Abu Hurayra ﵁ and His ~800 Students

From the pen of Ali ﵁ to the mouth of Abu Hurayra ﵁

In the previous article we saw the sahifa of Ali ibn Abi Talib ﵁ slide out of its sheath. A rolled parchment on the sheath of his sword, small in size, great in meaning. On it stood written what the Prophet ﷺ had ruled on diyat, on the protection of an escaped captive, on the forbidden zones around Medina. When Ali ﵁ was asked on the minbar whether anything else written from the Prophet ﷺ was in circulation among his Ahl al-Bayt, he shook his head, apart from that one sahifa. The message was clear: written material from the Prophet ﷺ existed. It was merely scarce, targeted, and in the hands of those who truly needed it.

But what then of the mass of transmissions that stand in our great collections. In al-Bukhari ﵀ alone more than seven thousand hadith. In Imam Muslim ﵀ around four thousand. In Ahmad ibn Hanbal ﵀, in his Musnad, nearly thirty thousand transmissions, after the sifting of duplicates amounting to about eight thousand distinct narratives. A bold calculator might ask: if most Companions ﵃ wrote nothing down, and if Ali ﵁ possessed only one small sahifa, where then does that enormous quantity of hadith come from.

The answer is a Yemenite young man who came to Medina in 7 AH, sat for three years at the feet of the Prophet ﷺ, and then spent more than forty years teaching. His name no one really knows; his kunya everyone knows. The man with the cat. Abu Hurayra ﵁.

And exactly here, on the threshold of his name, begins one of the oldest and most stubborn confusions in the history of Islamic text-critical thought. For Abu Hurayra ﵁ is, paradoxically, at once the Companion ﵁ with the largest number of transmitted hadith and the Companion ﵁ whose reliability has most often been called into question by outsiders. An Egyptian pamphleteer would dare to call him “the liar of the umma” in the twentieth century. A Hungarian orientalist had already suggested half a century earlier that a large part of the hadith corpus in fact rested on one dubious source.

In this article we turn the matter around. Not to deny that Abu Hurayra ﵁ transmitted many hadith. Al-Azami ﵀ cites in Studies in Early Hadith Literature the figure that Ibn al-Jawzi attributed to him: about five thousand three hundred and seventy-four transmissions. In the Musnad of Ahmad ibn Hanbal ﵀, 3848 stand in his name; after the deduplication of Ahmad Shakir, 1579 distinct narratives remain. We turn the matter around by asking one question his attackers never asked honestly. How many people heard directly from Abu Hurayra ﵁. And how many chains lead, for one and the same hadith, back to his mouth along different students.

The answer of the classical hadith scholars has been preserved in the rijal literature, the biographical science of the transmitters. About eight hundred. The classical counters, among them Baqi ibn Makhlad in his famous tafarrud survey and after him al-Bukhari ﵀, Ibn Hajar ﵀ and ad-Dhahabi ﵀, counted around eight hundred people who transmitted directly from Abu Hurayra ﵁, spread across Medina, Mecca, Sana’a, Kufa, Basra, Damascus, Egypt. No other Companion ﵁ in history has been heard and recorded by so many different people, in so many different places, in so many different tongues, and independently of one another. Whoever attacks Abu Hurayra ﵁ attacks no lonely voice. He attacks a choir.

A Yemenite boy in the Suffa

Abu Hurayra’s ﵁ story begins not in Medina, but in the southwestern hills of Yemen, in the tribe of Daws. His birth name has been transmitted variously by the biographical sources; ad-Dhahabi ﵀ lists some thirty variants, the most probable being 'Abd ar-Rahman ibn Sakhr al-Dawsi. The man said in his own time that he had felt his real name in the jahiliyya as a heavy thing to carry, and only under Islam experienced the beauty of a name with 'Abd ar-Rahman in it. The variation in the sources reflects the unimportance of the name problem; the man was without exception called by his kunya, and that kunya was Abu Hurayra. The father of the little kitten.

The story of that kunya has been handed down in all sobriety. As a boy he had a kitten he carried in his sleeve while tending his family’s sheep. Once he arrived in Medina, the Prophet ﷺ one day saw the young man with a cat on his arm and addressed him with the words “ya Aba Hurayra”. The name stuck. Simple, warm, and without any trace of majesty. A cat’s name for a man who would later, after the death of the Prophet ﷺ, become the widest mouth for the transmissions of his master.

