Islam: A Reader's Guide to the Qur'an, Hadith and Early Islamic History

A reader’s edition of a historical-critical review, written for anyone: Muslim, revert, skeptic, or simply curious. Nothing has been removed from the academic original. Everything has been explained.

Introduction: An Invitation

Let me tell you what this book is, and what it is not.

Somewhere around the year 850 of the common era — more than two centuries after the Prophet Muhammad died — a scholar named al-Bukhari was travelling across the Islamic world, collecting reports of what the Prophet had said and done. He would, according to tradition, examine hundreds of thousands of these reports. He would reject the overwhelming majority. The small fraction he kept became one of the most influential books in human history.

Stop and notice something strange about that sentence.

Two centuries.

Imagine that the earliest detailed accounts of Napoleon’s words and deeds were being written down, sorted, and authenticated only now — in our own lifetime — by scholars working from chains of oral testimony. “My teacher heard from his teacher, who heard from his grandfather, who heard from a man who served in the Grande Armée…”

Would you trust those accounts? Some of them? None of them? How would you even decide?

That is not a hostile question. It is the question. Muslims themselves asked it — within living memory of the Prophet, and in every century since. An entire science, one of the most sophisticated critical traditions in the pre-modern world, was built by Muslim scholars precisely to answer it. And in the last hundred and fifty years, modern historians — armed with manuscripts, coins, inscriptions, papyrus documents, and new analytical methods — have asked it again, and reached a bewildering variety of answers.

This book walks through the whole argument. All of it.

It is written for the reader who knows nothing in advance. Every technical term will be explained the moment it appears. Every historical figure will be introduced. Every piece of evidence will be held up to the light so you can see what it is, where it came from, why historians care about it — and, just as importantly, what it cannot tell us.

And it is written for every kind of reader at once. If you are a Muslim, nothing here asks you to abandon your faith; this is a book about what history can and cannot establish, and it is scrupulously careful about the difference. If you are a new revert trying to understand where your religion’s rules actually come from, this book is a map. If you are an atheist or a Christian or simply curious, you need no belief of any kind to follow the evidence.

This book will not tell you what to conclude. It will show you the evidence, show you where scholars agree, show you where they fight, and leave the verdict to you.

The one rule that governs everything

The original academic review behind this book was checked — deliberately, adversarially — against three hostile readers: a traditional Muslim scholar, a Quranist (we will meet the Quranists properly in Chapter 1), and a secular historian. The idea was that no single perspective’s blind spots should survive unchallenged. That structure is preserved here, and near the end of the book (Chapter 18) you will watch all three critics attack the book’s own arguments — and see which attacks land.

But the deepest discipline in this book is simpler. It is a set of labels.

Every claim you will read falls into one of these categories, and the single most common failure in popular writing about this subject — on every side — is smearing these categories together:

Keep these labels in your pocket. You will use them on every page. When someone tells you “the Qur’an’s text was stable very early,” that claim belongs to a completely different universe than “the Prophet intended his sayings to be binding law forever.” The first can be tested against physical parchment. The second is a matter of faith and interpretation. Both may be important. They are not the same kind of important.

Come with me, then. Let’s examine the evidence together.

---

Editor's Note

Revert Way publishes this work in two editions. This is the Reader's Edition — the same evidence, the same arguments, and the same conclusions as the companion Academic Edition, re-expressed in plain language with every technical term and historical figure introduced on first use. Nothing has been removed and no evidence has been softened; the difference between the two editions is register, not content. Readers who want the full citation apparatus — precise page numbers, technical terminology, formal academic references — should read the Academic Edition linked above. Readers who want the same material without that apparatus should stay here.

Chapter 1 — What Exactly Are We Arguing About?

A word you need first: Hadith

Start with the word itself.

In ordinary Arabic, hadith simply means “speech,” “report,” “a thing said,” “news,” “a story.” Hold on to that — it will matter enormously later, in Chapter 5D, when we look at how the Qur’an itself uses the word.

Each of these reports comes in two parts, and the two parts have technical names you will meet throughout this book:

Muslim scholars would eventually build an entire critical science around interrogating both parts. Was every link in the chain a real person? Did each one actually meet the next? Was each one honest? Reliable of memory? And does the content itself make sense — does it contradict the Qur’an, or better-attested reports?

The collections

Over Islam’s first three centuries, these reports were gathered into great written collections. And here is where the story immediately becomes more interesting, because each major branch of Islam ended up with its own canon.

Sunni Islam — today the largest branch by far — reveres what became known as “the six books.” Their compilers, with their dates of death, were: al-Bukhari (died 256 AH / 870 CE), Muslim (d. 261/875), Abu Dawud (d. 275/889), al-Tirmidhi (d. 279/892), al-Nasa’i (d. 303/915), and Ibn Majah (d. 273/887).

(A quick word about those double dates, because you will see them on every page. Muslims count years from the Prophet’s migration — the hijra — from Mecca to the city of Medina in 622 CE. So “256/870” means the year 256 of the Islamic calendar, which corresponds to 870 CE. This book always gives both.)

Twelver Shia Islam — the largest Shia branch, which holds that leadership after the Prophet rightfully belonged to his family line through his cousin and son-in-law Ali — has its own “four books”: the collection of al-Kulayni (d. 329/941), that of Ibn Babawayh (d. 381/991), and two collections by al-Tusi (d. 460/1067).

Ibadi Islam — a small and very old branch, found today mainly in Oman and pockets of North Africa, which we will meet properly in Chapter 10 — has the Musnad attributed to al-Rabi’ ibn Habib, which reached its surviving form through a scholar named al-Warijlani (d. 570/1174).

Now look at those dates again, and let the first plain fact of this book sink in. It is a fact nobody on any side disputes, because it is established by the books’ own records of their compilation and transmission:

Every one of these collections, in the form that survives today, was put together between one and five centuries after the Prophet died.

The Prophet died in 632 CE. Bukhari died in 870. Al-Tusi died in 1067. Al-Warijlani died in 1174.

That gap — and what happened inside it — is what this entire book is about.

The argument is older than you think

You might assume that questioning the Hadith is a modern habit — something that arrived with Western universities and skeptical professors. It is not. This is our second plain fact.

One of the most influential legal minds in Islamic history was Muhammad ibn Idris al-Shafi’i (d. 204/820). Living nearly two centuries after the Prophet, he developed a theory of Islamic law that would shape Sunni thinking for the next thousand years — we will meet him again in Chapter 4A. Al-Shafi’i wrote a foundational book called al-Risala, “The Treatise.”

And in it, he argues — at length, and with some heat — against opponents in his own lifetime who accepted the authority of the Qur’an alone (Schacht, 1950, pp. 10–11).

Read that again. More than 1,200 years ago, within the classical Islamic world itself, there were Muslims saying: the Qur’an is enough; these reports are not binding. Al-Shafi’i thought they were badly wrong, and wrote to refute them. But he took them seriously enough to write.

The modern movement that revived this position is usually called Quranism, and its followers Quranists: Muslims who hold that the Qur’an alone should be the source of religious law, and that Hadith — whatever historical interest it may have — should not be treated as binding.

The modern movement has its own history, and it is worth knowing the names. It begins, by most accounts, with an Egyptian doctor named Muhammad Tawfiq Sidqi, who in 1906 published a provocative article in the reformist Cairo journal al-Manar. In Pakistan, Ghulam Ahmed Perwez built the Tolu-e-Islam (“Dawn of Islam”) movement around Qur’an-centered reform. In the United States, the Egyptian-American biochemist Rashad Khalifa founded the “Submitters,” a community rejecting Hadith outright. In Malaysia, the intellectual Kassim Ahmad wrote a book against Hadith authority that got him formally condemned.

Why did this old dispute roar back to life in the twentieth century? The scholars who study the movement (D. Brown, 1996; Musa, 2008) point to something simple and profound: mass literacy and direct access to the text. For most of Islamic history, ordinary believers encountered scripture through scholars. In the modern world, anyone can buy a translation, open it, and ask: where does it actually say that? The Quranist movement is what that question looks like when it becomes a community.

Three kinds of claims — never to be mixed

One more tool before we begin the investigation proper. Throughout this book, three types of claims are kept rigorously separate:

1. Historical-critical claims. When and how did this text come into being? These can be tested — with language analysis, with manuscripts, with physical evidence. Philology and archaeology have jurisdiction here.

2. Theological claims. What did God, or the Prophet, intend to be binding? No excavation, no manuscript, no laboratory can answer these. They belong to faith and to interpretation.

3. Legal-practical claims. What rules follow for how a Muslim should actually live? These depend on both of the above — plus a third ingredient, juristic method: the reasoning tools lawyers use.

Mixing these three is the single most frequent error in everything written on this subject — in apologetic writing (defending the tradition) and polemical writing (attacking it) alike. A historian who “disproves” a theological claim has misunderstood his own job. A believer who presents a theological conviction as a historical finding has made the mirror-image mistake.

This book will not make either mistake. When we cross from one territory into another, I will tell you.

Now — to the evidence.

Chapter 2 — The Silent Witnesses: Islam Before the Hadith Books

The problem, stated honestly

The puzzle at the heart of this chapter can be stated in two sentences.

The Prophet Muhammad died in 632 CE. The great canonical Hadith collections were compiled roughly between 850 and 915 CE.

That is a gap of more than two hundred years.

What was Islam actually like during that gap? How did people pray? What did they believe? What rules did they follow? And — the historian’s eternal question — how do we know?

Historians don’t rely on one type of evidence. Imagine trying to reconstruct an ancient civilisation. You might find a government document. A letter. A coin. A ruined building. A story written down centuries later by the civilisation’s own descendants.

Would you trust them all equally?

Of course not. And the same problem exists — acutely — for early Islam. We have three streams of evidence, and they are very far from equal:

Stream one: the internal Muslim narrative. The sira — biographies of the Prophet. The akhbar — historical reports. And the Hadith themselves. This stream is rich, detailed, vivid. It is also, every word of it, material that was compiled generations after the events it describes, and it suffers from exactly the same chain-of-transmission problems as the legal Hadith we are investigating. Using it uncritically would be circular.

Stream two: outsiders. Christians, Jews, Zoroastrians and others who lived through the Arab conquests and wrote about them. Their great advantage: they are genuinely contemporary — some wrote within two years of the Prophet’s death. Their great weakness: they are sparse, often confused, and almost totally uninterested in the questions we care about. A Byzantine monk watching armies emerge from Arabia wanted to know about politics and prophecy, not about how the newcomers prayed.

Stream three: things. Coins. Inscriptions carved in stone. Papyrus documents. The ruins of buildings. This is the historian’s gold, because things can be physically dated and things do not lie about their own existence. Their weakness is the mirror image of their strength: a coin can tell you what a government wanted stamped on its money; it cannot tell you what a theologian believed or how a jurist reasoned.

Let us now walk through the timeline, period by period, and see what each stream actually yields. This is the longest chapter in the book, and deliberately so — everything that follows depends on it.

The Prophet’s lifetime (610–632 CE)

Begin with a fact that surprises almost everyone the first time they hear it.

The only text that survives from the Prophet’s own lifetime is the Qur’an itself.

No biography of Muhammad was written while he lived. No contemporary diary, no letter describing him, no administrative record naming him survives from those twenty-two years of his mission. For the man himself, in his own time, we have the revelation he delivered — and nothing else.

What about outsiders? Imagine living in the Byzantine Empire around the year 634. News begins arriving from Arabia: a new religious movement, armies on the move, a prophet. You know almost nothing about these people. One Christian writer records what he has heard, in Greek, in a text of anti-Jewish polemic. Today we call that document the Doctrina Jacobi — “The Teaching of Jacob” — and it was written around 634 CE, barely two years after the Prophet’s death. It is the earliest known outside notice of Islam.

A generation later, in the 660s, an Armenian chronicle attributed to a bishop named Sebeos gives another early outside account.

Both are brief. Both are garbled in places. And both care chiefly about politics and about claims of prophecy — not about ritual (Hoyland, 1997, pp. 55–61, 124–130). They confirm that something enormous was happening. They tell us almost nothing about what it looked like from the inside.

The document that might reach back to the Prophet himself

There is, however, one internal document that deserves special attention — because even source-critical scholars, the professionally suspicious ones, judge it unusually early.

When the Prophet arrived in the city of Medina in 622, he found a fractious town of feuding Arab tribes and long-established Jewish clans. Some kind of agreement was drawn up organising relations among all of them — mutual defence, dispute resolution, who belonged to the new community. That agreement is known today as the Constitution of Medina (or the Charter of Medina).

We do not have the original parchment. The text survives embedded in a much later biography — the sira of Ibn Ishaq, as preserved by his editor Ibn Hisham. So why would any skeptical historian trust it?

Because of its Arabic.

The Constitution is written in an archaic, knotty, legal register of the language — strikingly different from the smooth storytelling prose that surrounds it in the biography. It reads like a genuine old legal document that a later author has copied in, not like something he composed. The great German scholar Julius Wellhausen argued this on linguistic grounds; the British Arabist R.B. Serjeant later made the case in detail (Serjeant, 1978; the argument is discussed further in Anthony, 2020, pp. 72–80).

If that judgment holds — and note the if — then this is a rare and precious thing: textual material plausibly reaching back to the Prophet’s own lifetime by a route completely independent of the later Hadith-transmission machinery.

Two honest caveats, exactly as the evidence demands. First, the Constitution documents inter-communal politics — pacts between tribes — far more than it documents ritual or worship. Second, its precise textual history is still debated among specialists. File it under: probably genuinely early; disputed in detail.

The Rashidun period (632–661 CE)

The Prophet died in 632. Leadership passed to a series of four successors — in Arabic, khalifa, which gives us the English word caliph. Sunni tradition calls these first four the Rashidun, “the rightly guided”: Abu Bakr, Umar, Uthman, and Ali. Their thirty years saw the explosive conquests that carried Arab armies into Syria, Iraq, Egypt and Iran — and, according to Muslim tradition, the collection of the Qur’an into a single standard text.

That traditional account runs roughly so: after battlefield deaths thinned the ranks of those who knew the revelation by heart, the first caliph Abu Bakr ordered the scattered revelation gathered; later, the third caliph Uthman, alarmed by regional differences in recitation, had a standard edition prepared and copies sent to the major cities.

How much of that story can a historian verify? Be careful here — the account itself is preserved in Hadith and later histories, which means it has the very transmission problems we are investigating. Most specialists, including fairly critical ones, treat its broad outline as plausible (Anthony, 2020). A more skeptical minority contests the details (Shoemaker, 2022 — we will meet him properly in Chapter 13).

But now the physical evidence begins to speak. And what it says is remarkable.

The manuscript hiding under a manuscript

In 1972, workers were renovating the Great Mosque of Sanaa, in Yemen — one of the oldest mosques in the world. In a loft between the roof and the ceiling they found a mass of old parchment: a hidden cache of ancient Qur’an manuscripts, apparently retired there centuries ago when they wore out.

One of those manuscripts turned out to be something extraordinary: a palimpsest.

A palimpsest is a recycled manuscript. Parchment — treated animal skin — was expensive, so scribes sometimes washed or scraped an old text off a page and wrote a new one on top. But the old ink leaves ghostly traces in the skin, and modern imaging techniques can recover the erased layer.

The lower, erased text of the Sanaa palimpsest is one of the oldest witnesses to the Qur’an in existence. And here is the finding that matters: although the erased text contains variant readings — differences in wording from the standard edition — it is recognisably the same Qur’an (Sadeghi & Bergmann, 2010; Sadeghi & Goudarzi, 2012). Not a different book. Not a radically other scripture. The same text, with variations of the kind textual historians expect in any early manuscript tradition.

The building that testifies

Now travel to Jerusalem, and stand before one of the most famous buildings on Earth.

The Dome of the Rock is the golden-domed, blue-tiled octagonal shrine that crowns the Temple Mount — the raised platform sacred to Jews as the site of the ancient Temple, and to Muslims as the Noble Sanctuary. It is not a mosque; it is a shrine, built over the rock from which, in Islamic tradition, the Prophet ascended on his night journey.

Who built it, and when? An Umayyad caliph named Abd al-Malik — remember his name; he is about to appear again — completed it in the year 72 AH / 692 CE. That makes it the oldest surviving major monument of Islamic architecture anywhere in the world.

Why should a historian of texts care about a building?

Because of what is written on it.

Ringing the inside of the shrine, in gold mosaic on green, runs a long band of Arabic inscriptions — about 240 metres of them. And those inscriptions quote the Qur’an. Since the building is dated — dated by its own foundation inscription, to the year — these mosaics are the earliest physically dated quotations of the Qur’an in existence.

Think about what that means. You can stand in Jerusalem today and read Qur’anic text that was set in mosaic just sixty years after the Prophet’s death — within the lifetime of people who could have met him as children.

What do the inscriptions prove? That the Qur’an existed, publicly and officially, in recognisable form, within one long lifetime of the Prophet — proclaimed on the most prestigious building the new empire could construct.

What do they not prove? Total textual fixity. The scholar Estelle Whelan, who studied the inscriptions closely, noted that they show minor wording variants from the later canonical text — small divergences, of the kind that suggest some fine-tuning of the text was still going on even after Uthman’s traditional standardisation date (Whelan, 1998). The big picture is stability; the fine detail shows a text still settling.

The argument from spelling mistakes

There is a third, stranger witness to the Qur’an’s early history, and it comes not from any single manuscript but from all of them at once.

The Dutch linguist Marijn van Putten noticed something about the earliest Qur’an manuscripts: their spelling. Early Arabic spelling was not standardised, and the Qur’anic text contains numerous idiosyncratic spellings — a word spelled one odd way here, another odd way there.

What he found was this: those quirks are identical across every regional manuscript tradition. A copy made in Syria, a copy made in Iraq, a copy made in Egypt — all reproduce the same peculiar spellings, in the same words, in the same places.

