Did Islam Spread by the Sword?

Updated September 8, 20268 min read

Summary

Islam was not spread simply by forcing conquered populations to become Muslim. Muslim armies fought for different reasons: some campaigns responded to attacks, military threats, or the killing of Muslim envoys carrying the message of Islam, while others arose from local political appeals or involved states seeking territory and political power.

Conquest changed political rule, not the population’s religion: Christians, Jews, Zoroastrians and Hindus remained non-Muslim for centuries, while conversion unfolded gradually over generations. Islam also spread across regions such as Southeast Asia without Muslim military conquest.

The Qur’an clearly rejects compulsion in religion: people are to be invited to Islam through wisdom and good counsel, while warfare is governed by separate rules concerning conflict and peace. Islamic rules of war protected non-combatants and ensured honor and ehthicality even in warfare. Conquest explains the spread of Muslim rule; it does not, by itself, explain the spread of Muslim belief.

Introduction

The claim that Islam was “spread by the sword” often compresses three different questions into one: why Muslim armies fought, what changed when they conquered territory, and whether conquered people were forced to become Muslim.

Muslim states did expand through warfare. But wars did not all begin for the same reason, and military victory did not ordinarily make the population Muslim. Across much of the Muslim world, political rule changed quickly while conversion took centuries. Non-Muslim communities remained in place, and Islam also spread widely where no Muslim army had conquered the population.

Forced conversions did occur. They are identifiable episodes within a much broader history, not a sufficient explanation for how Islam became a world religion.

Key terms

sūrah: a chapter of the Qur’an — the standard division of the text, of which there are 114, each identified by a name rather than a number.
jizya: a per-head tax paid by non-Muslim subjects in place of military service and the alms owed by Muslims, in exchange for protection.
hadith: a transmitted report of what the Prophet Muhammad ﷺ said, did, or tacitly approved.
ṣaḥīḥ: “sound” — a hadith graded as meeting the highest bar for authenticity of transmission (as distinct from the collection titles Sahih al-Bukhari and Sahih Muslim).
ḍaʿīf: “weak” — a hadith with a defective chain of transmission, which cannot on its own establish a legal ruling.

Why were Muslim armies fighting?

There was no single cause behind every Muslim war or conquest. Historians distinguish the causes of particular campaigns from the reasons the early conquests succeeded and from the later religious history of the conquered populations.1

The Prophet’s ﷺ northern campaigns already show different circumstances. Traditional Muslim accounts connect the expedition to Mu’tah in 629 to the killing of al-Harith ibn Umayr, an envoy carrying a message toward Byzantine territory.2 The expedition to Tabuk in 630 followed reports that Byzantine forces were gathering against Medina; early sources themselves record that the intelligence proved exaggerated, and no battle took place.3

Later Muslim campaigns could begin for still other reasons. In 827, for example, the Byzantine rebel commander Euphemius sought military assistance from the Aghlabid ruler of North Africa; his appeal became the immediate occasion for the Muslim invasion of Sicily.4

None of this means that every Muslim conquest was defensive. Muslim states also pursued expansion, territory and political power. The point is narrower: “conquest” describes what happened; it does not by itself explain why a particular war began, still less prove that its purpose was to force the population to convert.

Conquest changed the government, not the population’s religion

When Muslim armies took a territory, the immediate change was normally political: one ruling authority was replaced by another. The inhabitants did not thereby become Muslim.

Non-Muslims could retain their religion, accept Muslim political authority and pay the jizya, a tax imposed on non-Muslim subjects and discussed separately in The Jizya. A system built around a tax on non-Muslims presupposed that non-Muslim populations would continue to exist under Muslim rule.

The Prophet’s ﷺ instructions to commanders expressly allowed settlement without conversion:

“Invite them to accept Islam… If they refuse to accept Islam, demand from them the jizya. If they agree to pay, accept it from them and hold off your hands.” — Sahih Muslim 45215

Acceptance of Muslim rule therefore did not require acceptance of Islam. Political submission and religious conversion were separate outcomes.

