Muhammad ibn Ishaq ﵀ and the First Systematic Maghazi
A Second Surviving Folio
In the previous article we took into our hands the nuskhah of Suhail ibn Abi Salih ﵀, a hadith manuscript from the second century AH in which a pupil of the Tabi’i generation set down his father’s transmission on parchment in an ordered form and passed it on to his own circle. It was a small book, modest in size, but great in its significance, because it supplied us with a direct material artefact from a period of which part of the orientalist critique had maintained for more than a century that it knew no written transmission. In that article we dissected the isnad line of the nuskhah, pointed out the substantive parallel with the Sittah transmissions, and showed how firmly a physical document, verifiable in its material state, with an unbroken chain of transmission reaching into the modern editions, can close a discussion on a single point.
This article follows a second artefact from roughly the same sphere of transmission, but from a different scholarly discipline. Where the nuskhah of Suhail ﵀ belongs to the hadith corpus, the folio that Mohammad Mustafa Azami ﵀ reproduced in Appendix IV of Studies in Early Hadith Literature belongs to the Maghazi literature, that is, the genre that records the battles, the embassies and the campaigns of the Prophet ﷺ in ordered form. The folio is a preserved fragment from al-Maghazi of Muhammad ibn Ishaq ibn Yasar ﵀, the Medinan scholar who in the first half of the second century AH made the sira genre into a systematic science. The folio is not from the later redaction of Ibn Hisham ﵀, the work that circulates in modern editions as Sirat Ibn Hisham and that most readers know as their gateway to Ibn Ishaq ﵀. It is from an older, parallel line of transmission, and as we shall see further on, Azami names the transmitter of precisely this folio by name and isnad: Muhammad ibn Salamah al-Harrani ﵀.
The significance of this folio reaches beyond the Maghazi science itself. A considerable part of what has counted since the nineteenth century as critical sira and hadith scholarship rests on an assumption about the early transmission that comes under pressure from an artefact such as this. In the tradition of Ignaz Goldziher and later Joseph Schacht it has been argued, with great influence, that the early Islamic transmission, including the isnad apparatus, would to an important degree be a later back-projection, constructed by scholars of the second and third centuries who cast the past into shape after the fact. A physically surviving folio from the work of a scholar who died in 151 AH, in an isnad transmission documented by the early biographical literature, is exactly the type of material that tests such assumptions. In this article we follow the life, the method and the pupils of Ibn Ishaq ﵀; we describe the Appendix IV folio and what precisely Azami establishes about it; and we place the finding in the wider context of the textual criticism that distrusts the early sira.
Who Was Ibn Ishaq ﵀
Muhammad ibn Ishaq ibn Yasar al-Muttalibi ﵀ was born in Medina, according to the classical biographical literature, around 85 AH. His grandfather Yasar was a freedman who in the time of Abu Bakr ﵁, after the capture of Ayn at-Tamr in Iraq in 12 AH, had been brought to Medina as a prisoner of war and had been freed there by Qays ibn Makhrama of the Banu Muttalib. Through this manumission the household of Yasar attached itself legally to the Quraishi clan of the Banu Muttalib, and the grandson would later carry the al-Muttalibi in his name as a sign of that affiliation. The family was literate and showed a pronounced interest in the Maghazi material; of Yasar’s sons Musa and Abd al-Rahman it is recorded in the biographical literature that they were themselves transmitters of reports about the battles, so that the young Muhammad grew up in a household in which the accounts of Badr, Uhud, the Battle of the Trench and the treaty of Hudaybiyya were familiar matter. All of this is classical biography, preserved in Ibn Sa’d and in the later rijal literature; it is not the material for which we rely on Azami, and we keep the distinction clean.