Spelling quirks do not spread like that by coincidence. Independent scribes writing from memory or dictation would each produce their own quirks. The only good explanation for shared quirks is descent: every surviving copy of the Qur’an goes back to one single written master copy (van Putten, 2022).

This is a powerful argument, and note what kind it is: purely linguistic, independent of the manuscripts’ physical dating, independent of the Muslim tradition’s own account — and it lands on the same conclusion. The text was unified, in writing, very early.

What about government?

Later histories describe the second caliph, Umar, establishing a diwan — a register for paying stipends to soldiers and members of the community. It is the kind of institution an expanding conquest state would need. No actual diwan document survives from Umar’s own period, so we cannot verify it directly; but the description is consistent with the earliest surviving Arabic papyri — which belong to the next period, and to which we now turn.

The Umayyad period (661–750 CE)

In 661, after civil war (much more on that in Chapter 10), power passed to the Umayyads — the first dynasty of Islam, ruling from Damascus for nearly a century.

This period is the stage for one of the most consequential theories in the modern study of Islamic law. Its author was Joseph Schacht (1902–1969), a German-born scholar who ended his career at Columbia University, and whose 1950 book The Origins of Muhammadan Jurisprudence reshaped the entire field. Schacht’s thesis, in brief: the regional legal schools of early Islam — in Kufa in Iraq, in Medina in Arabia — first developed their law through judges’ own discretionary reasoning, called ra’y (“opinion”), and only later anchored that law, retrospectively, in Hadith attributed to the Prophet. Law first; prophetic pedigree afterwards.

Is he right? Hold the question; Chapter 13 stages the full battle. What matters here is that the Umayyad period is where his theory lives — and that four streams of physical evidence bear directly on what religion and government actually looked like in that century.

Witness one: tax receipts

Why on earth would historians get excited about tax receipts?

Because of what a tax receipt is: a dated, mundane, unfakeable-in-retrospect piece of ordinary administration. Nobody forges a receipt for a few sheep to deceive historians thirteen centuries later.

Papyrus — the ancient world’s paper, made from an Egyptian reed — survives beautifully in Egypt’s dry climate. And administrative Arabic papyri survive from as early as 22 AH / 643 CE — barely a decade after the Prophet’s death. (The foundational corpus was assembled by the Austrian papyrologist Adolf Grohmann; the volume of surviving documents rises sharply from the 710s, under an energetic governor of Egypt named Qurra ibn Sharik, whose office correspondence survives in quantity.)

What are these documents? Tax receipts. Requisition orders. Official letters. The paperwork of a functioning state.

What do they prove? That an Islamic polity with a working fiscal administration existed remarkably early.

And what do they not tell us? Almost anything about worship. The papyri are, in the academic review’s dry phrase, “almost entirely silent on ritual detail.” They corroborate the existence of the Islamic state — not its liturgy.

Witness two: money

Coins are a historian’s dream: mass-produced, officially sanctioned, hard to destroy completely, and very often dated.

The earliest coins minted under Muslim rule (the 660s to the 690s) are strange objects to modern eyes. They mostly kept the old imperial imagery — the portrait of a Byzantine emperor, the fire-altar of Zoroastrian Persia — with short Arabic or Islamic phrases squeezed in. New management, old branding.

Then comes a turning point that we can date to the exact year.

In 77 AH / 696–697 CE, the caliph Abd al-Malik — the same ruler who had just completed the Dome of the Rock — reformed the coinage of the empire root and branch. All images were removed. No emperor, no fire altar, no pictures at all. In their place, nothing but writing: the shahada — the Islamic declaration of faith, “There is no god but God; Muhammad is the messenger of God” — together with short Qur’anic phrases (overlapping with the wording of Qur’an 112:1–4 and 9:33).

Pause on what this object is. A coin passes through every hand in the empire. Whatever is stamped on it is, by definition, official, public, and standardised.

So Abd al-Malik’s reformed coinage is the clearest numismatic evidence — numismatics is the study of coins — that a specific, recognisable, textually stable core of Islamic creed was official and public by the end of the first Islamic century. And it is evidence completely independent of any Hadith transmission (Hoyland, 1997; Anthony, 2020).

Take a moment with that. Sixty-five years after the Prophet’s death, before a single canonical Hadith collection existed, the words “There is no god but God; Muhammad is the messenger of God” were jingling in the pocket of every merchant from Córdoba to Samarkand. Whatever else remains uncertain about the first century — and much does — the creed was not improvised later. It was already on the money.

Witness three: buildings, and which way they point

Muslims pray facing Mecca. The direction of prayer is called the qibla, and in a mosque it is marked by a niche in the wall called the mihrab; the preacher’s pulpit beside it is the minbar.

The earliest congregational mosques — at Kufa and Wasit in Iraq, at Fustat in Egypt (the garrison town that became old Cairo), and the great monuments of Damascus and Jerusalem — already show this basic equipment of communal prayer: mihrab, minbar, a qibla wall. Institutionalised, architectural, communal prayer toward Mecca, from the Umayyad period.

Now an honest wrinkle — and watch how the evidence labels work.

Specialist studies of early mosque orientation (surveyed in the work of the historian of Islamic science David King and the art historian Sheila Blair) show that early mosques’ aim toward Mecca was often measurably imprecise — off by noticeable angles. What does that mean? It is exactly what you would expect from a period when the practical astronomy of finding a direction across thousands of kilometres was still being worked out. Developing competence, not instant perfection.

Some fringe revisionist writers have seized on this imprecision to claim that early Muslims were really praying toward a different holy city altogether. You should know that this claim exists — and you should know its status: it is rejected by the mainstream architectural-historical consensus. It is not a live scholarly debate, and it should not be mistaken for one.

Witness four: the biography material that might be early

One more stream, this time textual — and it introduces a method you will meet again and again.

Urwa ibn al-Zubayr (d. 94/713) was a nephew of the Prophet’s wife Aisha, and one of the earliest people known to have collected accounts of the Prophet’s life. His material was transmitted onward through a scholar named al-Zuhri (d. 124/742).

Can we recover what Urwa actually taught, underneath a century and a half of retelling? A group of modern scholars believes we partly can, using a technique called isnad-cum-matn analysis — Latin-Arabic jargon for a simple idea: take all the surviving versions of one report; compare their chains of transmission and their wording, together; map where the versions converge and diverge; and use that map to triangulate where and when the report actually originated, the way comparing family trees and shared mutations lets biologists reconstruct ancestry.

Applying this method, Harald Motzki — and separately Andreas Görke and Gregor Schoeler — argued that a recoverable core of the Urwa material plausibly goes back to the late first Islamic century (Görke & Schoeler, 2008; Motzki, ed., 2000). That is one of the most optimistic datings on offer for any biographical material.

But mark its evidence label carefully: this is a probabilistic reconstruction, an inference from patterns — not a physically verified fact like a coin or a carbon date. Sophisticated, serious, and contested.

The bridge book

At the very end of the Umayyad era stands a book that is neither the old world nor the new, and understanding it will pay off for the rest of this book.

Malik ibn Anas was the great jurist of Medina, and around the 150s AH / 770s CE he compiled the Muwatta (“The Well-Trodden Path”). Open it and you find an extraordinary mixture: hadith of the Prophet, yes — but woven together with opinions of the Companions (the Prophet’s contemporaries), and with something Malik calls the practice of the people of Medina — the living custom of the Prophet’s own city, cited as authority in its own right. And all of it presented without insisting that every ruling be backed by a formal chain of transmission.

The Muwatta is the crucial transitional witness: something different in kind both from the undocumented living practice that came before it, and from the strict, chain-obsessed canonical Hadith genre that came after it (Dutton, 1999).

The early Abbasid period (750–850 CE)

In 750, revolution: the Umayyads were overthrown by the Abbasids, a dynasty claiming descent from the Prophet’s uncle Abbas, who moved the capital east to their new city of Baghdad.

In this period, the collection of Hadith intensifies into a formal movement, and the legal schools harden into institutions (Melchert, 2006).

One contested idea belongs to this chapter’s edges. The historian Fred Donner has argued that earliest Islam began as a broad, ecumenical “Believers’ movement” — open to righteous monotheists generally, including some Jews and Christians — which only later sharpened into a confessionally bounded “Islam” with hard edges (Donner, 2010). Note the label: this is a thesis about communal identity more than about legal content, and it is not a consensus view. It matters here because, if true, it would imply that a good deal of later legal-ritual definition was still taking shape in this period.

The verdict of Chapter 2: known, probable, possible, unknown

Resist the urge to compress everything above into one slogan — “early Islam is well documented!” or “we know nothing!” Both slogans are wrong. The evidence sorts into four distinct confidence tiers, and collapsing them into each other is the most common error in traditionalist and Quranist argument alike.

KNOWN — physically dated, non-textual evidence:

An Islamic state with a functioning tax administration existed by the 640s CE. (The papyri.)

A stable creedal formula — the shahada, plus short Qur’anic phrases — was official, public, and textually fixed enough to stamp on the empire’s money by 696 CE. (Abd al-Malik’s coin reform.)

A Qur’anic text recognisably continuous with today’s was in circulation, and being quoted in monumental mosaic, by 692 CE. (The Dome of the Rock.)

Congregational prayer toward Mecca was institutionalised in architecture by the same period. (The mosques.)

PROBABLE — strong converging inference, not directly physically verified:

The broad structure of practice — five daily prayers, fasting in Ramadan, almsgiving (zakat), pilgrimage (hajj) — was operating in some recognisable form already in the Rashidun period. The inference rests on the internal narrative’s consistency plus the early institutional evidence above; no contemporary outside or physical source independently dates this exact configuration.

The Constitution of Medina’s core clauses probably preserve genuinely early, Prophet-era material, on the linguistic grounds we saw (Serjeant, 1978).

POSSIBLE — consistent with the evidence, but not established:

That the specific ritual details later fixed in Hadith — the exact number of prayer units, the precise loud-and-silent structure of the prayers, the exact almsgiving thresholds — were already fixed in their final canonical form this early, rather than still varying by region and only standardised during the transition from Umayyad to Abbasid rule.

That the Urwa/al-Zuhri biography material preserves substantially accurate detail, rather than early-but-already-legendary elaboration.

UNKNOWN — no currently available method can resolve this:

The exact wording used in prayer during the Rashidun period.

Whether the regional legal variation of the Umayyad era reflected divergent memories of one original practice — or independent local elaboration from a much sparser common core.

How much of the Companion legal opinion embedded in Malik’s Muwatta reflects genuine first-century transmission, as opposed to second-century back-projection.

This four-tier verdict is, in itself, the most defensible historical conclusion this book will offer about the early period. It is not seriously disputed that the creedal and administrative core is secure, while the ritual-legal detail is not. Any claim — traditionalist or Quranist — that goes further than this, in either direction, currently exceeds the evidence (Motzki, ed., 2000; Robinson, 2003).

Hold that asymmetry in your mind. The whole rest of the book is, in a sense, an argument about what to do with it.

Chapter 3 — Two Words That Are Not the Same Word: Sunna and Hadith

There is a distinction coming that sounds like pedantry — and turns out to carry half the debate on its back.

People — including many Muslims — use the words Sunna and Hadith as if they were interchangeable. Historically and conceptually, they are not.

Sunna, in its earliest usage, meant something like “the trodden path”: normative precedent, established practice, the way things are properly done. The word is older than Islam — pre-Islamic Arabs used it for tribal custom. And in the earliest Islamic legal writing, by the so-called “ancient schools” of law in Kufa and Medina, sunna referred to living, communally transmitted practice — how the community actually prayed and lived, passed on by doing and watching, not necessarily pinned to any specific written report about the Prophet (Schacht, 1950; Hallaq, 1997, pp. 10–15).

Hadith, by contrast, is — as we saw in Chapter 1 — a discrete, textualised report with a chain of names attached. A literary genre, with a compilation history of its own.

One is a way of life, carried in bodies and habits. The other is a text, carried in books.

Did practice exist before the books? Yes — and both sides agree

Now ask the question that matters: did Sunna — the living practice — exist independently of the canonical Hadith?

The evidence supports a qualified yes. Communal ritual practice — praying, fasting — plainly existed a century or more before the canonical Hadith collections that were later cited as its textual warrant. We established that in Chapter 2, and Chapter 14 will confirm it from the other end.

This simple chronological fact is the historical kernel inside one of the traditionalists’ most important concepts: mutawatir amal — “mass-transmitted continuous practice.” The idea runs like this. A single written report could, in principle, be forged. But how do you forge the way an entire community prays — a practice performed publicly, five times a day, by everyone, taught by every parent to every child, in every city, in every generation, without interruption? Continuous mass practice, the argument goes, is a category of evidence distinct from, and in principle independent of, any individual hadith chain (Hallaq, 2005, pp. 47–75).

It is a genuinely powerful argument. Sit with it for a moment before we complicate it.

Schacht’s darker reading of the same fact

Joseph Schacht — the scholar we met in Chapter 2 — looked at the very same chronological fact and read it differently.

For Schacht, “Sunna” in the earliest period was a somewhat elastic label. When the early legal schools of Kufa or Medina said “the sunna,” he argued, they often meant their own school’s accumulated legal reasoning — the settled practice of their local tradition — dressed in the prestige of the word. Only later, as the Hadith movement gained strength and demanded documentary proof for everything, was that school doctrine retrospectively textualised, given chains of transmission, and attributed to the Prophet himself (Schacht, 1950).

On this reading, the Hadith did not record the Sunna. The Hadith rebranded it.

Where the two readings meet — and where they cannot

Now notice something important, because it is a model for how this whole book works.

Both readings agree on the chronology: Sunna-as-living-practice came before Hadith-as-literary-genre. That is common ground between a devout traditionalist and the field’s arch-skeptic.

Where they part is on continuity of content. Did the living practice of, say, 680 CE faithfully preserve the practice of 632 — which the later Hadith then faithfully recorded? Or had practice already drifted, region by region, with the Hadith then canonising the drift?

And here is the uncomfortable, honest conclusion: this question remains largely unresolved — and is probably unresolvable in full. Why? Because no contemporary record of the “ancient schools’” actual content survives from before it was turned into Hadith. The very evidence that could settle the question is the evidence whose reliability is in question.

Keep this pattern in view. You will see it again.

And now that we know what the reports claim to be, it is time to watch them take physical form — from whispered memory to sacred canon.

Chapter 4 — From Private Notebooks to Sacred Canon

How does an oral report become a canonical text? Not in one leap. The transmission of hadith moved through several overlapping stages — not a tidy sequence, but roughly this arc:

Stage one: memory and mouth. Oral transmission among the Companions — the Prophet’s contemporaries — and the Successors, the next generation, who had known the Companions but not the Prophet.

Stage two: the notebook. Private written notes, used as memory aids rather than published books. Scholars call this the “hypomnema model,” borrowing the Greek word for a personal aide-mémoire (Schoeler, 2009, 2011) — think of a preacher’s private sermon notes, never meant for circulation.

Is this stage just a theory? No — we have a physical survivor. The Sahifa of Hammam ibn Munabbih is a small written collection set down by a student of the Companion Abu Hurayra; its author died around 101/719. The text survives, and was edited and published by the scholar Muhammad Hamidullah in 1953. A sahifa is simply a “sheet” or written document — and this one is material confirmation that hadith were being written down, privately, within the first Islamic century.

Stage three: regional schools. Kufa, Basra, Medina, Syria — each centre developing its own legal tendencies and, over time, its own characteristic hadith emphases (Schacht, 1950; Melchert, 2006).

Stage four: the critics arrive. From the second Islamic century (the 700s CE), scholars began systematically interrogating the chains. Was this transmitter honest? Did he really meet the man he claims to quote? Was his memory sound in old age? A whole biographical literature grew up to answer such questions — ilm al-rijal, literally “the science of men”: vast dictionaries assessing thousands of individual transmitters.

Stage five: canon. In the third Islamic century (the 800s CE) come the great topically organised, chain-vetted compilations — Bukhari, Muslim, and the rest.

Why these six books?

Out of a far larger contemporary output, why did six Sunni collections become “the canon”?

The first thing to say: it was a documentable historical process, not an instant coronation — Chapter 14 tells that story in full, and it will surprise you.

The second thing: the compilers themselves were ferociously selective. Al-Bukhari reportedly examined an enormous pool of circulating reports and retained only a small fraction, by his own criteria. And the compilers’ criteria differed. Muslim’s collection, for instance, arranges the versions of a hadith by the strength of their chains — a different method from Bukhari’s arrangement by legal topic.

And the third thing — an uncomfortable one for any claim that the six books contain “the best” material by some neutral standard. Many rival collections of the same era simply did not survive. Some vanished entirely; others survive only as quotations embedded in later works. A book survived if the scholar who compiled it had students who transmitted it, whose own works were transmitted in turn — a chain of institutional luck.

Statisticians have a name for the trap this creates: survivorship bias. We can only compare the books we have. “What survived” and “what was best” are not automatically the same thing (J. Brown, 2007, ch. 1).

Books, though, are only half the story. The other half is people — the question of who, after the Prophet, had the right to say what Islam required. That question has a history of its own.

Chapter 4A — Who Speaks for God? The Evolution of Authority

The problem no community can escape

While the Prophet lived, religious authority had an address.

His authority combined two things that later tradition would carefully distinguish: revelation — the Qur’an, understood as God’s own speech — and Sunna, his personal example and judgment. During his lifetime, nobody needed a theory separating them. Both flowed from one living man. If you had a question, you could walk across Medina and ask him.

Then, in 632, he died. And revelation — on Islam’s own terms — ended forever.

What happens to a community built around a living oracle when the oracle is gone? The sociologist Max Weber gave the problem a name we will explore in Chapter 15 — the “routinisation of charisma.” Everything in this chapter is, in one sense, the historical record of how the Muslim community redistributed that unified personal authority into durable, transmissible forms.

