Empire of Refugees
North Caucasian Muslims and the Late Ottoman State
Vladimir Hamed-Troyansky



ON A SCORCHING DAY in the late summer of 1863, fishermen in the Ottoman port of Trabzon, in the southeastern corner of the Black Sea, noticed several boats on the horizon. The boats carried Muslim refugees fleeing the Russian conquest of their lands in the Caucasus. Dehydrated and disoriented, Circassian families disembarked in the port. The next day, more boats sailed into the harbor. Dozens of boats kept arriving daily not only in Trabzon but also in Samsun and Sinop. By the end of autumn, Circassian refugees exceeded the resident population in Ottoman port cities on the Black Sea. Inns, schools, and mosques were filled to the brink, and refugees slept in covered bazaars, stables, and the streets. Deadly epidemics of typhus and smallpox broke out, devastating refugee communities. The ports in northern Anatolia started redirecting boats with refugees to Istanbul, Burgas and Varna (in Bulgaria), and Köstence (in Romania). The onset of winter did not stop the arrival of refugees, and the following year even more refugees disembarked on Ottoman shores. Between 1863 and 1865, up to half a million Circassians fled the Caucasus for the Ottoman Empire.1 It was the largest refugee crisis that the Ottomans had experienced by then.

Migration from the Caucasus continued for the next half century. Abkhazians, Chechens, Ingush, Balkars, Karachays, Ossetians, Avars, Lezgins, and other Muslim communities left their native mountains. One constant in late Ottoman history was the continuous arrival of Muslim refugees from Russia. While the earlier groups of refugees had fled an ethnic cleansing perpetrated by the Russian military in the North Caucasus, the latter parties were pushed out by Russia’s new civil reforms and settler colonial policies. Between the 1850s and World War I, approximately a million North Caucasian Muslims had left the tsardom in what was one of the greatest displacements in Russian imperial history.

The Ottoman government maintained an open-door policy for North Caucasian refugees. They arrived when the Ottoman Empire was steadily losing territory and population in the Balkans and North Africa. Muslim refugees fit neatly into the Ottoman government’s agenda to stem demographic decline, revitalize the economy, and solidify the imperial hold on far-flung provinces. Within two generations, North Caucasian refugees were resettled throughout the Ottoman Empire in the following fourteen countries today: Turkey, Syria, Jordan, Lebanon, Israel, Iraq, Georgia, Bulgaria, Romania, Serbia, Kosovo, Greece, Cyprus, and North Macedonia. The Ottomans considered settling Circassian refugees also in Albania, Bosnia, Montenegro, and Libya.2 Temporary refugee camps existed in Palestine, and some North Caucasians moved, without Ottoman support, to Egypt.

The successive Muslim migrations turned the Ottoman state into an empire of refugees. In addition to North Caucasians, hundreds of thousands of Muslims from Crimea, the Balkans, the South Caucasus, and Crete, as well as smaller groups from North Africa, Central Asia, and Afghanistan, arrived in the Ottoman Empire as refugees. Many parts of the empire became a refugee country, where one was more likely to hear Circassian or Abkhazian than Turkish, Arabic, or Greek. The Ottoman Empire fashioned itself as a refuge for Muslims displaced in the age of European imperial conquest and colonialism. Meanwhile, the resettlement of Muslim refugees changed the empire from within and was a harbinger of population transfers and forced homogenization that befell the Middle East and the Balkans in the twentieth century.


Empire of Refugees is a history of migration that examines how North Caucasian refugees transformed the late Ottoman Empire and how the Ottomans managed Muslim refugee resettlement. This book advances several arguments. First, between 1860 and World War I, the Ottoman government had constructed a refugee regime, which coexisted with, but was distinct from, the Ottoman immigration system. The Ottoman Refugee Commission (Ott. Tur., Muhacirin Komisyonu), founded in 1860, implemented the refugee regime. The Commission was responsible for settling Muslim refugees from Russia, arriving in the aftermath of the Crimean War of 1853–56 and the Caucasus War of 1817–64, and Ottoman Muslims displaced during the Russo-Ottoman War of 1877–78, the Balkan Wars of 1912–13, and World War I. Having settled between three and five million Muslim refugees in total, the Commission presided over the demographic, economic, and social transformation of the remaining Ottoman territories, especially Anatolia. The Ottoman refugee regime built on the Ottoman Immigration Law of 1857, which had set the terms for immigration into the empire for anyone, irrespective of their faith, and the Land Code of 1858, governing land ownership and tenure. After the 1860s, the vast majority of immigrants in the Ottoman Empire were Muslim refugees. The Commission developed a set of additional policies and subsidies specific to refugee needs, inaugurating a regime of expectations—of protection and settlement by Muslim refugees and of obligations and loyalty by the Ottoman government.

The Ottoman refugee regime preceded and has a distinct genealogy from the contemporary international refugee regime. The modern refugee regime is a product of the United Nations and is anchored by the Convention Relating to the Status of Refugees, better known as the Refugee Convention of 1951. It has roots in the interwar era, when the League of Nations implemented ad hoc procedures to resolve refugee crises, arising out of the collapse of the Ottoman and Russian empires.3 The legal status of a modern refugee is derived from one’s citizenship in a nation-state. In recent years, historians demonstrated that the ideology of modern humanitarianism and such practices as population exchange, refugee transfer, and territorial partition, which were central to the interwar refugee regime, had roots in the Middle East.4 Humanitarian crises in the post-Ottoman world defined global conversations about protecting refugees. This book suggests a further historiographical corrective. The nineteenth-century Ottoman Empire, struggling mightily with European annexations, created its own nonwestern and nonsecular system of categorizing, sheltering, and resettling refugees. The status of refugee, or muhajir, was based not on one’s subjecthood or citizenship but on facing religious persecution and seeking refuge in the sultan-caliph’s domains. A refugee being Muslim, while not a codified requirement, was an expectation and raison d’être of the Ottoman refugee regime.

