In April 1926 the qadi Ibn Bulayhid issued a fatwa, and the Ikhwan — the tribal fighters who had just conquered the Hejaz for Abd al-Aziz ibn Saud — set about leveling the tombs of the Baqi cemetery in Medina. Under the domes lay the graves of the Prophet's daughter, his grandson, his wives, and his companions, and for something like a thousand years Muslims had come to stand at them and pray beside them. The Wahhabi verdict held that the standing and the praying were idolatry, that a dome over a grave was a road to shirk, the one sin the Quran says God does not forgive. So the domes came down.

To most of the Muslims alive in 1926 that was desecration. Visiting the graves of the righteous, seeking their intercession, marking a saint's day — until roughly the middle of the nineteenth century a majority of Muslims, or close to it, practiced a version of Islam in which those acts were the ordinary texture of devotion, as the scholar Omid Safi puts it, with no clean wall between a "mystical" Islam and an "orthodox" one. The men who razed al-Baqi were one current acting on one reading, against practices the bulk of the tradition had blessed for centuries.

A monolith does not demolish its own shrines. And a tradition with no center does not produce a fight this bitter over precisely which acts cross the line into idolatry, because the argument is over where the line falls, and the everything-goes reading says there is no line.

The disagreement is real and old, and often violent. It runs along a small number of identifiable fault lines — where religious authority sits, how geography and prior culture shaped practice, how modern states built official creeds, whether mysticism belongs at the center or beyond the wall, and how rival renewal movements answer the same modern crisis. Each line has a settled anchor and a live contest.

I get asked, when I explain this, which of the two errors is the worse one, the alarmist's monolith or the apologist's everything-goes. The honest answer is that they are the same error facing opposite directions. Both refuse to look at where the disagreement actually falls. What follows is an attempt to look.

Executive Summary

Islam is internally plural, and the pluralism has a shape. It runs along a small set of axes, each with a settled anchor and a live contest, and the reader who can name the axes can put down two opposite and equally common errors in a single move.

The first error is the monolith — 2 billion people running one fixed program, their disagreements cosmetic. The second is the apologist's mirror of it, the claim that the authentic Islam is innocent of whatever ugly thing was quoted at it. Its stronger form drops the essence altogether and holds that Islam is infinitely flexible and has no center, so that nothing can be located inside it or outside it. Neither version looks at the map.

The map has five main axes. They are a frame imposed here to make the argument legible, and the tradition does not sort itself this way. Authority: the Sunni-Shia divide is a structural disagreement about where religious authority legitimately sits, and it propagates into different sources of law and, on the Shia side, a living clerical hierarchy Sunnism never built. Geography: the largest Muslim populations are not Arab, and regional Islams differ in substance through a named legal mechanism, custom admitted as an ancillary source with its own explicit edge. State-engineering: the two loudest official Islams, Saudi Wahhabism and the Iranian velayat-e faqih system, are, in the reading of the scholars who study them, modern state constructions, each claiming an ancient inheritance it does not have. Mysticism: Sufism's very standing is one of the tradition's contests, mainstream for most of Islamic history and condemned by revivalists, so "the moderate Islam we like" and "a heresy" both misfile it. Renewal: modernist reform and the Salafi revival are rival answers to the same modern crisis, sharing a genealogy and a textual move, which is why grading one enlightened and the other backward misdescribes both.

The tradition has a settled center, small and real. Its own boundary document, the 2004 Amman Message, names the shared core and forbids mutual excommunication, and the classical machinery draws its lines on identifiable anchors. But the center is narrow. Universal consensus, Kamali notes, is a claim the factual evidence has hardly ever borne out. Most of the tradition lives in the licensed contest around that core.

This is Part 2 of three. Part 3 takes the map into the question the Jewish reader wants answered, where the law itself was settled early and what varied was its administration, its enforcement, and what it was like to live under.

"Monolith" and "Anything Goes" Are the Same Analytical Failure

Two readers never meet, and they are the reason this series exists. One reads the newspaper and concludes that Islam is a fixed civilizational essence. Bernard Lewis wrote in 1990 that the confrontation was "no less than a clash of civilizations," the reaction of "an ancient rival against our Judeo-Christian heritage, our secular present, and the worldwide expansion of both" (the polemical Lewis of that essay, and Part 3 leans on the careful historian of The Jews of Islam for the opposite service), Samuel Huntington built that intuition into a foreign-policy paradigm three years later, and the commentariat after 2001 inherited a world in which "Islam" is a single actor with a single will. The other reader, hearing any ugly verse or ruling quoted, replies that this has nothing to do with real Islam, that the authentic thing is innocent of it. The two despise each other. They are running the same program.

Edward Said diagnosed the mechanism on the alarmist side, the way Western discourse converts Islam's variety into a monolithic, unchanging essence, a caricature standing in for a civilization. Aaron Hughes caught it on the apologist side, and his is the sharper observation for our purposes, because it is less expected. The "authentic Islam" that conveniently matches liberal values, Hughes argues, is the mirror image of Orientalist essentialism. Both reify Islam into a thing with one essence. One says the essence is violence, the other says the essence is peace, and neither has looked at where Muslims actually disagree, because both decided in advance that the disagreement is noise.

The disagreement they both dismiss is the tradition's oldest feature, and one it built on purpose.

Islamic legal theory has a word for licensed disagreement, ikhtilaf, and it draws a hard line through it. Mohammad Hashim Kamali, among the leading living authorities on Islamic legal theory, separates ikhtilaf proper, where each opposing view rests on valid evidence, from khilaf, disagreement that rests on none. Al-Shafi'i drew the line by subject, forbidding disagreement over what a clear text has settled and permitting it where the evidence is open to interpretation, against the Quran's warning to those who "are divided amongst themselves and fall into disputations after receiving clear signs" (Q 3:105). Islamic law is often described, in the formulation Kamali reports, as a diversity within unity, unity as regards basic principles and diversity regarding details, with ikhtilaf tolerated as a matter of principle and no one entitled to pressure a jurist out of his convictions. The survival of multiple living schools for twelve centuries is the evidence that the tradition institutionalized the argument instead of ending it.

What keeps that from collapsing into "anything goes" is that the tradition specifies where the argument is allowed. The governing distinction is level of evidence. A ruling grounded in a definitive text, qati, admits essentially no legitimate disagreement. A ruling resting on speculative evidence, zanni, is the proper field of ikhtilaf. Consensus, ijma, is a recognized source, but it is subservient to the Quran and the Sunna and can never overturn an explicit injunction, and it carries a specific technical power. Once a genuine consensus forms behind a speculative ruling, that ruling hardens into a binding one, which is the exact machine that converts an open question into a closed one. Beneath all of it sits a further category, the anchors a Muslim is held to know by necessity, and the final section takes up where they fall.