Jerusalem provides an early example

Jerusalem gives a particularly vivid example when the city surrendered in 637–638 to Umar ibn al-Khattab. He was not a minor commander: he was the second Muslim caliph—the head of the Muslim state and the political leader of the Muslim community. Yet Christian sources remember his arrival for its remarkable simplicity. The Byzantine chronicler Theophanes, who was hostile to the Muslim conquerors, describes the caliph entering in rough camel-hair clothing and refusing finer clothes offered by Patriarch Sophronius. Muslim tradition adds the well-known account that Umar and his attendant took turns riding a single mount, and that Umar approached Jerusalem on foot while his attendant rode.6

A later Christian account preserved by Eutychius of Alexandria makes the point even more sharply. Sophronius invited Umar to pray inside the Church of the Holy Sepulchre. Umar refused and prayed outside, explaining that if the Muslim caliph prayed inside, later Muslims might use that act as a reason to take the church and turn it into a mosque. The account was written centuries later, so it cannot establish every detail of the scene. But what it preserves is striking: Muslim tradition remembered the ruler who had just taken Jerusalem deliberately protecting a Christian church from appropriation by Muslims themselves.6

Conversion took centuries

If conquest itself had been a program of mass conversion, religious change should have followed military victory quickly. Instead, the evidence shows a much slower process.

Richard Bulliet’s Conversion to Islam in the Medieval Period reconstructed conversion timelines from Muslim personal names in biographical dictionaries. For Iran, Iraq, Syria, Egypt and Spain, he found gradual S-curves extending across roughly three to four centuries rather than sudden changes at conquest.7

Egypt provides a particularly clear comparison. Quantitative research using tax records and later census data places the conversion of the Coptic population between 641 and about 1200.8 The conquest took about two years; the transition to a Muslim majority took roughly five centuries.

Bulliet’s graphs count those who eventually converted, not the whole population. Syria and Egypt still contained substantial Christian populations at the end of the medieval period.7 Later scholarship has revised particular regional curves, especially for al-Andalus, but the broader pattern remains gradual conversion over generations rather than immediate mass conversion.

Political conquest was rapid. Religious change was not. That is not the pattern expected if armies were ordinarily being used to make conquered populations profess Islam.

Non-Muslim communities survived—and sometimes flourished

The evidence goes beyond communities simply surviving conquest. Egypt’s Copts have remained continuously present since before the Muslim conquest. Syriac and Assyrian Christians endured in Iraq and Syria, Zoroastrians remained in Iran, and Hindus remained the majority across most of the Indian subcontinent despite centuries of Muslim dynastic rule.

India is especially instructive. Muslim dynasties ruled large parts of the subcontinent from the Delhi Sultanate beginning in 1206 through the end of Mughal rule in 1857. Yet the 2011 census still recorded roughly 80 per cent of India’s population as Hindu and 14 per cent as Muslim.9

And some communities did more than survive. Under the Abbasids, the great Jewish academies of Babylonia became centres of authority for Jewish communities across the Islamic world; Oxford scholarship describes their prestige and influence as unprecedented. The Church of the East likewise entered a period of exceptional power and influence after its patriarchate moved to Baghdad, while Syriac Christians played a central role in the Abbasid translation movement.10

In some places, non-Muslims even had reason to prefer the incoming Muslim rulers to the government being displaced. Visigothic Spain had subjected Jews to severe persecution and forced conversion; when Muslim forces entered in 711, significant elements of Jewish communities gave them allegiance. In Egypt, the Coptic Church had suffered Byzantine pressure to accept imperial doctrine. After the conquest, its patriarch Benjamin returned from exile, and the Christian chronicler John of Nikiu says that Amr ibn al-As preserved church property during his rule. Later Byzantine reconquests could even drive non-Chalcedonian Christians to seek refuge in Muslim-held territory.11

This does not mean every conquered population welcomed Muslim armies, or that every Muslim ruler governed justly. The point is simpler and more important: the ruler being defeated was not the same thing as the people he ruled. For some Jews and Christians, a change to Muslim rule could mean relief from persecution by the previous state; for others, it allowed their communities not merely to remain, but at times to flourish.

Islam also spread without Muslim conquest

The “sword” explanation also fails where Islam spread without an invading Muslim army.

The Islamisation of the Indonesian archipelago developed through trade, Muslim merchant communities, preaching, marriage and the conversion of local rulers rather than an Arab conquest. Indonesia had roughly 242 million Muslims in 2022.12

Similar commercial and scholarly networks carried Islam through the Malay world, across trans-Saharan routes into West Africa and along the Indian Ocean coast of East Africa. Muslim communities also developed in China through merchants travelling by land and sea.

Conquest therefore cannot be the general explanation for Islam’s spread as a religion: some of the world’s largest Muslim populations developed where no Arab conquest occurred.

Belief and warfare in the Qur’an

The Qur’an treats compelled belief and warfare as separate questions. Belief is not to be forced; invitation is through persuasion. Fighting, where commanded, has its own limits and is not presented as a mechanism for manufacturing belief.