Second, refugee resettlement accelerated the collapse of the empire in the Balkans and fortified Ottoman rule in the outlying regions of the Levant and Anatolia. Historians explain the disintegration of the Ottoman Empire by geopolitics, including European annexations throughout the nineteenth century and the empire’s defeat in the Balkan Wars of 1912–13 and World War I, and by cultural forces, especially the sectarianism and nationalism that tore through the empire’s social fabric.5 By revisiting late Ottoman history through the prism of migration and political economy, I show that Muslim refugee resettlement had a profound impact on both the survival and the demise of the Ottoman Empire. Ottoman refugee resettlement was an ambitious project of the Azizian (1861–76) and Hamidian (1876–1908/9) governments. The Ottoman government expected Muslim refugees to become productive farmers. It happened for some but not all. The economic well-being of refugee communities depended on Ottoman support. That support could come in different forms: favorable legislation, free land, financial aid, roadway construction, and military backing. How much help the government provided and whether refugees could use it explain why refugee resettlement had vastly different outcomes throughout the empire.

This book examines refugee resettlement in three core parts of the Ottoman Empire: the Balkans, Anatolia, and the Levant. In the northern Balkans, the Ottoman government’s settlement of Muslim refugees from Russia stoked anti-Ottoman sentiments among Balkan Christians. The lack of state funding for refugee households undermined the efficacy of the resettlement program and pushed some refugees toward banditry. The ensuing violence of Circassian gangs, especially against Christians, inflamed social tensions, leading up to the Bulgarian uprising of 1876 and the Russo-Ottoman War of 1877–78. The war ejected the Ottomans from half of their European territories, which they had held for over five centuries. Meanwhile in central Anatolia, Ottoman financial support for refugees was equally limited, but the government backed North Caucasians in their conflict over land with Turkic nomads. Refugee villages survived, well hidden among the mountains, yet their economies stagnated because of a lack of Ottoman infrastructure. In contrast, North Caucasian refugees in the Levant took advantage of Ottoman land reforms and the state-sponsored Hejaz Railway and built thriving villages on the edge of the desert. North Caucasian refugees founded three of the four largest cities in modern Jordan, including its capital city of Amman. The Ottoman government favored refugee farmers, and, through their settlements in the Levant and central Anatolia, it expanded authority into the sparsely settled parts of the empire.

The resettlement of Muslim refugees was key to the disintegration of the Ottoman Empire in several ways. Resettling Muslims from Russia strained Ottoman imperial and provincial budgets at a time of fiscal emergency. The Ottomans took out their first international loans during the Crimean War of 1853–56 and kept borrowing from European banks, in part to pay for the settlement of Muslim refugees. In 1875, the Ottoman state defaulted on its debts, which ushered in the European-controlled Ottoman Public Debt Administration, shrinking Ottoman sovereignty and fiscal control. Yet borrowing for refugee resettlement continued. Furthermore, the resettlement of Muslim refugees intensified intercommunal strife. In almost every place of settlement, refugees came into conflict with local populations over land. Those conflicts soured relations between North Caucasians and their settled and nomadic neighbors, whether Christian, Muslim, or Druze, while helping the Ottoman government to entrench a new land regime in its reformist drive to centralize the empire. Finally, after 1878, the resettlement of Muslim refugees went hand in hand with displacing and dispossessing Ottoman Christians. The Hamidian government altered demographic ratios on the margins of the empire to forestall further loss of territory. While the policies were not consistent and varied from region to region, the general message was clear. The government welcomed foreign Muslims and trusted them to become loyal subjects of the sultan at a time when it viewed its Christian subjects with suspicion. The resettlement of Muslim refugees lay at the heart of the Ottoman government’s ambition to preserve the imperial project.

Third, Russia attempted to control Muslim migration to consolidate its authority over the Caucasus. Russia’s migration policies in the Caucasus helped to manage its “Muslim empire,” which stretched from Crimea and the Caucasus to the Volga region, through the Ural Mountains and into the Kazakh steppe, Central Asia, and Siberia.6 Muslims had long been Russia’s second-largest religious community after Orthodox Christians. By World War I, the Russian tsar counted more Muslim subjects than the Ottoman sultan did. Russia’s migration policies underwent a transformation. During the Caucasus War, the Russian government abetted, promoted, and sponsored Muslim emigration. Between 1862 and 1864, the tsarist military perpetrated ethnic cleansing, expelling western Circassians into the Ottoman Empire. After 1867, however, tsarist authorities changed course to discourage and restrict emigration as a way to keep their new Muslim subjects inside Russia. Simultaneously, after 1861, Russia vigorously opposed the return of North Caucasian refugees from the Ottoman Empire. Within the Caucasus, Russia pursued the policies of forced relocation of North Caucasians from highlands to lowlands and colonization of their territories by Cossacks and Christian peasants. Tsarist migration policies redrew the demographics of the North Caucasus and solidified Russia’s control over its newest region near the borders of the Ottoman Empire and Iran. The North Caucasus is the tsarist empire’s last major acquisition remaining within today’s Russian Federation, unlike other non-Russian regions that had been annexed in the eighteenth and nineteenth centuries and have since become independent, including Poland, Finland, Ukraine, the Baltic states, the South Caucasus, and Central Asia.