What the young Ibn Ishaq ﵀ found in Medina was more than a family interest. He entered the student circles of the city in a period in which a second and third generation after the Sahaba was ordering the inheritance, and in which the first systematic shape of the hadith science was being built up by a handful of great masters. The foremost of these masters was Muhammad ibn Shihab az-Zuhri ﵀, the first systematic compiler of the hadith, whose life and work have been followed elsewhere in this series. Al-Zuhri ﵀ was at that moment a leading teacher in Medina, a Quraishi of the Banu Zuhra with direct lines to the households of the Prophet ﷺ, and he had in his circle a stream of pupils who came to him for various sub-disciplines. The young Ibn Ishaq ﵀ entered this circle as one of the pupils who came above all for the Maghazi material.
For the Maghazi other masters were available as well. Hisham ibn Urwa ﵀, son of Urwa ibn az-Zubayr ﵀, was one of the principal sources for the early events. His father Urwa ﵀, a nephew of Aisha ﵂ through the side of her sister Asma, belonged to the first Tabi’in who had written down the life story of the Prophet ﷺ in fragments and passed it on to their sons; through that line Hisham ﵀ was himself a great-nephew of Aisha ﵂. Asim ibn Umar ibn Qatada ﵀, a scholar from an Ansari tribe, supplied the Medinan eyewitness transmission that had been preserved in his own family, with the details about Badr and Uhud as his ancestors had experienced them. Abdullah ibn Abi Bakr ibn Hazm ﵀ supplied the transmission that his father, the same Abu Bakr ibn Hazm to whom Umar ibn Abd al-Aziz ﵀ had addressed the well-known hadith compilation decree, had recorded in his official register. Yazid ibn Ruman ﵀, a freedman of the Quraish, added a line of transmission that he had received from Urwa ﵀.
To these masters Ibn Ishaq ﵀ added a series of others in his years of study. The later biographical literature, in particular Tahdhib at-Tahdhib of Ibn Hajar al-Asqalani ﵀ and Siyar A’lam an-Nubala of adh-Dhahabi ﵀, registers a large number of names of scholars from whom he received transmission. The geographical spread of that list is important for what follows. He received not only from Medinans, but also from Meccans, from Egyptians who had come to the Hijaz to perform the Hajj and on that occasion taught in the Prophetic mosque, and from Iraqis who stayed in Medina for trade or administration and brought their transmission with them. He was, in other words, a scholar who drew into his web of knowledge the broad circle that at a given moment converged in Medina.
The Method
What distinguished Ibn Ishaq ﵀ in his manner of working from his predecessors in the Maghazi material was not the presence of isnad, but its systematic character. Before him a tradition already lay in place: Urwa ibn az-Zubayr ﵀ had ordered the material in fragments, and al-Zuhri ﵀ had passed these fragments on in his circle in a wider arrangement. Maghazi works of contemporaries and predecessors were also already in circulation; Azami himself treats in Studies in Early Hadith Literature, among others, the Maghazi of Musa ibn Uqbah and of Aban ibn Uthman as early works in this field. It is therefore incorrect to say that nothing existed before Ibn Ishaq ﵀. What the article proposes here is an interpretation: that Ibn Ishaq ﵀ brought these building blocks into a more coherent whole than his predecessors, with a chronological structure that fixed the entire biography of the Prophet ﷺ, from his birth to his passing, including the battles and embassies, in one running framework, and with, for each event, a traceable reference to the source from which it derived.
One may read this work, by way of analytical ordering, in three layers. The first layer was the chronological backbone, from the Mubtada wa al-Mab’ath (the Beginning and the Sending) through the Meccan years to the Hijra, and from there in a year-by-year structure through the whole Medinan period. The second layer was the substantive elaboration per event, with the motives, the geographical context, the participants, the outcomes and the spiritual significance. The third layer was the source attribution, in which he indicated for each individual account where it came from, sometimes in the form of a complete chain up to a Sahabi ﵁, sometimes in a collective formulation in which he brought several sources together. When he had something not from a direct chain but from a general transmission, he said so. When he had something from one single source, he said so. When he had several sources that reinforced one another, he said that too. This three-layer reading is a device of this article to make his method intelligible, not a formulation that stands this way in the sources themselves.
