The seven ahruf: Umar ﵁ and Hisham ﵁ at the mosque

One skeleton, many mouths

In the previous article we watched the scribes of the Prophet ﷺ at their work, from parchment to shoulder blade to palm stem, until by the end of his life the revelation lay in written form scattered across dozens of houses of Medina. We saw how every verse was dictated, placed and read back, and how in this way a single written skeleton-text arose that had been fixed by dozens of hands at once.

That written skeleton is a certainty which in this article we must widen, not narrow. For alongside that one written skeleton-text there lived, from those same first years in Medina, several authorized oral recitations of that same Quran side by side. Not as different books. Not as rival editions. But as different ways, licensed by the Prophet ﷺ himself, of placing the same verse in the mouth. The principle under which they stood carried a name that the umma has preserved in its memory from the Prophetic age onward. It was called the seven ahruf.

In this article we return to Medina, to a morning in the mosque on which a man named Umar ibn al-Khattab ﵁ hears another man recite in a form he does not recognize. The scene itself is short, but the conclusions it compels carry the whole rest of the textual history of the Quran. Without the ahruf-licence the revelation would have remained bound to a single dialectal tongue, and no believer outside that tongue could have placed the Book in his own mouth without first disowning his speech. Without that licence there would also have been no Hudayfah ﵁ who decades later would come riding to Medina from the front with the cry of alarm that would lead to Uthman’s ﵁ standardization, to which we return later in this series, for there would then never have been a multiple recitation-practice to be alarmed about. What is licensed here is the space within which everything that comes afterward has played out, and it is a licence that did not rise from the mouth of a scholar or a caliph, but from the mouth of the Prophet ﷺ himself.

The scene in the mosque

The report comes from the mouth of Umar ﵁ himself. Al-Azami ﵀ records it in his eleventh chapter in the most compressed form: Umar ﵁ and Hisham ﵁ differed over a verse of Soera al-Furqan, Umar ﵁ asked Hisham ﵁ who had taught him that passage, and Hisham ﵁ answered: the Prophet ﷺ. That is all the chapter on the textual history of the matter requires, and for it al-Azami refers to al-Bukhari ﵀ in the section Fada’il al-Qur’an and to Muslim ﵀ in the section Salat al-Musafireen. The full, extended rendering of the scene, with the recitation in the prayer hall, the seizing of the cloak and the twofold confirmation by the Prophet ﷺ, does not stand in that compressed chapter. It lives in the broader classical tradition itself, and specifically in the longer version that al-Bukhari ﵀ recorded in his Sahih in Fada’il al-Qur’an, with parallel chains in the sections Khusumat and Bad’ al-Khalq, alongside the separate chain of Muslim ﵀. A hadith with this degree of multiple isnad-confirmation is of the highest calibre in the science of hadith. What follows in these paragraphs is that full tradition, drawn from its classical source, and what by it we know with certainty.

Umar ibn al-Khattab ﵁ was present in the mosque, still during the lifetime of the Prophet ﷺ, and although the report gives no exact date, it situates the scene unmistakably in the Medinan period, presumably in the later years, when Hisham ibn Hakim ﵁ was by then living in Medina as a believing Companion ﵁. It is precisely this that makes the scene so telling, for Hisham ﵁ was no obvious source of a strange recitation: he came from a well-to-do Quraish family, was the son of Hakim ibn Hizam ﵁ and had come to Islam at the conquest of Mecca, so that he did not come from some distant tribe or region but was as Quraishi in origin as Umar ﵁ himself. What Umar ﵁ heard that morning therefore did not come from a strange tongue that one might easily have mistrusted, but from the mouth of a man of his own city and his own tribe.

Umar ﵁ heard him recite. What Hisham ﵁ took into his mouth was Soera al-Furqan, the chapter that begins with a declaration about the Criterion that Allah ﷻ had sent down to His servant. Soera al-Furqan is one of the Meccan surahs of which Umar ﵁, in his long years in the presence of the Prophet ﷺ, had heard many recitations. He knew the surah. He knew the word order. He knew the sounds in which the Prophet ﷺ had recited it.

And precisely here came the shock, for Hisham ﵁ did not recite the surah in the form Umar ﵁ had heard from the Prophet ﷺ. In several places, so the report says, there occurred sounds and words that Umar ﵁ did not know, and these were no glaring deviation, no reversed meaning, no falsification of the content, but other modes, other vocalizations, another register, in a passage that Umar ﵁ had anchored in his memory in a different way.