Ottoman and Russian migration policies toward North Caucasian Muslims were starkly different, yet they pursued the same goal of consolidating imperial authority. The Ottoman policy was inclusive toward foreign Muslims because it benefited the Ottoman state. The government used refugees to increase the population of a shrinking empire, to bring unused land into cultivation, to rein in nomads, and to strengthen the empire’s hold on Christian-majority frontier regions. The Russian government excluded those North Caucasian Muslims whom it perceived as opponents of Russian rule during and after the Caucasus War. Both empires developed a sectarian logic, equating one’s religious identity with their loyalty to a co-religionist state. The Russians, who had a hard time conquering and suppressing rebellions in the North Caucasus, assumed that indigenous Muslims, especially those who had already left, would be loyal only to the sultan, not the tsar.

This transimperial history of migration explores how the Ottoman and Russian governments tried to control mass movements of people and also how Muslim refugees responded to policies designed to limit their mobility. Refugees were not mere pawns, expelled and resettled at governments’ will, but active agents of history. Most North Caucasians became farmers, transforming economies and landscapes of entire regions and shaping local forms of capital accumulation. Some refused to settle where the Ottoman government had sent them or abandoned their refugee villages. Many petitioned the government and litigated in courts, or joined the army or militias. North Caucasian refugees were transimperial subjects—not solely because they crossed borders but also because of the decisions they made in navigating their displacement and resettlement, negotiating with Russian or Ottoman authorities, and articulating their North Caucasian identity in exile.7 They and their descendants helped to shape the history of the modern Middle East. Today, the Circassian diaspora in Turkey is estimated at between two and three million people and is the country’s second largest non-Turkish minority after the Kurds. Up to 100,000 descendants of North Caucasian refugees live in Syria, 30,000 in Jordan, over 10,000 in Iraq, and 4,000 in Israel.8


The migration and resettlement of North Caucasian Muslims were part of what is often called the first wave of globalization, between 1870 and 1914. Enabled by improvements in transoceanic transportation, migration in this period conjures images of mass voluntary immigration of Europeans to overseas colonies and nation-states. The proliferation of new forms of capitalism, agricultural expansion, and commodification of land accelerated the frontier settlement, mostly by white immigrants, in the Americas, southern Africa, and the Pacific region.9 An inextricable, yet frequently omitted, part of this globalization was another migration—global displacement of indigenous communities, complete with the destruction of their landscapes, legal dispossession, and forced labor.10 The late imperial era unleashed mass forced migration. The North Caucasians’ migration was both of these stories. The Russian army displaced this Muslim community from its homeland, and Christian settlers took over its lands. Meanwhile, having arrived as refugees in the Ottoman Empire, North Caucasians soon became settlers themselves, occupying and working the land that had been claimed by other communities. North Caucasian refugees were victims of the expansion of the Russian imperial frontier southward, through conquest and settlement, and then also helped the Ottoman state to push its internal frontier into territories of nomadic communities on the empire’s margins.11

The term that North Caucasian refugees used to describe themselves was muhajir (Ar., muhājir, pl. muhājirūn; Ott. Tur., muhacir, pl. muhacirler). The Arabic term muhājir is derived from hijra, which denotes the journey of the Prophet Muhammad from Mecca to Yathrib (Medina) in 622 CE. The Prophet Muhammad’s companions who undertook this journey to preserve their nascent religious movement were the first muhajirs. Throughout Islamic history, Muslim communities that had left, or been expelled from, their homelands used this term in emulation of the Prophet’s companions. By the nineteenth century, the term acquired anticolonial and anti-imperialist sentiments, as many regions across the Muslim world were occupied by the European empires. The flight of Muslims for refuge in the Ottoman Empire, the world’s strongest sovereign Muslim state and the seat of the caliph, acutely reverberated throughout the Caucasus, the Balkans, North Africa, and beyond. The present-day translations of refugee in Turkish, Arabic, and Russian are, respectively, mülteci, lājiʾ, and bezhenets. None of these terms were commonly used to refer to North Caucasians between the 1850s and World War I. I will use muhajir throughout the book as well as the terms refugee, immigrant, and emigrant, which all capture different aspects of what being a muhajir entailed, when discussing relevant stages of muhajirs’ experiences.

The English-language term refugee came into popular usage in the aftermath of the collapse of the Ottoman and Russian empires. Previously used to refer to religious and political exiles, it then described Armenian survivors of the genocide and refugees of the Russian civil war who were stranded away from their homeland, stateless, and increasingly seen as a global responsibility.12 It was defined in the United Nations Refugee Convention of 1951 as someone who, “owing to well-founded fear of being persecuted for reasons of race, religion, nationality, membership of a particular social group or political opinion, is outside the country of his nationality and is unable or, owing to such fear, is unwilling to avail himself of the protection of that country” or “unable or, owing to such fear, is unwilling to return to it.”13

The terms muhajir and hijra defined how the Ottoman state conceived of Muslim refugee migration and resettlement and how North Caucasians perceived their displacement. They challenge the artificial distinction between forced and voluntary migration, still prevalent in the study of global migration. Hijra is not only a religious concept but also a distinct type of global migration that draws its legitimacy and language from early Islamic history and comprises different kinds of Muslim displacement, including ones caused by imperial conquest, ethnic cleansing, and partitions over the past two centuries. The North Caucasian hijra stemmed from Russia’s expulsions of Muslims from the Caucasus but also encompassed emigration as a way to escape poverty and to live under the rule of a Muslim sovereign.

In recent decades, historians have demonstrated astounding mobility between the Ottoman and Russian empires. Muslim pilgrims from Russia and its protectorates in Central Asia made an annual hajj, or pilgrimage, to the holy cities of Mecca and Medina, where some stayed as long-term residents.14 Intellectuals from Istanbul, Crimea, Kazan, and Bukhara traveled widely within the Turkic world, which spanned the two empires.15 The two governments captured and exchanged prisoners of war.16 In the Danubian borderlands, Bulgarian, Gagauz, Ukrainian, and Moldavian peasants crossed the Russo-Ottoman border back and forth, fleeing epidemics and tax collectors, and, in the Caucasus borderlands, Armenian, Greek, and Laz villagers and Kurdish and Turkic nomads traversed the porous frontier along with their herds and contraband goods.17 Empire of Refugees focuses on the final half century of the autocratic empires of the Romanovs and Ottomans, when transborder migration increased dramatically. The Russo-Ottoman borderlands were being remade through displacement, and most migrants crossing the border were refugees.

