SECTARIANISM IN SYRIA'S CIVIL WAR

SECTARIANISM IN SYRIA’S CIVIL WAR Fabrice Balanche A Geopolitical Study Featuring 70 Original Maps SECTARIANISM IN SYRIA’S CIVIL WAR SECTARIANISM IN SYRIA’S CIVIL WAR FA B R I C E B A L A N C H E a geopolitical study featuring 70 original maps by fabrice balanche & mary kalbach horan he opinions expressed in this book are those of the author and not necessarily those of he Washington Institute for Near East Policy, its Board of Trustees, or its Board of Advisors. All rights reserved. Printed in the United States of America. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the publisher. © 2018 by The Washington Institute for Near East Policy THE WASHINGTON INSTITUTE FOR NEAR EAST POLICY 1111 19TH STREET NW, SUITE 500 WASHINGTON, DC 20036 www.washingtoninstitute.org DESIGN: 1000colors.org CON T EN TS Table of Maps vii Acknowledgments ix Introduction xi I THE ROLE OF SECTARIANISm IN THE SYRIAN CONFLICT 1 | Sectarian Lines Are Drawn and Redrawn II 3 2 | Assad’s Syria, hen and Now 31 3 | Syrian Kurdistan under the PYD 51 4 | Radicalization of the Sunni Arab Rebellion 68 5 | Four Postwar Scenarios 94 THE ORIGINS OF SYRIA’S SECTARIANISm 6 | A Divisive Evolution About the Author 109 Inside Back Cover TABL E OF M AP S  xii 1. Geography of the Syrian Population 2004 2. Sectarian and Ethnic Distribution in Syria 2011 3. Arab Tribes in Syria 4. Geography of the Syrian Uprising, Fall 2011 5. Sectarianism and Frontlines, may 2017 6. military Situation and Sectarianism, April 2013 7. military Situation and Sectarianism, may 2017 8. Homs 2011: Sectarian Distribution 9. he 2011 Damascus Uprising  xiv  10  5  9 10. Baniyas Sectarian Distribution 2011 11. Syrian Cities: Two models 12. military Situation, may 2017 13. Syrian Population Growth 1940‒2010 14. Refugees by Sect, June 2017 15. Syrian Refugees and IDPs, may 2017 16. Resident Syrian Population by Sect 17. Population by Sect in the Regime Area 18. Population by Zone, 2013 and 2017  12  18  20  11  19  21  22  23  23 19. Potential migration from Rebel and IS Areas, may 2017  27 20. Syrian Army Progression in Damascus Area, march 2013‒may 2017  33 21. he Syrian Army Surrounds East Ghouta 22. Syrian Army Damascus Strategy, may 2017 23. Latakia Sectarian Distribution 2011 24. From Alawite Stronghold to Statelet 25. Jabal al-Druze and the Southern Front, march 2017 26. East Aleppo vs. West Aleppo, January 2016 27. Syrian Army Ofensive East Aleppo, Fall 2016 28. Struggle for the manbij‒Azaz Corridor, Winter 2016 29. Ethnic Divisions in Northern Syria 30. PYD Expansion in Northern Syria, may 2017 31. Rojava Population and Future Administrative Organization 2016  54 32. Arabization of Village Names in Northern Syria 33. Change in Irrigated Land, Northeast Syria 2001‒2009 34. Syrian Kurds Break Encirclement, April 2017  xiii  4  6  7  vii   34  35  37  39  45  52  40  46  47  53  55  59  57  60 Partition with Rebellion in Damascus 54. Potential Iranian Routes to the mediterranean, August 2017 55. Russia’s Strategy in Syria 56. Russia’s Strategy in the Levant: Energy and Turkish Siege  103 57. Sectarian Distribution in Levant and Potential Statelets 58. he New hirty Years War 59. Sectarian Distribution in the Levant at End of Ottoman Empire  111 60. Centers and Peripheral Zones in the Levant at End of Ottoman Empire  112 61. Old City of Damascus at End of Ottoman Empire 62. he Alawite State 1920‒1936 63. Administrative Divisions in Syria 64. Baathist Planning 1963–1990: Development of the Periphery 65. Periphery Cleavage Since the 1990s 66. Syrian Population 2010 by Age and Gender 67.  97 Demographic Growth in Syria by District 1970‒1981  126 68.  98 Demographic Growth in Syria by District 1994‒2004  127 69. Illiteracy in Syria by Province 2004 70. Internal migration in Syria 1990 –2010 36. Syrian Rebel Groups by Ideology, march 2016  69 37. Syrian Rebel Powerbrokers by Number and Ideology, march 2016  70 38. Syrian Rebel Powerbrokers by Front, march 2016 40. military Developments in Northwest Syria, march/April 2017  74 41. military Situation in Northern Syria, may 2017 42. Close Allies of Jabhat al-Nusra, June 2016  71  73 39. military Situation in Southern Syria, may 2017  76  75 43. Jabhat al-Nusra and Coalitions, Winter 2015‒Fall 2016 44. Hayat Tahrir al-Sham Territorial Organization in Idlib Area, August 2017  80 45. Hayat Tahrir al-Sham Strongholds in Idlib Area, August 2017  81 46. military Situation in Syria and Iraq, may 2017 47. Energy Production in Syria  86 48. Arab Tribes and military Situation in Northeast Syria, April 2016  88 49. Syria’s Ethnic and Social Structure 50. Unity with Regime in Damascus  98 51. Unity with Rebellion in Damascus 52. Partition with Regime in Damascus  99 53. 35. Rojava as a Future Corridor to the Sea  99  83  77  viii   102  105  114  104  113  120  123  128  124  129  101  122 AC k NOw L Ed gM E N TS M y PrOFOuNd ThANkS to all those who helped and supported me in the realization of this study. First, the publications team at he Washington Institute was instrumental in bringing this project to fruition: Editor George Lopez did a wonderful job organizing the text and inessing the prose while scrupulously checking that all information was properly referenced. Publications director mary Kalbach Horan spent an extraordinary amount of time redesigning and editing the maps, ensuring the coherence of names down to the smallest and most obscure Syrian village. Research assistants marine Barjol, James Bowker, and Jackson Doering helped me translate my original French drafts into readable English. heir fresh eyes forced me to clarify ideas for the many readers who have not spent their professional lives analyzing Syria in minute detail. Furthermore, this study is the result of immense teamwork among the Institute’s numerous scholars and experts. martin Kramer, David Pollock, David Schenker, Andrew Tabler, Ehud Yaari, and Aaron Zelin read and reread various chapters with great attention to detail and endeavored to enrich them. Patrick Clawson coordinated this team, greatly stimulating and encouraging my research over the past two years. I am also grateful to executive director Robert Satlof for his hospitality and kindness. Syria is a diicult subject with violent polemics, so any scholar who tackles it depends on the support of the institution hosting him. I have always found this support at he Washington Institute. Finally, I would like to express my deep gratitude to all the Washington Institute administrators, researchers, assistants, interns, and donors who helped and supported me during my stay. Fabrice Balanche January 2018  ix  I N T r Od u CT ION A S SyrIA’S seemingly interminable war drags on, nagging questions about its initial causes and current dynamics have yet to be fully answered, particularly in comparison to other regional crises. Why did Bashar al-Assad’s regime not fall quickly like Hosni mubarak’s did in Egypt? Why has the Syrian army not fractured like muammar Qadhai’s in Libya? And why has the ighting persisted for so long? In many ways the Syrian conlict has been taken out of the hands of Syrians themselves, becoming a proxy war between regional and international forces that often exploit the country’s divided society for their own beneit. Yet this sectarian fragmentation was not created when the war began in 2011; it came from an inherited Ottoman millet system whose traits were accentuated by the “divide to reign” policies of Bashar’s late father, Haiz. he war has compelled Syrians to cling to their sectarian identities more tightly, whether out of socioeconomic self-interest or simply to survive. Examining these identities is therefore crucial to answering the most fundamental questions about the ongoing upheaval. he main objective of this study is to foster a fuller understanding of the role that sectarianism has played in Syria’s war, and to reassess the notion that the regime’s divisive eforts singlehandedly transformed the secular and democratic revolution of 2011 into the brutal sectarian conlict of today. Since the beginning, sectarianism has been a major factor in the conlict, perhaps even more so than geopolitical drivers and other issues, but many analysts and participants have deliberately marginalized it. When sectarianism is discussed, it is often framed in terms of the regime consolidating its power. Certainly, Assad has used all means necessary to stay in power, including tactics that exacerbate interfaith tensions, but neither he nor his father created these divisions—they only exploited the preexisting vulnerabilities of Syrian society. Analyzing these vulnerabilities can shed light on the war’s current course, including the most salient military, political, demographic, and economic trends. It can also yield lessons about state failure and other ills alicting the wider middle East. Focusing on sectarianism does not mean one should see it lurking behind every door in Syria, but it does need to be granted a more prominent place in discussions about the war and Syria’s future. A SECTArIAN CIVIL wAr he general concept of sectarianism has been abundantly explained by a multitude of authors. I have relied in particular on maxime Rodinson, who used his keen understanding of the interplay between cultural and socioeconomic parameters to describe this phenomenon in the middle East.1 he deinition of sectarianism employed in this study is wide, denoting not only religious communities but also ethnic and tribal groups. In that sense, “sect” describes any social group whose members share a common identity and are able  xi  to create a strong solidarity link. he sect becomes a political player when its leaders decide to compete for authority at the state level, using sectarian solidarity to seize mulk (political power). Indeed, medieval scholar Ibn Khaldun deined any such collective as an asabiyya, a solidarity group created with the express goal of taking political power.2 In the 1980s, michel Seurat applied Khaldun’s theories to Syria in ways that are deeply relevant to the current war.3 Detailed analysis of such theories can be found in Part II of this study, “Origins of Syria’s Sectarianism,” but they inform Part I as well, which focuses on more practical questions about how the war irst erupted, why it has persisted, and how it is reshaping the country’s territory and population. In simplest terms, sectarianism is both a cause and consequence of the initial revolt—and its subsequent failures. Chapter 1 goes immediately to the heart of the matter, assessing the role of sectarianism in the early uprising and the behavior of diferent com- 1. GEOGRAPHY OF THE SYRIAN POPULATION 2004 N al-Malikiyah W Amuda E Kobane Qamishli Ras al-Ain S Tal Abyad Afrin Manbij Azaz HASAKA al-Bab ALEPPO RAQQA IDLIB al-Thawra LATAKIA Maarat al-Numan Jableh TARTUS DEIR AL-ZOUR HAMA Baniyas Salamiya Mayadin Safita HOMS al-Qusayr Abu Kamal Palmyra Nabek Provincial border 2004 POPULATION 4,500,000 DAMASCUS AREA Zabadani Golan Heights Damascus Shahba SUWAYDA DERAA 2,000,000 1,000,000 100,000 10,000 1,000 Jayrud Douma Kiswah TYPE OF LOCALITY Urban Busra DAMASCUS AREA Rural 0 100 km SOURCE: 2004 CENSUS xii   FABrICE BAL ANChE 2. SECTARIAN AND ETHNIC DISTRIBUTION IN SYRIA 2011 N W TURKEY E S HASAKA ALEPPO RAQQA IDLIB LATAKIA DEIR AL-ZOUR HAMA Tartous TARTUS HOMS Mediterranean Sea LEBANON IRAQ DAMASCUS QUNEITRA Golan Heights ISRAEL SUWAYDA DERAA JORDAN 0 100 km Sunni Christian Turkmen Province center Alawi Shia Kurdish Province border Druze Ismaili Bedouin International border S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r   xiii 3. ARAB TRIBES IN SYRIA N Tigris River W TURKEY E Kobane Tal Abyad Afrin Qamishli Ras al-Ain S Adwan Manbij ALEPPO Fadan RAQQA Sabkha Lake Assad Quayar Afadla IDLIB HASAKA Jabbour al-Shadadi Walda Mediterranean Sea Tay Shammar Baggara al-Washeb LATAKIA Haddadin Mawali HAMA Fadan DEIR AL-ZOUR Ougaidat TARTUS HOMS Salamiya Bani Khalid Fawaira LEBANON Sbaa al-Shour Hassana Palmyra Euphrates River Abu Kamal Ruwala al-Swaileh al-Shaga DAMASCUS Ruwala IRAQ al-Abdullah Golan QUNEITRA Heights al-Hassan SUWAYDA ISRAEL DERAA Sedentary area Nomadic area xiv  al-Ghiati al-Masaid al-Sherfat Sbaa JORDAN 0 Tribe Province border Province center Tribal territory International border City 100 km  FABrICE BAL ANChE munities as the civil war escalated. he use of the term “civil war” may shock some readers, since it goes against the conventional reading of the conlict as a ight between a dictatorial regime and the people. he Assad regime is certainly a brutal dictatorship that has slaughtered many innocent civilians, and this study does not aim to elide that harsh reality. At the same time, though, the chapters that follow describe how (and why) a large portion of Syria’s population supports Assad, whether out of self-interest, fear, or some combination of factors. And on the other side, rebel groups have attacked civilian populations that they consider allies of the regime simply because of their sectarian ailiation. Syria is therefore in the midst of a civil war—one that is mostly sectarian, sometimes even in the purely religious sense of the word given the involvement of the Islamic State, al-Qaeda ailiates, and foreign Shia militias.4 UN oicials themselves have described the crisis as a civil war since as early as 2012, so it is hardly inappropriate for others to do the same.5 SyrIA’S dIVISION ALONg EThNOrELIgIOuS LINES Civil wars provoke intense population movements, and Syria is no exception. more than 6 million residents have led the country as refugees, and just as many have become internally displaced persons (IDPs), meaning that more than half the current population of 23 million has been transplanted in one way or another.6 Syria’s sectarian distribution has also been modiied by ethnic cleansing in diferent areas, whether under the aegis of the regime, the rebels, the Islamic State, or the Kurds. he most diverse and populous area is the one controlled by the regime’s army. Although many of this zone’s residents are Alawites—that is, followers of the Alawi branch of Shia Islam, the same sect as the Assad family—the majority are Sunnis, and many minorities have led there as well.7 At various points during the war, Sunni IDPs from Idlib, Aleppo, and Homs have sought refuge in the coastal region, the stronghold of the Alawites. Displaced residents of Damascus city have led to Jabal al-Druze or the Druze/Christian suburb of Jaramana. In contrast, rebel zones are almost exclusively Sunni Arab. he few minorities who lived in these areas have led or been forced to convert to Sunnism. In the northern border zone controlled by the Kurdish-led Syrian Democratic Forces (SDF), an Arab-Kurdish blend remains, but “Kurdiication” eforts continue—belying the pluralistic discourse of the Democratic Union Party (PYD), the dominant Syrian Kurdish faction that shares close ties with Turkey’s Kurdistan Workers Party (PKK).8 Syria is thus divided into three main areas: the multisectarian regime zone, S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r  the Kurdish-controlled SDF zone, and the Sunni Arab rebel zone. Each of these zones has signiicant internal issures as well. he latter is the most fragmented, with several hundred competing rebel groups and some areas still held by the Islamic State. he regime zone has some institutional coherence, but the various community and tribal militias belonging to the National Defense Forces have introduced fragmentation at the local level.9 he NDF still respect Assad’s authority, but they often engage in violent inter-militia rivalries for control of territory and resources. he Kurdish-controlled zone is the most centralized, despite the PYD’s oicial discourse about local democracy and federalism. Kurdish militia forces are everywhere, and the Arab militias within the SDF are mainly treated as tokens or substitutes rather than full partners. his new geography will likely determine the structure of postwar Syria. It also informs the structure of this study—each of these three zones is covered by its own chapter. Between march 2013 and December 2016, the main fronts were stabilized along sectarian lines, recalling Lebanon’s civil war in 1975–1990. In the west, the regime managed to secure a strip of territory stretching from Latakia to Suwayda, with key strategic points in the hands of religious minorities, enabling Assad to keep the local Sunni majority under control. he SDF controlled the Kurdish zone, while rebel groups held sway over a patchwork zone that became almost entirely Sunni Arab. he 2015 Russian intervention moved the lines somewhat, enabling regime forces to leave their minority strongholds and reconquer certain Sunni Arab territories deemed hostile to Assad. he recapture of Aleppo in late 2016 was a particularly important turning point, and the regime has since been expanding its grip in other areas such as the central and southern desert. In the Damascus suburbs formerly held by the rebels, most inhabitants have stayed in place once regime forces retake their neighborhoods, even in Sunni strongholds. he army’s destruction of Daraya and dramatic siege of East Ghouta likely frightened many locals into submission, and the brutal tactics used by certain rebel groups no doubt drove them into the regime’s arms as well. After years of war, most civilians seek security above all else, and the fractured opposition has been largely unable to provide it given the prevalence of radical actors. In short, the regime’s counterinsurgency eforts have been quite effective, at least in territorial terms.10 Assad does not seek to win hearts and minds, but rather to break the opposition’s neck, just as his father did forty years ago when crushing a muslim Brotherhood revolt.11 And while political and territorial cleavages cannot be  xv dismissed when examining the nature of this ight, sectarianism is the key to understanding its roots and nearterm prospects. AN INEVITABLE CONFLICT Ever since I began my research in Syria in 1990, the country’s sectarian divisions have been glaringly evident. Back then, the Alawite community was the backbone of Haiz al-Assad’s classic patronage system: in exchange for political support, they were given material beneits that only increased their sectarian solidarity. he regime knew very well how to agitate the Islamist threat in order to strengthen Alawite loyalty. Yet this privileged relationship increasingly frustrated portions of the majority Sunni population, especially those who sufered deteriorating living conditions. Civil war seemed inevitable in the long term. When Bashar succeeded his father in 2000, the system continued to run on inertia for years, much like Yugoslavia after the death of marshal Tito.12 Eventually, however—after a decade of rapid population growth, socioeconomic deterioration outside the cities, and lapsed structural reforms—Syria’s disaster inally struck. he long-brewing revolt in the rural areas and urban peripheries had been masked for years by ultimately supericial progress such as the opening of private banks and the huge increase in automobile ownership.13 While xvi  city dwellers lived in relative luxury, the countryside sufered, and deep sectarian issures were reactivated by economic frustration. most analysts did not see this growing unrest and spoke of sectarianism as a rapidly disappearing vestige of local medievalism. he same error of analysis persisted at the beginning of the civil war, based on the lawed notion that Alawites could reject Assad and embrace the “popular revolution” without sectarian implications. Only a few analysts took the sectarian parameter into consideration, such as Fouad Ajami.14 more recently, Nikolaos Van Dam denounced the Western diplomatic tendency to downplay sectarian issues and dismiss the fears of religious minorities who faced a revolt largely overtaken by radical Islamists.15 he chapters that follow emphasize the role of sectarianism not to encourage a future partition of Syria based on these divisions, but simply to describe the reality of a protracted civil war. Concerned members of the international community cannot hope to stop the violence unless they open their eyes to its core causes. A NOTE ABOuT METhOdOLOgy Since irst visiting the country in 1990, I have built a geographic information system (GIS) that incorporates around 6,000 Syrian localities, decades worth of data on ethnoreligious identity in these areas, and population data from various Syrian censuses conducted since 1960. he ethnic and religious data has been generated by cross-referencing three types of sources: a hand-drawn ethnographic map of Syria created by French mandate authorities, topographical maps from the 1960s (i.e., before the government Arabized many placenames), and author surveys conducted in numerous Syrian communities. An individual GIS has also been built for each major city, broken down by neighborhood. Taken together, this research shows that the ethnic and religious identity of Syrian villages remained virtually unchanged between the French mandate and 2010. he sectarian composition of urban neighborhoods is also well known because little population mixing occurred in the cities—for the most part, the only residents living in mixed neighborhoods are wealthy people. hese trends, coupled with a GIS database that has been repeatedly reined over the years, allow for the production of highly accurate ethnographic maps, which in turn make it feasible to assess sectarian distribution even amid wartime upheaval. herefore, unless otherwise indicated, all demographic statistics and projections ofered in this study are derived from this database and the author’s related work.  FABrICE BAL ANChE fikraforum/view/from-qamishli-to-qamishlo-a-trip-to-rojavas-new-capital. NOTES 1. See his book Islam, Politique et Croyances (Paris: Fayard, 1993). 2. Ibn Khaldun, Les Prolégomènes (Paris: Librairie Orientaliste Paul Geuthner, 1863). 3. See Michel Seurat, L’Etat de Barbarie (Paris: Seuil, 2012). 4. Fabrice Balanche, “Communautarisme en Syrie: lorsque le mythe devient réalité” [Communitarianism in Syria: when the myth becomes reality], Confluences Méditerranée 89 (Spring 2014), https:// www.cairn.info/revue-confluences-mediterranee-2014-2-page-29.htm. 5. Asked if the situation could be described as a civil war, UN undersecretarygeneral for peacekeeping operations Herve Ladsous replied, “Yes, I think we can say that.” See “Syria Conflict Now a Civil War, UN Peacekeeping Chief Says,” Reuters, June 12, 2012, http:// www.reuters.com/article/us-syria-crisisun-idUSBRE85B11V20120612. 6. For more on these figures, see the “Demographics Overwhelmed by War” section of chapter 1. 7. Souhail Belhadj, La Syrie de Bashar alAsad (Paris: Belin Litterature et Revues, 2013). 8. See Fabrice Balanche, “From Qamishli to Qamishlo: A Trip to Rojava’s New Capital,” Fikra Forum, April 13, 2017, http://www.washingtoninstitute.org/ 9. In 2012, Iran helped Assad create the paramilitary NDF, which the regular army then tasked with protecting regime strongholds at the local level. This strategy has proven to be a mixed blessing for Assad, as discussed in later chapters. 10. Fabrice Balanche, “Insurgency and counterinsurgency in Syria” (in French), Geostrategic Maritime Review 2 (Spring–Summer 2014), http://www.joshualandis.com/blog/insurgency-andcounterinsurgency-in-syria-by-fabricebalanche. 11. Martin van Creveld, The Changing Face of War: Combat from the Marne to Iraq (New York: Presidio, 2008). 12. Fabrice Balanche, La region Alaouite et le pouvoir Syrien [The Alawite area and Syrian power] (Paris: Karthala, 2006). 13. Fabrice Balanche, “The new lion of Damascus will not make an economic tiger with Syria” (in Spanish), Revista Culturas no. 8: Ten years of Bashar alAssad (Sevilla, September-October 2010), http://revistaculturas.org/el-nuevo-leon-de-damasco-no-transformarasiria-en-un-tigre-economico. 14. See his book The Syrian Rebellion (Stanford, CA: Hoover Institution, 2012), particularly the section beginning on page 229. 15. Nikolaos van Dam, Destroying a Nation: The Civil War in Syria (London: I. B. Tauris, 2017). S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r   xvii part i ThE rOLE OF SECTArIANISM IN ThE SyrIAN CONFLICT 1. S E CTA rI A N L I N E S Ar E dr Aw N A Nd rEdr Aw N w hEN the revolt began in 2011, it was driven by many different factors, resulting in numerous interpretations to explain what was happening. Looking at the uprising from a geographical perspective, for example, one could see a clear issure between the center and the periphery. he initial protests in Deraa province were motivated by sociopolitical frustration accumulated against the regime for decades. A drought had undermined local agriculture since 2005, and the state’s lack of interest in this peripheral area ripened the conditions for revolt. Yet this explanation, while valid and analytically useful, does not tell the whole story. Even as their neighbors in Sunni Arab Deraa rose up, residents of neighboring Jabal al-Druze did not join the protest movement, despite their equally degraded economic situation. his lack of revolutionary solidarity was due largely to the ancestral sectarian divide between the Sunnis of the south- ern agricultural region known as the Houran and the Druze of the adjacent mountainous area (for more on these and other ancestral divides, see Part II). To be sure, the geography of the revolt’s irst months is very important to understanding the reasons for the ongoing war, and the irst part of this chapter examines these territorial issures in detail. As will be seen, however, sectarian factors quickly gained ascendancy over territorial, economic, and political factors. he second part of the chapter therefore focuses on the behavior of Syria’s six main sects—Sunni Arabs, Alawites, Druze, Christians, Ismailis, and Kurds—during the uprising. hese sects do not have the same cohesion they did in the nineteenth century, but their ties are strong enough to inluence the behavior of large communities and give some of them access to resources denied to others. At a time of war, Syria’s sectarian identities have been awakened more powerfully than ever, with noncombatants forced to choose a camp as they seek a safe place to ride out the ighting. In this manner, the conlict has greatly disrupted the country’s population distribution. more than half the people have left their homes—sectarian iefdoms have become safe havens for those caught in mixed areas afected by violence, and ethnic cleansing has been carried out in some locales. Accordingly, the third part of this chapter seeks to establish a new geography of the Syrian population by charting this ongoing demographic upheaval. In April 2013, Syria’s battle map largely aligns with the map of its sectarian distribution. As described in the Introduction, Sunni rebel and jihadist forces control much of the eastern and central countryside, which by this point is almost entirely populated by Sunni Arabs. Kurdish forces control most of the Kurdish-majority northern border zone. And the regime controls the Alawite coast, Jabal al-Druze, and similar territories where minorities passively or actively support Assad. Damascus and the other major cities (Aleppo, Homs, Hama, and Latakia) are either wholly or partially held by  3  his forces, with sectarian neighborhood boundaries often deining the frontlines. Of course, a host of crucial nonsectarian factors complicates this picture (e.g., the class-based, urban/rural divide that characterized many wartime developments in Aleppo), so they are examined as well. TErrITOrIES OF ThE rEVOLT In 1979, scholar Elisabeth Picard noted that while Syria was experiencing sectarian violence as a result of an uprising by Sunni Islamists (i.e., the muslim Brotherhood), its situation was not comparable to that of Lebanon, which was in the throes of a full-blown sectarian civil war.1 As she noted, Syrians did not seem to express their sectarianism with the same vigor as the Lebanese at the time, and Haiz al-Assad was able to crush the unrest before it escalated into a nationwide movement. Today, however, the question of Syrian sectarianism must be reconsidered—analysis of the revolt’s geography Turkey KURDISH POPULATION QAMISHLI KURDISH POPULATION KURDISH POPULATION ALEPPO Sunni enclaves in Alawite territory IDLIB LATAKIA HASAKA RAQQA ARAB TRIBES Jabal al-Turkmen Jabal al-Akrad Periphery of Alawite territory 4. GEOGRAPHY OF THE SYRIAN UPRISING Fall 2011 TYPE OF TERRITORY DEIR AL-ZOUR HAMA Hostile Abandoned periphery BANIAS Support RASTAN HOULE TARTUS Idle PALMYRA Tal Kalakh HOMS AL-QUSAYR Free Syrian Army base Iraq ZABADANI Mediterranean Fighting Lebanon Abandoned urban periphery City Primary road Secondary road SUWAYDA Abandoned periphery Israel 4  Foreign support to fighting DAMASCUS Desert DERAA Jordan  FABrICE BAL ANChE 5. SECTARIANISM AND FRONTLINES, May 2017 TURKEY PYD PYD HASAKA PYD Mediterranean Sea ALEPPO SYRIAN ARMY RAQQA IDLIB REBELS LATAKIA IRAQ DEIR AL-ZOUR IS HAMA Tartous TARTUS HOMS SYRIAN ARMY IS LEBANON DAMASCUS QUNEITRA Golan Heights REBELS SUWAYDA ISRAEL JORDAN DERAA 0 JORDANIE 100 km Sunni Christian Turkmen Province center Alawite Shia Kurdish Province border Druze Ismaili Bedouin International border Frontline S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r  suggests that the country has major (perhaps irreconcilable) sectarian issures, despite oicial denials by the regime and opposition alike. he demonstrations that swept Latakia, Baniyas, and Homs in 2011 occurred only in the Sunni districts of those cities. In the suburbs of Damascus, most Druze, Christian, Ismaili, and Alawite districts did not join the protest movement, but outlying Sunni towns did. he demonstrations in Jdeidat Artouz, a Druze/Christian area, and Bab Touma, the capital’s main Christian quarter, were largely undertaken by Sunni residents. In response, the older Druze and Christian inhabitants mobilized against what they considered aggression by newer inhabitants who practiced a diferent religion. meanwhile, the Sunni inhabitants of Douma, Deraa, and other rebellious districts around Damascus rose to defend their territory against regime aggression. For them, the state could no longer be considered an extension of the Syrian people, if it ever had been during the Baath/Assad era; instead, they saw it as a hostage of the Alawite minority, and therefore a foreign element. he Alawite Fallback Syria’s Alawites occupy a more or less continuous territory between the mediterranean Sea and the Orontes River, including the Jabal al-Ansariyya range and surrounding plains. Additional Alawite settlements are scattered east of the Oron 5 6  6. MILITARY SITUATION AND SECTARIANISM, April 2013 N TURKEY W Tigris River Euphrates River Qamishli E Ras al-Ain Kobane S Tal Abyad Azaz Afrin Manbij al-Bab IDLIB ALEPPO HASAKA Lake Assad RAQQA Jisr al-Shughour Ariha LATAKIA Maarat al-Numan DEIR AL-ZOUR HAMA Salamiya TARTUS Palmyra Mediterranean Sea HOMS BA NO N Abu Kamal LE tes, created in the nineteenth century when the large landowners of Homs and Hama reconquered farmland on the edge of the steppe using Alawite sharecroppers. Prior to the creation of modern Syria, local Alawites were ostracized by Sunni and Ottoman authorities, who did not consider them true muslims. Alawites were forbidden from even entering cities—a harsh edict in light of the fact that minority Christians were permitted to live there. It was only with the French mandate (1920–1945) that Alawites began to settle in the cities. And when Alawite military oicers such as Haiz alAssad asserted themselves at the outset of the Baath era in the 1960s, their coreligionists took numerous positions of authority in these cities. Today, Latakia is a majority Alawite city, as are Baniyas, Jableh, and Tartus. Yet some key locales in the western Alawite heartland are dominated by other sects. In Talkalakh, the majority of the population remains Sunni, while in Homs, Alawites comprise only a quarter of the population. hey are practically absent from Hama, Jisr al-Shughour, and maarat al-Numan, whose Sunni residents are openly hostile toward the minority. Such Sunni animosity has grown over the past few decades because of the preferential treatment Alawites receive from the state. his favoritism is a crucial part of the regime’s security system,2 an arrangement whose contours and inherent vulnerabilities have been put into DAMASCUS QUNEITRA IRAQ Golan Heights ISRAEL DERAA SUWAYDA INHABITED AREAS OF CONTROL 100 km 0 JORDAN MINORITIES Syrian Army Shia minorities Main road Province border PYD (Kurdish) Christian Secondary road International border Rebels Kurdish Desert Province center Disputed Turkmen City  FABrICE BAL ANChE 7. MILITARY SITUATION AND SECTARIANISM, May 2017 N W TURKEY Tigris River Qamishli E Kobane S Azaz Afrin Ras al-Ain Tal Abyad Manbij HASAKA al-Bab ALEPPO Lake Assad IDLIB Tabqa Jisr al-Shughour al-Shadadi RAQQA Maarat al-Numan LATAKIA DEIR AL-ZOUR HAMA Salamiya Rastan TARTUS Mediterranean Sea HOMS Mayadin Shaer gas field Euphrates River al-Qusayr Abu Kamal Palmyra LEBANON Zabadani DAMASCUS IRAQ Douma QUNEITRA Golan Heights ISRAEL DERAA SUWAYDA INHABITED AREAS OF CONTROL Syrian Army UNINHABITED AREAS OF CONTROL 100 km 0 JORDAN MINORITIES Syrian Army Shia minorities SDF (Kurdish) SDF (Kurdish) Christian International border Secondary road Other rebels Other rebels Kurdish Province center Desert Islamic State Islamic State Turkmen Locality S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r  Province border Main road sharp relief during the current war. Alawite military oicers are the heart of this system, which is based on direct loyalty to the Assad family. Over the years, the growing density of Alawite cadres in the regime ensured that every Alawite family had a potentially exploitable link within the state, assuming they demonstrated allegiance to Assad. Although individuals from other sects could (and did) access similar privileges, the elder Assad’s strategy was to make the Alawite community a loyal monolith while keeping Syria’s Sunni majority divided.3 And like the Capetian dynasty in France, he sought to designate his son as his successor in advance to avoid any dispute after his death. Interestingly, Bashar al-Assad paid less attention to the Alawite community upon his ascension—he married a Sunni and neglected to visit Qardaha, his ancestral home and stronghold, preferring to spend summer weekends in Palmyra and Latakia with members of the Sunni bourgeoisie. Yet the war has forced him to reconsider the value of having a faithful community at his back. It has also convinced him that, as an Alawite, he will never be able to obtain the Sunni majority’s acceptance without the use of force. he sectarianism that structures Syria’s society and regime is thus a doubleedged sword, especially now that the smokescreen of Baath idealism has long dissipated and the social pact has been broken in rural areas by the welfare state’s  7 failure. Leading up to the war, economic diiculties increased for the general population as the state withdrew from many areas, but Alawites in western Syria tended to retain their privileges thanks to their better integration in the power networks that dominated the security branch. hey had more opportunities to join the public sector, obtain real estate, secure agricultural loans, and take advantage of exemptions (e.g., for digging illegal water wells on al-Ghab plain). Such favoritism exasperated Sunnis, and many of those who rose up in 2011 simply sought to rebalance the scales rather than overturn the entire system. In Baniyas, for instance, one of the demonstrators’ main demands was for the regime to create 3,000 jobs at the local oil reinery in order to curb massive unemployment among Sunni youths. And in Latakia, prominent Sunnis summoned to the governor’s oice after riots expressed their bitterness that all positions of authority in the local administration were occupied by Alawites.4 he repression of protests by Alawite militias (shabbiha) and intelligence services only increased Sunni rejection of the ruling community. he may 15–19, 2011, crackdown on Talkalakh was particularly provocative—several thousand of the town’s Sunni residents were forced to take refuge in Lebanon, which Sunnis interpreted as a deliberate act of ethnic cleansing meant to beneit Alawites. By that point, even if one rejected the no8  tion that the state had become hostage to the Alawites, the sect was clearly the irst beneiciary of regime largesse.5 From the Alawite perspective, however, this largesse is a matter of existential necessity, not (or not merely) material luxury. Well before the current conlict, Alawites regarded their home enclave in western Syria as a safe haven in the event of mass uprisings, civil war, or other displacements. In 1939, Alawites from the Ottoman province of Alexandretta immigrated to the Aleppo area after the former was ceded to Turkey. In subsequent decades, Alawites who resided elsewhere in Syria often used their wealth or favorable positions to make real estate investments in the coastal region, partly out of attachment to their homeland but mainly as an investment in their safety. heir region became the main terminus of internal emigration after the Baath took power in 1963, and the positive migration balance continued in the 1994 and 2004 censuses. Some of this movement involved Alawite retirees returning to their birthplace, but many of their descendants moved west as well when they were unable to integrate into Damascus, Aleppo, and other cities with large Sunni populations. During the muslim Brotherhood insurgency of 1979–1982, Alawites were systematically targeted in such cities, and large numbers led toward the coast. Additionally, thousands of Alawite villagers from Idlib province were forced to seek refuge in Latakia because Islamists were threatening their isolated hamlets in the middle of the Sunni countryside. In light of these precedents, it came as little surprise when the Assad regime sent arms to Alawite villagers around Jisr al-Shughour in June 2011 after cracking down on protests in that Sunni-majority town. Such actions reinforced the sectarian divides and territorial discontinuity between Alawites and Sunnis—a phenomenon that can be seen not only on the national level, but within individual cities.6 Divided and Encircled Cities Sectarian distribution within Syria’s major cities tends to follow one of two models. he irst is the divided city model, exempliied by Latakia, Baniyas, Homs, and other urban areas where two main communities share most of the territory. Typically, one of these communities is composed of original residents, while the second community comprises rural denizens who settled in or near the city later. In the cities mentioned above, the mix is Sunni/Alawite, but in other locales such as Aleppo and Raqqa, one inds an Arab/Kurdish mix as well. In both types, potentially tense urban cleavages are tempered by the presence of Christian districts and mixed middle-class neighborhoods that are more amenable to coexistence, perhaps because they are linked to the state and the bourgeoisie. Each city also has a central souk where all faiths can mix, though individual communities maintain their own (sometimes conlicting) commercial interests as well. he second model is that of the encircled or garrisoned city, such as Damascus and Hama. In these cases, Alawites and other urban minorities did not have a suficient demographic reservoir to counterbalance the inlux of rural Sunnis, so the regime felt compelled to install more of its clients there, often in the form of large military garrisons and their families. Some cities are a mix of the two archetypes: until recently, for instance, Aleppo was both divided and encircled, with sharply deined Arab and Kurdish districts as well as loyalist populations that were deliberately settled in strategically vital neighborhoods. he resulting urban agglomerations looked more like groups of rival territories than cities, so their dramatic fragmentation during and after 2011 was hardly surprising. Damascus is the typical model of the encircled city. Haiz al-Assad endeavored to build good relations with the capital’s bourgeoisie in order to bolster the regime, and his son extended this policy. On the security front, the elder Assad surrounded the capital with Alawite military camps that could seal it of from the rest of Syria in case of an uprising. In contrast, Homs has long been internally divided into neighborhoods with strong sectarian identities, and the bloody confrontations  FABrICE BAL ANChE To Aleppo 8. HOMS 2011: Sectarian Distribution To Salamiya Sunni Alawite Christian Industrial area Mixed area al-Waar Cathedral Deir Baalba al-Qousour Church al-Bayda The Great Mosque al-Abbasiyah Mosque 10 Orontes River Homs in 1936 5 3 2 al-Muhajireen Street 4 Zahra 1 Fight al-Arman 6 To Tartus 8 al-Fardos 7 al-Ghouta 2 Governorate 3 Khalid ibn al-Walid Mosque 4 al-Hamidiyah 5 Karm al-Sham 6 Bab Drib 7 Bab al-Sebaa 8 al-Mahatta 9 Station 9 Karm al-Zeitoun Nazha Bab al-Amr 1 Karm al-Loz University Akermeh To Palmyra 10 Industrial area 0 1000 m To Damascus S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r  al-Khalidiya Sources: Topographic maps 1942, Tourist map 1996, and personal inquiry seen during the initial phase of the uprising unfolded right on the boundaries between the Alawite and Sunni districts (for more on the sectarian situations in Damascus and Homs, see chapter 2). Fragmentation Prevented an Even Wider Revolt he severe insecurity brought on by the war has only highlighted Syria’s preexisting fragmentation, manifested on the local level by loyalty to neighborhoods and on the regional level by sharply deined sectarian refuges (e.g., Alawites led to their coastal enclave; the Sunnis of the Houran and the Druze of Jabal al-Druze drew battle lines in the south). Indeed, the Syria built by the Assad regime is not so much a nation-state as a territorial state, even a multi-territorial state—a characterization that applies to the relatively quiet regime zone as well. To be sure, the central administration had authority over the whole country prior to the war, unlike in Lebanon, where certain territories traditionally operated outside state control. Yet the regime is in perpetual negotiation with local societies, and this process bypasses the institutional channel of elected assemblies, since they are not representative.7 Instead, the president and his appointed provincial governors bargain with representatives of tribes, communities, or economic groups through a patron-client relationship. When the current uprising began, Assad replaced governors who  9 To Aleppo To Aleppo Maaraba 9. THE 2011 DAMASCUS UPRISING Tel Machaekh Barada River Douma Centers of power BARZAH Harasta Territory loyal to regime NEW DAMASCUS RUKN AL-DIN Qudsaya AL-QABUN Famous square To Beirut Arbin Jabal Qasioun JOBAR AL-MUHAJRIN DUMMAR Urban area in 2011 Zamalka Abu Rumaneh Presidential Palace al-Malikiyah Jabal Mezzeh al-Mazraa al-Qassa Umayyad Square Shaqba Old city Abassyn Square Cultivated area Kafr Batna OLD CITY Noncultivated area University EAST GHOUTA Douwayla MEZZEH To Beirut Uprising Locality KAFR SOUSEH AL-SHAGHOUR Damascus border Mleiha AL-MIDAN Mezzeh military camp Jaramana Principal road Presidential airport QABOUN District YARMOUK Babila N Moadamiya WEST GHOUTA W Daraya E S Set Zaynab al-Hajar al-Aswad Sahnaya 10  0 To international airport 5 km To Deraa  FABrICE BAL ANChE To Latakia N W 10. BANIYAS SECTARIAN DISTRIBUTION 2011 E S 5 HOSPITAL To al-Qadmus Sunni Alawi Mixed area M e d i te r rane an S e a Mosque Baniyas, 1936 CULTURAL CENTER Street Fight POST OFFICE TOWN HALL Sources: Topographic map 1942, Tourist map 1996, and personal inquiry To Marqab 0 400 m To Tartus S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r  failed to reestablish dialogue with local notables. In Hama, for example, he sacked the governor in June 2011 and installed a native of the province, at odds with the traditional policy of appointing non-natives in order to avoid collusion. he uprising apparently forced Assad to prioritize “territorial capital” in an attempt to restore order—that is, holding onto the territory of a restive province by appointing a native with a large local network was more urgent than preserving the political status quo by appointing another unpopular non-native. As will be discussed at greater length in chapter 2, however, whenever the revolt stalled in certain areas, it often had as much to do with old cross-sectarian differences boiling over as with Assad making strategic adjustments, at least during the war’s early years. For instance, after the uprising began in Deraa, it spread rapidly throughout the Sunni Houran region but then stopped short at the largely Druze/Christian province of Suwayda. Within that province, the Jabal al-Druze region had previously stood against central authorities on several occasions: in 1925, the Great Syrian Revolt against French occupation began in these mountains under the direction of Druze leader Sultan al-Atrash; in 1967, local Druze rose against the Baath regime, and their province nearly sufered the same destructive fate Hama would two decades later; and in 2001, Druze in the area began protesting the Assad  11 11. SYRIAN CITIES: TWO MODELS LATAKIA A Divided City DAMASCUS A Controlled City NEIGHBORHOODS Sunni Kurdish Alawite Druze Christian Urban mixed Political center Military base Communication axis Market (central/secondary) Suburban mixed regime following a neighbor’s quarrel with Sunni Bedouin, spurring severe suppressive measures from Damascus. Despite this rebellious background, however, Jabal al-Druze saw only a few small anti-regime demonstrations after march 2011, and most residents did not join the Sunni opposition movement sweeping in from Deraa. In fact, some of them formed a pro-Assad militia as 12  the war escalated. meanwhile, extremist elements among the Sunni rebels ostracized the few Druze who tried to join them, kidnapped Druze villagers, assassinated certain Druze notables, and engaged in full-scale military operations against Jabal al-Druze in fall 2012 and August 2014. Similarly, Aleppo’s fragmentation delayed the protest movement’s eforts to gain a foothold there and impeded rebel eforts to capture northern Syria’s main metropolis. he provincial capital was not afected by the rebellion until July 2012, in part because the Aleppo bourgeoisie— who had been punished by the regime for years after supporting the past muslim Brotherhood revolt—put a damper on local demonstrations. he delay also highlighted how much the urban popula- tion distrusted the countryside. When the opposition called for Aleppo to arise on June 30, 2011, asking all of the province’s inhabitants to converge on the city, it made a serious mistake: it underestimated the strength of the urban/rural cleavage. Since the French mandate, Aleppo’s identity had been built not only in opposition to Damascus, but also against the countryside and outlying towns.  