He arrived late. In the seventh year after the Hijra, after the taking of Khaybar, somewhere in 628 or early 629 of the Christian calendar. The Prophet ﷺ was then already around sixty, his message had left the toiling-and-sowing stage of the first years in Medina far behind, and the community in Medina had grown into a well-defined state with legal rulings, military campaigns, and public administration. Abu Hurayra ﵁ got exactly three years with the Prophet ﷺ. No more.

Three years for five thousand hadith. It seems mathematically absurd, and it is precisely that point that the attackers of Abu Hurayra ﵁ have always played. Yet their arithmetic rests on an assumption that any serious study of the sources refutes: that Abu Hurayra ﵁ would only have passed on what he had personally heard from the mouth of the Prophet ﷺ, and only at the moments when he was present. Both premises are false.

First the second one. Abu Hurayra ﵁ was not an ordinary Companion ﵁ who conducted his trade by day or worked his fields and appeared in the Mosque only at prayer times. He was a man of Ahl as-Suffa. A term that literally translated means “the people of the shelter”, and that refers to a group of Companions ﵃ without means who literally lived in the Mosque of the Prophet ﷺ, in a spot behind the prayer ground covered by palm branches, sleeping on the ground, in the clothing they wore. They had no house, no wife, no trade. What they had was time and proximity. Two things that in the transmission of hadith are unmatched in value.

Abu Hurayra ﵁ himself, in a transmission recorded by Imam al-Bukhari ﵀ in his Sahih, put forward this position by way of explanation. When Marwan ibn al-Hakam, or one of the other critics of the Umayyad era, suggested that he transmitted too much, he answered roughly thus:

In one answer the whole attack is dismantled. Abu Hurayra ﵁ was, in modern terms, a full-time student. The Muhajirun had to resume Meccan trade in a new city. The Ansar had to maintain the fields of Medina. Both groups came to the Prophet ﷺ when their work allowed it, and went home again when duty called. Abu Hurayra ﵁ went nowhere. He slept in the Suffa, he ate from the hand of the Prophet ﷺ, he sat in his circle from sunrise until the last evening prayer. For three years. Almost without interruption.

Whoever occupies such a position does not need to be personally, bodily present at every verse and every saying in order to know it. For, and here comes the second correction, the Prophetic community was no isolated collection of individuals. It was a living network in which every saying of the Prophet ﷺ was passed on among dozens of Companions ﵃ within hours, sometimes minutes. When the Prophet ﷺ pronounced a hukm one morning in a particular majlis, that hukm had been made known to the absent through the present before the midday prayer. Abu Hurayra ﵁ could thus have been present at the oral repetition by Abu Bakr ﵁, by Umar ﵁, by Ubayy ﵁, and pass on that repetition with as much authority as if he had heard the saying from the Prophet ﷺ himself. In the science of hadith this is called a mursal Sahabi transmission, and it counts unanimously as valid. No classical scholar has ever maintained that a Companion ﵁ might only transmit what he had heard from the Prophet ﷺ with his own ears.

With those two corrections the so-called “Abu Hurayra problem” shrinks to a shadow. Three years of full-time presence in a network where every Prophetic saying circulated within hours, combined with a lifelong devotion to passing on what he had received and supplemented in those years. Five thousand transmissions is then no longer an impossibility, but the expected result of a teaching career of forty years in Medina, fed by three years of proximity and supplemented by what he took in from his direct fellow Companions ﵃. A normal outcome for a man who spent his whole life on preserving and passing on.

The wave of attacks, old and new

Yet it is precisely at this man that the orientalist and modern sceptical literature has aimed its heaviest artillery. We must name this honestly, not to give the attackers the stage, but to understand why that classical count of eight hundred transmitters is so important. A dam only becomes visible when one knows the fury of the river.