King Abdullah II of Jordan issued the Amman Message in November 2004 and convened the scholarly conference that produced its Three Points the following July, and over the year that followed it drew 552 endorsements from scholars in 84 countries. It recognized the validity of eight legal schools — the four Sunni madhhabs, the Shia Jafari and Zaydi, the Ibadi and the Zahiri — plus Ashari theology, authentic Sufism, and true Salafi thought. And in the same breath it forbade takfir, the declaring of any of them an apostate, and barred anyone lacking the training and method from issuing fatwas. One text, both moves. The inside is plural. The edge is real. We will come back to this document, because it is the closest thing the argument has to a keystone: the settled endorsement of a broad scholarly consensus, which the tradition takes seriously even where no court enforces it. A king convened it, and kings convene things for reasons. Jordan had an obvious interest in denying takfiri movements their license. So take the document for the one thing it can carry, evidence of what a very broad body of scholars will put their names to in public.

The apologist's favorite proof-text is a line attributed to the Prophet: "the difference of opinion in my ummah is a mercy." It gets quoted to mean that every position is equally valid, that the tradition blesses disagreement without limit. The trouble is that the line is a fabricated hadith. The hadith critics went looking for its chain of narration and found none at all, no transmission sound, weak, or forged. A case for internal pluralism built on a forged prophetic saying is a weak case, and worth refusing even when it points the right way. The tradition does bless a certain bounded disagreement as a mercy, and it says so in its own sound sources. That distinction, between the disciplined version and the free-floating abuse of it, is one the final section will need, so hold the chainless line here and watch for what the tradition put in its place.

None of this makes the tradition tidy. The anthropologist Talal Asad described Islam as a discursive tradition, and its orthodoxy as a relationship of power to truth, the ongoing contest over whose reading is authoritative. That is a tradition with a fiercely contested center, which is a different object from a tradition with no center at all. Shahab Ahmed pressed the same point from the other side, insisting that any honest account of Islam has to hold together the fact that it is at once internally contradictory and coherent, and that treating the contradiction as a defect to be explained away is itself the error. The monolith and the everything-goes are two ways of refusing the contradiction.

The Sunni-Shia Divide Is a Dispute Over Where Authority Lives

The most reliable way to get the Sunni-Shia divide wrong is to date it to a single week in 632 and call it a family quarrel that never healed. The quarrel was real. When the Prophet died, one camp held that the community should choose his successor and another held that authority belonged to Ali, his cousin and son-in-law, and to the house of the Prophet. But the reduction to a succession grudge misses what the disagreement became, which is a structural dispute about where religious authority legitimately resides, and that dispute reaches all the way down into law, theology, and the shape of the clergy.

Wilferd Madelung's study of the early caliphate cuts against the frozen-in-632 picture from inside the sources. The classical Sunni theory of an elected caliphate, he shows, is "obviously posterior to the succession," a doctrine worked out afterward to justify what had happened. The Quran, he notes, grants the Prophet's kin an elevated station and advises the faithful to settle some affairs by consultation, "but not the succession to prophets." Both sides were reading a text that did not simply hand them the answer, and the electoral principle Sunnis treat as obvious was itself a construction. Madelung himself leans toward the Alid case, but the point borrowed here is the narrow one, that the electoral principle came after the event it justified, and that much is not seriously contested.

From that root the two systems diverge on doctrine that has nothing to do with old grievances. Twelver Shiism holds the imamate to be a divine office, its holder fixed by explicit textual designation, nass, rather than chosen, and the imam is held to be divinely protected from error, possessed of isma, infallibility, which Sunni Islam refuses to any figure after the Prophet. The imam is held to be God's proof on earth, a metaphysical necessity, so that no age can be without one. That doctrine forced a crisis and then a structure. The twelfth imam, in the Twelver account, did not die but entered concealment, the occultation, ghayba, its minor phase running from 874 to 941 through four successive deputies and its major phase from 941 to the present. With the imam hidden, his juristic authority devolved onto qualified scholars, the mujtahids, and this is why Twelver Shiism grew a living, hierarchical clergy that Sunnism never developed.

That clergy is younger than it looks. Usuli Twelver doctrine requires every believer to follow a living marja al-taqlid, a source of emulation, and the requirement that he be living means authority cannot fossilize in dead founders the way it does around the four Sunni school-eponyms. A marja rises by peer recognition of superior learning and by mass emulation rather than by appointment, and the religious tax his followers pay him, khums, gives senior Shia clergy a financial independence Sunni scholars generally lack. But the mature institution is recent. Murtada al-Ansari, who died in 1864, is generally identified as the first universally recognized marja, and the Usuli school that requires all of this only defeated its Akhbari rivals in the late eighteenth century. This is the first appearance of a pattern worth naming now, because it recurs. The forms of Islam that most loudly claim ancient authority are frequently the youngest.

The divide reaches ordinary law through the sources each side will use. Sunni jurisprudence runs on the Quran, the Sunna, consensus, and analogy, qiyas. The Twelver Jafari school, named for the sixth imam Jafar al-Sadiq, draws on the Quran, the Sunna, consensus, and reason, aql, and characteristically rejects qiyas. Kamali records the Imami position that analogy is "pure conjecture," the guidance of the Quran, the Sunna, and the rulings of the imams being held sufficient. The point wants stating precisely, because the Usuli do accept certain limited analogies by way of reasoned effort, so this is a rejection of qiyas as a formal source and not a ban on all reasoning. On consensus the Shia narrow it further, counting only the agreement that discloses the view of the Prophet's family. Different sources produce different law. The Jafari rules on temporary marriage, on the fifth-tax, and on inheritance shares depart from the Sunni schools.

The binary conceals a spectrum even on the Shia side. Zaydism, the Fiver branch concentrated in northern Yemen, drops most of what makes the contrast sharp — no infallible imam, no hidden imam and no occultation, no father-to-son requirement, the imamate earned by a learned descendant of Ali who rises in arms — which is why it is usually called the Shia branch closest to Sunni Islam. Ismailism seceded from the common Imami trunk over which son of Jafar al-Sadiq carried the imamate, and its Nizari line keeps a living, present imam, today the Aga Khan, the exact mirror of the Twelver hidden one. A tradition that runs from a Zaydi imam-in-arms to a Twelver imam-in-occultation to an Ismaili imam-in-the-room holds a range of answers to one question about where designated authority went, which is more than a binary can carry.

None of which makes Shiism a deviation, whatever the takfiri polemic says, and here the tradition has ruled on itself. The Amman Message recognizes the Jafari and Zaydi schools by name as valid Islam and declares excommunicating their adherents impermissible, a judgment carried by scholars across the Sunni world. The demographic shape is worth keeping in view too, since it corrects a Western reflex to treat Shiism as the tradition's center of gravity. Shia Muslims are between a tenth and an eighth of the total, on Pew's counting somewhere between 154 and 200 million people, most of them in just four countries, with roughly four in ten of the world's Shia living in Iran alone. The divide is doctrinal, and it is lopsided. And it built institutions, which is what a live argument about authority does over time.

