Did Islam Spread by the Sword? A Critical Look at Forced Conversions

The question of forced conversions to Islam in history is a cornerstone of the centuries-old “spread-by-the-sword” narrative that has been (and continues to be) used to demonize Islam and Muslims. However, many leading present-day historians have challenged this narrative. They recognize that there have been cases of forced conversion, but also that these were rare, exceptional, occurred in particular contexts, and in violation of the Qur’anic prohibition of this practice. This article provides a cursory perusal of some of the arguments that have been used to discredit this narrative and examines three cases of forced conversion to Islam in history: the spread of Islam in South Asia, the Ottomans’ devshirme system, and Imam Yahya’s “Orphans’ Decree” in Yemen.

Historians describe Islam as one of the youngest of the world’s major religions, having emerged only in the early 7th century CE. Within a century of Prophet Muhammad ﷺ passing away, the early Muslims established an empire stretching from modern-day Spain in the west to India in the east. Many of the peoples who were conquered in this process eventually embraced Islam, laying the foundations for today’s global Muslim community, which is widely regarded as the world’s second-largest and fastest-growing religious group.^[1]

The tremendous numerical success of what European Christendom regarded as a ‘wicked religion’ has confounded observers for centuries. The only conceivable answer, in the minds of many, was (and is) that Islam must have initially spread through its inherent “bestial cruelty,” as Peter the Venerable put it in the 12th century.^[2] From this perspective, the present-day size and growth of the Muslim community is a product of episodes of history during which Islam was “spread by the sword.”

A key “fact” of this spread-by-the-sword narrative is the notion of forced conversions of non-Muslims to Islam. This is part of a constellation of questionable “facts” that have been repurposed many times in history. Peter the Venerable, though he was generally an advocate for the academic study of Islam (for the purpose of refutation), penned his scathing words in the context of the Crusades. In later centuries, European colonizers (such as the British in India) attempted to convince their non-Muslim subjects of their own benevolence by comparing it to the cruelty―real or imagined―of earlier Muslim conquerors.^[3] Today, the Islamophobia industry’s association of Islam and Muslims with barbarity and terrorism can be neatly justified by presenting this as the latest episode of Islam being spread “by the sword.”^[4]

The notion of Islam being “spread by the sword” can be traced to the Crusades and remained a cornerstone of European Christians’ anti-Islam polemics for centuries. It was picked up in the late 19th and early 20th centuries by Orientalist scholars such as Sir William Muir,^[5] many of whom―as British colonial officials and/or active Christian missionaries―were in a position to benefit from the vilification of Islam to non-Muslim audiences (i.e., “divide et impera”). The Orientalists crystallized and legitimized pre-existing traditional oral history―or mythistory, to use William McNeill’s term^[6]―and in doing so “translated [historical memory] from myths to facts with a rational scientific attitude.”^[7]

However, even in the ranks of the Orientalists there were those, such as Sir Thomas Arnold and De Lacy O’Leary, who chopped away at the narrative of forced conversions to Islam. O’Leary wrote in 1923 that “the legend of fanatical Muslims sweeping through the world and forcing Islam at the point of the sword upon conquered races is one of the most fantastically absurd myths that historians have ever repeated.”^[8]

The renowned historian Marshall Hodgson, in his pioneering work The Venture of Islam, articulated essentially the same position.^[9] More recently, Ira Lapidus wrote in A History of Islamic Societies that “European scholars believed that conversions to Islam were made at the point of the sword and that conquered peoples were given the choice of conversion or death. It is now apparent that conversion by force, while not unknown in Muslim countries, was, in fact, rare. Muslim conquerors ordinarily wished to dominate rather than convert, and most conversions to Islam were voluntary.”^[10]

In addition to the historians cited throughout this article, Jamal Malik, Jonathan Berkey, and Kevin Barrett are some of the many other historians who have challenged and discredited the “spread-by-the-sword” narrative.^[11]