The Hungarian orientalist Ignaz Goldziher published in 1890 the second volume of his Muhammedanische Studien. It was accepted in the Western university world almost immediately as the standard work on hadith, and remained so in academic circles far into the twentieth century. Goldziher’s thesis was that the great majority of what stands in the canonical hadith collections are not authentic sayings of the Prophet ﷺ, but inventions of the second century AH, projected onto early Companions ﵃ to lend the inventions ecclesiastical authority. In that story Abu Hurayra ﵁ played a specific role. He was, according to Goldziher, the favourite “peg” for later forgers, because his name was known early on as the source of a large quantity of hadith. One extra hadith in his name did not stand out among the thousands of others.

Goldziher’s methodology, and that of his successor Joseph Schacht in The Origins of Muhammadan Jurisprudence of 1950, was in essence inverted. They did not take the sources as their starting point and analyse them, but took a presupposition, that most hadith was forged, and searched the sources for confirmation. When a hadith was transmitted through several chains, that was in their reading no indication of authenticity but of a well-organised forgery machine. When a hadith was transmitted through one chain, that was an indication of arbitrary invention. No pattern could refute their thesis, because every pattern was counted in advance as evidence for the thesis.

Once established in Europe, Goldziher’s view also seeped into parts of the Egyptian intellectual world of the twentieth century. Mahmud Abu Rayyah published in 1958 a work entitled Adwa’ ‘ala as-Sunnah al-Muhammadiyya, in which he pointed to Abu Hurayra ﵁ personally as the chief suspect of mass fabrication. The thesis was taken over almost word for word from Goldziher, only in Arabic dress. The book caused a shockwave in Muslim circles and was refuted with great care by a whole series of scholars, among them Mustafa as-Siba’i in As-Sunnah wa Makanatuha and Muhammad Abu Shahbah in Difa’ 'an as-Sunnah. But the pamphlet of Abu Rayyah, in its fleeting form, reached more readers than the solid refutations, and the story of “Abu Hurayra the liar” lives on in certain modernist corners to this day.

The attack always rests on three pillars, which we had better name so that we can refute them calmly.

First pillar: the figure. Abu Hurayra ﵁ transmitted so many hadith that it would be mathematically impossible for him really to have heard them all from the Prophet ﷺ. We have already undone this argument above. With three years of full-time presence in a network where every Prophetic saying circulated within hours, and with forty years of teaching afterwards, the figure is not only possible, it is to be expected.

Second pillar: the relative lateness. He arrived only in 7 AH, three years before the death of the Prophet ﷺ. How then could he transmit more than Companions ﵃ who had known the Prophet ﷺ since Mecca, such as Abu Bakr ﵁ or Ibn Mas’ud ﵁. The answer, treated by Ibn Hajar ﵀ in al-Isaba, is that it was precisely the earliest Companions ﵃ who after the death of the Prophet ﷺ were flooded with administrative and military responsibilities. Abu Bakr ﵁ and Umar ﵁ were Caliphs; they had no time for systematic hadith teaching. Their transmissions are scarce not because they knew little, but because their public duties halted their oral collecting. Abu Hurayra ﵁, by contrast, did virtually nothing after 11 AH for forty years except teach.

Third pillar: Aisha’s ﵂ correction. The attackers quote with relish the transmissions in which Aisha ﵂ corrects Abu Hurayra ﵁ or suggests that he transmits too much. They forget, however, to quote the context. Aisha ﵂ corrected all Companions ﵃ when she believed they had remembered something wrongly; she was known for that and her authority was acknowledged for it. Her correcting of Abu Hurayra ﵁ was no indictment but collegial oversight. Moreover, and this is the crucial point, Abu Hurayra ﵁ answered her corrections always with the same goodwill. In the transmission in Imam Ahmad ﵀ in his Musnad he says in one of those encounters:

No defence, no reproach, no attempt to cover himself before her. Only a reminder of the simple cause of his broader memory. The Prophet ﷺ himself had pointed him to this gift in the same period, and recorded in Sahih al-Bukhari ﵀ stands the scene in which the Prophet ﷺ taught him that he could hold what he took in by spreading out a cloth, without ever forgetting. A prayer for memory, a presumed extension of his capacity, and a promise. Nothing that has anything to do with fabrication or weakness of opinion.

The attacks were in essence a mirror effect. The attackers projected their own distrust, their own twentieth-century secularisation, onto a man of the seventh century who was unanimously held reliable by his own contemporaries. No Companion ﵁, no tabi’i ﵀, no classical imam ever accused Abu Hurayra ﵁ of fabrication. Those who did do so did it twelve hundred years later, in a language and a world the man had never spoken or known.