Indonesian Islam and Saudi Islam Diverge Where Custom Becomes Law

If the authority axis is the tradition's deepest fault line, geography is its widest, and it is the one that most directly demolishes the monolith. The four Sunni schools of law are stratified across the map almost like dialects. Hanafi law predominates across South and Central Asia, Turkey, the Balkans, and the Caucasus. Maliki holds North and West Africa. Shafii runs down East Africa and across Southeast Asia. Hanbali, the school of the Saudi state and the one Western coverage treats as the Sunni default, governs Arabia, and it is the smallest of the four, barely one percent of the world's Sunnis, with a majority only in Saudi Arabia and Qatar.

The demographic picture is the same shape. Roughly 240 million Muslims live in Indonesia, about an eighth of the global total and nearly all of them Shafii, which makes the single largest concentration of Muslims on earth non-Arab and Southeast Asian. Around fifty-nine percent of Muslims live in the Asia-Pacific region and about a fifth in the Middle East and North Africa. The great majority of Muslims worldwide are not Arab. These numbers settle where most Muslims live and what most of them practice. They do not settle which forms of Islam a state has built and funded into prominence, which is a different question and the one the fourth section takes up. But an analyst who reaches for Riyadh or Tehran to answer the first has reached for the loud statistical margin and skipped the bulk of the field.

The variation is not cosmetic, because the histories that produced it were not. Islam reached the Indonesian archipelago through merchants and Sufi teachers from around the twelfth and thirteenth centuries, spreading by the conversion of local rulers rather than by conquest, and the Sultanate of Demak displaced the Hindu Majapahit only in the 1520s. The Wali Songo, the nine saints credited with Islamizing Java, did their work by harmonizing the new faith with the animist and Hindu-Buddhist and customary forms already there, so that older cultural material was reframed as Islamic instead of erased. Clifford Geertz, mapping Javanese religion in 1960, split it into the syncretic abangan of the rural mass, the scripturally observant santri, and the court-gentry priyayi with its Hindu-Buddhist inflection. Later scholars, Andrew Beatty and Ahmad Najib Burhani among them, have pushed back on the neatness — priyayi looks more like a class than a creed, and most Javanese live in the grey zone between santri and abangan. The categories are contested. The variation they were drawn to describe is not.

Custom, urf, called adat in the Malay world, is a recognized ancillary source in Islamic legal theory. A custom that does not contradict the Quran or the Sunna is urf sahih and can specify or qualify a ruling where no definitive text governs. A custom that conflicts with them is urf fasid and is thrown out. The maxim al-adah muhakkamah, custom is authoritative, was one of the classical legal maxims and was written into positive law in the Ottoman Majalla of 1869. It carries its own limit inside it. The Maliki jurist al-Qarafi held that a ruling built on a custom lapses when the custom changes, and that custom can never override an explicit command of the Quran or the Sunna. Regional Islam is structured license. The Maliki and Hanafi schools give custom the most weight of the four, which is part of why North African and South and Central Asian practice looks so localized — though the largest localized Islam of all, Indonesia's, runs on Shafii law, so the mechanism is not the school alone.

Follow that license across regions and you find the same devotional act blessed in one place and damned in another, inside the tradition. North African maraboutism, the veneration of local living and dead holy men at tombs sited by springs and caves and trees, was long read by colonial ethnographers as a pre-Islamic Berber survival, a reading later scholarship has largely dismantled. In South Asia the identical shrine-and-saint devotion is piety to the Barelvi movement and pre-Islamic superstition to the Deobandi movement, two currents born within a generation of each other in the same colonial society. In West Africa the Sufi orders — Qadiriyya, Tijaniyya, and the Senegalese Muridiyya — are the organizational spine of the whole regional religion. Each of these is a live intra-Islamic argument about how much of the local past a Muslim may keep.

The apologist reaches for this material to prove that Islam is anything a community wants it to be, and overshoots. Regional variation runs on a named legal mechanism with an explicit edge, custom admitted only where it does not contradict a definitive text and thrown out where it does. Indonesian Islam and Saudi Islam genuinely differ in substance, in the law they run and the customs they carry, and both sit inside the same tradition, arguing under the same usul about where custom stops.

Saudi Wahhabism and Iran's Velayat-e Faqih System Are Two Modern State Constructions

Part 1 left two doors deliberately shut, the modern movement that grew in Hanbali soil and the doctrine of the guardianship of the jurist, and promised to open them here. This is the room. Every claim below belongs to a named scholar who studies the system in question, and the argument is about the history of institutions.

In 1744, at Diriyah, a local amir named Muhammad ibn Saud and a preacher named Muhammad ibn Abd al-Wahhab struck a pact. David Commins, whose history of the mission is the standard account, records the compact in plain terms. The shaykh held that God might compensate the amir with booty and legitimate taxes greater than what he took from harvests, and Commins calls it "the origin of the pact between religious mission and political power that has endured for more than two and half centuries." It did not condense out of the Arabian air.

What that contract founded was a departure from mainstream Hanbalism. Cole Bunzel, the leading current historian of the movement, describes Wahhabism as a radical eighteenth-century revival that leaned on the medieval Hanbali Ibn Taymiyya and his student Ibn Qayyim but re-articulated them, told its followers to set aside the Hanbali legal manual then standard in Najd, and practiced takfir, the excommunication of professed Muslims, on a scale its predecessors did not. Wahhabism entered the tradition as a fight. The ulama of its own day treated it as a schism, and one early refutation came from inside the founder's own family, his brother Sulayman ibn Abd al-Wahhab, in the al-Sawaiq al-Ilahiyya attributed to him, which charged the movement with excess takfir and the misuse of Hanbali principles, a charge the Shafii mufti of Mecca, Zayni Dahlan, later amplified.

The state came later, and the scholars who study it describe what the state did to the creed in the vocabulary of engineering. Nabil Mouline titles the relevant chapter of his study of the Saudi clerical establishment "Routinization and Institutionalization of Hanbali-Wahhabism," tracing how Abd al-Aziz ibn Saud drew on the Wahhabi scholars and then, in Bunzel's phrase, tamed them into the state structure. The clearest single proof is that the founding king turned his guns on his own zealots. The Ikhwan who had razed al-Baqi rose against him when his use of modern technology and his halt to cross-border raiding at the colonial frontiers offended their literalism, and he defeated them at the battle of Sibila in 1929, in an hour's fighting Commins calls brief and almost anti-climactic, though it took nearly a further year to finish the revolt. Raison d'etat beat doctrine. The apex clerical body that governs Saudi religious life today, the Council of Senior Scholars, is a state creation of 1971. Madawi al-Rasheed puts the whole transformation in a phrase worth quoting because it is hers. Under the state, she writes, Wahhabism became a "religious nationalist" project, turning personal piety into a public project and a moral community under a political center, manufacturing a legitimacy and a rationale for a pious nation.