Here the temperament of Umar ﵁ comes into view, for the Prophet ﷺ himself had once said that when the devil saw Umar ﵁ walking down one road, he chose another for safety’s sake. Umar ﵁ was not the man to nurse a question in silence, nor the man to acquiesce wordlessly to an unfamiliar variant and work out at home for himself how to place it. What he heard was for him no abstraction, but the Book of Allah ﷻ in the mouth of a Companion ﵁, in a form he did not know from the Prophet ﷺ, and that he did not let lie.

He waited until Hisham ﵁ was finished with his recitation, until the taslim had been pronounced and the prayer had been completed, and what he then did is recorded in the broader tradition with physical precision. He walked up to Hisham ﵁, seized him by the cloak that hung over his shoulder, and dragged him out of the mosque into the street. There was no quarrel in the prayer hall itself and no interruption of the gathered community, but there was an unmistakable, bodily arrest of one Companion ﵁ by another Companion ﵁, in full view of whoever was still departing.

Between the mosque and the house in which the Prophet ﷺ was, there lay only a few minutes’ walk. Umar ﵁ bridged that distance with Hisham ﵁ in his grip. The physical indignation of the man being led across the street, and the physical resolve of the man leading him, formed the shortest possible prelude to what followed on the other side of the threshold.

The Prophet ﷺ received them, and what was exchanged between those three men was later retold in detail by Umar ﵁ himself, because he did not want the umma to preserve the scene in a fleeting summary but in its precise wording. At his insistence the hadith entered the canonical collections of al-Bukhari ﵀ and Muslim ﵀ in almost verbatim form, and we render it here in the English wording that stays as close as possible to the Bukhari formulation.

At this fragment we must walk through a number of things slowly, for in its brevity it carries the whole theology of the seven ahruf.

The Prophet ﷺ takes no side in the sense that Umar ﵁ expected, for he does not interrogate Hisham ﵁ over an alleged error and he does not chide Umar ﵁ over his zeal either. He makes one single move: he asks both men to repeat their reading in his presence, first Hisham ﵁ and then Umar ﵁, and behind both recitations he sets precisely the same seal. Thus it was revealed. Not this is also permitted, not this is allowed within the margin, not this is current in the Quraishi tongue, but twice, in wholly identical wording: thus it was revealed. Both recitations come, by the testimony of the Prophet ﷺ himself, directly from the revelation, and neither of them stands one rung below the other.

Only afterward comes the explanation, and it is no academic definition and no list of seven dialects, but a simple statement of a divine concession with a practical instruction to close. This Quran was sent down in seven ahruf. So recite whatever of it is easy for you. Place in your mouth what you can place in your mouth, runs the sense, for the text was given spacious enough to suit different mouths.

Umar ﵁, by his own later testimony, understood at that moment something he had not understood before. He was, to approach his own words, a man who had heard the Book from the mouth of the Prophet ﷺ in one particular form and who had built his trust on that form. What the Prophet ﷺ showed him here was that trust in his own form did not mean the other form was untrustworthy. The revelation had more channels than his ear in his vicinity had caught. And those several channels were, all without exception, heavenly authorized.

What the seven ahruf are

Before we go further, we must first ask what the word ahruf actually means. It is the plural of the Arabic harf, which literally means letter or edge and in a wider reading denotes something like a manner of pronunciation or a modality, and the scholars have devoted, from the first centuries after the revelation, a whole chapter of their science to its precise scope. As-Suyuti ﵀ records in his al-Itqan that there exist roughly forty different scholarly positions on what the seven ahruf precisely denote.

Forty is a large number for the interpretation of a single hadith, and it deserves some attention that it ran so high. It shows first of all that the scholars did not proceed lightly, for a single obvious meaning would never have produced forty documented positions. It shows next that the matter itself was not simple, and that the Prophet ﷺ in his compressed statement left behind no conclusive technical definition, but gave a principle and left its further elaboration to the scholars after him. For the textual history of the Quran that is an important datum in itself, for it means that whoever today asserts firmly that he knows exactly which seven the seven ahruf are, claims more certainty than the classical tradition ever allowed itself.

We need not walk here through all forty views, but it is worth setting two of them against each other, so that the reader tastes how open the question truly is. Part of the scholars have argued that seven here should not be read merely as an arithmetical number, because the number seven in classical Arabic usage, besides its literal value, can also denote a rounded generous quantity, as when in Arabic one says seventy or seven hundred to express a multitude. In that reading the Prophet ﷺ did not necessarily point to precisely seven separate dialects in a closed enumeration, but to a divine concession that extends within a bounded but generous margin.