But the real answer to all the attacks lies not in argumentation. It lies in the figures. In the names. In the geography. In the immense network of people who independently of one another drank from his mouth and wrote down his words in a dozen cities. For that we must now, with the classical rijal literature in hand, take a walk through the world of the Tabi’in.

Eight hundred transmitters, six provinces

The figure “about eight hundred” is no rhetoric. It is a classical count. The biographical science of the transmitters, the rijal literature, has preserved from generation to generation the names of those who transmitted directly from Abu Hurayra ﵁. Baqi ibn Makhlad recorded in his tafarrud survey how many unique transmissions each Companion ﵁ carried, and in the same tradition the transmitters of Abu Hurayra ﵁ were counted at around eight hundred. Later biographical reference works, such as Tahdhib al-Kamal of al-Mizzi and Tahdhib at-Tahdhib of Ibn Hajar ﵀, have recorded those names each with their own birth and death dates, their own geographical stations, their own further students and their own share in the canonical collections.

Eight hundred. Let us allow this figure to sink in for a moment. In a time without a printing press, without mass communication, without universities in the modern sense, one man passed on directly, in his lifetime, to about eight hundred expert transmitters, each preserved with name, station and life dates in the biographical science. Such breadth, so minutely registered, has scarcely a second example in the history of oral teaching.

Who were those people. Where did they live. Here the conversation about Abu Hurayra ﵁ becomes vivid, because the transmitters did not sit in one circle around his feet. They streamed out from Medina across the whole known Muslim world of the first century. What follows is a walk past some of the best-known names from the classical rijal tradition. Not each of them appears in al-Azami’s ﵀ own list; that list, to which we return shortly, counts the people who preserved his hadith in writing. The names below come from the broader biographical tradition of famous students.

In Medina itself was his core group. There sat his closest students, who sometimes followed him for decades. One of the best known is 'Abd ar-Rahman ibn Hurmuz al-A’raj ﵀, born around 40 AH and died in 117 AH, a tabi’i who is mentioned in the sources as “transmitting directly from Abu Hurayra ﵁”. It is through him that the famed Muwatta of Imam Malik ﵀ received a considerable part of its Abu Hurayra hadith. Another Medinan student, al-A’la’ ibn 'Abd ar-Rahman ﵀, born around 50 and died in 139 AH, had according to the sources several nusakh of hadith that were famous in Medina, and he used to tell copyists that they must copy his book in its entirety or not at all. Habib ibn Salim al-Ansari ﵀, Muhammad ibn Ziyad al-Qurashi ﵀ from the tabaqa of about 35 to 120 AH, 'Awn ibn 'Abd Allah ibn 'Utba ﵀ from the generation up to about 110 AH; all these names appear in the classical rijal literature with the standard notation “transmitting from Abu Hurayra ﵁”. Each of them in turn had his own circle of students, sometimes dozens of men, who in their turn passed it on again.

In Mecca Abu Hurayra ﵁ likewise had students. Although he himself lived in Medina, he came to Mecca regularly for Hajj and Umrah, and when Meccan students arrived in Medina in the season of the pilgrimage, they sat with him too. A key name here is ‘Ata’ ibn Abi Rabah ﵀, born in 27 and died in 117 AH, the great mufti of Mecca in his generation, of whom the classical sources record that he even encouraged his students to write down their transmissions and provided them with ink and paper for it. Through ‘Ata’ ﵀ a considerable part of Abu Hurayra’s ﵁ corpus flowed to the Meccan school, and from there via Sufyan ibn 'Uyaynah ﵀ and al-Shafi’i ﵀ to the wider Islamic world.

In Yemen, in the old capital Sana’a, sat the most crucial of all Abu Hurayra students for the story that will follow later in this series in the article on the Sahifa of Hammam. Hammam ibn Munabbih ﵀, brother of Wahb ibn Munabbih ﵀, a tabi’i who was born around 40 AH and died in the first half of the second century AH. He wrote down so many hadith from Abu Hurayra ﵁ that his collection, the so-called Sahifa of Hammam ibn Munabbih ﵀, carried a name of its own in the transmission and has been preserved as a complete manuscript into the twentieth century. We return to that sahifa at length in the article on the Sahifa of Hammam.