The Rashidun: rule by precedent and judgment

The first four caliphs (632–661) governed through a blend of three things: consultation — shura in Arabic; the accumulated weight of senior Companions’ precedent; and their own reasoned judgment — that word ra’y again — on matters the Qur’an did not address.

Two famous examples involve Umar, the formidable second caliph.

Conquered Iraq presented a problem no revelation had anticipated: vast, rich farmland. Should it be divided among the conquering soldiers as spoils? Umar ruled no — the land would stay with its cultivators, who would pay a land tax, the kharaj, funding the state. A momentous ruling, with no direct scriptural text behind it.

And during a famine, Umar suspended the prescribed punishment for theft — reasoning that starving people driven to steal food should not be punished as ordinary thieves.

Was this an early form of authoritative legal reasoning — what later jurists would call ijtihad? Or simply pragmatic governance? Be careful: that distinction belongs to later legal theory. The actors themselves probably never drew it.

The Umayyad century: two temperaments form

As the community spread across an empire, two great legal temperaments crystallised — and their names will follow us through the rest of the book:

Ahl al-Ra’y — “the people of reasoned opinion” — centred on Kufa in Iraq, associated with early masters like Ibrahim al-Nakha’i and Hammad ibn Abi Sulayman (the teacher of Abu Hanifa, founder of the Hanafi school). They gave relatively greater weight to structured, disciplined reasoning.

Ahl al-Hadith — “the people of Hadith” — centred on the Hijaz, the region of Mecca and Medina. They gave relatively greater weight to transmitted reports and continuous local practice.

These labels are the genuine historical origin of the divide — not a later caricature (Schacht, 1950; Hallaq, 1997, pp. 16–20).

And connecting the regions, a remarkable institution: al-rihla fi talab al-ilm — “the journey in search of knowledge.” Through the second Islamic century, scholars criss-crossed the empire, from Spain to Central Asia, collecting reports and cross-checking them against what other regions had. Think of it as an informal, empire-wide peer-review network — a disciplining mechanism against purely local, undocumented practice. It fed directly into the great collection movement of Chapter 4.

Al-Shafi’i: the great synthesiser

Into the quarrel between the two temperaments stepped the man we met in Chapter 1: Muhammad ibn Idris al-Shafi’i (d. 204/820).

In al-Risala, he proposed a peace treaty in the form of a hierarchy. Legal authority, he argued, flows through four ranked sources:

The Qur’an.

The Sunna — which al-Shafi’i now defined, fatefully, as authenticated Hadith specifically. Not vague communal practice; documented reports, with chains.

Ijma — the consensus of qualified scholars.

Qiyas — analogy: extending a revealed ruling to a new case that shares its underlying rationale, reasoning strictly within revealed principle.

Free-floating personal opinion — raw ra’y — was demoted; disciplined analogy took its place (Hallaq, 1997, pp. 21–35). The synthesis did not end regional variation overnight. But it supplied the methodological template that the crystallising schools would each adapt.

The four schools take institutional shape

A madhhab is a school of law — a complete, teachable system of legal method and accumulated rulings. Sunni Islam settled into four. And here is a detail worth savouring: none of them was really created by its founder alone. Each reached settled form through students who systematised the master’s method:

The Hanafi school — through Abu Yusuf (d. 182/798), chief judge of the empire, and al-Shaybani (d. 189/805), who wrote down and extended Abu Hanifa’s positions.

The Maliki school — through the Muwatta itself, and its later massive elaboration in the Mudawwana compiled from Ibn al-Qasim’s teaching by Sahnun (early 800s).

The Shafi’i school — directly from al-Shafi’i’s own writings, refined by students such as al-Muzani.

The Hanbali school — from the rulings of Ahmad ibn Hanbal (d. 241/855), the arch-traditionalist of Baghdad, collected and systematised chiefly by his students and later by al-Khallal (d. 311/923). The delicious irony: Ibn Hanbal himself resisted producing a systematic code — he distrusted system, preferring case-by-case adherence to transmitted text. In a strict sense, the “Hanbali school” is more his students’ creation than his own.

Did the gate ever close?

An older generation of Western scholars — Schacht, and the legal historian N.J. Coulson — told a tidy story: around the 4th Islamic / 10th Christian century, the “gate of ijtihad” (independent legal reasoning) swung shut. From then on, jurists could only practise taqlid — following the established doctrine of their school. Islamic law, on this telling, froze.

In 1984, the legal historian Wael Hallaq published an article that took the story apart. Examining the actual output of later jurists, he showed continued, institutionally channelled, genuinely independent legal reasoning running right through the later centuries. The “closure of the gate,” he argued, was a retrospective simplification — a myth of rigidity — not a datable event (Hallaq, 1984).

Why does this matter to our debate? Because it cuts both ways. It tells traditionalists that their own tradition was more flexible than its stern self-image suggests. And it tells Quranists that the “frozen, ossified orthodoxy” of their polemic is partly a scholarly artefact. The reality was more dynamic than either side’s rhetoric assumes.

Meanwhile, on the other paths: Shia and Ibadi authority

Twelver Shia Islam solved the authority problem differently — and its solution had a built-in crisis.

While the Imams lived — the divinely designated leaders from the Prophet’s family line, twelve in all by Twelver count — authority rested in a living, present guide: analogous in kind, though not in scope, to the Prophet’s own combined role. You could, once again, ask.

Then, Shia doctrine holds, the twelfth Imam did not die but was hidden by God from the world: the occultation (ghayba). A “minor occultation” from 260/874, during which he communicated through four successive agents; then the “major occultation” from 329/941 — full concealment, to last until his messianic return at the end of time.

With no accessible Imam, who decides? Authority passed to qualified jurists acting as the Imam’s general deputies — niyaba amma. And here is a fact that startles many Shia and non-Shia alike: the fully developed modern institution of the marja al-taqlid — the “source of emulation,” one of a small number of supreme jurists whose rulings ordinary believers are expected to follow — is essentially a nineteenth-century development, generally associated with the consolidation of authority around Murtada al-Ansari (d. 1281/1864). It is not a feature of early, or even medieval, Shia practice (Newman, 2000).

Ibadi Islam, meanwhile, developed around its own imamate tradition and transmitted scholarship through regional teaching circles called halaqas — largely independent of both Sunni school-formation and Shia Imami theology.

The shape of the whole story

Step back and look at the arc: unified authority in one living Prophet → dispersal into regional custom → discipline through cross-checked transmission → formalisation into ranked hierarchies of sources → institutionalisation into transmissible schools.

That general trajectory is not seriously disputed by anyone. What specialists fight over — bitterly — is how much authentic early content survived the journey through each stage. That fight is Chapter 13.

Chapter 4B — Why Hadith Had to Be Invented (Even If It Wasn’t)

That chapter title is deliberately provocative, so let me immediately explain it.

Chapter 4A traced how authority moved from the Prophet to the schools. This chapter asks a different question: why. What pressures made some hadith-like institution — a written, citable record of the founder’s example — close to inevitable for a religion administering an empire?

And notice the crucial point before we start: this question is completely independent of whether any specific report is authentic. We are asking why the container had to exist — not what filled it.

The pressures

Imperial scale. Within a century of the Prophet’s death, the Muslim community governed territory stretching from Iberia — Spain and Portugal — to Central Asia. Hold that against the Prophet’s own polity: Medina, a single city-state. A legal and administrative system operating across three continents needed standardised, citable, portable rules that did not depend on physical access to a living authority. A textualised, transmissible corpus of reports fills exactly that gap.

Judges needed something to cite. The empire appointed judges — qadis — across its new provinces. Picture one of them, in newly conquered Central Asia, facing a commercial dispute involving contract customs no one in Medina had ever imagined. He needs a ruling — and he needs a basis for it that is more than his personal say-so. That need created demand for reports of Prophetic or Companion precedent. Real or attributed.

Administrative standardisation. Taxation. Inheritance division. Marriage law. The freeing of slaves. All of it now for a vastly larger and more diverse population than Medina’s — and all of it requiring more granular rules than the Qur’an’s general ethical-legal statements supply. Note this carefully: on any reading — traditionalist, Quranist, or secular — the Qur’an leaves that functional gap for something to fill. The disagreement is about what should legitimately fill it.

Political legitimacy. Umayyad and Abbasid rulers — and every rebel and rival who challenged them — had powerful incentives to wrap their policies and succession claims in prophetic precedent. This is precisely the mechanism that Ignaz Goldziher documented in the 1890s (his story comes in Chapter 13): hadith as a currency of legitimacy contests, not merely as neutral historical memory.

Sectarian rivalry. Once competing communities existed — proto-Sunni, proto-Shia, Kharijite/Ibadi (Chapter 10) — each needed to justify its position on leadership and law. Each therefore had reason to develop, or to privilege, a body of authoritative reports supporting its own reading. The result: the divergent, sect-specific corpora we have already glimpsed.

And the schools themselves. Once formal madhhabs existed, each needed a citable evidence base to compete for students, adherents, and official patronage. School-formation both responded to the pressure and accelerated it.

The astonishing comparative pattern

Now widen the lens, because Islam turns out not to be unusual here. It turns out to be typical.

Rome. Roman law began as the localised, custom-based ius civile of a small city-state. As Rome became an empire, the law expanded into a vast apparatus of jurists’ opinions — the responsa of legal giants like Gaius, Ulpian, and Papinian — finally systematised in the sixth century CE in the Emperor Justinian’s Corpus Juris Civilis. A many-centuries-later codification of accumulated expert opinion. Not a text handed down complete at Rome’s founding.

Judaism. The Written Torah is a covenant document; running a community’s daily life requires operational law for far more situations than its explicit text covers. The solution was the Oral Torah — a body of law held to accompany the written one — codified into the Mishnah around 200 CE, once rabbinic authority needed a citable, transmissible basis. The timing is telling: the Temple in Jerusalem had been destroyed in 70 CE, and with it the priestly centre of authority. (Much more in Chapter 12.)

Christianity. Canon law developed even more explicitly as an administrative response to imperial-scale church governance: the decisions of councils, the decretals of popes, the precedents of the Church Fathers, accumulating for a millennium and finally systematised in the twelfth-century Decretum Gratiani. Recognisably parallel in function — if not in content — to the ninth-century Hadith canonisation.

Among historians of religion and law, the pattern is not seriously disputed: every scripture-founded tradition that outgrew its founding community’s original scale developed some analogous secondary, jurist-mediated legal corpus — and did so within a roughly comparable multi-generational timeframe after the founding period.

What the comparison proves — and what it carefully does not

Handle this conclusion with precision, because both sides are tempted to misuse it.

A functional explanation of why such an institution was likely to emerge is compatible with that institution’s content being anywhere from substantially accurate to substantially invented. So the comparison does not settle whether the Hadith corpus’s specific content is authentic (Chapter 13) or theologically legitimate (Chapter 5).

What it does responsibly establish is this: the mere existence of a large secondary legal corpus, growing over the centuries after a scripture’s revelation, is not — by itself — evidence of illegitimacy or fabrication. It is simply what happens, sociologically, to essentially every scriptural legal tradition that survives contact with imperial scale. Weber’s routinisation-of-charisma framework (Chapter 15) describes the same phenomenon at the level of religious authority generally.

A Quranist who says “the Hadith corpus grew late, therefore it is fraudulent” has overreached. A traditionalist who says “the corpus was functionally necessary, therefore it is authentic” has overreached identically. The container had to exist. The contents must be judged on their own evidence.

Chapter 5 — “This Day I Have Perfected Your Religion”: The Completeness Puzzle

We have arrived at the heart of the whole debate. Everything before this chapter was preparation.

State the puzzle as sharply as it can be stated.

The Qur’an repeatedly describes itself as complete. Detailed. Clarified. Sufficient. And yet classical jurists in all three traditions — Sunni, Shia, and Ibadi — hold that Hadith is necessary to practise Islam correctly.

How can both be true? Is one side simply misreading?

A rigorous answer has to start where the Qur’an starts: with Arabic words. Because here is the first discovery — popular writing on this subject routinely flattens four distinct Arabic terms into the single English word “complete.” They are not one word. And they do not make one claim.

A working vocabulary: five words for “complete” that aren’t the same word

1. Kamal — completion, perfection. This is the word in the most famous verse of the whole debate, Qur’an 5:3: al-yawma akmaltu lakum dinakum — “Today I have perfected your religion for you, completed My favour upon you, and am pleased with Islam as your religion.”

Classical Islamic scholarship preserved, for many verses, an account of the circumstances in which the verse came down — a genre called asbab al-nuzul, “the occasions of revelation.” For 5:3, the majority of classical commentators (al-Tabari, the great tenth-century exegete; Ibn Kathir, the fourteenth-century one) report a very specific occasion: the ninth of the month Dhu’l-Hijja, in year 10 of the Islamic calendar, at the plain of Arafah, during the Prophet’s Farewell Pilgrimage — his last, just months before his death.

That anchoring shapes the classical Sunni reading: “perfected” marks the completion of Islam’s core creedal and ritual framework at that point in history. Mecca’s idolaters were no longer a threat; the pilgrimage rites had been purified. It is a temporal-historical completion marker — not a claim that every future legal question had been textually pre-answered (Ibn Kathir, on this verse).

Now watch the ground shift. Twelver Shia exegesis — for instance al-Tabataba’i’s great modern commentary al-Mizan — links 5:3 to a different moment on the same journey home: the event of Ghadir Khumm, a gathering at a desert pool where, in Shia understanding, the Prophet publicly proclaimed Ali’s authority over the community. On that reading, what was “completed” was the announcement of leadership.

Pause on what just happened. Even the flagship completeness verse is subject to major sectarian disagreement — not merely over how far its claim reaches, but over what it refers to in the first place.

2. Tafsil / fussilat — detailed exposition. From the Arabic root f-s-l, whose core sense is “to separate, to distinguish parts.” It appears in 6:114 (“a Book explained in detail”), 12:111 (“a detailed explanation of everything”), and 41:3 (“a Book whose verses are set out distinctly”). The root meaning is laying something out part by part, element from element. That is a claim about the text’s internal clarity and organisation — not self-evidently a claim about covering every future legal contingency. Quranists read the “everything” of 12:111 as procedural comprehensiveness. The classical commentators al-Razi and al-Qurtubi read “everything” as bounded by the surrounding context — creed, guidance, the lawful and the unlawful — rather than unlimited.

3. Tibyan — clarification. From the root b-y-n. The word of 16:89: “a clarification of all things.” It denotes making something evident, plain — not necessarily supplying its full operational instructions.

4. Tasrif — varied exposition. From the root s-r-f, “to turn, vary, diversify.” The word of 18:54: “We have certainly diversified in this Qur’an, for the people, every kind of example.” This describes rhetorical and pedagogical variety — teaching one lesson through many parables. It is a distinctly different claim from either “completeness” or “detail,” and classical commentators never read it as a legal-sufficiency claim. Honesty requires flagging: its inclusion in popular Quranist verse-lists is comparatively weak, and some careful Quranist writers themselves omit it.

5. Bayan — elucidation. The same root as tibyan (b-y-n); the word of 16:44 and 75:19. And here is the pivot of the traditionalist case. Verse 16:44 assigns an act of clarification partly to the Messenger himself: “…that you may make clear — li-tubayyina — to people what has been sent down to them.”

Sit with the implication. If bayan — clarifying — is itself a function the Qur’an delegates to the Prophet, then the Qur’an’s own self-described clarity need not exclude a further, Prophet-mediated layer of explanation. On the contrary: the text would be anticipating one.

Is this a settled dictionary-level victory for either side? No. The Japanese scholar Toshihiko Izutsu — a pioneer of studying the Qur’an’s vocabulary as interconnected semantic fields, in his 1964 book God and Man in the Qur’an — showed that these roots occupy overlapping but distinguishable regions of meaning, and which region a given verse activates depends on context that the commentary tradition itself never uniformly fixed (Saeed, 2006, ch. 2).

The eleven completeness verses, one by one

Popular debate cites “the completeness verses” as an undifferentiated bundle. Rigour requires laying them out individually — because, examined one at a time, they turn out to be of visibly uneven strength.

A few threads to pull from the table.

On 6:38 — “We have not omitted anything from the Book” — the majority classical commentators (al-Tabari, al-Razi, al-Qurtubi, Ibn Kathir) read “the Book” as the Preserved Tablet: in Islamic theology, the heavenly tablet on which God has recorded all things, every leaf that falls and every destiny. On that reading the verse says nothing about the earthly Qur’an’s legal coverage at all. The Quranist reading — and it is also a live minority position among some modern reformists — takes the plain surface sense and puts the burden of proof on whoever wants to narrow it.

And notice, laid out verse by verse, the unevenness: 5:3, 12:111, and 16:89 carry real interpretive weight for the Quranist case; 18:54 and 39:23 are comparatively weak once their specific Arabic terms are examined individually rather than dissolved into a generic impression that “the Qur’an calls itself complete.”

So what have we actually learned? That “the Qur’an says it is complete” is not one claim but a family of different claims, made with different words, of different strengths — and that the strongest of them still leave the decisive question (complete for what?) unanswered by the text alone. Hold that thought; the next three chapters will attack it from three more directions.

A third reading nobody expects

There is one more voice at this table, and it belongs to neither camp.

The historical-critical scholar Nicolai Sinai reads these self-referential claims through the lens of the Qur’an’s own era — the world scholars call Late Antiquity, the centuries of transition from the classical to the medieval world. Scriptural texts of that era, Sinai observes, characteristically assert their own sufficiency as guidance against unbelief — that is the rhetorical genre — not sufficiency as an exhaustive legal code (Sinai, 2017, chs. 2–3).