Another, equally classical line maintains that the number here was indeed a factual designation. In one well-known elaboration of that line, connected to the tradition of Abu Ubayd, seven Arabic dialect-zones are named, those of Quraish, of Hudhayl, of Thaqif, of Hawazin, of Kinanah, of Tamim and of Yemen, with the Quraishi form as basis because in it the revelation had originally sounded, and the remaining six as added concessions. We name this enumeration here expressly as one view among the forty, not as the settled meaning of the seven ahruf. Other scholars have named other sets of seven, still others have connected the number to seven kinds of variation instead of seven tribes, and a classical voice such as that of as-Suyuti ﵀ has precisely enumerated the whole series with care, without proclaiming any one of them decisive. The precise content of the seven remained contested, and that ought to remain visible in an honest treatment.

What however remains standing in every one of the forty interpretations, and what for the textual history is the only truly necessary point, can be summarized in one sentence: the seven ahruf are not seven different Qurans, but a bounded multiplicity of recitation-modes in which one and the same revelation could legitimately be recited. In content, in meaning and in law, there is no difference between the ahruf, and the variation lies solely in the linguistic processing, in a different vowel, a different synonym or a different verb form in which two Arabic tongues strive toward exactly the same thing. What the one tongue does with the classical example aqbil, the other does with ta’al, and both say come here without either of them saying anything other than the other. The Arabic body of language was spacious enough to carry two equivalent words for the same concept, and the ahruf-licence made both into equally legitimate bearers of the revelation.

A second distinction we must voice here already, because it comes to stand central in the continuation of this theme: the seven ahruf are not the same as the seven qira’at. The ahruf are a Prophetic, licensing principle, the spacious room within which several oral recitations of the revelation are legitimate, whereas the seven qira’at form a much later systematization, from the third and fourth centuries AH, in which the scholar Ibn Mujahid ﵀ and his successors identified within the Uthmanic skeleton seven canonical recitation-traditions, each connected to an eponymous scholar from centres such as Kufa, Basra, Mecca, Medina, Damascus and Hims. The qira’at therefore move within the Uthmanic skeleton, within one written form, whereas the ahruf extend over a broader space that in the pre-Uthmanic period was present in the oral transmission, and to that later codification of the seven canonical readings under Ibn Mujahid ﵀ we return at length later in this series.

Examples of ahruf-variation

To understand what we are speaking of here, we must for one moment come out of the abstract and ask ourselves what kind of variation actually fell under the ahruf-licence, and what Hisham ﵁ had had in his mouth that morning at which Umar ﵁ was so startled.

We do not possess a verbatim record of precisely which verses of Soera al-Furqan Hisham ﵁ recited that morning, nor an account of his deviating sounds, for the tradition preserves the principle and not the philological details of that one recitation. What we do have are other transmitted examples of ahruf-variation within verses that in the Mushaf itself already admit a multiple reading, and they show the nature of the phenomenon more clearly than any abstract description could.

The most cited example comes from Soera al-Fatiha, the verse that in every prayer is pronounced by every Muslim.

In this verse the fourth word, in its pronounced form, stands on the boundary between two equally ancient recitations both licensed by the Prophet ﷺ. The one recites malik, with the long a, and that means owner or possessor, whereas the other recites malik, with the short a, and that means king or sovereign. In the written skeleton form of the Uthmanic Mushaf both readings stand under the same row of consonants, for the skeleton in its earliest form contains no marking of the long vowel between m and l, and so the same three letters carry both words at once. Both forms were taught by the Prophet ﷺ, both were passed on by different Companions ﵃ in the transmission-chains, and both are represented to this very day in the canonical qira’at.

The relation of meaning between owner and king of the Day of Judgment is not that of a contradiction, for both are true: Allah ﷻ is on that Day both the Owner of its events and the King over its justice. The two ahruf are here not two statements that contradict each other, but two shadings of the same theological reality, each fully true and each with its own accent, and whoever in his prayer says the one misses nothing of what is enclosed in the other, just as little as whoever says the other misses anything of what lies in the one. The revelation is, to borrow al-Azami’s ﵀ own distinction, not variant-wise vulnerable but lexically rich.

