How Islam Travelled
What came first, what came after, and how each school reached the countries it holds today, the routes, the mechanisms, and the dates.
What came first, what came after, and how each school reached the countries it holds today, the routes, the mechanisms, and the dates.
Drawn in parallel lanes, because the sequence is the point. Popular accounts imply Sunni and Shia Islam split once, early, and then developed alongside one another. What the lanes show is that the Sunni canon and the end of the Shia imamate happen in the same generation, and that both are two centuries downstream of the political rupture.
↔ Scroll the figure sideways
Snapshots, what actually existed, and when
The clearest way to see the sequence is to ask what a Muslim alive in a given year would have recognised.
Schools did not spread because they persuaded. They spread because something carried them. These mechanisms account for nearly every arrival on the map, and knowing which one operated in a given place explains far more than any doctrinal comparison.
A schematic, not a geographic map, regions are placed roughly by position and sized by nothing. What matters is the arrows: who fed whom, by what mechanism, and when. Filter by tradition to read one story at a time.
↔ Scroll the figure sideways
Every tradition, in full, origin, mechanism, hop by hop with dates, and why it stuck. Where a date is contested or a causal story is scholarly inference rather than documented event-history, it says so.
The strongest evidence for the whole thesis is the set of places where a region changed school. Doctrine is held constant, the arguments of Mālik and Abū Ḥanīfa did not alter, and only politics varies. If doctrine drove distribution, these changes should not happen. They happen repeatedly.
A madhhab does not lose an argument. It loses appointments.
The Awzāʿī school was the official school of both Syria and al-Andalus. It is completely extinct, not refuted, not absorbed after a debate, but starved of judgeships in Iberia by an Umayyad preference for Mālik, and in Syria by a Shāfiʿī judge who offered material incentives to memorise Shāfiʿī texts. By the eleventh century, none of its adherents could be found anywhere.
Egypt changed official school four times in six centuries, Shāfiʿī, then Ismāʿīlī Shia under the Fatimids, then Shāfiʿī again under Saladin, then four parallel chief qāḍīs under the Mamluks, then Ḥanafī courts imposed over a Shāfiʿī population by the Ottomans in 1517. Same territory, same people, five legal regimes. Doctrine explains none of it.
The modern mechanisms are different in kind, and the scholarship has shifted noticeably in the last fifteen years, away from state funding as the master explanation, and towards migration, local institutional capacity, and the restructuring of religious authority by media. Ranked by the weight the literature actually gives them.
Albania
Saudi and Gulf institutions distributed roughly a million Qur'ans in Albania and funded some two hundred mosques after 1991. Albania today reports among the lowest religious salience of any Muslim-majority country in the world.
The lesson is not that money does nothing. It is that money buys infrastructure, and infrastructure only converts into adherence where local institutions were already strong enough to use it. Widely circulated totals for Saudi religious spending (the “$75 billion, 1,500 mosques, 2,000 schools” figure) are weakly sourced; Mandaville and Hamid, who wrote the standard study, decline to give a consolidated total at all.
Region by region, the dominant school today, roughly when it arrived, and the mechanism that carried it. Read the right-hand column and the pattern is unmistakable.
One caution that applies to this entire table: no census of madhhab affiliation exists anywhere in the world. Pew measured the Sunni/Shia split only. Everything finer than that is inference from institutional presence and scholarly consensus, not survey data.
↔ Scroll the table sideways
“The madhhab map is a map of medieval states and trade routes” is the claim this whole piece has been building. It is substantially true. It is also routinely overstated, and the honest version is more interesting than the slogan.
The early-modern empires, mechanism, dates and paperwork
Guy Burak does not merely assert the correlation; he periodises it. His claim is that from the fifteenth century something genuinely new appears, “the era of the state madhhab”, in which dynasties determined not just who judged but which opinions were valid. He roots it in post-Mongol political culture, where rulers claimed a mandate to legislate.