That forced conversions to Islam have been rare and exceptional in history is no surprise to those familiar with the Qur’ānic principle: “Let there be no compulsion in religion, for the truth stands out clearly from falsehood” (2:256).^[12] The renowned classical mufassir (commentator) of the Qur’ān, Ismāʿīl ibn Kathīr (d. 1373), said that this passage means, “Do not force anyone to become Muslim, for Islam is plain and clear, and its proofs and evidence are plain and clear. Therefore, there is no need to force anyone to embrace Islam.”^[13] Muslims have a collective responsibility to share the message of Islam, but the normative way to do this has been clearly described in the Qur’ān: “Invite all to the Way of your Lord with wisdom and kind advice, and only debate with them in the best manner” (16:125).

The second option has been derogatorily termed “dhimmitude” by Bat Ye’or.^[15] Ye’or and others have argued that the requirement of paying the jizya and holding dhimmi status was a form of “compulsion in religion.” Ye’or’s position has been challenged by eminent historians, including Bernard Lewis―who cannot be credibly accused of apologetic revisionism, given that his own views in other contexts are far from “Muslim-friendly”―and Chase Robinson.^[16]

The dhimma system was “in conformity with the usual custom of all mediaeval societies where non-dominant religious communities were concerned” and, for comparison, “there is no doubt that the history of the dhimma compares favorably with the treatment of non-Christians in Europe during most of the premodern era.”^[17] Though there are certainly examples of abuse of this system―such as by the Fātimid “mad caliph” al-Hākim (r. 996-1021)^[18]―these are rare and “often occurred on the outer fringes of the Islamic world, especially in the presence of an urgent menace coming from outside.”^[19]

‘Umar is generally regarded as having been one of the most stringent Muslim leaders, and Jerusalem was a very important city for Muslims. It could be expected, then, that if the early Muslims were so keen to forcibly convert non-Muslims to Islam, ‘Umar certainly would not have made an exception for Jerusalem. Similar examples are scattered throughout Islamic history, such as the agreement (in 713) between ʿAbd al-ʿAzīz ibn Mūsa and Theodemir that “they [i.e., Theodemir’s followers in Spain] will not be coerced in matters of religion, their churches will not be burned, nor will their sacred objects be taken from them…”^[21]

It was in part due to such protections that “for at least two centuries the majority of the inhabitants of the Islamic empire were non-Muslims.”^[22] Furthermore, according to Hugh Kennedy, forced conversion to Islam was “almost impossible” following the early Muslim conquests, as Muslims were a small minority in newly-conquered areas―perhaps about 10% of the population in Egypt and 20% in Iraq. “In these circumstances, forcing unwilling people to convert was out of the question.”^[23]

In the regions conquered by Muslims by 732 (i.e., in the first century after Prophet Muhammad ﷺ), Islam did not become a majority religion until 850-1050. Nearly all of Iran, for example, had been conquered by 705; however, empirical research by Richard Bulliet has shown that it was only in the mid-9th century that the Muslim population of Iran reached 50%, and it took nearly another century for that figure to hit 75%.^[24] As some historians have pointed out, “if forced conversion to Islam had been the impetus behind the conquests, they were a miserable failure.”^[25]

It has even been argued that, far from forcing non-Muslims to convert to Islam, many Muslim rulers actually preferred to rule over non-Muslims and collect the jizya from them. One reason for this was that the zakāt (charity that is obligatory upon Muslims) collections were often redistributed locally in the provinces and could only be used in certain ways, but the jizya was sent to the central treasury in the capital, was paid in cash, and could be used at a ruler’s discretion.^[26]

