Angelika Neuwirth. Encyclopaedia of the Qur’an. Editor: Jane Dammen McAuliffe. Volume 2, Brill Academic Publishers, 2003.
Preliminary Reflections About the Redaction and Canonization of the Qurʾan
Methodological dilemmas
Any assessment of qurʾanic form and structure depends on the position chosen by the researcher as to the redaction and the canonization of the qurʾanic corpus (for a recent analysis of western views on the collection of the Qurʾan, see Motzki, Collection). Two apparently irreconcilable positions are currently infelicitously blocking each other in qurʾanic scholarship: on the one hand, there is the historico-critical approach which is oriented to older, more traditional biblical scholarship. It focuses on the development of the Qurʾan and views it as concomitant to that of its transmitter. It assumes the historicity of the basic Islamic traditions about the genesis of the Qurʾan, though sometimes tends to cling too closely to the reports contained in the biography of the Prophet (sira) and thus unduly re-historicizes the Qurʾan. On the other hand, there is the counter-position of John Wansbrough’s hyper-skeptical revisionist approach informed by a more modern trend in biblical scholarship, namely Formgeschichte, as well as semiological approaches that reject the traditional narrative altogether. This approach projects the role hitherto ascribed to the Prophet and the first caliphs in the redaction process onto an anonymous committee assumed to have assembled a century or more later. In A. Rippin’s words: “Canonization and stabilization of the text of the Qurʾan goes hand in hand with the formation of the community. A final fixed text of the scripture was not required, nor was it totally feasible, before political power was firmly controlled; thus the end of the second/eighth century becomes a likely historical moment for the gathering together of oral tradition and liturgical elements leading to the emergence of the fixed canon of scripture” (Literary analysis, 161). This approach, which not only dismisses the sira but also rigorously de-historicizes the Qurʾan, and, by confining itself to the macrostructure of the canonized final version, disregards the distinctive internal literary structures of the Qurʾan (q.v.), has been criticized for its mechanistic argument. Thus, J. van Ess comments: “Generally speaking I feel that the author [i.e. J. Wansbrough] has been overwhelmed by the parallel case of early Christianity. Islam comes into being at a time and in surroundings where religion is understood as religion of the Book. This understanding had been prepared by the developments in Judaism and Christianity, as well as in Manichaeism. Canonization was no longer something novel. It was expected to happen. This, in my view, suffices as a justification of the process in Islam taking place so rapidly” (Review of J. Wansbrough, 353). This article argues for a third way: a shift in focus from a “canon from above” to a “canon from below,” and a reading of the Qurʾan which studies the sura (q.v.) as a communication process and thus respects this redactionally-warranted unit as a genuine literary text.
Canonization and the problem of the “sura” as a unit
Several recent studies on the Qurʾan have focused anew on the problem of its canonization, making this a central issue in qurʾanic research. What these studies have called into question is the traditional account of the redaction and publication of a unified and authorized final version of the Qurʾan through which the text came to occupy the status of a scripture bearing an intrinsic logic of its own. By focusing on this final phase and ranking it as the crucial event in qurʾanic genesis, an epistemological course has been set: The literary image of the Qurʾan as reflecting a text still in progress and thus displaying a unique micro-structural diversity due to its evolution out of an extended process of a liturgical communication, becomes blurred, being eclipsed by its macro-structural weight and the social importance of the henceforth normative corpus and its ideological implications for the construction of the community’s identity.
According to the dominant Islamic tradition, the Qurʾan owes its authoritative final version to the redaction carried out by a committee summoned by the third caliph, ʿUthman b. ʿAffan (r. 23-35/644-56). The creation of this codex does, it is admitted, impose on the suras a sequence that, until then, had not been fixed. In many cases it also incorporates passages that had been transmitted in an isolated manner into completely new contexts. The committee clings faithfully, however, to the text material whose authenticity is warranted by reliable oral and/or written tradition, taking into consideration the entire corpus of the qurʾanic revelations available at the time. The performance of the committee is, therefore, traditionally identified as an act of collection (jamʿ), one accomplished in perfect accordance with the concept of its commissioner, ʿUthman, who is reported to have imposed on the redactors—apart from observing some linguistic cautions—no further task than that of gathering all the extant parts of the Qurʾan. The traditional account of the collection of the Qurʾan accords with the evidence offered by the text itself, since the new codex, which does not claim any chronological or theological rationale for the sequence of the single units (suras)—which appear to be arranged according to merely technical external criteria—does display inextinguishable traces of its compilation as a collection. On the surface, it presents itself as a corpus of unconnected texts of considerable structural diversity, not allowing for an immediate classification under one particular genre.
The traditional reports identify political constraints as the explanation of, and justification for, the admitted fact that the collection was carried out somewhat hastily and thus had to proceed in a rather mechanical fashion. Although other redactions had to be suppressed, the sequences of suras in two of them (the codices [maṣaḥif] of Ibn Masʿud [d. 32/653] and Ubayy b. Kaʿb [d. ca. 19/640 to 35/656]) are known to us. Both seem to have considered suras 1, 113 and 114 to be not part of the corpus, but rather prayers to be uttered concomitant with the recitation of the Qurʾan (q.v.). The official redaction and publication of the standard text neither completely extinguished the memory of extant variants, later known as qiraʾat shadhdha, nor precluded the emergence of further variants. Indeed, a number of reading traditions of the entire Qurʾan (qiraʾat mutawatira), which, in many instances, diverge—although not substantially—from each other have come down to us. Seven of these (the so-called “seven readings,” al-qiraʾat al-sabʿ) even received canonical status through Ibn Mujahid’s (d. 324/936) scrutinizing selection of admissible Qurʾanic text forms. Although these have since enjoyed an equal status in the scholarly and the cultic tradition (ʿilm al-qiraʾa, ʿilm al-tajwid) only two have survived and are still in use in modern times, namely the reading of Ḥafṣ ʿan ʿAṣim (current in the Islamic east) and that of Warsh ʿan Nafiʿ (current in the western Islamic world). Since modern audio media have further enhanced the status of the former, contemporary qurʾanic scholarship usually refers only to the Ḥafṣ text.
Yet, with the ʿUthmanic consonantal fixation of the text, a decisive course had been set with regards to its structure, which gave rise to a problematic development: namely, the joint codification of loosely composed passages and often unframed, conceptually isolated communications—so characteristic of the Medinan “long suras” (ṭiwal al-suwar)—together with the complex poly-thematic structures and mnemonic, technically sophisticated pieces that comprise the short and middle-sized suras resulted in a most heterogeneous ensemble, a fact that did not remain without consequences. Once these elements melded to form a comprehensive and closed corpus, a codex (muṣḥaf, q.v.), they became neutralized as to their liturgical Sitz-im-Leben and their communicational context in the emergence of the community. Previously defined text-units distinguishable through reliable devices such as introductory formulas and markers of closure were, it is true, retained by the redaction process and labeled “sura.” They lost much of their significance, however, for, in the same codex there were now other units also labelled as “suras,” but whose constituent passages had not come to form a coherent literary structure and thus invalidated the structural claim raised by those suras that were neatly composed. The neatly-composed suras eventually ceased to be considered integral literary units conveying messages of their own and mirroring individual stages of a process of communication. On the contrary, once all parts had become equal in rank, arbitrarily selected texts could be extracted from their sura context and used to explain other arbitrarily selected texts. Passages thus became virtually de-contextualized, stripped of the tension that had characterized them within their original units. Genuine text-units lost their literary integrity and could be mistaken for mere repetitions of each other.