If Sinai is right, both maximalist readings are anachronisms: the Quranist who reads “clarification of all things” as a legal-code claim, and the traditionalist who reads it as needing a Hadith-shaped supplement, are both asking a question the verses were never answering.

Label it honestly: a plausible but unproven genre-based reading — which would, if accepted, complicate both sides simultaneously.

Two questions to close the chapter, answered honestly

Does the Qur’an anywhere anticipate a future written Hadith corpus as binding legislation? No verse states this explicitly, in those terms — a textual fact both sides can, and should, agree on. The traditionalist case rests on inference from the obedience and clarification verses (fully catalogued in the next chapter). The Quranist case rests on an inference from the completeness verses that such a corpus is excluded. Both are inferences. Neither is a direct statement.

Did the earliest generation read these verses the way later jurists did? This cannot be established — in either direction. No commentary survives from the Companions’ generation independently of later transmission. The earliest systematic exegesis — material attributed to the Prophet’s cousin Ibn Abbas, transmitted through later, contested chains — already postdates the four-source legal model of Chapter 4A by generations. The honest answer is: we do not know. And any claim on either side about “what the first generation understood” exceeds the evidence.

Chapter 5A — “Obey the Messenger”: Every Authority Verse in the Qur’an

If you have ever watched this debate conducted online, you have seen it reduced to a duel of proof-texts. One side fires “Obey God and obey the Messenger!” The other fires back “The Messenger’s duty is only to convey!” Neither side seems to notice that the Qur’an contains many more verses bearing on religious authority — and that they fall into distinct categories making distinct claims.

So here is the full catalogue. Every category, every verse, with the traditional and Quranist readings side by side. Take your time with it; this table is one of the load-bearing walls of the whole book.

A word on the Banu Nadir affair, since it anchors one of the table’s most contested rows. The Banu Nadir were a Jewish tribe of Medina, expelled after a conflict with the Prophet; verse 59:7 addresses the distribution of their abandoned property — war spoils. “Whatever the Messenger gives you, take; whatever he forbids, refrain from” is, in its immediate context, about that distribution. Classical jurists lifted the sentence into a general principle of Prophetic legislative authority — one of the most consequential interpretive moves in Islamic legal history. Whether that lift is legitimate is precisely what the two columns above disagree about.

Two structural observations

Lay the verses out this way — rather than duelling with single proof-texts — and two things become visible that neither side’s pamphlets usually admit.

First: the traditional case is cumulative. It does not rest on any single verse. It draws converging support from at least five distinct categories in the table — obedience, judgment, clarification, permitting/forbidding, exemplar. That is a real structural strength. And the standard Quranist method — rebutting verses one at a time — does not fully neutralise a cumulative case, even where individual rebuttals (like the one about 59:7’s original context) are textually well grounded. Ten weak beams can hold a roof that no single beam could.

Second: the Quranist case is strongest exactly where the traditional reading has to stretch. Wherever tradition extends a verse’s plain historical context into a general legislative principle — 59:7’s spoils dispute; 16:43’s scripture-keepers — that extension is an interpretive move, which traditionalists must argue for in each case rather than assume. This book flags every such extension rather than granting it silently.

Chapter 5B — One Word to Rule Them All: What Does “Din” Actually Mean?

Return now to the flagship verse, 5:3: “Today I have perfected your din.”

Every English translation you have ever seen renders din as “religion.” And the entire completeness debate is then conducted as if “religion” — in the modern, Western, privatised sense of the word: a compartment of life involving worship and belief — were obviously what din means.

It is not that simple. And the complications matter.

The Arabic root d-y-n occurs roughly ninety times in the Qur’an across its various forms — the count comes from the standard concordance of the Qur’an compiled by Muhammad Fu’ad Abd al-Baqi, the reference work scholars use to find every occurrence of a word. Ninety occurrences give us a large sample. And the sample shows a word with a much wider, and stranger, range than “religion.”

The full range of the word

Judgment, recompense. The Qur’an’s very first chapter calls God “Master of yawm al-din” — the Day of Judgment (1:4; also 82:17–19). Here din plainly means judgment, reckoning, the settling of accounts. No connection to “religion” in any institutional sense.

Submission, indebtedness. The root connects to the verb dana — “to submit, to owe obedience, to be indebted to” — a pre-Islamic Arabic usage rooted in social subordination and debt relationships. The philologist Arthur Jeffery, in his 1938 study of the Qur’an’s loanword vocabulary (The Foreign Vocabulary of the Qur’an), traced a plausible additional background: the Aramaic and Middle Persian word den, used in Zoroastrian religious-legal contexts to mean “religion/law/custom.” The Qur’anic sense may fuse the native Arabic “submission and debt” meaning with an already-current Persian institutional sense of “religion-as-law” (Izutsu, 1964, pp. 219–229; Jeffery, 1938).

A legal-administrative system — with no religious content whatsoever. And now the occurrence almost nobody discusses, and arguably the most important one in the whole debate. In the Qur’an’s telling of the Joseph story, 12:76, Joseph — serving in Pharaonic Egypt — “could not have detained his brother fi dini al-malik”: “under the king’s din.” The din of a pagan Egyptian king. Here the word denotes a foreign monarch’s legal-administrative code. A secular legal system, full stop. This single verse demonstrates that din’s semantic range includes purely legal-administrative systems as a live, native sense — not only creed or worship.

Religion in the institutional sense. Din Allah (God’s religion), din al-haqq (the religion of truth), din al-qayyima (the upright religion) — at 9:36, 12:40, 30:30, 98:5. The sense closest to the conventional translation: a total system of worship, creed, and communal identity.

“Islam” as a proper name. “Truly the din with God is Islam” (3:19); also 3:85 — and 5:3 itself.

“Way of life.” A distinctly modern gloss. You will find it in translators like Muhammad Asad, and pressed to its limit in the political theology of twentieth-century Islamist thinkers — above all Abul A’la Maududi, for whom din denotes a comprehensive socio-political order embracing law, economy, and governance. This is a live and influential modern reading. But specialists in Qur’anic semantics (Izutsu, 1964) classify it as an ideological extension of the word’s classical range — not a straightforward restatement of it.

So what, exactly, did God “perfect”?

Given that range, “I have perfected your din for you” admits at least four non-equivalent construals — and the text alone does not choose between them:

1. Completion of the religious-social order as a functioning whole, at a specific historical moment. This is the majority classical reading, anchored by al-Tabari and Ibn Kathir to the Farewell Pilgrimage — and note, it is fully compatible with continued operational elaboration afterward.

2. Completion of a legal system specifically. Press the 12:76 sense — din as legal-administrative code — onto 5:3, and the verse moves much closer to the Quranist “legal completeness” reading. But most linguists would raise a caution here: importing a sense from an unrelated narrative context (a foreign pagan king’s administration) onto God’s own din is a stretch that requires independent justification. It is not a straightforward parallel.

3. Completion of the covenant — the submission-and-indebtedness relationship between God and the community. The construal closest to the root’s oldest meaning.

4. (Twelver Shia) Completion via the Ghadir Khumm announcement. On this reading, what was completed is a matter of authority and governance structure — who leads — not legal content at all. A third construal, distinct from both the traditionalist and the Quranist readings.

The honest verdict

5:3’s scope is underdetermined by the word’s semantics alone.

The most plausible reading, weighing everything: the verse marks the completion of Islam as an institutional religious-social order — creed, core ritual, basic communal law, identity — at a specific moment in sacred history. But the word’s demonstrated “legal code” sense in 12:76 means the stronger Quranist reading cannot be dismissed on dictionary grounds alone. It simply requires an additional, contestable interpretive step — treating “God’s din” as activating the same narrow sense the Qur’an applies to a pagan king’s administration. Some Quranist writers take that step without flagging it as a choice. This book flags it.

And a final word to each of our three readers. To the traditionalist: the majority gloss is well supported by the verse’s occasion of revelation — but it should not be treated as foreclosing all broader senses a priori. To the Quranist: the “legal completeness” construal needs an explicit, independently justified argument for importing 12:76’s sense — not a silent assumption. To the secular historian: “what was completed” is, in the end, a question about divine reference and intention that no document or excavation can settle. History can establish only two things: that the word’s range is wider than “religion,” and that the verse is tied, by the tradition’s own account, to a specific and datable event.

Chapter 5C — Was Muhammad a Lawgiver? Seven Functions, Pulled Apart

“Did Muhammad possess independent legislative authority — or only authority mediated through revelation?”

The question sounds simple. It is not — because the phrase “Prophetic authority” bundles together at least seven different functions, and they need to be separated before the question can even be asked properly.

The date-palm story

Now for the single clearest data point on this whole question — and, delightfully, it is not a Qur’anic verse at all. It is a hadith. A well-attested one, by mainstream Sunni standards, found in Sahih Muslim.

The story. The Prophet, newly arrived in Medina — a merchant of Mecca, not a farmer — passed cultivators hand-pollinating their date palms: climbing the trees to transfer pollen, as Iraqi and Arabian farmers had done for millennia. He remarked that perhaps it would be better not to. Out of deference, they stopped.

That year the harvest failed.

When he was told, he reportedly said: antum a’lamu bi-umuri dunyakum — “You know better the affairs of your worldly life.”

Why does a story about date farming matter so much? Because of what mainstream Sunni orthodoxy does with it. This report is the classical proof-text by which Sunni theology explicitly denies that the Prophet possessed unlimited independent authority in all matters. His guaranteed reliability — theologically termed isma, protection from error — is restricted, on this account, to matters of din (religion, in whichever of Chapter 5B’s senses is operative) and does not extend to dunya — worldly, technical affairs. Farming. Medicine. Military tactics, on some accounts.

This distinction is astonishingly under-used, by both sides, in popular debate. Watch what it does:

It shows that mainstream Sunni theology does not hold the totalising position — “everything he ever said or did is binding law” — that Quranist polemic sometimes attributes to it. The tradition itself draws a boundary.

And it shows, equally, that traditionalists do claim a strong, non-trivial authority — bounded to religious matters. A boundary, note, which the completeness verses (Chapter 5) and the obedience verses (Chapter 5A) do not, on their own wording, clearly draw for the reader. The boundary comes from theology, not from the verses’ surface text.

How far does the protection extend? Three traditions, three answers

How totalising Prophetic authority is held to be differs across the traditions — and mark the evidence label on everything in this paragraph: theological doctrine, not historical finding.

Twelver Shia doctrine of isma — extending to the Imams as well as the Prophet — is generally more comprehensive than mainstream Sunni positions. The dominant Sunni theological schools (Ash’ari and Maturidi — the two great frameworks of Sunni systematic theology) typically restrict guaranteed infallibility to the transmission of revelation and religious guidance, allowing ordinary human fallibility in worldly and technical judgment; the date-palm hadith is their classical proof-text. Ibadi theology historically holds a comparatively more restrained view of prophetic authority, consistent with its broader caution about claims of human perfection.

What can history — as opposed to theology — establish? Only this, and it is secure: Muhammad functioned, observably, as judge, military commander, and treaty-negotiator of the Medinan polity. The biographical tradition, the outside notices, and the Constitution of Medina agree on that much. Everything beyond it is theology layered onto history.

And a closing word to the three readers. To the traditionalist: the extension of 7:157 and 59:7 into general legislative principles is a real interpretive move, not a direct textual statement — argue it; don’t assume it. To the Quranist: 66:1 cuts both ways. Yes, it shows the Prophet’s personal judgment was correctable. But by the same token it shows the text distinguishing his personal missteps from a category — revealed guidance — that it does not similarly correct. That is some evidence against collapsing all his authority into “mere conveyance.” To the secular historian: whether any of this constitutes binding law for later generations is not a question archaeology can answer. Ever.

Chapter 5D — The Qur’an’s Favourite Word for Itself

Remember, from Chapter 1, that hadith in ordinary Arabic simply means “speech, discourse, report, story.”

Quranists often cite two verses — 45:6 and 77:50, “in what hadith after this will they believe?” — as the Qur’an itself dismissing the Hadith literature. Traditionalists wave the argument away as a pun. Both responses are too quick.

Instead of cherry-picking, let us do what serious analysis requires: look at every one of the twelve occurrences of the noun hadith in the Qur’an. No filtering, by either side.

Reading the pattern — without apologetics in either direction

Count it up. Of the twelve occurrences:

Nine — 12:111, 18:6, 39:23, 45:6, 52:34, 53:59, 56:81, 68:44, 77:50 — refer to the Qur’an’s own discourse: either explicitly self-referentially, or as the object of the “what discourse after this?” rhetorical formula.

One — 31:6 — refers to frivolous talk explicitly opposed to the Qur’an.

One — 4:87 — refers to divine truthfulness generally.

And 7:185 is ambiguous between the patterns.

None of the twelve refers to a technical genre of post-Prophetic reports about the Prophet’s sayings and deeds. Why not? For the plainest historical reason imaginable: no such named literary genre existed yet at the time of revelation. You cannot condemn — or endorse — a genre that has not been born.

Pause here for a moment, because this is one of those places where both sides of a bitter argument are standing on the same solid ground without noticing. The count is not in dispute. The twelve verses are the twelve verses. Everything contested lies in what a reader is entitled to do with the pattern.

Now hold two conclusions in your head at once, because intellectual honesty demands both:

Against the maximalist Quranist claim. “The Qur’an explicitly names and condemns Hadith literature” is a philological overreach. A word’s earlier, general usage does not preclude a later, narrower institution built on independent theological grounds — the obedience and clarification verses of Chapter 5A. The technical sense of “hadith” postdates the text; reading it back in is anachronism.

Against the maximalist traditionalist dismissal. “The word hadith in the Qur’an has no bearing on this debate at all” understates something genuinely striking: the Qur’an repeatedly, consistently describes itself using the very word that was later applied to a competing source of authority. Nine times out of twelve, “hadith” is the Qur’an’s word for itself. That is, at minimum, rhetorically suggestive — even where it falls short of proof (Izutsu, 1964; J. Brown, 2014, pp. 92–95).

To the three readers, one last time. Traditionalist: the self-referential pattern is real; do not wave it away as a coincidence of common vocabulary. Quranist: retro-projecting a later technical meaning onto an earlier generic noun is a genuine philological error; do not present it as settled exegesis. Secular historian: the pattern itself is plain textual fact. What normative weight a reader gives it is a hermeneutical question — a question of interpretation — not a linguistic one.

Chapter 6 — “Do Not Write Anything From Me”: The Strangest Report in the Canon

Prepare for one of the oddest facts in this entire story.

The best-known report on the writing of hadith — found in Sahih Muslim, in the Book of Zuhd, with parallels in the Musnad of Ibn Hanbal — has the Prophet saying:

“Do not write anything from me except the Qur’an… whoever has written anything from me other than the Qur’an should erase it.”

Read that again, slowly. A hadith — transmitted, written down, canonised in one of the two most authoritative collections in Sunni Islam — prohibiting the writing of hadith.

The tradition preserved, at its own heart, a report against itself.

The tension deepens

It gets more tangled. Because the same tradition also preserves reports pointing the other way:

The Companion Abdallah ibn Amr ibn al-As is reported to have kept a written collection of the Prophet’s words, called al-Sadiqa (“The Truthful”) — with the Prophet’s reported approval (Ibn Hanbal; Abbott, 1967).

And there are hadith explicitly commanding transmission: “Convey from me, even one verse.”

So which was it? Write, or erase?

Cook’s solution

The historian Michael Cook studied the prohibition reports in a detailed 1997 article. His conclusion is elegant, and once you see it you cannot unsee it.

We know — from abundant evidence — that in the second Islamic century (the 700s CE), Muslim scholars were engaged in a genuine internal controversy about whether religious knowledge should be committed to writing or kept oral. Some feared writing would freeze and corrupt what memory kept alive; others insisted writing was the only safeguard.

And what do we find in the hadith corpus? Reports of the Prophet prohibiting writing… and reports of the Prophet endorsing writing.

Cook’s reading: the prohibition reports most likely crystallise that second-century debate — each camp projecting its position back onto the Prophet’s authority — rather than recording one consistent Prophetic policy (Cook, 1997).

The classical tradition had its own harmonisation, and it deserves stating fairly: al-Nawawi — the great thirteenth-century commentator on Sahih Muslim — read the prohibition as an early, situation-specific measure: in the earliest days, revelation and the Prophet’s ordinary words risked being mixed on the same scarce writing materials, so writing anything but Qur’an was banned until the danger passed.

What did the first caliphs do?

The reports about the four Rashidun caliphs and hadith-writing form a fascinating study in evidence quality. Watch the labels.

Abu Bakr — the first caliph — is reported to have collected about five hundred hadith… and then burned them, anxious about transmitting error. A dramatic story, invoked eagerly by both sides. But now check the sourcing: it appears only in al-Dhahabi (d. 748/1348) and Ibn Kathir (d. 774/1373) — writers six to seven centuries after the event — and it is absent from the earlier standard histories (al-Tabari’s great History; Ibn Sa’d’s biographical dictionary, the Tabaqat). Both Musa (2008, pp. 15–20) and Juynboll (1969, ch. 1) flag the chain as weak and the attestation as late. Savour the irony: a report used by both sides as historical evidence is itself unreliable by the tradition’s own chain-criticism standards.

Umar — the second caliph — is better served by the sources. His caution is broadly attested: he discouraged excessive narration of hadith, and reprimanded the Companion Abu Hurayra — the single most prolific hadith narrator — for narrating too freely. But Umar’s caution remains open to the dual reading that haunts this whole book: was it early wariness toward an emerging practice? Or early custodianship of it — quality control by a believer in the enterprise?

Uthman — the third — is barely associated with hadith-recording policy at all. His documented role concerns Qur’anic standardisation (Chapter 2); reports linking him to hadith caution are sparse and comparatively weak.