The North Caucasian refugee migration was part of mass population displacements in late Ottoman history. Lord Curzon, Britain’s foreign secretary, famously referred to demographic changes in the Balkans in the early twentieth century as the “unmixing of peoples,” a phrase that later became associated with international negotiations over post-Ottoman demographics in Lausanne in 1923.18 That unmixing, which homogenized the maps of modern-day Turkey, Greece, and Bulgaria, is remembered as the apotheosis of ethnic nationalism, although it also had a significant religious component. The religious “unmixing” of communities in the Russo-Ottoman borderlands had begun earlier. Since the late eighteenth century, Muslims from Russia had been fleeing to the Ottoman Empire, while Ottoman Armenians, Greeks, and Bulgarians had been emigrating to the tsarist Caucasus, Crimea, and Bessarabia.19 The displacements of Muslims and Christians evidenced gradual sectarianization in the Russo-Ottoman borderlands, whereby one’s religious identity became the basis for one’s political identity. The religious sorting in the nineteenth-century borderlands laid the groundwork for more comprehensive, demographic engineering, justified in ethnonational terms in the early twentieth century.20 The migration of North Caucasian Muslims was a turning point in the history of displacement in the Middle East and Eastern Europe. The expulsion of most western Circassians from the Caucasus, followed by their resettlement throughout the Ottoman domains, confirmed a dangerous idea that a modern state could swiftly move a large population elsewhere for its benefit. This book excavates the origins of state-managed population transfer, which would soon be implemented for different ends, including in the 1923 Greek-Turkish population exchange and Stalin’s deportations of Chechens, Ingush, Karachays, and Balkars from the Caucasus in the 1940s.

Muslim hijra to the Ottoman Empire occurred alongside two other migrations. Their terminology draws on the same Arabic root (h–j–r). The first one was the emigration of Ottoman subjects from the Levant to the mahjar, as diasporic lands were known to Arabic speakers. Between 1880 and 1914, about half a million Lebanese and Syrians—mostly Christians but also Jews, Druze, and Muslims—emigrated to South and North America.21 As foreign Muslims fled for refuge in the Ottoman Empire, many Ottoman non-Muslims sought to escape poverty and discrimination at home. The second migration was tahjīr, or tehcir when transliterated from Ottoman Turkish, used to describe deportations by the Ottoman government of Ottoman Greeks in 1914 and Armenians and Assyrians in 1915. The ethnic cleansing, and eventually genocide, of Anatolia’s Christians was unfolding parallel to the immigration of Muslims. Those were mutually reinforcing processes in the state-driven overhaul of Anatolia’s demographics, homogenizing it as a Muslim land, soon to be reimagined as a Turkish one.

The displacement of Muslims from the Caucasus proceeded apace with the Jewish exodus from the Pale of Settlement, in the territories of Poland, Ukraine, Belarus, Moldova, Lithuania, Latvia, and Russia. As with the Crimean Tatars and Circassians in previous generations, discriminatory policies and violence, culminating in pogroms, drove out Jewish communities. Between 1881 and 1914, about 1.5 million Jews emigrated, primarily to the United States, but several tens of thousands made their way to Ottoman Palestine, forming the bulk of early Zionist immigrants of the modern aliyah, or migration to Palestine.22 The Jewish aliyah holds many parallels to the Muslim hijra in the specific directionality and religious mandate of emigration as well as in their origins in persecution and mass flight.

North Caucasian refugees included both free and enslaved people, and their migration transformed late Ottoman slavery in many ways. Circassian, Abkhazian, and Georgian slaves had long been trafficked to the slave markets of Istanbul and Cairo, where they were sold into elite urban households, including the Ottoman imperial harem.23 By the mid-nineteenth century, the Black Sea slave trade was all but gone, and new slaves from the Caucasus were few in the Middle East. The mass arrival of North Caucasian refugees in the 1860s changed that. The Caucasus communities had practiced various forms of bondage, rooted in household and agricultural servitude. Many upper-status muhajirs brought with them their serfs and slaves. The sudden supply of thousands of enslaved Circassians bloated the Ottoman and Egyptian slave markets and depressed prices, making Circassian slaves more accessible to a greater range of urban households and significantly hampering abolitionist efforts in the Middle East.24 Moreover, North Caucasian forms of slavery were transplanted into the Ottoman countryside, where many muhajirs continued to be owned by other muhajirs.25 For many North Caucasian Muslims, their resettlement in the Ottoman domains meant their continued enslavement.


Muslim refugees from the North Caucasus are the main subjects in this book. The North Caucasus, a mountainous region between the Black Sea and the Caspian Sea, sustains remarkable ethnic and linguistic diversity. By the mid-nineteenth century, the region was home to several dozen ethnic groups. Much of the indigenous Muslim population in the region spoke Northwest Caucasian and Northeast Caucasian languages, which are unrelated to their neighbors’ languages, such as Russian, Armenian, and Persian (of the Indo-European language family); Turkish and Azerbaijani (of the Turkic language family); and Georgian (of the Kartvelian language family, endemic to the South Caucasus). The tenth-century Arab historian and geographer al-Masʿudi aptly called the Caucasus the “mountain of tongues.”26 The region’s rugged terrain ensured spectacular isolation of many villages, proliferating languages and dialects spoken nowhere else.