Hence, with its final official canonization, the Qurʾan had become de-historicized. Not the process of its successive emergence as mirrored in the text, but the timeless, eternal quality of its message had become its brand. This made the understanding of the Qurʾan all the more dependent on the sira, a corpus that, although transmitted and codified separately, had been grafted on the Qurʾan by its readers and listeners from early times. Prophetic tradition, in its development of haggadic meta-history, thus took the place that intra-qurʾanic history should legitimately have occupied, i.e. the history, however sparse the chronological evidence, of a liturgical and social communication process, that took on a distinctly textual shape in the Qurʾan and is reflected in the structure of the suras. Further literary investigation into the micro-structure of the Qurʾan, which might reveal the still-traceable traits of that history, remains an urgent desideratum.
As M. Mir (The sura) has stressed, Muslim exegetes have only recently rediscovered the most prominent micro-structure of the Qurʾan, namely the sura as a unit containing meaning, a concept long neglected in Muslim circles and generally dismissed as irrelevant in western scholarship. Exceptions to this dismissal have more recently appeared (cf. M.A.S. Abdel Haleem, Context; A. Neuwirth, Zur Struktur; id., Symmetrie; id., Koran; id., Images; id., Erste Qibla; id., From the sacred mosque; id., Qurʾanic literary structure; A.H.M. Zahniser, Word of God; id., Sura as guidance; M. Sells, Sound, spirit and gender; id., Sound and meaning; A.H. Johns, Qurʾanic presentation; and S.M. Stern, Muhammad and Joseph).
Reflections of a canonical process
The older suras in particular seem to mirror a development which in its essential traits reflects a canonization from below, as characterized by Aleida and Jan Assmann (Kanon und Zensur). These two scholars distinguish between a canon described as power-oriented and one that relies on a particular source of meaning, not least on the charisma of the transmitter of a message. According to the Assmanns’ theory, “whenever the message is preserved to survive beyond the situation in which the original group was directly interacting, it will usually undergo a profound change in structure. The message gains a new appearance through scripturalization and moreover through institutionalization.” In the case of the Qurʾan, then, a canon from below certainly precedes the canon from above. The latter comes about only with the authoritative final redaction, which became necessary to counteract the pressure of a reactionary tendency towards provincialization and fragmentization. The canon from below has thereby changed into a canon from above, a development comparable to that in early Christianity when the official Church contracted a pact with political power.
To discern the textual signs of a canon developing from below, we may draw on the new approaches developed in recent biblical studies, principally those of the American scholar Brevard S. Childs, who has proposed an understanding of the genesis of a canon as a process of growth. Canon in this context no longer covers the officially codified final form of a text, but rather signifies the “consciousness of a binding covenantal character deeply rooted in the texts” (C. Dohmen, Biblischer Kanon, 25) that is affirmed by the continuous references of later emerging text-units to a text nucleus and by the recurrent instances of intertextuality mirrored in the text-units developing around the nucleus. Even at the point where the genesis of a text conceived as a canonical process has come to a close with the end of the text’s growth, its final form will not be a harmonious presentation but will leave the roughness caused by the organic growth un-leveled. The final shape only re-locates interpretation, which, until then, had taken place in productive additions or changes within the text, and which henceforth takes place through exegesis and interpretation separate from the text.
Methodological conclusions
The following presentation of qurʾanic form and structure is based on these observations. At the same time it represents an attempt to comply with a provocative demand proffered by A. Rippin (Qurʾan as literature) that the Qurʾan should be studied by (a) situating it in its literary tradition and (b) situating it as the focal point of a readers’ response study. But, diverging from Rippin’s proposal, we will not go so far as to replace an immediately traceable intra-qurʾanic context with a speculative biblical or post-biblical one in order to provide the appropriate literary tradition. Nor will we embark on reconstructing a post-qurʾanic reader-response from the exegetical literature. Rather, what we shall analyze—on the basis of individual suras—is the qurʾanic communication process as taking place between speaker and listeners. The reader-response is thus replaced by a listener-response, the concept of the “implied reader” is modified into that of the “implied listener.” Situating the Qurʾan in its literary tradition will be realized through the investigation of its peculiar referentiality, not stopping short at the notice of particular instances of a biblical background, but proceeding to examine the position of the sura as a stage in an extended canonical process.
This article will discuss the language and style of the Qurʾan in general and on this basis the individual literary genres assembled in the Qurʾan will be surveyed in terms of form and content. To present such an inventory of the building blocks or “enjeux” (Ger. “Gesätze”) of the suras is a useful propaedeutic step towards the literary assessment of the Qurʾan, although hardly any of the enjeux themselves appears as a self-sufficient communication, i.e. as a complete sura. Rather, they are integrated in complex ensembles and thus, to be adequately understood, must be viewed in their wider context. The discussion will therefore survey the contextuality, i.e. the diverse combinations of individual enjeux displayed in individual suras.
Now, the Qurʾan has never been conceptualized or intended as a primarily literary corpus whose purpose was to convey information to, or serve the re-education of, its readers. Rather, it has manifested itself—until its final publication—as a continuous hermeneutical process reflecting, and simultaneously conditioning, the attitudes of its listeners towards the message. The literary ensembles—suras—thus constitute essentially liturgical units that have developed not so much through the textual growth of the corpus as through a liturgical or communicational process that transpired within the emerging Islamic community. Their “history” can therefore be plumbed out only by closely considering the process of conveying the message, i.e. by surveying the subsequent changes in communication techniques and the hints at the performative framework, in terms of time, space and protagonists involved, as mirrored in the self-referential passages of the Qurʾan. Only such a synopsis of the literary and the communicational, i.e. liturgical development, will enable us to pursue the canonical process which finally produced the corpus as we have it today.
Linguistic, Stylistic, and Literary Character of the Qurʾan
Diversity of views
An early debate about the question of qurʾanic language—Meccan vernacular (Vollers, Volkssprache) or poetic koine (ʿarabiyya, Nöldeke, Neue Beiträge; Geyer, Zur Strophik) was decided in favor of the latter, though occasional linguistic interferences reflecting the Ḥijazi vernacular are still discernible beneath the amendments later supplied by the classical philologists. Still, the style and language of the Qurʾan have often been dismissed as defective, with verdicts ranging from Th. Nöldeke’s “Sündenregister” (Neue Beiträge, 5-23) imposing upon the Qurʾan grammatical rules that were developed at a later date, to L. Kopf ’s (Religious influences, 48) denigration of the Prophet’s stylistic talents, to R. Blachère’s (Histoire, ii, 187-241, esp. 204-36) reaffirmation of Nöldeke’s influential critique. Although recognizing the division of the text into three sections from the Meccan period and one from the Medinan period, based upon predominantly stylistic considerations, and thus admitting a poetic character for the earlier suras as against a more prosaic one for the later suras (Nöldeke, GQ, esp. i, 66-75; 143-4), Western qurʾanic scholarship has for a long time failed to draw due methodological conclusions and to analyze the qurʾanic texts in an accordingly complex manner. An attempt to broadly survey the literary qualities of the Meccan part of the corpus was undertaken by Neuwirth in several studies (see bibliography).
