies.³⁶See ‎Cortese, “The Majmūʿ al-Tarbiya between Text and Paratext”‎ from which most information on this case study are taken. The twelve manuscripts are shortly described on p. 163f. The collection, including its extensive preface, is attributed to Muḥammad b. Ṭāhir al-Ḥārithī (d. 584/1188), a high-ranking religious dignitary within the Tayyibi-Ismailis in Yemen. The collection entails ḥaqāʾiq, reserved exclusively for the sworn and initiated members of the community (see above example 1). According to the preface of the compiler, his collection should serve as a literary canon for religious education and instruction. At the same time, the compendium was intended to preserve and transmit the authoritative Fatimid and Tayyibi texts. The collection has been preserved in many multiple-text manuscripts, the oldest of which dates from 1121/1709 CE. As Cortese shows, the Majmūʿ al-Tarbiya played an important role, particularly in the religious life of the Indian Daʾudi Tayyibi Ismailis, where it eventually gained high normative status.

Although the textual contents of the compendium remained largely unchanged during its transmission, the compilation nevertheless underwent considerable transformation. Thus, the approximately fifty-one texts appeared in different arrangements and groups, not always in the same order. Some scribes also added pages containing new short texts.

However, the most important sources that shed light on the reception and readers’ use of the work are its paratexts. In general, these textual elements, which are abundant in Arabic manuscripts, accompany the main text and comprise a great variety of types and functions. There are two basic types of paratexts: those that relate to the main text itself (ḥāshiya, pl. hawāshī, endocentric) or those that relate to its external environment (exocentric).³⁷Terminology according to ‎Moulin, “Rand und Band”‎ For typology and terminology, also in the Arabic tradition, see ‎Brinkmann, “Introduction,” 14–18‎; ‎Gacek, Arabic Manuscripts, 79‎ (commentaries), 202–3 (preface, incipit). In addition, some paratexts combine both endocentric and exocentric elements.

Many endocentric, text-related paratexts applied by scribes include headings, introductions (incipits), prefaces, colophons, and tables of contents. Endocentric paratexts may be short, such as glosses, corrections, or lexicographical explanations; conversely, they can be detailed and systematically applied by a scholar who comments on, discusses, and explains the text, or cross-copying variants found in other manuscripts.

The exocentric basic type of annotation, on the other hand, is often characterized by its documentary character. These so-called manuscript notes (overarching term: taʿlīq, taʿlīqāt) may be reading notes, study records (samāʿāt), transmission licences (ijāzāt), information on the purchase price of the manuscript, autobiographical notes, and statements by scribes, owners, buyers, readers, and other actors.³⁸See ‎Déroche, Islamic Codicology, 311–29, 330–44‎. In both chapters, the author presents a series of key elements, as title pages, ex libris, table of content, colophon, dates, formulas, dees of endowments (waqf), certificates (as samāʿāt, ijāzāt and more). ‎Görke and Hirschler, Manuscript Notes as Documentary Sources, 9–16‎. These paratexts, if analyzed systematically, provide clues to, or even deeper insight into, the respective social environment of the manuscript, its individual “biography” and provenance, as well as its context of use.³⁹See ‎Liebrenz, Empire of Books‎.

Manuscripts of Majmūʿ al-Tarbiya contain a variety of endocentric paratexts, thus mirroring different needs of teaching, learning, studying the instructive works and excerpts. In addition, added prefaces, tables of contents, author names, headings often highlighted in colors, serve both as ornaments and finding aids. According to Cortese, sometimes parts of the content move from the main text to the margins, and paratexts from the margins to the body of the texts. These dynamics between the interior and exterior of the text block thus bears witness to the fluidity “between text and paratext,” causing gradual transformation of the textual content of the compendium.⁴⁰See ‎Cortese, “The Majmūʿ al-Tarbiya between Text and Paratext,” 163–67‎; ‎Brinkmann, Marginal Matters, 45–47‎.

Related research in other fields of Arabic literary studies has emphasized the active role of scribes and scholars—often anonymous—in composite textual traditions, as exemplified by Kalīla wa-Dimna, the globally disseminated anthology of ethical and practical advice. Examples from the work’s transmission history, meticulously analyzed by Beatrice Gruendler and Khoulud Khalfallah, show that certain scribes can be designated as “copyist-redactors.” This type of agent, or even co-author, can intervene by means of paratexts. For example, a new, detailed, multi-part text block may be added at the beginning to redefine the purpose and meaning of the collection. Using the method of cross-copying, the scribe can add text passages from various source manuscripts to his or her own copy of the work,⁴¹Women, too, could be professionally trained and active as scribes, although evidence for this is rare: Thus, Ṣafiyya (d. 1911), daughter of Muḥammad ʿAlī al-Hamdānī was educated by her father as a manuscript copyist: ‎Klemm, “A Library in One Volume,” 390‎. Examples of her work are kept in the manuscript collection of the IIS. thus playing a key role in the historical evolution of the text collection.