In Kufa sat the students who, through the presence of Ali ﵁, Ibn Mas’ud ﵁ and the whole migration of prominent Companions ﵃ from Medina in the time of Umar ﵁ and Uthman ﵁, had built up a school of their own. There too Abu Hurayra ﵁ was not passed over. When Kufan students such as Sa’id ibn Jubayr al-Asdi ﵀, born around 46 and died in 95 AH, came to Medina, they sat with Abu Hurayra ﵁ in the same circle in which they also sat with Ibn 'Abbas ﵁ and Ibn 'Umar ﵁. Sa’id ﵀ would later compile a commentary on the Quran on the basis of what he had taken in at Medina, and in that commentary Abu Hurayra transmissions appear in several places.

In Basra, the other great city of Iraq, the pillar was al-Hasan al-Basri ﵀, born in 21 and died in 110 AH. He transmitted, according to the classical sources, from very many Companions ﵃ and possessed many books, among them a Quran commentary that he dictated to his pupils. Whether al-Hasan ﵀ also heard Abu Hurayra ﵁ personally is, incidentally, disputed among the classical scholars; some consider his transmission from Abu Hurayra ﵁ mursal, that is, through an intermediate link. Either way, al-Hasan ﵀ fed, through his own students, such as Qatada ibn Di’ama ﵀, Hisham ad-Dastawa’i and countless others, a gigantic second and third generation of Abu Hurayra transmissions. A Basran side branch through Muhammad ibn Sirin ﵀, born in 33 and died in 110 AH, supplied yet another separate stream, and his name does indeed appear in al-Azami’s ﵀ own list of those who preserved the hadith of Abu Hurayra ﵁ in writing. Ibn Sirin ﵀ was known for his caution and his refusal to tolerate loose books in his house without checking, and precisely for that reason his chain carries extra weight in authority. Even a man like him, with the strictest views on what might circulate in writing, transmitted hadith of Abu Hurayra ﵁.

In Damascus and Sham we find the third great stream. Makhul al-Shami ﵀, born around 55 and died in 118 AH, compiled according to Ibn al-Nadim a Kitab as-Sunan and a Kitab al-Masa’il fi al-Fiqh, and in both works appeared Abu Hurayra transmissions he had received through different chains. Rabi’ah ibn Yazid ad-Dimashqi and the whole circle around Imam al-Awza’i ﵀ in the second century AH would build on this stream.

In Egypt a student group of its own likewise arose. Abu Hurayra ﵁ himself was in Medina, where he served for a time as deputy governor under Marwan ibn al-Hakam, but his transmissions reached the valley of the Nile early. The sources record that many Egyptian tabi’in took in his hadith, whether by sitting with him in Medina during the Hajj, or through the first generation that moved from the Hijaz to al-Fustat, and their transmissions form the core of what would later resound in the circle around al-Layth ibn Sa’d ﵀ and Imam ash-Shafi’i ﵀ in his Egyptian period.

Six provinces, then. Medina, Mecca, Yemen, Iraq with Kufa and Basra together, Sham, and Egypt. Not three or four men who sat in one room with the master and monopolised his transmissions. Around eight hundred registered transmitters spread across the whole territory in which Islam was alive in the second half of the first century AH. Who could ever have organised such a network as a deliberate lie.

The law of the multiple isnad

But this is not even the strongest argument. We have so far spoken of Abu Hurayra ﵁ as source, and of his students as channels. The real wonder of the hadith system lies in what happened afterwards: how one and the same saying of the Prophet ﷺ, after being passed on through Abu Hurayra ﵁ to ten different students, was then passed on through those ten students to hundreds of further transmitters, and finally came together in the canonical collections through dozens of independent chains.