The Iranian case runs in parallel. The baseline, in Abrahamian's account, is a clergy ambivalent and divided about the state through the long centuries of the occultation, shunning it, grudgingly accepting it, or embracing it outright. Because the doctrine holds that only the sinless Hidden Imam rules legitimately, all government in his absence is imperfect by its own reckoning. Ruhollah Khomeini's doctrine of velayat-e faqih, the guardianship of the jurist, broke from that. Traditional guardianship covered the legally incapable, minors and widows and the insane, and, as Ervand Abrahamian notes, no Shia writer had explicitly contended that the senior clergy held the authority to control the state. The nearest antecedent, Mulla Ahmad Naraqi in the early nineteenth century, stretched the jurist's guardianship toward public matters but stopped well short of a jurist-ruled state, and Moojan Momen judges that the most any pre-Khomeini Shia writer claimed was that kings and rulers should be guided in their actions and policies by the jurists.

In his 1943 Kashf al-Asrar, Khomeini still wavered on jurist-rule. He argued it categorically only in a set of lectures delivered in Najaf in 1970, published as Islamic Government, and it was enshrined in the constitution after the 1979 revolution, so marginal at the time, on Arjomand's account, that the initial draft constitution did not mention it. Said Amir Arjomand describes Khomeini as having revolutionized Shia political theory, and describes the further step of 1988, velayat-e motlaqeh, the absolute guardianship, as a further revolutionary turn born of the state's own governing crises. In that reading the ruling jurist's authority became as incumbent on the believer as prayer itself, extending even to suspending the primary obligations of the faith, the fast and the pilgrimage, in the interest of the state, and it was written into the revised constitution of 1989. A doctrine that can suspend the pillars in the name of the state, one Khomeini argued into the tradition in 1988, stands thirteen centuries after the imams it claims to serve.

And the Shia establishment largely rejected it, which is the surest sign that it is one position inside the tradition. Nearly the entire senior clergy of 1979, Grand Ayatollah Khoei and Ayatollah Shariatmadari among them, held that Islam does not require the clergy to govern. The largest living marja, Ali al-Sistani in Najaf, keeps the quietist line to this day, the jurist as guide rather than executive. The doctrine's most authoritative internal critic was Hossein-Ali Montazeri, Khomeini's own designated successor, who argued that the guardianship must rest on a social contract with the people and be constrained by an elected body's power to remove an unfit jurist, and who was stripped of the succession in March 1989.

Both are modern constructions that a state built and staffed and bent to its purposes, each departing from a longer tradition it claims to continue, Wahhabism from mainstream Hanbalism and the quietist Sunni ulama, the Iranian system from centuries of Twelver quietism. That is why the two loudest official Islams are such poor guides to the field. Their prominence is a twentieth-century fact about oil revenue and state power. Neither won the field on doctrine, and the reader who mistakes Riyadh or Tehran for Islam has mistaken two engineered exceptions for the rule.

Sufism's Standing Is Itself One of the Tradition's Contests

Sufism gets filed wrong in two opposite drawers, and the drawers cancel out. To one Western sensibility it is Islam's liberal wing, the tolerant, apolitical, poetry-writing alternative the West should cultivate against the hardliners. To the revivalist current it is barely Islam at all, a sump of shrine-worship and innovation. Both readings fail, and they fail on the same facts.

Sufism is a dimension that runs through both Sunni and Shia Islam, as Omid Safi puts it, a set of practices and teachings a Muslim takes on top of being Sunni or Shia. The headline shorthand that lists "Sunni, Shia, and Sufi" as three parallel sects gets it wrong from the first word, because Sufism is an orientation carried inside the other two. Part 1 mapped where it sits in the architecture and how al-Ghazali made it orthodox for Sunnis.

The liberal-wing reading has an institutional history. The RAND Corporation, in Civil Democratic Islam (2003) and Building Moderate Muslim Networks (2007), sorted the Muslim world into fundamentalists, traditionalists, modernists, and secularists and recommended that the United States promote the moderates and back Sufi networks as a counterweight to the militants. It was a policy proposal, and it rested on a stereotype that scholars of Sufism reject outright. Sufi orders were the organizational backbone of some of the largest armed jihads of the modern era — Abd al-Qadir al-Jazairi's Qadiriyya war against the French in Algeria, Imam Shamil's Naqshbandi resistance to Russia in the Caucasus, the Sanusiyya and Omar Mukhtar against the Italians in Libya. Sufi shaykhs' descendants held military and administrative office under the Mughals. A tradition that produced Abd al-Qadir and Shamil was never a ready-made quietism waiting to be enlisted, and "moderate" was a label dropped onto Sufism from outside.

The heresy reading is the revivalist verdict, and it is a real position with real force, but it is one position among others. Salafi and Wahhabi currents condemn the Sufi repertoire — visiting shrines, seeking a saint's intercession, celebrating the Prophet's birthday, the ceremonies of audible remembrance — as shirk and innovation against strict monotheism. That charge has hardened into demolition and murder, from the razing of the Baqi tombs to the campaigns against Sufi shrines and worshippers that Part 1 noted in Sinai and Timbuktu, and on to the shrine bombings at Data Darbar in Lahore and Sehwan Sharif in Sindh. The salient fact for our map is the direction of the violence. Sufis are among the primary targets of jihadist currents, which places them as contested terrain inside the tradition, a long way from the neutral moderate middle the policy shops imagined.

And they are contested from inside Sunnism as much as from outside it, in the Barelvi and Deobandi argument the geography section already set out. The demographics cut against both the fringe reading and the third-sect reading. On Pew's 2012 survey, formal membership in an order is uncommon across most of the Muslim world, but it reaches ninety-two percent in Senegal and more than one in ten in Bangladesh, Russia, Tajikistan, Pakistan, Malaysia, Albania, and Uzbekistan, with the Tijaniyya alone spanning much of West Africa. Ordinary Muslims, where they have been asked, mostly count Sufis as inside. The tradition has spent centuries unable to agree on whether its own mystical heart is the center of devotion or a deviation from it, and the disagreement itself is the accurate description of Sufism's place.

Reform and Salafism Are Rival Answers to One Modern Question

The last axis is the one where Western sympathy most wants to pick a winner, and picking a winner is the mistake. On one side stands modernist reform, the current that wants Islam re-read in the light of reason and modern conditions. On the other stands the Salafi revival, the current that wants Islam returned to the practice of the salaf al-salih, the first three generations. The reflex is to read the first as the enlightened team and the second as the throwback. The history of the two makes that sorting harder to hold than it looks.