Both the Majmūʿ al-Tarbiya, a compendium of Ismaili religious works, and the collection of fables Kalīla wa-Dimna, which offers wisdom and advice to a wide audience, feature rich examples of cross-copying in composite textual traditions. In both oeuvres, editorial activities, interventions and rewriting by scribes lead to the gradual transformation of a self-contained composite textual tradition in multiple-text manuscripts.⁴²See Gruendler and ‎Gründler and Khalfallah, “Anthologizing Kalīla wa-Dimna.”‎ Analyzing the role of agents who have hitherto been overlooked in research sheds new light on the transmission history of a collection.⁴³It should be noted, however, that in medieval Islamic civilization transmission and authentication of seminal texts from different disciplines/fields of learning were governed by strict standards in order to ensure the coherence and continuity of tradition. On study books in the form of multiple-text manuscripts from the fields of law, religion, philosophy, and the rational sciences see ‎Endress, “One-Volume Libraries and the Traditions of Learning in Medieval Arabic Islamic Culture.”‎

Notebooks as Knowledge Repositories

Personal notebooks constitute another type of majmūʿāt,⁴⁴In their introduction to the volume ‎Personal Manuscripts‎, David Durand-Guédy and Jürgen Paul assign notebooks to the (large) group of personal manuscripts with mixed content, attribute to them the following constitutive characteristics: Compiled and used by one person (sometimes several persons), heterogenous content, wholly or largely lacking internal organization, openness of the writing process, aggregation of content, i.e., earlier entries can be added to at a later date. and were, as documented in some cases, referred to as “majmūʿa”⁴⁵See ‎Klemm, “A Library in One Volume.”‎ My following study focuses on note-taking and commentary practices of the compiler: “At the High End of Learning: Note Taking and Commentary Practices of a Nineteenth-Century Ismaili Scholar in India.” or “majāmīʿ” by their owners.⁴⁶See ‎Bauden, “Data Overload and Information Management in the Mamluk Period,” 53‎. Below, I discuss the notebook of a scholar from Ismaili Bohra Tayyibis, an ancient merchant community in the Indian Ocean region.⁴⁷The Bohras are an old and prospering Ismaili Muslim merchant community in Northwest India with a high degree of spatial, social, and cultural mobility in the Indian Ocean between East Africa, Arabia, and Southeast Asia. As their closely connected fellow believers, the Yemeni Tayyibis, the Bohras trace their origin back to the Fatimids (297–567/909–1171).

The owner of the notebook was Sayyidī Muḥammad ʿAlī al-Hamdānī (1833–98) from Surat/Gujarat. In his small community, he was a highly respected scholar and poet and committed agent in religious-political affairs.⁴⁸The notebook ‎MS Ismaili Studies 1662‎ is part of the Hamdani Collection, a library that migrated with its owner from Yemen to Gujarat in India during the late eighteenth century CE. It remained in the possession of the family for seven generations, until the late Professor Abbas Hamdani, Univ. of Wisconsin, donated it to the Institute of Ismaili Studies in 2008 and 2015. The manuscript 1662 contains 105 numbered folios and measures 19 x 14 cm. For further description, examination and contextualization, see ‎Klemm, “A Library in One Volume”‎; ‎Klemm, “At the High End of Learning.”‎ The paper of his majmūʿa bears a British watermark dated to the year 1862. It contains an abundance of heterogeneous material including complete texts, short excerpts, text snippets, lists, tables, numbers, calculations, symbols, and sketches, all created over time, with pages either filled or blank. Gradually filling it with a wide variety of materials, al-Hamdānī used his notebook as a repository, drawing on a broad spectrum of disciplines and fields of knowledge, including Islamic and Ismaili traditions, the Old and New Testaments, Judeo-Christian religious history, the traditions of ancient Greece, Rome, and Iran, as well as astrology and prognostication, and, at times, esoteric sciences and magic. In several tables, cross-cultural calculations and conversions of calendar systems and dates appear, as well as comparisons of terminologies, names of celestial bodies, stellar constellations, units of measurement, currencies, and more. The notebook also contains evidence of his engagement with mathematics (trigonometry and geometry), chronometry, and physics.