A simple example can illuminate this. Take any well-known hadith of Abu Hurayra ﵁, for instance the saying “Deeds are according to intention” or the transmission about the seven under the shade of Allah ﷻ on the Day of Judgment. Whoever looks up the takhrij, that is, the source comparison, of such hadith in a work like Tuhfat al-Ashraf of al-Mizzi or in the modern digital databases, sees beneath the text ten, fifteen, sometimes twenty-five different isnads. All of them begin with Abu Hurayra ﵁. All of them run through a different student. Some via al-A’raj ﵀ in Medina, others via Hammam ﵀ in Sana’a, yet others via Abu Salama ibn 'Abd ar-Rahman ﵀, or via Sa’id ibn al-Musayyib ﵀, or via ‘Ata’ ﵀ in Mecca, or via Muhammad ibn Sirin ﵀ in Basra. And from there they split further.

Picture that vividly for a moment, as a purely illustrative calculation model. One hadith at one moment from the mouth of the Prophet ﷺ. Passed on to a handful of Companions ﵃, among them Abu Hurayra ﵁. Passed on by Abu Hurayra ﵁ to ten of his students, in ten different decades, in ten different gatherings, in sometimes ten different cities. Suppose each student passes the hadith on to twenty of his own students, and those twenty each to ten. In three generations a network of branchings would arise running into the hundreds, all independent of one another, all going back to the same Prophetic saying through the same Companion ﵁. The exact number does not matter; what matters is the order of magnitude that even a modest branching already yields.

When we then lay those chains against each other in the canonical collections, and all versions agree textually, the probability that we are dealing with a fabrication has become statistically negligible. No forger, however clever, can put the same lie into the mouths of two hundred unknown students in six different provinces at the same time without someone somewhere noticing that the different versions cannot have a common origin. The science of hadith calls the property such a hadith has mutawatir or mashhur, and those terms are not polemical but technical. They describe a degree of transmission in which doubt is methodologically impossible.

Al-Azami sets out the same phenomenon in modern text-scientific terms. When a text occurs through several independent manuscript chains, in different geographical regions, in different manuscripts, in a nearly identical textual form, then the most economical explanation for that agreement is that the text really existed in that form at one point in time and spread from there along all the chains. This is, ironically, exactly the same methodology applied in Western textual criticism to classical Greek and Latin works. When Cicero occurs in thirteen independent manuscripts in nearly the same form, we accept the text as authentically Ciceronian. When a hadith of Abu Hurayra ﵁ occurs in two hundred independent chains in nearly the same form, the same acceptance should follow according to the same methodology. What Goldziher did was deliberately not apply his own text-critical standard for the Bible and the classics to the hadith. That is not scholarship; that is selective use.

The irony of the orientalist attack is thus that it attacks precisely that one Companion ﵁ for whom the counter-evidence is most overwhelming. An attack on a Companion ﵁ who was heard by only one or two students would at least have a semblance of plausibility. An attack on Abu Hurayra ﵁, whose corpus was passed on independently by around eight hundred transmitters in six provinces, is as if one claimed that the Roman Senate never existed, while we cite hundreds of thousands of inscriptions in hundreds of thousands of places at once.

The test of Marwan

There is in the old biographical literature one scene that condenses this whole question of Abu Hurayra’s ﵁ reliability into a single tableau, and which we cannot leave undiscussed. It is the scene of the test by Marwan ibn al-Hakam, the young Umayyad statesman who later, after his cousin Mu’awiya ﵁, would himself become Caliph, but who in the years when Abu Hurayra ﵁ taught in Medina was governor of Medina on behalf of his cousin. Marwan ibn al-Hakam was not a man of great trust among the Companions ﵃; he had had run-ins with several senior Sahaba in his younger years, and his name is not mentioned in the sources with an honorific. But precisely for that reason his test is all the more eloquent.

The transmission, given by al-Khatib al-Baghdadi ﵀ in Taqyid al-‘Ilm and ad-Dhahabi ﵀ in Siyar a’lam an-nubala’, goes roughly as follows. Marwan was sceptical about the enormous quantity of hadith Abu Hurayra ﵁ was passing on in Medina. He suspected that the man was inventing his transmissions on the spot, or at least passing them on with varying wording from one occasion to the next, which in the strict hadith doctrine would be an indication of unreliability. To test that, he devised a ruse. He had Abu Hurayra ﵁ come to him one evening and asked him to dictate a number of hadith. Behind a curtain sat one of Marwan’s scribes with ink and parchment, and noted everything the old Companion ﵁ said, word for word. Abu Hurayra ﵁ knew nothing of it and simply spoke as always.