The word Salafism does not name one thing in the scholarship. Henri Lauziere, whose conceptual history is the standard treatment, points out that the literature carries two incompatible definitions — one describing a movement of Islamic modernism associated with Jamal al-Din al-Afghani, Muhammad Abduh, and Rashid Rida, known for rational and progressive views, and another, dominant today, describing a purist rigorism that condemns exactly that kind of rationalist reform. Two opposite currents, one shared name. And the shared name was not even self-chosen. The label modernist salafiyya was popularized by the French orientalist Louis Massignon and pinned onto the Abduh school afterward. Afghani and Abduh never called themselves Salafi at all.

The genealogy is not a coincidence, because one man sits at the fork. Rashid Rida began as a modernist student of Abduh and, across the 1900s and 1920s, moved toward a scripturalist, Ibn-Taymiyya-centered position, working to rehabilitate the Wahhabis whom his teacher's generation had held at arm's length. Rida's journal, al-Manar, fed the Muslim Brotherhood. One lineage seeded movements that later commentary treats as enemies. Reform and Salafism are branches of a single trunk grown apart, and the trunk is the encounter with European power and the felt crisis of Muslim decline that both were built to answer.

They even reach for the same tool. Both currents reject taqlid, the binding imitation of the established schools, and both demand a return to the Quran and the Sunna as the authentic ground, which is to say they contest identical terrain and walk off it in opposite directions. The modernist turns the reopened door toward context and purpose, invoking the objectives of the law, the maqasid, and the public interest, maslaha, to justify rereading old rulings for democracy and science and human rights. The Salafi turns the same reopened door toward the letter, taking the practice of the first generations as the standard against which all later accretion is measured. Same premise, reject the dead weight of the schools and go back to the sources. Opposite destinations.

Each side has real names and real institutions, which is how you can tell they are movements and not moods. The modernist line runs from Afghani and Abduh and Rida through Sayyid Ahmad Khan and Muhammad Iqbal in British India to Fazlur Rahman, and its living exponents include figures such as Khaled Abou El Fadl, Abdullahi An-Naim, Mustafa Akyol, Tariq Ramadan, and Abdulaziz Sachedina, who argue in various registers for reconciling Islamic law with pluralism and human rights. Its founding organs were journals, al-Urwa al-Wuthqa, published by Afghani and Abduh in Paris in 1884, and Rida's al-Manar, launched in Cairo in 1898. The Salafi line runs from Ibn Taymiyya through Ibn Abd al-Wahhab and Rida to Nasir al-Din al-Albani, often called the spiritual father of the contemporary movement, with Ibn Baz and al-Uthaymin as twentieth-century authorities and the Islamic University of Madinah and its peers as the training institutions that carry the scholarship across borders.

And the current that presents itself as the oldest is, as a comprehensive ideology, among the youngest. Lauziere's larger argument is that Salafism as a total program is a twentieth-century creation tied to decolonization. Nor is Salafism a synonym for violence, whatever the collapse of "Salafi" into "jihadi" suggests. Quintan Wiktorowicz's much-used typology divides the current three ways, into purists who reject political activism for propagation and teaching, politicos, and jihadists, and the purists are the ones the collapse most obviously misdescribes.

The impulse to sort all of this into a good, moderate, promotable Islam and a bad, backward one to be defeated has a bureaucratic pedigree. It is the logic of the RAND studies, promote the modernists and set them against the fundamentalists. Mahmood Mamdani answered it in a line that has outlived the policy: the sort into good Muslim and bad Muslim is a political projection, not a religious reality. Reform and Salafism are rival renewal projects working the same textual ground with opposite answers, and an analyst who crowns one of them has stopped describing the argument and joined it. The job of a map is to show where the roads run. It was never to tell you which town deserves to win.

Where the Pluralism Actually Stops

Everything so far has been the pluralism. A reader this far in is primed to conclude that the tradition is all contest and no core. That conclusion is the apologist's error wearing the analyst's clothes, and the tradition itself refuses it.

It refused it, most cleanly, in the document that proved the pluralism. The Amman Message was produced to draw a boundary, and it draws one. Its Second Point states the shared core directly, in the tradition's own voice — that God is one, that the Quran is the revealed word of God, that Muhammad is His prophet and messenger, together with the five pillars and the six articles of faith. And it locates all permitted disagreement in the furu, the ancillary branches, and not in the usul, the foundations. This is where the mercy of difference belongs, and where it is legitimate. Section 1 flagged a forged version of that idea, the chainless hadith that "the difference of opinion in my ummah is a mercy," deployed to license anything at all. What the Amman Message invokes is a juristic maxim, a different kind of thing entirely. A maxim carries no chain to grade, so it stands or falls on the machinery around it, and here the machinery is explicit: the disagreement is bounded to the furu and endorsed by a real consensus of scholars. The forged line makes disagreement infinite. The maxim makes it a licensed zone with a fence around it, and the usul build that fence, not the saying. Keeping those two straight is most of the discipline this whole series is trying to teach.

On Pew's global survey, affirmation of the one God and the prophethood of Muhammad is close to unanimous across the dozens of countries polled. Ramadan fasting runs between ninety-four and ninety-nine percent in four of the six regions surveyed, and about half that in Central Asia. Belief in fate runs between eighty-eight and ninety-three in four of five, with Southern and Eastern Europe at fifty-seven. The creed is affirmed at rates almost nothing else in social life reaches, and the practices track it closely outside the post-Soviet margins. And the same survey shows the edge is genuinely contested at the Shia and Sufi lines. In five of seven Middle Eastern and North African countries polled, four in ten Sunnis or more said Shias are not Muslims, and acceptance of Sufis as fellow Muslims ran from about three-quarters in South Asia down to under a fifth in Central Asia. The core holds near-unanimously. Its outer boundary is argued village by village.

The classical machinery draws that line in older language. The hard anchors have a name, al-malum min al-din bi-l-darura, what is known in the religion by necessity, and it is the knowing denial of those anchors that takes a person out of the fold. Disagreement over anything downstream of them does not. Al-Ghazali, in his treatise on the boundaries of belief, reduced the boundary-defining foundations to three — the oneness of God, the prophethood of Muhammad, and the reality of the Last Day — and set the trigger for excommunication narrowly, at giving the lie to something the Prophet is definitively known to have taught, warning that a man who throws the charge of unbelief around risks it landing back on himself. The counterintuitive point here comes from Kamali. Universal consensus, the strongest thing the tradition has for drawing a boundary, almost never actually forms. Its absolute classical terms, he writes, have "hardly been fulfilled by conclusive factual evidence," and it can never overrule an explicit text of the Quran or the Sunna. The settled center is real. It is also small. Most of the tradition lives in the vast licensed field around it, which is the anti-monolith point and the anti-apologist point folding into a single finding.