One relevant case led to the Umayyad general Jarrāh ibn ʿAbdullāh al-Hakami (d. 730) being removed from his post by the caliph, ʿUmar ibn ʿAbd al-Azīz. Al-Hakami had been ruling in Khurasan, and he actually prevented his subjects from embracing Islam and continued to demand the jizya even from those who converted to Islam. “Indeed, you only embraced Islam to avoid paying it,” he said to them. ʿUmar ibn ʿAbd al-Azīz removed him from his post, saying, “Allāh only sent Muhammad ﷺ as a preacher and not as a tax collector.”^[27]

Conversion could be effectively discouraged for other reasons and in other ways. An example is that of Haydar ibn Kāwūs (d. 841), a general of the ʿAbbāsid caliph al-Muʿtasim and ruler of a small principality in Central Asia called Ushrusana. Though he was a Muslim, he had made an agreement with his non-Muslim subjects to not permit daʿwah within Ushrusana and he kept his word, punishing two Muslims for preaching in his jurisdiction. He then faced trial for doing this in Muʿtasim’s court.^[28]

Perhaps the most widely-circulated examples of forced conversion to Islam in history come from South Asia. The American historian Will Durant asserted in 1935 that “the Mohammedan conquest of India is probably the bloodiest story in history,” explaining that “millions of Hindus were converted to Islam by sword during this period” (i.e., 800-1700 CE).^[29] (Interestingly, Durant also stated that “most history is guessing, and the rest is prejudice.”)^[30] This position was popularized by British Orientalist “administrator-historians” (e.g., Henry Elliot) in India in the 19th and early 20th centuries, and has been borrowed and further developed by nationalist or even Hindutva-sympathetic historians such as R.C. Majumdar and Koenraad Elst, respectively.^[31]

Furthermore, Yohanan Friedmann has pointed out that the phrases found in Persian historical sources that Elliot and his followers have used as evidence for their position are ambiguous and can be interpreted in different ways. “They submitted to Islam,” for example, could refer to Islam the religion, the Muslim state, or the “army of Islam,” and a contextual reading usually supports one of the latter two interpretations.^[35] Hence, a phrase intended to mean that a group submitted to a Muslim army under duress could easily―and perhaps deliberately―be misinterpreted as meaning that the group was forced or pressured to submit to the religion of Islam.

Richard Eaton points out that if the spread-by-the-sword position were accurate, “one would expect that those areas exposed most intensively and over the longest period to rule by Muslim dynasties―that is, those that were most fully exposed to the ‘sword’―would today contain the greatest number of Muslims.” However, according to the earliest reliable census, the population of Muslims was significantly higher in Punjab and Bengal (70-90%), both historically on the fringes of Indo-Muslim rule, than it was in the Gangetic Plain (10-15%), the heartland of Muslim rule.^[36] Rowena Robinson points out that Eaton has also convincingly argued that “Islam spread among Bengali cultivators during a period when the [Muslim] rulers were actually opposed to conversion.”^[37]

The Ottomans’ devshirme system is another example of forced conversion to Islam. In this system, young Christian boys were systematically taken from their families, converted to Islam, and trained to serve in the empire’s bureaucracy or in the sultan’s personal military force, the Janissaries. However, this same system often “provided religious minorities with unfettered access to the highest government positions.”^[38] An example is that of Sokullu Mehmet Pasha (d. 1579), a Slav from Bosnia who rose through the bureaucracy to become the empire’s grand vizier, a position from which he was able to support Bosnia’s Christian community, though he himself remained Muslim.^[39]

Thus, the same devshirme system that forcibly converted some non-Muslims to Islam could at times also play a role in preserving non-Muslim communities and protecting them from mass forced conversion. This may, in part, explain why conversion to Islam proceeded so slowly in some parts of the Ottoman Empire. The region that makes up modern-day Albania, for example, was gradually conquered by the Ottomans over the course of the 15th century, but conversion to Islam only really took off nearly 200 years later.^[40]