A year passes, a new occasion is awaited. Marwan has Abu Hurayra ﵁ invited again, to dictate the same series of hadith as at the first session. Again a scribe sits behind a curtain, again Abu Hurayra ﵁ knows nothing. He dictates. Marwan then lays the two records side by side.

The whole matter sits in that one sentence. A man who lets a year pass between two dictation sessions, and in both sessions gives the same transmission word for word, cannot possibly be inventing his material on the spot. Whoever invents on the spot, varies. Whoever repeats word for word identically across the distance of a year has learned the material, memorised it, and preserved it tenaciously. Marwan himself, no friend of Abu Hurayra ﵁ in the political relations, had to concede afterwards that he had been mistaken in his scepticism.

The value of this episode is great, for it does not come from the pen of Abu Hurayra’s ﵁ admirers. It comes from the paperwork of his powerful critic. Marwan’s investigation had yielded nothing, and the honesty of the sources lies in the fact that they preserved that result and did not suppress it. A forging generation would have smuggled the whole test away. The Islamic transmission culture, sharp to the point of discomfort, has preserved even the test of Abu Hurayra’s ﵁ critics and passed it on to us, because in that test itself lies enclosed the confirmation of his reliability.

We have already seen the same methodology with the commission of Uthman ﵁ in the Quran cycle. Uthman’s ﵁ critics claimed that he had removed verses, and the Islamic sources did not suppress that accusation, but reported it and refuted it by pointing to its impossibility in a community where hundreds of living eyewitnesses sat. Here, in the case of Abu Hurayra ﵁, the methodology does the same. The attack is not suppressed; the attack is written down, with the outcome that contradicts the attack.

Catalogue versus chains: an exercise in arithmetic

Let us, to close the substantive discussion, do one arithmetical exercise that places the attention in the right proportions. Suppose a present-day reader comes by chance upon a hadith of Abu Hurayra ﵁, for instance the well-known saying: “When the Mu’azzin has pronounced the iqama, do not run to the prayer”. He wants to know through which chains this hadith has come to us.

The takhrij in Tuhfat al-Ashraf and in modern works such as Jam’ al-Fawa’id teaches him that this hadith occurs in several canonical collections, among them Sahih al-Bukhari, Sahih Muslim, Muwatta Malik and Musnad Ahmad. In those collections together, several different isnads are cited. All beginning with Abu Hurayra ﵁. But they split already at the second generation: some chains run through Abu Salama ibn 'Abd ar-Rahman ﵀, others through Sa’id ibn al-Musayyib ﵀, yet others through al-A’raj ﵀ or through Hammam ﵀ or through another direct student. And at each of these second-generation nodes the chains split again. Thus a single hadith, even before it reaches Imam Malik ﵀ or Sufyan ﵀, fans out into a considerable number of independent transmission lines.

And this is a rather ordinary hadith of Abu Hurayra ﵁, not even one of the most cited. Whoever wished to apply this exercise systematically to his whole corpus sees that the total number of transmission lines coming back to his name exceeds every practical count by far. An unsurveyable web of chains, returning to one Companion ﵁ in Medina, with at every node a registered transmitter whose reliability has been catalogued in the biographical works. The precise number is unknown and also does not matter; what matters is the scale.

It is exactly this that makes the science of hadith so solitary a phenomenon in human history. No text from antiquity, no text from the Middle Ages, no text from the modern age has come to us on this scale along so many independent chains and so many verified transmitters. Whoever attacks the hadith on grounds of scepticism should be consistent and also declare Plato, Aristotle, Cicero, the gospels and the whole of classical literature unreliable on the basis of much weaker transmission data. No serious scholar does that. And so no serious scholar should be able to dismiss a man like Abu Hurayra ﵁ as a fabricator.

The principle of preservation: from mouth to reed pen

Here, at the end of the article, we come to the transition to the next one. We have seen that Abu Hurayra ﵁ taught in Medina, that around eight hundred transmitters heard from him directly, that his transmissions have come to us through dozens of independent chains, and that even the most sceptical critics of his time, after testing, found no reason for distrust. But there was something else happening in the same decades, something that added a second layer of safeguard to the oral transmission.