In the western Caucasus, specifically on the Black Sea coast and in the fertile Kuban region, lived Circassians, Abazins, and Abkhazians (see map 2), who spoke closely related but mutually unintelligible Northwest Caucasian languages. Circassians, or the Adyghe in their native language, were divided into twelve communities: Abzakh, Bzhedugh, Hatuqwai, Mamkhlegh, Natukhai, Temirgoi, Yegeruqwai, Zhaney, Shapsugh, Ubykh, Besleney, and Kabardian, each with its own dialect.27 The first ten communities, the western Circassians, were the ones who bore the brunt of the expulsions in the early 1860s, and their descendants form the majority of the North Caucasian diaspora in the Middle East today. To the south of western Circassians, along the Black Sea coast leading toward Georgia, lived Abazins and Abkhazians, whose diaspora is collectively known as Abaza in the Middle East. Moving eastward along the northern slopes of the Caucasus Mountains, through mountain ridges and valleys, one could find six large communities: Turkic-speaking Balkars and Karachays; Kabardians, or eastern Circassians; Ossetians, speaking an Iranian (Indo-European) language; and the closely related Ingush and Chechens, speaking Northeast Caucasian languages. In the Northeast Caucasus, where the mountains slowly descend toward the Caspian Sea and the Nogai steppe, lies Dagestan. If Circassian, Chechen, and Ossetian are terms for ethnic groups, Dagestani is a multiethnic regional designation. Dagestan, or the “land of mountains,” is home to over thirty ethnic groups. Many of them, including Avars, Dargins, Lezgins, Laks, Tabasarans, and Tsakhurs, speak Northeast Caucasian languages, and others, like Nogai Tatars and Kumyks, speak Turkic languages. Most North Caucasian languages did not have a written tradition in the nineteenth century. The dominant literary language among Muslim communities in the region had been Arabic, despite not being native to anyone in the Caucasus, and many elites also knew Ottoman Turkish and Russian. In addition to native North Caucasian communities, many Russians, Ukrainians, Armenians, and Ottoman Greeks settled in the region by the mid-nineteenth century.

Map 2. Ethnic groups in the Caucasus, c. 1850.

The vast majority of North Caucasians were Sunni Muslims. Only Abkhazia, Kabarda, and Ossetia had native Orthodox Christian communities, and southern Dagestan was home to Shiʿi Azerbaijanis and Mountain Jews. Almost all North Caucasian refugees to the Ottoman Empire were Sunni Muslims. By no means was the practice of Islam homogeneous in the region. Some Muslim communities traced their origins to the Arab conquest of Derbent in Dagestan in the seventh century. Others embraced Islam gradually, through contacts with the Crimean Khanate, the Ottoman Empire, and Iran. The anti-Russian resistance and the spread of Sufi orders in the eighteenth and nineteenth centuries accelerated the conversion of many communities to Islam, especially in the Northwest Caucasus. A common refrain in literature is that when Circassians arrived in the Ottoman Empire their Muslim identity was still new and less salient than their Circassianness. This view risks projecting ideals of twentieth-century secular intellectuals and reifies a false notion that there is a specific way to be a Muslim, or a good Muslim. The muhajirs’ faith, while certainly only one part of their identity, played an outsized role in their displacement and resettlement because of what the Russians and the Ottomans made of it, and it also helped muhajirs to articulate their loss and their place in history and in their new empire.

I collectively describe these refugees as North Caucasians, while acknowledging that they came from vastly different cultural backgrounds and geographic settings. Many studies of migration focus on a single ethnic group, whereas my approach is purposefully multiethnic, as it helps to examine how the refugee identity was constructed in the nineteenth century and how the Ottomans and the Russians managed population diversity. In the mid-nineteenth century, most North Caucasians would describe themselves as Muslims and then identify specific communities and kins to which they belonged. The tsarist government referred to Muslims in the region as mountaineers (Rus., gortsy), an identity that was Orientalized and romanticized in nineteenth-century Russian literature.28 Ottoman authorities usually used the term Circassians (Tur., Çerkesler; Ar., Sharākisa), an ethnonym for the largest group of refugees, as an umbrella designation for all North Caucasian Muslims. Only in the early twentieth century did diasporic intellectuals in Istanbul popularize the idea of a multiethnic North Caucasian identity, based on their origins in the same region. I further include Muslim refugees from Abkhazia and the Zaqatala region, south of the Caucasus Mountains, who emigrated to the Ottoman Empire alongside their kin communities from the north. In this book, you will meet many western Circassians, Abkhazians, Kabardians, Chechens, and Dagestanis, who, in that order, formed the largest muhajir communities from the Caucasus in the Middle East. Wherever possible, I will identify ethnic and subethnic identities of refugees.

This book builds on the rich scholarship that has situated Muslim refugee migration within the broader Ottoman and Middle Eastern history. Kemal H. Karpat stressed the importance of Muslim migrations in Islamicizing and Turkifying Anatolia.29 Muslim refugees were long presented as perpetrators of violence, especially in the post-1878 Balkan historiographies, or as victims of European nationalism and imperialism in Turkish scholarship.30 The latter approach has occasionally veered toward minimizing, or even denying, the Ottoman genocide of Christians during World War I, as if the suffering of one community justifies or explains the horrors that the state would inflict on another.31 In recent decades, historians have scrutinized the work of the Ottoman Refugee Commission and the impact of global capitalism on Ottoman settlement policies.32 Reşat Kasaba placed nineteenth-century refugees into the longer narrative of Ottoman history, wherein the government managed its “moveable empire” by increasingly suppressing nomad, migrant, and refugee mobility.33 Meanwhile, Dawn Chatty found the arrival of Muslim refugees from the Caucasus and the Balkans as a starting point in the history of mass displacement in the modern Middle East, preceding that of Armenians, Palestinians, Jews, Kurds, Iraqis, and Syrians.34

Figure 1. Circassians in Istanbul.
Photograph by Abdullah Frères, between 1880 and 1900. Library of Congress, Prints and Photographs Division. LC-DIG-ppmsca-03805.