The English wording below follows the cited translations, with “God” used for the divine name.

Belief is not to be compelled

“Let there be no compulsion in religion” (Sūrat al-Baqarah 2:256), Dr. Mustafa Khattab, The Clear Quran

﴿لَآ إِكْرَاهَ فِى ٱلدِّينِ﴾ [سُورَةُ البَقَرَةِ: ٢٥٦]

“Had your Lord so willed ˹O Prophet˺, all ˹people˺ on earth would have certainly believed, every single one of them! Would you then force people to become believers” (Sūrat Yūnus 10:99), Dr. Mustafa Khattab, The Clear Quran

﴿وَلَوْ شَآءَ رَبُّكَ لَـَٔامَنَ مَن فِى ٱلْأَرْضِ كُلُّهُمْ جَمِيعًا ۚ أَفَأَنتَ تُكْرِهُ ٱلنَّاسَ حَتَّىٰ يَكُونُوا۟ مُؤْمِنِينَ ۝٩٩﴾ سُورَةُ يُونُسَ

“Invite ˹all˺ to the Way of your Lord with wisdom and kind advice, and only debate with them in the best manner” (Sūrat al-Naḥl 16:125), Dr. Mustafa Khattab, The Clear Quran

﴿ٱدْعُ إِلَىٰ سَبِيلِ رَبِّكَ بِٱلْحِكْمَةِ وَٱلْمَوْعِظَةِ ٱلْحَسَنَةِ﴾ [سُورَةُ النَّحۡلِ: ١٢٥]

Fighting has separate limits

“Fight in the cause of God ˹only˺ against those who wage war against you, but do not exceed the limits. God does not like transgressors” (Sūrat al-Baqarah 2:190), Dr. Mustafa Khattab, The Clear Quran

﴿وَقَـٰتِلُوا۟ فِى سَبِيلِ ٱللَّهِ ٱلَّذِينَ يُقَـٰتِلُونَكُمْ وَلَا تَعْتَدُوٓا۟ ۚ إِنَّ ٱللَّهَ لَا يُحِبُّ ٱلْمُعْتَدِينَ ۝١٩٠﴾ سُورَةُ البَقَرَةِ

“If the enemy is inclined towards peace, make peace with them. And put your trust in God.” (Sūrat al-Anfāl 8:61), Dr. Mustafa Khattab, The Clear Quran

﴿وَإِن جَنَحُوا۟ لِلسَّلْمِ فَٱجْنَحْ لَهَا وَتَوَكَّلْ عَلَى ٱللَّهِ﴾ [سُورَةُ الأَنفَالِ: ٦١]

The rules of war protected non-combatants

After a woman was found killed during an expedition, the Prophet ﷺ “forbade the killing of women and children” (Sahih al-Bukhari 3015; also 3014, Sahih Muslim 4547–4548 and Muwatta Malik 967).13

Another report gives the reason. Seeing soldiers around the body of a woman, the Prophet ﷺ said, “This one was not among those who were fighting,” then ordered Khalid ibn al-Walid not to kill “any children, or women, or any labourer” (Sunan Ibn Majah 2842).14

The governing distinction is combatant versus non-combatant, not Muslim versus non-Muslim. The woman’s religion is irrelevant to the ruling; the labourer is protected because he is not fighting. Standing instructions to commanders also prohibit treachery, mutilation and the killing of children.5

A report attributed to Abu Bakr, the first caliph, similarly tells armies sent to Syria not to kill women, children, the elderly or infirm; destroy inhabited places or fruit trees; slaughter livestock except for food; or disturb monks in their monasteries. Its chain is weaker than the Prophetic reports above, so it is evidence of reported early practice rather than the basis of the ruling.15

Forced conversions did occur

There were periods in which Muslim rulers did use force to impose religious conformity.

The Almohads in twelfth-century North Africa and al-Andalus forced Jews and Christians to profess Islam. Their policy broke with the earlier arrangement under which those communities had lived for centuries as non-Muslim subjects.16

From 1501, the Safavid state used coercion in the transformation of predominantly Sunni Iran into a Twelver Shi’i state. Sunnis faced pressure, persecution and, in some places, forced conformity to Shi’i practice.16

These cases are important precisely because they are identifiable as forced conversion. Muslim rule and forced conversion are not the same historical category; otherwise there would be no meaningful contrast between these episodes and the centuries in which large non-Muslim communities remained under Muslim government.