Another case of forced conversion in Islamic history is that of the Orphans’ Decree issued by Imam Yahya al-Mutawakkil (d. 1948) in the early 20th century. After the First World War, the crumbling Ottoman Empire recognized Yahya as its successor in Yemen. Yahya then laid claim to “Greater Yemen,” parts of which were ruled by the British or Yahya’s political rivals. As a leader of the Zaydī community, Yahya also re-introduced Zaydī law, part of which contained the “Orphans’ Decree” requiring his government to forcibly convert orphaned Jewish children to Islam.^[41]

However, a closer look at the case of the Orphans’ Decree reveals that though Imam Yahya himself introduced the decree, he turned a blind eye to the smuggling of Jewish orphans out of Yemen to avoid conversion, and on some occasions even facilitated this process.^[42] The Jews who fled from Yemen exaggerated the number of children forced to convert in an attempt to earn sympathy for the cause of helping more Jewish children escape.^[43] However, the guardians of many Jewish children helped them to escape to Imam Yahya’s jurisdiction rather than from it, and they found refuge with the very regime that was ostensibly bent on forcing them to convert to Islam.^[44]

Imam Yahya, for his part, implemented the decree selectively, and in doing so he revealed his reason for introducing it in the first place: it had little to do with a desire to forcibly convert his Jewish subjects to Islam, and more to do with asserting his authority in an unstable political environment in post-WWI Yemen.^[45] Though a reductionist narrative of the Orphans’ Decree may immediately present it as a case of Islam being “spread by the sword” among Yemen’s Jews, it is noteworthy that Jewish sources describe Imam Yahya in very favorable terms.^[46] At the same time, “Yemeni Jewish writings discuss the forced conversion of Jewish orphans explicitly but are reluctant to mention voluntary conversions.”^[47]

The preceding case studies present examples of forced conversion to Islam in history―in the case of South Asia, there are sufficient grounds to believe that some forced conversions likely occurred. Any of these could easily be used to justify the 800-year-old narrative that Islam was spread by the sword. And yet, even a cursory investigation of each case reveals thought-provoking details that cannot be used to justify any of these cases, but can—and should—remind us that historical events are rarely as simple as we too easily believe.

Even most advocates of this narrative must certainly be aware that reality is more complicated than that. In fact, Peter the Venerable himself, even as he was citing the Muslims’ “bestial cruelty” as the only possible cause for the spread of Islam, had privately asked a colleague to write a refutation of Islamic teachings “to consider and provide for weak members of the Church, who are commonly led astray or even unthinkingly won over by trivial arguments.”^[48] Clearly, Peter had evidence that at least some non-Muslims were responding favorably to Islamic preaching.

In this article, I have deliberately made minimal use of examples of the peaceful spread of Islam in many―and, at least at some point in history, in all―regions of the world. Southeast Asia and West Africa, for example, are two of the regions with the largest Muslim populations in the world today, but neither has ever been significantly subjected to “the sword” by Muslims.^[49]

I have done this because the purpose of this article is not to gloss over or cover up the fact that there have likely been actual cases of forced conversion to Islam in history. These may have been rare, exceptional, and in contradiction to clear Qur’ānic directives, as has been discussed above, but this certainly does not mean that they did not occur or that we can afford to turn a blind eye to them.

In fact, the underlying message of the discussion is precisely the opposite. We need to stop glossing over history by subscribing to shallow (and often demonizing) narratives such as “Islam was spread by the sword” or, for that matter, that classical Muslim societies were utopic or that the sword never played a role in the spread of Islam.^[50] All of these are myths, with the truth to be found somewhere in the messy gray area between them―a gray area that conscientious and truth-seeking people must roll up their sleeves to diligently explore.

[1] Conrad Hackett and Michael Lipka, “Why Muslims are the world’s fastest growing religious group,” 6 April 2017, http://www.pewresearch.org/fact-tank/2017/04/06/why-muslims-are-the-worlds-fastest-growing-religious-group/.

[2] Karen Armstrong, “We cannot afford to maintain these ancient prejudices against Islam,” The Guardian, 18 September 2006, https://www.theguardian.com/commentisfree/2006/sep/18/religion.catholicism.