Qurʾanic composition fared even worse. Since the sensational hypothesis presented by D. Müller (Die Propheten) claiming a strophic composition for the suras was dismissed without further scrutiny by subsequent scholarship (Nöldeke, GQ) the possibility that “a firm literary hand was in full control” of the composition and structure of individual suras has been virtually excluded. Disclaimers (adduced by Rippin, Review of Neuwirth) range from Goldziher’s (Introduction, 28, n. 37) statement, “Judgments of the Qurʾan’s literary value may vary, but there is one thing even prejudice cannot deny. The people entrusted… with the redaction of the unordered parts of the book occasionally went about their work in a very clumsy fashion,” to Wansbrough’s (QS, 47) “… ellipsis and repetition [in the Qurʾan] are such as to suggest not the carefully executed project of one or of many men, but rather the product of an organic development from originally independent traditions during a long period of transmission.” Although Nöldeke’s work still built on the reality of the suras (admitting, of course, subsequent modifications), the hypothesis of an artistically valuable composition—be it of the qurʾanic corpus or of the single suras—has since been negated, and existing literary forms have been considered to be the result of a haphazard compilation.
The problem of periodization
As against the view just mentioned, through micro-structural analysis, structures do become clearly discernible beneath the surface. These structures mirror a historical development. Indeed, observations about style and structure complemented by thematic considerations have induced Western scholars (Weil, Historischkritische Einleitung; Nöldeke, GQ [repeated by Blachère, Le Coran; id., Histoire]) to declare a division of the text into three sub-sections from the Meccan period and one from the Medinan period, thus further developing the distinction between Meccan and Medinan text-units already made by Muslim traditional scholarship. Although the assumption (also held by Bell, Qurʾan; id., Introduction) of “a historical progression at work between the diverse sections, i.e. that stylistic and thematic considerations can be translated into historical conclusions” has been contested (Rippin, Review of Neuwirth), it should nonetheless be noted that stylistic developments in any literature, once attained, are not deemed reversible. Since Nöldeke’s division still proves useful as a working hypothesis, it appears worthwhile to further scrutinize his observations. As a first step in that direction, Neuwirth (Studien) has tried to establish a critical basis for determining verse structures by scrutinizing the verse divisions of the “standard Ḥafṣ text” through consultation with other traditional schemes. The crucial procedures demanded in order to reach a valid periodization are, however, more complex, and they have to proceed from a thorough investigation of qurʾanic rhyme to that of verse and then to that of paragraph structure in relation to the diverse semantic units.
Rhymes and verse structures as a criterion of relative chronology
The poetical structure of the Qurʾan is marked by rhyme endings of the verses. A description of these rhymes in toto is a necessary pre-requisite for the analysis of the composition of a sura, since only a synopsis of all the rhymes figuring in the Qurʾan will allow us to isolate sequences of rhymes and to examine their relation to semantically coherent groups of verses. Such a classification has been undertaken for the Meccan parts of the Qurʾan by Neuwirth (Studien). There, a significant difference was noted between those suras classified as early Meccan (whose endings comprise some eighty types of rhyme), as middle Meccan (seventeen types of rhyme endings) and as late Meccan (five types of rhyme endings). The diversity of rhymes is, of course, related to the style at large: The suras commonly considered the oldest, i.e. those that display sajʿ, rhymed prose in the strict sense—short units rhyming in frequently changing sound patterns reiterating the last consonant and based on a common rhythm—are made up of monopartite verses containing one colon each (see for the colometric structure, Neuwirth, Zur Struktur; id., Studien), e.g. Q 70:8-9, yawma takunu l-samaʾu ka-l-muhl/wa-takunu l-jibalu ka-l-ʿihn. Longer compositions, whose style is too complex to be pressed into short sajʿ phrases, usually display a bipartite (two cola) structure, e.g. Q 54:42, kadhdhabu bi-ayatina kulliha fa-akhadhnahum akhdha ʿazizin muqtadir, or even pluripartite (more than two cola) verse, e.g. Q 37:102, fa-lamma balagha maʿahu l-saʿya qala ya bunayya inni ara fi l-manami anni adhbaḥuka fa-nẓur madha tara qala ya abati fʿal ma tuʾmaru sa-tajiduni in shaʾa llahu mina l-ṣabirin. The relative length of the verses should not be dismissed as simply conditioned by a more or less complex content. Rather, the transition from sajʿ speech to a more ordinarily flowing, though still poetically tinted, articulation attests to the transformation of an adherence to the standard pre-Islamic (jahili) tradition into a novel literary paradigm that may be considered as a genuine qurʾanic development marking a new stage in the history of the Arabic literary language.
Proportions between verse groups as a criterion
Bell (Qurʾan, 71) claimed that “many suras of the Qurʾan fall into short sections or paragraphs. These are not of fixed length, however, nor do they seem to follow any pattern of length. Their length is determined not by any consideration of form but by the subject or incident treated in each.” This claim is, however, no longer tenable. Bell’s perception of the Qurʾan—not unlike that held by Nöldeke and many later scholars—relies heavily on the imagination of a written text and completely neglects the oral character of the majority of the Meccan compositions. The principally liturgical function of the qurʾanic texts, however, presupposes texts that are easily memorized and which, as long as writing is not involved, are dependent on mnemonic-technical devices. An analysis of the structure of the verses of the Qurʾan in terms of their division into segments and the relationship between the grammatical structure of each segment and the thematic contents carried out by A. Neuwirth (Studien)has resulted in a typology of sura structures. Most Meccan suras display fixed sequences of formally and thematically defined verse groups distinctly separated by a change of rhyme or other clearly discernible, sometimes formulaic markers of caesurae. A group of two verses may be adduced at Q94:7-8, fa-idha faraghta fa-nṣab/wa-ila rabbika fa-rghab (new rhyme, strictly parallel structure); a group of three verses is Q 90:8-10, a-lam najʿal lahu ʿaynayn/wa-lisanan wa-shafatayn/wahadaynahu l-najdayn (new rhyme, identical subject); a group of four verses is Q 90:1-4, la uqsimu bi-hadha l-balad/wa-anta ḥillun bihadha l-balad/wa-walidin wa-ma walad/laqad khalaqna l-insana fi kabad (ensuing change of rhyme, oath cluster with assertion); a group of five verses is Q 99:1-5, idha zulzilati l-arḍu zilzalaha/wa-akhrajati l-arḍu athqalaha/wa-qala l-insanu ma laha/yawmaʾidhin tuḥaddithu akhbaraha/bi-anna rabbaka awḥa laha (ensuing change of rhyme, apocalyptical scenery succeeded by an eschatological process); a group of six verses is Q 75:1-6, la uqsimu bi-yawmi l-qiyama/wa-la uqsimu bi-l-nafsi l-lawwama/a-yaḥsabu l-insanu allan najmaʿa ʿiẓamah/bala qadirina ʿala an nusawwiya bananah/bal yuridu l-insanu li-yafjura ama-mah/yasʾalu ayyana yawmu l-qiyama (group made up by 2 + 2 + 2 verses, held together by concatenation; ensuing change of rhyme, the group is followed by two further groups of six verses: 2 + 4, 2 + 2 + 2); a group of seven verses is Q 56:81-7 (polemics against adversaries of the Qurʾan), followed by another group of seven verses (Q 56:88-94) presenting the eschatological retribution; a group of eight verses is Q 93:1-8, wa-l-ḍuḥa/wa-l-layli idha saja/ma waddaʿaka rabbuka wa-ma qala/wa-la-l-akhiratu khayrun laka mina l-ula/wa-la-sawfa yuʿṭika rabbuka fa-tarḍa… (ensuing change of rhyme, oath cluster with three assertions); groups of nine verses are Q 73:1-9, 10-18; for groups of ten verses and more cf. Neuwirth, Studien, 186 f.