FABrICE BAL ANChE Aleppo’s large and well-deined Kurdish districts also help explain why much of the city remained passive—indeed, this ethnic factor likely prevented a general revolt in most of the northern provinces. he few protest movements that erupted in Kurdish-majority cities such as Qamishli and Ras al-Ain were mainly intended to obtain advantages for their community alone, such as the naturalization of stateless Kurds known as bedoon. many Kurds feared the specter of severe regime repression if they joined the main revolt—an understandable concern given their bitter memories of a previous regime crackdown on Kurds in Hasaka province. In 2004, Assad responded to local Kurdish riots by deploying elite Republican Guard forces to the area, resulting in multiple deaths and a mass Kurdish exodus to northern Iraq. He also bought the help of local Arab tribal leaders in this endeavor. he violence did not inspire wider Syrian sympathy for the Kurds, likely due to an Arab nationalist perspective that cast the Hasaka uprising as irredentism. he Assad regime was no doubt eager to take advantage of such sentiment, posing as a guarantor of national unity in the face of a “foreign Kurdish conspiracy.” In short, Syria’s sectarian divisions helped fracture or neutralize the protest movement in various parts of the country. By spring 2012, the opposition had become militarized in the face of unfaltering regime repression, and the nation- al death toll rose from a dozen per day8 to more than a hundred.9 he army soon withdrew from most Kurdish territories and the hostile Sunni Arab countryside, concentrating its eforts on large cities. And while outside actors such as Iran, Hezbollah, and Russia would eventually play a massive role in how the rest of the war played out, Syria’s longstanding sectarian issures are greatly responsible for the shape of its current battle lines, and its uncertain prospects for future reuniication or stability. MINOrITy SOLIdArITy, SuNNI FISSurES As of 2010, Syria’s population was roughly 65% Sunni Arab, 15% Kurdish, 10% Alawite, 5% Christian, 3% Druze, 1% Ismaili, and 1% Twelver Shia.10 Given this huge demographic advantage, one might argue that the Sunni Arabs should have won the war quickly. Yet the Assad regime is still in power and on the ofensive, and while it holds only half of the country as of this writing, this territory remains the most populous zone (10 million inhabitants out of 16 million as of 201511). Obviously, Russian military assistance and the inlux of 40,000–50,000 foreign Shia militiamen were major elements in the regime’s turnaround. Yet the Sunni Arab rebels received substantial outside help as well, including air support, heavy weapons, and legions of foreign ighters, so their failure to make S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r  further headway is at least partly rooted in nonmilitary factors. A closer look at the loyalties, internal cohesion, and often-competing interests of Syria’s various communities is therefore warranted. Divided Sunni Arabs Syria’s Sunnis do not constitute a single, uniied community. In total they represent 80% of the population, but given the Arab/Kurdish ethnic divide, the Sunni Arab element is closer to 65%.12 Sunni Arab communities are themselves divided between practicing believers, atheists, secularists, followers of Suism, quietists, and radicals who wish to impose sharia law and strict (mis)interpretations of Islam. hey are further split by regional and tribal/clan loyalties. In political terms, while the Sunni Arab bourgeoisie maintains close ties with the clergy, inancing pious foundations and charitable associations, it does not represent a uniied alternative to the Assad regime, whether because of internal divisions, regional rivalries, or a desire to stay out of politics altogether. For example, the historical rivalry between the business classes of Aleppo and Damascus—long fostered by Assad and his father before him—keeps these inluential Sunni Arabs apart, with Aleppines accusing Damascenes of exploiting their proximity to the halls of power in order to unfairly monopolize markets. meanwhile, the entrepreneurs who started small- and medium-size business- es under Syria’s initah (openness) policy have been more likely to support opposition movements, at least compared with the larger establishment irms they compete against. In the 1970s and 1980s, such businessmen formed the social base of the Sunni muslim Brotherhood. Yet years of economic liberalization have considerably enriched them, pushing some away from Islamist sentiment. For years, many wealthy Sunnis have feared their working-class coreligionists on the peripheries of the cities, where the strictly conservative Salai strain of Sunni Islam tends to dominate. he feeling of religious belonging is generally stronger among the working classes and others who feel left out of the regime’s wealth redistribution, civil service appointments, and military promotions, which often favor Alawites. hese sectarian feelings are more apparent in areas where Sunnis and Alawites live in close proximity. For example, the spring 2011 demonstrations in Deraa—an almost entirely Sunni province—were not overtly sectarian,13 but working-class Sunnis in the mixed city of Baniyas hit the streets with clearly deined sectarian demands.14 To be sure, a great many Sunnis likely feel strong resentment toward a regime they view as Alawite and therefore illegitimate—the question is, how many of them are willing to risk their interests or lives to act on these grievances? When the current uprising broke out, the brutal repression of the 1982 muslim Broth 13 erhood revolt was still a source of fear as much as anger. moreover, the “Alawite” regime is not entirely Alawite—many Sunni oicials are represented within the state apparatus, and powerful Sunni economic elites likewise had little reason to revolt given the beneits they received from Assad’s liberalization policies (apart from a handful of notables who ran afoul of the Assad family’s personal interests). And for many ordinary Sunni civilians, the growth of radical Islam within the rebellion spurred them to either support the regime or take a careful wait-and-see approach. Given these sentiments, the Sunni bourgeoisie quickly distanced themselves from an opposition movement they deemed dangerous for the business climate—a fear that was conirmed in summer 2012 when Sunni rebels plundered Aleppo’s industrial zone.15 he Alawite Monolith Not all of the Alawite community is behind Assad. Some Alawite notables joined the opposition, such as Aref Dalila, an academic and political activist who was arrested in 2001 and imprisoned for years on charges of sedition. And Syria’s main marxist opposition faction, the Party of Communist Action, recruited heavily among Alawite youths as early as the 1980s. Yet unlike in the past, Alawite opposition igures have not joined armed rebel factions during the current war. here are 14  no Alawite brigades in the Free Syrian Army (FSA) or the Islamist militias— Alawite conscripts who deserted the army have instead taken refuge abroad to escape the ighting. While some Alawites no doubt hate the dictatorial regime as much as other groups do, they also worry that their loved ones would become victims of rebel retaliation if Assad falls. hey are convinced that they have everything to lose from regime change, fearing they might sufer the same fate as Saddam Hussein’s cadres in Iraq. As such, defending their economic and political interests has become secondary to what they perceive as an existential struggle against an Islamist, anti-Alawite rebellion. hese fears are not the fruit of regime manipulation or propaganda, at least not entirely—they are rooted in the Alawite community’s long history of persecution. After years of being isolated in the mountains or exploited by large estate holders on the surrounding plains,16 the Alawites took their revenge with the arrival of the Baath regime. hey gained access to the highest military and civilian roles, and the community as a whole beneited from Baath-driven development policies that favored the western Alawite region. Even so, they have always lived in fear of Sunni uprisings given the Alawite displacements that occurred during the 1979 muslim Brotherhood revolt,17 and the same scenario took place on a national level when the current conlict was irst unfolding. As early as fall 2011, for example, Sunni insurgents in Homs began daily bombing of the city’s Alawi neighborhoods, with the aim of expelling Alawites from a city where many regarded them as intruders.18 Some observers speculated that the regime deliberately let the situation in Homs deteriorate so that sectarianism would fracture the local revolutionary movement. And in areas where the regime resorted to direct, violent repression, paciist demonstrators were quickly overtaken by militarized opposition as people picked up weapons to defend themselves. hese armed elements then organized by sect; as in many other conlicts worldwide, violence created stark dilemmas in which people had to make tough choices with group consequences. his is hardly surprising in an environment where both the regime and the opposition instrumentalized the country’s latent sectarianism. While one can debate which side was more responsible for this atmosphere of incitement, the result is that Syria was caught in a spiral of sectarian violence. he behavior of other minorities further illustrates this problem. he Druze: From Wait-and-See to Self-Defense During the irst year of the uprising, the Druze remained largely neutral. Despite having front-row seats to the revolt and repression in neighboring Deraa province, and despite their previously discussed history of rebelliousness, most of the Druze in Suwayda province and mt. Hermon remained passive. he situation was so calm that Syrian army troops responsible for repressing Sunni unrest in Deraa often took breaks in Druze villages.19 A few demonstrations did break out in Suwayda in spring 2011, organized by local lawyers, but their demands centered on respect for human rights and the release of certain Druze prisoners, the latter of which was achieved. Local Druze oicials and sheikhs refused to participate in the wider revolt, though they also refrained from making statements supporting the regime. he Druze relationship with the regime has a complicated history. hey were excluded from power after the 1966 coup, when many Druze military oicers and other igures who had denounced the Baath Party’s communist drift were forced into exile, mainly in Iraq.20 Since then, the regime has built numerous military camps in Suwayda province, largely stafed by soldiers from the Alawite coast. Oicially, their function was to monitor the Golan Heights, but in reality they were intended to monitor the Druze population. During the 1970s and 1980s, the Jabal al-Druze region was deprived of public investment, which caused a strong exodus toward Damascus and abroad.21 Yet remittances from diaspora Druze and investments by those who moved to the capital helped bring Jabal al-Druze out of its isolation.  FABrICE BAL ANChE In the 1990s, the Druze returned to the regime’s good graces, which contributed to their economic advancement. A member of the Atrash family was appointed to head the powerful ministry of Local Government, which led to a windfall of public investment for the Druze, including enhanced services and the creation of thousands of administrative jobs. Even so, Druze were still much less prevalent in the public sector than Alawites, and relatively few of them served in the army or intelligence services. As the civil war escalated in 2011– 2012, Druze began to organize themselves into defense groups in the suburbs of Damascus and Jabal al-Druze because their neighborhoods were being attacked by the FSA and other rebel forces. he Druze/Christian town of Jaramana sufered several bloody attacks in 2012.22 From then on, Druze areas of southern Syria were armed and clearly on the regime side. Lebanese Druze leader Walid Jumblatt repeatedly asked his Syrian counterparts to support the Sunni opposition, but they did not heed his call because they were convinced that the Assad regime was their best protection. One exception was the Druze community in Jabal alSummaq north of Idlib, which ofered vocal (if not material) support to the rebels.23 Yet this was simply a matter of self-preservation, as they are located deep within opposition-controlled ter- ritory. Even after giving such pledges of support, they were still forced to convert to Sunnism in 2015 by the local al-Qaeda ailiate that dominates most rebel territory. he Ismailis: Revolutionary Disillusionment Syrian Ismailis are mainly concentrated in the Salamiya area east of Hama, where they organized anti-regime demonstrations in the irst months of the revolt. he regime responded in force, but not as brutally as it did in Homs or Hama, likely because it did not want this small but strategically located minority to join the Sunni opposition. most Ismailis originated in al-Qadmus, masyaf, and other portions of the coastal mountains, but Alawites drove them west toward the plains during the nineteenth century. he center of the community then became Salamiya, where the Ottomans used them to defend the cultivated lands around Homs and Hama against Bedouins. he population is poor but educated, largely thanks to missions inanced by the Agha Khan (their spiritual leader and chief of a well-endowed foundation). Politically speaking, many Syrian Ismailis have been seduced by marxist ideology over the years, and the Syrian Communist Party–Political Bureau recruited heavily among them. In the 1970s and 1980s, Salamiya became a iefdom of the left-wing opposition, S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r  and the Assad regime monitored the area closely. Yet the 2011 uprising did not gain much traction there, at irst because the Ismailis wanted to conine themselves to peaceful political opposition, and later because they feared the rise of Islamists among the rebels. he many human-rights activists in the Ismaili community have since become disillusioned with the revolution and retreated into sectarian solidarity.24 he Shia: First Victims of the Jihadists As a sect that maintains a low proile due to their small numbers in Syria, Twelver Shia have always stood outside the halls of power, sharing few common interests with those who practice the Alawite ofshoot of their faith. Yet the situation changed after Sunni jihadist groups began attacking small Shia communities between Raqqa and Deir al-Zour in 2012. In one high-proile incident, the Shia population of Hatlah village on the Euphrates River was massacred by jihadists on June 13, 2013.25 hese communities resulted from Iranian proselytism in Syria since the 1990s, when Tehran conferred its protection on all traces of Shia culture there. his included the Sayyeda Zainab shrine on the outskirts of Damascus and two important locations in Raqqa: Uwais al-Qarni mosque and the Ammar ibn Yasir shrine.26 Using these locales as a base, Iranian mis- sionaries sought to convert the surrounding Sunni populations. heir eforts reached a few thousand people, eventually spurring Sunni religious authorities to ask for Assad’s intercession against Shia proselytism in 2008.27 To avoid angering these authorities, the president oicially halted the Iranian campaign. Early in the current war, attacks on Syrian Shia were part of the reason why Lebanese Hezbollah intervened on Assad’s side even before their Iranian patron asked them to. In autumn 2011, Sunni rebels raided Shia villages around al-Qusayr, stirring the deep clan links that unite Shia on both sides of the Syria-Lebanon border. Hezbollah could not resist the pressure to act, which emanated most strongly from its social base in Lebanon’s north Beqa Valley.28 he Christians: Fear of Sharing Iraq’s Fate Syrian Christians accounted for less than 5% of the population when the war began29 and were scattered throughout the country. most of them lived in urban areas, with notable concentrations in Aleppo, Homs, Damascus, Latakia, and Hasaka. his was the only way to sustain their communities because the rural areas are dominated by a conservative and often-unwelcoming strain of Sunni Islam. he era of Christian missions also played a role in this distribution pattern, providing converted Syrians with access  15 to modern education and hence more remunerative urban jobs. Since the country gained independence from France in 1945, the Christian proportion of the population has shrunk to a third of its former percentage due to a fertility rate twice as low as that of muslims and an emigration rate twice as high. he aging community has thus been growing weaker by the decade, a trend exacerbated by their proliferation of denominations. Around 36% of Syrian Christians are Orthodox Greeks, followed by Orthodox Armenians (22%), Catholic Greeks (12%), and Catholic Armenians (11%). he remainder are a mix of Orthodox and Catholic Syrians, maronites, Protestants, Assyro-Chaldeans, and other subgroups.30 most Christians belong to the middle class, working as merchants, liberal professionals, or civil servants. hey are not well represented in the army or intelligence services. Because of this, they tend to feel helpless today, with many becoming easy prey in a wartime environment where abductions and organized crime are commonplace. Since 2011, Christians have left the country in even greater numbers than before, seeking refuge in Lebanon, Armenia, Europe, and North America. hose who stayed behind have generally supported the Assad regime. Some intellectuals participate in the opposition, such as michel Kilo and George Sabra, but most of these individuals hail 16  from a former generation of left-wing opposition that is no longer relevant to Christians under sixty. In January 2012, the Catholic Archbishop of Aleppo, Jean-Clement Jeanbart, told his followers that “Assad must be given a chance,” calling him “a loyal and sincere man.”31 He also expressed fears that the rise of Islamism might lead to a new dhimma system under which non-muslims would become second-class citizens, a concern likely shared by most of his brethren. hey do not want to leave their country permanently, but they have seen Iraqi Christians transit Syria en route to the West for years, so they take the possibility of forced light very seriously. Armenians were particularly traumatized by the Sunni revolt, viewing the main rebel faction in the opening phases of the war, the FSA, as a proxy of Turkey. As early as march 2011, the Armenian party Tashnag organized demonstrations in favor of the regime. hey later set up a militia to defend the Armenian districts of Aleppo, even retaking the old Christian quarter of Jdaide after it fell into rebel hands in August 2012. Since then, more Christians have sought to protect their neighborhoods and villages by joining the regime’s National Defense Forces militias. he Kurds: Quest for Autonomy he regime and the Kurds have been using each other from the start of the uprising, sharpening the social and terri- torial issures between Arabs and Kurds in the process. After the previously described 2004 revolt, many Kurds lost all conidence in their Arab countrymen, viewing them as complicit in the regime’s repression. Rather than joining forces with Sunni Arab rebel groups in 2011, they focused on their own goal of forming an autonomous Kurdish territory in the north, even cooperating with Assad when it suited their purposes. For its part, the regime seemed to realize the temporary beneits it could derive from this Kurdish quest for autonomy, particularly in terms of keeping the opposition fractured. his may explain why its response to Kurdish anti-regime demonstrations in the towns of Qamishli, Amuda, and Afrin was not as harshly repressive as its crackdown on Homs or Hama. At the same time, the regime maintained close contact with the Kurdistan Workers Party (PKK) in Turkey, informally authorizing its return to Syria in spring 2011,32 more than a decade after ousting the group. his move indirectly bolstered Syrian Kurdish factions that were ailiated with the Turkish group (see chapter 3 for a fuller discussion of the Assad-PKK connection). he number of Kurds in Syria is often underestimated by analysts, who tend to cap them at 10% of the population. In fact, they are closer to 15%. In 2010, “purely” Kurdish territories were rather small (Afrin and Kobane), with most Kurds living with Arabs in mixed territories.33 his intermixture decreased during the war, and despite reports of forced relocations and other actions against Arab civilians,34 Kurdish authorities have so far avoided fullscale ethnic cleansing in their zones of control. Around two million of Syria’s three million Kurds live in a strip of territory along the Turkish border. Over the course of decades, however, the extreme poverty that characterized much of this rural zone spurred the other million to move south, mainly to Aleppo and Damascus. After one or two generations in an Arab milieu, many of these Kurdish families became Arabized; in fact, the regime seemed to leave Kurdish territories destitute for this very purpose, to encourage Arabization. his phenomenon was more visible in Damascus than Aleppo, where Kurds still maintained their ethnic identity and close ties with their home villages in the years leading up to the war. As early as 2012, the Kurds formed militias to control their territories and prevent Arab rebel forces from entering. By autumn 2013, they had formed a government in Jazira “canton” under the aegis of the Democratic Union Party (PYD), which commanded the People’s Defense Units (YPG) and had become the main political and military force of “Western Kurdistan.” hen as now, their objective was to build their own zone modeled on the Kurdistan Regional  FABrICE BAL ANChE Government (KRG) in northern Iraq. To achieve this, they seem to hope that neither Assad nor the rebels will win a deinitive victory, thus keeping the regime too weak to reassert itself in the north and the opposition too preoccupied to challenge their autonomy project. Open Kurdish hostility toward the armed Arab opposition was evident as early as autumn 2012, when Syrian Kurdish Peshmerga forces fought FSA units in the vicinity of Afrin and near the Kurdish district of Ashraiya in Aleppo. he situation deteriorated further that December, with Salaist groups led by Jabhat al-Nusra entering Syria from Turkey and attacking the north-central Kurdish town of Ras al-Ain. From that point on, the PYD moved closer to Assad in the hope of securing military support. In doing so, the Kurds became strategic assets for the regime, which has relied on their presence along the northern frontier to deprive Arab rebels of their bases and supply lines in Turkey. meanwhile, the U.S. support the Kurds have received against the Islamic State (IS) since summer 2014 has permitted them to double their territory in Syria. hey were able to link the cantons of Kobane and Qamishli by seizing Tal Abyad. hey also managed to take the city of Hasaka in 2015 after local regime forces grew too weak from ighting IS. heir ideal goal is to merge their eastern territory with the Afrin canton to the west, making their so-called “Rojava” region a contiguous entity along the entire border. As will be discussed further below and in chapter 3, military developments since 2015 and local demographic realities make that scenario increasingly unlikely, though a narrow Assad-controlled corridor was established between these areas in march 2017. Territorial Fragmentation most maps of the war divide the country into four major areas of control: the regime zone, the Kurdish zone, the IS zone, and the “other rebels” zone. Yet this vision of Syria can be misleading if extrapolated too far because the fragmentation has been much more pronounced from the earliest stages of the uprising. Sunni Arab rebel territory has the most issures; it consists of small, discrete areas controlled by local warlords. Even the regime zone is no longer wholly run by a centralized administration like it was in the past. he state administration is locked in competition with local National Defense Force militias that oversee key areas while the army is occupied on other fronts. For example, the young militiamen in Jabal al-Druze will only perform their military service locally, defending threatened Druze zones such as the foot of mt. Hermon. his system of decentralized protection has been quite efective during recent phases of the war because local populations are S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r  more motivated to defend their land than conscripts from other regions. Yet it may prove problematic in the long term, since the national army cannot intervene and defend a given area if locals do not support the regime. As for the Kurds and IS, their methods of territorial control have been broadly similar, at least in organizational terms: they tend to delegate local management to civilians while keeping militia garrisons and broader military oversight in the hands of central authorities. his approach enabled the PYD and IS to establish centralized control over a united territory early on. Even in the Afrin canton that remains separate from the rest of Rojava, local authorities do not question the political control exercised by the PYD because the area has long been a stronghold for the party, and because the population is 99% Kurdish. Yet Kurdish forces cannot exercise that level of authority in the various Arab-majority areas they passed through on the way to the IS “capital” of Raqqa in the Euphrates Valley. hus, even as the PYD dominates the Arab militias that have partnered with it to form the Syrian Democratic Forces (SDF), the group has still been obliged to delegate power to local Arab chieftains when moving beyond Kurdish-majority areas. It is unclear whether this blurring of sectarian lines is a temporary tactical gambit or an indicator of longer-term cooperation between Kurds and local Arabs. Whatever the case, Arab tribes could decide to take further advantage of the situation once IS forces are defeated, playing the PYD and Assad against each other and forming a semi-autonomous bufer zone of their own between Rojava and regime territory. Territorial fragmentation also occurred at the city level as the war progressed. For example, many neighborhoods of Damascus were recomposed into sectarian bunkers. he new dividing lines cut residents of from each other and encouraged the return of local economies, which only reinforced the cleavages. Finally, several waves of dramatic population movement during the war have emerged from—and exacerbated—Syria’s territorial fragmentation. Whenever one sect has established military control over a given area, it has led to the departure of some of the local population, whether voluntarily or through ethnic cleansing. For example, the Islamic State took the latter approach, pushing out native Christians, Shia, and other non-Sunni minorities from whatever territories it conquered. he Kurds of Raqqa city, who accounted for about 20% of the population,35 virtually all migrated. Taken together, these profound upheavals and internal issures could portend new political borders in the future Syria, and perhaps even neighboring countries.  17 12. MILITARY SITUATION, May 2017 dEMOgrAPhICS OVErwhELMEd By wAr By march 2017, more than 465,00036 people had been killed or gone missing during the war. And as mentioned in the Introduction, another 7 million had led the country as refugees as of September 2017,37 while 6 million more had become internally displaced persons (IDPs). According to UN projections,38 Syria’s resident population should have been around 18.5 million by the middle of 2015, before the huge migration to Europe that summer. Yet the author’s calculations indicate that the resident population at that time was more like 16 million—the UN underestimated the number of refugees (4.2 million instead of the more accurate 5.3 million) and did not take the wartime reduction of fertility rates into account.39 he resident population was still around 16 million as of September 2017 due to continued mass emigration over the previous two years. Put another way, more than half of all Syrians have left their homes, either as refugees or IDPs. Tellingly, much of this population redistribution has occurred along clear sectarian lines, on both the national and local level. Tigris River TURKEY Qamishli Ras al-Ain Kobane Azaz Afrin Tal Abyad Manbij HASAKA al-Bab IDLIB al-Shadadi RAQQA ALEPPO Jisr al-Shughour Maarat al-Numan LATAKIA Tabqa DEIR AL-ZOUR HAMA Salamiya Rastan TARTUS Mediterranean Sea al-Qusayr Mayadin Shaer gas field HOMS Palmyra Abu Kamal Euphrates River LEBANON Zabadani DAMASCUS IRAQ al-Dumayr N QUNEITRA W Golan Heights S SUWAYDA ISRAEL E DERAA 100 km 0 JORDAN AREAS OF CONTROL INHABITED UNINHABITED Syrian Army Syrian Army Province border Main road SDF (Kurdish) SDF (Kurdish) International border Secondary road Other rebels Other rebels Province center Desert Islamic State Locality Islamic State he Battle of Numbers Before further it, the sources 18  examining this redistribution and drawing conclusions from question of problematic data needs to be clariied. For much  FABrICE BAL ANChE of the war, the opposition and regime have engaged in a battle of numbers regarding civilian populations in their areas of control, at times misleading the international organizations entrusted with providing humanitarian assistance to these communities. he regime no longer publishes demographic statistics, but it does provide such information to the UN Oice for the Coordination of Humanitarian Affairs (OCHA), which moved its Syrian headquarters to Damascus in 2015. Consequently, there are serious doubts about the number of IDPs reportedly living in the regime zone. OCHA’s estimates for this population are often based on data from the regime, which has ample cause to inlate the numbers in order to receive more humanitarian help.40 For its part, the opposition—more precisely, the Syrian Interim Government based in Gaziantep, Turkey—created a special “Assistance Coordination Unit” (ACU) to provide demographic data at the provincial level, but only for “liberated areas.” Up until OCHA moved to Damascus, opposition data served as a key source for UN agencies to determine humanitarian needs in Syria. As with regime data, however, ACU statistics were not always accurate, particularly with regard to IDPs. he opposition often inlated these numbers not only to secure more aid, but also to demonstrate that most of the population lived within its areas of control. 13. SYRIAN POPULATION GROWTH 1940–2010 Population (in millions) 20 15 10 5 0 1940 1960 Syrian population igures are manipulated in high places because they are weapons of war. he most glaring example of this occurred in Aleppo. When the army surrounded the city’s rebel-held eastern districts in August 2016, the opposition claimed that the area contained around 300,000 civilians, while the UN put forth a similarly high estimate of 250,000.41 hese igures were taken up by Western oicials and media outlets who denounced the humanitarian crisis in the rebel enclave. After the army captured East Aleppo four months later, however, there turned out to be only 100,000 civilians there.42 moreover, during the author’s June 2014 research mission on Syrian refu- S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r  1981 2004 2010 gees in Turkey, members of medecins Sans Frontieres responsible for helping residents in Aleppo indicated that only 200,000 civilians remained in the rebel part of the city. It is diicult to believe that this number could have increased to 250,000–300,000 by August 2016, since the regime began bombarding East Aleppo in summer 2013 speciically in order to make civilians lee and isolate the rebels. Apart from occasional ceaseires, this bombardment was continuous up until the rebel districts fell, causing drastic deterioration in East Aleppo’s security situation and local economy. In short, thousands of civilians were leeing the area’s ever-tightening encirclement, not moving in. he opposition’s potential reasons for inlating Aleppo’s numbers are no mystery: to obtain more aid and, later, a UNbrokered ceaseire. Yet why did the UN, which is supposed to publish reliable igures, validate the rebel numbers? French journalist Georges malbrunot found the most likely answer in December 2016, when he interviewed a UN oicial from Geneva who admitted that such inlation “was necessary to help the insurgents.”43 In other cases, the UN’s estimates were in line with reality but opposition groups and international humanitarian organizations challenged its data, sometimes doubling or tripling the numbers. During the battle for the Damascus suburb of Daraya in 2016, the UN estimated that 4,000 people were under siege, but the local council announced more than 8,000,44 and the NGO Save the Children said 12,000.45 most Western and Arab media reports used the highest estimate, but when the rebels negotiated their exiltration to Idlib that month, only 4,000 people were in Daraya, including a thousand ighters. Such examples show why observers must be very cautious about relying solely on igures provided by sources involved in the conlict, including international organizations who may be using compromised data. Rapid Population Growth Interrupted In 1947, Syria’s population was 3.46 million.46 By the eve of the 2011 upris 19 ing, it exceeded 20 million. his strong growth was encouraged by multiple Syrian governments, who saw it as a strategic interest. Early on, population expansion was a response to the ArabIsraeli wars, which in the Baath regime’s view had caused the dismemberment of historic Syria (the Bilad al-Sham) by more populous neighbors. Syrian families’ desire for children was matched by the state’s desire for a large population to resist external predations. his growth began to decline in the mid-1990s, however. In previous years, the annual growth rate exceeded 3%, essentially doubling the population every two decades. Yet according to Syria’s Central Bureau of Statistics, it fell to 2.6% between the last two censuses (1994 and 2004). Four provinces even fell below 2% during this period: Tartus, Latakia, Suwayda, and Hasaka. he case of Hasaka is quite surprising because fertility rates in this poor Kurdish-majority region remained very high. As described above, however, poverty and underdevelopment spurred many residents to leave the province, in line with the Baath regime’s efort to push Kurds southward toward cities where they were more likely to become Arabized. Indeed, Syrian Kurds have been victims of Arab nationalism since the country’s independence. In 1962, the pre-Baath government conducted a census in the northeast that took ethnic and religious ailiation into account. Afterward, around 120,000 20  of the area’s nearly 1 million Kurdish residents were stripped of Syrian citizenship and classiied as foreigners. In 2011, these bedoon were estimated to number around 300,000.47 meanwhile, as part of a massive irrigation program in the Euphrates basin, the regime pushed Arabs into Kurdish regions in order to reduce the latter’s demographic weight and isolate them from Kurdish areas in southern Turkey. Population growth decreased in Syria’s Alawite and Druze regions as well, mainly due to a fall in fertility rates among these religious minorities. By 2004, the rate was 2.1 children per woman in the Alawite-dominated coastal region, 1.8 in Jabal al-Druze, and 1.8 for the widely dispersed Christians.48 hese minority communities no doubt felt threatened by the higher fertility rates seen among the country’s Sunni Arab majority, particularly in the north and the Euphrates Valley (e.g., 5.5 children per woman in Raqqa and 6.2 in Deir al-Zour49). he Alawitedominated Assad regime has therefore counted on its ability to divide the Sunnis in order to ofset this massive demographic disadvantage. he civil war slowed population growth even further—abruptly and drastically. he majority of Syrian men have been mobilized on one side or another, greatly afecting fertility rates among all communities. Economic precariousness, unending violence, and population dis- placements have reduced the birthrate as well, and the deaths of so many men on the battleield ensure that these effects will be felt far into the future. It is impossible to determine precisely how the growth rate of individual communities has been altered by the war; one can only extrapolate on the trends of previous decades. Two things are clear, however: (1) the country’s sectarian demographics have been far more afected by casualty, displacement, and emigration rates than declining birthrates; and (2) while some minorities have increased their share of the population inside Syria due to these factors, Sunni Arabs retain their wide majority. Huge Population Movements By analyzing and extrapolating from the author’s previously described GIS statistics and other data, one can reach rough estimates of Syria’s current population distribution. he greatest challenge is to locate the internally displaced people. Around 6.3 million IDPs remained in the country as of January 2017, comprising nearly 40% of all current residents. Some 80% of these IDPs reside in regime-controlled areas. As noted previously, these numbers may be inlated given OCHA’s reliance on questionable data sources. Whatever the case, IDPs have arrived from both opposition-held territory 14. REFUGEES BY SECT, June 2017 Kurdish 9% Christian 8% Turkmen 2% Shia minorities 2% Arab Sunni 79%  FABrICE BAL ANChE 15. SYRIAN REFUGEES AND IDPs, May 2017 N W E S 3,000,000 TURKEY 1,200,000 SYRIA 6,300,000 240,000 1,010,000 IRAQ LEBANON 1,200,000 660,000 Government figure UNHCR figure 120,000 EGYPT JORDAN 0 Source : UNHCR, Syria Regional Refugee Response, May 2017 S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r  200 km and other regime areas that were ravaged by ighting. Some of them may have come from diferent provinces, while others probably just changed neighborhoods within the same city because of urban violence. In general, IDPs go where they have acquaintances or relatives and where they feel they will be safe. his creates diferent migration routes that greatly depend on religious and/or ethnic identity. Christians led to Christian-majority areas such as Wadi al-Nasara; many Alawites in Damascus returned to their ancestral villages on the coast; the Druze of Jaramana led to Suwayda. In contrast, Sunni Arab IDPs have not hesitated to head for non-Sunni areas such as Jabal al-Druze or Tartus. his inlux of Sunni refugees has altered the sectarian distribution of some minority strongholds, though it also shows that the war is not strictly sectarian, since IDPs base their movements on multiple factors. By the end of 2015, more than 80% of the Syrians who had led abroad were Sunni Arab—hardly a surprise given the community’s majority status and the fact that most of the ighting had taken place inthe Sunni Arab-dominated areas that spawned the rebellion. he Christian population has been even more decimated by wartime emigration, in large part because members do not have their own sectarian refuge inside the country, unlike Alawites and Druze. Christians represent around 10% of all  21 16. RESIDENT SYRIAN POPULATION BY SECT Christian 5% 2011 Kurdish 14% Alawite 10% Turkmen 1% Christian 3% Kurdish 16% Shia 1% Ismaili 1% JUNE 2017 Alawite 13% Turkmen 1% Shia 1% Ismaili 1% Druze 3% Druze 4% Arab Sunni 61% Arab Sunni 65% Sunni Syrian refugees (600,000–700,000) but only 5% of the country’s prewar population (1.2 million), meaning that half of the community has emigrated since 2011 alone. Armenians in particular have been afected. Out of the 150,000 who lived in Aleppo before the war, only a few tens of thousands are left. most went to Armenia, where they do not need a visa, or to the United States, France, or Canada. Kurds comprise around 10% of Syrian refugees as of fall 2015. most of them went to Iraqi Kurdistan, where they were well received. Even Kurds 22  Druze Shia Twelvers Turkmen from Damascus and Aleppo have chosen to take refuge in the KRG rather than closer destinations like Lebanon or Turkey. In addition to ethnic ailiation, the greater possibility of inding work there was a powerful attraction; even after the KRG fell into economic crisis in 2014, it remained a beacon of prosperity compared to the Kurdish zone in northeast Syria (whose economy was a disaster) and the Kurdish areas in southern Turkey (where the central government has made clear that Syrian Kurds are not welcome). Finally, Alawites, Druze, Shia, and Alawite Christian Ismailis left the country at lower rates than Sunnis because their territories have largely been spared by the war. Yet those who lived in areas conquered by the rebels or IS had to lee. Population Concentrated in Regime-Controlled Areas As mentioned previously, the regime controls only around half of Syrian territory as of September 2017, but this zone contains around two-thirds of the remaining population. Other factions control large swaths of territory but not nearly as many people. Ismaili Kurdish At its peak, IS held almost half of Syria, but most of its territories were sparsely populated desert areas, and only two million people lived under its rule. Once the group lost manbij, Jarabulus, and al-Bab, this igure was reduced by half a million, and the number dropped further amid subsequent defeats in 2017. he areas held by other Sunni Arab rebel factions (mainly in the northwest, the south, and the Ghouta district east of Damascus) have lost the most inhabitants because they are the least safe. Russian and Syrian airstrikes have prevented all semblance of normal life there, and  FABrICE BAL ANChE 18. POPULATION BY ZONE May 2017 17. POPULATION BY SECT IN THE REGIME AREA Syrian Army June 2017 15 % SDF (Kurdish) 13 % Other rebels 5% 6% 7% Islamic State 21% 65 % 2% 58% 2% March 2013 6% Syrian Army SDF (Kurdish) 18 % Rebels 7% Sunni Druze Shia Twelver Alawite Christian Ismaili 41% Kurdish Disputed 34 % S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r   23 rivalries between various rebel groups guarantee further insecurity. Paradoxically, the area held by IS was safer for many Sunni Arabs because the group’s centralized authority helped maintain local security. In both the IS zone and other rebel areas, however, many minorities and secular Sunnis have led whenever extremists have sought to impose their rigorous interpretations of Islamic law (though the departure of non-practicing Sunnis has been ofset somewhat by the arrival of thousands of foreign Sunni jihadists). As for the Kurdish zone, its population has fallen to less than two million. Economic diiculties and the PYD’s monopoly on political life have led around half a million people to leave during the war. Inside the regime-controlled zone in western Syria, Sunni Arabs remain the majority population, in line with their demographic dominance in the country as a whole. Yet religious minorities now account for 42% of the population there, a major increase. most everywhere Assad’s army goes, it receives support from local