Ali — the fourth — is credited, in later Shia and some Sunni sources, with keeping a private written document: variously described, including material later associated with legal maxims, and with the disputed “Sahifa of Ali.” Its authenticity and scope are contested across Sunni and Shia scholarship — partly, it must be said, for sectarian reasons: a written document in Ali’s hand matters differently depending on your view of Ali.

So the tradition’s own memory of its first century is divided — write, don’t write, burn, preserve. Which sharpens a question that has been waiting quietly since Chapter 2: why does one of Islam’s two sources carry a divine guarantee of preservation, while the other was left to human hands?

Chapter 7 — One Text Gets a Divine Guarantee. The Other Doesn’t. Why?

Put the question as sharply as it can be put.

The Qur’an says of itself, in verse 15:9: “We have sent down the Reminder, and We will surely preserve it.” An explicit divine promise of preservation.

No comparable promise exists — anywhere in the Qur’an — for reports about the Prophet.

If both the Qur’an and the Hadith function as binding religious authority… why does one carry a divine preservation guarantee and the other rest entirely on human transmission?

This chapter presents each side’s strongest form of the argument. Not the straw man. The steel man.

The strongest Quranist argument

The asymmetry, Quranists argue, is not rhetorical decoration. It is physically measurable — and Chapter 2 already showed you the measurements.

The Qur’an’s consonantal text — the basic skeleton of Arabic letters — stabilised within the first Islamic century, with comparatively limited variation afterward. The witnesses: the Sanaa palimpsest (Sadeghi & Bergmann, 2010; Sadeghi & Goudarzi, 2012), the Dome of the Rock inscriptions, and van Putten’s single-archetype spelling argument (van Putten, 2022).

The Hadith corpus, by contrast, reached canonical form through a multi-century process, with an evolving methodology, producing different canons for different sects — Sunni six, Shia four, Ibadi one.

One stabilisation was fast, unified, and materially verifiable. The other was slow, contested, and sectarian.

Now add the theology: if God explicitly promises to preserve one text and is silent about the other, then treating both as equally binding law erases a distinction the Qur’an itself appears to draw. Why would God guarantee the preservation of the lesser-needed source and leave the equally-binding one to the mercies of human memory?

The strongest traditional response

The traditionalist reply is not a dodge. It is a category distinction — and it has real teeth.

Verse 15:9’s promise, the tradition says, concerns the revealed text specifically. And Hadith was never claimed to belong to that category. No classical scholar ever said hadith reports were divinely preserved. Quite the opposite.

Hadith’s authority was always claimed to rest on something explicitly human: the isnad — the chain of named, examined, fallible transmitters — corroborated by continuous communal practice. And because the transmitters were human, the classical scholars openly acknowledged the corpus was fallible. That acknowledgment is not a modern concession wrung from embarrassed apologists. It is the founding premise of the entire classical enterprise.

Think about what the hadith sciences — ulum al-hadith — actually are. Scholars built a critical apparatus precisely because they knew transmission could fail: biographical dictionaries assessing thousands of narrators; technical grading of every report as sahih (sound), hasan (good), or da’if (weak); refined categories for broken chains, hidden defects, suspicious wording. You do not build a quality-control system for a product you believe is divinely guaranteed.

On this reading, the asymmetry is not a contradiction. It is a division of labour the tradition always recognised: revealed text is preserved by God; reports about the Prophet are preserved — imperfectly — by people, and were graded accordingly.

Where this leaves us

Notice what kind of disagreement remains once both steel men have spoken.

The Quranist says: the asymmetry shows the two sources were never meant to carry equal weight. The traditionalist says: nobody ever claimed they carry equal weight in the same way — one is God’s speech, the other is graded human testimony about God’s messenger.

At bottom, this is a disagreement about whether “equally binding” is even an accurate description of the traditional position — and that is a question of legal theory and theology, not of history. The historian can measure the two transmission processes, and has. The verdict on what the measurement means for religious authority lies beyond history’s jurisdiction.

But there is one place where the abstract question of authority becomes concrete — physical, daily, universal. Five times a day, a billion and a half people enact the answer with their bodies. Time to take the prayer apart, piece by piece, and ask where each piece came from.

Chapter 8 — Anatomy of a Prayer: What Comes From Where

Every practising Muslim prays. Five times a day, in a choreography so familiar it feels eternal: the standing, the bowing, the prostration, the seated recitation, the turn of the head to each side.

Now ask the historian’s question of that choreography. Where does each element actually come from — textually?

This chapter dissects Islamic ritual, element by element, and files every piece under one of six labels: explicit Qur’an; reasonable inference from the Qur’an; Hadith only; later fiqh (jurisprudence — the work of the law schools); regional custom; or historically unknown.

A few terms first, so the tables read smoothly. A rak’a (plural rak’at) is one complete unit of prayer — one cycle of standing, bowing, prostrating. The adhan is the call to prayer sung from the mosque; the iqama is the shorter call made just as the prayer begins. The tashahhud is the recitation made while sitting during prayer. Qunut is a special supplication inserted in certain prayers. Witr is an extra night prayer; tarawih is the special congregational night prayer of Ramadan. And janaza is the funeral prayer.

The master table

The full comparison across the traditions

Now the same territory in higher resolution — with the Sunni, Shia, Ibadi, and Quranist positions side by side, and the earliest material evidence noted where any exists.

The pattern — and the honest fight over what it means

Run your eye down either table and one pattern holds consistently: the broad occasions and general obligation of prayer are not seriously disputed as early and Qur’anically grounded — while nearly every element of prayer’s internal choreography is Hadith- or fiqh-derived.

This is where the traditionalist plays the strongest card in the hand — the one from Chapter 3. On the traditional account, that choreography was transmitted as lived communal practice — parents teaching children, congregation modelling for congregation, in every generation without a gap — before and alongside its later writing-down in Hadith. It was not invented by ninth-century compilers out of nothing.

And the honest reply: whether “transmitted practice” and “textualised report” deserve equivalent evidentiary weight is precisely where the debate remains open. We will hear this argument tested to destruction in Chapter 18.

Case studies: the disputed details, one by one

The Basmala, aloud or silent. Should “In the name of God, the Most Gracious, the Most Merciful” be recited audibly at the start of the opening chapter in congregational prayer? This is a dispute that is not Qur’an-versus-Hadith at all, but Hadith-versus-Hadith, entirely inside Sunni Islam. Shafi’i practice recites it aloud in the audible prayers; Hanafi and Maliki practice does not, or recites it silently. Each school supports its position with a different set of hadith — some of which the opposing school’s own hadith-critics grade as weak (J. Brown, 2009, pp. 199–200). A clean demonstration that “the Hadith position” on a ritual element is frequently not even singular within one branch of Islam.

Reciting behind the imam. Must a worshipper silently recite the opening chapter even while the prayer-leader recites it aloud? Disputed across the schools on the basis of conflicting hadith. And here is the detail to savour: al-Bukhari himself — the great compiler — wrote a dedicated monograph on this single question (Juz’ al-Qira’a khalf al-Imam), defending the position that the worshipper must recite. Which proves something important about chronology: hadith-based legal disagreement at this level of fine grain was already a live, book-length dispute among the top specialists in the ninth century. It is not a later degeneration.

The prostration of forgetfulness (sujud al-sahw). Lose count of your prayer units? Add one by mistake? The tradition prescribes a compensatory prostration — entirely Hadith-derived (Sahih al-Bukhari; Sahih Muslim, Book of Forgetfulness), with the schools disagreeing over its exact triggers and whether it comes before or after the closing salutation. A granular example of fiqh elaborating procedure with no Qur’anic anchor whatsoever.

Shortening versus combining — the cleanest contrast in the book. Watch these two side by side, because nothing illustrates the Qur’an/Hadith distinction better.

Shortening prayer while travelling — qasr — has explicit Qur’anic warrant. Verse 4:101: “When you travel in the land, there is no blame on you if you shorten the prayer, if you fear the disbelievers may harm you.” (Classical exegetes disagree whether the fear-clause is a genuine condition; the majority view drops it, permitting shortening on any journey.)

Combining two prayers into one time-window — jam’ — has no Qur’anic textual basis whatsoever. It rests entirely on Hadith: Sahih Muslim reports the Prophet combining the noon and afternoon prayers, and the sunset and night prayers, while travelling — and, in some reports, even while not travelling, for reasons jurists have debated ever since.

Quranists generally accept qasr — it is explicit — while treating jam’ as the paradigm case of Hadith supplying a rule with zero Qur’anic anchor. The contrast is textually pristine: the Qur’anic warrant for the adjacent practice throws the absence of warrant for this one into brilliant relief.

Friday prayer (jumu’a). Verse 62:9 explicitly commands hastening to “the remembrance of God” when the call is made on “the day of congregation,” and leaving off trade — a rare case where the Qur’an itself names a specific weekly congregational occasion. But everything else — the sermon’s structure, the timing, and the striking fact that Friday prayer replaces the noon prayer rather than adding to it — is Hadith- and fiqh-elaborated. Even whether attendance is individually obligatory (fard ayn) or communally sufficient (fard kifaya — satisfied so long as enough people attend) remains a live juristic disagreement.

Women’s position and participation. Neither separate rows for women, nor the specific spatial arrangements found in later mosque architecture and legal manuals, have explicit Qur’anic basis. The relevant material is entirely Hadith- and later-custom-derived — and actual practice has varied considerably by region and period. This is a point under increasing discussion in contemporary reformist and historical scholarship on gendered ritual space (Duderija, 2011).

The funeral prayer (salat al-janaza). Structurally unlike the daily prayers — no bowing, no prostration; a sequence of standing recitations punctuated by takbirs (declarations of “God is greatest”). It has no Qur’anic textual basis at all, and no rak’at in the ordinary sense. It is entirely Hadith- and fiqh-derived — including a disagreement across the schools about the number of takbirs (four is most widely attested, though some early reports describe more). Reflect on what this is: a ritual practised universally, at every Muslim funeral on Earth, resting on no Qur’anic text whatsoever. Traditionalists explain it under the Prophet’s exemplary conduct — the uswa of Chapter 5C. Quranists cite it as the paradigm case of a required ritual created wholesale after the Qur’an.

Moving the index finger. During the seated tashahhud, should the worshipper raise the index finger? Move it continuously? Once? Not at all? A well-documented dispute resting entirely on differently worded hadith, further split by regional custom. A perfect miniature of how far Hadith-based elaboration can descend — into sub-gestural detail — with no Qur’anic anchor, and with no school claiming otherwise.

Beyond prayer, briefly — but completely

Zakat calculation. The Qur’an commands almsgiving (2:43) and names its recipients (9:60) — but gives no rates and no thresholds. The percentages and the nisab — the minimum wealth at which the duty triggers — are entirely Hadith/fiqh-derived. For traditionalists, this is the paradigm case of mujmal legislation: the Qur’an legislates in summary form, and the Sunna supplies the operational detail. For Quranists, it is a gap God left to communal discretion.

The hajj sequence. Verses 2:196–203 and 22:26–29 outline the pilgrimage’s sanctity and core rites — but not its full choreography. The traditional warrant for the rest is the hadith “Take from me your rites” (Sahih Muslim).

Stoning for adultery (rajm) — and the caliph Umar’s extraordinary “verse of stoning” speech (Sahih al-Bukhari, Book of Hudud) — is treated in full in Chapter 11. It deserves its own stage.

The apostasy punishment. Hadith-based — and in apparent tension with the Qur’an’s 2:256, “no compulsion in religion.” A qualifying minority tradition restricts the punishment to political treason rather than mere change of belief (Saeed & Saeed, 2004).

Women’s dress. Verses 24:31 and 33:59 give general directives; the specific coverage requirements are a matter of continuing juristic disagreement.

The Mahdi traditions — reports about a divinely guided figure who will appear near the end of time. A detail that surprises many Muslims: these traditions are absent from both Bukhari and Muslim, and contested even within the other four Sunni collections (J. Brown, 2009, pp. 96–97).

---

Chapter 9 — “We Found Our Fathers Doing It”: The Problem of Inherited Religion

The verses

The Qur’an contains a recurring scene. A messenger brings revelation. And his hearers answer, in essence: but this is not what our fathers did.

The Qur’an’s judgment on that answer is withering. Read the passages in full context, with their classical interpretation — because context is exactly what this chapter is about:

2:170. “When it is said to them, ‘Follow what God has sent down,’ they say, ‘Rather, we follow what we found our forefathers doing’ — even if their forefathers understood nothing and were not guided.” The classical commentators — al-Tabari at their head — place this in the mouths of Meccan polytheists refusing Muhammad’s monotheism. The “forefathers” here are pre-Islamic idolaters.

31:21. Near-identical wording, same audience.

5:104 and 7:28. The same refusal pattern, again addressed to pagan interlocutors defending idolatrous or superstitious practice. The context of 7:28 is memorably specific: the pre-Islamic custom of circumambulating the Ka’ba — the cube-shaped sanctuary at Mecca — naked, a practice its defenders justified as ancestral and even divinely ordained.

21:52–54. Abraham confronting his own father’s idol-worship — the Qur’an’s paradigmatic story of principled rebellion against inherited religious error. “We found our fathers worshipping them,” say the idolaters. Abraham’s reply: “You and your fathers have been in manifest error.”

43:22–24. The pattern generalised, as a law of history: “We found our fathers upon a way, and we are guided by their footsteps” — spoken, the Qur’an says, by “the affluent ones” of every community to which a warner was ever sent.

The interpretive fork

Now the question on which everything turns.

Across all six passages, the classical exegetical tradition is consistent: the rejected “forefathers” are pre-revelation or anti-revelation communities — idolaters, deniers of a prophet’s message. Not, in the first instance, later generations of the same revealed community elaborating on an already-accepted revelation.

Is that scope restriction sound exegesis? Or a self-serving narrowing that lets the tradition dodge an uncomfortable mirror?

Be honest about what the text itself supports. The verses’ grammar favours the narrow reading: each passage addresses named prior communities defending named erroneous content. But their rhetorical structure — one identical pattern, illustrated six times, capped by 43:22–24’s explicit generalisation to every community in history — arguably supports treating the critique of the reasoning (“our fathers did it” as sufficient warrant, full stop) as a standing principle rather than a period-bound description.

The text alone does not settle which. That fork is exactly where the modern debate stands.

A twist: Islam has been making this critique of itself for centuries

The polemics on both sides usually forget something. Criticism of unreflective following has a long and distinguished history inside mainstream, Hadith-accepting Islam.

The Arabic term is taqlid — literally “placing something around one’s neck,” like a collar or a pendant: accepting a ruling on another’s authority without examining the evidence.

Ibn Taymiyya (d. 728/1328) — the fierce, brilliant, controversial Damascus theologian — and his student Ibn al-Qayyim (d. 751/1350) were both firmly traditionalist Hanbalis, fully committed to Hadith’s authority. And both argued forcefully against rigid taqlid of any single school’s rulings wherever those rulings collided with a sound hadith or a clear Qur’anic text. A scholar’s duty, they insisted, is to follow the strongest evidence, not the school’s inertia. Ibn al-Qayyim’s I’lam al-Muwaqqi’in is the classical statement of the argument.

Six centuries later, the Egyptian reformers Muhammad Abduh (d. 1905) and Rashid Rida (d. 1935) — the circle that produced the journal al-Manar, where, remember, the first modern Quranist article appeared — extended the same critique of blind school-following as part of their Salafiyya reform program. And note carefully: they did so without rejecting Hadith’s authority as such. Their demand was that each generation re-examine transmitted rulings against the primary sources.

Why does this history matter? Because it shows that the Qur’an’s forefather-critique has been mobilised, repeatedly, inside the hadith-accepting tradition, against that tradition’s own complacency. The Quranist application of those verses against Hadith-inheritance itself is a further, more radical extension of a move Islamic scholarship has long made against itself. It is not an argument imported from outside.

The sociological reality — for everyone

Now step back from texts to people.

How do most Muslims — Sunni, Shia, or Ibadi — actually acquire their religion? The answer, uncontroversially: as an integrated package. Family, community, school, local custom. Almost no one independently audits which element of their practice rests on Qur’anic text, which on a Hadith ruling, which on a school’s juristic elaboration, and which on pure local custom.

And before anyone feels superior: this is not distinctive to Islam. It describes religious transmission generally — as Chapters 12 and 15 will show in detail.

Given that, it would overreach to apply verses rhetorically aimed at pagan forefather-worship directly onto the transmission of practices that trace, however indirectly, to the founding revelation and its earliest community. That is the traditionalist’s fair point.

The Quranist’s fair counterpoint: the verses target an epistemic method — uncritical inheritance — not merely a category of content (idolatry). And a method-critique applies wherever the method operates, regardless of the content’s pedigree. The taqlid-critique tradition above shows even insiders half-agreeing.

And now the turn of the knife — in both directions. The same critique applies, with equal force, to Quranist communities’ own transmission of interpretive method from movement founders — Perwez, Khalifa, Kassim Ahmad (Chapter 1) — to their followers. A second-generation Submitter who accepts Khalifa’s interpretive framework on communal trust is doing exactly what the framework condemns. Traditionalist critics raise this point, and they are right to.

One more honesty requirement: no survey or historical evidence establishes which transmission mode — direct textual engagement or institutional inheritance — actually predominates in any community, in any period. Claims about this, on both sides, typically exceed the evidence.

What the psychologists add

Set scripture aside entirely for a moment. Modern psychology has a good deal to say about why inherited religious content feels self-evidently true to those raised inside it — regardless of its actual historical warrant. Four findings, each independent of Islam and of religion in general:

Authority bias. The robustly documented tendency to weight a claim by the perceived authority of its source rather than by independent evaluation. The classic probe is Stanley Milgram’s obedience experiments (1963), in which ordinary people administered what they believed were dangerous electric shocks because a man in a lab coat told them to. The relevance: reports attributed to the Prophet, or rulings attributed to a revered scholar, will be accepted with reduced scrutiny by hearers positioned as subordinate learners. That is not a comment on the reports’ truth. It is a comment on the hearers’ filters.