A number of Abu Hurayra’s ﵁ students wrote it down. Not as improvised notes, not as a memory aid for personal use, but as a systematic collection, hadith after hadith, in a continuous manuscript that circulated in the transmission under a name of its own. The Arabic term for such a manuscript is sahifa, in the plural suhuf, and we have already met that word in earlier articles: with the Sahifa of Ali ﵁, and with the commission of Uthman ﵁ in the Quran cycle.

The most famous and historically most important of all the suhuf of Abu Hurayra transmissions is that of Hammam ibn Munabbih ﵀, the Yemenite tabi’i from Sana’a whom we already named above as one of his students. Hammam ﵀ wrote down 138 hadith directly from Abu Hurayra ﵁ in a sahifa that he spread to his own students. That sahifa has, in a wonder of manuscript continuity whose scale was only truly measured in the twentieth century, come down to us in two independent manuscripts. One in Damascus, one in Berlin. And when one looks up those same 138 hadith in the Musnad of Imam Ahmad ﵀ and in the Sahihayn, they turn out to return there, with the right minimal variations explained by the multiple isnad, in nearly the same wording.

In other words: the sahifa of Hammam ﵀ from the first century AH, written down from the mouth of Abu Hurayra ﵁ in Medina or in Sana’a, can be found again a hundred and eighty years later almost word for word in a Musnad from Baghdad. That is a textual continuity for which modern philology knows no analogy. The critical edition of this Hammam sahifa, compiled from the two manuscripts in Damascus and Berlin, is the work of Muhammad Hamidullah in the twentieth century. Al-Azami, for his part, published in the second part of Studies in Early Hadith Literature another early collection from the circle of Abu Hurayra ﵁: the Nuskhah of Suhail ibn Abi Salih, forty-eight transmissions recorded via his father Abu Salih al-Samman from Abu Hurayra ﵁. We return to the Hammam sahifa at length in the article on the Sahifa of Hammam, with the history of the two manuscripts, the comparison with the canonical collections, and what this means for our view of the transmission of the hadith as a whole.

For now it suffices to point to what al-Azami quietly and matter-of-factly recorded in Studies in Early Hadith Literature. In his survey of who did not only pass on the hadith of Abu Hurayra ﵁ orally but also put it into writing or possessed it in writing, stands a short, well-defined row of names: Hasan ibn 'Amr ad-Damari, Bashir ibn Nahik, 'Abd al-'Aziz ibn Marwan, Abu Salih as-Samman, Hammam ibn Munabbih ﵀, Marwan ibn al-Hakam, Muhammad ibn Sirin ﵀, Sa’id al-Maqburi ﵀, 'Ubaid Allah at-Taimi and ‘Aqbah ibn Abi Hasna’. Ten men who supplied the proof in ink. The Sahifa of Hammam ﵀ is thus not an exception that confirms the rule. It is the best-preserved example of what in the time of Abu Hurayra ﵁ was done by a whole series of his transmitters. The principle of preservation was active. Memory was no loose mud blowing away in the wind. It was taken firmly in hand, on reed pen and parchment, in Medina, in Sana’a, in Mecca, in Basra, in Kufa, in Damascus, in al-Fustat. And from those written nuclei grew, in the second and third centuries AH, the great canonical literature that we hold in our hands to this day.

Whoever attacks Abu Hurayra ﵁, then, attacks not one man. He attacks an eight-hundred-voice choir, which across a territory as large as the Roman Empire in its heyday independently sang the same melody, and in many places at once fixed that same melody in ink. It is no scholarship to wave that symphony away as unreliable. It is a compelling spiritual choice, made before the evidence was consulted. And like every choice that precedes the evidence, it unmasks not the source, but the critic himself.

In the next article we leave the circle of Abu Hurayra ﵁ and follow a third great stream of transmission, that of Abdullah ibn Abbas ﵁, the cousin of the Prophet ﷺ who in the sources was called the Sea of Knowledge. We see how this young Meccan, after the death of the Prophet ﷺ, went from door to door to question the Companions ﵃, how he had what he heard written down into books by camel-loads, and how he left behind in Mecca a school whose students would carry the tafsir and hadith science of the second century. Once again a written tradition, once again a network of independent chains, this time fanning out from Mecca to Khorasan.