Refugee migration from the North Caucasus, despite being among the largest displacements in Middle Eastern and Russian history, did not attract scholarly interest until the late twentieth century. In the North Caucasian diaspora, research on Ottoman resettlement long faced both institutional and self-imposed censorship.35 For much of Turkey’s republican era, North Caucasians had to be careful not to draw attention to their non-Turkishness. Beginning in the late 1980s, Turkish historians, including many of North Caucasian descent, published works that delved into the geography and demographics of North Caucasian resettlement, based on Ottoman documents.36 To challenge the common perception among the Turkish public that the Circassian leadership was disloyal to Mustafa Kemal’s cause in the 1920s, several works emphasized contributions of refugee communities to the Turkish National Movement.37 Similarly, North Caucasian writers in the Arab world proceeded cautiously, careful to center the commitment of their non-Arab minorities to the Syrian and Jordanian nation-states and to popular pan-Arab causes, such as Palestinian liberation.38 In recent decades, Turkish and Jordanian scholars produced pathbreaking work on the complexity of North Caucasian identities in the contemporary Middle East.39

In the Soviet Union, discussion of tsarist-era displacements of North Caucasians, although not entirely taboo, was not welcomed. It challenged the Soviet narrative of “brotherhood of nations” and risked reopening old wounds in a region brutalized by Stalin’s deportations of entire ethnic groups during World War II. The glasnost of the late 1980s blew the lid off, allowing scholars to mine Russian archives and to share their findings in the 1990s and 2000s.40 The Russian-language monographs, mostly published in the autonomous North Caucasus republics, focused on migrations of specific ethnic groups—Circassians,41 Karachays and Balkars,42 Chechens,43 and Dagestanis44—and included a few overviews of the multiethnic diaspora in the Middle East.45 By and large, the existing historical scholarship in Turkish, Arabic, and Russian prioritizes the early stages of migration: displacement from the Caucasus, arrival in the Ottoman Empire, and settlement by the Ottoman Refugee Commission. This book brings scholarship in Turkish, Russian, and Arabic in conversation with each other, while focusing on migration from the entire North Caucasus region between the 1850s and World War I and on what happened after refugee settlement in the Ottoman Empire.

Empire of Refugees is based on extensive research in over twenty public and private archives, primarily in Turkey, Jordan, Bulgaria, Georgia, the United Kingdom, and Russia, including the autonomous North Caucasus republics of Kabardino-Balkaria, North Ossetia-Alania, and Dagestan, as well as in Romania, Armenia, Azerbaijan, and the United States. It adopts top-down and bottom-up approaches to migration to explore different facets of refugeedom, or how refugees have experienced their displacement.46 This is a history of how the Ottoman and Russian empires managed migration, for which I use a wide variety of state-produced evidence: on the Ottoman side, documentation by the government in Istanbul and provincial (vilayet), subprovincial (sancak), and district (kaza) authorities, and on the Russian side, correspondence between the administration of the Caucasus Viceroy in Tiflis (now Tbilisi, Georgia) and North Caucasus provincial (oblast’) and district (okrug) authorities. To examine the political economy of refugee resettlement, I draw data from different types of Ottoman registers that have recorded the population, land allotments, financial aid, and tax payments of refugee communities, as well as from shariʿa court records. During my fieldwork, I was fortunate to gain access to a full set of Ottoman land records for the Amman region, under the purview of Jordan’s Ministry of the Interior; minute provincial data on refugee resettlement on the Black Sea coast of Bulgaria and Romania, preserved in yet uncatalogued boxes in Bulgaria’s National Library in Sofia; and documents of the Ottoman Refugee Commission, only recently made available at the Ottoman Archives in Istanbul. This book further relies on Russian and British consular records; travel accounts of European and American visitors to the Ottoman Empire; and newspapers in Ottoman Turkish, Russian, and Bulgarian.

This is equally a social history of migration, exploring how refugees challenged the two empires and built anew. Refugees often appear silent in the imperial archive: their individualities sacrificed for collective generalizations and their predicament summarized by officials in how they understood it.47 The resettlement authorities’ efficiency depended on reducing refugees to numbers in their paperwork: how many arrived and therefore how much land to allot and how many oxen to distribute and coffins to make. I use several types of evidence to bring out refugees’ migration experiences. First, I collected dozens of private letters that muhajirs wrote to their families, within the Ottoman domains and across the Ottoman-Russian border, in Ottoman Turkish and Arabic. Some of these letters have been preserved in archives in Nalchik, Vladikavkaz, and Tbilisi. I searched for the rare letters in private hands through communal listservs and by word of mouth and am very grateful to families who shared with me copies from their private collections in Amman and Zarqaʾ in Jordan and Kizilyurt in Dagestan.48 Private letters, which muhajirs did not intend for the authorities to see, reveal how they sought to improve their lot in the Ottoman Empire and preserve ties with their families in the Caucasus.49 Second, I examined petitions that North Caucasians sent to Ottoman and Russian authorities. These documents, while carefully crafted by their writers and curated by the state, testify to refugees’ most urgent needs during their migration and resettlement and their expectations from the two empires. Petitions varied widely: from individual complaints to communal requests on behalf of hundreds of refugees, and, depending on their content, soliciting justice from the sovereign, the government, or local officials. Finally, I interviewed forty descendants of North Caucasian refugees in Jordan and Turkey about how their families remember Ottoman resettlement, and their recollections richly supplement the archival findings about their communities’ histories.