Conformity. Solomon Asch’s famous line-judgment experiments (1951): participants, asked to compare line lengths, would deny the plain evidence of their own eyes to agree with a unanimous group giving the wrong answer. Directly relevant to why communal ritual practice, once established, resists individual re-examination — regardless of its textual warrant.

Social identity theory. Henri Tajfel and John Turner (1979) showed that group membership functions as identity — and religious affiliation is among the most powerful identity markers humans have, independent of doctrinal content. Revising an inherited practice is therefore costly in identity terms even for a believer who, on reflection, finds its textual warrant thin. The cost is not intellectual. It is social and existential.

Cultural transmission theory. The “epidemiology of representations” model developed by Dan Sperber and Pascal Boyer (1990s–2000s) asks: which ideas survive transmission across generations? Answer: ideas that are memorable, socially reinforced, and mildly counterintuitive spread reliably — largely independent of their truth-value (Sperber, 1996).

Now, the crucial disclaimer, stated with full force: none of this adjudicates whether any specific Islamic practice is textually warranted. Psychology explains mechanism, not merit. And — this cannot be said too often — the mechanisms operate on all sides of this debate, including within Quranist communities, whose adherents are subject to the same biases with respect to their own founders and conventions. There is no methodological high ground here. There is only the discipline of knowing one’s own filters.

Six traditions, one pattern

Finally, widen the lens once more. The pattern — inheriting an integrated package of scripture plus interpretive tradition plus communal custom, rather than scripture in isolation — recurs across every major scriptural tradition on Earth. A preview here; the full treatment is Chapter 12:

Rabbinic Judaism transmits Written and Oral Torah as one integrated whole, from birth.

Karaite Judaism — the Jewish scripture-alone movement — transmits its scripture-alone method as its own inherited communal norm. No less sociologically inherited for rejecting the rabbinic content.

Catholicism transmits scripture inseparably from church tradition and the hierarchy’s teaching authority.

Protestantism, despite its founding principle of “scripture alone,” transmits denomination-specific confessional interpretation — Lutheran, Reformed, Baptist — alongside the biblical text. In practice, “scripture alone” functions as an inherited interpretive lens.

Eastern Orthodoxy explicitly elevates continuous, embodied liturgical and patristic tradition to a status functionally inseparable from scripture — the least text-centric major Christian branch.

Mormonism — the Church of Jesus Christ of Latter-day Saints — is the instructive limiting case. Founded in 1830 on a new, textually fixed revelation (the Book of Mormon), explicitly rejecting inherited denominational tradition… and within two centuries it generated its own dense inherited apparatus of continuing revelation, institutional hierarchy, and communal custom.

Dwell on that last case. Even movements founded explicitly as a corrective return to textual immediacy tend, sociologically, to generate a comparable inherited-tradition layer within a few generations. Weber’s routinisation of charisma, again (Chapter 15).

The implication for Quranism’s own long-term trajectory is direct, and it is offered here as comparative observation, not as gotcha: on the historical evidence, no scripture-alone movement has yet proven exempt from the very process it critiques in others.

Chapter 10 — How Islam Split — and What the Splits Did to Hadith

To understand why different Muslims have different Hadith collections, you need the story of how Muslims came to be different from each other at all. What follows is that story — compressed, but complete.

The wound: succession

The Prophet died in 632 without — on the Sunni account — designating a successor, or — on the Shia account — with his designation of Ali ignored. That disagreement is the oldest wound in Islam, and everything in this chapter flows from it.

Crisis came to a head a generation later. The third caliph, Uthman, was assassinated by mutineers in 656. Ali became caliph; Mu’awiya, the powerful governor of Syria and Uthman’s kinsman, demanded justice for the killing and withheld allegiance. Civil war followed — the first fitna, the trauma at the root of Islamic sectarianism.

The Kharijites — and the Ibadis

In 657, at the Battle of Siffin, Ali’s and Mu’awiya’s armies fought to a bloody standstill, and the two sides agreed to submit the dispute to human arbitration.

A faction of Ali’s own army revolted at that decision. Judgment belongs to God alone, they insisted — not to negotiators. They walked out of Ali’s camp, and history knows them by that act: the Kharijites, “those who went out.”

The Kharijites held the most rigorist — and, initially, the most textually minimalist — position on legitimate leadership: the best Muslim should lead, whoever he is, and grave sin disqualifies. Their militant wings fought everyone and were largely destroyed. But a moderate remnant survived, and became Ibadi Islam — which endures today in Oman and parts of North Africa. In relative geographic isolation, the Ibadis developed their own hadith corpus (the Musnad of al-Rabi’ ibn Habib, from Chapter 1) and their own criteria for evaluating narrators.

The Shia

The party of Ali — shi’at Ali, whence “Shia” — developed around the conviction that leadership belonged to the Prophet’s family line, through Ali and his descendants: the Imams.

That conviction shaped their hadith corpus structurally: Twelver Shia collections privilege narrations transmitted through the Imams — the family line being, on Shia premises, the protected channel of the Prophet’s true teaching.

Two facts about Shia hadith scholarship deserve their own spotlight.

First: formal technical grading of hadith — sorting reports into sound and weak, in the Sunni manner — arrived comparatively late in Shia scholarship: with Ibn Tawus (d. 664/1266) and al-Allama al-Hilli (d. 726/1325), and partly through exposure to Sunni method (Kohlberg, 1987; Newman, 2000).

Second: centuries later, Shia scholarship fought its own great internal war over its own books — the Akhbari–Usuli dispute of the seventeenth and eighteenth centuries. The Akhbaris held that the Four Books were reliable wholesale — no grading needed; the reports of the Imams are simply followed. The Usulis insisted on chain-criticism and rational scrutiny of every report. The Usulis won, decisively — which is why the modern Shia clerical establishment, with its ranked jurists and its marja system (Chapter 4A), exists at all (Gleave, 2007).

The Sunnis — and a forgotten fifth school

Sunni Islam consolidated around acceptance of all four Rashidun caliphs and, eventually, the four legal schools of Chapter 4A. But remember a smaller fifth: the Zahiri school, whose greatest voice was the Andalusian polymath Ibn Hazm (d. 456/1064). The Zahiris were hadith-literalists: they accepted Hadith’s full authority while rejecting qiyas — analogy — as human presumption. Texts, and only texts. The school eventually died out as an organised madhhab, but its existence matters for our map: it proves that “more Hadith” and “more reasoning” were independent dials, not one dial.

The rationalists who still used Hadith

One more group scrambles every neat binary. The Mu’tazila — a rationalist theological school prominent from the eighth to tenth centuries, famous for asserting the Qur’an’s createdness and human free will — were notably skeptical of solitary hadith reports as a basis for theology. A report carried by a single chain — the technical term is khabar al-wahid — yields, they argued, only probability; and creed must rest on certainty.

And yet, in matters of law, the same Mu’tazila were generally far more accepting of those same reports.

Let that complication register. Anyone who wants “rationalist” and “hadith-reliant” to map onto opposing camps — a mapping both modern polemics find convenient — has to reckon with a school that was both at once, depending on the subject matter.

Politics all the way down

Beneath the whole map runs the methodological fault-line of Chapter 4A — Ahl al-Hadith versus Ahl al-Ra’y — and beneath that runs politics.

The succession crisis. Uthman’s assassination and the first civil war. The Umayyad–Abbasid revolution. Later regional fragmentation. Each of these upheavals generated retrospective doctrinal and legal material tied to competing legitimacy claims (Goldziher, 1971; Schacht, 1950). Whose virtues do the hadith praise? Whose sins do they record? Which caliphs do they legitimate? The competing corpora are not incidental to sectarian history. They are structurally woven into it.

Did fragmentation bury a universal core?

A natural closing question: did all this splitting obscure some original, universal, Qur’anic core?

Notice first that historical method alone cannot fully answer it — because the question presupposes a judgment about what that core substantively is, which is itself the contested matter.

But something can be said, and it is worth saying precisely. The broadest theological core — monotheism, prophethood, the five basic practices — remained shared across all three major traditions despite everything: despite civil wars, rival caliphates, competing canons, and a millennium of polemic. The divergence concentrated in legal and ritual detail.

Sectarian development affected operational specifics far more than the shared theological core. Whatever else the splits did, they did not produce three religions.

They did, though, produce three sets of books — each community certifying its own reports. And the strangest evidence about those books comes not from their rivals, but from inside the books themselves.

Chapter 11 — The Witnesses Against Themselves

Some of the most awkward evidence in this entire debate does not come from skeptical professors or Quranist pamphlets. It sits inside the most authoritative Hadith collections — preserved there, faithfully, by the tradition itself. This chapter examines the two most famous cases, and what the classical scholars did about them.

The verse that isn’t there: Umar and the stoning verse

Both Sahih al-Bukhari and Sahih Muslim — the two most authoritative books in Sunni Islam after the Qur’an — preserve a speech by the caliph Umar, delivered from the pulpit in Medina near the end of his life.

In it, Umar declares that a verse prescribing stoning for adultery had been revealed — that the Prophet had applied it, and that Umar himself feared a time would come when people would say, “We do not find stoning in the Book of God,” and would thereby abandon an obligation God had sent down.

Stop and take in what this report is. It is an internal, canonical admission — transmitted through the tradition’s own most rigorous channels — that material some early Companions believed to be Qur’anic revelation did not survive into the mushaf, the standard written Qur’an (Burton, 1977; Motzki, ed., 2000).

The classical tradition did not hide this report. It theorised it. The solution is a doctrine with a formidable name: naskh al-tilawa duna al-hukm — “abrogation of the recitation without abrogation of the ruling.” The idea: God sovereignly withdrew the verse’s text from the Qur’an, while leaving its legal force standing. The wording gone; the law intact.

Assess that doctrine with both eyes open. Is it logically coherent? Yes — nothing contradictory in it, granted the premise of a God who edits His revelation. Is it empirically falsifiable? No — no conceivable evidence could disconfirm it. And does it sit comfortably beside the completeness verses of Chapter 5 — “We have not omitted anything from the Book”? That tension, at least, is real, and the reader is entitled to feel it.

The stranger case: Aisha and the breastfeeding verses

The second case is less famous outside specialist literature — and, if anything, starker.

Background, briefly: in Islamic law, breastfeeding creates kinship. A child nursed by a woman becomes her kin for marriage purposes — her milk-children cannot marry her biological children. The legal question was always: how many feedings establish the bond?

Aisha — the Prophet’s wife, and one of the most important transmitters in all of Sunni Hadith — reported the following (Sahih Muslim, Book of Suckling): the Qur’an originally specified ten known breastfeedings to establish the bond; this was later abrogated to five; and — here is the sentence to read twice — the Prophet died while the “five” ruling was still being recited as Qur’an by some.

Textual instability, on this report, persisting up to the Prophet’s death itself. Reported by his own wife. Preserved in one of the two most authoritative collections. Graded sahih — sound — by that collection’s own criteria.

And that grading is precisely what makes the report so evidentially awkward for maximalist completeness claims: it cannot be waved away as a weak report without breaking the very grading system the tradition relies on everywhere else.

The eighth-century contradiction wars

Perhaps you are now thinking: surely someone noticed these tensions before modern scholars came along?

They did. In the eighth and ninth centuries.

Rationalist critics of Hadith — including circles adjacent to the Qur’an-only opponents al-Shafi’i answered in Chapter 1 — compiled lists of self-contradicting reports as an argument against the corpus’s reliability. The contradiction critique is not a modern discovery. It is one of the oldest arguments in Islamic intellectual history.

And the tradition answered it — in writing. The scholar Ibn Qutayba (d. 276/889) wrote a dedicated book, Ta’wil Mukhtalif al-Hadith — “The Interpretation of Conflicting Hadith” — responding case by case. His toolkit: restrict each report to its original context; sequence apparently conflicting reports chronologically (the later abrogating the earlier); harmonise where possible.

How persuasive is the toolkit? Honestly: uneven — it depends on the case. Some harmonisations are elegant and convincing; others visibly strain. But the book’s existence proves the essential point: the tradition knew its own tensions, engaged them publicly, and considered them answerable. Whether they were answered is a judgment this book leaves with you.

One more inconvenient truth: authenticity was never neutral

A final observation before we leave the internal evidence.

Sunni and Shia collections diverge precisely on the narrations that matter most to their quarrel — reports bearing on succession, on the merits and failings of specific Companions, on the family of the Prophet. Each tradition’s authentication machinery, applied by sincere scholars, systematically validated the reports its community needed and questioned the ones it didn’t.

What does that show? Not that any particular report is false. It shows something subtler and more important: “authenticity,” in practice, has never been a confessionally neutral historical judgment. Historians across the field acknowledge this, whatever their view of any individual disputed report. Keep it in mind whenever anyone — on any side — cites a grading as if it were a laboratory result.

Chapter 12 — The Same Story, Three Times: Judaism, Christianity, Islam

Is the Qur’an-and-Hadith structure unique? Or does it have siblings?

The comparison is illuminating in both directions — it shows what is common to all scriptural civilisations, and it isolates what is genuinely distinctive about the Islamic case. Take the traditions in turn.

Judaism: the Written and the Oral

Rabbinic Judaism — the mainstream of Jewish life for two millennia — rests on a doctrine strikingly parallel to the Sunna: the Oral Torah. Alongside the Written Torah given to Moses at Sinai, the tradition holds, God gave an oral teaching — transmitted from master to disciple for well over a thousand years before being codified.

The codification came around 200 CE, when Rabbi Judah ha-Nasi compiled the Mishnah — the foundational written digest of the oral law. Centuries of rabbinic commentary on the Mishnah then produced the Talmuds (Neusner, 1986).

The structural parallel to Islam is close: scripture, plus a later-codified oral tradition claiming to descend from the founding revelation, becoming the operative basis of daily law. Even the trigger is parallel — remember Chapter 4B: the Mishnah’s codification followed the destruction of the Temple in 70 CE, which removed the priestly centre of authority, just as Hadith canonisation followed the Prophet’s death and the empire’s expansion.

The Jewish Quranists: the Karaites

And Judaism produced its own Quranism.

In the eighth century CE — in the very Islamic world where these debates were raging — a Jewish movement arose rejecting the Oral Torah and the Talmud’s binding authority in favour of the Written Torah alone. Its founding figure, by tradition, was Anan ben David; the movement became Karaite Judaism (“scripturalists,” from the Hebrew for reading).

The Karaites are a near-exact structural precedent for modern Quranism: same move, same target (a post-scriptural oral-turned-written legal corpus), same slogan-logic (back to the text alone).

What happened to them? They flourished for a time — the ninth to eleventh centuries were a golden age — and then persisted as a minority for over a millennium, never displacing majority Rabbinic practice, down to small communities today (Nemoy, 1952; Polliack, ed., 2003).

One lesson from their story deserves underlining, and it cuts in an unexpected direction: minority status, in this comparative case, tracks institutional consolidation at least as much as evidentiary merit. The Rabbanites had the academies, the courts, the communal infrastructure. The lesson: do not treat either tradition’s numerical size — majority or minority — as, by itself, evidence for or against its textual arguments. Truth is not decided by headcount, in either direction.

Christianity: four variations on the theme

The New Testament itself emerged from decades of oral transmission before the written Gospels — a formation history studied minutely by scholars such as James Dunn (2003) and Bart Ehrman (2016), and directly relevant comparative material for evaluating oral transmission generally.

Within Christianity, the scripture-and-tradition question then split four ways — and each way illuminates a corner of our debate:

Catholicism holds scripture and sacred tradition — the councils, the Church Fathers, the ongoing teaching authority called the magisterium — as jointly authoritative and inseparable. Structurally, this is the closest Christian analogue to the Qur’an-plus-Sunna model.

Protestantism raised the banner of sola scriptura — “scripture alone” — in the sixteenth-century Reformation (MacCulloch, 2003), rejecting tradition’s co-authority in principle. Structurally analogous to Quranism. But observe Protestantism in practice: Lutheran, Reformed, and Baptist communities each transmit their own dense confessional interpretation alongside the biblical text. “Scripture alone” has functioned, observably, as an inherited interpretive lens — which bears directly on whether any scripture-alone movement, Quranism included, can fully escape the interpretive-tradition problem it identifies in others.

Eastern Orthodoxy goes further than Catholicism: continuous, embodied liturgical and patristic “holy tradition” is elevated to functional inseparability from scripture. The least text-centric major branch — a useful opposite pole to Quranism on the spectrum.

Mormonism, once more, is the limiting case (Chapter 9): founded 1830 on new, textually fixed revelation, explicitly anti-traditional — and generating, within two centuries, its own thick apparatus of continuing revelation, hierarchy, and custom. A concrete, datable illustration that “return to the text” movements are not exempt from generating their own secondary tradition layer.

The asymmetry that must not be softened

Comparison can slide into false equivalence — “every tradition has this, so the details don’t matter.” Resist that slide, because one asymmetry here is genuinely different in kind, and honesty requires stating it plainly rather than diplomatically.

The Qur’anic consonantal text — on the broad scholarly consensus assembled from manuscripts, inscriptions, and linguistics in Chapters 2 and 7 — stabilised unusually quickly, with comparatively limited variation afterward.

Set the comparators beside it. The New Testament’s manuscript tradition shows substantially greater variation across its history — thousands of manuscripts, hundreds of thousands of variants, most trivial but some significant. And the rabbinic Oral Torah was, by its own tradition’s account, uncommitted to writing for some four centuries after the traditional Sinai revelation.

Now place the Hadith corpus on the map. Its transmission — oral and increasingly written across two to three centuries, filtered through an evolving critical method, issuing in sectarian corpora of substantially different content — is a different kind of process from the Qur’an’s rapid stabilisation… and a materially similar kind of process to the Mishnah’s formation (and, in transmission mode if not in content, to the Gospels’).