Robson, whose summary Azami cites in the sixth chapter of Studies in Early Hadith Literature, points out in this connection that Ibn Ishaq ﵀ did indeed employ the isnad apparatus in his work, albeit not always with the strictness that the later hadith critics would demand of a pure hadith transmission. That distinction, between the norm of the Maghazi narrator and the norm of the hadith critic, is important for understanding his later position in the tradition, and we return to it further on.
A methodological detail concerns his dealings with living witnesses. Ibn Ishaq ﵀ gathered part of his material not exclusively from scrolls or from sessions with great masters, but also from direct conversations with the children and grandchildren of Sahaba ﵃. In Medina, in the years in which he was building up his work, numerous families were still living whose father or grandfather had played a role in one of the Prophetic battles or in one of the Medinan events. He went to these families, he asked what had been preserved in their household about the event, he noted down what they told with the indication of the source, and he integrated this material into his work where it fitted, alongside the transmission he had received through the scholarly circles. This manner of working, the direct familial inquiry, has sometimes been underestimated in the modern scholarly literature; one might compare it with what a present-day collector of oral history does when he goes to the surviving witnesses of an event to set their version beside the known version. That comparison is a modern figure of speech of this article, not a statement of the sources; it serves only to make the nature of his work vivid. What is certain is that he used the time in which the generations after the Prophet ﷺ were still alive, and that he collected from them transmission that would otherwise have been harder to recover.
Between Medina and Baghdad
The early years of Ibn Ishaq ﵀ passed in Medina itself, in the shadow of his masters and in the growth of his own circle. But as with al-Zuhri ﵀ before him, Medina was not his final station. When his work approached a certain completion, he began to travel. He stayed for some time in Egypt, where he had meetings with the scholars present there and exchanged transmission. He stayed in al-Jazira, the upper Mesopotamian region that in his time harboured a considerable community of learning. He stayed in al-Hira, the city in the south of Iraq where the old Christian and Arabic traditions of letters had met one another.
Toward the end of his life, in the years around 150 AH, he arrived in Baghdad. The Abbasid state had come into being in 132 AH through the fall of the Umayyads, and in 145 AH al-Mansur had had the new capital Baghdad laid out at a place where the Tigris and the Euphrates come close together. Al-Mansur was a caliph who actively surrounded himself with scholars from all corners of the world of Islam, not because he accepted them in all their theological judgements, but because he understood that a capital without a scholarly infrastructure would not remain a capital. According to the classical account he had Ibn Ishaq ﵀ come from Medina to Baghdad, and asked him to draw up a short overview of the Maghazi for his son and intended successor al-Mahdi.
The anecdote, which has been preserved by several early historians, gives a picture of the relationship between the caliph and the scholar. Al-Mansur asked for a work that he considered suitable for the education of his successor. Ibn Ishaq ﵀ delivered a text, or rather a structured summary of his larger work. The caliph, so it is told, looked into the document and judged that it was too extensive; he wished a more concise version. Ibn Ishaq ﵀ shortened it, and the short version was taken into the court archive for the instruction of al-Mahdi. Whether this short version has come down to us is disputed, but that Ibn Ishaq ﵀ lived in Baghdad in his last years and there passed his work on to a wider circle stands firm. This episode is classical biography, preserved in Ibn Sa’d and in the later historiography of Baghdad, not material from Studies in Early Hadith Literature.
He died in Baghdad in 151 AH, at an age of about sixty-six years. Azami notes this date of death, 151 AH, where he brings Ibn Ishaq ﵀ into the isnad debate. He was buried in the Khayzuran cemetery, one of the early Baghdad burial grounds, where in the course of the second and third centuries AH numerous scholars would find their final resting place. To the city where he died he gave the first coherent sira text that circulated in its circle of learning, and from that circle a number of his principal recensions would depart for the other centres of the world of Islam.