Christian, Alawite, Druze, Ismaili, and Shia communities. In contrast, advances by Sunni Arab rebels and IS jihadists earlier in the war inevitably triggered the departure of these same populations. Only the Druze of Jabal al-Summaq remained in rebel territory, but they were forced to convert to Sunni Islam, making them a fragile exception 24  that proves the rule. In all, the population of rebel-held territory is now more than 98% Sunni Arab. many Turkmens in these areas have stayed put and constitute the main minority there, but this is unsurprising given their intense antiAssad sentiment and heavy involvement in opposition militias.50 At the national level, religious minorities constituted around 25% of the population as of fall 2015, compared to 20% before the war, while the Sunni Kurds made up 15% and the Sunni Arabs 60%. he current proportions should be the same because the violence has been concentrated in Sunni Arab areas, spurring higher emigration rates from there. his represents a radical transformation in the distribution of Syria’s population, to the beneit of non-Sunni minorities, whose concentration in the regime zone makes their long-term presence more viable. In addition, the war is not over—millions more Syrians will likely lee the country as new military campaigns unfold, and Sunni Arabs who support the opposition or IS will continue to make up the bulk of these refugees. In the Euphrates Valley, Sunni Arabs themselves are the most likely drivers of these future refugee lows, since some tribes will probably seek retribution against their co-religionists who sided with IS. Although these population movements exacerbate sectarian fragmentation in many parts of the country, they actually reinforce diversity in other ar- eas. Notwithstanding the many community-speciic motivations described at length throughout this chapter, there is no clear-cut pattern behind this process—its mechanisms must be understood individually. he irst driver of population displacement is of course the violence that has engulfed Syria since 2011. Economic deterioration and a desire to evade mandatory military service are primary motivators as well. Fighting, Repression, and Conscription As described previously, the regime has pursued a very basic and ruthless counterinsurgency strategy of violently separating rebels from civilians in order to better eliminate the opposition.51 his is the main cause of displacement on Syria’s various fronts. he army has applied the principle to all areas held by the rebellion, prioritizing major urban districts such as East Aleppo and the western suburbs of Damascus. Beginning in late 2013, the army massively bombarded rebel areas of Aleppo while gradually encircling them, with the objective of spurring civilians to lee. When the regime inally retook the last of these Sunni Arab districts in December 2016, the population of East Aleppo had been reduced from more than one million as of July 2012 to less than 100,000. Like many before them, the last civilians to be evacuated went to rebel-controlled Idlib—a journey with no possibility of return because most of them are related to rebel ighters and are consequently on the regime’s blacklist. his same method of displacing and/ or banishing rebel-sympathetic populations has been practiced in other locales too, including militarily important ones. he most prominent examples are in the Homs area, where the army retook the Bab al-Amr district in spring 2012 and the town of al-Qusayr in may 2013, expelling Sunni Arab civilians and armed rebels alike. hose who led al-Qusayr are now refugees in Lebanon, with little possibility of return, since they are all considered opposition supporters. Similarly, the regime’s winter 2013–2014 reconquest of the Qalamoun border region forced most locals to seek refuge in Lebanon. Assad’s focus on expelling rather than wooing restive populations helps explain the extent of the exodus from Syria since 2011—and suggests that the majority of these refugees will not be permitted to return. In all likelihood, only those who can prove that they left for economic or safety reasons will be allowed back in. his is why tens of thousands of Syrian refugees in Lebanon rushed to their local embassy in June 2014 to cast absentee ballots in that month’s Syrian presidential election, so that they could prove their loyalty in the hope of going home eventually. In addition to the millions of ordinary civilians caught up in the ighting and forced to lee, tens of thousands of  FABrICE BAL ANChE opposition activists have been repeatedly displaced throughout the war. he repression that began in march 2011 led some to seek refuge in opposition-held territory. But as Islamist extremists took over much of the armed rebellion, these secular/moderate activists had to lee abroad. Not all of them made it out— for example, lawyer and human-rights activist Razan Zaitouneh was kidnapped in Douma and likely murdered by the Islamist group Jaish al-Islam in December 2013.52 And early regime opponent Raed Fares, an independent radio host in Kafr Nabl, was arrested several times by al-Qaeda ailiate Jabhat al-Nusra for daring to broadcast songs against the group’s wishes; his life was spared once he pledged to respect Nusra’s brand of “Islamic order” and cease airing songs.53 many other young Syrians led abroad after deserting the army or evading conscription, regardless of their religion. Numerous families have left the country one or two years before assigned conscription dates in order to spare their children from service in the war. he same phenomenon is occurring in the Kurdish and rebel zones, where conscription systems have also been established.54 Economic Deterioration Syria’s eroding economic situation during the war has been pushing much of the middle and professional classes into exile, particularly those individuals who have preexisting networks abroad. his includes physicians, engineers, technicians, and university professors, many of whom are resettling in Europe or the Persian Gulf countries. he Syrian pound lost more than 90% of its value between 2011 and 2017, and wages have not been adjusted suiciently to compensate for the drop. he situation deteriorated sharply during winter 2014–2015, when inlation caused the exchange rate to rise from 300 pounds per U.S. dollar to nearly 500.55 Syrian families had additional impetus to emigrate after German chancellor Angela merkel opened her country’s doors to refugees in summer 2015; many of these “economic” refugees sold everything they had to pay for their family’s passage to Germany. Yet this low was largely halted after the European Union reached agreements with Ankara to require visas from any Syrians attempting to enter Turkey, the main refugee route to Europe. Some portions of north Syria, such as Jazira, have been hit especially hard by economic problems due to the blockade imposed by Turkey and the KRG.56 In response, many Syrian Kurds have taken refuge in Iraqi Kurdistan, where conditions are much more favorable. Despite its own economic and political diiculties, the KRG has generously welcomed its neighbors, who are glad to take the ample work opportunities created by a higher standard of living and a bustling manual labor market. S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r  Ethnic Cleansing in Rebel and IS Territories In areas held by the rebels, ethnic homogeneity is de rigueur, and native Christians, Druze, and Shia have been forced to leave or convert. While some districts initially sought to demonstrate tolerance toward minorities (e.g., Jabal al-Summaq), such sentiment largely evaporated once Jabhat al-Nusra gained prominence in the rebellion and began to take over these areas, in some cases massacring non-Sunni Arabs.57 At times, the regime and opposition have essentially held each other’s sectarian pockets hostage. For example, the Shia enclave of al-Fua and Kefraya northeast of Idlib was long surrounded by rebels who could easily seize it, but they were restrained by the fact that Hezbollah had simultaneously encircled the rebel enclaves of madaya and Zabadani north of Damascus. In April 2017, the remaining population of madaya and Zabadani was evacuated to Idlib province, while most civilians in al-Fua and Kefraya were transferred to Aleppo, though Hezbollah and National Defense Force units are still occupying the Shia enclave.58 In the territories controlled by IS, there were few non-Sunnis to begin with. Al-hawra, a Euphrates town that overlooks a major dam, had a majority Christian and Alawite population due to the inlux of employees from the ministry of Irrigation, but these minorities led in spring 2013 when rebels seized Raqqa. he small Christian community in Raqqa (around 1,000 people) likewise led once IS moved in and started imposing its repressive policies. IS began to pressure the city’s Kurds as well, partly because they were considered to be allies of the PYD to the north, and more generally because of the jihadist group’s tendency to deine even fellow Sunnis as potential enemies if they are not Arab or do not follow its warped interpretation of Sunni orthodoxy. IS cadres also played on Syria’s existing Arab/Kurdish issures to impose themselves on local populations, just as they did in Iraq. On the Turkish border, the group sought to awaken old territorial disputes and exploit the widespread Arab view of Kurds as immigrants from Turkey who have no real property rights in Syria. he resultant tensions produced an all-out military siege in Kobane, while Kurds in Tal Abyad were expelled in spring 2013 by IS-backed Arab tribes. When the PYD later retook the city, it did not expel the native Arab population, though it has conducted such displacements in other areas (see above and chapter 3). Non-Kurds in Rojava Since winning a key victory against IS forces in Kobane in January 2015, the PYD has continued to expand its territory, to the detriment of both IS and other rebels in the Azaz corridor to  25 the west. Advancing from the heavily Kurdish territories of Afrin, Kobane, and Qamishli, the PYD began taking mixed and non-Kurdish territories in a bid to establish contiguity between its eastern and western cantons. Its forces took Tal Abyad in spring 2015, spearheaded the SDF operation to push IS out of manbij in August 2016, and led the siege of Raqqa in 2017. he PYD has also taken non-Kurdish territories such as al-Shadadi in order to cut of IS routes to Iraq and gain control over local oil wells. Based on the 2004 census and the author’s projections, the PYD-controlled Rojava zone would have held around 2.6 million inhabitants if the war had not broken out. Factoring in the movements of refugees and IDPs, Rojava’s actual population was around 1.8 million in fall 2015, only half of them Kurds. In the eastern Jazira and Kobane cantons, Kurds still hold a slight majority, but not in the western Afrin canton.59 Depending on where one draws its borders, Afrin canton is either 30% or 40% Kurdish. Some areas on the fringes of Rojava’s cantons (e.g., Azaz and Jarabulus) are even more heavily populated by Arabs and Turkmens. Accordingly, the more the PYD expands its territory, the more it will have to integrate non-Kurdish populations. his is the case in the manbij area, where Kurds represent less than a quarter of the 26  population. Yet PYD leaders still hope to merge Afrin and Kobane, despite the August 2016 Turkish intervention in the area of Jarabulus, Azaz, al-Bab. heir aim is not just to unify Rojava, but also to “re-Kurdify” the area. Close attention to local toponymy and maps published under the French mandate indicates that a signiicant portion of the population between these two cantons is of Kurdish origin. hese Arabic-speaking Kurds could choose to reconnect with their roots if the PYD became the dominant political actor there. he area’s demography might also be signiicantly altered if Arab refugees do not return to northern Syria in the same numbers as displaced Kurds. Displacements in the Regime Zone As described previously, the regime zone has the most diverse sectarian mix in Syria, welcoming IDPs from all denominations. At the local level, however, this mixing has its limits, and some cases of ethnic cleansing have occurred. Sunni Arabs still make up the zone’s majority population, but they have been expelled from some areas and will probably not be permitted to return after the war. he case of al-Qusayr is emblematic. Between the Sunni rebel takeover in 2011 and the Shia Hezbollah takeover in spring 2013, around 90% of the town’s mixed population left. Since then, only the Christian and Shia residents have returned; Sunni Arabs are generally forbidden from coming home unless they have proven their loyalty to the regime by joining the army and ighting rebels. Similarly, Sunni Arab residents of alHussein village at the foot of the Crac des Chevaliers have not returned either. hey rose up against Assad in 2012, offering the medieval castle to rebel forces that were bombing nearby Christian and Alawite villages. hey left once the tide of battle turned. After the regime recaptured the castle in spring 2014, local Christian and Alawite militias destroyed many of the village’s dwellings to prevent Sunnis from coming home. Yet observers should be careful to distinguish cases like this—where local minorities engaged in ethnic cleansing out of revenge or perceived self-preservation— from cases in which central government forces have expelled rebel populations as part of a deliberate military strategy (as in East Aleppo). Apart from displacements, the regime has sought to keep a lid on Sunnis within and abutting its area of control by ensuring that minorities dominate strategic points throughout western Syria, including the Alawites in maan, Ismailis in Salamiya, Christians in maharda, and Shia in mazraa. Yet the Sunni enclave stretching between Rastan and Houla—right in the middle of the regime zone—fell to the rebels at the beginning of the war and has remained under their control ever since. CONCLuSION Syria’s civil war is not driven exclusively by sectarianism, but that appears to be the most essential factor today. After six years of war, all of the country’s main confessional minorities either support the Assad regime or have ceased major hostilities against it, while the anti-Assad rebellion is almost exclusively Sunni Arab. True sectarian coexistence is limited, and ethnic homogenization is under way in many areas. he regime zone is heavily mixed on the whole, and is home to a Sunni majority, but individual areas within it are being increasingly segregated. And on the most basic neighborhood/village level, sectarian mixing is almost entirely absent except in areas belonging to economic and political elites. Large cities remain mixed, but communication between residents from diferent sects is low, and sectarian districts are often barricaded for fear of attacks from other communities (apart from the heavily secured Damascus city center). Ultimately, the situation is in line with trends observed in Syria since the 1980s, when the muslim Brotherhood revolt brought latent sectarian issures back to the fore. he demographic weakening of religious minorities to the beneit of the Sunni Arab majority is one of the war’s primary causes—and the regime’s main obstacle in regaining control over more  FABrICE BAL ANChE 19. POTENTIAL MIGRATION FROM REBEL AND IS AREAS, May 2017 N W Tigris River TURKEY Qamishli E Kobane S Ras al-Ain Tal Abyad Azaz N Manbij Afrin HASAKA W E al-Bab S IDLIB Jisr al-Shughour RAQQA ALEPPO Euphrates Valley 1,200,000 al-Tabqa Idlib-West Aleppo al-Shadadi LATAKIA 1,200,000 DEIR AL-ZOUR HAMA Rastan Rastan pocket Salamiya 100,000 TARTUS HOMS Mediterranean Sea Mayadin Shaer gas field al-Qusayr Abu Kamal Palmyra Euphrates River LEBANON Zabadani DAMASCUS Damascus suburbs 300,000 IRAQ Douma QUNEITRA Golan South Heights 400,000 ISRAEL SUWAYDA DERAA JORDAN AREA CONTROLLED BY 100 km 0 RESIDENT POPULATION Syrian Army SDF (Kurdish) Other rebels Islamic State S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r  Province border Main road International border Secondary road Province center Desert 1,000,000 500,000 100,000 Locality territory. Assad probably cannot stabilize Syria under his rule unless he further rebalances the population’s size and territorial distribution in favor of his minority constituents. Given the breadth of the country’s demographic disparity (in terms of sheer numbers and fertility rates), this would mean expelling millions more Sunni Arabs and preventing their return once peace is restored. Such a scenario is hardly far-fetched: the regime has already managed to drive out nearly 5.5 million of them, and the reconquest of Sunni Arab strongholds such as Idlib, Deraa, Jarabulus, and the Euphrates Valley would likely generate new mass migrations. Even if many Sunni refugees do manage to return, the regime has shown that it can maintain support (grudging or not) among large communities of native and displaced Sunnis in western Syria so long as it keeps them divided and dependent on its largesse. Yet that would probably be a temporary ix at best, as shown by the aftermath of past uprisings and crackdowns. Whatever the case, the international community can no longer do without a sectarian analysis of the Syrian revolt. he initial uprising was only partly mixed, and militarization quickly led to the full exclusion of minorities (the Kurdish factor complicates but does not invalidate this point, as will be seen in chapter 3). Alawites, Christians, Druze, and other groups do not believe the op 27 position’s promises that they would be safe in a free and democratic Syria, nor do they believe Western promises of international protection. In February 2012, French foreign minister Alain Juppe delivered the following declaration: I call for the participation of Christians and all other communities in the creation of a new and democratic Syria where all citizens will have the same rights and duties. Who can believe that the rights of minorities are better protected by bloody dictatorships than by democratic regimes? If questions persist about the future, I wish to tell the Christians of the Orient... that France will not abandon them.60 Yet today’s France is not the France of Napoleon III, which sent boats to protect the Christians of Lebanon from massacres in the 1860s. Although the French and other Western governments have provided arms and air support to the opposition at various points in the war, they did not take robust action when the regime crossed their supposed redlines (e.g., after chemical weapons attacks against civilians in August 2013). he Syrian people took note, viewing this inaction as a sign of Western weakness and setting aside any illusions they may have had about Western protection. In sum, the revolt’s sectarian aspects have too quickly been brushed under the carpet in favor of viewing the conlict as a classic political challenge by a revolutionary population against a dictatorial 28  regime. Syria is deep in the throes of the same ethnoreligious fragmentation already experienced in Lebanon and Iraq. Of course, Syria’s minorities want democratic reforms just as much as their Sunni Arab countrymen—but not at the cost of their marginalization or elimination. many of them fear that applying democratic principles in the Syrian context would eventually usher in a new dictatorship by the Sunni majority, similar to post-Saddam Iraq’s slide toward Shia majority rule and repression. To protect themselves from this scenario, minorities with their own viable territory may choose partition in the end, much like the Kurds have sought to do. Extending the war could even result in Syria’s minorities taking up arms against each other, on the model of Lebanon’s civil war. Whether the Assad regime stays or falls, the country will not escape further ethnic cleansing and territorial fragmentation. Notes 5. Hanna Batatu, Syria’s Peasantry, the Descendants of Its Lesser Rural Notables, and Their Politics (Princeton University Press, 1999). 6. Fabrice Balanche, “La fragmentation spatiale en Syrie: entre patrimonialisme et communautarisme rampant” [Spatial fragmentation in Syria: between patrimonialism and rampant communitarianism], Revue de l’Economie Méridionale, 2005, https://www. academia.edu/769788/La_fragmentation_spatiale_en_Syrie_Spatial_fragmentation_in_Syria. 7. 8. 9. Fabrice Balanche, “Les municipalités dans la Syrie Baathiste” [Municipalities in Baathist Syria: administrative deconcentration and political control], Revue Tiers Monde 193, January–March 2008, https://www.cairn.info/revue-tiersmonde-2008-1-page-169.htm. In March 2012, the UN estimated that 9,000 people had been killed since the beginning of the uprising. See “En Syrie, l’ONU parle de 9000 morts en un an” [In Syria, the UN speaks of 9,000 deaths in one year], Le Monde, March 27, 2012, http://www.lemonde. fr/proche-orient/article/2012/03/27/ syrie-bachar-al-assad-se-rend-ahoms_1676516_3218.html. By November 2012, the UN estimate had climbed to 40,000 deaths. See “Le bilan des violences en Syrie dépasse les 40 000 morts” [The toll of violence in Syria exceeds 40,000 dead], Le Monde, November 22, 2012, http://www.lemonde. fr/proche-orient/article/2012/11/22/lebilan-des-violences-en-syrie-depasseles-40-000-morts_1794102_3218.html. 1. See her article “Existe-t-il un problème communautaire en Syrie?” [Is there a community problem in Syria?], Maghreb-Machrek (1979). 2. Souhail Belhadj, La Syrie de Bashar alAsad [The Syria of Bashar al-Assad] (Paris: Belin Litterature et Revues, 2013). 3. Fabrice Balanche, La region Alaouite et le pouvoir Syrien [The Alawite area and Syrian power] (Paris: Karthala, 2006). 10. For details on how the author arrived at the demographic statistics cited in the study, see the “Note on Methodology” at the end of the Introduction. 4. According to author interviews with residents of Baniyas, April 2011. 11. Fabrice Balanche, “Ethnic Cleansing Threatens Syria’s Unity,” PolicyWatch 2528 (Washington Institute for Near East Policy, December 3, 2015), http:// www.washingtoninstitute.org/policyanalysis/view/ethnic-cleansing-threatens-syrias-unity. 12. Syria’s Turkmen population (1%) is included in these estimates. 13. See Kelly McEvers, “Revisiting the Spark That Kindled the Syrian Uprising,” NPR Morning Edition, March 16, 2012, http:// www.npr.org/2012/03/16/148719850/ revisiting-the-spark-that-kindled-thesyrian-uprising. 14. See Fouad Ajami, The Syrian Rebellion (Hoover Institution, 2012), p. 196. See also Christopher Phillips, “Sectarianism and Conflict in Syria,” Third World Quarterly 36, no. 2 (2015). 15. Suleiman Al Khalidi and Angus McDowall, “Hard Choices for Syrian Industrialists in Ruins of Aleppo,” Reuters, October 2, 2017, https://www.reuters. com/article/us-mideast-crisis-syria-economy-insight/hard-choices-for-syrianindustrialists-in-ruins-of-aleppo-idUSKCN1C71B8. 16. Jacques Weulersse, Le pays des Alaouites [The country of the Alawites] (Institut Français de Damas, 1940). 17. Fabrice Balanche, La region Alaouite et le pouvoir Syrien [The Alawite area and Syrian power] (Paris: Karthala, 2006), p. 285. 18. French journalist Gilles Jacquier was killed by one such strike on January 29, 2012. At the time, Paris accused the Assad regime of organizing a vendetta against international journalists in Homs, and much of the French media tended to echo this line (apart from individual reporters such as Georges Malbrunot, who questioned the claim that Jacquier had been hit by regime fire). In July 2012, the rebels finally acknowledged that they had killed Jacquier in error, but the  FABrICE BAL ANChE French media generally ignored the revelation; only a few outlets such as Le Figaro reported it, e.g., see Malbrunot’s article “Jacquier : l’enquête française pointe les rebelles syriens” [Jacquier: the French investigation points to the Syrian rebels], Le Figaro, July 17, 2012, http://premium.lefigaro. fr/international/2012/07/17/0100320120717ARTFIG00525-jacquier-l-en quete-francaise-pointe-les-rebelles-syriens.php. 19. Mazen Ezzi, “A Static Revolution: The Druze Community,” in Friederike Stolleis, ed., Playing the Sectarian Card: Identities and Affiliations of Local Communities in Syria (Beirut: Friedrich Ebert Stiftung, 2015), pp. 39–70, http://library.fes.de/pdf-files/bueros/ beirut/12320.pdf. 20. Nikolaos van Dam, The Struggle for Power in Syria: Politics and Society Under Asad and the Ba’th Party (London: I. B. Tauris, 1996). 21. Cyril Roussel, Les Druzes de Syrie: Territoire et mobilité [The Druze of Syria: territory and mobility] (Beirut: Institut Français du Proche-Orient, 2011), p. 263, http://www.ifporient.org/ node/1024. 22. Human Rights Watch, ‘He Didn’t Have to Die’: Indiscriminate Attacks by Opposition Groups in Syria (March 22, 2015), p. 44, https://www.hrw.org/sites /default/files/reports/syria0315_ForUpload.pdf. 23. “Les druzes de Syrie pour la révolution, mais pas pour la rebellion” [Syrian Druze are for the revolution, but not for the rebellion], L’Orient-Le Jour, September 10, 2012, http://www.lorientlejour.com/ category/Moyen+Orient+et+Monde/ article/777148/Les_druzes_de_Syrie_ pour_la_revolution,_mais_pas_pour_ la_rebellion.html. 24. “Anna, Syrienne: ‘Nous avons perdu notre patrie et nos rêves de liberté’” [Anna, a Syrian: “We have lost our homeland and our dreams of freedom”], L’Orient le Jour, March 11, 2014, http:// www.lorientlejour.com/article/857609/ anna-syrienne-nous-avons-perdu-notrepatrie-et-nos-reves-de-liberte.html. 25. “Syrian Observatory: Qaida-Linked Militants Blow Up Shiite Hussainiyah,” Naharnet, June 16, 2013, http://www. naharnet.com/stories/en/87102. 26. See Myriam Ababsa, “Les mausolées invisibles: Raqqa, ville de pèlerinage chiite ou pôle étatique en Jazîra syrienne?” [The invisible mausoleums: the city of Raqqa between Shia pilgrimage and Syrian statism], Annales de Géographie 110, no. 622 (2001), pp. 647–664, http://www.persee.fr/doc/geo_00034010_2001_num_110_622_1706. See also Andrew J. Tabler, “Catalytic Converters” New York Times Magazine, April 29, 2007, http://www.washingtoninstitute.org/policy-analysis/view/ catalytic-converters. 27. Author interview with a French diplomat in Damascus, 2008. 28. Martin Klasta, “Le Hezbollah en Syrie: la résistance redéfinie?” [Hezbollah in Syria: resistance redefined?], MaghrebMachrek 218 (April 2014), pp. 85–98, https://www.cairn.info/revue-maghrebmachrek-2013-4-page-85.htm. 29. Youssef Courbage, “Ce que la démographie nous dit du conflit Syrien” [What demography tells us about the Syrian conflict ], Slate, October 15, 2012, http://www.slate.fr/story/62969/syrieguerre-demographie-minorites. 30. Ibid. 31. “‘Il faut laisser une chance à Assad,’ selon l’archevêque catholique d’Alep” [“Assad must be given a chance,” says Catholic archbishop of Aleppo], France 24, January 14, 2012, http:// www.france24.com/fr/20120113-syrieil-faut-laisser-une-chance-president- S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r  assad-selon-archeveque-catholiquealep-jeanbart. 32. Aimad Hesso and Cyril Roussel, “Les alliances des Kurdes de Syrie: un jeu de dupe?” [The alliances of the Syrian Kurds: a fool’s game?], Moyen-Orient 33 (Jan–Mar 2017), http://www.areion24. news/produit/moyen-orient-33. 33. Fabrice Balanche, Atlas du ProcheOrient Arabe [Atlas of the Arab Middle East] (Paris: Presses de l’Université Paris-Sorbonne, 2011), p. 36. 34. Amnesty International, “Syria: US Ally’s Razing of Villages Amounts to War Crimes,” October 13, 2015, https://www.amnesty.org/en/press-releases/2015/10/syria-us-allys-razing-ofvillages-amounts-to-war-crimes. 35. Estimate based on the author’s field surveys. 36. Jack Moore, “Syria War Death Toll Hits 321,000 With Further 145,000 Missing: Monitor,” Newsweek, March 13, 2017, http://www.newsweek.com/sixth-anniversary-syrian-conflict-looms-war-monitor-says-465000-killed-or-567181. 37. According to the Office of the UN High Commissioner for Refugees, 5.2 million Syrians were registered as refugees in Middle Eastern countries as of September 2017. Yet 6 million is a more realistic number—the UN figure underestimates the situation because many refugees in the region decide not to register for one reason or another (e.g., some fear being arrested and taken back to Syria, as has happened frequently in Lebanon; others are wealthy and do not see the point of registering). Adding the nearly 1 million refugees in Europe pushes the overall count to around 7 million. See the UNHCR’s “Syria Regional Refugee Response Inter-agency Information Sharing Portal,” http://data.unhcr.org/ syrianrefugees/regional.php. 38. See “Annual mid-year population, United Nations estimates: 2006–2015,” in UN Department of Economic and Social Affairs, Statistics Division, Demographic Yearbook 2015 (New York, 2014), https://unstats.un.org/unsd/demographic/products/dyb/dyb2015.htm. 39. The UN’s demographic projections tend to treat Syria as a uniform entity. To generate more accurate projections, one must take into account the vast differences between the country’s various provinces, particularly when it comes to sharply contrasting rural vs. urban fertility rates. 40. For instance, see OCHA’s “Humanitarian Needs Overview 2017,” December https://docs.unocha.org/sites/ 2016, dms/Syria/2017_Syria_hno.pdf. 41. OCHA, “Syria: ‘An estimated 200,000300,000 people are at risk of besiegement,’ says UN Humanitarian Chief,” July 21, 2016, https://www.unocha.org/ legacy/top-stories/all-stories/syria-estimated-200000-300000-people-are-riskbesiegement-says-un-humanitari. 42. “En silence, les rebelles quittent Alep entre des soldats russes” [In silence, rebels leave Aleppo between Russian soldiers], Le Figaro, December 22, 2016, http://www.lefigaro. fr/international/2016/12/22/01003-2 0161222ARTFIG00229-en-silence-lesrebelles-quittent-alep-entre-des-soldats-russes.php. 43. Ibid. 44. “Syrian government steps up siege of rebel-held Damascus suburbs: state media, rebels,” Reuters, February 6, 2016, http://www.reuters.com/article/ us-mideast-crisis-syria-daraya/syriangovernment-steps-up-siege-of-rebelheld-damascus-suburbs-state-mediarebels-idUSKCN0VF0RT. 45. “12,000 people trapped in Syria refugee camp by bombs, shells and bul 29 lets,” Guardian, May 13, 2016, https:// www.theguardian.com/global-development/2016/may/13/12000-peopletrapped-in-syria-refugee-camp-bybombs-shells-and-bullets. 46. Anne-Marie Bianquis and Mohamed al-Dbiyat, “La population Syrienne: un tournant démographique?” [The Syrian population: a demographic turnover?], Méditerranée, 81, no. 1 (1995), pp. 81–90, http://www.persee.fr /doc/medit_0025-8296_1995_num_81 _1_2878?q=population%20syrienne. 47. “Damas promulgue un décret de naturalisation d’habitants Kurdes” [Damascus promulgates a decree of naturalization of Kurdish inhabitants], Agence France-Presse, April 7, 2011, http://www.lemonde.fr/proche-orient/ article/2011/04/07/damas-promulgueun-decret-de-naturalisation-d-habitants-kurdes_1504426_3218.html. 48. Courbage, “Ce que la démographie nous http://www.slate.fr/story/62969/ dit,” syrie-guerre-demographie-minorites. 49. Ibid. 50. Aymenn Jawad Al-Tamimi, “The Factions of North Latakia,” Syria Com- 30  ment, December 10, 2015, http: //www.joshualandis.com/blog/the-factions-of-north-latakia. 51. David Galula, Counterinsurgency Warfare: Theory and Practice (New York: Praeger, 2006). 52. Nabih Bulos, “Top Syrian Rebel Leader Reported Killed in Airstrike,” Los Angeles Times, December 25, 2015, http://www.latimes.com/world/ middleeast/la-fg-syria-rebel-death20151225-story.html. 53. Laura Pitel, “Syria: Jabhat al-Nusra seizes two prominent activists in raid on opposition radio station,” Independent, January 10, 2016, http:// www.independent.co.uk/news/world/ middle-east/syria-jabhat-al-nusra-seizetwo-prominent-activists-in-raid-on-opposition-radio-station-a6805186.html. 54. According to author surveys conducted in Iraq and Syria (March 2017) and in Lebanon (June 2015 and June 2017). 55. Author interviews in Syria, June 2015. 56. Ankara and Erbil each have their reasons for blockading the Syrian Kurdish zone; these tensions are discussed at length in chapter 3. 57. For example, see Aymenn Jawad AlTamimi, “The Massacre of Druze Villagers in Qalb Lawza, Idlib Province,” Syria Comment, June 15, 2015, http:// www.joshualandis.com/blog/the-massacre-of-druze-villagers-in-qalb-lawzaidlib-province. 58. “Death toll from Aleppo bus convoy bomb attack at least 126: Observatory,” Reuters, April 15, 2017, https:// www.reuters.com/article/us-mideastcrisis-syria/death-toll-from-aleppo-busconvoy-bomb-attack-at-least-126-observatory-idUSKBN17H04Y. 59. According to the Kurdish administration, Afrin canton includes the Afrin district, Azaz district, Jarabulus district, northern al-Bab district, and northern Manbij district. Yet most of this territory is under the control of Turkey-backed rebels, not the Kurdish-led SDF. 60. See his op-ed “Les chrétiens d’Orient et les printemps arabes” [The Christians of the East and the Arab spring], L’Orient le Jour, February 29, 2012, https://www. lorientlejour.com/article/747414/Les_ chretiens_d%27Orient_et_les_printemps_arabes.html.  FABrICE BAL ANChE 2 . A S SAd ’S S y r I A, T h E N AN d NOw A S OF SuMMEr 2017, the Assad regime’s eforts to retake territory had given it control over roughly half the country, stretching from Latakia in the northwest to Suwayda in the southwest and to portions of the Euphrates River in the east. Damascus has been the centerpiece of this resurgence: since their nadir in march 2013, Assad’s forces have reestablished their grip on the suburbs, and the capital is no longer threatened by rebel pockets on the outskirts. Further south, Jabal al-Druze remains a stronghold for the army, while the Alawite coastal region to the west has been largely undisturbed by the rebellion since 2015. In the north, Aleppo city was reconquered in December 2016 after more than four years of ierce ighting. he regime also retains control over small but important enclaves in the east such as Deir alZour, Hasaka, and Qamishli, indicating that it still intends to retake all of Syria in the long term after eliminating rebel strongholds in the west. Does Assad have the means to match such ambitions? Answering this question requires more than assessing the regime’s military capabilities—it also means taking a closer look at the demography and loyalties of the various local populations under consideration. Currently, almost all of the territories that remain outside Assad’s control are either Sunni Arab or Kurdish. he territories that have remained faithful to the regime are dominated by Alawites and other religious minorities. Yet from these minority zones, Assad’s forces have been able to reassert their hold over certain areas with large communities of Sunni Arabs, who still constitute the majority population in the regime zone. his approach echoes the classic counterinsurgency strategy of relying on a loyalist minority—in Syria’s case more than one minority, each manning strategically crucial places within the regime zone, as described in chapter 1—while slowly reasserting one’s sway over the rest of the population by force, fatigue, or enticements. hrough tribal allegiances, eco- nomic interests, and other clientelist networks, Assad has been able to keep local Sunni Arabs under control during the war and return key territories to the regime’s bosom. Examining how he was able to do so—and how his late father established minority control before him—is central to understanding how he might go about trying to reconquer the rest of the country despite facing long demographical odds. Accordingly, this chapter focuses on Assad’s strategies in the ive main sectors of regime-held Syria: Damascus, the Alawite coast, Jabal al-Druze, the central Homs-Hama area, and Aleppo. dAMASCuS rEMAINS ThE kEy he phrase “Whoever holds Damascus holds Syria” has been attributed to Haiz al-Assad, and his son has adopted the same view. he regime’s elite troops have defended the capital well throughout the war, and while Sunni rebel forces seized the suburbs early on, they were never  31  able to approach the central districts. he midan quarter was temporarily occupied by the rebel Free Syrian Army in summer 2012, but Assad’s forces quickly regained control. Although al-midan is a Sunni neighborhood, it is largely populated by middle- and upper-class families who seemingly did not feel any solidarity with the rebels. Anecdotal observations gathered by the author in April 2011 and October 2013 hinted that this social cleveage was present in many Syrian cities. In Latakia, for instance, a middle-class Sunni friend declared that he did not want to march in the streets with people from the city’s poorest Sunni neighborhood, Ramel al-Filistini, where the protests began. more important, the geography of the revolt has continually shown that social divisions can be stronger than sectarian solidarity among urban Sunnis.1 Since march 2013, the army has recaptured most of the lost territory near Damascus, conining the rebels to enclaves that have been inexorably reduced. Holding the capital gives Assad some of the internal and external legitimacy he needs to cast himself as Syria’s leader. Damascus is a multiethnic city where each of the country’s sectarian communities is represented, allowing the regime to maintain contact with the various networks that structure Syrian society. Half a century of political, administrative, and infrastructural centralization has made Damascus the main hub for all of these networks, and Assad fully understands this dynamic.2 Urban Planning to Control Insurgency Since the 1970s, the army has exerted considerable territorial inluence on the capital and surrounding region. Large military camps occupy southern and western Damascus, including one that hosts the Republican Guard, an elite corps of 30,000 mostly Alawite troops. Oicially, the regime claims that this large-scale military presence is necessary to protect the capital from Israel, since the Golan Heights is less than ifty kilometers away. Yet the real objective was well understood by the strategy’s architect, Haiz al-Assad: namely, to better control an area that he regarded as indispensable to holding all of Syria. Upon his ascension to power, he reinforced his grip on the city by installing tens of thousands of Alawite soldiers and oicials there, along with their families. Damascus thus became the irst heavily Alawite city in Syria. In 1947, the capi32  tal’s population of 600,000 included only 300 Alawites.3 By 2010, they comprised more than 500,000 of the city’s ive million inhabitants, which meant that a quarter of the country’s entire Alawite population resided in the greater Damascus area.4 he current distribution of communities around Damascus best illustrates the regime’s system of control. Alawites and other religious minorities occupy areas of critical importance. Senior regime oicials live in malki, around Assad’s private residence. Some lowerlevel civil servants live in mezzeh 86, a large urban district that spans several aluent neighborhoods, while others reside in small Druze/Christian towns in the suburbs (Jdeidat Artouz, Jaramana, and Sahnaya). he Druze and Christian lifestyle is more compatible with that of the Alawites (who allow alcohol consumption, unveiled women, etc.) than the conservative Sunni localities of the Ghouta district (e.g., Douma, Daraya, Zamalka), which became rebel strongholds early on. Before and during the war, the Alawite and Druze/Christian districts were permitted to extend their boundaries close to the strategic points that connect Damascus with the rest of the country, in the process cutting the Sunni areas of from each other and the rest of the rebellion. his was the case in Jaramana, the large Druze/Christian suburb that developed along the route to the inter- national airport. he regime’s apparent willingness to allow Jaramana to expand corresponds with its strategy of separating the Sunnis in the East and West Ghouta neighborhoods with a non-Sunni bloc. his was useful in preventing the rebels from encircling the city and cutting it of from the airport. In the northeast, public housing in Dahiyat al-Assad, maarat mahmoud, and Adra is home to mostly non-Sunni industrial public-service employees who constitute a loyalist bulwark for that part of the capital. In the southwest, the density of military camps and Druze/Christian localities helps protect the roads to Beirut, Quneitra, and Deraa. As for the Sunni localities of moadamiya, Daraya, and Babila, they are bordered to the south by the Druze/Christian belt, a large population of Alawite military and civil servants, and a ring road that was built not only to ease traic, but also as a line of defense against rebellious suburbs. Other aspects of the city’s planning it this strategy as well. he Damascus municipality is surrounded by a wide peripheral boulevard and lanked by wide avenues that create ruptures in the urban space. hrough this “anti-insurgency urban planning,” the regime likely sought to optimize the movement of armored vehicles, which would in turn discourage large street demonstrations. he old city of Damascus was a victim of this strategy in the late 1970s, as parts of the old souks were razed to make way for a new shopping district with wide streets. he regime’s error was to allow the growth of informal suburbs, which created a labyrinth of narrow streets more conducive to demonstrations. In addition to the obstacles posed by the regime’s urban planning, the rebels also failed to make inroads into downtown Damascus because much of the local Sunni population was, for social and economic reasons, not supportive. As mentioned previously, many middleand upper-class residents there were hostile to the rebels, who tended to hail from lower-income neighborhoods.5 he old alliance between Haiz al-Assad and members of the capital business class was reinforced by Bashar’s economic liberalization policies, so they seemingly had little desire to revolt. Only al-midan district, a former muslim Brotherhood stronghold, rose up in summer 2012, but the army quickly retook it. Turning the Tide in Damascus, 2012–2016 When the national security headquarters in Damascus was bombed on July 18, 2012,6 killing several prominent regime igures, the rebels seemed close to seizing the capital, and Western oicials were predicting that the regime would collapse in a matter of days or weeks. By 2016, however, the local military situation had been completely reversed—the remaining rebels in the area were surrounded by army forces and allied Shia  FABrICE BAL ANChE 20. SYRIAN ARMY PROGRESSION IN DAMASCUS AREA, March 2013–May 2017 vers Alep Kafr al-Awamid vers Palmyre Ain al-Fija Dimas Barada River AL-TALL Dahiyat al-Assad Masaken al-Dimas vers Beyrouth Adra Maarat Mahmoud DOUMA Barzah Qura al-Assad Hawsh Mubaraka Harasta QUDSAYA Mayda al-Qabun Rukn al-Din Dummar al-Balad DAMASCUS Arbin Jobar Hawsh al-Shalaq Beit Siwa Zamalka Yafour al-Nashabya Douwayla Mezzeh Jarba al-Abbadah al-Midan Yarmouk Moadamiya al-Bahariya Utaya Saqba Marj al-Sultan Jaramana al-Mulayhah Qaysa Babila Deir al-Asafir DARAYA QATANA Sayyida Zaynab Jdeidat Artouz Harran al-Hawamid Hatite al-Turkman Sahnaya Artouz al-Ghezlaniya N Khan al-Sheikh W al-Kiswah vers Quneytra 0 10 km E S vers la Jordanie PROGRESSION Urban area Province center Main road March 2013 Rebel held Industrial zone District center Secondary road July 2016 Ceasefire Rural area Town Airport Military base Damascus municipality May 2017 S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r   33 To Aleppo AL-TALL Adra Maarat Mahmoud Adra industrial area BRIGADE 39 21. THE SYRIAN ARMY SURROUNDS EAST GHOUTA DOUMA Barzah PROGRESSION Hawsh Nasri Harasta Mayda al-Qabun March 2013 Hawsh al-Shalaq Arbin Nov 2014 June–July 2016 Bayt Siwa DAMASCUS Jobar Utaya Zamalka Nov 2016 –Jan 2017 Saqba Old City Rebel July 2016 Local agreement May 2017 al-Bahariya Jarba al-Nashabya Hawsh al-Malban Spring 2013–Summer 2014 East Ghouta surrounded Syrian Army old offensive al-Abbadah Syrian Army ongoing offensive Jaramana al-Mulayhah Jan–Aug 2014 Yarmouk Babila Marj al-Sultan Marj al-Sultan military airport al-Otaiba Dec 2015 Qaysa May 2016 Urban area Rural area Deir al-Asafir Industrial zone al-Shahba Military base Sayyida Zaynab Province center Haran al-Awamid Hatite al-Turkman District center Town Damascus Int'l Airport BRIGADE 158 Damascus municipality N al-Ghezlaniya W Highway E Principal road S 0 10 km Airport To Suwayda 34   FABrICE BAL ANChE 22. SYRIAN ARMY DAMASCUS STRATEGY, May 2017 To Saidnaya To Homs To Palmyra Kafr al-Awamid Ain al-Fija Minority stronghold AL-TALL Barada River To Beirut Military base Maraat Mahmoud Dimas Adra Dahiyat al-Assad Syrian Army offensive since Spring ‘16 Masaken al-Dimas DOUMA Barzah Qura al-Assad QUDSAYA Jobar Dummar al-Balad DAMASCUS Mezzeh 86 Ceasefire Zamalka Saqba YA Jarba al-Nashabya Strategic road al-Abbadah Road closed Marj al-Sultan al-Mulayhah K OU Syrian Army progress, July '16 to May '17 al-Bahariya Utaya Douwayla RM DARAYA Hawsh al-Shalaq Arbin Bayt Siwa Jaramana Moadamiya Rebel held Mayda al-Qabun Rukn al-Din Yafour Hawsh al-Mubaraka Harasta Qaysa Frontline Babila Urban area Deir