These distinct verse groups often form part of clear-cut patterns of proportions. Thus, Q 75 is built on the following balanced verse groups: 6 + 6 + 6 + 6 + 5 + 5 + 5; Q 70 is made up of 6 + 7 + 7 + 7 + 7 + 9; Q 79 entails two groups of nine verses, its proportions being strikingly balanced: 5 + 9 / 6 + 6 + 6 / 9 + 5. Q 51 is made up of groups of 9 + 14 + 14 + 9 + 7 + 7 verses. Similar cases are found in many of those early Meccan suras that exceed some ten verses, proportion being obviously a mnemonic device required in a situation where memorizing without written support was demanded from the listeners (see below for a further discussion).
The clausula phrase
Any similarity to sajʿ is given up when verses exceed the bipartite structures. In these cases, the rhyming end of the verses follows the stereotypical -un, -in-pattern that would hardly suffice to fulfill the listeners’ anticipation of a resounding end to the verse. A new mnemonic-technical device is utilized, solving the problem. This device is the rhymed phrase, a syntactically stereotyped colon which is distinguished from its context insomuch as it does not partake in the main strain of the discourse, but presents a kind of moral comment on it, as “… give us full measure and be charitable with us. Truly God will repay the charitable” (…fa-awfilana l-kayla wa-taṣaddaq ʿalayna, inna llaha yajzi l-mutaṣaddiqin, Q 12:88), or else refers to divine omnipotence and providence, as “… that we might show him our signs. Truly he is the hearer, the seer” (… li-nuriyahu min ayatina, innahu huwa l-samiʿu l-baṣir, Q 17:1). An elaborate classification of the rhymed phrases has been provided by Neuwirth (Zur Struktur) on the basis of sura 12, a text particularly rich in clausulae that, hardly by mere coincidence, display a large number of divine predicates (al-asmaʾ al-ḥusna). Although it is true that not all multipartite verses bear such formulaic endings, and occasionally do contain ordinary short sentences in the position of the last colon, still, clausula verses may be considered to be a characteristic developed in the late Meccan period, and present in later verses. The presence of clausulae should not be considered as a purely ornamental phenomenon due to the merely stylistic moods of the speaker and thus devoid of significance for periodization. On the contrary, their appearance marks a new and irreversible development: The clausula serves to turn the often-narrative discourse of the extended suras into paraenetical appeals, thus immediately supporting their theological message. They therefore betray a novel narrative pact between the speaker and his audience, the consciousness that there is a basic consensus on human moral behavior as well as on the image of God as a powerful agent in human interaction, a consciousness that has of course been reached only after an extended process of the community’s education (Neuwirth, Referentiality; id., Qurʾan, crisis and memory).
Orality, scripturality, and the canonical process
In spite of the etymology of its earliest self-designation (qurʾan < Syriac qeryana, i.e. recital, pericope to be recited in services), far too often the Qurʾan is implicitly considered to be a literary work, imagined as “authored by Muḥammad,” as becomes apparent from all the critiques which blame the text for not fulfilling particular literary standards. Since the quest for an “Urtext” has long been prevalent in historical-critical studies, qurʾanic speech has usually been investigated according to the criteria of written compositions with no relation to oral performance. This view has been met with criticism in more recent scholarship, which has demanded that the quest for “original meaning” be replaced by a consideration of the Qurʾan’s socio-cultural context as necessary for its interpretation (Martin, Understanding the Qurʾan). Denny (Exegesis and recitation, 91) criticized the neglect of the “ritual-recitational dimensions of the Qurʾan” and Graham (Beyond, 80) stressed “the abiding and intrinsic orality of the Qurʾan as a scriptural book of revelation and authority.” “Oral composition” such as has been claimed for ancient Arabic poetry by Zwettler (Oral tradition) and Monroe (Oral composition) on the basis of the thesis presented by M. Parry in 1930-2 (The making of Homeric verse) and followed by Lord (in The singer of tales), although not immediately applicable in the case of the Qurʾan, is still in need of debate. According to Parry and Lord, “oral poetry” is characterized by its composition during performance, a procedure which is supported by a thesaurus of formulaic phrases. In some cases this may apply to the Qurʾan (see below), but can hardly be proved for the bulk of its corpus. Many early suras (e.g. Q 73 and 74) that surely were composed without the support of writing attest to their origin in nocturnal vigils (q.v.) rather than public performances. Later suras (from the so-called Raḥman period onward, see Watt-Bell, Introduction; Nöldeke, GQ), composed of multipartite verses with little poetic shaping and thus devoid of effective mnemonic-technical devices, strongly suggest an immediate fixation in writing if they were not initially written compositions.
To investigate the full scope of this development one has, however, to go beyond the mere technical aspects. It is noteworthy that, although the distinction between two decisive periods for the genesis of the Qurʾan (a qurʾan phase and a kitab phase, the latter implying the use of writing as a mnemonic-technical device to preserve the text) has been accepted in historic-critical qurʾanic scholarship as a whole (Watt-Bell, Introduction; Nagel, Vom Koran zur Schrift; Robinson, Structure), the double self-representation of the qurʾanic text has never been explored under the perspective of its implications for the canonical process. One has to keep in mind, however, that the terms qurʾan and kitab denote very different concepts. The first points to a communal event in progress involving a multiplicity of dramatis personae—a speaker reciting a message received from an “absent” commissioner that he is to communicate to a plurality of listeners. It thus stresses a horizontal human interaction. This dynamic, thanks to the striking phenomenon of qurʾanic self-referentiality, is mirrored clearly in the early suras themselves, which have preserved lively scenarios of the reception of the qurʾanic revelation. The second concept focuses on the hierarchical quality of a transcendent message presupposing a vertical relationship between an “author” (or his spokesperson) and the “reader” (or the worshipper). Thus the notion of a kitab in itself clearly implies a strong claim of canonicity. Indeed, it was realized as such by the early community who first observed kitab as a transcendent scripture, on the one hand manifested in the texts held sacred by the adherents of the older religions (i.e. tawrat, injil), and, on the other hand, being communicated to them in subsequent messages (ḥadith, Q 51:24; 20:9; nabaʾ, Q 15:51; 26:69; 38:21) to form narrative pericopes within the more complex liturgical recitals (qurʾan). They only later realized kitab to be the entelechy of their own growing corpus of divine communications. What was qurʾan in the beginning, then, developed into kitab in the end; so a similar claim of canonicity cannot, in principle, be excluded for the term qurʾan either, which in later usage comes very close to that of kitab. In turn, the Muslim kitab preserves much of its “qurʾan-ness” since throughout the process of revelation the presence of the listeners is maintained, the believers among whom, i.e. the community even step into the text, not only as protagonists in new scenarios of salvation (q.v.) history but as conscious voices in an ongoing debate. Thus the entirely vertical relationship between the sender and the recipients, which prevails at the close of the qurʾanic development, i.e. after the completion of the corpus, is not really relevant to the preceding stages. The direct or indirect reference to the notion of kitab thus may serve as a reliable guide when tracing the ongoing process of canonization in the qurʾanic development.