The Ahmadiyya movement identifies as Muslim, follows the Quran, and prays in the Islamic forms, and mainstream currents nonetheless place it outside, holding that its founder's reading breaches the finality of prophethood, the doctrine that Muhammad is the last prophet, anchored in a specific verse (Q 33:40). Here a caution matters, because the doctrinal boundary and the state's coercive one are different things. When Pakistan wrote the Ahmadiyya exclusion into its constitution in 1974, a theological line hardened into legal machinery, and the analyst who conflates the two has made an error the tradition itself does not require. The mirror case is the Nation of Islam, which used Islamic vocabulary while teaching a mortal, incarnate God and skipping the pillars, and which sat outside the anchors by its own doctrine, until Warith Deen Mohammed, after 1975, repudiated those teachings and led the community into Sunni orthodoxy. A boundary that a whole movement can cross, in a known year, on specific doctrine, is a real threshold, drawn on the anchors and on nothing softer.

Talal Asad's caution is that orthodoxy works as a relationship of power, the outcome of who holds the authority to define correct practice, so the honest job is to describe where and how the tradition draws and contests its line and to leave the title of true Islam unawarded from the outside. Shahab Ahmed's caution runs the other way, against pretending the center is tidier than it is, and he presses it partly against Asad. Hold both anyway and you arrive where the map was always heading. Islam is genuinely plural, along every axis above, and it is genuinely bounded, with a small hard core its own scholars can name and a vast contested field around it. That is the premise of everything that follows.

So put the two errors down together, now that the map is drawn. The monolith cannot survive the Baqi demolition, or the Sunni-Shia argument over the sources of law, or the crushing of the Ikhwan by the state that armed them. The everything-goes cannot survive the Amman Message, the finality of prophethood, or the plain fact that most of the world's Muslims practice a regional Islam with an explicit rule for where custom stops. The tradition is plural, and the pluralism has edges, and both halves are the tradition's own, drawn in its own documents by its own scholars.

That is not a small thing to carry into the argument that comes next. Part 3 turns the map toward the Jewish reader, and it inherits this section's finding as its first premise. The legal frame the Jews lived under was settled early and it held. What moved was everything downstream of it, the administration and the enforcement and the temper of the man doing the collecting, across fourteen centuries and three continents in which Jews lived under it, prospered under it, and suffered under it. The record is plural because the application was plural. Anyone who offers you one verdict on it, a golden age or an unbroken persecution, is selling the monolith again, in a costume you might not recognize at first. We will take the map into that country next, and read the record the way this series reads everything. Which layer. Whose reading. Under which school.