Other examples run along comparable lines, and the classical works on the science of recitation have collected them abundantly. A verb form that in the one tongue was pronounced with a short a, could in a neighbouring tongue sound with a short i without the meaning shifting; a final sound that in the one dialect was inflected with imala, a slight leaning toward the i-sound, could in the other dialect be pronounced without that leaning; and a supplementary waw or fa at the beginning of a sentence could be present in the one recitation and not in the other, each time without the legal or theological sense shifting by it. On the same line lies the well-known double form in which one and the same question formula was carried by the one recitation in the third and by the other in the second person, again without the sense of the question changing, and in a number of verses the one recitation carried a synonym that in another was replaced by an equivalent synonym, with preservation of the full meaning.

Equally important here is what the ahruf are not. They are no different legal rulings for the same case, no different testimonies about the same historical event, no different depictions of Allah ﷻ or of His Prophets ﵈ or of the Law, but solely phonetic and lexical shadings within one theological whole. Whoever must choose between two ahruf therefore never chooses between truth and error, but always between two pronunciation-forms of the same truth.

For this reason al-Azami in his treatment deliberately avoided the word variant for the ahruf and chose the word multiple, for a variant suggests that there is one correct original reading and beside it a deviant, defective copy of it, whereas a multiple precisely suggests that there exist several, all equally correct readings at once within a space that the author of the text himself has licensed. For the Quran unmistakably the second holds and not the first, and in that one distinction of word lies the whole difference between a text that stands under suspicion and a text that overflows with richness.

The divine wisdom of the concession

Why would Allah ﷻ send the Quran down in several ahruf and not in one single dialectal form, and why would a Book that claims universal truth allow for its own recitation divergent pronunciation-forms instead of fixing itself in one polished norm. The answer comes not from the scholars after the fact, but from the tradition itself, in a second hadith that rests not with Umar ﵁ but with Ubayy ibn Ka’b ﵁, the man who later in this series, at the cry of alarm of Hudayfah ﵁ and Uthman’s ﵁ standardization, comes forward as the master of recitation in Syria.

Ubayy ibn Ka’b ﵁ passed on more than one report about the ahruf-concession, and the classical sources preserve them separately. In the version that Muslim ﵀ recorded in his Sahih, in the section Kitab as-Salat, Jibreel ﵇ came to the Prophet ﷺ when he was in the vicinity of Banu Ghifar, and there unfolded the conversation we render below.

In a second report, likewise through Ubayy ﵁ but recorded by Ibn Hanbal ﵀ in his Musnad, the place of the conversation is another and the declaration of Jibreel ﵇ sounds still more literal. There the Prophet ﷺ met him at the mira’-stones, on the edge of Medina in the vicinity of Quba’, and there fell the sentence that captures the wisdom of the concession in one image: I have been sent to a nation of the illiterate, among whom the old prowling sheikh, the aged woman and the young child. So command them to recite the Quran in seven ahruf.

Herein lies the wisdom of the concession, for the revelation came down upon an Arab people that did not possess one uniform tongue, and although Quraish formed the centre, Quraish did not stand alone: outside Mecca there lived countless tribes whose mouths had been shaped by birth and custom toward other phonetic patterns. The tradition captures that multiplicity not in a list of tribes, but in three human faces, the prowling greybeard, the aged woman and the young child, and precisely that choice is telling. It is the worn-out man whose tongue could no longer revise its form over the years, it is the old woman whose speech habits had been ingrained a lifetime long, and it is the child that had already learned its first sounds in a local dialect; three people to whom a strict Quraishi norm would have made the Book nearly unpronounceable. The concession lifted the threshold away precisely for such mouths, without distinction by tribe or region.

What Jibreel ﵇ brings forward in this report is therefore no pragmatic compromise and no merely pedagogical accommodation, but a divinely revealed principle: the Quran is licensed in several modes, so that no Arab would experience the Book as unattainable. The revelation bends in part toward the ear of the hearer, so that the hearer may then bend with his whole life toward the revelation, and the ahruf are the threshold that lays the Book low enough for every house, in every tribe and in every region of Arabia.

This principle explains why the ahruf-licence came to full flowering not in the Meccan but in the Medinan phase. In Mecca the group of believers was small, geographically concentrated and to a high degree Quraishi in origin, so that the practical necessity for multiple ahruf remained limited. From the Hijra that changed profoundly, for in Medina the Aws and the Khazraj were added, two tribes of southern, Yemeni origin who had settled generations earlier in the Hijaz, and in the years after Badr and after the Conquest of Mecca soldiers and converts streamed in from virtually all Arab tribes. The ahruf-concession made it so that the Quran could be carried in every new mouth, without the hearer having first to disown his own tongue in order to be able to be a believer, and thus the unity of the text was preserved not despite the multiplicity of tongues, but precisely through a revelation that had taken that multiplicity into itself from the very start.