Political pressure could still encourage conversion

Absence of forced conversion does not mean political rule had no effect on religious choices. Muslim government could make conversion advantageous through taxation, access to office, patronage or proximity to a Muslim ruling class. The jizya created a financial incentive in some periods, including measurable effects among poorer converts; the evidence is discussed in The Jizya.8

Some rulers also imposed discriminatory or coercive policies short of a general forced-conversion program. Aurangzeb, for example, restored the jizya in Mughal India in 1679, and his reign included the destruction of particular Hindu temples.17

But pressure, incentive and forced profession of faith are different claims. The historical test remains the same: conversion was usually gradual, substantial non-Muslim populations persisted, and the explanation cannot account for regions such as Indonesia where Islam spread without Muslim conquest or a jizya-paying subject population.

Conclusion

The phrase “spread by the sword” collapses three different histories into one.

Muslim armies fought for different reasons: some campaigns answered attacks or reported threats, some arose from local political appeals, and others were expansion by states seeking power and territory. When conquest occurred, it ordinarily changed the ruling authority rather than the population’s religion. Non-Muslim communities could remain for centuries; some Jewish and Christian institutions flourished, and some groups oppressed by previous rulers cooperated with or preferred the new order. Conversion then followed its own, much slower history—and in many regions spread without Muslim conquest at all.

Forced conversion did occur, and political rule could create pressure to convert. Neither fact makes forced conversion the general mechanism by which Islam spread.

Conquest explains the spread of Muslim rule. It does not, by itself, explain the spread of Muslim belief.