That placement — the Qur’an in one category of textual history, the Hadith in another, with the Hadith’s category shared by the Oral Torah and the Gospel tradition — is one of the most historically defensible, non-polemical distinctions available to the Quranist position. It is a finding a secular historian can endorse without adopting anyone’s theology.

Chapter 13 — The Detectives: A Century and a Half of Modern Scholarship

Time to meet the modern scholars properly — the names that have been appearing in citations all through this book. Think of this chapter as a guided tour of a long detective story, in which each investigator inherits the case files, overturns some of his predecessor’s conclusions, and hands the mystery on.

Ignaz Goldziher (Hungarian; his Muhammedanische Studien published 1889–90, translated as Muslim Studies) opened the modern case. His discovery: hadith content correlates suspiciously with later controversies. Disputes that only arose generations after the Prophet’s death — over dynasties, over theology, over rival cities’ prestige — turn out to have hadith conveniently supporting each side. Reports, he concluded, were being generated by the conflicts, then projected backward. As a general phenomenon — leaving aside any individual report — this finding has never been overturned, and remains the shared foundation of the entire field.

Joseph Schacht (1950 — you know him from Chapters 2 and 3) took the argument into law, with two celebrated theses. First: isnads grow backward. Compare early and late citations of the same legal doctrine, and the chains get longer and more complete over time — later sources claim earlier and fuller pedigrees than the earlier sources themselves claimed. Second: legal doctrine generally preceded its prophetic documentation — the “ancient schools” reasoned first, and the hadith arrived afterward to certify.

G.H.A. Juynboll (1983) refined the machinery with the common link method: map every chain of a given hadith, and they often converge on a single transmitter — decades after the Prophet — from whom all versions radiate. That person, the “common link,” is on this method the report’s likely origin point, not merely its busiest transmitter.

Harald Motzki (2002) turned the tools against the skeptics. Using isnad-cum-matn analysis — the chain-plus-wording triangulation you met in Chapter 2 — he argued that some material can be dated earlier than Schacht allowed, in significant cases into the first Islamic century.

Behnam Sadeghi (2008) refined the method’s rigour further — his “traveling tradition test” — and found the results cut both ways: case by case, sometimes earlier than Schacht, sometimes later than Motzki.

Joshua Little (doctoral research at Oxford and subsequent publications) represents the method’s newest, most skeptical generation: re-dating several hadith the Motzki school had treated as early — including a close, high-profile scrutiny of the reports about Aisha’s marriage age. Mark his position on the map precisely: a corrective from within the isnad-cum-matn tradition — not a return to Schacht’s sweeping claims.

Gregor Schoeler (2009, 2011) contributed the transmission model from Chapter 4: the private-notebook middle path, neither purely oral nor properly published.

Jonathan Brown (2007, 2009, 2011, 2014) — among the field’s most influential current voices — documented the corpus’s variable reliability, defended the classical hadith sciences as a methodologically serious critical enterprise (against caricatures of credulity), and coined a distinction this book has used repeatedly: a hadith’s literal truth (did he say those words?), its historical truth (does it reflect something he really taught?), and its effective truth (did the community treat it as true, with real consequences?) are three different properties, and a report may have any combination.

Wael Hallaq (1997, 2005) — from the legal-history side — argued that jurists often constructed doctrine first and anchored it in hadith afterward; you met his demolition of the “closed gate” myth in Chapter 4A.

Fazlur Rahman (1965, 1966) — the great Pakistani modernist, who ended his career at Chicago — offered believers a middle path: read Hadith as largely the early community’s “living tradition,” projected back onto the Prophet — historically invaluable, normatively instructive, but not verbatim court transcript. His “double movement” hermeneutic: first understand a ruling in its original context; then carry its principle — not necessarily its letter — into the present.

And in Qur’anic studies specifically, four names complete the cast. Nicolai Sinai (2017): the compositional-history synthesis, and the “Late Antique rhetoric” reading of the completeness verses (Chapter 5). Marijn van Putten (2022): the single-written-archetype spelling argument (Chapter 2) — the great moderating counterweight to maximal skepticism about the Qur’an’s text. Sean Anthony (2020): the balanced synthesis of documentary and literary evidence. Stephen Shoemaker (2022): the field’s prominent maximal skeptic, arguing the Qur’an’s compilation extended later than the standard narrative — a genuinely minority position in current Qur’anic studies, and one that has drawn specific methodological criticism for its handling of the inscriptional and manuscript record.

The scoreboard: where the field agrees

After a century and a half, genuine consensus exists on five points:

Isnads alone cannot guarantee authenticity. A perfect chain can be forged precisely because everyone knew what a perfect chain looked like.

A significant proportion of the corpus reflects later doctrinal and political dispute. Goldziher’s foundation still stands.

Purely oral, unfixed-transmission models are too simple. Writing entered early (Hammam’s sahifa; Schoeler’s notebooks).

Formal canonisation is a ninth-century phenomenon — even where some underlying material is much older.

Wholesale acceptance and wholesale rejection are both empirically unsupportable. The corpus is neither transcript nor fabrication. It is a mixture — and the mixture must be sorted case by case.

Where the field fights

How much can be dated early with real confidence? Motzki–Sadeghi optimism versus Little–Berg skepticism, case by case.

Is isnad-cum-matn analysis even sound? Herbert Berg (2000) pressed the sharpest objection: the method may be partially circular — it reconstructs early transmission from chain-and-text patterns, but if forgers fabricated chains and texts together, the patterns themselves are contaminated.

Regional legal diversity versus genuine early transmission — which better explains the corpus’s formation?

And in Qur’anic studies: how early was full textual stabilisation? The moderate consensus (van Putten, Anthony, Sinai) versus Shoemaker’s late dating.

And the honest bottom line

The authenticity of the great majority of individual hadith is not resolvable to consensus with currently available evidence.

And here is the harder sentence: this is likely a permanent evidentiary limit — not a temporary gap awaiting one more manuscript discovery (Motzki, ed., 2000; J. Brown, 2009). The materials that could settle it — dated first-century documents of hadith transmission — in all probability never existed in the required form.

Some historical questions do not get answered. They get understood. This is one of them.

And yet the books at the centre of it all did not merely survive the arguments — they rose above them, to a status approaching untouchability. How that happened is a story with dates, places, and names.

Chapter 14 — How the Canon Became THE Canon

An assumption almost everyone makes without checking — Muslim and non-Muslim alike: that Sahih al-Bukhari and Sahih Muslim were recognised as uniquely authoritative from the moment their ink dried.

They were not. And the story of how they became canonical is one of the best-documented — and most under-told — episodes in this whole history.

The dedicated study is Jonathan Brown’s The Canonization of al-Bukhari and Muslim (2007), and its central finding can be stated in one sentence: the “two Sahihs’” elevated status — the doctrine that their contents, as a body, are more reliable than any other collection, approaching consensus-backed certainty — developed gradually, over the tenth to twelfth centuries. That is one to three hundred years after the books were compiled.

What drove the ascent? Three forces, all documentable:

Networks. Regional scholarly networks championed the two books — above all in Nishapur, a great scholarly metropolis in what is now northeastern Iran (Muslim’s home city, and a centre of hadith scholarship), and from there across the widening Sunni world.

Curriculum. The two collections were adopted as teaching and certification texts — the books a student studied, and was certified in, to become a scholar. Canonicity, here as everywhere, followed the classroom. What the schools teach becomes, within a generation, what everyone “has always known” to be authoritative.

Theology — after the fact. Later scholars, most influentially al-Nawawi (the thirteenth-century commentator you met in Chapter 6), articulated a formal doctrine: the community’s universal acceptance of the two books — in Arabic, talaqqi bi’l-qabul, “reception with approval” — itself constituted a form of consensus (ijma), theologically guaranteeing their contents as a body.

Now the detail that changes how you see everything: that strong guarantee-claim was not made by Bukhari or Muslim themselves, nor by their contemporaries. It was constructed generations later, about them. The compilers offered their books as their own rigorous best; the halo was added by posterity.

And the rest of the six-book canon? Later still — and contested. The very membership of the list was fought over: for the sixth slot, Ibn Majah’s Sunan competed against al-Darimi’s Musnad and even Malik’s Muwatta in different regional canons, before broad convergence settled on the list we know (J. Brown, 2007, chs. 2–3).

Why does this chapter matter so much? Because of what it establishes about the nature of canonical authority: it was a historically constructed, gradually achieved status — not a property self-evidently present in the texts from the moment of compilation.

Both sides must reckon with that. The traditionalist cannot treat the canon’s authority as timeless and self-attesting — it demonstrably was not. And the Quranist cannot treat canonisation as a smoke-filled-room conspiracy — it was a slow, public, contested, documentable scholarly process. History, as usual, is more interesting than either polemic.

Step back now from the particulars — the books, the chains, the schools — and ask what kind of process all of this was. It turns out there are thinkers whose entire careers were spent studying exactly this kind of process.

Chapter 15 — The Machinery of Belief: Three Sociologists

Three thinkers — none of them writing about Islam in particular — supply the machinery for understanding how everything in this book happened, without settling whether any specific report is true. Their ideas have been working quietly through earlier chapters; here they get their proper introductions.

Max Weber (1864–1920), the German founding father of sociology (his masterwork translated as Economy and Society, 1978), gave us the phrase this book has used repeatedly: the routinisation of charisma. A founder’s authority is charismatic — personal, immediate, unrepeatable. When the founder dies, the community faces a stark choice: dissolve, or convert that authority into durable forms — offices, institutions, and above all texts — that can be transmitted across generations. Routinisation is that conversion.

Seen through Weber’s lens, some mechanism functionally like Hadith — a written, citable record of the founder’s normative example — is close to a sociological near-inevitability for any religious movement that outlives its founder by more than a generation. And now, the caution this book has attached every time: the lens explains why the institution was likely to emerge. It says precisely nothing about whether any given report inside it is authentic. Weber explains the bottle, not the wine.

Peter Berger (1929–2017), in The Sacred Canopy (1967), coined the concept of plausibility structures: the social machinery — schools, scholarly lineages, communal ritual, the sheer daily company of fellow believers — that sustains a worldview’s felt self-evidence for people who did not independently derive it. Why does inherited religion feel, from inside, like direct perception of truth rather than mediated inheritance (Chapter 9)? Berger’s answer: because an entire social world is continuously, invisibly holding it in place. Remove the structures — emigrate, lose the community — and beliefs that felt like bedrock can suddenly feel optional. The structures were doing more work than the arguments.

Pierre Bourdieu (1930–2002), in his Outline of a Theory of Practice (1977), added the concept of habitus: knowledge that lives in the body. How you pray — the precise fold of the hands, the depth of the bow, the rhythm of rising — is transmitted pre-reflectively: by watching, imitating, being physically corrected as a child. Not by reading manuals.

Bourdieu’s relevance is surgical. He supplies the mechanism behind the traditionalist claim of Chapters 3 and 8 — that prayer’s choreography was carried by lived practice, body to body, independent of and prior to its textualisation in Hadith. Embodied transmission is real, and it genuinely is a different channel from books.

And now the closing discipline, one final time: none of this sociology adjudicates authenticity. It explains why textualised normative authority of some kind was likely to emerge, and why inherited practice feels self-evidently authoritative to those raised within it — on every side of this debate. Whether the feeling tracks historical accuracy is a separate question. The machinery of belief runs equally well on true beliefs and false ones. That is exactly what makes it machinery.

Armed with the evidence, the scholarship, and the machinery, we can now afford one flight of imagination — a question historians ask when they want to find out what a piece of the past is really carrying.

Chapter 16 — Thought Experiment: The Day the Hadith Disappeared

Historians sometimes learn the most from questions about worlds that never existed. So run this one — carefully labelled as a historical thought experiment only, not a theological argument for anybody:

Suppose every Hadith collection had been lost by the year 900. All of them — Sunni, Shia, Ibadi. What could historians reconstruct of Islam?

The available evidence would be: the Qur’an; archaeology; inscriptions; manuscripts; papyri; non-Muslim contemporary sources; and living communal practice. Everything from Chapter 2, minus the report literature.

What would survive

The core theology. Monotheism, prophethood, revelation, final judgment — all fully recoverable from the Qur’an itself, corroborated by the coins and inscriptions.

The broad shape of practice. The obligation and general timing-pattern of prayer; fasting in a lunar month; almsgiving; pilgrimage to Mecca — recoverable from the Qur’anic text plus the early documentary and architectural corroboration (the mosques, the papyri, the early external notices).

The state. The basic administrative and legal existence of an early Islamic polity — the papyri alone secure it.

The Book itself. A text of the Qur’an close to the present one — because, as Chapters 7 and 13 established, the manuscript and inscriptional evidence for the Qur’an is independent of Hadith transmission. The Sanaa palimpsest does not need Bukhari. The Dome of the Rock does not need Muslim. The Qur’an’s textual history stands on its own material feet.

What would be lost

The choreography of prayer. The rak’at counts, the specific wording, the loud-and-silent structure — everything Chapter 8 filed under “Hadith only” and “fiqh.”

The numbers. Zakat’s rates and thresholds — gone.

The pilgrimage script. The full hajj sequence — gone.

The Prophet’s life. Virtually all biographical detail of Muhammad beyond what the Qur’an itself references — and the Qur’an, remember, references remarkably little of it directly.

The law. Virtually all specific rulings in family law, criminal law, commercial law.

The end times. The entire genre of eschatological detail — the Mahdi traditions and their kin.

And — the subtle one — the keys to the Qur’an itself. The interpretive apparatus: the “occasions of revelation,” the asbab al-nuzul of Chapter 5, on which much traditional commentary relies to contextualise otherwise ambiguous verses. Without the report literature, verse 5:3 would still say “this day I have perfected your religion” — but no one would know which day, or where, or in what circumstances. The verse would keep its words and lose its anchor.

What the experiment actually reveals

The twist that makes the exercise worth running is this.

Notice what the two sides would say about this counterfactual world — and notice that they would agree on the facts of it. Both would agree the loss would be enormous for legal and biographical detail. Both would agree the theological core and the Qur’an itself would survive intact.

Their disagreement is about the meaning of the loss. The traditionalist says: this shows how indispensable the Sunna always was — losing it would cripple the religion’s practice, which is precisely why God’s community preserved it. The Quranist says: this shows how much of the religion’s operational detail never had revealed status in the first place — what survives the loss is exactly what God guaranteed, and what perishes is exactly what humans added.

The counterfactual, being a historical exercise, cannot settle that — because that is a question about what was legitimately binding revelation, which is theology. The experiment’s real yield is the map itself: a precise inventory of which parts of Islam rest on which foundations. That inventory, at least, both sides can share.

Chapter 17 — How Do We Know What We Know? The Method, Audited

One chapter before the finale, an audit. Take the book’s most load-bearing claims and ask of each, explicitly: how do we know that — and how securely?

The Qur’an’s early textual stability. How we know: physically dated manuscripts — the Sanaa palimpsest, and the Birmingham fragments (Qur’an leaves held at the University of Birmingham, from the Mingana collection, whose parchment radiocarbon-dates to the mid-600s CE); the dated Dome of the Rock inscriptions; and van Putten’s independent spelling-based archetype argument. Status: consensus on substantial early stability; minority dissent (Shoemaker) on the precise timeline. Among the most secure findings in the book.

Any individual hadith’s authenticity. How we know — or try to: isnad-cum-matn analysis; narrator biography (ilm al-rijal); comparison of wording across versions. Status: genuinely disputed, case by case. No general method commands field-wide agreement on results. The permanent limit of Chapter 13.

Pre-canonical ritual practice. How we know: papyri, architecture, and external notices for the broad outline; textual sources only — with all their dating problems — for legal and procedural detail. Status: secure in outline; speculative beyond it. The four-tier verdict of Chapter 2.

The stoning verse and the breastfeeding reports. How we know: preserved within the most authoritative Sunni collections themselves — hence textually secure as reports. The underlying historical claim — that such wording was once genuinely revealed text — is unfalsifiable by any available method. Note the two-layer structure: certainty about what the tradition said, permanent uncertainty about whether what it said was so.

Canon formation and reception. How we know: the compilers’ own colophons (the scribal notes recording a book’s completion and transmission); the biographical dictionaries; citation patterns across later scholarship (J. Brown, 2007). Status: well documented as a gradual process — among the most secure historical findings in this entire book.

Notice the audit’s shape: the book’s confidence is highest exactly where the evidence is physical or bibliographical, and lowest exactly where it depends on the transmitted memory of the transmitters being transmitted about. That shape is not an accident. It is the method.

---

Chapter 18 — Turning the Guns on Ourselves

A book that spends three hundred pages judging everyone else’s arguments owes its readers one final exercise: the strongest available critique of its own. The original review was, remember, checked against three hostile readers. What follows is what each would say about the book you have just read — at full strength, no straw men — and then a reckoning of which blows land.

The traditional Muslim scholar’s objection

“You have repeatedly treated ‘continuous communal practice’ and ‘chain-transmitted report’ as comparably uncertain (Chapters 3 and 8). But classical legal theory drew a careful distinction you have blurred: between mutawatir transmission — carried by so many independent people, in every generation, that coordinated fabrication is rationally implausible — and ahad transmission, carried by one or a few chains. Nobody ever claimed the mutawatir core rests on any single fallible chain. Collapsing the two categories for rhetorical symmetry misrepresents the tradition’s own, more careful, internal grading.

“And your handling of the stoning and breastfeeding reports (Chapter 11) concedes too little. Abrogation-of-recitation is not an ad hoc rescue device: it is a coherent, textually attested category — preserved, after all, within the most rigorously vetted collections. You called it ‘unfalsifiable’ as though that settled something. Most of theology is unfalsifiable. That is not a defect of theology; it is the definition of it.”