Empire of Refugees is divided into three parts, each representing a specific stage in the refugees’ journey. Part I, “Refugee Migration,” provides a bird’s-eye overview of displacement from the Caucasus and immigration in the Ottoman Empire in the late imperial era.Chapter 1 examines what led to mass expulsions and emigration of Muslims from the North Caucasus to the Ottoman Empire between the 1850s and World War I. It traces the evolution of Russia’s emigration policies for Muslims. Chapter 2 investigates what happened to North Caucasian refugees when they arrived in the Ottoman domains. I demonstrate how, over the years, the Ottoman Refugee Commission built up a refugee regime that guaranteed admission to displaced Muslims. It remained a state agency, and its settlement of muhajirs increasingly reflected the Ottoman government’s political objectives.

Part II, “Refugee Resettlement,” zooms into the Ottoman resettlement of North Caucasians in the Balkans, the Levant, and Anatolia. Within those regions, I focus on provinces that accepted the highest numbers of refugees, respectively the provinces of Danube, Damascus, and Sivas. Chapter 3 explores the resettlement in the northern Balkans, particularly the steppe region of Dobruja, now split between Bulgaria and Romania. Between 1860 and 1878, it was a bustling Circassian and Abkhazian refugee country, but, without sufficient support, refugee economies floundered in these northernmost districts of the Ottoman Empire. During the Russo-Ottoman War of 1877–78, almost all North Caucasians fled the Balkans with the retreating Ottoman army, prohibited to return by the victorious Balkan national governments. Chapter 4 follows some expellees from the Balkans across the Aegean and Eastern Mediterranean, via port cities in Lebanon and Palestine, to arid Transjordan, where Circassians and Chechens founded refugee villages. In what became the Ottomans’ southernmost region of refugee resettlement, muhajirs invested in real estate, traded with bedouin, and attracted Syrian and Palestinian merchants to the booming village of Amman, founded in 1878. Chapter 5 moves to the snowy mountain valley of Uzunyayla in central Anatolia, where refugees from different ethnic groups built their own Little Caucasus between 1860 and World War I. This chapter traces the remarkable history of the Circassian Khutat family, who had been searching for an ideal place to settle in the Ottoman Empire.

Part III, “Diaspora and Return,” examines how North Caucasians navigated their lives between the Ottoman and Russian empires. Chapter 6 explores how muhajirs forged the North Caucasian diaspora. By the early twentieth century, muhajir intellectuals in Istanbul published the first Circassian-language newspaper, established the first Circassian schools, and tried articulating what it meant to be Circassian, North Caucasian, Muslim, and Ottoman. They also engaged in a transimperial debate about Muslim emigration from Russia and whether it should continue. Chapter 7 brings the book full circle by tracing the little-known history of Muslim return migration from the Ottoman Empire to Russia. The tsarist army instituted a ban on the return of North Caucasians, which was never rescinded and, in new iterations, continued into the Soviet and post-Soviet eras. Yet despite Russian and Ottoman opposition to return migration, about 40,000 Chechen, Abkhazian, and other refugees clandestinely returned to the Caucasus.


1. For contemporary accounts of the Circassian refugee crisis of 1863–65, see House of Commons, Papers Respecting the Settlement of Circassian Emigrants in Turkey; Dulaurier, “La Russie dans le Caucase”; Berzhe, “Vyselenie gortsev s Kavkaza.”

2. Presidential State Archives of the Republic of Turkey, Ottoman Archives (T.C. Cumhurbaşkanlığı Devlet Arşivleri Başkanlığı, Osmanlı Arşivi, Istanbul, hereafter cited as BOA) MVL 1016/36 (2 muharrem 1282, 28 May 1865); TŞRBNM 19/13 (24 zilkade 1280, 1 May 1864); TŞRBNM 25/120 (15 safer 1281, 20 July 1864); İ.DH 953/75394 (27 şaban 1302, 11 June 1885).

3. Skran, Refugees in Inter-War Europe.

4. Robson, States of Separation; Watenpaugh, Bread from Stones; Watenpaugh, “League of Nations’ Rescue”; White, “Refugees and the Definition of Syria.”

5. Akin, When the War Came Home; Aksakal, Ottoman Road to War; Campos, Ottoman Brothers; Gingeras, Sorrowful Shores; Makdisi, Culture of Sectarianism; Reynolds, Shattering Empires.

6. Crews, For Prophet and Tsar; Campbell, Muslim Question; Kefeli, Becoming Muslim; Meyer, “Speaking Sharia to the State”; Ross, Tatar Empire; Tuna, Imperial Russia’s Muslims.

7. On transimperial subjects, see Rothman, Brokering Empire; and on transimperial Muslims, see Meyer, Turks across Empires.

8. Besleney, Circassian Diaspora, 31–32; Jaimoukha, Circassians, 101–22; Katav and Duman, “Iraqi Circassians.”

9. On the critique of the Eurocentric view of late imperial migrations, see McKeown, “Global Migration.”

10. Amrith, Crossing the Bay of Bengal; Datta, Fleeting Agencies.

11. For the idea of the internal frontier, see Rogan, Frontiers of the State; for how the Ottomans’ “civilizing attitude” led to refugee settlement on the internal frontier, see Adamiak, “To the Edge of the Desert,” esp. 17–61.