Back to The Qur’an and the Messenger ﷺ

Notes

Fred M. Donner, The Early Islamic Conquests (Princeton University Press, 1981), explicitly distinguishes the causes of the conquests from the causes of their success and from subsequent Arab migration. Donner’s later work likewise treats the conquests as the expansion of a newly organised state while examining how centrally directed that expansion was.
The killing of al-Harith ibn Umayr as the immediate background to Mu’tah belongs to the traditional Muslim historical account, preserved in the sira and maghazi literature. See Ibn Ishaq/Ibn Hisham, The Life of Muhammad, trans. Alfred Guillaume (Oxford University Press), and Ayman S. Ibrahim, Muhammad’s Military Expeditions: A Critical Reading in Original Muslim Sources (Oxford University Press, 2024), ch. 7. The early narrative sources postdate the events and should be read as historical traditions rather than contemporary diplomatic records.
Al-Tabari, The History of al-Tabari, vol. IX, The Last Years of the Prophet, trans. Ismail K. Poonawala (SUNY Press, 1990), pp. 47–48 and translator’s note 341. The note records al-Waqidi’s report that news reached the Prophet ﷺ of Byzantine forces gathering and threatening Medina, but that the information proved exaggerated. No Byzantine army was encountered at Tabuk.
On Euphemius’ appeal to the Aghlabids and its role in the beginning of the conquest of Sicily in 827, see Theresa Jäckh, “827: The Chronicon Salernitanum on the Muslim Conquest of Sicily,” Transmediterranean History 1.2 (2019), and “827: al-Nuwayri on the Beginning of the Muslim Conquest of Sicily,” Transmediterranean History 3.1 (2021). Greek, Latin and Arabic accounts differ on Euphemius’ motives, but agree on his attempt to secure Aghlabid assistance.
Sahih Muslim 4521. The full report also prohibits embezzling spoils, breaking pledges and mutilating the dead. It instructs commanders not to grant terms in the name of God and His Prophet ﷺ when their own guarantee would suffice, because breaking a human pledge is the lesser wrong.
Theophanes, Chronographia, writing in the early ninth century from a hostile Byzantine perspective, describes Umar entering Jerusalem in rough camel-hair garments and refusing finer clothing offered by Patriarch Sophronius; see the translated extract collected by Loyola University Chicago, “Theophanes”. The account that Umar and his attendant shared one mount and that the attendant was riding as they approached Jerusalem belongs to later Muslim historical tradition and is cited here as tradition rather than contemporary testimony. The refusal-to-pray account is preserved by Eutychius of Alexandria (Sa’id ibn Batriq, d. 940), roughly three centuries after the conquest. See Denys Pringle, The Churches of the Crusader Kingdom of Jerusalem, vol. III (Cambridge University Press), pp. 9–10, citing Eutychius. Its late date is why the text uses it as an illustrative remembered ideal rather than proof of the precise events.
Richard W. Bulliet, Conversion to Islam in the Medieval Period: An Essay in Quantitative History (Harvard University Press, 1979), includes conversion curves for Iran, Iraq, Egypt, Tunisia, Syria and Spain. Bulliet infers conversion dates from Muslim given names in biographical dictionaries. His graphs count eventual converts, not entire populations—important in Syria and Egypt, where large minorities did not convert. Later scholarship has revised particular regional curves, especially for al-Andalus, but the general pattern remains gradual conversion over several centuries rather than sudden change.
Mohamed Saleh and Jean Tirole, Taxing Identity: Fiscal Policy and Conversions in Early Islam, Toulouse School of Economics, 5 December 2018, testing their model against “Egypt’s conversion to Islam between 641 and 1200.”
Census of India 2011, religion tables: approximately 79.8 per cent Hindu and 14.2 per cent Muslim. The Delhi Sultanate is conventionally dated from 1206 and Mughal rule ends in 1857; Muslim sovereignty varied considerably by region and period.
On Jewish life under the Abbasids, see Marina Rustow, “Jews under Medieval Islam,” Oxford Bibliographies in Jewish Studies, which notes that the Babylonian academies attained “unprecedented prestige and influence” over Jewish communities across the Islamic world; see also Eve Krakowski, “Jews and Diaspora in the Medieval Islamic Middle East,” The Oxford Handbook of the Jewish Diaspora. On the Church of the East, see Philip Wood, “The Church of Baghdad: A New Past for Christian Iraq”, describing the Abbasid period after the transfer of the catholicosate to Baghdad as one of “unprecedented power and influence.” On Syriac Christians in the translation movement, see R. Y. Ebied, “The Syrian impact on Arabic literature,” The Cambridge History of Arabic Literature.
For Visigothic Spain, William D. Phillips Jr and Carla Rahn Phillips, A Concise History of Spain, ch. 3, note both Visigothic oppression of Jews and the allegiance given to Tariq’s forces by significant elements of Jewish communities. The broader history includes royal forced conversions of Jews in Visigothic Spain; see The Cambridge History of Christianity, ch. 7. For Egypt, the Chronicle of John of Nikiu records Benjamin’s return after exile under the Romans and says that Amr “took none of the property of the Churches”; English translation. Modern scholarship cautions against treating Coptic attitudes as uniform, while confirming prior imperial pressure on the anti-Chalcedonian church and the flourishing of Coptic language and literature after the conquest; see Roger S. Bagnall (ed.), Roman Egypt, ch. 6. On non-Chalcedonian Christians later seeking refuge in Muslim territory during Byzantine reconquest, see Jean Richard, “The Eastern Churches,” The New Cambridge Medieval History.
On Southeast Asian Islamisation through merchants, trade networks and conversion of rulers, see standard histories of Southeast Asian Islam; for a concise modern discussion, see Stelios Michalopoulos, Alireza Naghavi and Giovanni Prarolo, “Trade and Geography in the Spread of Islam,” The Economic Journal 128 (2018). Pew Research Center, using Indonesian Ministry of Religious Affairs data released in 2022, reports roughly 242 million Muslims in Indonesia.
Sahih al-Bukhari 3015 and 3014; Sahih Muslim 4547–4548; Muwatta Malik 967.
Sunan Ibn Majah 2842, graded hasan sahih and sahih li-ghayrihi in the collection’s apparatus.
Muwatta Malik 968, graded mauquf da’if. Mauquf means the report ends with a Companion rather than reaching the Prophet ﷺ; da’if means its chain is weak. It is cited here only as reported early practice, while the legal principle is established by the Prophetic reports above.
For the Almohads, see Maribel Fierro, “Conversion, ancestry and universal religion: the case of the Almohads in the Islamic West,” Journal of Medieval Iberian Studies 2.2 (2010), and “The Religious Policy of the Almohads,” in The Oxford Handbook of Islamic Theology. For Safavid Iran, see Encyclopaedia Iranica, “Shi’ism in Iran Since the Safavids”; coercion was especially severe in the early Safavid period, although the longer conversion of Iran to Shi’ism was more complex and gradual.
Aurangzeb reimposed the jizya in 1679. Temple destruction under his reign is historically documented but was not a uniform policy toward every Hindu temple; particular demolitions were often entangled with political rebellion and imperial control. See Satish Chandra, “Jizyah and the State in India during the 17th Century,” Journal of the Economic and Social History of the Orient 12.3 (1969), and Richard M. Eaton’s work on temple desecration in premodern India.

Qur’an verses cited

2:190 — fighting restricted to those who fight, with limits on transgression
2:256 — no compulsion in religion
8:61 — if the enemy inclines to peace, accept peace
10:99 — belief cannot be forced
16:125 — invite through wisdom, good counsel and argument