The Many Pupils and Their Recensions
What made Ibn Ishaq ﵀ a figure of lasting significance was not only the work he produced, but the geographical spread of his pupils. He taught in Medina, in Egypt, in al-Jazira, in al-Hira and in Baghdad, and in each of those centres a circle of pupils arose that took over his work in a recension of its own. The biographical literature registers a large number of pupils by name, and the number of people who received hadith or Maghazi material from him in one or more sessions certainly lies higher. Azami remarks in this connection that many versions of this work existed; in a note to his Appendix I he speaks, with reference to the introduction of Guillaume’s translation, of at least fifteen versions of Ibn Ishaq’s work.
For the transmission history of his work a few recension transmitters are of particular importance. The first is Ziyad ibn Abdullah al-Bakka’i ﵀, a pupil from Kufa who in his younger years had sat at Ibn Ishaq’s feet in Medina and who had returned to Kufa with a complete redaction of al-Maghazi in his baggage. This al-Bakka’i recension, which has not come down to us directly in its original form, formed the basis on which Abdul-Malik ibn Hisham ﵀ in Egypt composed his own edition some decades later. Ibn Hisham ﵀ took over the text of Ibn Ishaq ﵀ through the al-Bakka’i transmission, removed a number of passages that he did not consider suitable for publication, added philological explanations where the text might be unclear to his readers, and brought the work out in a new whole as his own Sirah. This is the Sirah that from the third century AH onward circulated throughout the world of Islam under the name Sirat Ibn Hisham, and that in the modern age, since the Göttingen edition that Wüstenfeld published around 1859 and the dozens of Arabic editions that appeared thereafter, has been the standard work on the life story of the Prophet ﷺ.
A second recension transmitter is Muhammad ibn Salamah al-Harrani ﵀, and it is this name that stands central for this article, for it is his recension that Azami reproduces in Appendix IV of Studies in Early Hadith Literature. We return to this at length in the next chapter. At this point it suffices to establish that Muhammad ibn Salamah al-Harrani ﵀ represents his own line of transmission of Ibn Ishaq’s work, documented by the biographical literature, and that he must not be confused with another transmitter with a similar-sounding name who will come up shortly.
That other transmitter is Salama ibn al-Fadl al-Abrash ar-Razi ﵀, a pupil from Rayy in the north of Iran, who passed on the longest of all known recensions. Azami names him expressly where he treats the work of Ibn Ishaq ﵀ in the fifth chapter: he cites how Yahya ibn Ma’in spoke of a man in Baghdad who transmitted the Kitab of Salama from Ibn Ishaq, and he notes that Salama’s books on the Maghazi were counted among the most complete. On this Salama recension Muhammad ibn Jarir at-Tabari ﵀ relied in the third century AH when in his monumental Tarikh ar-Rusul wa al-Muluk he quoted extensively from Ibn Ishaq ﵀. The quotations in at-Tabari ﵀ give us a control point on the text, alongside the Ibn Hisham redaction and the other recensions. Muhammad ibn Salamah al-Harrani ﵀ and Salama ibn al-Fadl al-Abrash ﵀ are two different men; it is an easy pitfall to merge them through the shared sound of their names, and we keep them carefully apart.
Besides these transmitters the tradition has also preserved a recension of its own under the name of Yunus ibn Bukayr ﵀, likewise a pupil from Kufa, whose redaction differs on a number of points from the al-Bakka’i version and which has been transmitted in fragments in later works and manuscripts. To the list may further be added Ibrahim ibn Sa’d al-Zuhri ﵀, Abdullah ibn Idris ﵀ and Jarir ibn Hazim ﵀, all pupils with lines of transmission of their own that were taken up in later works. The sum of all these recensions is that the original work of Ibn Ishaq ﵀ fanned out through a large number of independent channels into the third century AH, so that whoever wishes to reconstruct the text in its entirety can compare these recensions with one another and approach the original work from their common core. The Ibn Hisham redaction is the best known among them, but it is, as Azami shows precisely in Appendix I when he sets it beside the recension of Muhammad ibn Salamah al-Harrani ﵀, not identical in all details to the text as other transmitters passed it on.