The “enjeux” or building blocks of the sura (“Gesätze,” structurally definable verse groups)
Since the appearance of A. Welch’s article (Ḳurʾan) in 1981, further attempts at a classification of the “enjeux” have been put forward. Contrary to Welch—who is skeptical of the intra-Meccan periodization and thus reluctant to discuss the forms according to their successive emergence —, Neuwirth (Studien), in an extensive study of the qurʾanic literary forms of Meccan suras, does consider this periodization—i.e. the approximately chronological sequence of suras (Entwicklungsreihen) presented by Nöldeke and accepted by Schwally and Blachère—as still valid and useful as a working hypothesis. Unlike Welch’s article, which praises Bell’s atomization of the sura as an important step forward, Neuwirth’s study insists on the significance of the sura as a literary unit although conceding that many Meccan suras have undergone developments (Fortschreibungen) during their liturgical use, and that Medinan suras constitute a case of their own. It is, however, assumed that the Meccan sura in its final composition is an intended unit that reflects a natural growth, not a haphazard combination of diverse elements. The acceptance of the sura as an intended unit following verifiable compositional patterns that are important for the understanding of the ensemble of “enjeux” enables the perception of structural developments, which, again, make possible a rough periodization of the suras as units as well as of their “enjeux.”
The following list comprises only the main types of “enjeux,” focusing on the early manifestations of the particular elements. On the whole, Meccan and Medinan suras consist of the same building blocks; a few elements that appear in Medinan suras exclusively will be discussed at the end of the list (for a more exhaustive discussion, see Neuwirth, Studien, 187 f. and 238 f.).
Oaths and oath clusters (introductory and intra-textual sections)
From among the forty-three suras ascribed by Nöldeke to the first Meccan period, seventeen are introduced by oaths. In eight instances, oaths appear within suras. Two types of oath formulas can be distinguished: a group introduced by waw alqasam (fifteen times in introductory sections, three times within suras) and another introduced by la uqsimu bi- (twice in introductory sections, five times within suras). The particular importance of the introductory sections of the qurʾanic suras for the entire composition has not been discussed on any systematic level. Still, observations concerning the beginning of the suras have led to quite far-reaching hypotheses about the special brand of Muḥammad’s prophethood: i.e. the early suras betray a close relationship to the utterances of the pre-Islamic soothsayers (q.v.; kuhhan, sing. kahin), and may even be considered the most reliable evidence for kuhhan speech itself.
Now, the specimens of kuhhan sayings that have been transmitted in early Islamic literature are not always assuredly genuine, nor have they been studied regarding their literary form. Theories about their relation to qurʾanic speech, therefore, still lack a methodological foundation. Neuwirth (Der Horizont; id., Der historische Muhammad) has presented some preliminary observations about the relationship between kahin expression and the early suras. Whereas oaths still bearing traces of legally binding commitments are found sporadically in the Qurʾan—mostly in the context of solemn pronouncements invoking God as witness for the truth of a statement—the oaths appearing in the early Meccan suras are completely devoid of any legal connotation, but form clusters that serve exclusively as a literary device. This is affirmed by several formal characteristics, the most striking of which is the multiplicity of the objects invoked. Unlike in the case of legally binding oaths, these are not of a superior order (God, the life of the speaker, etc.) but, rather, are objects chosen from the empirical realm. A second characteristic is the limitation of the oaths to the standard formula wa- X or la uqsimu bi- X followed by an assertion, a “statement,” usually worded inna Y la- Z, not implying any allusion to a legally binding commitment on the part of the speaker. The oath clusters may be classified as follows:
a) Oath clusters of the type wa-l-faʿilat:Q37:1-3; 51:1-4; 77:1-4; 79:1-5, 6-14; 100:1-5. These oaths, which do not explicitly name the objects to which they refer, but only allude to them by qualifying them as being moved in different successive motions, have been considered the most intricate by both Muslim exegetes and Western scholars. Displaying a metaphorical language distinctly different from that of the rest of the corpus, they have come to be known as particularly enigmatic, not so much because of the few undeniable lexical and grammatical ambiguities, but because of a more fundamental difficulty: their pronouncedly profane imagery (horses on their way to a raid [ghazwa], clouds heavy with rain) which seems inconsistent with the overall purport of the suras as documents of religious discourse.
b) Oath clusters alluding to sacred localities and the abundance of creation: Q52:1-6; 90:1-3; 95:1-3. The localities mentioned refer to particular theophanies, thus functioning as symbols of divine instruction. The one locality constantly mentioned is Mecca (q.v.); it appears once alone (Q90) and twice (Q 52 and 95) in combination with Mount Sinai (q.v.) as the second site. In all three oath clusters an immediately recognizable semantic coherence between the oath formulae and the following text passage is missing, thus delaying the anticipation of a solution to the enigma posed which is disclosed only at the end of the sura: theophanies, i.e. divine communications, necessitate an account be rendered on the day of judgment.
c) Oath clusters relating to cosmic phenomena and liturgically significant time periods of the day and the night are found at the beginning of a number of suras: Q85:1-3; 86:1-3; 89:1-4; 91:1-7; 92:1-3; 93:1-2; they appear within suras in: Q51:7-9; 86:11-12.
What justifies the classification of suras with introductory oath clusters as a type of their own is not so much the observation of such obvious traits as common topics or patterns of composition as it is the immanent dynamics dominating these suras. With regards to form, this particular quality is due to the accumulation of parallel phrases in the introductory section creating a rhythm of its own. Structurally speaking, it is based on the anticipation of a solution to the enigma that is aroused in the listeners’ minds by the amassed metaphorical elements, an enigma that is not immediately comprehensible or even plausible to them. It is this dynamization of the entire sura created by the introductory oath clusters that is the main characteristic of this text group.
In the case of (a), the faʿilat-clusters, the anticipation of an explication of the ideas presented in the cluster in an oblique metaphorical way through their empirically known prototypes is fulfilled only at the end of the sura (or the first main part). The metaphorically projected catastrophe is none other than the eschatological dissolution of creation. In the case of oaths referring to (b), symbols of creation and instruction, the anticipation of the ideas of judgment and account is suspended in a similar way and fulfilled only at the end of the sura, or again, at the end of the first main part. Suras introduced by oath clusters referring to (c), cosmic phenomena and liturgically significant day and night phases, respectively, betray a somewhat different structure of anticipation. They are characterized, it is true, by a hymnical (or polemical) tonus rectus that remains audible throughout the entire sura. However, in both types it is the ever-stressed opposition between created beings in terms of moral behavior, structurally prefigured through the contrast of light (q.v.) and darkness (q.v.), that arouses the anticipation of a final affirmation of unity personified in the creator, a unity that alone gives meaning to the oppositions extant in the realm of created beings. Indeed, the concluding sections, in speaking of the believers’ nearness to the divine speaker, lead back to the experience of divine unity felt in liturgy and Qurʾan recitation to which the images in the introductory section (liturgical time phases) allude.
In the later suras, the anticipation aroused by the oaths is fulfilled immediately, without suspense, in the ensuing statement (Q 36:2, object: al-qurʾan al-ḥakim; Q 38:1, al-qurʾan dhi l-dhikr; Q 43:2, al-kitab al-mubin; Q 44:2, al-kitab al-mubin; Q 50:1, al-qurʾan al-majid; Q 68:1, al-qalam wa-ma yaṣṭuruna), all of which are followed by assertions related to revelation. The oath clusters have thus developed from functional units into merely ornamental devices. In these later and more extended suras, where the primary function of the oaths, i.e. arousing tension toward the explication of the initial enigma, has become faint, the attention of the listener can thus concentrate on particular—structurally important—images bearing symbolic value. It is not by mere coincidence that the standard incipit characteristic of so many later suras develops from one of the types of early oath clusters: In the end, the image of the book (al-kitab)—which had constituted the object of most of the early Meccan intra-textual oaths (Q 56:75 f.; 81:15 f.; 84:16 f.; 86:11 f.) but appeared less frequently in the introductory part (Q 52:2-3)—alone remains in use, the most abstract of all the different symbols used, essentially no more than a mere sign. The book is thus the only relic that survives from among a complex ensemble of manifold accessories of revelation, originally comprising cosmic, vegetative, topo-graphic, cultic and social elements. The book as the symbol of revelation par excellence successively acquires the dignity that it has preserved until the present day to represent the noblest emblem of Islamic religion.