Al-Hamdānī used his notebook for both storing and further engaging with his materials. In his majmūʿa, he entered into a conversation with his previously selected and written excerpts and quotes. To interact with these texts, he made use of the margins, as well as the space above the headings and between the lines. Indeed, his commentarial activities form another meaningful layer of annotations in the notebook.

Thanks to the careful preservation of his manuscript⁴⁹Restored in the 1950s by his son Fayḍ Allāh b. Muḥammad al-Hamdānī (Surat). we can virtually watch its owner at work. Al-Hamdānī studied many of his copied texts with great interest in their transmission history and variance. Thus, he sometimes cross-copies variants from other manuscripts or adds quotations from works of the Fatimid and Tayyibi literary tradition, accompanied by references to the author and work. His readings and comments prove that he approached the written tradition with a decidedly critical and historically-oriented approach. Al-Hamdānī’s marginal and interlinear commentaries show that he examined and commented on both his text extracts and the manuscripts he used. In his notes, he checked ownership statements and colophons. Sometimes he mentioned the names of scribes, locations, and dates of copy of a manuscript he had studied. In the only first-person statement contained in the notebook he documents the highly remarkable Yemeni-Indian biography of a valuable old manuscript:

“I found a manuscript of the Kitāb al-Shajara by the Dāʿī Abū Tamām in the library of the daʿwa [in Surat]. The copyist Muḥammad b. ʿAlī b. ʿAbdallāh al-Hindī completed the copy on 12 Muḥarram of the year 509 hijra.”

In the continuation of the paratext, he mentions, by name and titles, the owner of the manuscript in Yemen, the chief Qāḍī in the realm of the Ismaili-Tayyibi Ṣulayḥids in Yemen. He also quotes an official statement (tawqīʿ) in the manuscript stating that he passed the manuscript on to his successor.⁵⁰‎MS Ismaili Studies 1662‎ f. 22 r.; for further historical contextualization see ‎Klemm, “A Library in One Volume”‎ as well as id. ‎“At the High End of Learning.”‎

Hamdānī´s majmūʿa is an authentic and living testimony of his work and his engagement with the materials he copied. His notes and his intellectual interaction with them provide valuable insight into his intellectual interests, working methods, and practices.

Conclusion

Research into majmūʿāt in general and Ismaili in particular is still in its early stages. Large quantities of Arabic multiple-text and composite manuscripts remain undocumented and unexplored to this day.

In every majmūʿa there is at first an invisible relationship between the compiler, his text selection, and the physical structure of the manuscript. Occasionally—though by no means always— one of these parameters becomes apparent, for example, through a note referring to the compiler, a distinctive selection of works, or a conspicuous textual variant. Whenever this occurs, an opportunity arises to reveal some of the secrets of the artefact—such as the idiosyncrasy of its composition, or the agenda of the commissioner or copyist, or the social and cultural function of the majmūʿa within a specific historical environment.

The three case studies presented here sought to demonstrate meaningful examples of these research possibilities.

(1) Once identified, the text types composed in the majmūʿa can, as in the case of the Ismaili manuscripts from Syria, provide insight into the social and religious status of the manuscripts’ owners; the type and function of the book; the context of the manuscript’s use; the community’s networks of literary transmission in the course of its history; and, last not least, as treasure troves of works that might have been otherwise lost.

(2) Textual collections, together with their paratexts, often reveal their scholarly context. Compilers, copyists, and/or scholars, acting in most cases as unrecognized and unappreciated “copyist-redactors” (Gruendler and Khalfallah), shape the fluidity of a text compilation, its gradual evolution, lines of transmission, and regional variants.

(3) Through the careful analysis of a notebook that appears at first glance to be haphazardly filled with excerpts, quotations, and notes taken from a seemingly disparate mixture of various sources, the intellectual preferences of a scholar become apparent. His note-taking sheds light on his methodology, as well as his commentarial and textual practices. Moreover, the corpus of texts to which he had access reveals the bibliographic resources used by a scholar at the time (in this case, among the nineteenth-century Bohra Tayyibi Ismailis in northwest India).

Another significant aspect of majmūʿāt is their role in preserving and transmitting texts that have been handed down rarely or would otherwise have been lost.⁵¹See ‎Löhr, “Off the Record”;‎ ‎Klemm, “Hidden Literary History,” 201–6‎. In short, majmūʿāt offer valuable insights into often personal selections and representations of text collections and their contexts in the spheres of production, transmission, and reception. They are essential sources that can significantly deepen our knowledge of Arabic literary history. This applies not only to the examples mentioned here, which stem from specific genres, milieus, and contexts, but also to all Arab-Islamic manuscript cultures, with their respective literary traditions and religious and cultural contexts.