The Azami Folio
We now come to the heart of this article. In Appendix IV of Studies in Early Hadith Literature Azami reproduces a plate of a manuscript folio from al-Maghazi of Ibn Ishaq ﵀, with beside it the corresponding printed text from the Sirat of Ibn Hisham ﵀. The caption to the plate is sober and precise: a page from al-Maghazi of Ibn Ishaq, with the corresponding printed text in the Sirat of Ibn Hisham. It concerns, then, a folio whose content Azami sets beside the familiar Ibn Hisham text, so that the reader can see the two transmissions side by side.
The discussion that belongs with it does not stand in the Appendix itself but in Appendix I of the same book, in the twenty-fifth of the nuskhah descriptions that Azami gives there. There he names the work expressly: al-Maghazi of Ibn Ishaq is a well-known book, transmitted by various pupils, and one can compare the version of Ibn Hisham with that of Muhammad ibn Salamah, to which he refers for the plate. The isnad of the manuscript he gives precisely with it: Abd Allah ibn al-Hasan al-Harrani from al-Nufaili from Muhammad ibn Salamah from Ibn Ishaq. In this chain the name Yunus ibn Bukayr ﵀ does not occur; the folio belongs to the recension of Muhammad ibn Salamah al-Harrani ﵀, not to that of Yunus. This is a detail that one easily remembers wrongly, and we place the emphasis on it here, because the entire evidential force of the folio hangs on the correctness of this isnad. Azami notes moreover that the manuscript was read aloud to the great traditionist and historian al-Khatib al-Baghdadi in the year 454 AH, a datum that places the text within a documented history of transmission.
It is important to say precisely what Azami does and does not do with this folio, for an earlier reading of this material has ascribed to him things that are not there. Azami supplies no extensive palaeographical treatise with this folio. He does not date the hand of the script to the second or third century; the dating he gives concerns the transmission, namely the reading before al-Khatib al-Baghdadi in 454 AH. His actual remark on the comparison of the two versions is short: he points to very slight differences between the manuscript and the Ibn Hisham text, and to one important variation in the isnad, namely the al-Harrani chain. Of an argument that no later addition whatsoever would be found in the folio, there is no trace in Azami. On the contrary: in another place in the same book, where he discusses the phenomenon of copyists who add text, he refers precisely to this Appendix IV folio as a clear example of a copyist who inserted two lines even before the sentence was completed. In Azami, then, the folio does not illustrate the absence of scribal insertion, but precisely a case of it. We take this over as he gives it, for the honesty of the argumentation weighs heavier than a smoother but incorrect presentation.
What the folio does prove must be formulated with this nuance. It is a tangible manuscript folio that contains a passage from al-Maghazi of Ibn Ishaq ﵀ in the recension of Muhammad ibn Salamah al-Harrani ﵀, with an isnad confirmed by the biographical literature, and which through a documented transmission, up to the reading before al-Khatib al-Baghdadi in 454 AH, is connected with the early period. It does not stand alone. When a passage from Ibn Ishaq’s work is present in the recension of Muhammad ibn Salamah al-Harrani ﵀, is present in the Ibn Hisham redaction, and is present in the at-Tabari ﵀ quotations through Salama ﵀, and when these versions agree with one another on the main lines, with identifiable and explicable small variations, then the common core of that passage can with great certainty be ascribed to Ibn Ishaq ﵀ himself. The common core was not invented by one transmitter, for it appears in recensions that were developed in different centres, Harran, Kufa, Egypt and Rayy. That a single transmitter could add text on a single point, as Azami himself points out with this folio, takes nothing away from that common core; it is precisely by setting such cases side by side that one can distinguish the addition from the core. This multiple-recension argument is the article’s own reasoning, built on the building blocks that Azami supplies, not a conclusion that he writes down in this form.