Eschatological passages (introductory and intra-textual sections)
Clusters of idha-phrases
Five suras (Q 56:1-6; 81:1-13; 82:1-4; 84:1-5; 99:1-3) start with idha-phrase-clusters, most of which have a distinct internal structure: Q 81:1-13: six pairs of verses; Q 82:1-4: two pairs; Q 56:1-6: two groups of three verses. Idha-clusters are also encountered within suras, e.g. Q 56:83 f.; 75:26 f.; 79:34-36; 100:9-11. They are typologically related to the oath clusters as they build up a pronouncedly rhythmical beginning to the sura or part of the sura; here, however, the tension is resolved immediately in the closely following apodosis. In their particularly concise and poetically tinted syntactical structure (idha + noun + verb instead of the standard prose sequence of idha + verb + noun), these clusters (ranging from two to twelve verses) present apocalyptic scenes depicting the dissolution of the created cosmos on the last day. It is noteworthy that the highly rhythmical idha-phrases never exceed mono-partite verse structures and thus contribute to the pronounced sajʿ character of the early suras. In some cases the idha-phrases are not confined to natural and cosmic phenomena but proceed to depict the preparations for the final judgment (the blowing of trumpet, positioning of the throne, opening of the account books etc.). Yawma may also serve the function of the conjunction idha: Q 52:9-10; 79:6-7.
Eschatological processes
In terms of grammar, the idha-phrases constituting the protasis of a conditional period are followed by equally stereotyped apodoses referring to the foregoing with the adverb yawmaʾidhin (e.g. Q 69:15; 79:8; 99:4, 6). These “eschatological processes” depict the behavior of people in the apocalyptic setting and their separation into the groups of the blessed and the condemned (Q 56:7).
Diptycha: Descriptions of the hereafter Continuing (in grammatical terms) the apodosis of the eschatological period, these descriptions of the hereafter are strictly divided into two counterparts. Introduced by fa-amma… wa-amma (Q 101:6-7, 8-9) or wujuhun… wujuhun (Q 80:38-9, 40-2), they juxtapose the situation of the believers in the paradisiacal garden (q.v.; janna) with that of the disbelievers (kuffar) or evildoers (fasiqun and the like) in the tribulations suffered in the fire (q.v.; nar) of hell (q.v.; jahannam). It is noteworthy that both depictions are particularly rich in imagery and together form a double image, consisting of either an equal number of verses (e.g. Q 51:10-4, 15-9: five verses each) or of two verse groups displaying a proportional relation to each other (e.g. the just of Q 69:19-24 as against the evildoers of 69:25-37, seven and fourteen verses, respectively). As such, they remind us of the closely juxtaposed pictorial representations of both sections of the hereafter depicted in Church iconography, thus suggesting the designation of “diptycha.”
Flashbacks
Not infrequently, diptycha comprise recollections of the particular behavior of the inmates of the two abodes during their worldly life, serving to justify their eschato-logical fate. These are stereotypically introduced by innahu kana (Q 69:33), and they are sometimes interspersed with direct speech, e.g. yaqulu ya laytani (Q 69:25). Some of them merge into a catalogue of virtues to be emulated (Q 32:15-7) or vices to be avoided (Q 83:29-33). Independent flashback passages are Q 56:88-94; 75:31-5; 78:27-30; 84:13-5; subgroups of verses within passages are Q 52:26-8; 56:45-8; 69:33-4; 74:43-6; 83:29-32.
Signs (Ayat)
Signs implied in nature
Several descriptions of the “biosphere,” of copious vegetation, fauna, an agreeable habitat for humans, the natural resources at their disposal, and the like, are incorporated into paraenetic appeals to recognize divine providence and accept divine omnipotence, since all these benefits are signs (q.v.; ayat) bearing a coded message. If they are properly understood, they will evoke gratitude and submission to the divine will (Graham, The wind). The perception of nature, which, in pre-Islamic poetry, is a first step to the heroic defiance of its alien roughness, has, by middle Meccan times, crystallized into the image of a meaningfully organized habitat ensuring human welfare and arousing the awareness of belonging. Extensive ayat passages in the strict sense, with their explicit designation of “signs,” do not occur before the second Meccan period; they are, however, preluded by enumerations of divine munificence, as in Q 76:6-16; 77:25-7; 79:27-32; 80:24-32; 82:6-8; 88:17-20; 90:8-10. Often recalling the imagery of the psalms, ayat passages serve to express the progressive change in paradigm concerning the perception of nature. They soon become stock inventory: Q 15:16-25; 25:45-50; 36:33-47; 50:6-11; 14:32-4; 35:9-14, 27-8; 40:61-6; 41:37-40; 42:28-35; 45:12-5. Although signs do occur in polemical contexts (Q 21:30-33: a-wa-lam yara…; Q 78:6:a-lam najʿal…; Q 79:27-33: a-antum ashaddu khalqan ami l-samaʾu banaha…; Q 88:17: a-fa-la yanẓuruna…), hymnical ayat predominate.
Closely related to the hymnical ayat is the hymn as such. Sections praising God’s benevolence, omnipotence and his deeds in history occur predominantly in introductory sections (early: Q 87:1-5; 96:1-5; later: Q 67:1-4 introduced by a doxology; Q 35:1-2). They are also found distributed within the suras (early: Q 53:43-9; later: Q 32:4-9; 25:61-2 introduced by a doxology “tabaraka”; Q 39:62-6). Loosely related to the hymn in a structural sense, but serving a different purpose—namely to present a moral example for the community—is the catalogue of virtues which appears already in early suras and is frequent in later texts (Q 23:57-61; 25:63-76; 42:36-43). Its counterpart is the catalogue of vices which can be traced through the entire corpus (Q 104:1-2; 18:103-5; 53:33-7; 68:8-16).
Signs implied in history: retribution legends
Short narratives—the invasion of Mecca (Q 105); the Thamud (q.v.) myth (Q 91:11-5); the story of Pharaoh (q.v.; Firʿawn) and Moses (q.v.; Musu, Q 79:15-26)—or ensembles of narratives like that in sura 51 including: Abraham (q.v.; Ibrahim) and Lot (q.v.; Luṭ, Q 51:34-7), Moses and Pharaoh (Q 51:38-40), the ʿAd (q.v.; Q 51:41-2), the Thamud (Q 51:43-4), Noah (q.v.; Nuḥ, Q 51:46)—or evocations of stories (suras 51, 53, 69, 73, 85, 89)—occur from the earliest suras onward. The latter sometimes form lists (suras 51, 53, 69, 89). Longer narratives are introduced by the formula known from ayat in nature: a-lam tara…, later by wa-idh (faʿala)…, i.e. they are assumed to be known to the listeners. It is noteworthy that the longer narratives which occur in the first Meccan period are split into equal halves, thus producing proportionate structures (e.g. Q 79:15-26, six plus six verses; Q 51:24-37, seven plus seven verses; and 68:17-34, nine plus nine verses). This remains the rule in later narratives as well. Narratives successively develop into retribution legends or punishment stories (Horovitz, KU, “Strafle genden”), serving to prove that divine justice is at work in history, the harassed just being rewarded with salvation (q.v.), the transgressors and the unbelievers punished by annihilation. At the same time, legends that are located in the Arabian peninsula may be read as reinterpretations of ancient notions of deserted space: sites lie in ruins no longer due to preordained natural processes, but to a fair equilibrium—maintained by divine providence—between human actions and human welfare. Deserted sites acquire a meaning, voicing a divine message. The often-proffered view that it is the retribution legends that are signified with the qurʾanic phrase “the seven reiterated (utterings),” (sabʿan mina l-mathani, Q 15:87) has been called into question by Neuwirth (Der Horizont). From Surat al-Ḥijr (Q 15) onward, retribution legends no longer focus predominantly on ancient Arabian lore but increasingly include biblical narratives: Q 15:49-77 offers a detailed narrative about Abraham and Lot, followed by a shorter report about the People of the Thicket (q.v.; aṣḥab al-ayka) and those of al-Ḥijr (aṣḥab al-ḥijr).