al-Asafir Rural area Sayyida Zaynab QATANA Jdeidat Artouz Haran al-Hawamid Hatite al-Turkman Industrial zone Sahnaya Airport Artouz Province center al-Ghezlaniya District center Town Khan al-Sheikh al-Kiswah To Quneitra militias, and they had lost hope of being rescued by Western-backed ighters to the south, whose focus had shifted from the Assad regime to the Islamic State. One of the main reasons for this reversal was robust support from Iran and Hezbollah, who have treated Damascus To Deraa and especially its airports as a strategic gateway for transferring weapons to the Lebanese militia. heir willingness to commit Shia forces to the capital is also explained by the presence of the Sayyeda Zainab shrine, a major Shia holy place that attracted hundreds of S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r  0 10 km Damascus governorate To Suwayda thousands of pilgrims annually before the war. Although the population of the shrine’s neighborhood is predominantly Sunni, the local Shia minority has been able to keep the district in the regime’s hands with major help from Hezbollah, which deployed numerous forces there early on. Whenever a rocket falls or car bomb explodes anywhere near the district, the news reverberates throughout the Shia world, helping Iran and Hezbollah mobilize new ighters by playing on widespread fears that the shrine will be eradicated (much  35 like the prominent Samara mosque in Iraq was destroyed by Sunni extremists in 2006). Another major reason for the opposition’s setbacks in the capital was its inability to unite the two parts of the Ghouta and cut of the road to the airport. When the rebels irst went on the offensive, Jaramana was strongly defended by the army and local Druze members of the National Defense Forces militias, and the population stayed put amid rebel car bomb attacks and rocket assaults. Once the tide turned, the army used Jaramana as a launching point to expand its grip on both sides of the airport road and encircle the Sunni parts of the Ghouta. he regime’s Ghouta siege has at times been accompanied by food blockades and aerial bombardment intended to scare surviving civilians into leeing— the same strategy applied in Aleppo. he campaign against Daraya was exemplary in this respect. In 2012, the southern suburb was home to around 80,000 people, but by August 2016—when the rebels negotiated the evacuation of the last ighters and civilians to Idlib—the number had fallen to 4,000.7 One objective of the regime’s brutal campaign was to make an example out of Daraya in order to facilitate a modus vivendi with localities where rebels were less aggressive. Babila, moadamiya, Qudsaya, al-Qabun, and Barzah all agreed to ceaseires in order to avoid total de36  struction. Indeed, locals in Qudsaya asked the rebels to leave8 so that they could spare themselves further sufering. Once the ighters were evacuated, some residents were granted amnesty and integrated into a local police force. he army then air-dropped lealets over other opposition-held areas instructing to them choose between the Daraya and Qudsaya formulas. One rebel who departed Qudsaya described this stark choice: “Finally, they literally said to us, ‘Either you get the hell out of this town or we completely destroy the place’... hey literally said that. hey’d destroy the town and then we’d have to leave anyway, like in Daraya. So a group of us decided we’re better of leaving, we didn’t want the town to be destroyed.”9 In all, the army was able to regain about half of East Ghouta from April to December 2016. hese victories were facilitated by a fratricidal war between the rebel groups Jaish al-Islam, Failaq alSham, and the Fustat Army (led by alQaeda ailiate Jabhat al-Nusra).10 he irst group had previously exercised quasi-hegemony over East Ghouta, but the death of founder Zahran Alloush in December 2015 weakened the militia, and his brother mohammed failed to live up to local and international expectations upon succeeding him. mohammed was quickly marginalized at the Geneva peace talks in January–march 2016, and the political and military opposition lost whatever unity it had once cultivated. Assad Now Sure of His Power in the Capital Area Today, services are functioning normally in the Damascus city center, and if not for the sound of artillery shelling against rebel holdouts from mount Qasioun, the war would likely seem nonexistent to most residents. he international airport has been back in operation since 2015, and the roads to Homs, Deraa, Suwayda, and Beirut are relatively safe. In opposition-held East Ghouta, the army is using intra-rebel divisions to gradually gain ground, alternately bombing neighborhoods and providing humanitarian aid in order to win over the local population. he regime is also well aware that the crossing points into the Ghouta are a signiicant source of revenue for whichever rebel faction controls them, so it has sometimes favored one faction to provoke clashes with others, as in the spring 2016 conlict between Failaq al-Rahman and Jaish al-Islam. Going forward, the regime’s negotiations with Jaish al-Islam will likely intensify as pressure on the group mounts. he surrender of Douma, the de facto rebel capital in the Ghouta, would mark the end of the enclave. hese and other developments in Damascus can only boost Assad’s conidence, even though much of the country remains out of his control and his army is still struggling to preserve the gains that Russian air power has made possible in other areas. It would therefore be diicult for the international community to push him out of power under the current circumstances. his would be true even if moscow were willing to step aside and permit his ouster, since Assad does not need the Russian air force to hold the capital. ThE ALAwITE STrONghOLd Syria’s coastal Alawite region has been staunchly loyal to Assad since the beginning of the war, and for good reason. Although some Alawite intellectuals joined the opposition and sparked challenges to the Assad family, the bottom line remains the same: most Alawites see the war as an existential ight against a Sunni Islamist threat. he Alawite region stretches from the Turkish border south to the Lebanese border, and from the mediterranean Sea east to the Orontes River, more or less corresponding to the former Alawite state created by France in 1920. In 2011, the local population was about two million, of which Alawites comprised 70%, Sunnis 20%, Christians 10%, and Ismailis 1%. he Sunnis mainly reside in the cities and in three large rural enclaves: Jabal al-Akrad, Jabal Turkman, and south of Baniyas. he majority of Christians have settled in the cities of Latakia, Tartus, and Saita. Wadi al-Nasara, near the historic Crac des Chevaliers, became the main rural  FABrICE BAL ANChE home for Christians in Syria, particularly after many began to lee Homs during the war. At the Turkish border, Armenian villages cling to the slopes of Jabal al-Akrad; one of them, Kasab, is among the last vestiges of Western Armenia, and it beneits much from the support of the Armenian diaspora. As for the Ismailis, they are concentrated in al-Qadmus and masyaf and live on good terms with the Alawites. memories of the wars that ravaged the area in 1920 have faded; the Ismaili population is no longer large enough to threaten the Alawites, and they have found common ground in their shared animosity toward Sunni Islamists. Since the war began, the Alawite region’s population distribution has been modiied somewhat by the arrival of internally displaced persons from Aleppo, Idlib, and Damascus. many Alawite and Christian families have returned to their villages of origin, along with signiicant numbers of non-hostile Sunnis. he UN Oice for the Coordination of Humanitarian Afairs estimated that as of fall 2016, about 600,000 IDPs lived on the coast.11 Indeed, Latakia, Tartus, and Saita are experiencing a construction boom due to high demand from new homeowners. Yet the inlux of Sunnis might destabilize the Alawite region over time if it gives terrorist operatives cover to iniltrate the area. he Islamic State conducted a series of bombings there S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r  in may 2016 (causing 170 deaths in Tartus and Jableh) and September 2016 (50 dead). Such attacks could eventually spur reprisals against Sunni refugees. 23. LATAKIA SECTARIAN DISTRIBUTION 2011 To Kassab and Turkey N W E To Mashqita Demsarkho S Datur To Ibn Hani Besnada Ramel al-Shamali Hamam Knenas Bustan Rihan Mediterranean PORT AREA al-Jumhuriyah Mashru al-Reji To Damascus Governorate Port Sheikh Dahir Zyraa Mashru al-Thani University To Damascus al-Sleibeh Railway Tobacco station factory MILITARY AREA al-Tabiyat INDUSTRIAL AREA Ramel al-Filistini Ramel al-Janubi Kornich al-Janubi 0 1 km Sunni Christian Cathedral Great Mosque Latakia, 1936 Alawi Unpopulated Church Mosque Street Mixed he Latakia Corridor he jihadist threat has moved further away from Latakia since the recapture of Jabal al-Akrad in winter 2015–2016, but the city and surrounding areas are not immune to sectarian conlict. In 2010, the city’s population of around 400,000 was roughly 50% Alawite, 40% Sunni, and 10% Christian (mostly Orthodox).12 Alawites occupy the northern and eastern suburbs, while Sunnis live downtown and in the southern suburb of Ramel al-Filistini, the city’s poorest area. Christians inhabit the “American district,” named after the local Evangelical Presbyterian school established by Americans. In this historically Sunni city, old urban residents still view Alawites as foreigners. Up until the French mandate, the city had no Alawite residents apart from household servants. more than two decades later, Alawites still constituted only 10% of the population, concentrated in the poor suburb of Ramel al-Shemali. Beginning in the 1970s, however, Haiz al-Assad’s “Alawitization” policy facilitated a dramatic demographic shift, and his sect became the majority in Latakia by the 1980s. he surrounding countryside is likewise divided between Sunni and Ala 37 wite villages. Traveling north toward Turkey, one inds that Sunnis are the majority; according to the 2004 census, they composed about 80,000 of the area’s 140,000 residents. he subdistricts of Rabia and Qastal maaf are mostly Sunni Turkmen, as are the coastal villages of Burj Islam and Salib al-Turkman. When the war began, the Turkmens of Rabia and Qastal maaf sided with the rebels, while those in Burj Islam and Salib al-Turkman remained neutral, likely because they were surrounded by Alawite villages. Sunnis also dominate the area east of Latakia, the northern part of Jabal alAnsariyya (including al-Hafah and its surrounding villages), and Jabal alAkrad. Although residents in the latter area have been of Kurdish origin since the middle Ages, none of them speak Kurdish anymore, and the area is considered essentially Arab. According to UN igures13 and personal communications with a municipal oicial in Latakia, 200,000 IDPs had led to Latakia city as of October 2016 (mostly Sunnis from Idlib and Aleppo), while another 170,000 went elsewhere in the province. In general, Alawite IDPs have led to Alawite villages and Sunnis to Sunni areas. Yet Jabal al-Akrad and Rabia were swept up in the ighting for a number of years, so most of the local Sunni women and children have led to Turkey by now, while most of the men joined rebel groups. 38  Beginning in spring 2012, the opposition seized control of Jabal al-Akrad and the area along the Turkish border up to Kasab village. In march 2014, jihadist groups operating from Turkey and Idlib province invaded Kasab and destroyed the Russian radar station atop Jabal Aqra. But they were unable to progress further southward, so they left the village that June.14 Similarly, rebel forces from Jabal al-Akrad briely occupied al-Hafah in July 2012, but the local population did not join them for fear of provoking regime retaliation. Despite these failed ofensives, the northwestern rebel zone continued to pose a real threat to regime control in Latakia. Accordingly, Assad created a new militia in 2015 called the “Shield of the Coast,” whose mission is to protect the area using young Alawites who refuse to ight outside their home province. he regime no doubt felt the need to protect “Alawistan” from ofensives like the one launched in July–August 2015, when rebels from al-Ghab plain advanced closer to Latakia and the underpopulated Alawite villages in northern Jabal al-Ansariyya. If the regime had failed to stave of such threats, Alawite soldiers could have taken matters into their own hands and renounced support for Assad. In winter 2016, however, the army regained the advantage in the Latakia area thanks to intense Russian air bombardment, and the rebels were driven out of the northeast coastal region. At some point the army will likely aim to retake Jisr al-Shughour, a city from which rebels have launched multiple attacks on the Alawite region in the hope of reaching the coast. A Possible Alawistan For the rebels, the (increasingly dim) prospect of gaining access to the sea is both strategic and symbolic. In march 2014, for example, Jabhat al-Nusra advanced from Kasab to the mediterranean in just a few days. Today, the group and its allies would very much like to control a major port such as Latakia or Tartus. Although opposition forces have since been pushed further east, any future rebel ofensives toward the coast would stand a better chance of success in Latakia than in Tartus. First, the road to Latakia is more accessible. Second, Alawites constitute 80% of the population in Tartus, compared to only 10% Sunnis and 10% Christians. he population between Tartus and Homs is predominantly Alawite as well, with a strong Christian minority; Sunnis are concentrated in the countryside around al-Hamidiyah and Talkalakh. hird, Lebanon’s proximity would complicate any jihadist push for Tartus, since Hezbollah and the Syrian army are stily controlling that frontier to prevent any cross-border Sunni coordination—a strategy exempliied by the may 2013 battle for al-Qusayr. To be sure, Iran and its Shia proxies regard the Latakia corridor as less of a strategic interest than Homs, Damascus, and the Golan Heights, so it is unclear how eager they would be to defend Tartus from any future rebel ofensives. Yet moscow has a strong interest in maintaining a presence along the coast; the Russian navy has a base in Tartus and plans to rebuild the former Soviet submarine base in Jableh twenty miles south of Latakia. As described previously, the Russian intervention and other developments have greatly diminished the prospects of the regime losing power in Damascus, yet it remains a distant possibility if geopolitical circumstances change. In that scenario, the regime could decide to retreat to the coast and form an Alawite statelet. Beginning in the 1970s, Haiz al-Assad built the area into a bunker where Alawites could take refuge if they lost power in the capital. his arrangement has proven particularly useful during the current war, since Alawite military personnel would not have agreed to continue ighting in Deir al-Zour, Damascus, or Aleppo if their families were not safe back home on the coast. In the event of regime collapse, the Alawites could bunker in their stronghold and perhaps extend it eastward depending on their capacity to retain portions of al-Ghab plain and the Homs countryside. Jabal al-Akrad and Jabal Turkman could serve as a further bufer, since they  FABrICE BAL ANChE N W 24. Idlib TURKEY FROM ALAWITE STRONGHOLD TO STATELET E S Kasab To Aleppo To Antakya Ariha Jisr al-Shughour TERRITORY Russia Burj Islam al-Haffah Salma Maarat al-Numan INFRASTRUCTURE Sunni Highway Christian Principal road Alawite Power station Jabal al-Zawiya Latakia Refinery Qardaha URBAN MAJORITY al-Suqaylabiyah Jableh Jabal al-Ansariyya Port Sunni al-Ghab plain Airport Baniyas Christian Hama Maharda Mediterranean Sea al-Rawda Alawite Ismaili International border Mixed Province border Masyaf Sheikh Badr al-Sawda Tartus Rastan Duraykish Safita POPULATION Wadi al-Nasara Tal Kalakh al-Hamidiyah OFFICIAL BORDERS Akkar plain POSSIBLE BORDERS 200,000 to 500,000 Pessimistic scenario 50,000 to 200,000 Optimistic scenario Less than 50,000 Frontline (May 2017) Homs Iran al-Qusayr Potential foreign aid LEBANON 0 To Baalbek 20 km Hezbollah S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r  To Damascus  39 25. JABAL AL-DRUZE AND THE SOUTHERN FRONT, March 2017 Mezzeh Mt Hermon Qatana Nov 2014 battle: JN and Druze militia Ruwala 2012–13: Car bombs, rebel bombardment Jaramana Sayyida Zaynab Sahnaya Jdeidat Artouz Arneh DAMASCUS Daraya Makhul Jun 2016: Rebels seize from IS Damascus Int’l al-Abdullah Marj Ruhayyil Hader Sep 2016: JN attacks Rebel route between Syria and Lebanon Apr 2016: IS attacks al-Ghiati MADINAT AL-BAATH Rebel smuggling route from Jordan Khalkhalah Quneitra UNDOF crossing Islamic State Leja base Sanamin Mar 2017: Syrian army seizes from IS Dama Tal Barakat Jun 2016: Syrian army seizes from IS Aug 2014: JN clashes with Sheikh Balous al-Hassan Izraa al-Shahba Yarmouk Martyrs Brigade JABAL AL-DRUZE al-Muzairab SUWAYDA al-Thaala 21 June 2016: Israeli airstrike on IS antiaircraft battery Jun 2015: Heavy JN attacks Mar 2017: Rebel offensive in Deraa DERAA Nasib Apr 2015: Rebels seize from Syrian army Busra al-Sham Salkhad Mar 2015: Rebels seize from Syrian army 0 20 INHABITED AREA OF CONTROL 40 km UNINHABITED AREA OF CONTROL RELIGION International border Province capital Syrian Army Syrian Army Sunni Province border District center Islamic State (IS) Islamic State (IS) Druze Golan DMZ Town Jabhat al-Nusra (JN) and other rebels Jabhat al-Nusra (JN) and other rebels Christian Border crossing Main road Shiite Military airport Secondary road Israel Notable event al-Hassan Arab tribe Informal route Commercial airport 40   FABrICE BAL ANChE were practically emptied of their civilian population during years of ierce ighting with the rebels. In theory, this “Alawistan” could extend over 12,000 square kilometers, including the provinces of Latakia and Tartus, the districts of masyaf and alSuqaylabiyah, and the western part of Homs province. It would be reduced to 6,000 square kilometers if the regime lost the mixed Alawite/Sunni territories north of Latakia, Homs, and al-Ghab plain. In the irst case, the population would be about 3 million; in the second, 2.5 million (since many Alawite, Christian, and Ismaili families from central Syria would likely lee there).15 Regime victories over the past year have made the Alawistan scenario increasingly unlikely, and Assad is still focused on retaking the entire country, not building a sectarian statelet.16 Yet the coastal bunker will remain a fallback option as long as he has Russia’s support. JABAL AL-druZE: CAughT IN ThE MIddLE he mountainous Jabal al-Druze region occupies a strategic position on the road to Damascus, so Assad has sought to keep its inhabitants in his fold despite their uncertain loyalties when the war began. he Sunni Arab opposition has made his job much easier by repeatedly mishandling and attacking the local Druze. Protecting the Capital’s Southern Flank In 2010, Syria’s Druze numbered around 700,000, or 3% of the national population. most of them lived in Suwayda: 90% of that province’s 375,000 inhabitants were Druze, 7% were Christian, and 3% Sunni. Another 250,000 resided in Damascus and its suburbs of Jaramana, Sahnaya, Ashraiyat Sahnaya, and Jdeidat Artouz. he remainder lived in mount Hermon (30,000) and the Jabal al-Summaq region of Idlib province (25,000 people spread over fourteen villages).17 With the exception of Jabal al-Summaq, most of these Druze areas are points of strategic support for the army. Jaramana played a key role in defending Damascus city at the height of rebel campaigns there. he Druze villages of mount Hermon have been a loyalist bastion on the capital’s southeastern lank, allowing the army to maintain lines of communication with the Golan Heights and prevent isolated rebel pockets from linking up with larger opposition-held zones. And the Jabal al-Druze area has given the regime an enduring land link to Jordan, as well as a launchpad for aircraft crucial to the defense of Deraa and a means of continually reminding rebels in the southwest that they are under threat of attack. Defending Jabal al-Druze requires little of the army; most of its protection is entrusted to a local militia of 10,000 S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r  men (mostly Druze with a few Christians) who know the terrain and are more motivated to defend their hometowns than ight for the regime in far-of Aleppo or Homs. In fact, keeping Druze conscripts inside Suwayda province rather than deploying them to other fronts was part of the tacit bargain that Assad reached with the community’s spiritual leaders (known as the Sheikhs of Akel) in order to maintain control over them.18 Sunnis vs. Druze As early as 2011, tensions emerged between the Druze and certain rebel factions. In Deraa, armed and unarmed opposition factions began using antiDruze slogans, describing them as heretics. One slogan even called Assad “son of the Druze,” a local pejorative. In a climate of growing Islamist incitement, some Druze villagers were abducted and then ransomed or murdered. hese included leading Druze dignitary Jamal Ezzeddine, who enjoyed good relations with the Deraa opposition; he and sixteen of his companions were abducted by Jabhat al-Nusra in December 2012 and killed a few months later. Similarly, Druze residents in Jaramana fell victim to assaults beginning in spring 2012. hat autumn, Jabhat al-Nusra led a rebel ofensive in Jabal al-Druze.19 In response, the Druze set aside their neutrality and formed a local militia with the regime’s help. From this point on, Jabal al-Druze was regularly attacked by Sunni Arab forces. In August 2014, rebels fought a serious battle against the Druze in al-Dana. One of the most active regime militias in that clash was led by Sheikh Balous, a former Druze policeman who had been very successful at recruiting local youths unwilling to ight outside Jabal al-Druze.20 And in June 2015, around the same time as a major opposition ofensive in northwestern Syria, rebels from a group called the Southern Front attempted to seize al-haala military base, the key to Suwayda’s defense. hey were pushed back by the army and local Druze militias—a huge turning point given that Druze cooperation with Assad would have been unimaginable in past years (e.g., in November 2000, the regime cracked down on a revolt in Suwayda, killing dozens of Druze and wounding hundreds more). hese rebel actions may seem baling given the Druze posture at the start of the uprising. Like other parts of Syria, Jabal al-Druze was the scene of signiicant anti-regime protests in 2011, echoing the region’s traditional air of deiance (e.g., as described in chapter 1, many inhabitants still brandish the proud memory of Sultan al-Atrash, the Druze leader who revolted against the French in the 1920s). Yet various opposition bodies, including the Syrian National Coalition, failed to meet the Druze community’s expectations, particularly the demand for secularism. Rather than  41 calling for a “secular state” following Assad’s ouster, the SNC instead adopted the idea of a “civil state,” which in Islamic discourse refers to governance by sharia, not secular law. Such ideological nuances exacerbated Druze mistrust of the Sunni Arab opposition, whom the regime had already portrayed as taking inspiration from medieval fatwas calling for genocide against the Druze. Even so, some Druze still decided to join the rebels at irst. In August 2011, Druze oicer maj. Khaldun Zeineddine deserted the army and created an anti-Assad armed group, the Sultan alAtrash Brigade.21 Although the group joined other rebel factions in Deraa and participated in several ofensives against regime forces in the Jabal al-Druze area, it was unable to recruit many Druze and eventually drew the ire of Jabhat al-Nusra. In 2013, members of the alQaeda ailiate arrested the Druze brigade’s members and sentenced them to death; they were eventually released and led to Jordan. his incident showed the Druze community in starkest terms that they were not welcome in the Sunni Arab rebellion. For those Druze who had joined the ighting, the jihadists did not consider their commitment to be sincere enough; in some cases they were threatened with death unless they converted to Sunnism (e.g., the Druze of Jabal al-Summaq were forced to convert in 2015, then destroy their own mausoleums22). 42  he Secessionist Temptation of Sheikh Balous Despite being rejected by the rebels, the Druze were initially hesitant to link their fate to Assad, since his fall would leave them defenseless. In mid-2015, at a time when the army seemed about to collapse, some Druze began to consider the option of opening Jabal al-Druze up to Jordan and making it an autonomous region under international protection. his idea was championed most prominently by Sheikh Balous, who had established one of the irst pro-regime Druze militias in 2012. He made his mark in 2014 during the battle for alDana, defending Druze territory from rebels. Afterward, he asked Assad to provide him and other militias with heavy weapons to defend Jabal al-Druze more efectively. He also began to focus on political issues, calling for reforms to address the high cost of living, corruption, and Druze conscription. By June 2015, his militia had increased to nearly 1,000 ighters,23 and he was reportedly receiving inancial aid from abroad, especially from Israeli Druze who were concerned about the fate of their Syrian coreligionists. Yet when Jabhat al-Nusra led the ofensive to seize al-haala military airport that month, his forces did not help defend the regime; instead, he called on the local population to seize army positions and government buildings. his gambit failed—three months later, Sheikh Balous was assassinated under mysterious circumstances, and his militia was dissolved. here is little doubt that the regime had him murdered for becoming too ambitious and encouraging the Druze to secede. he reasons for his move at al-haala are unknown. He may have believed that Assad’s army was on its way out of Jabal al-Druze given the successful rebel ofensives in Idlib province and around Deraa (e.g., the key southern town of Busra al-Sham had fallen that march). If so, he may have felt that the regime’s supposedly imminent fall might pave the way for him to become the leader of a Druze safe zone. Whatever the case, Russia’s intervention soon after his death completely changed the balance of power on the ground, so the Druze are unlikely to gamble on further secessionist attempts anytime soon. It will therefore be diicult, if not impossible, to detach Jabal al-Druze from the regime. he Druze will not be won over unless they are cut of from Damascus, and even then they would need very concrete assurances that international forces will protect them from Sunni Arab jihadist groups and spare them the fate of their co-religionists in Jabal al-Summaq. CENTrAL SyrIA: A SECTArIAN MOSAIC At the center of the regime zone lies a mixed bag of Christian, Alawite, Sunni, Ismaili, and Shia communities. his diversity is linked to the area’s complex history on the margin of Syria’s steppe, where nineteenth-century agrarian colonization attracted non-Sunnis from the coastal mountains and Shia from Lebanon’s Beqa Valley to villages near Homs. Christian communities in the area date back to pre-Islamic Syria; they were able to preserve their identity because their land lay next to swamps and was deemed too poor to interest muslim and Ottoman conquerors. he central region is dominated by Homs and Hama, rivals for centuries before the creation of modern Syria.24 Hama, a conservative Sunni city, dominated a vast countryside, and its residents mainly lived on land rent until Syria’s 1963 agrarian reform. In contrast, the cross-sectarian city of Homs invested more in industry and commerce beginning in the nineteenth century, so it was able to continue prospering under Baath land reforms a century later thanks to its diversiied economic base. Yet the new masters of Syria sought to upend the great landowning class of Hama, who had exploited their families for generations. Once these owners were deprived of land rent, they were quickly marginalized by the Baath regime. Hama’s muslim Brotherhood revolt in 1982 and the ierce repression that followed were a prelude to the bloody conlict Syria is experiencing today, with most combatants divided along urban/rural and minority/ Sunni lines.  FABrICE BAL ANChE Why Did Hama Stay Calm While Homs Ignited? When the uprising began in 2011, some demonstrations took place in Hama, but local Sunni Arab leaders quickly put them to an end. And unlike in crosssectarian Homs, no armed insurrection developed there. Two factors explain this relative calm. First, the memory of the 1982 massacre is still very present in Hama. In response to that revolt, Assad’s father killed at least 20,000 residents and completely razed the city center, so locals knew what to expect from the regime if they took up arms again. Apparently, their fear of Assad’s retribution outweighed their anger toward his repression. Second, the sectarian homogeneity of Hama’s population and bureaucracy seemed to defuse tensions. Although Alawites have a strong presence in the local civil service, many of the city’s top municipal and regime posts are occupied by native Sunni Arabs, not minority outsiders. Hama also lacks a distinct Alawite neighborhood—the city is largely unappealing to non-Sunnis, and those Alawites who do hold government posts there often prefer to reside in their towns and villages of origin when not performing their local duties. When the regime destroyed downtown in 1982, it erected a ive-star hotel on its ruins as a symbol of state power and modernity, but Haiz al-Assad did not seek to transform the city’s conservative Sunni roots. moreover, Hama’s subsequent political and economic marginalization deterred outsiders from moving there, which limited both the possibility of intercommunal friction and the visibility of regime favoritism toward Alawites. During the war, rebels have tried to seize the city several times. As of this writing, forces from Hayat Tahrir alSham (the radical coalition headed by Jabhat al-Nusra) are only about ten kilometers north of Hama, but they face a dense network of military bases built around the city’s perimeter in the 1980s and a western countryside full of Alawite, Christian, and Ismaili villages. But the regime’s greatest asset in Hama remains the population’s relative passivity throughout the war—an advantage it certainly has not enjoyed in Homs. Like most of central Syria, Homs is home to a mosaic of sects. As of 2011, its population was around 65% Sunni Arab, 25% Alawite, 10% Christian, and less than 1% Ismaili and Shia.25 As in the coastal cities, Alawites came to Homs from the former French mandate and settled in outlying districts to the south and east. Christians have a historic district in the city center from which they have spread eastward, reinforced by their brethren in rural areas. he city’s western half is wealthier than its eastern half, which is largely occupied by rural people from the margins of the steppe. he western half is also more of a religious mix, while the rest of the city’s S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r  neighborhoods tend to have more assertive community identities. In July 2011, clashes broke out in Homs and caused several dozen deaths. In most every case, these incidents took place on the borders between Alawite and Sunni districts. he army soon deployed to these hotspots, surrounding the northern Sunni neighborhoods that would become the rebellion’s stronghold in Homs. his resembles what michel Seurat wrote about Tripoli during the Lebanese civil war—clashes in that city did not take place within aluent neighborhoods, but rather between Sunni and Alawite districts where community endogamy and identity were strong.26 Initially, Homs rebels enjoyed strong external support, especially from Lebanon, their principal source of foreign weapons and ighters. he Bab al-Amr district, which rose up militarily against the regime, was known as a haven for smugglers operating between Syria and Lebanon before the war. hese smugglers often traveled through tunnels that begin several kilometers out in the countryside, bringing goods and weapons into the city while avoiding police checks. French journalist Edith Bouvier used one such tunnel to escape Homs in 2012, describing it as high and wide enough to ride a motorcycle through.27 he large Sunni Arab community of Tripoli was particularly supportive of the rebels next door. Saudi and Qatari networks used the Lebanese city as a base for organizing aid to the insurgents, and wounded rebels found relief at Gulf-based Islamic NGOs there.28 Yet the Gulf networks stopped operating in Tripoli after the Assad regime attacked rebel-held Homs in force and encircled the city in fall 2012. meanwhile, the Lebanese army and Hezbollah eventually took control of the border and prohibited further aid transfers to the rebels after the may 2013 fall of al-Qusayr, a former opposition stronghold on the northeastern frontier. Sectarian Slaughters As the war gained momentum in 2011, central Syria was racked by intercommunal confrontations. In Homs, demonstrations against the regime stopped at the edge of Alawite districts because clashes tended to multiply rapidly there. In the city’s suburbs, the Shia village of mazraa was regularly attacked on Fridays by local Sunni opponents.29 And when rebels seized the central districts, they ired mortars at the Alawite neighborhoods daily. Elsewhere, Christians and Shia led the nearby town of alQusayr in autumn 2011 after sectarian threats were broadcast from the loudspeakers of local mosques. Amid this tense climate, sectarian slaughters began to occur. One notorious incident took place in Houla on may 25, 2012, when 108 Sunni Arab civilians (including 49 children) were murdered by the regime’s shabbiha para 43 military.30 A week earlier, a dozen Alawite state employees had been massacred en route to Homs. In June of that year, 80 people were killed in the Sunni village of mazraa al-Qoubeyr; a month later, another 100 Sunnis were massacred in Tremseh. more recently, a score of Alawite civilians were killed in Zahra in may 2016. Alongside these incidents, thousands of individuals from every sect have been kidnapped in the area during the war and either ransomed or murdered.31 After years of violence, the frontlines in Homs became well deined, and many minorities moved into more homogeneous neighborhoods for protection. Yet attacks on Alawite neighborhoods continued. On October 1, 2014, terrorists bombed a school in Akrama, killing some forty people.32 And on February 21, 2016, car bombs killed iftyseven in Zahra.33 hese and other neighborhoods were later blocked of with concrete barriers to prevent outsiders from perpetrating further mass-casualty attacks (whether on foot or in vehicles). Alawites and Christians felt especially threatened, though Sunnis sufered reprisals as well. he Diiculty of Reconciliation In light of this brutal intercommunal violence, the return of mutual coexistence in central Syria is by no means guaranteed even if Assad’s forces retake the entire region. Although the regime 44  secured Homs after the last of the city’s rebel districts, al-Waar, was evacuated in march 2017, threats from other quarters persist. Just north of the city, incursions continue from the rebel-held pocket between Rastan and Houla. Similarly, Hama remained under threat from the Islamic State to the east until as recently as September 2017, when the group’s local stronghold was destroyed; the rebel coalition Hayat Tahrir al-Sham still looms to the north. he Rastan-Houla situation may prove to be particularly intractable. he roughly 100,000 residents of this Sunni Arab territory have eluded regime control since 2011, and the area gave birth to Liwa al-Farouk, a key brigade in the Free Syrian Army commanded by the famous rebel lieutenant Abdul Razzaq Tlass. Local ighters are now under the control of Hayat Tahrir al-Sham, which has been trying to break the siege and attack Homs and Hama. Previously, the regime seemed content with containing rebel advances in the area and negotiating. Rastan and Houla facilitated “reconciliation” eforts in Homs because the army was able to transfer many of that city’s rebels to the pocket, irst in April 2014 and later in September 2016. Yet the army may be targeting the pocket for a full-scale military campaign now that it has removed the IS presence in east Homs and Hama province. If the regime retakes Rastan-Houla, the restoration of trust between Sun- nis and other communities in central Syria will be diicult. Further east, Assad would also need to reestablish links with the Sunni tribes that have been supporting the Islamic State and conducting raids on Alawite and Ismaili settlements. he regime may task Defense minister Fahd Jassem al-Freij with leading this efort given his membership in the Haddadin tribe, which is based in east Hama province and has always been loyal to the Assad family. In 1982, the tribe participated in the bloody repression of Hama, and most of its members have been ighting with the regime during the current war as well. As in other parts of the country, these and other tribal cleavages have fragmented the opposition in central Syria, so Assad will likely use this to his advantage as he attempts to prevent a rebel resurgence there. ALEPPO’S dIVIdINg LINES After the rebels seized parts of Aleppo in July 2012, the city was split into the opposition-held east and the regimecontrolled west. his demarcation was virtually unchanged until 2016, when Assad’s forces retook the city. Yet the boundary was not drawn along strictly sectarian lines—Sunni Arabs were the majority on both sides of the city, so territorial and socioeconomic criteria played important roles in how the uprising played out there. Shifting Demography West Aleppo is considered the “legal” part of the city and has been populated by urbanites for generations. It includes the rich neighborhoods of alShahba and mohafaza, the lower-class Sunni district of Sulaymaniyah, and the middle-class public service district of alHamadaniyeh, where most of the Alawites live. In contrast, Aleppo’s eastern half is largely considered the “illegal” city, with unplanned neighborhoods that have spread widely due to rural exodus (a process called “inrifyeh,” a term used with disdain by old city dwellers). Rebels from the countryside found support in East Aleppo, whose residents have integrated poorly into the conservative city and face a persistent urban/rural divide. he working-class residents in these neighborhoods serve as labor for the vast industrial zone of Sheikh Najar on the city’s northeastern outskirts, run by the West Aleppo entrepreneurial class. In 2012, the population of Aleppo was 2.5 million: 1 million in the western district, another million in the east, and around 500,000 in the northern Kurdish neighborhoods of Sheikh maqsoud and Ashraiya. In October 2016, intense Russian bombing in East Aleppo sharply reduced the population; as described in chapter 1, only 100,000 people remained there by year’s end. Some of those who led found refuge in West Aleppo, in some cases replac FABrICE BAL ANChE 26. EAST ALEPPO vs. WEST ALEPPO, January 2016 al-Mallah Farms Kafr Hamra Central prison NORTH ALEPPO Bani Zaid WEST ALEPPO Regime Area Kurdish Area Sheikh Maqsoud al-Haydariya Railway station Governorate al-Jaziran Sakhur Nor Kiugh University Administrative city Sheikh Najar industrial area al-Azizieh MUNICIPALITY NEW ALEPPO Bab al-Nayrab EAST ALEPPO Military Academy Rebel Area Salah al-Din al-Hamadaniah Police Academy al-Nayrab Military industrial area Bus Station Ramouseh Sheikh Said N W E 0 S Old city Railway Highway Citadel Legal city Airport Principal road Principal location Illegal settlement Frontline Industrial area S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r  2 km ing middle- and upper-class residents who had left the city amid the ighting. For its part, Sheikh maqsoud’s population decreased to only 50,000 inhabitants by 2014,34 with the majority of the Kurdish population taking refuge in their villages of origin in Afrin, Kobane, or Hasaka. To be sure, sectarianism is not absent from the conlict in Aleppo. he Christian quarters remained uniformly loyal to the regime. Sheikh maqsoud rose up against the state, but for reasons diferent from those of Sunni Arab rebels; later, the Kurdish and Arab camps came into direct conlict when the former began cooperating with the army in certain campaigns (see chapter 3). And while Sunni Arabs were divided by clan rivalries based on blood ties or neighborhood solidarity, many of them became more uniformly and openly hostile toward Christians, Alawites, and Ismailis as the rebellion grew more radicalized. he Christian community was the irst victim of such sentiment, and its population fell from 200,000 to 40,000 between 2011 and 2016.35 Rebels targeted the Christian neighborhoods of al-midan, al-Azizieh, and Jdaideh, urging civilians to lee in order to seize their districts. Some neighborhoods resisted this encroachment—in summer 2012, for example, the Christian militia in al-midan halted the rebel advance in the district’s Armenian quarter.  