The Scepticism about the Early Sira
It is here that the wider significance of an artefact such as this becomes visible. Since the nineteenth century an influential current in the Western study of early Islam has fundamentally called the reliability of the Islamic transmission into question. Ignaz Goldziher argued that a large part of the hadith corpus reflected the theological and juridical struggles of the second and third centuries AH and was ascribed after the fact to the Prophet ﷺ and his companions. Joseph Schacht developed this line further for the law, and formulated a method in which the isnad apparatus was treated not as a control mechanism but as an instrument that had to lend credibility to a later formulation; according to his famous thesis, isnads grew backwards with time, in the direction of the Prophet ﷺ. Applied to the sira and the Maghazi, this attitude issues in the assumption that the detailed accounts of Mecca and Medina may not be read as early transmission, but as later constructions that present themselves as early.
It is against this attitude that Azami set his method in Studies in Early Hadith Literature. His book appeared in 1966 and 1967, out of a dissertation defended at the University of Cambridge. It addressed the generation of the early textual criticism, and its strength lies not in theoretical contradiction but in the showing of material. Azami lays manuscripts on the table, he compares isnads, he sets variants side by side, and he draws the conclusions that this material permits. The Appendix IV folio is a textbook example of this manner of working. It is not a theoretical argument but a physical document, with an isnad that one can trace, with a transmission history that reaches to a dated reading in the fifth century AH, and with a text that one can set beside the Ibn Hisham redaction. For whoever holds that the early sira transmission was erected only much later, this type of material is a touchstone: a second-century work preserved in several independent recensions, of which one lies before us in a manuscript with verifiable isnad, does not let itself be explained away without more ado as a late fiction.
Here we must, as an article, respect a chronological boundary that was crossed in the earlier version of this text. In the second half of the twentieth century the scepticism about early Islam was carried through in a radical form by a new generation of researchers. John Wansbrough argued in Quranic Studies of 1977 and in The Sectarian Milieu of 1978 that the canonisation of the Quran text would have come about only late, in the third century AH, and that the transmitted history of the earlier period was a retrospective mythologisation. In the same year 1977 Patricia Crone and Michael Cook, partly influenced by Wansbrough, published their book Hagarism: The Making of the Islamic World with Cambridge University Press, in which they largely set aside the early Islamic self-description and sought to reconstruct the origin of the community from non-Islamic sources, in particular Syriac-Christian and Armenian chronicles from the seventh century. The Sectarian Milieu is, for that matter, the title of a book by Wansbrough, not the name of a school, and Hagarism is a work of Crone and Cook alone; Wansbrough is not a co-author of it. These distinctions are not hair-splitting: it is precisely the carelessness with which names, dates and theses are mixed together that led the earlier version of this article astray.
What matters is that Studies in Early Hadith Literature is older than all these works. Azami’s book appeared a decade before Hagarism, before Quranic Studies and before The Sectarian Milieu. Azami could not discuss these authors in this book, and he did not do so; his Appendix IV is not a weapon against Crone, Cook or Wansbrough, and it would be an anachronism to present him as such. What one may say, and what this article therefore formulates as its own observation about the later context, is that the type of material Azami laid on the table in 1966 has since then only been supplemented. In the decades after these radical revisionists, the early Quran manuscripts from Sanaa and the folios from Birmingham, for the latter of which a very early age has been established by radiocarbon dating, have given rise to a scholarly discipline of their own that works with physical documents instead of theoretical reconstruction. The direction in which that material points lies in the extension of what Azami did with his Maghazi folio: it brings the discussion back from theory to the manuscript. Whoever holds the early Islamic literature to be a late construction must test his thesis against the documents preserved in the libraries, and those documents have proved more numerous and older than the most sceptical reconstruction supposed. This is emphatically the article’s inference about the wider history of research, not a statement ascribed to Azami or to his Appendix IV.
The Debate within the Tradition
It would be incorrect to create the impression from the foregoing that Ibn Ishaq ﵀ was never criticised within the Islamic tradition itself. He was criticised, and the criticism comes from the highest ranks of the hadith science. But it is a debate within the tradition, with substantive arguments about specific points, and not a judgement that dismisses the whole Maghazi material as fiction. The difference between these two kinds of criticism is essential for understanding his position in history.