12. Gatrell, Making of the Modern Refugee, 52–72; Ther, Outsiders, 388–98.

13. UN General Assembly, Convention Relating to the Status of Refugees, art. 1(A) (2) (28 July 1951).

14. Can, Spiritual Subjects; Kane, Russian Hajj.

15. Meyer, Turks across Empires; Khalid, Politics of Muslim Cultural Reform.

16. Smiley, From Slaves to Prisoners of War; Taki, Tsar and Sultan, 115–91.

17. Robarts, Migration and Disease; Balistreri, “Persistence of the Periphery.”

18. Marrus, Unwanted, 41; see also Brubaker, “Aftermaths of Empire.”

19. An important unpublished work on Russo-Ottoman migrations is Pinson, “Demographic Warfare.”

20. On works conceptualizing Russo-Ottoman and adjacent borderlands, see Goff and Siegelbaum, Empire and Belonging; Frary and Kozelsky, Russian-Ottoman Borderlands; Bartov and Weitz, Shatterzone of Empires; and on comparative empires, see Barkey and Hagen, After Empire; Brisku, Political Reform.

21. Balloffet, Argentina in the Global Middle East; Fahrenthold, Between the Ottomans and the Entente; Mays, Forging Ties; Pastor, Mexican Mahjar.

22. Klier, Russians, Jews, and the Pogroms, 382; Fishman, Jews and Palestinians, 49.

23. Barker, Most Precious Merchandise; Peirce, Imperial Harem.

24. On Ottoman abolitionism, see Toledano, Ottoman Slave Trade; Erdem, Slavery in the Ottoman Empire, 94–151.

25. Karamürsel, “Transplanted Slavery.”

26. Catford, “Mountain of Tongues.”

27. The exact number of, and boundaries between, Circassian communities are contested both in the Caucasus and in diaspora. For example, the Ubykh are sometimes considered a separate people. For an overview of different Circassian communities, see Richmond, Northwest Caucasus, 20–25.

28. See Grant, Captive and the Gift, 19–42; Layton, Russian Literature and Empire, 89–109.

29. Karpat, Ottoman Population; see also Toumarkine, Populations musulmanes balkaniques.

30. For example, Jagodić, Naseljavanje Kneževine Srbije, 38–43; Muchinov, Migratsionna politika na Osmanskata imperiia, 171–88; İpek, Türk Göçleri; Şimşir, Rumeli’den Türk Göçleri.

31. McCarthy, Death and Exile.

32. Cuthell, “Muhacirin Komisyonu”; Fratantuono, “Migration Administration”; Blumi, Ottoman Refugees.

33. Kasaba, Moveable Empire.

34. Chatty, Displacement and Dispossession.

35. For early works by North Caucasian diasporic writers, see Berkok, Tarihte Kafkasya; Mufti, Heroes and Emperors.

36. Aydemir, Göç; Bice, Kafkasya’dan Anadolu’ya Göçler; Eren, Türkiye’de Göç; Erkan, Kırım ve Kafkasya Göçleri; Habiçoğlu, Kafkasya’dan Anadolu’ya Göçler; Orat, Arslan, and Tanrıverdi, Kafkas Göçleri; Saydam, Kırım ve Kafkas Göçleri.

37. Berzeg, Türkiye Kurtuluş Savaşı’nda Çerkes Göçmenleri; Ünal, Kurtuluş Savaşı’nda Çerkeslerin Rolü.

38. Haghandoqa, Circassians; Ismaʿil, Dalil al-Ansab al-Sharkasiyya. For later works, see Mamsir Batsaj, al-Mawsuʿa al-Tarikhiyya li-l-Umma al-Sharkasiyya; Nashkhu, Tarikh al-Sharkas (al-Adigha) wa-l-Shishan.

39. On North Caucasian diasporic identities, see Besleney, Circassian Diaspora; Doğan, “Circassians in Turkey”; Kaya, Türkiye’de Çerkesler; Shami, “Prehistories of Globalization”; Shami, “Disjuncture in Ethnicity”; Yelbaşı, Circassians of Turkey. For recent works drawing on oral history, see Aksoy, Benim Adım 1864; Aksoy, Beyaz Köleler; Sunata, Hafızam Çerkesçe.

40. The early and groundbreaking work on displacement from the Caucasus is Dzidzariia, Makhadzhirstvo. On archival possibilities and limitations for scholars of the Caucasus and the Russo-Ottoman world, see Bobrovnikov, “Rossiiskie musul’mane posle arkhivnoi revoliutsii”; Meyer, “Guide to the Archives of Eurasia.”

41. Ganich, Cherkesy v Iordanii; Kasumov and Kasumov, Genotsid adygov; Kudaeva, Ognem i zhelezom; Kushkhabiev, Cherkesskaia diaspora; Kushkhabiev, Cherkesy v Sirii; Kushkhabiev, Ocherki istorii zarubezhnoi cherkesskoi diaspory; Polovinkina, Cherkesiia.

42. Borlakova, “Karachaevo-balkarskaia emigratsiia”; Dumanov, Adygskaia i karachaevo-balkarskaia zarubezhnaia diaspora; Kipkeeva, Karachaevo-balkarskaia diaspora.

43. Badaev, Chechenskaia diaspora; Garsaev and Garsaev, Chechenskie mukhadzhiry; Ibragimova, Emigratsiia chechentsev.

44. Abdullaeva, Vnutripoliticheskaia situatsiia v Dagestane; Magomeddadaev, Emigratsiia dagestantsev; Magomedkhanov, Dagestantsy v Turtsii; Murtazaliev, Literatura dagestanskoi diaspory Turtsii.

45. Aliev, “Severokavkazskaia diaspora”; Baderkhan, Severokavkazskaia diaspora.

46. On refugeedom and approaches to refugee history, see Banko, Nowak, and Gatrell, “What Is Refugee History, Now?”; Gatrell, “What’s Wrong with History?”; Tejel and Öztan, “Forced Migration and Refugeedom.”

47. See Gatrell et al., “Reckoning with Refugeedom.”

48. Hamed-Troyansky, “Letters from the Ottoman Empire.”

49. For inspiring use of private letters and family histories, see Stein, Family Papers; Alff, “Business of Property”; Seikaly, “Matter of Time.”