45 To Azaz, Afrin, and Turkey (closed) N W Queiq River S Anadan 27. SYRIAN ARMY OFFENSIVE EAST ALEPPO Fall 2016 Muslimiyah Alternative road to West-Aleppo (open since summer 2016) E Handarat Haritan Handarat camp al-Mallah Farms Central Prison AREA CONTROLLED BY Syrian Army To Bab al-Hawa, and Turkey SDF (Kurds) Castello Road Kafr Hamra Rebels Bani Zaid Sheikh Maqsoud Hanano SYRIAN ARMY TERRITORIAL GAINS al-Haydariya al-Shahba University 22 Sept to 30 Oct 2016 al-Midan Governor's Palace al-Jaziran Sakhur 23 Nov to 4 Dec 2016 al-Azizieh Administrative City Citadel Bab al-Nayrab Bustan al-Qasr Bab al-Maqam New Aleppo Military Academy al-Sukkari Police Academy Artillery School To Damascus via Salamiya Aleppo Airport Marjeh Salah al-Din al-Hamadaniah 5 Dec to 8 Dec 2016 al-Myassar Karm al-Dada Highway Principal road Hill 400 al-Nayrab Railway Sheikh Said Ramouseh To Damascus (closed) 9 Dec to 13 Dec 2016 Airport Urban area Industrial zone Rural area Khan Tuman 0 46  2 km Queiq River Principal road to West Aleppo (open since summer 2012) To Damascus via Salamiya  FABrICE BAL ANChE 28. THE STRUGGLE FOR THE MANBIJ−AZAZ CORRIDOR, Winter 2016 N W AREA CONTROL E Syrian Army KOBANE S Islamic State Feb 2016: Transfer of rebels through Turkey SDF al-Shuyoukh Tahtani Bulbul Islamic State Bab al-Salam Euphrates River AZAZ AFRIN Souran Sheikh Hadid Marea MAIN ETHNIC GROUP Arima Akhtarin Arab Sunni Kurdish Sunni Abu Qilqil AL-BAB al-Zahraa Turkmen Sunni Dec 2015: SDF crosses Euphrates to Manbij Tadef Darat Izza Mar 2016: Syrian ArmyHaritan begins to close Castello ring al-Dana Khan al Asal Atareb SDF (Kurdish) MANBIJ Feb 2016: Syrian Army closes Azaz road Nubl Sarrin al-Shemali Dabiq Tal Rifaat Jandeiris Other rebels al-Raai Maabatli HAREM Syrian army Ghandoura Sharran Bab al-Hawa OFFENSIVE To Gaziantep Raju To Antakya Other rebels JARABULUS TURKEY Rasm al-Harmel ALEPPO Ain Issa Tishrin Dam Arab Shia International border Khafsa Province border Kuweires Qorqanya al-Jarniya Deir Hafer Border crossing Province center AL-SAFIRAH Armanaz District center al-Jaboul Lake al-Fua Taftanaz Maskanah Hader Benan IDLIB Town Aleppo urban area Mar 2016: IS and rebels try to cut Aleppo road Lake Assad Main road Thawra Dam Saraqeb ARIHA THAWRA Tal Daman Khanaser Abu Duhur Ahsem Arab Druze To Hama S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r  0 10 20 km Secondary road Military airport open Military airport closed Bridge open Bridge destroyed  47 Retaking Aleppo Between summer 2012 and 2013, rebels encircled West Aleppo; the army seemed trapped, with the airport becoming its only means of communication with the outside. By may 2013, however, regime forces opened a road through the steppe between Aleppo and Homs, allowing for an inlux of fuel and giving hope to the inhabitants of the regime-held districts. From that date until summer 2016, the army slowly encircled the eastern districts. meanwhile, the air force launched continual strikes against East Aleppo to separate its civilian residents from the rebels, causing major casualties among both due to the indiscriminate nature of the bombing. On August 7, 2016, rebels from Idlib province briely succeeded in breaking through the siege lines and opening a corridor to East Aleppo via Ramouseh district. Yet this did not compensate for the July 28 loss of Castello Road, the main supply line for the city’s rebel districts. he Ramouseh passage remained under attack by Russian planes, which largely prevented the rebels from using it. A few trucks full of produce entered East Aleppo with much publicity on August 7, but this was essentially a onetime delivery that did little to change the daily reality for the rebel zone’s 100,000 inhabitants.36 meanwhile, the army opened a new road to supply West Aleppo from the north, taking a stretch of Castello Road in the process and greatly 48  reassuring the regime zone’s millionplus inhabitants. By October, the army had launched its inal ofensive against East Aleppo, retaking the rest of it in less than two months. many observers were taken aback by the campaign’s speed, but the timeline was not so surprising once it became clear that Russia and Turkey had already discussed the framework under which local rebels could negotiate their surrender. moscow also pushed Assad and his Iranian ally to let the 6,000 remaining rebels leave the city with their light weapons instead of eradicating them.37 he Kremlin seemed particularly keen on achieving this goal before the new American president came to oice in January 2017. Despite the intercommunal hostility witnessed in Aleppo during the war, the city’s fate shows how some key developments in the regime-held portions of Syria have been driven less by sectarian ailiation than by social divisions, whether between economic classes or along urban vs. rural lines. In other cities, however, these social divisions were usually trumped by sectarian concerns; for example, poor Alawites in Latakia did not join their impoverished Sunni “classmates” in demonstrations. CONTESTEd CITIES At various points in the war, the regime also held portions of four cities that lay outside its main zone of control: Deraa, Deir al-Zour, Hasaka, and Qamishli. hese cities ofer a useful illustration of the complex balancing game Assad has sought to play in dealing with sectarian and social factors during the war. he regime seemed to decide early on that it must cling to certain provincial and district capitals in order to legitimize its power, even when these cities became wholly surrounded by enemies. From these strongholds, Assad likely hoped to resume military ofensives using the “oil-spot strategy,” which entails circular expansion from a center. In each city, the regime had the support of the many public-sector employees who form its enormous national bureaucracy. Rebels besieged them and in some cases seized neighborhoods, but the army and security services concentrated their strength to prevent internal rebellion. his dynamic was also evident in provincial capitals that fell fully to the opposition. When rebels seized Idlib, for example, most of the regime’s local bureaucracy and constituents did not support them; instead, minorities and state employees were accused of collaboration and led to regime-held towns. hose who stayed behind were often treated harshly; many were killed by rebels for purportedly serving as “intelligence agents.”38 From this perspective, the regime bureaucracy in Syrian cities constitutes a multiethnic, cross-sectarian tribe of sorts, one that seeks to defend its interests as would any other tribe. Although most of the work done by these state employees is menial, it guarantees a ixed income that is out of reach for the impoverished lower classes that make up much of the rebellion.39 Sunnis held roughly the same number of government posts as Alawites when the war broke out,40 including top oicials such as Ali mamlouk, the powerful chief of the security services. Yet unlike the Alawites with Bashar al-Assad, the Sunni Arabs have no senior political, military, or tribal igure capable of unifying their diferent factions and counterbalancing Assad without crossing the line into outright rebellion or otherwise provoking retaliation. his limits their power. For his part, Assad relies on these Sunni networks within the bureaucracy to manipulate the population—they act as bridges to reconcile groups that sided with the rebellion, coaxing them back to the regime’s side. his was the case with Sheikh Nawaf al-Bashir, a powerful Sunni tribal leader who left Istanbul for Damascus in January 2017. By rallying to the regime, he showed that the Baggara tribe had shifted its support from the rebels to Assad. Yet regardless of the regime’s proven ability to exploit tribal, class, and political issures and coopt many of those who might oppose it, the fact remains that Sunni Arabs still form the large majority of the Syrian population—and the backbone of its current and past armed rebellions. Accordingly, forcing an Ala FABrICE BAL ANChE wite national construction onto wider parts of Syria while excluding other communities from power is not realistic in the long term; it may not even be viable in the multiethnic regime-held zone. he reason is simple: Alawites are a relatively small minority in western Syria and nearly nonexistent in the rest of the country. he Assad regime cannot take the same path as Iraq, where the Iranian-backed Shia majority has used the reins of power to appropriate most of the country for themselves. Notes 1. 2. 3. 4. 5. phone interviews with Syrian businesmen in Damascus, Latakia, and Aleppo; the answer was the same each time: most urban residents in the middle and upper classes considered the revolution dangerous because it came from the informal settlements populated by poor people of rural origin. who were motivated mainly by the prospect of seizing wealth from the city center. As one Sunni businesman in Latakia angrily put it, the uprising threatened to take away the benefits gleaned from twenty years of economic liberalization. 6. “Syria crisis: Three of Assad’s top chiefs killed in rebel bomb strike,” Guardian, July 18, 2012, https://www.theguardian. com/world/2012/jul/18/syria-rebels-killelite-damascus 7. “Syrian government steps up siege of rebel-held Damascus suburbs: state media, rebels,” Reuters, February 6, 2016, http://www.reuters.com/article/ us-mideast-crisis-syria-daraya/syriangovernment-steps-up-siege-of-rebelheld-damascus-suburbs-state-mediarebels-idUSKCN0VF0RT Fabrice Balanche, “Géographie de la révolte Syrienne” [Geography of the Syrian revolt], Outre Terre 3, no. 29 (September 2011), https://www. cairn.info/revue-outre-terre1-2011-3page-437.htm Fabrice Balanche, “Damas: une ville sous le Baath” [Damascus under the Baath], Moyen-Orient, April 2012, https://www.academia.edu/1514404/ Damas_sous_le_Baath_Damascus_under_the_Baath 8. Ibrahim Nouss, “La population de la République Syrienne: etude démographique” [The population of the Syrian Republic: demographic study], doctoral dissertation, The Sorbonne, 1951. 9. Fabrice Balanche, “L’habitat illégal dans l’agglomération de Damas et les carences de l’etat [Illegal housing in greater Damascus and the shortcomings of the state], Revue de Géographie de l’Est 49, no. 4 (2009), https:// rge.revues.org/1980 In April 2011 and following months, the author conducted numerous tele- Mustafa al-Haj, “Militants from Syria’s Qudsaya prepare to leave for Idlib,” Al-Monitor, October 14, 2016, https:// www.al-monitor.com/pulse/originals/2016/10/syria-qudsaya-regimesettlement-opposition-evacuation.html Peter Kenyon and Alison Meuse, “In A Syrian Suburb Cleared of Rebels, a Gradual Return to Everyday Life,” NPR Weekend Edition, December 25, 2016, http://www.npr.org/sections/parallels/2016/12/25/505304239/in-a-syriansuburb-cleared-of-rebels-a-gradual-return-to-everyday-life 10. Aron Lund, “Into the Tunnels: The Rise and Fall of Syria’s Rebel Enclave in the Eastern Ghouta,” Century Foundation, December 21, 2016, https://tcf.org/ content/report/into-the-tunnels 11. See the OCHA infographic “Syrian S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r  Arab Republic: Estimated people in need & IDPs per governorate,” Refworld, October 31, 2016, http://www. refworld.org/publisher,OCHA,MAP,SYR ,584ff2524,0.html 12. Fabrice Balanche, La region Alaouite et le pouvoir Syrien [The Alawite area and the Syrian power] (Paris: Karthala, 2006). 13. OCHA, “Syrian Arab Republic,” http:// www.refworld.org/publisher,OCHA,MA P,SYR,584ff2524,0.html 14. “Syrie: le village arménien de Kassab, victime ‘d’une épuration ethnique’” [Syria: the Armenian village of Kasab, an ethnic cleansing victim], France 24, March 24, 2014, http://www.france24. com/fr/20140324-syrie-village-armenien-kassab-epuration-ethnique-assad 15. Fabrice Balanche, “Le réduit Alaouite: un refuge de la Syrie ‘utile’?” [The Alawite stronghold: refuge for the useful Syria?], Carto 32 (November-December 2015). 16. Nicholas Heras, The Potential for an Assad Statelet in Syria, Policy Focus 132 (Washington Institute for Near East Policy, December 2013), http://www. washingtoninstitute.org/policy-analysis/view/the-potential-for-an-assad-statelet-in-syria 17. Cyril Roussel, Les Druzes de Syrie: Territoire et mobilité [The Druze of Syria: territory and mobility] (Beirut: Institut Français du Proche-Orient, 2011), http://www.ifporient.org/node/1024 18. Talal El Atrache, “The Druze in the Syrian Conflict,” Syria Comment, September 5, 2016, http://www.joshualandis.com/blog/druze-syrian-conflict-talal-el-atrache/ 19. It bears mentioning that this offensive was presumably conducted with arms supplied by the Jordan-based, Saudisupported “Military Operations Center” via sources in the Balkans. These weapons may originally have been intended for more “moderate” rebels but nevertheless fell into the hands of radicals. See Gareth Porter, “How America Armed Terrorists in Syria,” American Conservative, June 22, 2017, http:// www.theamericanconservative.com/ articles/how-america-armed-terroristsin-syria 20. Talal El Atrache, “Syrie: quel avenir pour les Druzes?” [Syria: which future for the Druze?], Monde Arab, September 19, 2015, http://magazine.com. lb/index.php/fr/monde-arabe-?issue_ id=200 21. Noam Raydan and Adam Heffez, “Why Syria’s Druse Are Sticking with Assad for Now,” Jerusalem Post, April 9, 2014, http://www.washingtoninstitute. org/policy-analysis/view/why-syriasdruse-are-sticking-with-assad-for-now 22. Aymenn Jawad Al-Tamimi, “Jabhat alNusra and the Druze of Idlib Province,” Syria Comment, January 24, 2015, http://www.joshualandis.com/blog/jabhat-al-nusra-druze-idlib-province/ 23. Nour Samaha, “Syria’s Druze reject calls for anti-government uprising,” Al Jazeera, September 14, 2015, http:// www.aljazeera.com/news/2015/09/syria-druze-reject-calls-anti-governmentuprising-150913111703291.html 24. Mohamed al Dbiyat, Homs et Hama en Syrie centrale (Damascus: IFEAD, 1995), p. 370. 25. Balanche, “Géographie de la révolte,” https://www.cairn.info/revue-outreterre1-2011-3-page-437.htm. These figures are a personal estimate distilled from the 2004 census and subsequent field surveys. In a 1995 article, Alain Chouet wrote that Alawites constituted 40% of the city’s population, but this estimate seemed exaggerated at the time. Moreover, natural growth patterns and emigration to Homs have not been in favor of the Alawites since then. See Chouet’s article “L’espace  49 tribal Alaouite à l’épreuve du pouvoir,” Maghreb-Machrek 147 (1995), pp. 83– 119, http://alain.chouet.free.fr/import/ Alaouite.pdf 26. See his essay, “Le quartier de Bâb Tebbâné à Tripoli: étude d’une assabiyya urbaine,” in Pierre Yves Pechoux (ed.), Mouvements communautaires et Espaces urbains au Machreq (Beirut: CERMOC, 1985), pp. 45–86. Seurat noted how Tripoli’s Sunni quarter (Bab Tebbane) and Alawite quarter (Bal Mohsen) were the scene of intense clashes. In fact, the two neighborhoods never really stopped fighting—they resumed their exchange of fire in 2005, and the Syria war has only exacerbated the tension. 27. Damien Pearse, “Edith Bouvier’s escape from Homs,” Guardian, March 3, 2012, https://www.theguardian.com/ world/2012/mar/03/edith-bouvier-escape-from-homs 28. Interviews with French NGO members in Tripoli, June 2014. 29. According to a Homs resident interviewed by the author in Damascus in November 2013. 30. “Houla: How a massacre unfolded,” BBC News, June 8, 2012, http:// www.bbc.com/news/world-middleeast-18233934 50  31. Babak Dehghanpisheh and Suzan Haidamous, “In Syria, Kidnappings on the Rise as Lawlessness Spreads,” Washington Post, April 21, 2013, https:// www.washingtonpost.com/world/mid dle_east/in-syria-kidnappings-ont h e - r i s e - a s - l a w l e s s n e s s - s p re a d s /2013/04/21/b0bb2f2e-a854-11e28302-3c7e0ea97057_story.html 32. “Double bombing kills at least 41 children at school in Syria,” Australian Broadcasting Corporation, October 1, 2014, http://www.abc.net.au/ news/2014-10-02/school-bombing-kills-41-children-in-syria/5784266 33. Yousuf Basil and Tim Hume, “In Syria, Dozens Killed as Bombers Strike in Homs and Damascus, Regime Says,” CNN, February 22, 2016, http://www.cnn.com /2016/02/21/middleeast/syria-civil-war/ index.html 34. According to a June 2014 author interview conducted in the border city of Kilis, Turkey, with members of Médecins Sans Frontières in charge of Aleppo. 35. Author interview with French scholar Frederic Pichon, Paris, March 2016. 36. Georges Malbrunot, “En silence, les rebelles quittent Alep entre des soldats russes” [In silence, rebels leave Aleppo between Russian soldiers], Le Figaro, December 26, 2016, http://www.lefiga- ro.fr/international/2016/12/22/0100320161222ARTFIG00229-en-silence -les-rebelles-quittent-alep-entre-dessoldats-russes.php 37. Ibid. 38. Aron Lund, “Assad’s Broken Base: The Case of Idlib,” Century Foundation, July 14, 2016, https://tcf.org/content/ report/assads-broken-base-case-idlib. Lund described the rebels’ methods: “Having broken into the local intelligence headquarters and released those prisoners who remained alive, [Jaish al-Fatah] also set about studying the documents captured there in order to weed out intelligence agents and informers. In May 2015, the group’s ruling council issued a list of some 750 wanted Assad supporters.” 39. Fouad Ajami described these class fissures well: “There were divisions of class and geography that still ran through the Sunni population, there was the age-old separation between city and country, and the middle classes of Aleppo and Damascus bristled at the idea that they would be led by provincials.” See his book The Syrian Rebellion (Hoover Institution, 2012), p. 91. 40. Souhail Belhadj, La Syrie de Bashar al-Asad [The Syria of Bashar al-Assad] (Paris: Belin Litterature et Revues, 2013).  FABrICE BAL ANChE 3 . S y rI A N k u r dI S TAN u N dE r T h E Pyd w hATEVEr the future may hold for the rest of Syria, the emergence of a partially autonomous Kurdish region has already become a reality. he boundaries of this “Rojava” are still uncertain and could wind up difering from the maximalist lines claimed by the Democratic Union Party (PYD). Yet the leading Kurdish faction has continued its eforts to absorb more territory, gather all Syrian Kurds under its banner, and make Rojava viable economically, indicating that the country’s sectarian fragmentation may be irreversible in the north. Looking back at the recent history of the Kurdish region, three trends stand out: the Assad regime’s forced “Arabization” of some inhabitants (which may now be reversed through a PYD “re-Kurdiication” campaign); the imposition of heavy economic restrictions, which have rendered it dependent on regime-controlled parts of the country; and the PYD’s rise to prominence out of the ashes of the 2004 Kurdish upris- ing, nearly a decade before the current civil war. Each of these trends will be examined below, along with their implications for the ongoing hostilities in various parts of northern Syria. ThE CurrENT SITuATION IN rOJAVA In October 2015, the PYD’s military branch, the YPG, partnered with a few local Arab militias to create the Syrian Democratic Forces, and the SDF have since helped to expand the territory under Kurdish control. In the northeast, Kurdish-led forces have reclaimed large swaths of ground from the Islamic State, including the provincial capitals of Hasaka and Raqqa. he rationale for this expansion seems to vary depending on where it occurs. As described in previous chapters, the PYD has advanced out of its Kurdish-majority heartland to conquer areas where Kurds are either the minority or not present at all. In some cases, such as the conquest of manbij in summer 2016, the goal is to establish full territorial contiguity between Rojava’s western canton (Afrin) and its two eastern cantons (Jazira and Kobane). In other cases, such as the thrust toward the IS “capital” of Raqqa, the PYD seems more concerned about countering current and potential threats to Rojava than expanding its contours. Whatever the PYD’s future plans, the SDF controlled about 25% of Syria as of October 2017. hese territories are home to around 2.5 million people, yet the Kurdish portion of this population fell below 50% with the occupation of large parts of the Arab Euphrates Valley. In the Jazira and Kobane cantons of Rojava itself, Kurds still constitute a slim majority (55%), while the area around Jabal al-Akrad (aka Kurd Dagh) in Afrin canton is almost 100% Kurdish. For demographic reasons, then, the Kurds likely have no desire to permanently integrate Arab-majority areas such as Raqqa city; rather, they launched their ofensive there because of the persistent IS threat to Rojava. But their plans  51  for al-Shadadi, an Arab-majority town south of Hasaka, are less clear. Although taking it met an obvious short-term military goal (cutting of the main IS route to mosul, Iraq), the town is ifteen times smaller than Raqqa and adjacent to signiicant oil wells, so the Kurds may decide that holding onto it indeinitely is worth the potential sectarian strife. In general, the Kurds may believe that they can ofset their demographic disadvantage in certain Arab-majority areas through ethnic cleansing, “reKurdiication” of Arabized Kurds, and alliances with Arab tribes that want to side with the strongest player. he PYD also hopes that some of the one million Kurds in Damascus and Aleppo will move to Rojava, but convincing them to do so will require drastic improvements in the north’s dire economic situation. dEFININg kurdISh IdENTITy & TErrITOry As mentioned in chapter 1, the toponymy of maps from pre-independence 29. ETHNIC DIVISIONS IN NORTHERN SYRIA N Tigris River W E Qamishli Euphrates River S TURKEY Kobane Jarabulus Afrin Azaz al-Raai Ras al-Ain Tal Tamer Tal Abyad Manbij HASAKA al-Bab Mediterranean Sea al-Malikiyah Harem al-Shadadi ALEPPO Kasab Lake Assad IDLIB RAQQA IRAQ Jisr al-Shughour LATAKIA DEIR AL-ZOUR 0 100 km HAMA Armenians Kurds Kurds and Arabs Province center Province border Turkmens Arabs Desert area (Arab population) Locality International border Assyrians Syria shows that the extent of ethnic Kurdish territory at the time was vast. Kurdish communities existed even in the coastal Alawite heartland—in the thirteenth century, for example, Sultan Baybars settled Kurdish tribes in Jabal al-Akrad in order to better control the Latakia-Aleppo axis.1 many other Kurdish settlements were built around historic fortresses such as Salah al-Din, marqab, and the Crac des Chevaliers; in fact, Arabs called the latter fortress “Husn al-Akrad,” or “Castle of the Kurds,” until 1950 (though they now call it “Qalaat al-Husn”). 52  Today, the Kurdish people residing in these locales have been Arabized to the extent that they no longer consider themselves Kurds, despite retaining Kurdish words in their Arabic dialect. moreover, Sunni Arab rebel groups were able to get a foothold in such areas during the current war—unlike in the Kurdish-majority north—highlighting the complexity of any efort to delineate “Kurdish territory.” Resisting Arabization Between 1945 and 1970, successive Syrian regimes systematically Arabi- zed village and city names in order to eliminate non-Arab religious and ethnic markers. his trend reached its peak under the Baath regime from 1963 to 1970. he names of hundreds of Kurdish villages were changed without the population’s consent, with the new name in some cases having no relationship to the original name. he application of this policy varied by region. In Aleppo province, for instance, numerous Kurdish village names were changed, but the Turkmen villages around Azaz were left unaltered. Why the diference? In all likelihood, the Baath regime wanted to support one Aleppo minority over the other, according to the old principle of “divide to reign.” In Latakia province, however, all Turkmen village names were Arabized in order to maintain the dominance of the regime’s fellow Alawites.2 In the northeast, Arabization efforts were rather limited. he dispersal of local housing and the multitude of  FABrICE BAL ANChE N W 30. PYD EXPANSION IN NORTHERN SYRIA, May 2017 E Tigris River S Euphrates River Qamishli TURKEY Kobane Jarabulus Ras al-Ain Tal Abyad Afrin Azaz al-Bab Tal Tamer Manbij HASAKA Harem Mediterranean Sea al-Malikiyah al-Shadadi ALEPPO IDLIB Lake Assad RAQQA IRAQ Jisr al-Shughour LATAKIA 0 100 km DEIR AL-ZOUR HAMA PYD EXPANSION OTHER ACTORS Inhabited | Uninhabited Inhabited | Uninhabited Jan 2015 Syrian army Province center Province border Jan 2016 Other rebels Locality International border Jan 2017 Islamic State Oil field Claimed Rojava border May 2017 hamlets in the area would have made comprehensive renaming extremely dificult, so the Baath settled for Arabizing the names of key locales. his approach also highlighted the regime’s lack of interest in the Kurdish countryside, which it largely abandoned over time. In addition, the prevalence of Kurdish identity there may have convinced Arab nationalists to retreat. Whatever the case, Kurds are now the majority population in three portions of northern Syria: the Kurd Dagh (Kurd mountains), including the small town of Afrin; the city of Kobane (aka Ain alArab) and surrounding countryside; and the large Jazira region stretching from Ras al-Ain to the Tigris River. Kurdish Identity in Damascus and Aleppo mapping the distribution of Syria’s Kurds raises thorny questions about people’s sense of belonging to an ethnic group. Which people are “truly” Kurds—only those who speak Kurdish? What about S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r  the Arabized Kurds for whom Kurdish is a second language? What about bilingual Kurdish/Arabic speakers who do not claim Kurdish identity? What about the children of mixed marriages between Kurds and Arabs, a practice that has contributed to the dilution of Kurdish identity and speech? he situation in Damascus best illustrates these dilemmas. In the Rukn al-Din and Barzah districts, many inhabitants are ethnically Kurdish but have neglected ties with their ancestral villages for three generations and are now well in- tegrated into capital society. Younger residents do not know how to speak Kurdish, and mixed marriages are numerous. In contrast, Kurds in Wadi Doumar— who arrived in the 1980s as construction workers to help build the city’s newest districts—tend to reside in exclusively Kurdish neighborhoods and remain attached to their rural roots despite living in a more urban setting for years. As for Aleppo, the Sheikh maqsoud district constitutes a Kurdish city within the city. Kurds made up 20–25% of Aleppo’s population before the war, mainly clustered in the northern parts  53 N W 31. ROJAVA POPULATION AND FUTURE ADMINISTRATIVE ORGANIZATION 2016 Tigris River E Derik Qamishlo Qamishli al-Malikiyah Euphrates River S Kobane Ain al-Arab TURKEY 30% Efrin Afrin Jarabulus Manbij Azaz Serekaniye Ras al-Ain Gire Spi 55% Tal Abyad KOBANE al-Bab ALEPPO Mediterranean Sea Lake Assad Jisr al-Shughour RAQQA IDLIB IRAQ LATAKIA 0 DEIR AL-ZOUR HAMA 100 km 2010 POPULATION BY CANTON ROJAVA ADMINISTRATION SYRIAN ADMINISTRATION Present Kurdish canton Canton border International border 900,000 Future Kurdish canton Canton center Province border 350,000 2010 Kurdish population by canton Rojava capital Province center CIZIRE 1,300,000 EFRIN KOBANE Serekaniye Kurdish name of the city. Since the 1970s, Aleppo has served as a natural outlet for the rural exodus from Kurd Dagh and the countryside around Kobane and Jarabulus. many Kurds preferred the city not only because of its numerous job opportunities, but also because it contained entire neighborhoods where they felt free to speak their language without discrimination. his was not the case in smaller towns such as manbij and alBab, where Kurdish minorities adopted a more discreet proile. Aleppo has even developed a Kurdish entrepreneurial bourgeoisie rooted in the city’s textile 54  55% CIZIRE (JAZIRA) al-Shadadi EFRIN Harem HESITSHE HASAKA and car repair industries. Kurdish auto shops gradually replaced those run by Armenians, while the Sheikh maqsoud district has become a vast informal industrial zone where hundreds of small apartment workshops supply textile wholesalers. Wadi Doumar and Sheikh maqsoud highlight the problem with calling Syria’s cities “mixed” or holding them up as models of sectarian coexistence. hese districts are essentially Kurdish mini-cities nestled within larger Arab cities. Kurdish residents could easily live Ras al-Ain Arabic name their entire lives without meeting people from Arab districts. hey watch international Kurdish television networks rather than Syrian or other Arabic channels, and they can content themselves with speaking Kurdish in their daily lives, since Arabic is only a foreign language taught at school. Indeed, the government has failed to implement special curricula for students with a diferent mother tongue, so all classes are taught in Arabic. Partly as a result of this language gap, many local teachers show little efort in their classrooms, which are usually overcrowded with forty or more children due to lack of investment from the ministry of Education.3 In response, Kurdish students often sidestep the dificulties of learning Arabic, leave school early, and ind jobs in garages or workshops. he level of education in these districts is lower than the national average, which has limited Kurdish access to the civil service and skilled jobs while transforming large areas into ghettos. Even so, a Kurdish intellectual middle class developed in Damascus and Aleppo over the years, and its members were ea FABrICE BAL ANChE N W 32. ARABIZATION OF VILLAGE NAMES IN NORTHERN SYRIA E Tigris River Euphrates River S Qamishli TURKEY Kobane Jarabulus al-Raai Afrin Ras al-Ain Tal Tamer Tal Abyad Manbij Azaz Mediterranean Sea HASAKA al-Bab Harem al-Shadadi ALEPPO Kasab al-Malikiyah Lake Assad IDLIB RAQQA IRAQ Jisr al-Shughour LATAKIA DEIR AL-ZOUR 0 100 km HAMA ETHNICITY Armenian Kurdish Desert area (Arab population) Turkmen Arab Arabization of village names 1958–1970 Assyrian Kurdish and Arab ger to integrate into wider urban society. For many of them, the Syrian Communist Party served as a vehicle for such integration. Unlike the Baath, the SCP did not promote Arab nationalism; in fact, Kurdish politician Khaled Bakdash was the party’s president for decades.4 At the national level, however, the notion of Kurdish integration was challenged by the presence of Abdullah Ocalan, the Kurdistan Workers Party leader who led to Syria from Turkey in 1979. He was granted refuge in Damascus until 1998, during which time he and his PKK cohorts launched cross-border attacks into Turkey with the Assad regime’s tacit approval. hese operations helped spread the Baathist idea that Syrian Kurds were of foreign origin, and that their status as “guests” (even those with Syrian citizenship) precluded them from claiming any political space in the Syrian nation. he regime eventually expelled the PKK amid Turkish military threats, but the perception of Kurds as outsiders did not fade. And while the group’s departure allowed for the emergence of other Kurdish political voices, the PKK’s inluence persisted via the PYD. S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r  Province center Province border Locality International border he 2004 Hasaka uprising and subsequent crackdown were perhaps the inal nails in the coin of Kurdish integration into Syrian society, preiguring the country’s general fragmentation as minorities retreated into religious and ethnic solidarities (see the PYD section of this chapter for more on that uprising). For many of the Damascus Kurds who felt compelled to lee the country during the current war, this meant taking refuge in Iraqi Kurdistan rather than other neighboring countries, where they believed they would be less welcome due to their ethnicity. ECONOMIC dEPENdENCE ON dAMASCuS he Kurdish territories of northern Syria sufer from chronic underdevelopment linked to years of Arabization policies and geopolitical problems with Turkey and Iraq. A slow economic opening with Ankara began in 2005, but it was not suficient to reverse the situation prior to the war. Developments since then have rendered Kurdish trade via Iraq and Turkey complicated or impossible, but the PYD has nonetheless maintained economic relations with the regime zone and hopes to increase its export options in the future.  55 Hasaka Treated Like an Internal Colony For decades, Syria’s borders with Turkey and Iraq were real barriers to economic activity in the northern provinces. he custom houses in Qamishli and Ras alAin were open only a few hours a day, and the Turkish consulate in Aleppo was the closest means of obtaining a visa. Until 2008, Syrians had to apply for visas to travel to Turkey, and those with Kurdish or Armenian names often faced delays in obtaining the necessary documents. As a result of such obstacles, Syrian trade with Turkey was meager in size and conducted mainly through Bab al-Hawa west of Aleppo rather than through Kurdish areas along the northern frontier. Similarly, the Iraqi border was often closed because the two countries had been enemies since the 1960s; the enmity was so strong that Iraqis were one of only two nationalities barred from holding a Syrian passport (Israelis being the other). Baghdad moved closer to Syria in the 1990s following the U.S. embargo, and Syrian markets beneited from increased trade. Yet the low of goods largely bypassed the northern Kurdish provinces; most items bound for Aleppo went through the eastern border crossing at Abu Kamal, while those bound for Damascus went through al-Tanf in the south. In light of these policies, the large Kurdish-majority province of Hasaka 56  (which has become known as the Jazira canton of Rojava) was completely isolated, despite occupying a long strip of the northeastern border. Damascus treated it like an internal colony. Before the war, Hasaka’s residents produced a third of Syria’s oil, most of its cotton, and half of its cereals, which helped to ensure the country’s food independence.5 In return, however, the central government exploited this wealth for its own purposes without reinvesting export proits back into the region. Local educational eforts lapsed as well. Technically, schooling was mandatory up to age fourteen in Syria, but many children in Hasaka province left well before then to help their parents with agricultural work. Tens of thousands of Hasaka residents were landless peasants who rented their services to large farms, so they could not aford to send their children to school. By 2004, government census data indicated that illiteracy had reached a record level of 30% in the province. Local Kurdish children often had great diiculty following a state-mandated curriculum taught in Arabic since that is not their mother tongue. Hasaka’s agricultural proletariat was (and in many ways remains) an anachronism in Syria. During and after the agrarian reforms of 1963–1970, large estates in most of the country were dismantled and redistributed to landless peasants as freehold land. In Hasaka, however, these reforms were only partially applied. Large estates were expropriated, but the land was not given to the peasantry. Instead, the state rented it to a relatively small number of individuals—usually regime cadres, including many of the same wealthy families that had previously owned it.6 Haiz al-Assad did not want Kurdish peasants to beneit from the land redistribution because that could have given them unprecedented political weight in the long term. By renting the land out, the Baath regime kept the Kurds in a perpetual state of dependence while also maintaining leverage on the former landowning class. Assad was able to “clientelize” these notables by implicitly threatening not to renew their leases. Hasaka was also left out of the socioeconomic development seen in other rural areas of Syria after the agrarian reforms, particularly the coastal Alawite region. he province was still lagging behind in public services before the war. Half the population resided in localities that did not have a municipality, so they had to go to the provincial capital for all of their administrative needs; more seriously, this meant there was no mechanism through which to carry out local development. After the 2004 revolt and crackdown, the regime launched a development plan for Hasaka, presumably believing it would limit future protests. Yet this plan was never fully implemented; projects were launched but not completed, including those aimed at addressing urgent water problems. Water Scarcity Agricultural practices in Hasaka remain very traditional, so the area has sufered dramatically from water scarcity since the beginning of the century. Overconsumption of water on the Turkish side of the border contributed to this problem, and the Assad regime made no serious attempts to resolve it. Unlike in the Euphrates Valley, where farmers receive cheap and abundant water due to major state irrigation projects, farms in northern Hasaka are supplied by private wells. In the years leading up to the war, drought and the tripling of the price of fuel (which farmers use for motorized water pumps) caused a signiicant reduction in Hasaka’s cultivated areas.7 In 2008, an extensive irrigation plan based on water from the Tigris River was announced, but the lack of political will from Damascus and the outbreak of the war blocked its implementation. In the future, water scarcity will be the main problem facing Rojava because agriculture is the population’s principal source of income. Kurdish administrators in Hasaka/Jazira have indicated that establishing a Tigris pump station is probably a priority,8 but fulilling that pledge would require an agreement with Iraq and Turkey before tapping into the transnational river. Bashar al-Assad obtained such permission from Turkey in  FABrICE BAL ANChE 33. CHANGE IN IRRIGATED LAND, NORTHEAST SYRIA, 2001–2009 Tigris River N 25%–43.2% decrease W E TURKEY Qamishli al-Malikiyah 0%–25% decrease S 0%–25% increase Ras al-Ain Tal Abyad Aleppo 25%–50% increase HASAKA Balikh River 50%–87.5% increase Uninhabited area IRAQ al-Thawra RAQQA Euphrates River River Province border District border Khabur River Province center DEIR AL-ZOUR Hama District center Mayadin Homs Abu Kamal 0 100 200 km Source: Ministry of Agriculture, Damascus 2008, but the war has changed all of the relevant parameters. Another option for Rojava is to develop irrigation plans that draw on the Euphrates dam at al-hawra, but that would create potentially serious issues with Arab farmers and tribes living downstream. If Rojava chooses to cut ties with the Assad regime, the Kurdish region’s bounty of arable land could help it become economically independent, since the Hasaka/Jazira region remains Syria’s largest producer of wheat and cotton. Yet these production levels are dependent on water, so the Kurds would need to reduce their consumption rates through better irrigation techniques. Local peasants still tend to use traditional gravity feed systems, which waste a lot of water. For instance, if farmers using this method want to get one cubic meter of water to their crops, they must draw seven times that amount from their source. With a sprinkler system, however, they would need to draw just two cubic meters.9 Prior to the war, few farmers used sprinklers in Hasaka province, and the regime bureaucracy dragged its feet on modernization eforts for years. Perhaps Rojava’s administrators will prove more eicient if given full autonomy to implement their own irrigation plans. Oil Wealth he Hasaka/Jazira region also holds signiicant oil reserves that could be a major asset for Rojava. In 2010, the local S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r   57 al-malikiyah and al-Shadadi oil ields provided around one-third of Syria’s total output of 383,000 barrels per day.10 Since then, that igure has collapsed due to lack of maintenance and closed pipelines, which made export impracticable. Nevertheless, restoring that low could give the Kurds energy autonomy; oil is already a major source of revenue for the Rojava administration, which sells it inside PYD territory and exports small amounts of it to the regime zone by truck. If Rojava can overcome the transportation hurdle, it could eventually export most of its production. hat hurdle is a tall one, however. Direct export via Turkey seems impossible given Ankara’s hostility toward the PYD, which it considers an arm of the PKK. Alternatively, if the original pipeline to the coastal terminal of Baniyas were reopened, the Kurds could sell oil to the Assad regime, though Damascus would surely oppose paying full price for “its own oil.” Rojava could also use the Iraqi Kurdish pipeline to Turkey, though that would require reaching an agreement with Baghdad and/or the Kurdistan Regional Government (KRG). he faction currently in power there, the Kurdistan Democratic Party (KDP), has very bad relations with the PYD due to decadeslong tensions with the PKK; the Iraqi and Turkish groups follow entirely different ideologies and are led by strong leaders who detest each other. Partly as a result, trade between the Syrian Kurds 58  and the KRG has been erratic, so establishing a stable oil export route is hardly a given. Still Firmly in Syria’s Economic Space hese transnational Kurdish tensions could cause long-term problems for Rojava because the KRG is its only direct overland connection with the outside world. he Turkish border is closed, much of the Euphrates Valley is controlled by IS (though perhaps not for much longer), and the reopened manbij-Aleppo road can only be considered a foreign connector so long as Rojava remains in the Assad regime’s good graces. When the Syrian army met up with the PYD/SDF south of manbij in February 2017, Kurdish authorities characterized it as a means of linking Afrin with the eastern cantons. he PYD would have preferred to control this link with its own troops rather than relying on the regime, but Turkish operations in the preceding weeks had blocked its westward march. Even so, the development gives Rojava a means of preventing Ankara from boxing it in politically or economically. Going forward, the manbij-Aleppo linkup will facilitate the circulation of goods between Afrin and the rest of Rojava. Fortunately for the PYD, Assad has a mutual interest in expanding economic relations with the Kurds, so he is unlikely to cut this low of anytime soon. Western Syria needs the cotton, wheat, and oil produced in Hasaka/Jazira, while the Kurds need to export their raw materials and import manufactured goods. his development also makes Rojava less dependent on the KRG for supplies; the northeastern passage to Peshkhabur, Iraq, is no longer the only international trade route open to it. To be sure, Kurdish wartime trade relations with western Syria were ongoing even before February’s Afrin linkup. Goods continued to circulate between Rojava and regime territory, with taxes levied by the army, Assad-allied militias, and even certain rebel/jihadist groups. For example, trucks transporting Jazira’s grain harvest to the regime zone had to pay a commission to IS forces when passing through their territory.11 And whenever Afrin was supplied with fuel from reineries in the Rmelan region east of Qamishli, Turkish-backed rebels in between the two areas used to take as much as half the cargo in “tolls.” Since February 2017, however, Kurds have been able to send fuel through the reopened army corridor between manbij and Afrin via Aleppo. Assad’s forces have kept their own tolls relatively low to incentivize the use of regime-controlled roads. Facilitating trade with Rojava helps Assad politically as well, keeping the Kurds under his economic inluence while also beneiting key associates and family members such as Rami makhlouf, owner of FlyDamas, the principal airline serving Qamishli. Despite being located deep within Rojava, Qamishli Airport is still under the Syrian army’s control. he PYD never tried to take it because it is an indispensable means of communication for Rojava. he two daily lights to Damascus are full of essential cargo, as are two weekly lights to Beirut and one weekly light to Kuwait.12 he airport is also the easiest way for local civilians to travel abroad. he nearest international airport is Erbil, a full day’s drive away in Iraq. Crossing the Tigris to Peshkhabur is a painstaking process because civilians cannot use the local barge, which is reserved for cargo trucks; customs formalities take up ample time as well. moreover, prior to the manbij-Aleppo linkup, Qamishli Airport was the only way for Rojava civilians to move to and from the western regime zone. housands of Kurdish students who attend universities in Aleppo, Damascus, Homs, and Latakia have therefore faced the prospect of not being able to return home during the war. Access to the capital is essential for medical reasons as well—Rojava hospitals are poorly equipped, and most of their medicines come from the regime zone.13 Public salaries and pensions from Damascus have also arrived by plane. In short, Kurdish authorities do not yet have the means to replace the Syrian state in many sectors, so they will likely do whatever is required to preserve the Qamishli and manbij links.  FABrICE BAL ANChE ful textile industry and was exported for great proit, so farmers were forced into strict regime-mandated production plans that excluded most other crops. Public oices supplied them with seeds and bought their whole crop at prices ixed by the state; they also received fertilizer from the huge chemical plant in Homs at low cost. meanwhile, the creation of local industrial enterprises was long forbidden Ideology vs. Economic Pragmatism he PYD takeover of Hasaka province in 2012, followed by the breakdown of territorial continuity with the regime zone when rebels seized the Euphrates Valley, completely disorganized the local economy. As described previously, Hasaka’s role had been limited to producing raw materials, including some 80 percent of Syria’s cotton.14 his “white gold” fed the country’s power- siderable oil reserves; most other electricity came from the Baath, Tishrin, and hawra dams on the Euphrates. By enforcing such economic dependency, the regime hoped to avoid any secessionist attempts by the Kurds—in striking contrast to the Alawite coastal region, which Bashar’s father endowed with all the infrastructure needed to establish an independent sectarian bunker if the regime ever lost power in Damascus. 