Imam Malik ibn Anas ﵀, Ibn Ishaq’s contemporary in Medina, had a number of reservations about his work. The anecdotes are transmitted at length in the biographical literature. Malik ﵀ is reported to have called Ibn Ishaq ﵀ al-dajjal min al-dajajila, “one of the deceivers”, a formulation whose sharpness may surprise the modern reader. The occasion lay, so far as the sources recover it, in a combination of factors. Ibn Ishaq ﵀ included in his work transmission from converts out of the Jewish rabbinic tradition in Medina, and with that, Israelite narratives (Isra’iliyyat) sounded through in his treatment of the pre-Islamic history of the prophets. For Malik ﵀ this was a methodological objection; he held that a Maghazi work had to confine itself to strictly Prophetic-Islamic sources. There was in addition a personal component: Ibn Ishaq ﵀ is said to have included in his work things about a number of Medinan families that were experienced within those families as hurtful, and Malik ﵀, who lived in the same city and moved among the same families, shared that displeasure.
The criticism of Malik ﵀ has been placed in its context by later generations. Adh-Dhahabi ﵀ in Siyar A’lam an-Nubala and Ibn Hajar ﵀ in Tahdhib at-Tahdhib have both argued at length that Malik’s reservation was not a general rejection of Ibn Ishaq’s reliability, but a specific criticism of a number of delimited points. Other great masters of the same generation praised Ibn Ishaq ﵀ in the very highest terms. Sufyan ibn Uyayna ﵀ in Mecca and Shu’ba ibn al-Hajjaj ﵀ in Basra belonged to his defenders; of the venerable title amir al-mu’minin fi al-Maghazi, “the leader of the believers in the Maghazi science”, more than one attribution has been transmitted in the biographical literature, so that we mention it here as an honour that fell to him in the tradition without fixing it to one speaker. The later consensus, as it is recorded in the rijal literature, was that Ibn Ishaq ﵀ as a transmitter of hadith was placed in the rank of the acceptable, that is, as a transmitter whose transmission was acceptable when supported by other sources; in the strictest Sahih material he was therefore handled with restraint, while in the Maghazi material, where his work formed the basis, he counted as the authority from whom every later historian set out.
The debate within the tradition about Ibn Ishaq ﵀ is in its nature a scholarly discussion between experts. It was about isnad criteria, about the type of sources that belonged in a Maghazi work, about the treatment of Israelite narratives, about the balance between completeness and strictness. It was not about the question whether the Prophet ﷺ had lived in Mecca, whether the Hijra had taken place, whether the battle of Badr had been fought in 2 AH. Those facts stood beyond all doubt for every participant in the debate, because they were confirmed through countless isnad chains and in countless parallel sources.
The modern scepticism, in the line of Goldziher and Schacht and in its later radical form, asserts something of a different order. It calls into question the reliability of the transmission itself, and in its most far-reaching form also the contours of the events. The type of criticism it conducts is not of the same kind as the criticism of Malik ﵀ on Ibn Ishaq ﵀, and the two debates may not be confused with one another in serious historical scholarship. Whoever adduces the criticism within the tradition as support for the modern scepticism misses the distinction between a methodological refinement within a scholarly discipline and a doubt about the existence of that discipline itself.
What the Folio Means
We return to the folio of Appendix IV to fix its significance in a final summary. The folio is a physical artefact, a manuscript page from al-Maghazi of Ibn Ishaq ﵀ in the recension of Muhammad ibn Salamah al-Harrani ﵀. Its isnad, Abd Allah ibn al-Hasan al-Harrani from al-Nufaili from Muhammad ibn Salamah from Ibn Ishaq, is confirmed by the biographical literature, and its transmission reaches to a dated reading before al-Khatib al-Baghdadi in 454 AH. Azami sets it beside the corresponding text in Ibn Hisham ﵀ and notes the very slight differences and the one isnad variation; he points out moreover that the folio contains a case of scribal addition, which underlines the honesty of his comparison. It is, with these nuances, a primary source in the strict technical sense of the word: a document connected with the early history of transmission by a direct material and documented trace, not merely by a literary reconstruction.