A related genre in terms of function, which also serves paraenetic purposes, is the parable (mathal)—the owners of the blighted garden (aṣḥab al-janna, Q 68:17-33); the good and corrupt trees (Q 14:24-7); the unbelieving town (Q 36:13-32; and cf. Welch, Ḳurʾan, 424). The particular relevance ascribed to parables is obvious from occasional introductory formulas such as wa-ḍrib lahum mathalan (Q 18:32; cf. 18:45). Parables are, however, less frequent than myths and historical narratives.
Salvation history narratives (occurring as complete suras and central sections)
Although initially embedded in catalogues of narratives of partly extra-biblical tradition, stories about major biblical figures like Moses, Jesus (q.v.) and a number of patriarchs known from Genesis gain a function of their own: They become the stock inventory of the central part of longer Meccan Suras. suras from the second Meccan period onward may indeed be read as the enactment of a service (see below). The appearance of biblical stories in the center fulfills the expectation of monotheistic worshippers demanding that the central position of a service should be occupied by the reading of scriptural texts, as is customary in other monotheistic services. These stories are explicitly referred to as elements of al-kitab; indeed, some suras identify themselves as drawing on a pre-existing more extensive text, i.e. as excerpts from a transcendent scripture. Such a book, obviously imagined as being unchangeable and comprehensive, presupposes a stream of tradition that has come to a standstill and became frozen, constituting a store of warranted knowledge. Qurʾanic reference to scripture therefore presupposes a certain stock of narratives existing in a previously fixed form and dispatched by the sender in single portions to form neatly composed pericopes to be inserted into a more extensive recital that also contains less universal elements such as the debate about ephemeral issues of the community. This ceremonial function of the biblically inspired narrative is underlined by introductory formulas, e.g. wa-dhkur fi l-kitabi (Q 19:16, 41, 51, 54, 56). At a later stage, when the particular form of revelation communicated to the Muslim community is regarded as constituting a scripture of its own, i.e. when community matters are acknowledged as part of salvation history, whole suras figure as manifestations of al-kitab.
Although the central position of the narrative in the middle and late Meccan suras is the rule, an exception is presented by Q 17:2-8. As has been argued by Neuwirth (Erste Qibla; id., From the sacred mosque), the particular composition of this sura may be due to its unique rank as a testimony of a cult reform, the introduction of the Jerusalem direction of prayer (qibla, q.v.). Other outstanding cases are Q 18 and Q 12, the latter of which contains the expanded narrative of Joseph (q.v.; Yusuf), which fills the entire sura (cf. Mir, The story of Joseph; Neuwirth, Zur Struktur). The phenomenon of recurring narratives, retold in slightly diverging fashions, has often been interpreted as mere repetition, i.e. as a deficiency. These forms deserve, however, to be studied as testimonies of the consecutive emergence of a community and thus reflective of the process of canonization. Their divergences, then, point to a successively changing narrative pact, to a continuing education of the listeners and the development of a moral consensus that is reflected in the texts (cf. Neuwirth, Negotiating justice). In later Meccan and Medinan suras, when a large number of narratives are presupposed as being well known to the listeners, the position acquired by salvation history narratives is occupied by mere evocations of narratives and debates about them (Neuwirth, Vom Rezitations-text).
Debate
Polemics
It has been argued that debate is one of the essential elements of the Qurʾan (McAuliffe, Debate). This is certainly true for the suras from the middle Meccan period onward. In early Meccan texts, polemical utterances are more often than not directed against listeners who do not comply with the exigencies of the behavioral norms of the cult. These listeners are reprimanded by the speaker in situ, e.g. a-famin hadha l-ḥadithi taʿjabun/wa-taḍḥakuna wala tabkun (Q 53:59 f.); a-raʾayta lladhi yanha/ʿabdan idha ṣalla (Q 96:9 f.). Sometimes curses are uttered against absent persons: tabbat yada Abi Lahabin (Q 111:1 f.) or against humankind in general:qutila l-insanu ma akfarah (Q 80:17); in other cases menaces are uttered against the ungrateful or pretentious: waylun li-… (Q 104:1; 107:4), and these may merge into a catalogue of vices (Q 104:1-2; 107:2-3, 5-7). Whereas in most of these early cases the adversaries are not granted an opportunity to reply: ma li-lladhina kafaru qibalaka muhṭiʿin (Q 70:36), later suras present the voices of both sides. Lengthy polemics are put forward against the unbelievers, sometimes in the presence of the accused (antum-addresses), more often, however, in their absence. During the middle and late Meccan periods, when the community had to struggle against a stubborn opposition, they needed to be trained in dispute. Meccan suras often begin and end with polemical debates, treating diverse points of dissent. In some cases, the absent adversaries are verbally quoted: qalu… (Q 15:6-7), while in other cases the simulation of a debate is presented, instructing the addressee and his listeners to react to a given statement of the adversaries with a particular response: wa-yaquluna… fa-qul… (Q 10:20). These instances—classified by Welch as “say-passages”—are to be regarded as virtual debates performed in the absence of one party of the discussants. As against these cases, there are qul-verses that do not refer to a debate, but serve to introduce prayers or religious mottos. Often polemics respond to the unbelievers’ rejection of the Qurʾan, again figuring at the beginning of suras (Q 15:1-3), the end of suras (Q 21:105-12) or in the conclusions to main parts of suras (Q 7:175-86).
Apologetics (closing sections, sometimes intra-textual)
Like polemics, apologetic sections frequently appear as framing parts of a sura. From early Meccan texts onward they mostly serve to affirm the rank of the Qurʾan as divine revelation, usually constituting the nucleus of concluding sections (early: Q 73:19; 74:54-5; 85:21-2; 87:18-9; later: Q 26:192-227). In later suras these concluding affirmations of the revelation tend to merge into exhortations of the Prophet (Q 11:109-23; 38:67-70; 76:23-31). It is noteworthy that affirmations of the revelation finally become a standard incipit of suras (Q 12:1-3; 13:1; 14:1-4; 28:1-3; 30:1-5; 32:1-3; 39:1-2; 40:1-4; 42:1-3; 45:1-6; 46:1-3), again often merging into exhortations (Q 41:1-8). In some cases, suras are framed by two affirmations of revelation (Q 41:1-5 and Q 41:41-54). In later developments, introductory affirmations are reduced to mere evocations of the book. By far the majority of these suras start with a pathetical evocation of the book, often introduced by a “chiffre” (Q 2:1; 3:1; etc.; see for the most plausible explanation of the initial “mysterious letters,” Welch, Ḳurʾan, 412-4). This incipit seems to hint at a newly achieved cultic function of the recited text which is no longer understood as the immediate communication of a divine message to the community, but as a recital from a sacred scripture assumed as pre-existing and only reproduced through recitation.