34. SYRIAN KURDS BREAK ENCIRCLEMENT, April 2017 N W in Hasaka. While the state opened two spinning mills there, most of the country’s cotton was processed in Aleppo and the coastal region. he province had no textile sector to speak of, and its agrofood industry was limited to a few artisanal dairies and lour mills for local needs. It had no reineries or plastics industry either. he Rmelan thermal power plant covered only strategic needs related to extracting the province’s con- E Tigris River Euphrates River S TURKEY Kobane Bab al-Salam Tal Tamer Azaz Manbij Hasaka al-Bab Aleppo Mediterranean Sea Katuniyeh Tal Abyad Afrin Bab al-Hawa Qamishli Ras al-Ain Jarabulus Peshkhabur al-Malikiyah Lake Assad Idlib Sinjar Mountains al-Shadadi Raqqa al-Tabqa IRAQ Jisr al-Shughour Latakia Deir al-Zour 0 100 km Hama AREAS OF CONTROL Inhabited | Uninhabited SDF Rojava commercial route Province center Province border Syrian army Closure Locality International border Sunni Arab rebels Border crossing Oil field Civilian airport open Islamic State New opening for Rojava S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r   59 35. ROJAVA AS A FUTURE CORRIDOR TO THE SEA Black Sea Istanbul Yerevan Ankara Caspian Sea Tabriz Diyarbakir Iskenderun ROJAVA w A la i t e s ta te ? Latakia PKK Qamishli Gaziantep Aleppo Jisr al-Shughour Qandil Mosul Erbil Tehran Sinjar Kirkuk KRG Hamadan UNSTABLE ARAB SUNNI AREA Mediterranean Sea ria state? rn Sy ste e W Baghdad Damascus Basra r Pe N si a n E S Claimed Rojava 60  ul G W 0 f 400 km Red Sea PKK stronghold PKK headquaters KRG control Official KRG Kurdish-majority areas (per Kurdish Institute of Paris) Since gaining control in Hasaka and other areas, the PYD has been pushing for a self-suicient economy to liberate itself from unequal relations with Damascus. It also rejects capitalism and seeks to promote the philosophy of PKK leader Abdullah Ocalan, who has long advocated marxist-leaning policies. At the moment, however, these ideas can only be implemented on a small scale in Rojava, so Kurdish authorities likely feel compelled to continue opening up their territory, exporting their raw materials, and trading for manufactured products (see the “PYD Path to Socialism” section below for more on this dilemma). he Kurdish communities in Rojava are very resilient and can accept spartan living conditions, but many residents have been leaving the region throughout the war. To stop this population hemorrhage, the PYD will need to develop the economy, which requires freer movement of goods internationally. In this context, opening a second commercial route to Iraq would strongly reinforce Rojava’s autonomy, and much more rapidly than the slow and uncertain construction of a self-suicient economy. A new land route to Kirkuk could break Hasaka’s dependence on the Peshkhabur border crossing, which remains under full KDP control and presents a host of political and practical obstacles. A new route could pass south of the Yazidi mountains, then through Sinjar and Tal Afar in western Iraq, and  FABrICE BAL ANChE inally on to Kirkuk through the Tigris Valley. All of these areas are now controlled by the Iraqi army and associated Shia militias.15 To be sure, the feasibility of opening this route is not yet assured. As of this writing, the Islamic State is still present south of Sinjar, while the KDP and PKK are locked in a standof over control of that city following violent clashes earlier in 2017. moreover, KDP leader masoud Barzani and Turkish president Recep Tayyip Erdogan both oppose any strategic axis that opens Rojava up to the outside world and reinforces its geopolitical importance for Iran. Kirkuk is partly under the sway of the Patriotic Union of Kurdistan (PUK), a KRG faction that is close with Tehran and Baghdad, unlike Barzani’s party. If the Syrian Kurds establish a direct corridor to the city, they might forge a link with Iran in the process, perhaps through Sulaymaniyah further east of Kirkuk. heoretically, Rojava could then become an Iranian transit route between Iraq, western Syria, and even the mediterranean coast, at least once U.S. forces leave eastern Syria. Although this is not the shortest potential westward route for Iran, it would have the advantage of circumventing IS strongholds along the Syria-Iraq border, where the terrorists are likely to take refuge after being expelled from Raqqa and mosul. Of course, all of these scenarios will likely become even more complicated in the wake of the September 2017 Iraqi Kurdish independence referendum, which spurred heavy backlash internationally, regionally, and domestically.16 Pyd hEgEMONy he PYD dominates the Syrian Kurdish political scene and has become a key military actor in the war, courted by both the United States and Russia. Its main goal is to unify the country’s Kurdish territories and then proclaim its autonomy within the framework of a future Syrian federation. Yet this goal, coupled with its apparent commitment to implementing the PKK’s communist ideology, is fraught with diiculties and could put the PYD at odds with local Arabs, policymakers in Washington, and entrenched business interests. he PYD’s Military Rise In march 2004, the Kurdish population in Qamishli rose up against the regime to protest discrimination, economic misery, and underdevelopment in their region. he revolt spread to other Kurdish areas of Syria, including Afrin, Jarabulus, Kobane, and districts of Aleppo and Damascus. In response, army forces closed Kurdish neighborhoods in the cities while police systematically arrested men over age ifteen. he regime also sent Gen. maher al-Assad’s Republican Guard to Hasaka province to quell the revolt, in partnership with lo- S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r  cal Arab tribes whom it had armed over the years speciically to keep the Kurds in check. Oicially, the crackdown killed 43, injured hundreds more, and resulted in about 2,000 arrests, but the actual numbers were probably much higher than these biased regime statistics.17 Several hundred activists escaped, mainly to Iraq Kurdistan, where they are still gathered in camps today (e.g., Sumel northeast of Dahuk). After the revolt, the regime promised the Kurds it would ameloriate their economic and political situation, but little was actually done. Public investment remained very limited. he previously stateless bedoon were oicially permitted to obtain Syrian nationality in 2011,18 but few Kurds seemed to receive their citizenship papers after requesting them.19 he only notable change was regime tolerance toward the creation of Kurdish political parties. Around a hundred were formed, most of them serving as extensions of individual clans. From the regime’s perspective, this political fragmentation was useful in keeping the Kurds under control. After the PYD was formally established, however, it took advantage of the regime’s new tolerance to expand its inluence, especially in Afrin and Kobane. he party was an ofshoot of the PKK,20 which had lost its inluence on Syrian Kurds after Ocalan and his cadres were expelled from Syria in 1998. According to scholar Jordi Tejel: Following the death of Haiz al-Assad in 2000...relations between Syria and the government of Turkish prime minister Recep Tayyip Erdogan began to improve rapidly. Dozens of PKK activists based in Syria were handed over to Turkish authorities, while former PKK ighters returning to Syria were given prison sentences ranging from one to ten years. Despite this context of adversity, Syrian PKK militants established the [PYD] in 2003, with two objectives: escaping state repression and maintaining the support of the PKK’s thousands of members and sympathizers.21 he 2011 national uprising reawakened Kurdish contestation of regime control in the north. As protests escalated, Assad did not want to multiply the number of potential fronts he faced, so he normalized relations with the PKK that spring in order to placate the PYD and its Syrian Kurdish constituents. his also served to counter Turkey, which openly supported Syria’s Sunni Arab opposition. he signing of a truce in 2011 between Tehran and the Iranian branch of the PKK, the Party for a Free Life in Kurdistan (PJAK), also helped the PYD, since the Assad regime quickly followed the cue of its Iranian patron by treating the PKK and PYD favorably. meanwhile, the KDP’s inluence on the Syrian scene began to wane, due in part to the Iraqi party’s ongoing tensions with the PKK and PYD. In 2011, President Barzani created the Kurdistan  61 National Council (ENKS in its Kurdish acronym), which was intended to bring together Syrian Kurdish parties while deliberately excluding the PYD. At irst this initiative seemed like a viable counterweight to the PYD thanks to the KDP’s considerable inancial resources and reasonable working relations with Baghdad. By 2012, however, Barzani was greatly distracted by conlicts with internal rivals and the Iraqi central government, which diverted attention and resources from Syria. he PYD has taken advantage of these factors to impose itself all the more during the war,22 despite the multiplication of other Kurdish parties in previous years, and despite the fact that rival factions believe it professes a retrograde and authoritarian ideology.23 Oicially, the PYD espouses Ocalan’s ideas about evolving from marxism-Leninism to socialist self-management, but in practice the party has tended more toward “democratic centralism,” which efectively amounts to monopolizing power. Its leaders advocate a federal Syria to justify Kurdish autonomy, but few Syrians outside Rojava have seriously proposed that option. Perhaps most important, the PYD is the only Syrian Kurdish party with its own militia. hus, whenever the rebel Free Syrian Army or jihadist groups like IS tried to take over Kurdish areas, the local population sought protection from the PYD because they had no one else to 62  turn to. he party’s contributions to defending Ras al-Ain against rebels coming from Turkey in winter 2012–201324 were a particularly potent boost to its inluence in Hasaka province. Afrin canton is still under threat from nonKurdish rebel groups and even Turkish troops,25 spurring many locals to back the PYD out of necessity. he PYD’s decision to ield its own Kurdish militia and launch ofensives outside its area of control also contributed to the FSA’s loss of prominence within the wider Syrian rebellion. Early in the war, the FSA had a signiicant number of Kurdish ighters, and its clashes with the PYD put these ighters in a very uncomfortable situation. he PYD depicted Kurdish FSA members as traitors to Rojava’s cause, while the Sunni Arab opposition grew increasingly wary of them, especially as radical elements began to take over the rebellion. By 2014, Jabhat al-Nusra was busy eliminating or absorbing secular rebel groups, and Kurds were leaving the FSA en masse, joining with Arab ighters to form the militia Jaish al-huwar (Army of Rebels). members of this militia often found themselves at odds with radical Arab groups; in February 2016 they played a major role in closing the Azaz corridor by taking minakh and Tal Rifaat from their former rebel colleagues with indirect help from the Syrian army and Russian aviation.26 Jaish al-huwar eventually joined the SDF, which is composed of former FSA Kurds, PYD/ YPG forces, and local Arabs. meanwhile, the battle for Kobane had another crucial outcome for the PYD besides boosting its local reputation: it made the group Washington’s main ally against IS from 2015 onward, much to the dismay of Turkey and the decomposing FSA.27 U.S. arms, air support, and other assistance have enabled PYD forces to extend their territorial control in the east and strengthen their inluence on the Kurdish population. moreover, Arab brigades that wish to ight IS in the Euphrates region have been compelled to join the PYD-led SDF coalition if they want to receive U.S. weapons. Some Arab militias have taken the alternative path of seeking Turkish rather than American support, for instance by backing Ankara’s September 2016 intervention in northern Syria. Yet this path is more limited than the SDF option because “pro-Turkish” rebels are bound by Ankara’s agreements with Russia. In the interest of protecting its client regime in Damascus, the Kremlin has restricted how far these brigades can advance into Syria, while the U.S.-backed forces have advanced as far south as Deir al-Zour on September 2017. he PYD’s Path to Socialism Oicially, the PYD aims to promote an original political and economic system in northern Syria based on selfsuiciency and local democracy. Its ap- plication has been limited because the Kurdish leadership is more concerned with the uniication of Rojava and the war against IS. Regardless, the new authorities are working to implement this PKK-inspired ideology on the ground. Abdullah Ocalan has written a series of books proposing a societal model for the Kurdish people, and his “Ocalanism” was directly inspired by American social theorist murray Bookchin, a radical leftist who wanted to break down capitalism through “libertarian municipalism.” Bookchin believed that capitalism’s fatal law lay in its conlict with nature—that is, by destroying the environment, capitalism will inevitably create a major crisis and consume itself. In his irst book on radical ecology,28 he recommended the decentralization of polluted metropolitan centers and pesticide-ridden industrial farms so that people could live on a smaller scale, produce their food locally, use renewable energy, and manage their own afairs. On the latter front, he recommended democratizing urban neighborhoods by empowering citizen assemblies. hese assemblies could then confederate at different levels: city, region, and so forth. hey would send delegates to confederal councils to coordinate and administer policy. Power would be based among the people, who would be directly represented at the top. In time, he theorized, confederal municipalities would become a counterweight to the nation-state, and  FABrICE BAL ANChE capitalism would naturally disappear. municipalities would expropriate major economic resources and “municipalize” the economy, which the community would take possession of. Bookchin never succeeded in popularizing his ideas among other radical leftists, anarchists, or ecologists. Yet Ocalan became his ideological disciple in 2004 after reading several of his works in prison. he PKK leader had been seeking a new ideology for the Kurdish people after renouncing marxismLeninism. At the 2005 PKK congress, Bookchin’s “libertarian municipalism” became the group’s oicial ideology. Today, the PYD seeks to implement Ocalan’s municipalist principles in Syria. Although the party does not oicially plan to build a Kurdish state there, it does seek to create a democratic and ecologically responsible society within the framework of a federal system. In 2013, the PYD began to administer territories under its control by establishing the movement for a Democratic Society (TEV-DEm), a coalition of civil associations and political factions such as the Syriac Union Party (SUP), the Syrian Kurdish Democratic Peace Party (PADKS), and the Kurdistan Liberal Union Party (PYLK). In November 2013, TEV-DEm announced the creation of its own autonomous administration for Rojava and its three cantons. At the head of each canton is a Kurdish prime minis- ter assisted by two vice prime ministers, most of whom are non-Kurdish (e.g., Sunni Arab or Christian). Although the PYD is an avowedly Kurdish party pursuing a clear sectarian policy, Rojava’s prime ministers are careful to respect diversity. he three cantons are supposed to be governed by an elected assembly that controls Rojava’s executive bureau, but elections have not yet taken place (as of this writing, TEV-DEm has scheduled them for January 2018). Any such elections would be contrary to Bookchin’s model of creating municipalities that elect delegates to confederal councils. In Rojava, such municipalities are known as communes (komun in Kurdish), each containing roughly 150 houses and around a thousand inhabitants. An elected communal council manages relations between individual villages and the established authorities; the latter oficials still run local public services such as water and electricity, since the administrative framework of the prewar municipalities has not disappeared. Ideally, new municipalities would arise naturally from the communes, but in reality the new and old structures exist in parallel. he communes deliver certiicates to the population for bread and fuel at low prices; they also supervise the local community and participate in its political education. his corresponds roughly with the village “committees” of Communist China. S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r  In addition, Rojava’s communes are supposed to organize economic life by promoting cooperatives. In the countryside, farmers are organized in groups of ifteen and asked to work together and exchange surplus production with other cooperatives, including in the cities. his practice is in line with the goal of designing communes to be self-suicient, with the eventual aim of eliminating traders and money while establishing a bartering system. Yet skepticism is warranted about the principles behind these measures and their application on the ground. Until recently, wartime disruptions had pushed Rojava’s population to organize a subsistence economy, and the Kurdish zone’s isolation created practical reasons to favor self-suiciency. Yet now that overland links are reopening, this policy can only be justiied on the ideological level. In the agricultural sector, the new authorities in Jazira canton want to reduce their share of cereal and cotton production to make room for activities that would make local communities better able to feed themselves, such as market gardening and arboriculture. To efect this change, large estates and public lands need to be entrusted to the population and organized in cooperatives. Yet the people seem unlikely to embrace such a system. he TEV-DEm program might seduce the landless peasants of Jazira, to whom the PYD plans to distribute former public domains, but it is unpalatable to existing owner-farmers, who would no doubt prefer to continue working individually. moreover, market gardening requires much greater personal investment than cereal farming, which is incompatible with the collectivist spirit the PYD has sought to inculcate. meanwhile, industry is almost absent from all of the cantons, mainly because the Assad regime preferred to keep things that way for “security reasons.” As described previously, only two cotton mills were built in the Hasaka/Jazira area; no other large-scale industrial facilities exist. To ill this gap and meet local needs, PYD authorities would like investors to develop agro-food and manufacturing industries. Yet attracting foreign investors into such an anti-capitalist system would be diicult. Entrepreneurship is encouraged in Rojava, but only within the framework of cooperatives. Similarly, engineers and technicians are needed to work for the “revolution,” but individuals with the necessary training tend to leave Rojava because salaries are too low there. moreover, many young men fear conscription and prefer to take refuge in Iraq. he middle classes in particular are experiencing this demographic hemorrhage, since liberal professionals and entrepreneurs are largely excluded from the economic system currently being set up. he application of Ocalan’s theories is still modest in Rojava’s economic sphere, as the PYD is aware that it risks  63 alienating a large part of the population, especially those who only rallied to the group for fear of IS. he reopening of land communications with the Assad regime zone has encouraged a return to the lucrative exportation of cereals and cotton. moreover, manufactured goods from the regime zone will likely lood Rojava markets before any local production could develop. Accordingly, local authorities may resort to protectionism to defend the cantons economically, perhaps by imposing tarifs and cutting of the western Syrian market. If the PYD’s cooperative economic system fails due to these pressures, the party would have two choices: coerce locals into accepting Ocalan’s theories or declare a “pause” in implementation due to wartime circumstances, much like Vladimir Lenin did with the Soviet Union’s New Economic Policy in 1920. In the irst scenario, the “communalization” of Rojava’s economy would entail the expropriation of property belonging to certain social groups, namely, constituencies deemed opponents of the PYD. his property would then be redistributed to the party’s own base with the objective of strengthening its inluence and eliminating the Assad regime’s. Such eforts would also indicate a separatist mindset, despite the federal model the PYD has been outwardly promoting. In the second scenario, a “pause” in economic collectivization would likely spur the PYD to renounce its intention 64  of changing Rojava society, and to further normalize relations with Damascus instead. he Kurdish cantons would then be reinstated in the Syrian economic space, and impediments to private initiative would be lifted. Whichever approach the party chooses, the local population—Kurdish and non-Kurdish—will be more inclined to accept the pursuit of some form of autonomy if their living conditions improve. he PYD and Arab Loyalty many inhabitants of PYD territory are not Kurdish, so the group will need to address any signs of internal sectarian revolt if it hopes to maintain its control over Rojava. his likely means allowing self-government for local Arabs and refraining from any attempts at forced Kurdiication. At present, the PYD’s governing policy appears lexible enough to accommodate such an approach, but once the Islamic State is defeated—that is, when the common enemy disappears—building strong relations with Arabs inside Rojava will pose a greater challenge. In their southward progress toward and past Raqqa, the SDF have received additional backing from anti-IS Arab clans. hese forces have joined up with the PYD-led coalition not out of any inherent ainity, but because the SDF is the lone repository for American weapons. For its part, the U.S. military has sought to avoid repeating the mistake of training Arab forces only to see them fall apart, though discrepancies in its statements about the SDF’s Arab contingent are puzzling. On August 31, 2017, Army general Stephen Townsend, commander of the Combined Joint Task Force–Operation Inherent Resolve, estimated the SDF’s total manpower at around 50,000 ighters and noted that half of them are Arab.29 On December 8, 2016, however, Air Force spokesman Col. John Dorrian stated that only 13,000 of the SDF’s 45,000 ighters were Arab.30 Other estimates of the Arab contingent are much lower, but even if Dorrian’s igures were accurate, they represented a surprisingly large increase since the SDF’s creation in October 2015, when Arabs constituted only 5,000 out of 30,000 total ighters.31 Dorrian and Townsend’s igures also seem at odds for reasons other than timing. It is diicult to believe that 25,000 Arab ighters are currently in the SDF because that would mean the number of Kurdish ighters has decreased from 32,000 to 25,000 since December 2016. Attributing that drop to losses from the Raqqa battle is not tenable—more likely, the U.S. military does not want to specify the actual number of Arab ighters because they are still a minority. Whatever the case, if the SDF gained 5,000 new ighters between December 2016 and August 2017 (from 45,000 to 50,000), and if all of them are Arab, then the current Arab contingent should be 18,000 at most, or around 36%—a far cry from “half.” Realistically, only some of the new recruits were Arabs, so their true proportion is probably onethird of the SDF. In August 2016, the launch of a new Turkish-supported rebel operation known as Euphrates Shield created dissension inside the SDF, with several Arab tribes in northwestern Syria leaving the PYD-led coalition to join the campaign. Among other objectives, Turkey sought to peel ethnic Turkmens away from the SDF, and it had the money and arms to do so. Going forward, the extent to which it can draw more Arabs from the SDF depends on whether it unduly favors Turkmen ighters over Arabs, and on how deeply it decides to intervene in Syria. Euphrates Shield oficially ended in march 2017 when the battle for al-Bab came to a close, but Ankara still hopes to push the SDF east of the Euphrates River. Another Turkish campaign has been underway since October, oicially against Arab jihadists in Idlib province, but in reality against the PYD’s Afrin stronghold.32 Selective Turkish support to some Arab factions could revive a tribal war for local power and delay the inal victory against IS. For the PYD, then, the big question is how to retain the loyalty of its Arab allies in the long term. he group has partly addressed this issue by permitting Arab self-governance in parts of Rojava and coopting local leaders in Arab councils while keeping military power  FABrICE BAL ANChE for itself. Arab intertribal divisions and the SDF’s monopoly on U.S. support have also kept many clans from turning against the PYD. In Afrin, the group appears to be receiving Russian support,33 giving local Arabs further incentive to partner with it. Yet the opposition regularly accuses the PYD of engaging in ethnic cleansing against Arab and Turkmen populations in Rojava, and the group’s leaders have only partially dispelled these concerns. he Kurds have clearly sought to strengthen their demographic weight in Rojava. In October 2015, Amnesty International denounced the PYD for preventing some displaced Arab villagers from returning to their homes near Tal Abyad and Tal Tamer.34 In response, the party claimed that it had to secure the area irst, since the proximity of IS forces made it impossible for civilians to return safely. By march 2017, the Arab village of Tal Fweida near Tal Tamer was still empty.35 his community had supported IS during its February 2015 ofensive against Assyrian villages along the Khabur River; when IS later withdrew, the Arab population followed. he same process apparently occurred around Tal Abyad when the SDF liberated it from IS in June 2015. Kurds have indisputably practiced ethnic cleansing in cases where local Arabs helped Islamist ighters displace Kurds and seize their properties. In spring 2013, after PYD forces briely captured Tal Abyad, IS ighters de- stroyed Kurdish villages around the city and expelled Kurdish civilians from the district. Such acts drew retribution when the PYD retook the city in 2015, despite reassurances to the contrary by the group’s leader, Salih muslim. In a 2013 interview, muslim had stated that his militias would ight jihadist groups but would not force out local Arabs, whether transplants or natives: here are three types of Arabs among us: there are those with whom we have always lived and fought alongside. We defend the brotherhood among these people. here are those who do not belong, Arabs who came from outside, other countries or the area, the jihadists who have burned our home and decapitated Kurds. Finally, there are the Arabs who were moved to Kurdistan by force by Haiz alAssad...to Arabize the area. hey are victims...and we advocate a peaceful solution for these populations. hose who can return to their hometowns should do so and the others can live in peace with the Kurds.36 muslim was referring to the “Arab belt” created in the 1960s when the regime moved Arab populations from the Raqqa area closer to the Turkish border. hat move involved around 25,000 people at most—a relatively small number given Rojava’s current population of around 2 million.37 Yet the Baath regime inlamed sectarian tensions disproportionately over the course of decades by showing favoritism toward local Arabs S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r  while neglecting Kurdish communities. Such tensions are no doubt complicating the PYD’s bid for hegemony in the area today. Economic dynamics could widen Arab/Kurdish issures as well. At present, Rojava is sufering under an economic embargo maintained by Turkey and the KRG. he restrictions have pushed many middle-class Kurds to emigrate, allowing the PYD to gain strength by pursuing control over a working-class economy rooted in handicrafts and agriculture. For those Kurds who remain in Rojava, accepting material sacriices is an acceptable cost for living in a region ruled by their PYD kinsmen. Yet the Arab population does not share this sentiment, so they would more willingly turn against the party if other political or economic alternatives arise. Critics of Kurdish rule emerged in manbij38 after the SDF seized it from IS in summer 2016, most likely because Turkey restricted humanitarian access to the area.39 In contrast, such access was encouraged after Turkey defeated IS in Jarabulus on August 2016. Ankara no doubt wants to show residents of PYD-held manbij and Tal Abyad that they can have a better quality of life under non-Kurdish leadership.40 he July 2017 assassination of manbij councilman Sheikh Shlash Ghaniemi41 shows the extent to which the area is being violently contested. In short, the local environment is complex, and any actions by outside players can have unforeseen repercussions. While arms and money might hold together a coalition against IS, establishing stability afterward will be a great challenge. Building better relations between Kurds and Arabs will not be easy, but it is the only way to forestall future sectarian clashes in the north. A dIFFICuLT FuTurE FOr kurdISh AuTONOMy many questions remain about the future of Rojava, and most of the answers will depend on how the region’s authorities handle local sectarian issures. he territory is stretched into a thin 800-kilometer-long strip, bordered to the north by a hostile Turkey and to the south by Arab populations who thus far refuse to recognize a reversal of power in the Kurds’ favor. he Kurdish population is not a majority in the Euphrates Valley, where the SDF is ighting IS, so Rojava authorities are unlikely to annex that area post-Raqqa. Yet they would certainly prefer that the Arab tribes who do assume control there are willing to maintain good relations with Rojava. In addition to snuing out the IS threat, this is likely a major reason why the SDF have been deployed into the Euphrates Valley—to cultivate and empower Arab allies. Inside Rojava, the PYD successfully imposed itself by force and through fear of IS. Now that the latter threat is wan 65 ing, however, the group will need to set up a viable government while mulling the major risks inherent in committing the population to a socialist project comparable to the Chinese Cultural Revolution of the 1960s. One can only be skeptical about the sustainability of such an economic system, even if the population is willing to accept the constraints of following the path traced by Ocalan. On the geopolitical level, the PYD seems caught on a narrowing path between the United States and Russia. American support against IS allows the Kurdish group to strengthen its territorial grip, placate local Arabs, and muzzle the various Kurdish opposition elements backed by the Iraqi KDP. Yet it is uncertain whether Washington will continue supporting the PYD once the Islamic State is eliminated, especially in light of Syrian Kurdish links with the PKK, a U.S.-designated terrorist group that has been at war with NATO ally Turkey since the 1970s. Russia will likely agitate the Turkish threat against Rojava if the Kurds fail to get along with Arab factions inside or outside their territory. At the same time, moscow appears eager to keep the PYD in play (albeit in weakened form) because the group is a useful lever against the Assad regime and Ankara. As for the PYD’s oft-proposed federal project for Syria, it is merely a chimera—the Kurds seem interested in it 66  only as a vehicle for establishing Rojava’s autonomy, while both the Sunni Arab opposition and the Assad regime reject it completely. In truth, PYD leaders are focused on maximizing the state of grace they have earned through ighting IS. hey are rushing to consolidate their power and to Kurdify Rojava, as seen in the expulsion of any hostile Arab populations that supported IS (whether directly or tacitly) and the implementation of Kurdish educational curricula. hose Arabs who want their children to receive an Arabic education now have to leave Rojava (private Christian schools are allowed to teach in Arabic, but only to Christians42; this policy is the PYD’s way of preserving its image as a protector of Christians). he future of Rojava depends on the PYD’s ability to manuever between Washington and moscow, obtaining beneits from both parties while avoiding military moves that alienate either one. Yet the group’s margin for error is increasingly slim. he August 2016 Turkish intervention in Jarabulus showed how quickly Russia could lose patience with the Kurds. And the intervention that Ankara launched in northern Idlib in October 2017 is unfolding near the borders of Afrin—another example of Russian-Turkish coordination unfavorable to the PYD. Turkey’s hostility will make it diicult for Syrian Kurds to achieve levels of development comparable to the KRG in Iraq. If the Assad regime ultimately prevails over the Sunni Arab rebellion and reestablishes control over more of Syria, the Kurds would be forced to withdraw into their sectarian strongholds of Afrin, Kobane, and Qamishli, where they might enjoy some autonomy but would have to use regime-controlled roads to communicate between cantons. form], in André Raymond, ed., La Syrie d’Aujourd’hui (Paris: National Center for Scientific Research, 1980), http:// books.openedition.org/iremam/744. 7. Fabrice Balanche, “Le programme de modernisation de l’irrigation en Syrie” [The modernization of irrigation systems in the Syrian northeast], Méditerranée 119 (2012), https://mediterranee.revues.org/6499. 8. According to author discussions with Abdullilah Arabo, vice president of the Movement for a Democratic Society (TEV-DEM), and Mohamad Dirok, vice president of economic policy for Jazira canton, Qamishli, March 2017. 9. According to a January 2011 survey conducted by the author under the auspices of the Plan for the Modernization of Irrigation in Syria, a joint venture between the Syrian Ministry of Agriculture and the French Development Agency. NOTES 1. 2. Jacques Weulersse, Le pays des Alaouites [The country of the Alawites] (Institut Français de Damas, 1940). Fabrice Balanche, “Clientélisme, communautarisme et fragmentation territoriale en Syrie” [Clientelism, communitarianism, and territorial fragmentation in Syria], A Contrario (March 2009), https://www.cairn.info/revue-a-contrario-2009-1-page-122.htm. 3. These judgments are based on the author’s personal observations and interviews conducted in Syria between 1990 and 2011. 4. Interestingly, the SCP has been divided between two camps since 1986: a Kurdish faction that supports the official party leadership (which Bakdash’s widow assumed after his death in 1995), and a Christian faction that has opposed some of the leadership’s policies (e.g., Bakdash’s decision to reject perestroika near the end of the Soviet era). The sectarian nature of the split is quite evident and reflects other minority trends in Syria. 5. “Statistical Abstract 2009,” Central Bureau of Statistics, Damascus, 2010. 6. Francoise Metral, “Le monde rural Syrien à l’ère des réformes (1958–1978)” [The Syrian rural world in the age of re- 10. “Syria: International Energy Data and Analysis,” U.S. Energy Information Administration, June 24, 2015, https:// www.eia.gov/beta/international/analysis.cfm?iso=SYR. 11. Author interview with a Syrian businessman in Beirut, March 2016. 12. Author interview with travel agency that books flights to Qamishli, March 2017. 13. Author interviews with doctors and pharmacists in Qamishli and Hasaka, March 2017. 14. “Statistical Abstract 2009,” Central Bureau of Statistics, Damascus, 2010. 15. Dexter Filkins, “Iran Extends Its Reach in Syria,” New Yorker, June 9, 2017, http:// www.newyorker.com/news/news-desk/ iran-extends-its-reach-in-syria. 16. Michael Knights and Bilal Wahab, “Resetting U.S. Relations with the Kurdistan Region,” PolicyWatch 2886 (Washington Institute for Near East  FABrICE BAL ANChE Policy, November 2, 2017), http://www. washingtoninstitute.org/policy-analysis/view/resetting-u.s.-relations-withthe-kurdistan-region. 17. Jordi Tejel, “Les paradoxes du printemps Kurde en Syrie” [The paradoxes of the Kurdish spring in Syria], Politique Etrangere (Summer 2014), https://www. cairn.info/revue-politique-etrangere2014-2-page-51.htm?1=1&DocId=9441 4&hits=10+8+5+. 18. “Syria’s Assad grants nationality to Hasaka Kurds,” BBC News, April 7, 2011, http://www.bbc.com/news/worldmiddle-east-12995174. 19. Author interviews with bedoon in Hasaka province, March 2017. 20. Kyle W. Orton, “The Error of Arming the Syrian Kurds,” New York Times, June 6, 2017, https://www.nytimes. com/2017/06/06/opinion/syria-kurdsisis-raqqa.html. 21. Jordi Tejel, “Syria’s Kurds: Troubled Past, Uncertain Future,” Carnegie Endowment for International Peace, October 16, 2012, http://carnegieendowment.org/2012/10/16/syria-skurds-troubled-past-uncertain-future/ e2nt. 22. Cyril Roussel, “Les Kurdes de Syrie et le projet du Rojava: rêve éphémère ou espoir durable?” [The Kurds of Syria and the Rojava project: an ephemeral dream or a lasting hope?], MaghrebMachrek 4, no. 222 (2014), http://www. cairn.info/revue-maghreb-machrek2014-4-p-75.htm. 23. “‘PYD Political Thought Resembles That of Baath Party’: Kurdish Politician,” ARA News, May 14, 2014, http:// aranews.net/2014/05/pyd-politicalthought-resembles-that-of-baath-party-kurdish-leader. 24. Mohammad Ballout, “Kurds Caught in Crossfire in Northwest Syria Battle,” Al-Monitor, November 14, 2012, http:// www.al-monitor.com/pulse/security/01/11/kurds-caught-in-the-crossfireof-battles-for-northwestern-syria.html. 25. Fehim Tastekin, “Turkey’s real target on way to Idlib: Kurds,” Al-Monitor, July 2, 2017, http://www.al-monitor.com/ pulse/originals/2017/06/turkey-syriakurds-real-target-on-way-to-idlib.html. 26. See Fabrice Balanche, “Kurdish Forces Bolster Assad in Aleppo,” Policy Alert (Washington Institute for Near East Policy, July 29, 2016), http://www.washingtoninstitute.org/policy-analysis/view/ kurdish-forces-bolster-assad-in-aleppo. 27. “Erdogan spokesman slams U.S. military for backing Syrian Kurdish militia,” Reuters, January 12, 2017, http://www. reuters.com/article/us-mideast-crisissyria-turkey-usa-idUSKBN14W0MZ. 28. Murray Bookchin (writing under the pseudonym Lewis Herber), Our Synthetic Environment (New York: Knopf, 1962). 29. U.S. Department of Defense, press briefing by Lt. Gen. Townsend via teleconference from Baghdad, August 31, 2017, https://www.defense.gov/ News/Transcripts/Transcript-View/Article/1297228/department-of-defensepress-briefing-by-general-townsendvia-teleconference-fro. 30. U.S. Department of Defense, press briefing by Col. Dorrian via teleconference from Baghdad, December 8, 2016, https://www.defense.gov/News/ Tr a n s c r i p t s / Tr a n s c r i p t - V i e w / A rticle/1025099/department-of-defensepress-briefing-by-col-dorrian-via-teleconference-from-bag. 31. “Kurdes et Arabes créent les Forces Démocratiques de Syrie” [Kurds and Arabs create Syrian Democratic Forces], L’Orient le Jour, Oct.r 12, 2015, https:// www.lorientlejour.com/article/949062/ kurdes-et-arabes-creent-les-forces-democratiques-de-syrie.html. 32. Aron Lund, “Turkey Intervenes in Syria: S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r  What You Need to Know,” IRIN, October 9, 2017, https://www.irinnews.org/ analysis/2017/10/09/turkey-intervenessyria-what-you-need-know. 33. Fehim Tastekin, “Is Turkey rattled by Russian-Kurdish deal?” Al-Monitor, March 24, 2017, http://www.almonitor.com/pulse/originals/2017/03/ turkey-syria-russian-troops-arrive-afrin. html. See also Barbara Slavin, “Syrian Kurdish leader: Moscow wants to work with us,” Al-Monitor, October 8, 2015, http://www.al-monitor.com/pulse/originals/2015/10/kurdish-leader-pyd-russia-support-syria.html. 40. Haid Haid, “Post-ISIS Governance in Jarablus: A Turkish-led Strategy,” Chatham House, September 2017, https:// www.chathamhouse.org/publication/ post-isis-governance-jarablus-turkishled-strategy. 41. “Arab tribal leader killed in bomb blast in Syria’s Manbij,” ARA News, July 22, 2017, http://aranews.net/2017/07/arabtribal-leader-killed-bomb-blast-syriasmanbij. 42. Author interview with Christian authorities in Qamishli, March 2017. 34. “‘We Had Nowhere to Go’: Forced Displacement and Demolitions in Northern Syria,” Amnesty International, October 13, 2015, http://www.amnesty.org/en/ documents/mde24/2503/2015/en. 35. Based on the author’s observations during a visit to the area. 36. “Kurds Seek Autonomy in a Federal Syria: Top Official,” Agence FrancePresse, December 2, 2013, http://www. mei.edu/content/kurds-seek-autonomy-federal-syria-top-official. 37. Jordi Tejel Gorgas, “Les Kurdes de Syrie, de la ‘dissimulation’ à la ‘visibilité’?” [The Kurds of Syria, from “concealment” to “visibility”?], Revue des mondes Musulmans et de la Méditerranée 115-116 (December 2006), https://remmm.revues.org/3022. 38. According to a June 2017 author interview with a Manbij resident who had been staying at a refugee camp in Lebanon since 2016. 39. “Manbij Civilian Council calls for international aid for IDPs, warns of humanitarian disaster northern Syria,” ARA News, May 20, 2017, http://aranews. net/2017/05/manbij-civilian-councilcalls-for-international-aid-for-idpswarns-of-humanitarian-disaster-northern-syria.  67 4. r A dICA LI ZAT ION OF T h E S u N N I A r AB rEBEL L ION N EArLy ALL OF the groups involved in the revolt against Bashar alAssad are Sunni Arab, and that descriptor is important because it deines their primary motivation for taking up arms: namely, defending the Sunni community by ousting a nonSunni regime. hey are of course driven by a host of social, economic, and political grievances as well, but these same grievances exist in all other Syrian sects without pushing them into open, sustained revolt. many Christians, Alawites, Druze, Ismailis, and Shia are dissatisied with the regime, but they have not risen up against it en masse. In contrast, large swaths of Syria’s Sunni Arab majority rebelled in 2011, and tens of thousands of foreign ighters have iniltrated Syria in the name of defending Sunni Islam, toppling the Alawite regime, and ighting its Shia allies, thereby reinforcing the insurrection’s sectarian character in a communal and religious sense. he Sunni rebels are divided into myriad ighting groups, drawing mainly from the country’s Sunni Arab majority but attracting many Turkmens as well, along with a multiethnic rainbow of foreign Sunnis. (Syrian Kurds are a special category, since they are more focused on safeguarding their Kurdish rather than Sunni identity and never fully embraced the goal of toppling the regime, seeking partial autonomy inside Assadruled Syria instead.) he Islamic State is the largest and most well structured of these groups. Some analysts may deem it iconoclastic to discuss IS in the same chapter as the “Sunni rebellion,” but the organization does its recruiting in the same ethnoreligious category as many radical jihadist rebel groups and espouses ideologies similar to those of Jabhat al-Nusra (JN), the al-Qaeda afiliate that has come to dominate much of the rebellion on the ground.1 (Note: JN changed its name to Jabhat Fatah alSham in 2016 and has since rebranded itself as part of a coalition called Hayat Tahrir al-Sham. But for simplicity’s sake, this chapter refers to the group as JN throughout, regardless of which time period is being discussed.) Indeed, prior to 2013, IS and JN were the same organization, and members of IS participated in rebel coalitions with more moderate factions.2 he subsequent war between IS and other Sunni groups was one of a series of fratricidal clashes that have come to characterize the rebellion in general.3 Of course IS does possess a number of unique traits, so this chapter discusses the group in a separate section and refers to the other factions collectively as “the Sunni rebellion” for convenience’s sake, rather than specifying “the non-IS Sunni Arab rebellion.” Sunni rebels control disparate territories that lack geographical continuity. Some are small pockets of resistance within areas controlled by the Syrian army, such as Rastan-Houla, mazraat Beit Jinn (close to the Golan Heights), and East Ghouta, but the largest rebel zones have not been fully surrounded because they lie on or near the borders with Turkey and Jordan, two countries that have supported the revolt. his agglomeration of territories has been  68  greatly reduced since summer 2012, when the rebels held some 80% of the country and seemed close to winning the war, while regime-controlled territory was highly fragmented. he situation is drastically diferent today—rebel areas are now separated from each other, while the regime controls a largely contiguous swath of territory in the western and central regions. Since 2011, various attempts to organize the Sunni rebellion have failed. he Free Syrian Army (FSA), an early umbrella group, lost most of its ighters to ephemeral, inefective coalitions formed with funding from Turkey, Qatar, and Saudi Arabia.4 In January 2015 these same countries helped create Jaish al-Fatah (the Army of Conquest), a more eicient JNled coalition that conquered the cities of Idlib province in a few short weeks and seriously threatened Latakia, Aleppo, and Hama. Yet the Russian intervention later that year put an end to the group’s expansion, and JN has since eliminated or alienated many other Sunni groups in its quest for hegemony over the rebellion. NuMBErS, IdEOLOgIES, & rEgIONAL FrONTS As the number of Sunni opposition groups continues to grow and the coalitions that house them change in composition and name, predicting the rebellion’s dynamics has become more and more diicult. Religious homogeneity has not been enough to give the rebels the military cohesion and uniied political identity they so sorely need. A closer look at the numerical, ideological, and geographical scope of this fragmentation can help observers better understand the rebellion and assess whether it still has a chance to prevail, or at least survive. A march 2016 report by Jennifer Cafarella and Genevieve Casagrande at the Institute for the Study of the War (ISW) categorized twenty-three of Syria’s hundreds of rebel groups as the main “powerbrokers” and “potential powerbrokers” in the armed opposition.5 Not much has changed since that assessment, despite the fall of Aleppo and other highproile military developments, so the ISW’s igures greatly inform the following discussion. At the time, these two powerbroker categories encompassed around 90,000 ighters in total. he report described a third category of groups with a few hundred ighters each. While most of the twenty-six factions in this category do not profess any ideology, several of them are linked to al-Qaeda: S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r  36. SYRIAN REBEL GROUPS BY IDEOLOGY, March 2016 POWERBROKER GROUPS 90,000 fighters 20% IDEOLOGY 25% Transnational Salafi-jihadist Syrian Salafi-jihadist Political Islamist 24% ALL REBEL GROUPS Minimum estimate: 100,000 fighters Secularist 31% ALL REBEL GROUPS Maximum estimate: 150,000 fighters 19% 20% 30% 45% 20% 28% 22% 16% BASED ON AUTHOR RESEARCH AND DATA FROM JENNIFER CAFARELLA AND GENEVIEVE CASAGRANDE, SYRIAN ARMED OPPOSITION POWERBROKERS, MIDDLE EAST SECURITY REPORT 29, MARCH 2016, INSTITUTE FOR THE STUDY OF WAR.  69 namely Jund al-Aqsa, Harakat al-Fajr alSham al-Islamiyah, Imarat al-Qawqaz i al-Sham, and a brigade called “al-Nusra Front al-Kavkaq.” he report also outlined a fourth category composed of hundreds of smaller groups with a few dozen ighters each. hese factions correspond to local clans, and their main objective is to protect their given neighborhood or village; they are incapable of launching ofensives. Estimating the total number of ighters in the third and fourth categories is diicult. he best approximation is between 10,000 and 60,000. In total, then, the Sunni rebellion could have anywhere from 100,000 to 150,000 ighters. he ISW usefully classiied all of these groups into four ideological categories: transnational Salai-jihadists (i.e., al-Qaeda-linked ighters), national Salai-jihadists, political Islamists, and secularists. he diference between national jihadists and political Islamists is more or less akin to the diference between Salaists and the muslim Brotherhood—in simpliied terms, the former seek strict application of Islamic law, while the latter tend to favor a state with an Islamic civil constitution but protections for religious freedom. As for the “secularists,” the term is applied very loosely because most of the ighters in this category are conservative muslims who do not actually want a secular government. Among the 90,000 “powerbro70  37. SYRIAN REBEL POWERBROKERS BY NUMBER AND IDEOLOGY, March 2016 TURKEY Tigris River NORTHEAST ALEPPO ? 