What this folio means in the Maghazi science lies in the extension of what the nuskhah of Suhail ﵀ means in the hadith science, and of what the manuscripts from Sanaa and the folios from Birmingham have come to mean in the Quran science. It brings a certain kind of discussion back from theory to the manuscript. As long as a researcher asserts that the early Islamic literature is not primary, and that its content would be a later construction, he can test that assertion against this type of artefact. A manuscript that contains a passage from a second-century work, in a transmission documented by the early biographical literature, is a datum against which a theoretical consideration must measure itself.
The methodological principle that Azami works out in Studies in Early Hadith Literature is strong in its simplicity. It comes down to consulting the manuscripts that exist, comparing the isnad transmissions that exist, setting side by side the textual variants that exist, and drawing from that material the conclusions it permits. Theoretical considerations about what an early Islamic community could or could not have transmitted are a second order of argumentation; the first order is what is physically present in the archives and libraries. Whoever holds to this principle arrives at a history that agrees in its main lines with the Islamic self-description and that leaves the most radical forms of the scepticism standing as unproven. This characterisation of Azami’s manner of working is a paraphrase of this article, not a literal sentence from his book, but it renders the spirit of Studies in Early Hadith Literature faithfully.
There lies in all this a certain irony. Islamic scholarship had, in its isnad discipline, developed a testing mechanism centuries before modern critical textual scholarship, which it deployed to establish what had been transmitted through which channels, with which variations and with which support. One might compare it with what European editorial science developed for the classical texts in the nineteenth century when it mapped the manuscript families; that comparison is a figure of speech of this article, not a historical descent. The reserve with which part of the modern criticism has treated the isnad mechanism as a disguise for fabrication looks, in this light, more like a prejudice than an inference from the sources, for in the sources themselves, and in a folio such as that of Appendix IV, it finds no support for that suspicion.
Toward Cairo
We close with a look ahead to what follows in the next article. While Ibn Ishaq ﵀ in Medina and later in Baghdad worked out the Maghazi science in its systematic form, something parallel was happening in another geographical pole of the world of Islam, in a different scholarly field. In what would later become Cairo, or rather in al-Fustat, which in these years functioned as the administrative seat of Islamic Egypt, lived a scholar of roughly Ibn Ishaq’s generation who performed a comparable work in his own discipline. Al-Layth ibn Sa’d al-Fahmi ﵀, born in 94 AH and deceased in 175 AH, was the Egyptian scholar who in his school of law, his correspondence and his teaching held fast, for an entire generation, a line of Prophetic transmission that was closely bound to the Hijaz but at the same time had a geographical and cultural context entirely its own.
The parallel between Ibn Ishaq ﵀ and al-Layth ﵀ has been underestimated in the existing historiography. Both belong to the later generations of the early transmitters, after the Sahaba. Both received their formation with masters who had learned directly from Sahaba ﵃. Both built up a circle of pupils that in the third century AH would grow into a geographically distinct scholarly tradition. What Ibn Ishaq ﵀ did in Medina for the Maghazi material, al-Layth ﵀ did in al-Fustat for the Prophetic transmission in the Egyptian context, and when we set their two bodies of work side by side, a picture arises of an Islamic scholarship that in its second century was not concentrated in one centre, but worked in a network of centres in parallel on the same material, in mutual contact and with mutually testable results.
The next article follows al-Layth ﵀ in his Egyptian context, with his correspondence to Imam Malik ﵀ in Medina, with his school of law that in the third century AH was displaced by the rise of the Shafi’i school, and with his position as the second pillar, beside Ibn Ishaq ﵀, of what we might call the non-Iraqi branch of the early science of transmission. The folio from Appendix IV is a first tangible witness of what was passed on in the second century AH along the Hijazi-Iraqi track; in al-Layth ﵀ we meet the same phase at another geographical pole, with the Nile instead of the Tigris and the Euphrates, and with a jurisprudence instead of a Maghazi corpus, but with the same underlying discipline of transmission, isnad testing and geographical spread on which the credibility of the whole second-century transmission rests.