Why the standardization did not destroy the ahruf

Later in this series, at the cry of alarm of Hudayfah ﵁ and Uthman’s ﵁ standardization, it comes up in detail what happened to the ahruf decades after the Hisham-Umar scene. In the years after the Prophet ﷺ the umma spread across Syria, Iraq, Egypt and beyond. Every garrison city had its own teacher, and every teacher passed on in his local recitation-practice the ahruf as he had learned them. What in Medina under the care of the Prophet ﷺ had functioned as a generous concession, became in the provinces, far from his presence, a source of confusion for the second and third generation of soldiers. The Syrian who had his recitation from Ubayy ﵁, and the Iraqi who had his recitation from Ibn Mas’ud ﵁, found each other’s sounds suspect in the camp on the Kura river. Out of that conflict Hudayfah ﵁ would come riding to Medina with the message that would lead to the Uthmanic standardization.

The question we must answer here, with our new knowledge of what the ahruf were in origin, lies at hand: did the Uthmanic standardization abolish the ahruf-licence, and did Uthman ﵁, by imposing one Mushaf in the Quraishi skeleton upon the umma, narrow a space licensed by the Prophet ﷺ down to one single reading.

The answer, as al-Azami works it out in his eleventh chapter, asks for a careful distinction. The standardization fixed one layer of the ahruf-licence in written form, namely the consonantal skeleton-layer in which the overwhelming majority of the ahruf-variation could be taken up, and that skeleton form carries in its classical shape, stripped of dots and marks, several pronunciation-possibilities that all fall within the licence. The example of Soera al-Fatiha 1:4 that we gave above shows that directly, for the same three letters m-l-k carry both malik and malik, and an Uthmanic Mushaf in skeleton form excludes neither of the two readings but takes them both implicitly into itself.

In those places where the ahruf-variation touched not merely the pronunciation but the skeleton itself, the Uthmanic committee did have to make a choice, and al-Azami notes that the total number of places at which the official five Uthmanic Mushafs differ among themselves in skeleton does not amount to more than roughly forty characters. Forty characters, over the full extent of the Mushaf, and at each of those places the committee deliberately chose the Quraishi reading as the reference norm, in line with the instruction that Uthman ﵁ gave to Zayd ﵁ and the three Quraishi committee members and to which we return later in this series, at his standardization. When you differ with Zayd, then write in the dialect of Quraish, for in their tongue it was revealed.

But what the Uthmanic committee did not do, was declare the other ahruf-recitations invalid. It established a written reference, one skeleton form, for one standardized public use in a dispersed caliphate. It left the oral recitation-traditions, within that skeleton form, untouched. The Syrian pupils of Ubayy ﵁ could continue their pronunciation passed on by him, insofar as it fitted within the written skeleton. The Iraqi pupils of Ibn Mas’ud ﵁ could, with the same restriction, continue their transmitted pronunciation. What became impossible was the skeleton-exceeding variation, for the skeleton now stood fixed. What remained possible was the skeleton-internal variation, and it, carried by transmitted isnad-chains, was continued until in the third and fourth centuries AH it was systematized by Ibn Mujahid ﵀ into seven canonical qira’at.

In other words: Uthman ﵁ congealed one of the ahruf into a written, publicly issued reference form, and granted the other ahruf their oral survival within that skeleton. The licence that the Prophet ﷺ had declared to Umar ﵁ and Hisham ﵁ was not withdrawn. It was, for the practical administrative need of an extended empire, concentrated into a form that could be verified in all cities at once, with preservation of its oral breadth for whoever had received it in an authorized isnad-chain.

The principle that carries the textual history

For the textual history of the Quran what has been told in this article carries four heavy conclusions, and we walk through them here one by one, because precisely in their coherence the thesis of this whole map becomes visible: that controlled variation, licensed by the Prophet ﷺ, is not the weakness but the strength of the tradition.

In the first place the ahruf-licence explains why in the earliest recitation-traditions of the umma there existed differences among themselves without one Companion ﵁ accusing another of deceit. Ubayy ﵁ and Ibn Mas’ud ﵁ did not teach the same pronunciation in their respective cities, and they knew this of each other too. Neither of them held that the other had disfigured the Book. Both knew that they had recited as the Prophet ﷺ had taught them, and both knew that the Prophet ﷺ in his life had handed different pupils different ahruf. What in later centuries was presented by sceptics as proof of a chaotic pre-Uthmanic text-transmission, is in reality proof of a documented Prophetic concession.