Rudolph Peters supplies the operational detail for the Ottoman case: exclusive Ḥanafī appointments, thirty-two sultanic orders on doctrine in the sixteenth century, and non-Ḥanafī qāḍīs demoted to deputies after 1517.
For the classical period the best version is Nurit Tsafrir's: “the identification of Ḥanafism with Baghdad and the ʿAbbāsids is the reason for its success” in penetrating Khurāsān and Transoxiana.
Burak's own periodisation limits the claim
Note carefully what follows from dating state doctrinal control to after 1400: the classical madhhab map (Mālikī Maghreb, Shāfiʿī Egypt, Ḥanafī Transoxiana) was not produced by that mechanism, because that mechanism did not yet exist. The most rigorous statement of the thesis is also the one that most restricts its scope.
Peters adds a second constraint that cuts against pure politics: the Ottoman arrangement worked because Ḥanafī doctrine permitted such stipulation, and other schools' did not. Doctrine bounded what politics was able to do.
Doctrine, custom, and the autonomy of jurists
1. Institutional primacy. Drawing on Ahmed El Shamsy's The Canonization of Islamic Law (2013), the argument is that a founder's success required methodological innovation plus influential students, and that socio-political factors were insufficient alone. Note that the founders often resisted state power, Abū Ḥanīfa refused the caliph's offer of a judgeship; Ibn Ḥanbal was flogged and imprisoned.
2. Schacht's older account attributes early regional legal difference to local variation in social conditions, customary law and practice, and offers no political explanation of school distribution at all.
3. The jurists'-law thesis. Hallaq's position that Islamic law was substantially autonomous of the state is the framework-level objection to the whole picture.
“States” and “trade routes” are not the same claim
The thesis is weakest precisely where it is stated most confidently. The maritime Shāfiʿī zone and Mālikī West Africa were shaped far more by trade, marriage and clerical lineage than by any state adopting a school.
Two facts settle it. The Hadrami sayyids had no state at all, and they supplied the religious authority of the entire Indian Ocean rim. The Ibāḍīs had a state, Oman, and a maritime empire reaching Zanzibar, and still failed to spread. If states were the master variable, those two results are backwards.
A note on sourcing: no single canonical one-line scholarly formulation of “the madhhab map is a map of medieval states” appears to exist. The commonplace circulates widely; the citable versions are the scoped, careful ones above.
The mechanism that actually explains the map is endowment, not patronage
Patronage is reversible: a ruler's preference dies with the ruler. The Almohads sponsored Ẓāhirism and burned Mālikī books, and when the Almohads fell, Ẓāhirism fell with them and Mālikism resumed. State sponsorship turns out to be neither sufficient nor durable on its own.
What is durable is the waqf. The Niẓāmiyya of Baghdad, founded 1065, carried a standing requirement that every teacher, preacher and librarian be Shāfiʿī in law and Ashʿarī in creed. An endowment is perpetual and its terms bind future administrators, so endowing a chair for a named school converts a momentary preference into an unalterable, self-financing, multi-generational salary line that outlives the dynasty that created it.
And where there was no state and no endowment, West Africa, a compact, teachable canon did the same work. The Risāla of Ibn Abī Zayd, the Mukhtaṣar of Khalīl, the fatwas of al-Wansharīsī: a curriculum small enough to cross a desert on a camel and complete enough to reproduce a school through clerical lineages without any state at all. Which is why Mālikism in West Africa survived the collapse of Songhay and French colonisation alike.
Every technical term used on this page, in one place. The first appearance of each in the text above carries its definition on hover or focus.
Dates corrected against sources during research, worth knowing if you have the older figures: the Fatāwā-yi ʿĀlamgīrī is 1664–1672, not 1667–75; the Rustamid imamate at Tāhert runs from 776/777, not 761 (which is Abū al-Khaṭṭāb's death); Said b. Sultan moved his capital to Zanzibar in 1840, with 1832 unverified; Qom's Shia foundation is given as 85/704 by WikiShia against the commonly cited 712.