4% NORTHWEST SYRIA 47,000 33% 35% 28% Mediterranean Sea ? HOULE-RASTAN Euphrates River LEBANON 5% 24% DAMASCUS AREA 20,000 IRAQ 71% 4% 4% 4% ISRAEL 87% TYPOLOGY OF REBEL FACTION SOUTH 23,000 INHABITED AREAS OF CONTROL 100 km 0 JORDAN UNINHABITED AREAS OF CONTROL Transnational Salafi jihadist Syrian Army Syrian Salafi jihadist PYD (Kurdish) PYD (Kurdish) Political Islamist Other rebels Other rebels Secularist Islamic State Islamic State Province border Syrian Army International border ? No data  FABrICE BAL ANChE 38. SYRIAN REBEL POWERBROKERS BY FRONT, March 2016 TURKEY NORTHWEST SYRIA Hizb al-Islam Fawj al-Awl al-Turkistani Jaysh al-Mujahideen Jabhat al-Nusra Nour al-Din al-Zenki Ahrar al-Sham Firqat al-Sultan Murad Jabhat al-Shamya Jaysh al-Naser Faylaq al-Sham Firqa 13 ALEPPO NORTHEAST ALEPPO Ahrar al-Sham Jabhat al-Shamya Nour al-Din al-Zenki Firqat al-Sultan Murad Firqa 13 Jaysh al-Naser Tigris River HASAKA RAQQA IDLIB LATAKIA DEIR AL-ZOUR HOULE-RASTAN Jabhat al-Nusra Ahrar al-Sham Harakat Tahrir Homs HAMA Mediterranean Sea TARTUS HOMS Euphrates River DAMASCUS AREA Jabhat al-Nusra Faylaq al-Rahman Aliwya Ahmed Abdou Jaysh al-Islam Ahrar al-Sham Jabhat al-Asl wal-Tanmya Ajnad al-Sham LEBANON IRAQ DAMASCUS QUNEITRA Golan Heights ISRAEL DERAA SUWAYDA SOUTH Jabhat al-Nusra Ahrar al-Sham Jaysh al-Islam Ajnad al-Sham Jaysh al-Awl Thalalaf Suquor al-Janoub Usoud al-Harb Faylaq al-Awl Seif al-Sham 100 km 0 JORDAN TYPOLOGY OF REBEL FACTION INHABITED AREAS OF CONTROL UNINHABITED AREAS OF CONTROL Fatah al -Sham Transnational Salafi jihadist Syrian Army Syrian Army Province border Ahrar al-Sham Syrian Salafi jihadist PYD (Kurdish) PYD (Kurdish) International border Faylaq al-Sham Political Islamist Other rebels Other rebels Province center Jaysh al-Naser Secularist Islamic State Islamic State Locality S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r  ker” rebels, ISF classiied some 20% as transnational jihadists, 31% as national jihadists, 24% as political Islamists, and 25% as “secularists.” When the thousands of rebels in the non-powerbroker categories were added, the “secularists” became the largest grouping, but they are also the most fragmented and therefore the least efective. he Sunni rebellion’s external supporters have failed to establish a single, uniied operational headquarters from which rebel coalitions can coordinate large-scale ofensives. Western governments, Saudi Arabia, and the Qatar-Turkey alliance each have their individual clients: the West mainly inances “secularists” while Riyadh, Doha, and Ankara fund political Islamists and Salaists. he operations rooms established in Jordan and Turkey do bring these external partners together in support of secularist and political Islamist factions, but the assistance coming from these centers is less important than the direct aid given by Arab Gulf countries, which has helped marginalize the secularists. he division among donors has combined with internal ideological diferences to cause multiple confrontations between rebel groups. In its various manifestations throughout the war, alQaeda ailiate JN has systematically eliminated groups that opposed its hegemony in the northwest, especially those linked to the FSA. In spring 2016, JN played a role in the bloody war that Jaish  71 al-Islam and Failaq al-Rahman waged against each other in the suburbs of Damascus, which allowed Assad’s army to retake half of East Ghouta over a sixmonth period.6 Later that year, ighting between Jund al-Aqsa and Ahrar alSham—both former members of the Jaish al-Fatah coalition—was largely responsible for the failure of a rebel ofensive against Hama.7 Disappearance of the FSA he attempt to transform numerous rebel groups into a genuine army, the FSA, fell apart as early as 2013. Saudi Arabia, Qatar, and other foreign donors prefer to inance groups directly according to their ideology, and Syria’s clan structure has fostered competition to obtain this external inancing. In September 2012, a Swiss journalist in Geneva provided the author with an edifying example of this disunity from Jabal al-Akrad: A Saudi delegation arrived with suitcases full of dollars to inance rebel groups. A faction had just taken possession of an observation post of the Syrian army, proving that it was eficient, and consequently that Saudi money was well used. he Saudis had ilmed the ight to report to donors, like humanitarian associations. A competing faction also wanted to prove its ability to beneit from the inancial windfall, but the attack on the military post on the Nabi Younes (near Slonfeh) resulted in a disaster: 72  half of the group was killed or injured because the attack had been ill-prepared, without any collaboration with the other factions since the objective was simply to assert themselves. he FSA General Staf, made up of defecting Syrian army generals, was never able to coordinate the operations of these groups, lacking the legitimacy and resources to exert such wide authority. In the end, “Free Syrian Army” became merely a label aixed to certain rebel groups by the military Operations Centers in Jordan and Turkey, which sought to coordinate rebel activities and foreign supply eforts.8 To receive the label, groups must adhere to a charter of “good behavior” and respect for secular and democratic values. In reality, most FSA groups are Islamist—only Firqa 13 (Division 13), formed early in the war by army defectors, objectively meets the secular-democratic criteria.9 his confusion has often extended to media reports about the FSA’s importance. many Syrians who stayed in rebel territory or led the country as refugees have told reporters that the FSA is present everywhere, but this impression was based on the mistaken notion that all rebels who are not from JN or IS are members of the FSA. he situation has been further complicated by reports of FSA ighters defecting to JN; in 2014, for instance, Rami Abdulrahman, head of the Syrian Observatory for Human Rights, noted that “dozens of ighters” had left the FSA-aligned Syrian Revolutionaries Front for JN during the latter’s takeover of Idlib province.10 he FSA regrouped into about ifty factions in 2016, by which time its presence inside Syria had been greatly reduced. Its only signiicant contingents were located in Aleppo city (which of course ended when the city fell that December), further north in Aleppo province around Azaz and marea (under Turkish protection since August 2016), in the greater Damascus area, and in Deraa province under the umbrella of the Southern Front (though some FSA factions in the south entered into a tactical alliance with the Islamist coalition Jaish al-Islam). Regional Fronts In the absence of a single command, rebel factions meet in regional coalitions; once they accomplish a given military objective, they usually separate and renegotiate their participation in a new operation. he coherence of these coalitions mainly depends on the sustainability of external inancing and the dominant group’s ability to maintain unity. he most sustainable and efective coalition so far has been Jaish alFatah, which JN created in the northwest in February 2015. It continues to grow thanks to its military successes and its coercive power over other groups. JN has been trying to reproduce this formula on other fronts as well. The Southern Front. Established in February 2014 by the Amman military Operations Center, this umbrella group comprised 23,000 ighters as of march 2016, dominated by ive “secularist” groups.11 Yet a report issued by researcher Aron Lund12 a month after the coalition’s creation argued that it existed only on paper, and that its ive main groups were merely a weak federation of village militias. Whatever the case, the front’s eforts to take Damascus in 2014 and then Deraa in June 2015 both failed. After September 2015, its military activity was limited to the rivalry with IS. damascus area. As of October 2017, an estimated 10,000 rebel ighters remained in the Damascus area, down from around 20,000 in march 2016. Hailing from Jaish al-Islam, Failaq al-Sham, and JN, most of them remain encircled by regime forces in East Ghouta. One by one they have lost towns in the western suburbs (e.g., Daraya, moadamiya, Qudsaya, Wadi Barada). hey had hoped to take the capital, but all their ofensives after summer 2012 failed, and they lost hope of reaching this goal or even being rescued from the south. Some factions in Daraya and Qudsaya therefore made deals to be exiltrated to rebel-controlled Idlib province—an arrangement the regime was willing to countenance because it could more easily reestablish its authority over  FABrICE BAL ANChE N DAMASCUS Mezzeh military airport W Mount Hermon LEBANON E Daraya 19 Dec 2015: Airstrike kills Hezbollah official Samir Kuntar Jaramana Qatana S Nov 2014: Battle between Jabhat al-Nusra and Druze militia Fall 2016: Syrian Army seizes Khan al-Sheikh Arneh Nabatieh Shebaa 28 Jan 2015: Hezbollah attacks Israeli convoy Sahnaya Sayyida Zaynab Damascus International Airport AREA CONTROLLED BY al-Kiswah al-Khyam Mazraat Beit Jinn 39. MILITARY SITUATION IN SOUTHERN SYRIA, May 2017 Khan al-Sheikh Marj Ruhayyil military airport Syrian Army Hader Islamic State (IS) Majdal Shams UNDOF CONTROL Buqata Mar 2017: Syrian Army seizes IS positions Israel MADINAT AL-BAATH Khalkhalah military airport Quneitra UNDOF crossing Sanamin RELIGION Sunni Mar 2017: Rebels seize IS positions in Lajat area Mazraat Amal GOLAN HEIGHTS ISRAEL Other rebels (including Jabhat al-Nusra) Druze 18 Jan 2015: Israeli airstrike kills Hezbollah and IRGCInkhil figures Christian Qazrin Shiite Safed International border Nawa Province border Shahba Izraa 1974 Golan ceasefire lines and DMZ Sheikh Maskin Sea of Galilee Border crossing Jamlah al-Shajara 21 Jun 2016: Israeli airstrike on IS antiaircraft battery JORDAN Dael Province center al-Muzairab al-Thaala military airport SUWAYDA Spring 2017: Rebels launch an offensive on Deraa District center Town Event DERAA Main road al-Qurayya Irbid 2 Mar 2016: Jordanian raid kills seven IS militants Secondary road Busra al-Sham Airport Salkhad 0 20 40 km S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r  Advances by IS affiliate Yarmouk Martyrs Brigade (LSY)  73 N W Rajo Bab al-Salam 40. MILITARY DEVELOPMENTS IN NORTHWEST SYRIA, March/April 2017 TURKEY Mar 22: Russian troops support PYD in Afrin E AZAZ S Marea AFRIN Tal Rifaat AREA CONTROLLED BY Jandeiris AL-BAB TURKEY Tadef Mar 29: Syrian army seizes Deir Hafer from IS AL-SAFIRAH al-Jinah Mar 31: Ongoing rebel offensive toward Kinsabba Deir Hafer Kefraya IDLIB Syrian army Other rebels MAIN ETHNIC GROUP PREWAR al-Fua Darkoush PYD (Kurdish) OFFENSIVE LAUNCHED BY Mar 16: U.S. strike reportedly hits mosque Qorqanya Kasab (closed) Islamic State Kuweires ALEPPO March 21–27: U.S. airstrikes kill al-Qaeda officials Other rebels Hama frontline 3/27/17 Bab al-Hawa HAREM Syrian army Nubl Apr 12: Scheduled evacuation of al-Fua and Kefraya JISR al-SHUGHOUR Saraqeb ARIHA Arab Sunni Arab Alawite Kurdish Sunni Christian Turkmen Sunni Arab Shiites Arab Druze Kinsabba MAARAT AL-NUMAN International border AL-HAFFAH Province border Border crossing LATAKIA QARDAHA Khan Sheikhoun Province center Apr 4: Chemical attack on Khan Sheikhoun Suruj JABLEH AL-SUQAYLABIYAH BANIYAS Town Souran Main road Secondary road MAHARDA Mar 27: Syrian army blocks rebel offensive MASYAF 74  District center Military airport open HAMA 0 10 20 km Notable event Urban area  FABrICE BAL ANChE Beit Sahem in southern Damascus. he Qalamoun-area towns of al-Dumayr and Jayrud reached a truce with the army, so neither can be considered a rebel stronghold anymore. residents of these suburbs if it ofered local rebels a chance to survive. he remainder are concentrated in East Ghouta because the army has retaken most other pockets, aside from Yarmouk and The Northwest. he focus of the rebellion is now in the northwest, where nearly 47,000 “powerbroker” ighters were active as of march 2016.13 he number has likely increased to around 60,000 as of this writing thanks to transfers of rebels from the Damascus area, Zabadani, Arsal, and al-Waar (a neighborhood of Homs). Jihadist and Islamist factions represent three-fourths of these 41. MILITARY SITUATION IN NORTHERN SYRIA, May 2017 N AREA CONTROLLED BY W E KOBANE Syrian Army S Islamic State JARABULUS TURKEY Other rebels al-Shuyukh Tahtani Bulbul Raju Kurdish forces Ghandoura Bab al-Salam Euphrates River MAIN ETHNIC GROUP Sharran AZAZ Maabatli Souran Dabiq AFRIN Sheikh Hadid Arab Sunni Sarrin al-Shemali al-Raai Kurdish Sunni MANBIJ Marea Turkmen Sunni Tal Rifaat Um al-Qura Jandeiris Qabasin Arima Arab Shiite Nubl al-Zahra AL-BAB Haritan International border Province border Rasm al-Harmel al-Dana HAREM Ain Issa Tishrin Dam Tadef Darat Izza Bab al-Hawa Arab Druze Abu Qilqil Khafsa ALEPPO Khan al-Asal Atareb Border crossing Province center Kuweires Qorqanya al-Jarniya Deir Hafer District center AL-SAFIRAH Armanaz Town Lake al-Jaboul al-Fua Taftanaz Aleppo urban area Maskanah Hader Benan Main road Lake Assad IDLIB Secondary road Saraqeb Military airport open ARIHA THAWRA Tal Daman Thawra Dam Khanaser Abu Duhur Ehsem S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r  0 10 20 km Military airport closed Bridge open Bridge destroyed  75 personnel. JN and its allies have essentially built an Islamic emirate in the Idlib area by gradually eliminating or integrating other groups. Since the 2015 Russian intervention, however, even rebels on this front have been forced to withdraw at times. heir territorial losses have not been large, but they have been strategic—key regime areas in Latakia, Hama, and Aleppo are no longer threatened. In the Aleppo area, JN and its allies gradually took over most rebel operations in 2016, and the local Fatah Halab coalition was seemingly marginalized. Led by JN forces, Jaish al-Fatah advanced toward Hama and Latakia in order to divert the army from Aleppo city. his gambit ultimately failed in a military sense once Aleppo fell, but it did help JN reairm its hardline stance against the Assad regime and attract other rebels disappointed by the lack of support from outside allies. Northeast Aleppo. he 5,000 rebel ighters present in the Azaz-marea pocket as of march 2016 were reinforced by thousands of pro-Turkish ighters from Idlib province once Operation Euphrates Shield began that August, including the Turkmen brigade Firqat Sultan murad. he main objective of their ofensives was to create a safe zone between Azaz, al-Bab, and Jarabulus and thus prevent the Kurds of Rojava from unifying their cantons along the northern border. 76  A few thousand Turkish soldiers supported these rebels, who were regrouped into the “Euphrates Shield” coalition. Turkey announced the oicial termination of Operation Euphrates Shield in march 2017, though it has reportedly launched a new campaign since then (see chapter 3). 42. CLOSE ALLIES OF JABHAT AL-NUSRA, June 2016 houla-rastan Pocket. A few thousand rebels are surrounded in this enclave between Homs and Hama. he leading group appears to be Harakat Tahrir Homs, a faction headquartered in Rastan and classiied as political Islamist.14 It competes with Jaish al-Sham in Talbisah and with a local coalition in Houla. Yet the situation is calmer in this region than in East Ghouta. he rebels occasionally launched northward ofensives to assist with the battle for Hama and link up with Idlib province, but to no avail. In the long term, the latter effort is their only chance of avoiding the fate of their brethren in Ghouta, especially if the regime takes direct action against this enclave. AL-QAEdA’S EVOLuTION ANd TAkEOVEr During winter 2013–2014, JN and other rebel groups pushed the Islamic State out of western Aleppo province, even as IS drove them out of eastern Aleppo and the Euphrates Valley. his war against IS was wrongly hailed as a second revolu- Suqur al-Sham Liwa al-Haqq Liwa al-Haqq (Homs) Jaish al-Sunna AHRAR AL-SHAM Fajr al-Khilafa Jabhat Ansar al-Din JABHAT AL-NUSRA Jund al-Aqsa Caucasus Emirate in al-Sham Jamaat Ahadun Ahad Junud al-Sham Fajr al-Sham Islamic Movement East Turkestan Islamic Movement Jamaat Ansar al-Islam Kataib Ansar al-Sham  FABrICE BAL ANChE 43. JABHAT AL-NUSRA AND COALITIONS, Winter 2015–Fall 2016 JAISH AL-NASER (Victory Army) ANSAR AL-SHARIA JAISH AL-FATAH (Army of Conquest) JABHAT AL-NUSRA JAISH AL-FATAH IDLIB (Idlib Army of Conquest) SYRIAN DEMOCRATIC FORCES (YPG and Revolutionary Army) S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r  FATAH HALEB (Aleppo Conquest) tion by many analysts,15 who spoke of a burst of “democratic forces” against the jihadists.16 In fact, it was only an internal conlict between rival jihadists: JN and its local allies wanted to eliminate IS before it could do the same to them. By getting rid of their main competitor in the northwest, they could focus on unifying the area’s other rebel groups by force. hose who refused were either expelled (e.g., Division 13) or eliminated (e.g., al-Hamza Front). As JN’s hegemony became inevitable, the group took control of the Jaish alFatah coalition.17 In spring 2015, Jaish al-Fatah’s advances in Idlib, Jisr alShughour, and Ariha shook the Assad regime. he jihadist contribution was fundamental to these victories; suicide bombers from JN’s fellow al-Qaeda afiliate Jund al-Aqsa played a major role in opening access to Idlib city. Since that successful ofensive, the six subgroups of Jaish al-Fatah have shared control of the province’s towns. Idlib city was divided into sectors, but JN and its closest ailiates appear to have full control there. In a may 27, 2015, interview with Al Jazeera, JN leader Abu muhammad al-Julani declared his intention to create an Islamic emirate in northwestern Syria,18 conirming analysis by expert homas Jocelyn one year prior.19 In pursuit of that goal, the group has used brutal methods similar to those of IS—the only diference is that executions in Idlib are not advertised as  77 openly as they were in the IS capital of Raqqa. most non-Sunnis have led the area, though as described in previous chapters, some were forced to convert. Unfortunately, even conversion did not prevent the massacre of twenty Druze in Qalb Lawzah in June 2015. JN ighters also regularly arrest local civil society activists, many of whom disappear.20 militarily speaking, JN and its afiliates are able to intervene pretty much anywhere in Idlib province without much opposition from whatever groups may be oicially controlling a given village or city.21 most maps illustrating the distribution of rebel factions tend to conine JN to two speciic areas in Idlib—the strongholds of Harem and Jabal al-Zawiya22—but the organization is actually present throughout the northwest. JN has espoused a consistent and clear political ideology for Syria. Territories under JN control generally respect the judgments issued by its courts and seem to respect the integrity and courage of its ighters—hardly the case for other rebel groups. hus far, JN has tolerated the work of foreign NGOs and the UN’s Oice for the Coordination of Humanitarian Afairs, but only because it needs these social services to help establish its political power. Humanitarian assistance is currently channeled through the Bab al-Hawa border crossing, which Ahrar al-Sham held until a July 2017 ofensive gave JN control over the border area. 78  Fertile Ground for Salaism JN’s strength lies in the geographic diversity of its recruitment, unlike the moderate groups and their highly localized recruitment. his approach has helped JN largely avoid the clan quarrels that often undermine the rebellion (except in Deir al-Zour, where it had to compete with IS for recruits from different tribes). Another strong point for the group is its ideological power, which helps it maintain legitimacy in many parts of the northwest despite its frequent brutality against locals. he Salafist strain of Islamist thought that JN espouses has been spreading throughout the countryside of Idlib and Aleppo for several decades via battalions of young imams trained in Saudi Arabia. his brand of Salaism is more relatable to the underdeveloped rural populations in northern Syria than the muslim Brotherhood’s ideology, which is more suited to the urban middle class. here are some exceptions. Kafr Nabl is often cited as an example of local democracy and secularism, widely highlighted by journalists visiting the north, while the “little moscow” of Jabal al-Zawiya has always been a leftist stronghold. Yet these exceptions are not reproducible at the provincial level, and Kafr Nabl’s free space is steadily being reduced. In January 2016, for example, JN nearly assassinated Raed Fares, a famous civil society activist and head of Kafr Nabl radio. He was forced to give a public mea culpa for broadcasting music on his station, and none of the local rebel groups protested the crackdown for fear of retribution from JN.23 Nevertheless, the radio station continues to operate, and locals still use the Syrian revolutionary lag during demonstrations.24 he Fake Break with al-Qaeda On July 28, 2016, JN leader Julani and al-Qaeda leader Ayman al-Zawahiri simultaneously declared that the two groups were no longer tied by an oath of allegiance.25 But this was merely a tactical maneuver to counter the July 15 U.S.Russian agreement to ight JN in Syria, which had greatly irked the Gulf states and Turkey. hese countries supported JN after its 2012 emergence in Syria, but in 2015 they began urging the group to break allegiance with al-Qaeda because the relationship was creating friction with Washington and complicating their eforts to supply the rebellion. Julani’s announcement a year later seemed to fulill that request, but a closer look at his speech and JN’s subsequent behavior points to a diferent conclusion. For one thing, he never explicitly stated that JN would break of ties with al-Qaeda. After praising the transnational group in the irst part of his speech, he announced the creation of a new group, Jabhat Fatah al-Sham, and stated that he had no “ailiation with an external organization,” implying al-Qaeda. Yet the radical Salaist ideology and terrorist methods that JN inherited from its parent group did not change at all afterward, despite the name change. moreover, the Zawahiri lieutenants who helped Julani’s group build a stronghold in northern Syria were not expelled from JN. Zawahiri’s remarks on the matter likewise included substantial qualiications. “You can sacriice organizational ties without hesitation if this endangers your unity,” he stated, reenacting the familiar farce of taqiyya (dissimulation) that terrorist organizations often use to keep up the low of external support while continuing their brutal operations.26 he timing of the supposed breakup was also telling. It was announced on the eve of a JN-led Jaish al-Fatah ofensive against Aleppo. With logistical and inancial support from Turkey, Qatar, and Saudi Arabia, the group’s forces temporarily broke the siege there, allowing a newly rebranded JN to present itself as liberator of the city’s eastern districts rather than a proxy of foreign jihadists. To be sure, the reopened Ramouseh corridor was impassable because of nonstop regime and Russian bombardment, and a Syrian army counterofensive that September soon closed the breach, leaving JN with considerable losses expended for a nearzero military outcome. Yet the group did achieve a key political objective by advancing into East Aleppo: the Obama administration was no longer willing to bolster Russia’s local ight against JN,  FABrICE BAL ANChE even indirectly, because this would have meant helping Assad win a major battle. In the end, however, various geopolitical developments since the fall of Aleppo indicate that JN’s 2016 rebranding was ultimately unsuccessful. he rapprochement between Turkey and Russia isolated the group on the international and national scene, and the arrival of the Trump administration seemed to usher in a more aggressive policy against it. At the Astana peace conference in January 2017, other rebel factions were speciically asked by the summit’s sponsors (Russia, Iran, and Turkey) to distance themselves from Jabhat Fatah al-Sham, which had been excluded from both the event and the associated ceaseire.27 he delegations were warned that they too would be considered terrorist groups if they refused to comply, and struck with the same vigor. For its part, Ahrar al-Sham refused to participate in the conference, but this may have been less a gesture of solidarity than a function of being surrounded by JN in Idlib province and therefore at the group’s mercy. he situation exacerbated Ahrar al-Sham’s internal tensions, since radical elements within the coalition had been pushing to merge with JN while other elements had sided with the group’s pro-Turkish leadership, which favored participating in Astana and avoiding public links with al-Qaeda. he West does not regard Ahrar alSham as a terrorist group, even though it has nearly the same ideology as JN. he oicers of both organizations were former fellow soldiers of Osama bin Laden, including Julani and Adel Radi Saqr al-Wahhabi al-Harbi for JN, and mohammed Bahaiya (aka Abu Khalid al-Suri) and Hashem al-Sheikh (aka Abu Jaber) for Ahrar al-Sham.28 Yet Ahrar’s internal issures have set the stage for more clashes and defections, and this is precisely what Assad and his allies want: inighting between weakened rebels that facilitates a regime takeover in Idlib. ThE IdLIB STrONghOLd After Raqqa, Idlib became the second provincial capital lost to the rebellion, falling to Jaish al-Fatah in spring 2015 along with the province’s two other major cities, Jisr al-Shughour and Ariha. Damascus had already lost control of the surrounding countryside in winter 2011–2012. According to the UN High Commissioner for Refugees, Idlib province had 2,080,000 inhabitants as of October 2017, including 984,000 IDPs.29 In sectarian terms, the fairly homogeneous population is around 95 percent Sunni Arab. To the north of Jisr al-Shughour, a Turkmen population lives close to the Turkish border. many Christians led during the countryside ighting of 2011–2012, while others left Idlib city in march 2015 as rebels took control. For example, the entire population of the Armenian village of Yacoubiyah left S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r  Syria and sought refuge in the NagornoKarabakh region.30 Ten Druze villages are scattered in the Jabal al-Summaq area; as mentioned previously, JN forced residents there to convert to Sunnism. he Shia towns of al-Fua and Kefraya have been surrounded by rebels since Idlib’s fall; residents proclaimed their support for the revolt early on, but JN cast doubt on their sincerity upon moving into the province and kept them under threat as leverage against Hezbollah. most civilians were later transferred outside the besieged cities in two waves (December 2016 and April 2017) thanks to an agreement with the rebels31 (see chapter 1). Before the war, 70 percent of Idlib’s population lived in the countryside, a fertile agricultural area that produces grains, olives, and winter vegetables, with individual irrigation allowing for additional crops in summer. For those who remained there, agriculture has provided a means to survive the turmoil. Yet the area had very little industrial activity prewar, and Idlib was long known as an administrative city, so the province depended heavily on Aleppo for manufactured products. Today, everything comes from Turkey, and supplies are severely limited because humanitarian convoys can only access the province through one crossing, Bab al-Hawa. For decades, the Assad regime practiced especially strict control of this part of the border due to the lingering Syrian claim on Turkey’s Hatay province. As a result, the Bab al-Hawa area was already crisscrossed by barbed wire, “no man’s land” zones, and mineields even before 2011. he situation became worse after 2015, when developments in the war greatly increased the refugee low and spurred Ankara to clamp down on Bab al-Hawa. Today, no crossings are permitted except in very select humanitarian cases—a sharp contrast with previous years, when Syrians could move in and out of Turkey whenever they needed refuge from the ighting. Downfall of Moderate Factions in Idlib By early 2016, more than sixty rebel groups were ighting in Idlib province and the western Aleppo area, ielding a total of some 60,000–90,000 combatants.32 JN was not the largest of these factions: its 10,000 ighters were only half that of Ahrar al-Sham at that time. Yet many other militias allied with JN out of fear, ideological sympathy, respect for the group’s battleield achievements, or other factors. In contrast, most of the secular and “moderate” Islamist groups were small, locally recruited units with a few hundred ighters or less each, so they had only marginal inluence despite accounting for as many as a third of all rebel forces in the area. he most important “moderate” unit was Division 13, a member of the Fatah Halab coalition. Supported by the United States and Turkey, the group claimed to ield 2,000 ighters who had battled the Syri 79 N W TURKEY Bab al-Salam 44. HAYAT TAHRIR AL-SHAM TERRITORIAL ORGANIZATION IN IDLIB AREA, August 2017 E S AZAZ Turkish-backed rebels AFRIN SDF SDF HTS LOCAL STRATEGY Jihad against Western countries AL-BAB Nubl Turkish border Border crossing TURKEY Bab al-Hawa HAREM Darat Izza Anadan Haritan Supply line ALEPPO al-Dana Syrian Army Kafr Nihal Salqin Sarmada AL-SAFIRAH Atareb IDLIB Darkoush Target HTS INTERNATIONAL STRATEGY Taftanaz Strong influence of Ayman al-Zawahiri Islamic Party of Turkestan Khirbet al-Joz IPT children’s training camp JISR al-SHUGHOUR Saraqeb ARIHA Yamadi TERRITORIAL CONTROL Abu Duhur Khanaser Town fully controlled Kansafra MAARAT AL-NUMAN al-Ziyarah Syrian Army Sinjar Town partially controlled HTS base LATAKIA Kafr Nabl Syrian Army Idlib rebel area Fundraising in Arab Gulf countries Tal Khanzir International border Qalaat al-Madiq Frontline Khan Sheikhoun Sarouja Sectarian line AL-SUQAYLABIYAH Souran BANIYAS MAHARDAH Main road Syrian Army Province center Mobilizing cause: destruction of Israel District center HAMA Islamic State 0 80  10 Town 20 km  FABrICE BAL ANChE Rajo N W Bab al-Salam 45. HAYAT TAHRIR AL-SHAM STRONGHOLDS IN IDLIB AREA, August 2017 TURKEY E AZAZ S Marea AFRIN Tal Rifaat AREA CONTROLLED BY Syrian army Jandeiris Kafr Nabl AL-BAB Anadan TURKEY Darat Izza Bab al-Hawa HAREM HTS and allies Kuweires Other rebels Kafr Ni’al Sarmada Salqin Haritan ALEPPO al-Dana Islamic State Tadef PYD (Kurdish) Atareb Deir Hafer AL-SAFIRAH MAIN ETHNIC GROUP PREWAR Darkoush Kasab Khirbet al-Joz Yamadi Taftanaz al-Fua IDLIB Arab Sunni Kurdish Sunni Kefraya Turkmen Sunni JISR al-SHUGHOUR Saraqeb ARIHA Arab Druze Khanaser Abu Duhur Kinsabba Arab Alawite Christian al-Ziyarah Kansafra MAARAT AL-NUMAN Arab Shiites Senjar AL-HAFFAH Kafr Nubul International border Province border LATAKIA QARDAHA Tal Khanzir Qalaat al-Madiq Border crossing Khan Sheikhoun Province center Suruj JABLEH AL-SUQAYLABIYAH District center Souran Town BANIYAS MAHARDAH Main road Secondary road HAMA MASYAF 0 10 20 km Military airport open Aleppo urban area S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r   81 an army in Aleppo and Kurdish forces in the Azaz-Jarabulus area, where they had been integrated into Turkey’s Euphrates Shield efort. moreover, after the Russian intervention, many small groups not ailiated with JN put aside their Islamist banners and other trappings and joined coalitions with nationalist names such as Jaish al-Nasr (the Army of Victory) and al-Jabha al-Shamiya (the Levant Front). hey did so not only to receive U.S. military aid, but also to resist JN’s hegemonic pressures. he regime’s December 2016 victory in Aleppo was a fatal blow to the already weakened moderate opposition, at least in the sense that they could no longer claim to represent a viable military or political alternative to Assad. he outcome strengthened the army by freeing up large numbers of troops, artillery, and Russian air capabilities, both to launch other ofensives and to better protect existing regime zones. It also led to additional Russian-brokered peace conferences in Astana, upping the diplomatic pressure on any Idlib rebels who opposed negotiating under the aegis of Assad’s military patron in moscow.33 At the January and may 2017 summits, Russia and Iran reached ceaseires with Turkey and select rebel groups, but these agreements were clearly of the “tactical, not strategic” variety, allowing the regime to consolidate its territorial gains and further divide the rebels. moscow in particular has seemed bent on using the 82  Astana process and its aftermath to present the rebels with a stark choice: assimilation or destruction. hose who choose the irst option would be integrated into the Assad regime’s political system and have the opportunity to proit from the money and power it has accumulated during the war. hose who continue to resist will be destroyed unceremoniously, in a replica of the method Russia used in Chechnya. JN has likewise used the Astana process to its advantage. In November 2016, the group attacked the Idlib-based forces of factions that sent delegations to the irst Astana conference, accusing them of collaborating with Russia—a claim that was not entirely unfounded given moscow’s objective of dividing the rebellion. And on January 28, 2017, soon after the second Astana conference and the fall of East Aleppo, JN spearheaded a reorganization of the rebellion in Idlib, merging with numerous other groups (e.g., Harakat Nour al-Din al-Zinki, Jabhat Ansar al-Din, Liwa al-Haqq, Jaish al-Sunna) to create Hayat Tahrir al-Sham (the Syrian Liberation movement, or HTS). his move also shook up Ahrar al-Sham. Although the organization itself did not join HTS, the majority of its ighters did, and its former leader, Hashem al-Sheikh, was tapped to lead the new coalition (though JN leader Julani remained in charge of the HTS military branch, ensuring his continued de facto control). Shortly after forming HTS, JN began to expel certain groups from its alliance and attack them, sometimes with adverse consequences. For instance, when it went after former ally Liwa al-Aqsa (formerly Jund al-Aqsa), various brigades merged with the Ahrar al-Sham coalition in protest, replacing many of the ighters Ahrar had lost to HTS. he two coalitions have since clashed regularly, whether for control of the Bab alHawa crossing or elsewhere As of August 2017, HTS constituted the largest rebel group in Idlib province and Syria as a whole.34 Of the 31,000 or so HTS combatants across the country, some two-thirds are situated in the Idlib area.35 he group often attracts additional recruits following military victories. In particular, more Ahrar al-Sham brigades joined up after a high-proile Idlib campaign in July 2017, including Usud al-Islam (Lions of Islam), based in the province’s southeast region, Usud al-maarat (Lions of maarat), based in maarat al-Numan, and a group in the small northern town of al-Dana. Further reinforcing HTS in Idlib have been ighters booted from enclaves retaken by the army, such as Zabadani, the al-Waar district of Homs, al-Qabun, Daraya, and others. Additionally, some 1,500 HTS ighters and their families may arrive from the Arsal area of Lebanon, where Hezbollah and Lebanese forces are conducting a campaign against Sunni jihadists. meanwhile, only one notable group has defected from HTS so far: Harakat Nour al-Din al-Zinki, which broke of to become an independent force in late July 2017. HTS Nodes of Control he group’s luid territorial presence is not indicative of its predominant strength in Idlib—HTS relies more on the potency of its network than on the accumulation of territory. In July 2017, it chased rival groups out of Idlib city and smaller towns such as maarat al-Numan, Saraqeb, and Atareb, all of which are former FSA strongholds. When occupying such communities, HTS notably practices discretion in order to avoid antagonizing locals. In the southern and eastern sections of the province, it has not moved beyond the regime military bases it conquered during the war, such as Abu Duhur. It also retains strongholds near Aleppo (in the northwestern suburbs), Hama (Khan Sheikhoun), and Latakia (Jisr al-Shughour), from which it can probe opportunities for expansion against the regime. Other areas not populated by Sunni Arabs are more diicult to control. In short, HTS does not seek territorial continuity but instead control of strategic points from which it can launch raids, including against villages such as Kafr Nabl, an anti-Islamist stronghold. Having elicited allegiance from local factions throughout the Idlib area, it can mobilize thousands of additional com FABrICE BAL ANChE 46. MILITARY SITUATION IN SYRIA AND IRAQ, May 2017 N W Tigris River Euphrates River TURKEY E Kobane Qamishli Manbij Hasaka S Afrin Aleppo Latakia Idlib Hama CYPRUS Tartus Raqqa Tal Afar Mosul Erbil Sinjar KDP al-Tabqa SYRIA Kirkuk Deir al-Zour Sulaymaniyah IRAN PUK Homs Palmyra LEBANON Ramadi Damascus Mediterranean Sea Baghdad Golan Heights Karbala IRAQ Najaf ISRAEL Amara JORDAN Basra SAUDI ARABIA EGYPT Persian Gulf KUWAIT AREAS OF CONTROL IN SYRIA AREAS OF CONTROL IN IRAQ Syrian army and militias Other rebels Iraqi army and militias Islamic State SDF (Kurdish) Islamic State KRG (KDP and PUK) PKK batants, as in the spring 2017 ofensive against army forces in Hama. he main HTS stronghold encompasses the border area with Turkey from Jisr al-Shughour to Bab al-Hawa. A lone weak point can be found in the Turkmen-inhabited countryside north of Latakia, a small Turkish protectorate that includes the Yamadi border crossing. Controlling the border is fundamental to the group’s assertion of dominance over Idlib province, since it gives HTS a monopoly over the transit of humanitarian aid. Various S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r  200 km 0 PKK-KDP groups engage in trade with regimeheld areas, but essential humanitarian aid comes almost exclusively from Turkey. moreover, if other factions lose access to weapons from Turkey, they may cross over to the regime side to avoid destruction. Major city Locality ThE ISLAMIC STATE Despite its ideological proximity to JN and other Islamist rebel groups, IS deserves a separate section because of its organizational originality and relative isolation. Unlike JN, IS was able to build  83 a uniied territory in eastern Syria that extended up to 90,000 square kilometers at its height in may 2015, after the capture of Palmyra. Combined with its conquests in Iraq, the group reigned over some 240,000 square kilometers, equivalent in size to the United Kingdom. his points to a fundamental strategic diference in how IS and al-Qaeda have approached the acquisition of territory—and, by extension, how they prioritize sectarian issues. he jihadis who formed IS generally believed in securing a large territory irst and then building international terrorist networks. his strategy emerged after bin Laden’s death and the dislocation of al-Qaeda’s network. Abu musab al-Zarqawi advocated it a decade ago when he headed the IS predecessor group al-Qaeda in Iraq, and it was likewise embraced by his disciple Abu Bakr al-Baghdadi, the self-proclaimed IS “caliph” who broke of relations with bin Laden’s successor Zawahiri. he strategy consists of taking root in a given area by joining whatever ight the local Sunni population is waging, rather than (or, in the case of IS, before) undertaking spectacular terrorist attacks against international interests. Zarqawi supporters led by Baghdadi followed that approach in proclaiming the irst “Islamic State” in Iraq in 2006. American troops were certainly one of their targets at the time, but their main enemies were Iraqi Shia and Kurds, based on the view that conlicts between local 84  communities outweighed the goals of universal jihad and destruction of Israel. As al-Qaeda’s chief ideologist, Zawahiri encouraged these more universal goals and criticized Baghdadi’s approach, arguing that it would lead to itna taiiyya (communal strife) with local Shia and distract Sunni Arabs from their supreme objectives. Both men had made clear that Shia were heretics who must be eradicated, but Zawahiri implied that their turn would come later, once more important goals were achieved.36 Nevertheless, when Baghdadi ordered the creation of a Syrian cell in 2011 with al-Qaeda’s blessing, his group had already established solid support bases in Iraq’s Sunni provinces of mosul and Anbar, both of which provided a steady stream of ighters and inancial resources. It had also set up its own tax system parallel to that of the Iraqi state. Local Sunni Arab residents felt marginalized by Baghdad and were still unhappy about the Shia community’s rise to power in 2003. Kurdish pressure on mosul, Kirkuk, and other disputed areas was a cause for concern as well. Although U.S. general David Petraeus had seemingly engineered the defeat of al-Qaeda in Iraq years earlier, this was a fragile victory that depended on maintaining the balance of power between the Shia-dominated central government and the northern Sunni Arab tribes who had greatly facilitated the counterinsurgency. When the last main contingent of American troops left Iraq at the end of 2011, Prime minister Nouri al-maliki began actively marginalizing Sunni Arabs, and this trend was reinforced by a Shia nation-building process. IS Blossoms in the Euphrates Valley In summer 2011, ighters who would eventually form JN began iniltrating eastern Syria, spurring Assad to deploy army units to Deir al-Zour province. Some of these ighters were Syrians who had gone to ight in Iraq after 2003 and were technically prohibited from returning home. When JN was formally established in early 2012, it was widely regarded as just one of many rebel groups that were proliferating as the opposition became increasingly militarized. Even as late as November 2012, the mainstream Syrian opposition protested when the United States designated JN as a terrorist group,37 while French foreign minister Laurent Fabius criticized the decision because JN was “doing a good job on the ground.”38 JN and proto-IS were a single outit until April 2013, when Julani, JN’s Syrian leader, had a falling out with Baghdadi, who hailed from Iraq. Various factors spurred the clash, including strategic divergence and inlated egos. he group’s Syrian members were apparently unhappy about being led by foreign ighters and wanted to command a movement that was ighting on their own soil. Yet Baghdadi argued that there was no longer any diference between Syria and Iraq; to prove his point, he declared that the groups were merging under a new title, the Islamic State of Iraq and al-Sham, invoking the ancient name for the territory occupied by modern Syria (the group would later shorten its name to just “the Islamic State,” reinforcing its desire to erase existing national boundaries). Julani rejected the move, so he and his supporters retained the name JN and the oath of allegiance to al-Qaeda. It was not an easy divorce because the two groups were deeply mixed in the provinces of Idlib, Aleppo, Raqqa, and Deir al-Zour. Unlike JN, however, IS did not have a signiicant presence in Damascus or Deraa province. And in winter 2013–2014, JN teamed with Ahrar al-Sham’s coalition, the Islamic Front, to try kicking IS out of Syria.39 