References

1. Mohammad Hashim Kamali. "The Scope of Diversity and Ikhtilaf (Juristic Disagreement) in the Shari'ah." Islamic Studies 37:3 (Autumn 1998), 315-338. Islamic Research Institute, International Islamic University Islamabad. JSTOR https://www.jstor.org/stable/i20836998; scan hosted at https://karamah.org/the-scope-of-diversity-and-ikhtilaf-juristic-disagreement-in-the-shariah/. Cited in Section 1 for the ikhtilaf / khilaf distinction, where khilaf is disagreement not resting on valid evidence (p. 317); al-Shafi'i's division of forbidden from permissible disagreement and the Q 3:105 warning (p. 325); and "a diversity within unity: that is, unity as regards basic principles, and diversity regarding details," with ikhtilaf tolerated as a matter of principle and no one entitled to pressure a jurist out of his convictions (pp. 315, 318).
2. Mohammad Hashim Kamali. Principles of Islamic Jurisprudence. Islamic Texts Society. Cited in Sections 1, 2, and 7 for qati/zanni, the ijma mechanism (zanni to qati) and its subservience to revelation, the Shia rejection of qiyas and confinement of ijma to the ahl al-bayt, and Kamali's point that the classical definition of universal ijma has "hardly been fulfilled by conclusive factual evidence."
3. The Amman Message. "The Three Points." https://ammanmessage.com/the-three-points-of-the-amman-message/ and /the-three-points-of-the-amman-message-v-1/. Cited in Sections 1, 2, and 7 for recognition of the eight schools plus Ashari/Sufi/Salafi thought, the forbidding of takfir and gatekeeping of fatwa, the shared-core Second Point, and the furu-bounded "mercy" maxim.
4. SeekersGuidance. "What Does It Mean for a Matter to Be Known in the Religion by Necessity?" https://seekersguidance.org/answers/islamic-belief/what-does-it-mean-for-a-matter-to-be-known-in-the-religion-by-necessity/. Cited in Sections 1 and 7 for al-malum min al-din bi-l-darura as the belief-boundary.
5. Islamweb, Fatwa 7158. "The difference of opinion among my Ummah is a mercy." https://www.islamweb.net/en/fatwa/7158/. Cited in Section 1 that the wording is a fabricated (mawdu) hadith with no known chain of narration.
6. alsalafiyyah.github.io, translating Fatawa al-Lajnah al-Da'imah no. 16113. Cited in Section 1 for the ruling that the saying "has no origin... no Isnad" (la asla lahu). This is a narrower verdict than #5's mawdu; the section reports the absent chain, which both agree on.
7. Robert Gleave. "Is Every Mujtahid Correct?" in Islamic Law in Context (Cambridge University Press). Cited in Section 1 for the taswib/takhtia framing of licensed juristic diversity (background).
8. Talal Asad. The Idea of an Anthropology of Islam (Georgetown CCAS, 1986; reprinted in Qui Parle 17:2, 2009, pp. 1-30). Cited in Sections 1 and 7 for Islam as a "discursive tradition" and, in the 2009 reprint's wording (p. 22), orthodoxy as "a relationship of power to truth." The 1986 printing (p. 15) reads "a distinctive relationship — a relationship of power."
9. Shahab Ahmed. What Is Islam? The Importance of Being Islamic (Princeton University Press, 2016). Cited in Sections 1 and 7 for the coherent-yet-contradictory account and the caution against overstating the center's tidiness.
10. Aaron W. Hughes. Islam and the Tyranny of Authenticity (Equinox, 2015). Cited in Section 1 that the apologist's "authentic Islam" is the mirror image of Orientalist essentialism.
11. Bernard Lewis. "The Roots of Muslim Rage." The Atlantic Monthly (September 1990). Cited in Section 1 as a representative voice of the monolith framing.
12. Samuel P. Huntington. "The Clash of Civilizations?" Foreign Affairs 72:3 (1993). Cited in Section 1 for building the monolith intuition into a foreign-policy paradigm.
13. Edward W. Said. Orientalism (1978) and Covering Islam (1981). Cited in Section 1 for the monolith-as-essence diagnosis.
14. Wilferd Madelung. The Succession to Muhammad: A Study of the Early Caliphate (Cambridge University Press, 1997), Introduction. Cited in Section 2 that the classical Sunni electoral-caliphate theory is "obviously posterior to the succession," and for the Quranic elevation of the ahl al-bayt.
15. Pew Research Center. Mapping the Global Muslim Population (October 7, 2009). https://www.pewresearch.org/religion/2009/10/07/mapping-the-global-muslim-population/. Cited in Sections 2 and 3 for the Shia 10-13% / 154-200M figure, the four-country concentration, and the Iran share. (Its 2009 Asia-Pacific figure of 62% is superseded in the body by #17's 59% as of 2020; MENA at about a fifth holds across both.)
16. Linda Walbridge (ed.). The Most Learned of the Shi'a (2001). Cited in Section 2 for the marjaiyya as an institution (background).
17. Pew Research Center. How the Global Religious Landscape Changed From 2010 to 2020, ch. 3, "Muslim population change" (June 9, 2025). https://www.pewresearch.org/religion/2025/06/09/muslim-population-change/. Cited in the Executive Summary and Section 3 for the ~2.0 billion global total as of 2020, the ~240M Indonesia figure, and the 59% Asia-Pacific share.
18. Pew Research Center. Muslim Population of Indonesia (November 4, 2010). https://www.pewresearch.org/religion/2010/11/04/muslim-population-of-indonesia/. Cited in Section 3 for Indonesia's Muslim share.
19. Clifford Geertz. The Religion of Java (University of Chicago Press, 1960). Cited in Section 3 for the abangan/santri/priyayi trichotomy.
20. Andrew Beatty. Varieties of Javanese Religion (Cambridge University Press, 1999); and A.N. Burhani, "Geertz's Trichotomy of Abangan, Santri, and Priyayi," Journal of Indonesian Islam. https://jiis.uinsby.ac.id/. Cited in Section 3 for the critique that the trichotomy over-freezes real variation.
21. al-Majalla / Mecelle (Ottoman civil code, 1869-1876), Art. 36 (al-adah muhakkamah). Cited in Section 3 for the codification of custom as an operative legal maxim.
22. IIUM Maliki-fiqh usul primer, "The Eleventh Source: Customs (Adat) and Urf," and Arriqaaq, "Usul al-Fiqh Made Easy: What is Urf." https://www.iium.edu.my/deed/lawbase/maliki_fiqh/usul12.html. Cited in Section 3 for urf sahih vs. urf fasid, al-Qarafi on the variability of custom, and the Maliki/Hanafi elevation of custom.
23. Oxford Islamic Studies Online. "West Africa, Sufism in." Cited in Section 3 for the Qadiriyya/Tijaniyya/Muridiyya as West Africa's organizational backbone.
24. GeoCurrents. "Deobandi Islam vs. Barelvi Islam in South Asia." Cited in Section 3 for the Barelvi/Deobandi split over shrine devotion.
25. David Commins. The Wahhabi Mission and Saudi Arabia (I.B. Tauris, 2006/2009). https://archive.org/stream/WahhabiMissionSaudiArabiaByProfessorDavidCommins. Cited in Section 4 for the 1744 Diriyah pact ("the origin of the pact between religious mission and political power") and the Ikhwan's defeat at Sibila (1929, pp. 90-92) — Commins calls the fighting "brief, almost anti-climactic," with the revolt taking nearly a further year to suppress.
26. Cole Bunzel. Wahhabism: The History of a Militant Islamic Movement (Princeton University Press, 2023), the Democracy Paradox interview (https://democracyparadox.com/2023/05/30/cole-bunzel-on-wahhabism/), and Bunzel's Q&A with the Stanford Islamic Studies program. Cited in Section 4 that Wahhabism was a radical 18th-century revival that departed from mainstream Hanbali practice, that Ibn Abd al-Wahhab drew on Ibn Taymiyya and his student Ibn Qayyim and took them in a more radical direction (the Stanford Q&A; the Democracy Paradox interview does not mention Ibn Qayyim), and that the state "tamed" the ulama into its structure.
27. Nabil Mouline. The Clerics of Islam: Religious Authority and Political Power in Saudi Arabia (Yale University Press, 2014). https://academic.oup.com/yale-scholarship-online/book/18816. Cited in Section 4 for the "routinization and institutionalization" of Wahhabism and the 1971 Council of Senior Scholars.
28. Madawi al-Rasheed. A Most Masculine State: Gender, Politics and Religion in Saudi Arabia (Cambridge University Press, 2013), ch. 1 "From Religious Revival to Religious Nationalism"; and "New Texts Out Now: A Most Masculine State," Jadaliyya. https://www.jadaliyya.com/Details/27877. Cited in Section 4 for the "religious nationalist" reading, in the author's own words.
29. Said Amir Arjomand. "Authority in Shiism and Constitutional Developments in the Islamic Republic of Iran," in The Twelver Shia in Modern Times (Brill, 2001), pp. 301-332; and The Turban for the Crown (Oxford University Press, 1988). Cited in Section 4 for the "revolutionary transformation of the traditional Shiite theory of authority" and for there having been "virtually no discussion" of the theory before summer 1979 (both p. 301), and for the 1988 velayat-e motlaqeh as a further revolutionary turn.