[3] Barbara Metcalf, “Too Little and Too Much: Reflections on Muslims in the History of India,” The Journal of Asian Studies 54, no. 4 (1995): 953-54.

[4] See, for example, Robert Spencer, The Politically Incorrect Guide to Islam (And the Crusades) (Washington, DC: Regnery Publishing, 2005), 107-17.

[5] William Muir, The Caliphate: Its Rise, Decline, and Fall (London, 1898; reprint Beirut: Khayats, 1963), 45

[6] William H. McNeill, “Mythistory, or Truth, Myth, History and Historians”, The American Historical Review 91, no. 1 (1986): 8.

[7] Amalendu Misra, Identity and Religion: Foundations of Anti-Islamism in India (New Delhi: Sage Publications, 2004), 223.

[8] De Lacy O’Leary, Islam at the Cross Roads (New York: E.P. Dutton and Co., 1923), 8.

[9] Marshall G. Hodgson, The Venture of Islam, Volume 1: The Classical Age of Islam (Chicago: The University of Chicago Press, 1974), 199.

[10] Ira M. Lapidus, A History of Islamic Societies (New York: Cambridge University Press, 2014), 271.

[11] Jamal Malik, Islam in South Asia: A Short History (Leiden: Brill, 2008), 183;  Jonathan Berkey, The Formation of Islam: Religion and Society in the Near East, 600-1800 (New York: Cambridge University Press), 162; Kevin Barrett, “Is Islam Reasonable?,” in Reasonable Perspectives on Religion, ed. Richard Curtis (Plymouth, UK: Lexington Books, 2010), 204.

[12] All quotations from the Qur’an in this article are taken from Mustafa Khattab, The Clear Quran: A Thematic English Translation of the Message of the Final Revelation (Lombard, IL: Book of Signs Foundation, 2016).

[13] Ismā‘īl ibn Kathīr, Tafsīr Ibn Kathīr (Vol. 2) (New York: Darussalam Publishers, 2003), 30.

[14] Michael Bonner, Jihad in Islamic History: Doctrines and Practice (Princeton, NJ: Princeton University Press, 2006), 90.

[15] Bat Ye’or, Islam and Dhimmitude: Where Civilizations Collide, tr. Miriam Kochan and David Littman (Madison/Teaneck, NJ: Fairleigh Dickinson University Press, 2002), 50.

[16] Bernard Lewis, “The New Anti-Semitism”, The American Scholar 75, no. 1 (2006), https://theamericanscholar.org/the-new-anti-semitism; Chase F. Robinson, “Review of “The Decline of Eastern Christianity under Islam, from Jihad to Dhimmitude: Seventh-Twentieth Centuries by Bat Ye'or, Miriam Kochan, David Littman””, Middle East Studies Association Bulletin 31, no. 1 (1997): 98.

[17] Bonner, Jihad in Islamic History, 91.

[18] Milka Levy-Rubin, “New Evidence Relating to the Process of Islamization in Palestine in the Early Muslim Period: The Case of Samaria,” Journal of the Economic and Social History of the Orient 43, no. 3 (2000): 263.

[19] Bonner, Jihad in Islamic History, 91.

[20] Hugh Kennedy, The Great Arab Conquests: How the Spread of Islam Changed the World We Live In (Philadelphia, PA: Da Capo Press, 2007), 91.

[21] Ibid, 315.

[22] William Cleveland and Martin Bunton, A History of the Modern Middle East (4th ed.) (Westview Press, 2009), 14.

[23] Hugh Kennedy, “Was Islam Spread by the Sword?: The Early Muslim Conquests Revisited,” The Yale Conference on Religion and Violence (Yale University, New Haven, CT, February 16, 2008).

[24] Richard W. Bulliet, “Conversion to Islam and the Emergence of a Muslim Society in Iran”, in Conversion to Islam, ed. Nehemia Levtzion (New York: Holmes & Meier Publishers, 1979), 36.