The Quranist’s objection

“Your Chapter 5A gives the traditionalists’ ‘repeated verb’ argument — obey God and obey the Messenger — more weight than comparable linguistic counter-readings receive. And your philological objection to reading 45:6 and 77:50 against the Hadith genre (Chapter 5D) is stated with more confidence than Izutsu’s semantic evidence actually forces. The rhetorical resonance across 39:23, 45:6, and 77:50 — the Qur’an calling itself the best hadith and challenging belief in any hadith after it — does not require the later technical sense to be operative for the parallelism to be meaningful. You conceded the pattern was ‘suggestive’ and then quietly gave it no work to do.

“And your Chapter 9 rejoinder — ‘the forefather-critique applies equally to Quranist transmission’ — is a tu quoque, a ‘you too.’ It does not engage the substantive claim: Qur’an-proximate transmission is at minimum auditable by any reader who can pick up the text. Inherited fiqh-and-Hadith transmission is not equally auditable without years of specialist chain-criticism training. Symmetry of psychology is not symmetry of epistemology.”

The secular historian’s objection

“Your Chapters 2 and 16 occasionally lean on ‘broadly plausible’ characterisations of the standard Rashidun narrative — the Qur’an’s compilation story, the collection committee — which a rigorously source-critical historian would flag as resting substantially on later, chain-bearing Hadith and historiographical material. Methodologically, that material is no more secure, a priori, than the legal Hadith you treat with consistent skepticism elsewhere. Apply the same standard evenly — not more leniently to historical narrative than to legal reports, just because the narrative is load-bearing for your chronology.

“And your comparative chapter (Chapter 12) understates how much the ‘Rabbinic versus Karaite’ and ‘Catholic versus Protestant’ framings are themselves artifacts of later confessional polemic — on both sides of each pair — in ways directly analogous to the Sunni/Shia/Quranist framings you are otherwise so careful to flag as contested. Your comparison cases deserve the same suspicion as your primary case.”

The reckoning

All three critiques land — at least partially. And the honest response is not to concede fully to any one of them, because each pulls in a different direction; it is to hold the tensions explicitly, in the open, where you can weigh them:

To the traditionalist: the mutawatir/ahad distinction is real, and it has genuine epistemic content — the book grants that its collapse anywhere in these pages was too quick. But the distinction’s evidential force still depends on a premise — that “communal practice” was itself accurately transmitted from the founding generation — which is not independently verifiable by non-circular means. The objection narrows this book’s claim. It does not eliminate it.

To the Quranist: correct — auditability by any literate reader is a genuine asymmetry between Qur’an-only and Hadith-inclusive method, and Chapter 7 already granted a version of it. The point deserves its full weight. But auditability is not the same property as historical accuracy — a text can be perfectly auditable and still under-determine practice, as Chapter 19 is about to show — and this book must not imply otherwise.

To the secular historian: yours is the most methodologically forceful of the three. This book has, at points, applied source-critical skepticism more consistently to legal Hadith than to biographical and historical narrative sharing the same transmission mode — and more sharply to the Islamic case than to its Jewish and Christian comparators. That inconsistency is named here plainly rather than smoothed over. Reader: carry it with you. Weigh it against every “broadly plausible” in Chapter 2, and every asymmetry claim in Chapter 12.

No comfortable verdict, then. Just the three tensions, held in the open — which is where they belong.

Chapter 19 — Conclusion: The Question That Remains When the Dust Settles

We began with a gap: two centuries between a Prophet’s death and the books that claim to record his voice. We have crossed that gap in both directions many times now — with coins and palimpsests, with grammarians and sociologists, with the tradition’s own most awkward reports and its most brilliant defenders.

No single finding in this book closes the Quranist–traditionalist debate. It would misrepresent the evidence to pretend otherwise. But the journey has not been circular. Five things can now be stated with reasonable confidence — each one earned, chapter by chapter:

1. The Qur’an and the Hadith have different kinds of textual history. On strong physical and linguistic evidence, the Qur’an’s transmission is of a different order of speed and stability than the Hadith corpus’s transmission (Chapters 7, 12, 13). One stabilised within a lifetime, in writing, from a single archetype. The other took form across centuries, through an evolving method, into different canons for different communities.

2. Part of the corpus demonstrably reflects the later community, not the Prophet. A significant, non-trivial proportion of the Hadith corpus reflects later doctrinal, legal, and political dispute rather than verbatim Prophetic speech. Among specialists this is not seriously contested — even as the proportion, and every specific case, remains fought over (Chapters 6, 11, 13).

3. The outline is early; the detail is not provable. Core Islamic practice in broad outline is well attested from a very early period by evidence independent of Hadith — while its precise legal and ritual detail is not (Chapters 2, 8, 16). The four-tier verdict — known, probable, possible, unknown — stands.

4. Canonicity was achieved, not born. Authority for specific collections was a gradually constructed historical achievement of the tenth to twelfth centuries — not a property the texts carried from compilation (Chapter 14).

5. Both maximalisms exceed the evidence. Full traditionalist confidence in the corpus’s overall reliability, and full Quranist confidence in reconstructing binding practice from the Qur’an alone, both outrun what the assembled evidence supports (Chapters 5, 8, 13, 18).

The better question

Ending with “so — are the Hadith authentic?” would only restate a question this book has shown resists a single verdict. There is a better question, and it is the one the whole investigation has been quietly assembling:

If the Qur’an genuinely claims some form of completeness — and Chapter 5 showed it claims something — what is the minimum amount of religious authority required beyond the text itself for it to function as a practicable religion? And where does the evidence leave that question?

Five answers are live in the literature. None is forced by the evidence alone. Here they are — each with its genuine strength and its genuine unresolved cost. This is the actual menu, laid on the table.

Answer 1: Zero additional authority. The strict Quranist position. The Qur’an, correctly read, is sufficient; every apparent gap — the exact rak’at counts, the zakat rates — reflects either insufficiently careful reading, or matters God deliberately left to communal discretion. The unresolved cost, documented in Chapters 7 through 9: self-identified Quranist communities have not, in practice, converged on a single Qur’an-only ritual system. Some pray five times, some three; forms differ. Which suggests — as a matter of observable fact, not polemic — that the text alone under-determines practice even for readers wholly committed to using nothing else.

Answer 2: Hadith as historical context — never as binding law. Consult the reports as evidence of how the earliest community understood the Qur’an — weighted like any other historical source, and never able to override the Qur’an’s plain sense or add wholly new obligations. This position draws on Jonathan Brown’s literal/historical/effective-truth distinction (Chapter 13) and sits close to Fazlur Rahman’s double movement and much contemporary reformist thought. A workable middle path. The unresolved cost: who adjudicates which hadith carry which kind of truth — once the classical system’s claim to finality has been set aside? The question of authority returns through the side door.

Answer 3: Continuous practice only — no textual Hadith. Accept that lived, uninterrupted communal practice (Chapters 3 and 8) transmits core ritual form independently of any written report — the mutawatir amal argument, Bourdieu’s habitus in religious dress — while declining to grant written reports independent legislative authority. This threads a real needle: it preserves prayer’s basic choreography, which demonstrably predates every hadith collection, while rejecting chain-based reports as a source of new law. The unresolved cost: it needs a workable criterion for distinguishing “the practice” from “the reports describing the practice” — and on inspection (Chapter 8), our knowledge of the practice’s early details is itself largely dependent on those very reports. The needle may not have an eye.

Answer 4: Graded, ranked authority — the classical model (Chapter 4A). Qur’an first; then mass-transmitted Sunna; then sound solitary hadith; then scholarly consensus; then analogy — each tier subordinate to the one above, with the schools disagreeing only over how strictly each tier is policed. This is the historically dominant answer across Sunni, Shia, and Ibadi jurisprudence alike — the three traditions differing chiefly in which corpus, and whose authority (the scholars’ consensus, or the Imams), occupies the tiers below the Qur’an. Its strength is fourteen centuries of institutional viability. Its unresolved cost is everything Chapters 11, 13, and 14 documented about the human, contested, gradually constructed character of the tiers themselves.

Answer 5: A living interpretive office rather than a fixed corpus. The Twelver Shia deputy-and-marja model (Chapter 4A) — and, in a different register, the classical Sunni system of judges and muftis. On this model, what ultimately binds is less a fixed body of reports than an ongoing, institutionally accountable interpretive process — with Hadith functioning as evidence submitted to that process, rather than as self-executing law. Its strength: it mirrors how legal systems actually work everywhere. Its cost: it relocates the authority question from texts to institutions — and institutions, as this book has shown at length, have histories.

Where this book stops — and why

This book does not choose among the five. Deliberately.

Choosing would require settling questions that history cannot settle: the true scope of din in 5:3; whether the “clarification” delegated in 16:44 licenses independent legislation; the nature and extent of the Prophet’s protection from error. Those are theological questions. Chapter 1 promised they would be treated as such, and the promise holds to the last page.

What this book has tried to establish is the more limited — and more useful — claim: the honest range of defensible answers is exactly the range above. Not “the Hadith are simply authentic.” Not “the Hadith are simply illegitimate.” A genuine spectrum of five positions, each with real textual support and real unresolved cost — on which further progress depends on argument within theology and law, not on any additional historical discovery this review can foresee.

And where the book could not reach a conclusion, it has said so, in so many words: the dating of most individual hadith; the precise moment ritual detail crystallised; whether 16:44’s clarification licenses legislation; what exactly was “completed” on that day at Arafah — or at Ghadir Khumm.

That was the discipline promised in the introduction: fact distinguished from consensus, majority from minority, dispute from resolution, theology from history — all the way through, and all the way down.

The evidence is now yours. So is the verdict.

---

References

(The complete scholarly apparatus of the original academic review, retained in full. Every work cited in this book appears here.)

Abbott, N. (1967) Studies in Arabic Literary Papyri II. Chicago: University of Chicago Press.

Anthony, S.W. (2020) Muhammad and the Empires of Faith. Oakland: University of California Press.

Asch, S.E. (1951) “Effects of Group Pressure upon the Modification and Distortion of Judgments,” in Guetzkow, H. (ed.) Groups, Leadership and Men. Pittsburgh: Carnegie Press.

Berg, H. (2000) The Development of Exegesis in Early Islam. Richmond: Curzon Press.

Berger, P.L. (1967) The Sacred Canopy: Elements of a Sociological Theory of Religion. New York: Doubleday.

Bourdieu, P. (1977) Outline of a Theory of Practice, trans. R. Nice. Cambridge: Cambridge University Press.

Brown, D.W. (1996) Rethinking Tradition in Modern Islamic Thought. Cambridge: Cambridge University Press.

Brown, J.A.C. (2007) The Canonization of al-Bukhari and Muslim: The Formation and Function of the Sunni Hadith Canon. Leiden: Brill.

Brown, J.A.C. (2009) Hadith: Muhammad’s Legacy in the Medieval and Modern World. Oxford: Oneworld.

Brown, J.A.C. (2011) “Did the Prophet Say It or Not?” Journal of the American Oriental Society 131(1), pp. 1–24.

Brown, J.A.C. (2014) Misquoting Muhammad. London: Oneworld.

Burton, J. (1977) The Collection of the Qur’an. Cambridge: Cambridge University Press.

Cook, M. (1997) “The Opponents of the Writing of Tradition in Early Islam,” Arabica 44(4), pp. 437–530.

Donner, F.M. (2010) Muhammad and the Believers. Cambridge, MA: Harvard University Press.

Duderija, A. (2011) Constructing a Religiously Ideal ‘Believer’ and ‘Woman’ in Islam. New York: Palgrave Macmillan.

Dunn, J.D.G. (2003) Jesus Remembered. Grand Rapids: Eerdmans.

Dutton, Y. (1999) The Origins of Islamic Law. Richmond: Curzon Press.

Ehrman, B.D. (2016) Jesus Before the Gospels. New York: HarperOne.

Gleave, R. (2007) Scripturalist Islam: The Akhbari Shi’i School. Leiden: Brill.

Goldziher, I. (1971 [1889–90]) Muslim Studies, trans. C.R. Barber & S.M. Stern. London: Allen & Unwin.

Görke, A. & Schoeler, G. (2008) Die ältesten Berichte über das Leben Muhammads: Das Korpus ’Urwa ibn az-Zubair. Princeton: Darwin Press.

Hallaq, W.B. (1984) “Was the Gate of Ijtihad Closed?” International Journal of Middle East Studies 16(1), pp. 3–41.

Hallaq, W.B. (1997) A History of Islamic Legal Theories. Cambridge: Cambridge University Press.

Hallaq, W.B. (2005) The Origins and Evolution of Islamic Law. Cambridge: Cambridge University Press.

Hamidullah, M. (1953) Sahifah Hammam ibn Munabbih. Damascus/Paris.

Hoyland, R.G. (1997) Seeing Islam as Others Saw It. Princeton: Darwin Press.

Ibn Qutayba (9th c.) Ta’wil Mukhtalif al-Hadith. Various editions.

Izutsu, T. (1964) God and Man in the Qur’an. Tokyo: Keio Institute.

Jeffery, A. (1938) The Foreign Vocabulary of the Qur’an. Baroda: Oriental Institute.

Juynboll, G.H.A. (1969) The Authenticity of the Tradition Literature. Leiden: Brill.

Juynboll, G.H.A. (1983) Muslim Tradition. Cambridge: Cambridge University Press.

King, D.A. (1999) World-Maps for Finding the Direction and Distance to Mecca: Innovation and Tradition in Islamic Science. Leiden: Brill.

Kohlberg, E. (1987) “Aspects of Akhbari Thought,” in Levtzion, N. & Voll, J.O. (eds.) Eighteenth Century Renewal and Reform in Islam. Syracuse: Syracuse University Press.

MacCulloch, D. (2003) The Reformation: A History. New York: Viking.

Melchert, C. (2006) Ahmad ibn Hanbal. Oxford: Oneworld.

Milgram, S. (1963) “Behavioral Study of Obedience,” Journal of Abnormal and Social Psychology 67(4), pp. 371–378.

Motzki, H. (ed.) (2000) The Biography of Muhammad: The Issue of the Sources. Leiden: Brill.

Motzki, H. (2002) The Origins of Islamic Jurisprudence, trans. M.H. Katz. Leiden: Brill.

Musa, A.Y. (2008) Hadith as Scripture. New York: Palgrave Macmillan.

Nemoy, L. (1952) Karaite Anthology. New Haven: Yale University Press.

Neusner, J. (1986) The Oral Torah. San Francisco: Harper & Row.

Newman, A.J. (2000) The Formative Period of Twelver Shi’ism. Richmond: Curzon Press.

Polliack, M. (ed.) (2003) Karaite Judaism. Leiden: Brill.

Rahman, F. (1965) Islamic Methodology in History. Karachi: Central Institute of Islamic Research.

Rahman, F. (1979 [1966]) Islam, 2nd edn. Chicago: University of Chicago Press.

Robinson, C.F. (2003) Islamic Historiography. Cambridge: Cambridge University Press.

Sadeghi, B. (2008) “The Traveling Tradition Test,” Islamic Law and Society 15(2), pp. 203–242.

Sadeghi, B. & Bergmann, U. (2010) “The Codex of a Companion of the Prophet,” Arabica 57(4), pp. 343–436.

Sadeghi, B. & Goudarzi, M. (2012) “Sana’a 1 and the Origins of the Qur’an,” Der Islam 87(1–2), pp. 1–129.

Saeed, A. (2006) Interpreting the Qur’an. London: Routledge.

Saeed, A. & Saeed, H. (2004) Freedom of Religion, Apostasy and Islam. Aldershot: Ashgate.

Schacht, J. (1950) The Origins of Muhammadan Jurisprudence. Oxford: Clarendon Press.

Serjeant, R.B. (1978) “The Sunnah Jami’ah, Pacts with the Yathrib Jews, and the Tahrim of Yathrib: Analysis and Translation of the Documents Comprised in the So-Called ‘Constitution of Medina’,” Bulletin of the School of Oriental and African Studies 41(1), pp. 1–42.

Schoeler, G. (2009) The Genesis of Literature in Islam, trans. S.M. Toorawa. Edinburgh: Edinburgh University Press.

Schoeler, G. (2011) The Biography of Muhammad: Nature and Authenticity, trans. U. Vagelpohl. London: Routledge.

Shoemaker, S.J. (2022) Creating the Qur’an. Oakland: University of California Press.

Sinai, N. (2017) The Qur’an: A Historical-Critical Introduction. Edinburgh: Edinburgh University Press.

Sperber, D. (1996) Explaining Culture: A Naturalistic Approach. Oxford: Blackwell.

Tajfel, H. & Turner, J.C. (1979) “An Integrative Theory of Intergroup Conflict,” in Austin, W.G. & Worchel, S. (eds.) The Social Psychology of Intergroup Relations. Monterey, CA: Brooks/Cole.

van Putten, M. (2022) Quranic Arabic: From Its Hijazi Origins to Its Classical Reading Traditions. Leiden: Brill.

Weber, M. (1978 [1922]) Economy and Society, eds. G. Roth & C. Wittich. Berkeley: University of California Press.

Whelan, E. (1998) “Forgotten Witness,” Journal of the American Oriental Society 118(1), pp. 1–14.

A note on the text. This reader’s edition preserves the complete evidentiary and scholarly content of the original historical-critical review — every argument, manuscript, inscription, coin, papyrus, scholar, debate, date, caveat, counterargument, and reference — re-expressed for the general reader, with all technical terms and historical figures introduced on first appearance. Positions cited within — on manuscript dating, isnad-cum-matn methodology, and early Islamic historiography — remain live scholarly disagreements, and should be read as provisional. The evidence labels defined in the Introduction apply throughout: where a claim is fact, consensus, majority view, minority view, disputed, theological, legal, customary, or speculative, the text says so — and where the honest answer is “unknown,” it says that too.