In the second place the ahruf-licence explains the pace and the nature of the Uthmanic standardization, for Uthman ﵁ stood not before a problem of a text gone corrupt, but before a problem of a licensed space without a guide. Decades after the Prophet ﷺ, with the ahruf-teachers dispersed across the provinces and their pupils who no longer knew the broader context, the original generosity threatened to turn into mutual accusations, and Uthman ﵁ did not thereupon shrink the ahruf but issued a geographical guide in written form, so that in every city the same skeleton form could be consulted, while he left the oral recitation-tradition to the competent teachers. Thereby he did not abolish the Prophetic concession but made it governable within the caliphate.

In the third place the ahruf-licence refutes a sceptical argument that is sometimes advanced, namely that before the Uthmanic standardization there existed countless rival Qurans of which Uthman ﵁ chose one and suppressed the rest. That picture is correct at no single point, for what existed was not a collection of competing books but one Quran in several recitation-modes licensed by the Prophet ﷺ, and Uthman ﵁ did not choose one Quran out of a stack of Qurans but fixed one written skeleton within which those already licensed modes could survive. The rivalry that the sceptics presume is a projection after the fact upon a community that knew from within that it possessed one Book and differed over it only in its oral breadth.

In the fourth place the ahruf-licence gives the umma its pronunciation-certainty to this very day. For the seven canonical qira’at that today are recited in mosques from Morocco to Indonesia are not private projects of medieval scholars. They are fixed isnad-chains of preserved ahruf-recitations, each going back to a Companion ﵁ who had heard his pronunciation directly from the Prophet ﷺ, and each within the Uthmanic skeleton. Whoever today in Cairo recites the qira’a of Hafs on the authority of Asim, and his cousin in Marrakesh the qira’a of Warsh on the authority of Nafi’, does not say a different Book than the other. He says the same Book in a modality licensed by the Prophet ﷺ, carried by a documented chain that goes back to the revelation itself.

Zayd ibn Thabit ﵁, the same man who would collect the first Suhuf under Abu Bakr ﵁ and later in this series would lead the committee of Uthman ﵁, made about this whole matter a statement that in four words puts the whole matter in its place. Al-qira’atu sunnatun muttaba’atun. Recitation is a followed sunna. Not a personal taste, not a philological preference, not a dialectal habit. A sunna that was passed on by transmission, in every link connected to a receiver who had taken it from his own teacher, back to the Prophet ﷺ. What in the first centuries of the umma existed of recitation-certainty, did not exist by the grace of consensus or majority decision. It existed by the grace of isnad.

What comes hereafter

We have in this article fixed the principle of the seven ahruf: a divine concession to the divergent Arabic dialects, licensed by the Prophet ﷺ in the scene in which Umar ﵁ led Hisham ibn Hakim ﵁ by his mantle to the dwelling of the Prophet ﷺ, and in which the Prophet ﷺ confirmed both recitations of Soera al-Furqan with the same four words. Thus it was revealed. We have seen that the precise content of the seven was never reduced by the classical scholars to one settled list, that even forty views could stand side by side, and that precisely that openness betrays no weakness but the honesty of a tradition that claimed no more certainty than it possessed. What stood firm in every view was that the ahruf were no rival Qurans but ways licensed by the Prophet ﷺ himself to recite one and the same revelation, and that the variation between them was controlled, authoritative and without loss of meaning. We have finally seen how the Uthmanic standardization granted their oral survival within a standardized written skeleton without abolishing their licence, so that the richness was preserved while the unity was guarded.

The space that on that morning in Medina was opened between Umar ﵁, Hisham ﵁ and the Prophet ﷺ carries from that moment to our own century every reciting prayer of every Muslim, in whichever ahruf-tongue he or she places it in the mouth. So recite whatever of it is easy for you.

We have now seen that the revelation could be recited in several authorized recitation-ways, and that this oral multiplicity was licensed by the Prophet ﷺ himself. But beside that living, orally-textual richness stood the written side of the same young state, and it is precisely that side which can be proven with tangible objects. For the same community that recited the Quran in seven aḥruf possessed a chancery that produced dated state letters, sealed with the seal of the Prophet ﷺ and sent as far as the courts of emperors and governors. In the next article we leave the reciting mouth for a moment and follow the writing hand: the sealed diplomatic letters of the Prophet ﷺ, preserved down to our own time, as documentary evidence that early Islam did not merely speak but also wrote.