30. Ervand Abrahamian. Khomeinism: Essays on the Islamic Republic (University of California Press, 1993). Cited in Section 4 that across centuries of debate no senior cleric had claimed the clergy's authority to run the state.
31. iran1400.org. "Wilayat al-faqih: How a Cold-Shouldered Idea Changed Iran's History." https://iran1400.org/content/wilayat-al-faqih-how-a-cold-shouldered-idea-changed-irans-history/. Cited in Section 4 corroborating the Naraqi antecedent, the 1943-vs-1970 evolution, and the doctrine's marginal status in 1979.
32. Ruhollah Khomeini. Hokumat-e Islami / Velayat-e Faqih (Najaf lectures, 1970) and Kashf al-Asrar (1943). https://www.iranchamber.com/history/rkhomeini/books/velayat_faqeeh.pdf. Cited in Section 4 as the primary texts dating the doctrine.
33. Sulayman ibn Abd al-Wahhab. al-Sawaiq al-Ilahiyya fi al-radd ala al-Wahhabiyya (late 18th c.). Cited in Section 4 as the earliest intra-Hanbali refutation of the Wahhabi movement (with Zayni Dahlan's later amplification).
34. Mulla Ahmad Naraqi. Awaid al-Ayyam (early 19th c.); and Moojan Momen, An Introduction to Shi'i Islam (Yale University Press, 1985), p. 196 (page located via a secondary citation, not confirmed against the book). Cited in Section 4 for the narrow pre-Khomeini antecedent of political guardianship, and for Momen's judgment that the most any pre-Khomeini Shi'i writer claimed was that rulers should be guided in their actions and policies by the jurists.
35. Saudi Royal Decree No. 137/1 (8 Rajab 1391 AH / 29 August 1971), establishing the Council of Senior Scholars. Cited in Section 4 for the Council's founding as a twentieth-century state institution. The decree itself was not opened for this piece; Mouline (#27) covers the Council's state-created character.
36. Shahrough Akhavi. "The Thought and Role of Ayatollah Hossein'ali Montazeri in the Politics of Post-1979 Iran," Iranian Studies; with Montazeri, Dirasat fi Wilayat al-Faqih (4 vols., 1988-90); and Oasis International, "Contemporary Critics of the Velayat-e Faqih," https://www.oasiscenter.eu/en/iran-critics-guardianship-jurisconsult. Cited in Section 4 for Montazeri's social-contract constraint on the guardianship, the removability of an unfit jurist by the elected Assembly of Experts, and his removal from the succession in March 1989.
37. Council on Foreign Relations. Backgrounder on Grand Ayatollah Ali al-Sistani. https://www.cfr.org/backgrounders/iraq-grand-ayatollah-ali-al-sistani. Cited in Section 4 for Najaf's continuing quietism.
38. Omid Safi. "The Characterization of Sufism as a Separate Sect... Is Inaccurate and Problematic." On Being. https://onbeing.org/blog/omid-safi-the-characterization-of-sufism-as-a-separate-sect-within-islam-is-inaccurate-and-problematic/. Cited in the lede and Section 5 that Sufism is a tendency within both Sunni and Shia Islam and that a majority of Muslims, "or close to that," practiced Sufi-oriented Islam until the mid-19th century.
39. Nile Green. Sufism: A Global History (Wiley-Blackwell, 2012). Section 5, for Sufism's centrality to Muslim social, political, and intellectual history (background).
40. Fait Muedini. "Sufism and Anti-Colonial Violent Resistance Movements." Open Theology 1 (2014). https://www.degruyterbrill.com/document/doi/10.1515/opth-2015-0003/html. Cited in Section 5 for the Sufi-led armed jihads (Abd al-Qadir/Algeria, Shamil/Caucasus, the Sanusiyya/Libya).
41. The Maydan (George Mason University). "The Problem of 'Political Sufism.'" https://themaydan.com/2018/12/problem-political-sufism/. Cited in Section 5 rejecting the apolitical-Sufism stereotype and for Naqshbandi-Mughal statecraft.
42. RAND Corporation. Civil Democratic Islam (Cheryl Benard, MR-1716, 2003/04) and Building Moderate Muslim Networks (MG-574, 2007). https://www.rand.org/pubs/monographs/MG574.html. Cited in Sections 5 and 6 as the bureaucratic form of the "promote the moderates / back the Sufis" and "promote the modernists against the fundamentalists" programs.
43. Pew Research Center. The World's Muslims: Unity and Diversity (August 9, 2012). https://www.pewresearch.org/religion/2012/08/09/the-worlds-muslims-unity-and-diversity-executive-summary/. Cited in Sections 5 and 7 for Sufi-order membership and acceptance figures, and for the shahada/Ramadan/fate anchors and the contested Shia/Sufi edges.
44. Wikipedia, reference-tier: "Demolition of al-Baqi," "2017 Sinai mosque attack," "Persecution of Sufis" [Sec. 5]; "Imamate in Shia doctrine," "Occultation (Islam)," "Marja'," "Zaydism," "Ismailism," "Ja'fari school" [Sec. 2]; "Madhhab," "Hanbali school," "Maliki school," "Islam in Indonesia," "Barelvi movement," "Traditional Berber religion" [Sec. 3]; "Guardianship of the Islamic Jurist," "Battle of Sibila," "Council of Senior Scholars (Saudi Arabia)," "Sulayman ibn Abd al-Wahhab" [Sec. 4]; "Islamic modernism," "Salafi movement," "Al-Urwah al-Wuthqa," "Nation of Islam" [Secs. 6, 7]. Where an academic anchor is named elsewhere in this list, these corroborate it. For a set of uncontested factual particulars they are the primary support: the occultation dates and the four deputies, al-Ansari's standing as the first universally recognized marja, the Shafii share of Indonesian Muslims, the Indonesian Islamization chronology, the founding dates of al-Urwa al-Wuthqa and al-Manar, and the Data Darbar and Sehwan Sharif bombings. None of those is contested in the literature, and none carries an argument on its own.
45. Henri Lauziere. The Making of Salafism: Islamic Reform in the Twentieth Century (Columbia University Press, 2016); author's "New Texts Out Now," Jadaliyya. https://www.jadaliyya.com/Details/34136. Cited in Section 6 for the two incompatible definitions of Salafism, the Massignon-coined label, and Salafism-as-comprehensive-ideology as a 20th-century construction.
46. Encyclopedia.com. "Salafiyya"; and IRF, "Muhammad Abduh and His Epistemology of Reform." https://www.encyclopedia.com/religion/encyclopedias-almanacs-transcripts-and-maps/salafiyya. Cited in Section 6 for the shared textual move (reject taqlid, return to the sources) and Rida as the pivot seeding both branches.
47. Quintan Wiktorowicz. "Anatomy of the Salafi Movement." Studies in Conflict & Terrorism 29:3 (2006); and Springer, "Quietism versus jihadism: the need to rethink contemporary Salafism" (2023). https://link.springer.com/article/10.1007/s11562-023-00525-w. Cited in Section 6 for the purist/politico/jihadi typology and its contested refinement. Wiktorowicz does not rank the factions by size; no relative-size claim is made in the body.
48. Mahmood Mamdani. Good Muslim, Bad Muslim: America, the Cold War, and the Roots of Terror (2004). Cited in Section 6 that the good/bad-Muslim sort is a political projection, not a religious reality.
49. Abu Hamid al-Ghazali. Faysal al-Tafriqa bayna al-Islam wa al-Zandaqa, trans. Sherman A. Jackson, On the Boundaries of Theological Tolerance in Islam (Oxford University Press, 2002). Cited in Section 7 for the reduction of the boundary-defining usul to three and the narrow takfir-as-takdhib criterion.
50. Nuh Ha Mim Keller. "Iman, Kufr, and Takfir." Cited in Section 7 for the traditional articulation of al-malum min al-din bi-l-darura (with #4).
51. Cambridge, The Ahmadis and the Politics of Religious Exclusion in Pakistan; and ScienceOpen / ReOrient, Pakistan's 1974 Second Amendment (doi:10.13169/reorient.9.2.0005). Cited in Section 7 for the Ahmadiyya boundary case (khatam an-nabiyyin, Q 33:40) and the caution that the doctrinal boundary and Pakistan's 1974 state-coercive one are distinct.
52. J. Fishman. "The Nation of Islam and the Muslim World." Religion Compass (2013). https://compass.onlinelibrary.wiley.com/doi/10.1111/rec3.12032. Cited in Section 7 for the Nation of Islam's self-exclusion and its post-1975 crossing into Sunni orthodoxy under Warith Deen Mohammed.
53. Farhad Daftary. The Ismailis: Their History and Doctrines (Cambridge University Press, 2nd ed. 2007). Cited in Section 2 for the Ismaili branch: the secession over which son of Jafar al-Sadiq carried the imamate, which ran from the common Imamiyya and not from a Twelver body that did not yet exist, and for the Nizari line's continuing living imam. (Daftary, p. 97, also establishes why the "Sevener" label does not fit that line: it names those who limited the imams to seven and stopped at Muhammad ibn Ismail.)

◇