Additional “enjeux” to be found in Medinan suras
Medinan suras have not yet been studied thoroughly as to their form and structure. Summary analyses are presented by Nöldeke (GQ), Bell (Qurʾan), Welch (Ḳurʾan) and Robinson (Discovering). Zahniser (The word of God; id., Sura as guidance) has discussed single suras. A systematic investigation of their building blocks is still lacking. It may, however, be stated that with a few exceptions (oath clusters, idha-phrase clusters), all the Meccan “enjeux” are met again in Medinan suras; the eschatological sections and the ayat, however, are no longer unfolded at length, but rather are summarily evoked. This should not be taken as a decisive shift in spiritual interest. Although new topics which occupy the focus of the community’s attention do emerge, the earlier topics remain present, since it is the partial corpus of the early suras (qiṣar al-suwar, later assembled in juzʾ ʿamma, Neuwirth, Koran) that is known by heart by the believers and serves as the textual basis for the emerging ritual prayers.
Regulations
Although occasional regulations—mostly concerning cultic matters—do occur in Meccan suras (Q 73:1-3 addressed to the Prophet, revised for the community in Q 73:20), more elaborate regulations concerning not only cultic but also communal affairs figure in the Medinan context (see Welch, Ḳurʾan). Their binding force is sometimes underlined by a reference to the transcendent source (kutiba ʿalaykum, Q 2:183-7; fariḍatan mina llahi, Q 9:60). Medinan regulations do not display any structured composition nor do they participate in neatly composed units; they suggest, rather, later insertions into loosely connected contexts.
Evocations of events experienced by the community
A new element appearing in Medinan suras is the report of contemporary events experienced or enacted by the community, such as the battle of Badr (q.v.) in 2/624 (Q 3:123), the battle of Uḥud in 3/625 (Q 3:155-74), the expulsion of the Banu Naḍir in 3/625 (Q 59:2-5), the siege of Khaybar in 7/628 (Q 48:15), the expedition to Tabuk in 9/630 (Q 9:29-35) or the farewell (q.v.) sermon of the Prophet in 10/631 (Q 5:1-3). It is noteworthy that these reports do not display a particularly artistic literary shaping. Nor do they betray any particular pathos. It does not come as a surprise, then, that, unlike the situation in Judaism and Christianity, where biblical history has been fused to form a mythical drama of salvation, no such “grand narrative” has arisen from the Qurʾan. A metahistorical blueprint of the genesis of Islam was constructed only later, through the sira (cf. Sellheim, Prophet).
Types of early Meccan suras
The spectrum of different ensembles is very broad in early Meccan times. Sura types range from mono-partite pieces: pure hijaʾ (Q 111), pure exhortations through the Prophet (Q 94), pure eschatological discourse (Q 95; 100; 101)—to bipartite ones: oath cluster (Q 92:1-13), eschatological section (Q 92:14-21)—to the later standardized tripartite sura: exhortations (Q 74:1-10), polemics (Q 74:11-48), affirmation of the Qurʾan (Q 74: 49-56). (See for their proportions, Neuwirth, Studien, 235-7.) Characteristic of this group as a whole is their striking self-referentiality. The suras mirror a scenario locally situated in a Meccan public place, most probably close to the Kaʿba (q.v.), taking into account their pronouncedly articulate references to sacred space and human behavior therein, as well as sacred time. The rites at the Kaʿba seem to be the Sitz-im-Leben of many early suras, the Kaʿba not only serving as the locale for the performance of their recitation, but its rites also marking particular times of the day respected by the community as ritually significant. In as much as these suras are memorized without any written support, their mostly distinct proportions are effective as mnemonic-technical devices.
Types of later Meccan suras
Things change substantially in later Meccan times. We may localize the caesura with Q 15, where, for the first time, an allusion is made to the existence of a particular form of service in which scripture functions as the cardinal section (cf. Neuwirth, Vom Rezitationstext; id., Referentiality and textuality). In these suras, the references to the Meccan ḥaram as the central warrant of the social coherence of the community have been replaced by new symbols. Instead of introductory allusions to liturgical times and sacred space we encounter an evocation of the book, be it clad in an oath (Q 36:2; 37:3; 38:1; 43:2; 44:2; 50:1) or through a deictic affirmation of its presence (Q 2:2; 10:1; 12:1; 13:1; etc.). Moreover, a new framework of the message in terms of space is realizable, and later Meccan suras have broadened the scope for the listeners, who are led away from their local surroundings to a distant landscape, the holy land, which becomes familiar as the scenery where the history of the community’s spiritual forebears has taken place. The introduction of the Jerusalem qibla is an unequivocal testimony to this change in orientation (Neuwirth, Erste Qibla; id. From the sacred mosque). In view of the increasing interest in the biblical heritage, it comes as no surprise that the bulk of the middle and late Meccan suras (twenty-seven instances) seems to mirror a monotheistic service, starting with an initial discursive section (apologetic, polemic, paraenetic) and closing with a related section, most frequently an affirmation of the revelation. These framing sections have been compared to the ecclesiastic ecteniae (initial and concluding responsoria consisting of pleadings for divine support recited by the priest or deacon with the community complementing the single addresses through affirmative formulas). The center of the monotheistic service and, similarly, of the fully developed sura of the middle and late Meccan period is occupied by a biblical reminiscence—in the case of the service, a lectio; in the case of the sura, a narrative focusing on biblical protagonists (Neuwirth, Vom Rezitations-text). Ritual coherence has thus given way to scriptural coherence, the more complex later suras referring to scripture both by their transmission through diverse processes of writing and by being themselves dependent on the mnemonic-technicalities of writing for their conservation. (For particular sequences of single “enjeux” and topics in these compositions, cf. the inventory in Neuwirth, Studien, 318-21.)
Types of Medinan suras
It is true that, already in later Meccan suras, the distinct tripartite composition often becomes blurred, with narratives gradually being replaced by discursive sections. Some compositions also display secondary expansions—a phenomenon that still needs further investigation. Yet, for the bulk of the middle and late Meccan suras, the claim to a tripartite composition can be sustained. In Medina, however, suras have not only given up their tripartite scheme, but they display much less sophistication in the patterns of their composition. One type may be summarily termed the “rhetorical sura” or “sermon” (Q 22; 24; 33; 47; 48; 49; 57 until 66); they consist of an address to the community whose members are called upon directly by formulas such as ya ayyuha l-nasu… (Q 22:1). In these suras, which in some cases (Q 59; 61; 62; 64) are stereotypically introduced by initial hymnal formulas strongly reminiscent of the biblical psalms, the Prophet (al-nabi, Q 33:6) appears no longer as a mere transmitter of the message but as personally addressed by God (ya ayyuha l-nabiyyu, Q 33:45) or as an agent acting synergetically with the divine persona (Allahu wa-rasuluhu, Q 33:22). As against these intended monolithic “addresses,” the bulk of the Medinan suras are the most complex. The so-called “long suras” (Q 2-5; 8; 9) cease to be neatly structured compositions but appear to be the result of a process of collection that we can no longer reconstruct. As pointed out earlier, a systematic study of these suras is still an urgent desideratum in the field.