Figures that circulate widely and should not be repeated as fact: Saudi religious spending totals; Tablīghī Jamāʿat membership (Pew's own range is 12–80 million, and no point estimate is defensible); “Barelvi 200 million”; the Shia share of Pakistan's population, for which no census figure exists and the 5–20% range is politically contested; and “20,000 Sunnis killed at Tabriz” in 1501, which is a chronicle figure, not a demographic one.
The state-madhhab thesis
- Guy Burak, “The Second Formation of Islamic Law”, CSSH
- Rudolph Peters on Ottoman judicial administration, paper
- Nurit Tsafrir, The Early Spread of Ḥanafism in Khurāsān
- Ahmed El Shamsy, The Canonization of Islamic Law (2013)
- Knut Vikør, Between God and the Sultan
Trade and the Indian Ocean
- Engseng Ho, The Graves of Tarim, the Hadrami sayyid mechanism
- COMPAS Oxford, “The Hadramis of the Indian Ocean”
- Stanford FSI, The Spread of Islam in West Africa, note it does not name the Mālikī school
Shia diffusion
- Encyclopaedia Iranica, Jabal ʿĀmel, al-Karakī, Shiʿites of Iraq
- The Safavid conversion of Iran
- The Ismaili diaspora · Seat of the Imamat, Portugal
Modern propagation
- Peter Mandaville & Shadi Hamid, Islam as Statecraft (Brookings, 2018)
- Carnegie, “Has Saudi funding spread Wahhabism?”
- CREST, British Muslims: Sectarian Movements
- Pew, Tablīghī Jamāʿat
- Olivier Roy, Globalized Islam; Gary Bunt, Hashtag Islam
Institutions
- Al-Niẓāmiyya of Baghdad and the Niẓāmiyya network
- Yossef Rapoport, “Legal Diversity in the Age of Taqlīd”, ILS 10:2 (2003)
- Timbuktu and Sankore
- Islamic University of Madinah (1961)
Demography
- Pew, Mapping the Global Muslim Population, Sunni 87–90%, Shia 10–13%, given deliberately as a range
- Madhhab shares throughout are directional estimates. Nothing counts madhhab affiliation anywhere.
Siddiqui, M. “How Islam Travelled.” mohammadsiddiqui.com.
Further reading
Six books to start with, chosen for a reader with no prior background. This is a different list from the sources above, which are there so the claims can be checked.
- Richard W. Bulliet, Conversion to Islam in the Medieval Period (Harvard, 1979)The curve that reframed everything: conversion was slow, and mostly happened centuries after the conquests. Method is contested, conclusion has held.
- Nile Green, Sufism: A Global History (Wiley-Blackwell, 2012)Treats the orders as social institutions that moved people and money, not as a mood. The best account of the mechanism this page calls the Sufi engine.
- Engseng Ho, The Graves of Tarim (California, 2006)Five centuries of Ḥaḍramī sayyid migration across the Indian Ocean, followed through genealogies and gravestones. The monsoon route made vivid.
- Michael Laffan, The Makings of Indonesian Islam (Princeton, 2011)How the world's largest Muslim population became Shāfiʿī, and what colonial scholarship did to the categories we still use for it.
- Peter Mandaville & Shadi Hamid, Islam as Statecraft (Brookings, 2018)The standard study of state religious diplomacy, and the reason the widely quoted Saudi spending totals are not repeated on this page.
- Nehemia Levtzion (ed.), Conversion to Islam (Holmes & Meier, 1979)Region-by-region case studies. Old, uneven, and still the fastest way to see how differently conversion worked in each place.
↑ Back to top
Siddiqui, M. “How Islam Travelled.” mohammadsiddiqui.com.
← Part II · The House and the Schools ↑ Top