[25] Richard C. Martin, “Conversion to Islam by Invitation,” in Sharing the Book: Religious Perspectives on the Rights and Wrongs of Proselytism, eds. John Witte Jr. and Richard C. Martin, (Eugene, OR: Wipf and Stock Publishers, 2008), 103.

[26] Kennedy, “Was Islam Spread by the Sword?”

[27] Ismā‘īl ibn Kathīr, Al-Bidāyah Wa’l-Nihāyah, ed. ʿAbd Allāh ibn ʿAbd al-Muḥsin Turkī (Al-Qāhirah (Cairo): Dār Hajr, 1997), 12:667.

[28] Kennedy, “Was Islam Spread by the Sword?”

[29] Will Durant, The Story of Civilisation: Our Oriental Heritage (New York: Simon and Schuster, 1935), 459.

[30]  Ibid, 12.

[31] Sudeshna Guha, “Negotiating Evidence: History, Archeology, and the Indus Civilisation”, Modern Asian Studies 39, no. 2 (2005): 403.

[32] See, for example, Derryl N. MacLean, Religion and Society in Arab Sind (Leiden: Brill, 1989), 25-27.

[33] Peter Hardy, “Modern European and Muslim Explanations of Conversion to Islam in South Asia: A Preliminary Survey of the Literature”, The Journal of the Royal Asiatic Society of Great Britain and Ireland 2 (1977): 185; the quotation at the end of this sentence is from the discussion about Hardy’s position in Richard M. Eaton, The Rise of Islam and the Bengal Frontier, 1204-1760 (Berkeley: University of California Press, 1993), 114.

[34] Rowena Robinson, “Modes of Conversion to Islam”, in Religious Conversion in India: Modes, Motivations, and Meanings, eds. Rowena Robinson and Sathianathan Clarke (New Delhi: Oxford University Press, 2003), 23.

[35] Yohanan Friedmann, “A Contribution to the Early History of Islam in India,” in Studies in Memory of Gaston Wiet, ed. Myrian Rosen-Ayalon (Jerusalem: Institute of Asian and African Studies, 1977), 322.

[36] Eaton, The Rise of Islam and the Bengal Frontier, 115.

[37] Robinson, “Modes of Conversion to Islam”, 26.

[38] Ibrahim Kalin, “Islam and Peace: A Survey of the Sources of Peace in the Islamic Tradition”, Islamic Studies 44, no. 3 (2005): 347.

[39] Ibid, 348.

[40] Krstic, Tijana. Contested Conversions to Islam: Narratives of Religious Change in the Early Modern Ottoman Empire (Stanford, CA: Stanford University Press, 2011), 21.

[41] Ari Ariel, “A Reconsideration of Imam Yahya's Attitude Toward Forced Conversion of Jewish Orphans in Yemen,” Shofar: An Interdisciplinary Journal of Jewish Studies 29, no. 1 (2010), 97.

[42]  Ibid, 103.

[43]  Ibid, 99.

[44]  Ibid, 103.

[45]  Ibid, 111.

[46]  Ibid.

[47] Bat-Zion Eraqi Klorman, “Muslim Society as an Alternative: Jews Converting to Islam,” Jewish Social Studies 41, no. 1 (2007): 90.

[48] A.J. Forey, Western Converts to Islam (Later Eleventh to Later Fifteenth Centuries), Traditio 68 (2013): 154.

[49] See, for example, Lamin Sanneh, Beyond Jihad: The Pacifist Tradition in West African Islam (New York, Oxford University Press, 2016), including a discussion of an established case of forced conversion to Islam in Ghana which most likely never even happened (p. 284).

[50] Hugh Kennedy (“Was Islam Spread by the Sword?”), for example, offers a simple but much more useful starting point to our investigation: “Islam was not spread by the sword but without the sword it would not have spread.”