As mentioned previously, they succeeded in chasing Baghdadi’s group out of Idlib and western Aleppo, but IS eliminated JN and other factions from Raqqa and Deir al-Zour, eventually taking over the entire Euphrates Valley and establishing Raqqa as its capital. With a third of their population illiterate, a birthrate of eight children per woman, and over 50% of their workforce tied to agriculture before the war, the provinces of Raqqa and Deir al-Zour sufer from low levels of development.40 he Baath regime failed to dissolve the  FABrICE BAL ANChE tribal system there, but it successfully manipulated local clans to its own ends by exploiting their reliance on patronage, including selectively distributed farmland and new irrigation programs that were highly coveted in this arid region (though threats of course played a role in their capitulation as well). From 1970 to 1990, the construction of alhawra Dam on the Euphrates and the creation of a vast irrigation system designed to develop eastern Syria absorbed nearly 20% of the national budget.41 Yet this development push was designed as a political strategy rather than an end in itself—once the regime had suiciently bought the loyalty of local residents, it cut much of the funding required to maintain and extend the irrigation program, leading to rampant water shortages and waste. his environment of high population growth, economic unrest, underdeveloped infrastructure, and archaic social patterns was particularly favorable to IS, which quickly eliminated local regime forces and secular opponents in the area, then imposed “authentic” Salaism on the largely Sunni Arab population. In fact, the traditional way of life practiced in much of Raqqa and Deir alZour before the war was not far removed from that preached by IS. he population already adhered to a particularly strict form of Islamic law, respecting the tenets of sharia more than the regime’s secular courts. Establishing Islamic courts was the group’s best means of seducing the people, particularly the lower classes, many of whom believed (at least at irst) that they could regain some of their dignity after years under a Baath legal system that tended to victimize or ignore them. According to anecdotal accounts from Syrian refugees, court justice in IS-held territory was faster at irst, rulings were executed immediately, and crime rates dropped noticeably,42 though any such improvements began to fade away quickly as corruption set in under IS rule.43 In any case, security is the primary demand of residents aflicted by years of ighting, and many of them do not care who provides it. IS was also careful to provide food to the population. Upon taking over the east, its irst actions included emptying the state’s grain silos, supplying bakeries, and compelling them to provide bread at modest prices; at one point bread cost less in Raqqa than anywhere else in Syria. Water was given freely to farmers on irrigated land, unlike the prewar years when drought, incomplete irrigation reform, and discriminatory economic liberalization hit agricultural areas hard. Over a two-decade period of instability, villages in northeast Syria expanded with no accompanying increase in public services or employment, while water scarcity and price increases prompted thousands of peasant farmers to leave their lands with no hope of inding work in the boomtowns.44 S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r  Paradoxically, this area is Syria’s main oil-producing region, but much like the Kurds in Hasaka province, residents of Raqqa and Deir al-Zour were immensely frustrated by the regime’s practice of taking local oil wealth without reinvesting it in local development. During the war, petrol sales have played a major role in the Islamic State’s inancial autonomy in the northeast. To be sure, IS received and redistributed signiicant funds from private donors and, at times, foreign governments, but the group did not want to become dependent on external resources, so it also taxed the movement of goods and coniscated the property of displaced populations.45 By spring 2015, Euphrates Valley oil wells between Deir al-Zour and Abu Kamal were providing it with an estimated $1–2 million in revenue per day.46 Crude and semi-reined oil from these wells was traded illegally to neighboring countries and even Syrian government territory, but local consumption generated the most revenue for IS. he group sold crude to small reineries that provided petrol to the roughly 2 million Syrians and 6 million Iraqis living in IS territory, who then used the fuel for vehicles, irrigation pumps, and generators. Yet Syrian oil production has steadily collapsed during the war, falling from 380,000 barrels per day in 2011 to less than 50,000 in 2014, and to nearly nothing outside Rojava by 2017.47 he international coalition against IS has bombed oil-well facilities and local re- ineries, reducing production to an unknown extent. here are no wells burning, so airstrikes have seemingly avoided targeting the wells themselves, hitting adjacent facilities instead. moreover, the reineries in IS territory are small and easy to rebuild; they are not comparable with Syria’s only oicial reineries in Baniyas and Homs, so IS or other actors could theoretically get them operational again if airstrikes are halted. Yet the group’s main source of income was not oil, but taxation on local trade and plundering of local resources, particularly the property of expelled residents.48 Such resources allowed IS to distribute bread and fuel to the people and pay high salaries to its ighters; as this economic largesse has dried up, so too has the group’s popular support. Assad and IS: Mutual Avoidance he Islamic State’s number one priority in Syria was to absorb Sunni Arab communities and rebel factions while creating a caliphate, not to topple the Assad regime. herefore, the group initially refrained from ighting the Syrian army. he situation changed in July 2014, however, when the regime’s air force launched signiicant bombing runs against IS forces for the irst time. hat shift, coupled with the group’s elimination or conversion of all other insurgent factions in the Euphrates Valley, apparently spurred a series of IS at 85 tacks against Assad’s military bases in Deir al-Zour and Raqqa. his included al-hawra air base, where the group’s quick success and brutal tactics traumatized the army—a day after state television had announced that the base was impregnable, IS ighters conducted suicide attacks to breach its heavy defenses and executed 150 soldiers after the battle. Previously, mutual avoidance between the regime and IS had led opposition leaders to accuse Assad of creating the jihadist group to divide and discredit the uprising. As proof of their claims, they cited the fact that the founders of IS had been released from Syrian prisons in 2011. he regime’s security services did in fact release known jihadists, and they undoubtedly believed that doing so would divide the opposition and scare of its Western supporters by accelerating the rebellion’s incipient radicalization. Yet most of these freed jihadists joined anti-Assad groups such as Ahrar al-Sham, not IS, and the regime certainly did not manufacture the jihadist movement or ship in the many foreign ighters that make up its ranks. Assad might have underestimated these operatives’ capabilities, but that does not necessarily mean he should be held accountable for their subsequent actions.49 Put another way, was King Abdullah II of Jordan aware of what he was doing when he released Abu musab al-Zarqawi during a general amnesty in 1999? 86  47. ENERGY PRODUCTION IN SYRIA N W Tigris River TURKEY E Euphrates River SUWAYDA OIL FIELD S Khabur Tishrin Suwayda Hasaka Aleppo Mediterranean Sea Thawra SHADADI OIL FIELD Raqqa Thawra Latakia Banias Baath RASAFA OIL FIELD Deir al-Zour Maharda Hama OMAR OIL FIELD SHAER GAS FIELD Bassel al-Assad Homs Palmyra LEBANON Damascus ISRAEL Golan Heights IRAQ Tishrin Deir Ali JORDAN 0 100 km DATA FROM INTERNATIONAL ENERGY AGENCY, DECEMBER 2016 Oil field Oil pipeline Hydroelectric power plant Prospective thermal power plant Gas field Gas pipeline Thermal power plant Oil terminal  FABrICE BAL ANChE In any case, the regime has had ample strategic reasons to avoid attacking IS as much as other groups during the war. First, IS forces are largely unable or unwilling to threaten the regime’s frontline between Latakia, Homs, Damascus, and Jordan; most direct clashes between the two have been limited to Deir al-Zour and Hasaka. In some areas, the steppe acts as a natural bufer between them. Second, IS has served a powerful motivation for hesitant or even hostile minority communities to join the regime’s forces, which they regard as the lesser of two evils. For example, when IS began threatening the Ismaili city of Salamiya east of Hama in spring 2014, the people petitioned Damascus for protection— despite their longstanding enmity toward the regime, their participation in anti-Assad riots since 2011, and their general refusal to enter military service. Assad replied that he would not help them unless they convinced the more than 20,000 local youths who had been dodging conscription to join the army or National Defense Forces militia. hey agreed, and Ismaili conscripts were soon trained and equipped to protect the city. he same scenario has been repeated in other locations threatened by jihadis. Even so, Assad has targeted IS whenever it threatens the western regime zone. In winter 2014–2015, IS began iniltrating the Damascus and Qalamoun region and demanding allegiance from local rebels. It also established networks in the south, convincing the Yarmouk martyrs Brigade to join it in spring 2014. It had bases north of Jabal al-Druze as well, at least until rebels seized Hawsh Hammad and remaining IS positions in the Leja area in march 2017. In addition, Palmyra has fallen into and out of the group’s hands over the past two years, usually depending on the army’s ability to divert troops from other lashpoints. IS forces captured the city in may 2015 to open the way to Damascus. he Russian intervention later that year and the entry of Iranian-led Shia militia reinforcements prevented the group from expanding past Palmyra and forced it to withdraw in march 2016. IS then retook the city that November when Assad’s forces were concentrating on Aleppo, but the regime pushed the group out again after winning that battle. Today, IS seems to be preparing refuges for its operatives throughout Syria so that it can continue to launch raids and terrorist attacks if the remaining “caliphate” in the east crumbles. Cleansing Kurds, Shia, and Christians As part of the group’s strategy of taking up local causes in order to enlist local support, IS forces in northeast Syria have sought to exploit a development that local Sunni Arabs regard as their greatest threat: the prospect of living under Kurdish control. Sunni Arabs tend to regard the wartime reversal of S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r  power in the area as unacceptable given their years of ruling over the Kurds. IS has therefore subjected Kurds in the Euphrates Valley and elsewhere to ethnic cleansing (e.g., in Tal Abyad in 2013). It does not matter that these Kurds are Sunni muslims—the group does not consider their form of Islam sincere, and the secular ideology of the main Kurdish powerbroker, the Democratic Union Party (PYD), is tantamount to apostasy for jihadis. Similarly, IS regards Syrian Shia as heretics and aims to eliminate them. his goal its with the general discontent that Syria’s Sunni majority feels toward decades of regime-sponsored Alawite domination, especially given the afinities between Shia and Alawi Islam. As described in chapter 1, Iranian missionaries fostered conversions in parts of the Euphrates Valley long before the war—some former Sunni communities became Shia, new Shia mosques sprung up in various towns, and a major Shia shrine was constructed in Raqqa. Once the area became oicial IS territory, however, the group massacred those Shia who had not already led, as seen in Hatlah village in June 2013. IS has targeted Christians in the Euphrates region as well. Raqqa, al-hawra, and Deir al-Zour have long been home to Christian communities, notably descendants of survivors from the 1915 Armenian genocide. In Deir al-Zour, IS destroyed the Armenian memorial complex and all Christian churches. he church in Raqqa was closed, and the few remaining Christian families were banned from practicing their faith. By this point, most Christians have seemingly led IS-held territory because their safety may come under threat at any time (e.g., Christian Assyrians in the Khabur Valley have been subjected to murderous raids and kidnappings50). Turning the Tribes Against IS As in Iraq, IS forces in Syria are made up of local ighters primarily recruited from lower-class Sunni Arab youths who dream of reversing the traditional tribal power structure for their own beneit. Yet the group has sought to iniltrate traditional tribal structures rather than overturn them completely, for instance by asking chiefs to marry their daughters to high-ranking IS oficers and send their sons to ight the group’s battles. As a result, tribes in eastern Syria have been driven in diferent directions by the Assad regime, outside actors, and their own self-interest. Untangling this complex web as part of a campaign to drive IS out of the Euphrates Valley will be diicult, potentially hindering any efort to pursue a counterinsurgency strategy based on Sunni Arab tribes, as General Petraeus did in Iraq a decade ago.51 In June 2016, Kurdish and Arab ighters from the Syrian Democratic Forces began a long southward cam 87 Tigris River al-Malikiyah Amuda Qamishli Peshkhabur TURKEY Kobane Jarabulus To Dohuk Ras al-Ain Tay Rmelan U.S. military facilities Tal Abyad Manbij Adwan Tal Tamer Balikh River Tishrin Dam Fadan al-Yarubiya To Mosul AREA CONTROLLED BY Jabbour al-Hawl To Aleppo 48. ARAB TRIBES AND MILITARY SITUATION IN NORTHEAST SYRIA, April 2016 Syrian Army Sinjar Mountain Islamic State HASAKA Khabur River To Mosul Other rebels To Aleppo RAQQA Walda Shammar Afadla Sabkha Thawra Dam Quayar PYD MAIN ETHNIC GROUP al-Tabqa Arab Sunni Baggara Kurdish Sunni al-Washeb Christian minority Fadan DEIR AL-ZOUR Sbaa To Salamiya Haddadin IRAQ Sbaa Mawali Secondary road Military airport open Military airport closed Bridge open Bridge destroyed International border Province border RAQQA Province center Manbij District center 20,000 100,000 HOMS al-Shour Fawaira POPULATION Ougaidat Euphrates River Bani Khalid To Homs Tribe territory Mayadin Jabal Abu Rujmayn Tribe name Main road 300,000 Palmyra Palmyra Russian military access Abu Kamal Ruwala N To Baghdad W E S To Damascus 88   FABrICE BAL ANChE paign to take Raqqa, launching attacks against IS positions in the Balikh Valley. Within a year, the SDF had surrounded the IS stronghold. Yet if the inal push to break the group’s inluence over the area is to succeed, the players involved will need a thorough understanding of the evolving role that Sunni Arab tribes have played there, irst under the Assad regime and then under IS. In Syria as in other countries, a tribe (qabila) is an ancestral network comprising anywhere from a few thousand to tens of thousands of members, subdivided into clans (ashayra). Although clans generally compete for control of their tribe, they show solidarity in the face of outside danger. During the Ottoman period, some tribal federations (e.g., the Shammar) were powerful enough to avoid paying imperial taxes, earning them the moniker “noble” tribes. hose who did pay were regarded as “common” tribes in the Bedouin hierarchy. many common tribesmen still harbor animosity against their “noble” counterparts after being dominated by them for centuries. While the noble tribes have retained their transnational nature (which includes strong patronage ties with Saudi Arabia), the common tribes are more rooted in the Syrian state and have largely abandoned their nomadic lifestyle. hese diferences shaped their responses to the 2011 uprising. When anti-regime protests irst broke out, Sunni Arab tribes with Saudi links took part, which explains why Deir al-Zour province rapidly shifted toward the opposition. Raqqa province remained loyal for longer because common tribes are more numerous there and had beneited from decades of Baath agrarian reform and development projects. Yet these projects allowed the regime to rent—not buy—the local tribes. Although the Raqqa region was loyal to Assad until as late as 2013, more recent agrarian policy failures and the inlux of money from Gulf countries ultimately pushed most Sunni Arab tribal leaders toward the opposition. Some sheikhs remained loyal and led to Damascus, however, where they enjoy protection because Assad hopes they will be useful if his forces retake the Euphrates Valley. he regime also employed military means to coopt tribes. he army began recruiting heavily in the area during the 1980s, and many residents came to view military service as a social elevator. hat same decade, Haiz al-Assad used the Haddadin tribe to quash the Sunni Arab revolt in Hama. Today, the Haddadin are helping his son ight rebel forces and protect the Aleppo supply road east of Hama, while prominent sheikh Fahd Jassem al-Freij serves as Bashar’s defense minister. Similarly, the younger Assad did not hesitate to distribute weapons to Arab tribes in Jazira when facing a Kurdish revolt in 2004. hese tribes—the Jabbour, Adwan, Tay, and Ougaidat— are still ighting alongside him today S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r  because they fear their enemies will take revenge if he loses power. his explains why the army has been able to control a pocket in the middle of Kurdish territory south of Qamishli. For its part, IS quickly integrated Sunni Arab tribes into its own system after establishing itself in eastern Syria.52 Local sheikhs were asked to pledge allegiance to the supposed “caliph”; those who complied were given oil wells, land, and other beneits, while those who resisted were attacked (e.g., IS killed 700 members of the Sheitat tribe in August 2014). Like the Baath regime before it, IS sought to create its own social base out of the traditional tribal hierarchy, and not just through the intermarrying mentioned above. If an elder sheikh was hesitant to cooperate, the group promoted a younger son or a secondary clan within his tribe. Indeed, the group mainly targeted adolescents, drawing them into camps for ideological indoctrination and military training. housands of young men have been radicalized and detribalized during the war, which will raise a serious rehabilitation problem once IS has been defeated. he group also used conlicts between tribes to impose its power. In Jarabulus, it supported the Tay against the Jais, forcing the latter to leave the city and seek refuge in Turkey. It is unclear what relationship the attackers had with the main Tay tribal stronghold on Syria’s far northeast border, though the war has at times spurred groups of tribesmen to leave their traditional territory. In August 2013, IS helped the Arab tribes of Tal Abyad defeat local Kurdish forces, expel Kurdish civilians from the district, destroy their villages, and redistribute their land to Arabs. Author interviews with local observers conirmed the group’s objective in such cases: to play on Sunni Arab fears of Kurdish irredentism. Going forward, even if all of the minority sectarian forces and foreign powers involved in the war make IS their main target, they will still need the local Sunni Arab population to reject the group if they hope to fully defeat it. Existing factors will help in this regard, such as local economic deterioration, the group’s heavy repression, and its gradual loss of legitimacy. Rallying the sheikhs will also require giving them money, political positions, and judicial immunity to replace what they received from IS. hey may be turned more readily given the group’s declining fortunes. he price controls in IS territory are no longer curbing inlation, especially as agricultural production decreases—a function of scarce fertilizers and pesticides, heavier taxation on farmers to compensate for decreasing oil revenue, and an irrigation system beset by infrastructure damage and mismanagement. Farmers in irrigated areas are subject to the same unpopular constraints they faced under Assad, but with less income. meanwhile, the myth of an IS 89 imposed Islamic order that provides justice to all of the faithful has faded. Recent interviews with refugees from Deir al-Zour and Raqqa provinces indicate that IS courts are just as corrupt as Assad’s, with the group’s members and their families receiving preferential treatment despite the occasional token execution of corrupt IS ighters. Conscription and enlistment of adolescents has provoked protests (e.g., in manbij in November 2015), spurring IS to increase repression to keep the population in check. Given their growing desire for revenge (intiqaam) and their traditional tendency to preserve tribal interests above all else, many individual Sunni Arabs and even entire clans are ready to help the SDF or Syrian army ight IS. For example, 200 members of the Sheitat tribe joined the army in Deir al-Zour after the 2014 massacre. Tribal collaboration also helped the regime retake Palmyra in march 2016, and tribal defections around Deir al-Zour and manbij multiplied as the army and SDF approached Raqqa. hus far, the United States has been backing the SDF via the Kurdish faction that leads it, the PYD. Such assistance— particularly the promise of air support and better weapons—is essential for encouraging tribes to join the anti-IS coalition. Arab tribes from the Fadan federation have already joined PYD forces in Raqqa province, while several Shammar 90  tribes in Hasaka province helped Kurdish units capture al-Hawl and al-Shadadi from IS in winter 2015. Despite the sharp sectarian divisions in other parts of the northeast, these tribes have always had good relations with the Kurds; for example, they refused to help Assad repress the Kurdish uprising of 2004. Yet they are relying on Washington to moderate the PYD’s hegemonic tendencies and ensure their own share of power once IS leaves. he same process of integrating Arab tribes into the SDF has been taking place in northern Raqqa province, but with many more obstacles. Some tribes remained iercely on the Islamic State’s side and still refuse to integrate into the SDF53 (e.g., the Afadla and Sabkha), while those who have been expelled from their lands by IS-backed tribes are not ready for quick reconciliation (e.g., the Jais and Sheitat). he level of violence has been so high since 2011 that traditional tribal measures for regulating it are no longer adequate—several clans and tribes will be forced to lee to avoid collective vengeance, such as the Tay in Jarabulus and the Sbaa in Sukhna (who originally helped IS capture Palmyra). FrOM FrACTurE TO FAILurE Today, the territory of the Sunni Arab rebellion is more divided than ever, much like the rebellion itself. All at- tempts to unify it have failed, whether by military or civilian groups. As early as 2012, local opposition committees tried to federate at the regional level and thus establish a nationwide organization, but this efort collapsed completely. he Syrian Interim Government set up in Gaziantep, Turkey, in early 2014 likewise aimed to bring the local committees together and build an administration for “liberated Syria,” but it failed as well, despite inancial support from Qatar. Powerful local authorities with their own discrete interests were the main factor in blocking the construction of a civil administration capable of unifying rebel territories, but resistance from individual rebel groups and destabilizing actions by the Assad regime also played a role. Since the Russian intervention and the reconquest of Aleppo, a number of isolated Sunni Arab areas where pro-rebel sentiment was strong have returned to the regime’s fold, such as Daraya, Qudsaya, and Wadi Barada. In some of these cases (like Qudsaya) they did so voluntarily—but only after witnessing the widespread destruction and depopulation of towns that fought to the bitter end (like Daraya). Assad’s attempts to regain control will be even more diicult in Sunni Arab regions where rebels hold larger swaths of territory, as in Idlib province. To win over villages there, the regime will need to await (or actively sow) internal divisions between rebel groups and communities, as happened in East Ghouta. In the south, Assad has established a stalemate with the regional rebel coalition, the Southern Front, which is disorganized, fractious, and likely unable to mobilize further large-scale ofensives. He may ultimately decide to regain inluence there by negotiation rather than conquest, even if that means empowering a small local warlord. Indeed, this template could probably be applied throughout much of Syria—Assad and his allies seem to believe that they have already come out on top because they can now negotiate peace individually with many of the remaining Sunni rebel groups. As for the two factions that would never surrender, JN and IS, the regime is capable of destroying them with continued outside help and seems bent on doing just that. Western actors looking to continue their intervention in Syria face similar complications in wooing local actors to their side, whether they limit their objectives to destroying IS or not. In general, any such eforts will necessitate some degree of involvement in local ethnic, religious, and tribal conlicts. In the north, for example, the Kurds are hardly ready to abandon their dream of uniting their three border cantons. Arab and Turkmen populations in these areas may therefore face a choice: accept their new minority status or leave the north. Convincing any of these parties to continue ighting IS post-Raqqa will require  FABrICE BAL ANChE acknowledgment of their local concerns, and even that may not be enough. Kurdish forces have little desire to advance much farther outside of Rojava, and no “moderate” Sunni Arab rebel partners exist outside the SDF, which remains under the YPG’s grip. he long-running battle for Deir alZour province is a good illustration of the Sunni Arab rebellion’s overall weakness, whether one looks at its numbers, popular appeal, or military efectiveness. U.S.-backed Arab rebel groups—namely, Usud al-Sharqiya (Lions of the East), Jaish maghaweer al-hawra (Rebel Commando Army), and Liwa Tahrir Deir al-Zour (Deir al-Zour Liberation Brigade)—number 2,000 ighters at most,54 so they clearly cannot defeat thousands of the most seasoned IS soldiers even if they were suiciently motivated to do so.55 heir June 2016 attempt to take Abu Kamal by air failed miserably when local tribes, fearing IS, did not carry out their expected revolt.56 Similarly, their spring 2017 ground advance from alTanf toward Abu Kamal was halted by a lash Shia militia ofensive. U.S. commanders have indicated a desire to transfer Arab rebels to al-Shadadi, where they could be integrated into the SDF ahead of launching an offensive in Deir al-Zour province from the north. Yet some of these ighters would likely defect to the regime instead of traveling to al-Shadadi.57 Besides, the PYD commanders who control the SDF refuse to support an ofensive headed by Liwa Tahrir Deir al-Zour, instead backing Quwat al-Sanadid (Sanadid Forces), a militia led by Shammar tribesman and close Kurdish ally Hamidi Daham al-Hadi. Without PYD collaboration, a U.S.-backed Sunni Arab force would struggle to challenge IS. Past attempts along those lines have ended in bitter failure, leading the United States to abandon training programs for such groups in order to concentrate its resources on the SDF. he PYD’s hesitance to bless an independent Arab force in northeastern Syria is partly rooted in its fears of a future competitor emerging once the Islamic State is defeated. he Kurds want to remain the only U.S. partner in the area, even if that means subordinating local Arab tribes. Such reasoning helps explain the PYD’s deployment against Raqqa, an almost wholly Arab city that could never be integrated into Rojava. Now that Raqqa is liberated, the Kurds realize that they may become a dispensable U.S. ally, so their calculations regarding Arab rebel groups could shift accordingly. Islamic Front] and Ahrar al-Sham have had an excellent working relationship with al-Qa`ida factions such as Jabhat al-Nusra, and regularly praise their contributions on the battlefield.” See “The Non-State Militant Landscape in Syria,” CTC Sentinel, August 27, 2013, https://www.ctc.usma.edu/posts/thenon-state-militant-landscape-in-syria. 3. Charles Lister, The Syrian Jihad: AlQaeda, the Islamic State, and the Evolution of an Insurgency (London: C. Hurst, 2015). 4. Aron Lund, “Syria’s Salafi Insurgents: The Rise of the Syrian Islamic Front,” Swedish Institute of International Affairs, March 2013, www.ui.se/globalassets/ui.se-eng/publications/ui-publications/syrias-salafi-insurgents-therise-of-the-syrian-islamic-front-min.pdf. 5. 2. Aron Lund, “Who Are the Soldiers of the Islamic State?” Carnegie Middle East Center, October 24, 2014, http://carnegie-mec.org/diwan/57021?lang=en. According to Aron Lund, “The [Syrian S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r  11. Cafarella and Casagrande, “Syrian Armed Opposition Powerbrokers,” http://www.understandingwar.org/report/syrian-armed-opposition-powerbrokers-0. 12. Aron Lund, “Does the ‘Southern Front’ Exist?” Carnegie Middle East Center, March 21, 2014, http://carnegie-mec. org/diwan/55054 13. Cafarella and Casagrande, http://www. understandingwar.org/report/syrian-armed-opposition-powerbrokers-0. 14. Ibid. 6. Aron Lund, “After Zahran: Rising Tension in the East Ghouta,” Carnegie Middle East Center, April 19, 2016, http://carnegie-mec.org/diwan/63376. 15. For example, see quotes from Thomas Pierret in Helene Sallon, “En Syrie, la ‘deuxième révolution‘ des insurgés” [In Syria, the ‘second revolution’ of the insurgents], Le Monde, January 2014, http://www.lemonde.fr/ proche-orient/article/2014/01/10/ladeuxieme-revolution-des-insurges-syriens_4345623_3218.html. 7. Saleem al-Omar, “Islamist Groups Ahrar al-Sham and Jund al-Aqsa Go to War,” Atlantic Council, October 13, 2016, http://www.atlanticcouncil. org/blogs/syriasource/islamist-groupsahrar-al-sham-and-jund-al-aqsa-go-towar. 16. Jean-Pierre Filiu, “La Syrie, ‘destination de rêve’ pour djihadistes Français” [Syria, a ‘dream destination’ for French jihadists], Souria Houria, April, 8, 2014. https://souriahouria.com/la-syrie-destination-de-reve-pour-djihadistes-francais-par-jean-pierre-filiu/. 8. Aron Lund, “The Free Syrian Army Doesn’t Exist,” Syria Comment, March 16, 2013, http://www.joshualandis. com/blog/the-free-syrian-army-doesntexist/. 17. “Syria: How a new rebel unity is making headway against the regime,” BBC News, May 1, 2015, http:// www.bbc.com/news/world-middleeast-32540436. 9. Thanassis Cambanis, “The Syrian Revolution Against al Qaeda,” Foreign Policy, March 29, 2016, http://foreignpolicy. com/2016/03/29/the-syrian-revolution- 18. See “[FULL] Interview Al Jazeera with Syaikh Abu Muhammad al Jaulani (27 Mei 2015),” YouTube https://www.youtube.com/ video, NOTES 1. Jennifer Cafarella and Genevieve Casagrande, “Syrian Armed Opposition Powerbrokers,” Institute for the Study of War, March 2016, http://www. understandingwar.org/report/syrian-armed-opposition-powerbrokers-0. against-al-qaeda-jabhat-al-nusra-fsa/. 10. Mariam Karouny, “Al Qaeda group seizes bastion of Western-backed rebels in Syria’s Idlib region,” Reuters, November 1, 2014, http://www.reuters. com/article/us-syria-crisis-nusra-idUSKBN0IL2QT20141101.  91 watch?v=1EVaBgSAJ3o. 19. Thomas Joscelyn, “Leaked audio features Al Nusrah Front emir discussing creation of an Islamic emirate,” FDD’s Long War Journal, July 12, 2014, http://www.longwarjournal. org/archives/2014/07/leaked_audio_ of_al.php. 20. Aymenn Jawad Al-Tamimi, “The Massacre of Druze Villagers in Qalb Lawza, Idlib Province,” Syria Comment, June 15, 2015, http://www.joshualandis.com/ blog/the-massacre-of-druze-villagersin-qalb-lawza-idlib-province/. 21. Marie Kostrz, “Idleb sous la férule de Jabhat Al-Nusra” [Idlib under the rule of Jabhat al-Nusra], Orient XXI, February 15, 2016, http://orientxxi.info/ magazine/idleb-sous-la-ferule-de-jabhat-al-Nusra,1202. 22. “Control of Terrain in Syria: September 14, 2015,” Institute for the Study of War, http://www.understandingwar. org/map/control-terrain-syria-september-14-2015. 23. Richard Spencer, “Al-Qaeda’s Syrian branch seize citizen journalists,” Telegraph, January 10, 2016, http://www. telegraph.co.uk/news/worldnews/ middleeast/syria/12091812/Al-Qaedas-Syrian-branch-seize-citizen-journalists.html. See also the February 2016 interview with Fares by “100eyes,” available at https://www.youtube.com/ watch?v=l0Co90T7uzs. 24. For example, see the Facebook group “Lovers of Rebels Kafr Nabl,” https://www.facebook.com/groups/ 309240689152345. 25. Thomas Joscelyn, “Analysis: Al Nusrah Front rebrands itself as Jabhat Fath Al Sham,” FDD’s Long War Journal, July 28, 2016, http://www.defenddemocracy.org/media-hit/thomas-joscelyn-analysis-al-nusrah-front-rebrands-itself-asjabhat-fath-al-sham/. 92  26. Charles Lister, “The Dawn of Mass Jihad: Success in Syria Fuels al-Qa’ida’s Evolution,” CTC Sentinel, September 7, 2016, https://ctc.usma.edu/posts/ the-dawn-of-mass-jihad-success-in-syria-fuels-al-qaidas-evolution. 27. “Conférence d’Astana sur la Syrie: pas de percée mais un ‘optimisme prudent’” [Astana conference on Syria: no major progress, but cautious optimism], France 24, January 23, 2017, http://www.france24.com/fr/20170123conference-astana-texte-commun-optimisme-rebelles-syrie-syriens-bacharassad-russie. 28. Thomas Joscelyn, “Jihadists in Syria honor Mullah Omar, praise Taliban’s radical state,” FDD’s Long War Journal, August 4, 2015, http://www.longwarjournal.org/archives/2015/08/syrian-jihadists-honor-mullah-omar-praise-talibans-radical-state.php. 29. “Whole of Syria Protection Needs Overview (PNO), 2018: Idleb Governorate,” UNHCR, October 12, 2017, https://reliefweb.int/report/syrian-arab-republic/ whole-syria-protection-needs-overview-pno-2018-idleb-governorate. 30. As discovered during the author’s April 2015 visit to Nagorno-Karabakh. 31. “Syria, militants set to evacuate 4 towns as part of deal: Sources,” Press TV, April 12, 2017, http://www.presstv. com/Detail/2017/04/12/517709/SyriaFoua-Kefraya-Zabadani-Madaya. 32. Cafarella and Casagrande, “Syrian Armed Opposition Powerbrokers,” http:// www.understandingwar.org/report/ syrian-armed-opposition-powerbrokers-0. 33. Fabrice Balanche, “Will Astana Displace Geneva in the Syrian Peace Process?” PolicyWatch 2754 (Washington Institute for Near East Policy, January 20, 2017), http://www.washingtoninstitute.org/policy-analysis/view/will-as- tana-displace-geneva-in-the-syrianpeace-process. 34. Fabrice Balanche, “Preventing a Jihadist Factory in Idlib,” PolicyWatch 2856 (Washington Institute for Near East Policy, August 31, 2017), http://www. washingtoninstitute.org/policy-analysis/view/preventing-a-jihadist-factoryin-idlib. 35. Daveed Gartenstein-Ross, “Tens of Thousands of Syrian Rebels Poised to Regroup,” The Cipher Brief, October 12, 2017, https://www.thecipherbrief. com/article/middle-east/tens-thousands-syrian-rebels-poised-regroup. 36. Gilles Kepel, Terreur et Martyr, Relever le Defi de la Civilization [Beyond terror and martyrdom: the challenge of civilization] (Paris: Flammarion, 2009). 37. Michael Gordon and Anne Barnard, “U.S. Places Militant Syrian Rebel Group on List of Terrorist Organizations,” New York Times, December 10, 2012, http://www.nytimes.com/20 12/12/11/world/middleeast/us-designates-syrian-al-nusra-front-as-terrorist-group.html. 38. “Pression militaire et succès diplomatique pour les rebelles Syriens” [Military pressure and diplomatic success for Syrian rebels], Le Monde, December 13, 2012, http://www.lemonde.fr/procheorient/article/2012/12/13/syrie-pression-militaire-et-succes-diplomatiquepour-les-rebelles_1805889_3218.html. 39. Georges Malbrunot, “Le nouveau visage de la rebellion Syrienne” [The new face of the Syrian rebellion], Le Figaro, January 13, 2014, http://www.lefigaro. fr/international/2014/01/13/0100320140113ARTFIG00585-le-nouveauvisage-de-la-rebellion-syrienne.php. 40. Fabrice Balanche, Atlas of the Near East: State Formation and the Arab-Israeli Conflict, 1918–2010 (Leiden: Brill, 2017). 41. Myriam Ababsa, Raqqa, Territoires et Pratiques Sociales d’une Ville Syrienne [Raqqa, territories and social practices of a Syrian city] (Beirut: Institut Français du Proche-Orient, 2009), http://books. openedition.org/ifpo/1260. 42. Author interviews with Syrian refugees in Turkey, June 2014. 43. Author interview with Deir al-Zour refugees, Beirut, March 2016. 44. Fabrice Balanche, ”Le programme de modernisation de l’irrigation en Syrie” [The modernization of irrigation systems in the Syrian northeast], Méditerranée 119 (2012), https://mediterranee. revues.org/6499. 45. “Caliphate in Decline: An Estimate of Islamic State’s Financial Fortunes,” International Centre for the Study of Radicalisation and Political Violence, February 17, 2017, http://icsr.info/ wp-content/uploads/2017/02/ICSRReport-Caliphate-in-Decline-An-Estimate-of-Islamic-States-Financial-Fortunes.pdf. 46. “Isis Inc: How oil fuels the jihadi terrorists,” Financial Times, October 14, 2015, https://www.ft.com/content/b823 4932-719b-11e5-ad6d-f4ed76f0900a. 47. “Syrie: l’economie et le secteur petrolier reduits a neant” [Syria: the economy and the oil sector are destroyed], Agence France-Presse, May 30, 2017, http://www. lefigaro.fr/flash-eco/2017/05/30/9700220170530FILWWW00134-syrie-l-economie-et-le-secteur-petrolier-reduits-aneant.php. 48. “Financing of the Terrorist Organisation Islamic State in Iraq and the Levant (ISIL),” Financial Action Task Force, February 2015, http://www.fatf-gafi. org/documents/documents/financingof-terrorist-organisation-isil.html. 49. Nikolaos van Dam, Destroying a Nation: The Civil War in Syria (London: I.  FABrICE BAL ANChE B. Tauris, 2017), pp. 80–81. 50. “Islamic State frees 19 of 220 Abducted Assyrian Christians,” BBC News, March 1, 2015, http://www.bbc.com/news/ world-middle-east-31685931. 51. David Petraeus, “Learning Counterinsurgency: Observations from Soldiering in Iraq,” Military Review, January–February 2006, http://www.au.af.mil/au/ awc/awcgate/milreview/petraeus1.pdf. 52. Haian Dukhan and Sinan Hawat, “The Islamic State and the Arab Tribes in Eastern Syria,” E-International Relations, December 31, 2014, http:// www.e-ir.info/2014/12/31/the-islamicstate-and-the-arab-tribes-in-easternsyria. 53. Skype interview with a journalist from northern Syria, October 2, 2017. 54. Shelly Kittleson, “Syrian rebel group just trying to get home,” Al-Monitor, April 21, 2017, http://www.al-monitor. com/pulse/originals/2017/04/syriasouth-armed-opposition-group-advance-isis.html. 55. Sam Heller, “A Deadly Delusion: Were Syria’s Rebels Ever Going to Defeat the Jihadists?” War on the Rocks, August 10, 2017, https://warontherocks. com/2017/08/a-deadly-delusion-weresyrias-rebels-ever-going-to-defeat-thejihadists. 56. Liz Sly, “Islamic State Routs Pentagon-Backed Syrian Rebels in Fresh Setback for U.S. Strategy,” Washington Post, June 29, 2016, https://www. washingtonpost.com/world/middle_ east/islamic-state-routs-pentagonbacked-syrian-rebels-in-fresh-setbackfor-us-strategy/2016/06/29/41833a163e0b-11e6-9e16-4cf01a41decb_story. html. 57. Ryan Browne, “Russia and Syrian Regime Seeking to Poach U.S.-Backed Fighters,” CNN, August 3, 2017, http:// edition.cnn.com/2017/08/03/politics/ russia-assad-poach-us-backed-fightersat-tanf/index.html. S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r   93 5 . F Ou r P O S T wAr S C EN ArIOS T and sectarian identity are not permanent historical constructions. In times of peace and prosperity, a process of sectarian dissolution usually takes place, but in wartime, the security provided by sectarian communities takes precedence over other relationships. Decades ago, the Baath coup in Syria seemed to point toward a future marked by socially and territorially integrated nationalism. Instead, however, the new rulers established a system of power that favored the revival of sectarianism. Baath leaders insisted that ethnic and religious diferences were private, but this did not stop them from exploiting sectarianism to dominate society. Before Haiz al-Assad became president, he often condemned his Baath comrades as “sectarian,” yet he himself illed the army chain of command with members of his Alawite clan in order to seize power. he 2011 uprising was a long-brewing reaction to the failure of these disErrITOry torted national construction methods inherited from the 1960s (in-depth analysis of which is laid out in Part II of this study). Yet by putting forward a starkly Sunni Arab identity for postwar Syria, the opposition has taken a path similar to the regime’s, generating a wave of violent exclusions that could result in the country’s partition. Allowing the Sunni Arab majority to govern may seem just in abstract terms, but the fact is that the Kurds and other minorities cannot accept such dominion today because most of the Sunni Arab rebellion has been radicalized against them. Despite the glaring realities on the ground in Syria, a great hypocrisy continues to reign whenever Western observers discuss—or, rather, choose not to discuss—the role of sectarianism in the middle East. For example, Lebanon has perhaps the most sectarian political system in the world and sufered ifteen years of civil war because of it, not to mention many years of foreign occupation and domestic strife since then. he appointment of a senior oicial is a mat- ter that mobilizes all sects because unoicial quotas must be respected. Yet when a Westerner speaks with Lebanese citizens, they are more likely than not to dismiss sectarianism outright or call it a Western fabrication aimed at weakening Arabs. his contradiction is diicult to understand at irst, but researchers can grasp the problem and examine it objectively once they realize two things: irst, that political ideology is usually just a smokescreen in the middle East, and second, that sectarianism remains a major driver of regional politics and social organization. hus, the dogma of the “Syrian Arab nation” has not withstood the current war, and the clues that reveal this Baathist deceit have been clear for decades to those who really know Syrian society. EXPLOSION Or rEuNIFICATION? As stated at the outset of this study, the fate of Syria is no longer Syria’s to decide. Its reuniication or partition will largely depend on the regional and/or  94  international consensus that emerges after the war and the decisions of the foreign powers that have so extensively intervened there. he international community is always cautious about redrawing borders because dividing a country is a bad example for others and can exacerbate the conlicts it seeks to remedy (or create new ones). Regional powers are even more opposed to touching Syria’s borders because doing so would beneit the Kurds irst. Turkey and Iran are iercely opposed to oicial Kurdish autonomy. Admittedly, Ankara came to accept the idea of a semiindependent Kurdistan in Iraq. But the September 2017 statehood referendum may have soured that tolerance, which was mostly based on masoud Barzani’s willingness to partner with Turkey on numerous ventures while rejecting its bitter enemy, the Kurdistan Workers Party. his is certainly not the case with Rojava and the Democratic Union Party in Syria, since Ankara regards the PYD as an arm of the PKK. For its part, Iran believes that an inde- pendent Syrian Kurdistan could serve as an unwelcome precedent for Iranian Kurds.1 Washington has good relations with the dominant Kurdish factions in Iraq and Syria, and past administrations have already applied the ethnic-partition solution elsewhere (e.g., Yugoslavia in 1994). Yet promoting a Kurdish state in Syria headed by the PYD/PKK does not seem like a viable U.S. option for a host of reasons, including the PKK’s status as a U.S.-designated terrorist group, the Trump administration’s opposition to the Iraqi Kurdish statehood vote, and the regional turmoil ushered in by that referendum. he Russians also support partition when it serves their geopolitical interests. In neighboring Georgia, moscow has backed the autonomous republics of Abkhazia, Adjara, and South Ossetia in order to weaken leaders whom it regards as too close to the West and NATO. he same fate probably awaits Ukraine. In Syria, however, moscow’s objective seems to be preserving the country’s territorial integrity as much as possible rather than partitioning it, since Damascus is a strong regional ally. Yet regardless of international and regional views on the subject, partition already exists on the ground in Syria, so the question has become whether any of the conlict’s actors is capable of reunifying the country. Despite their cooperation with the regime on various matters, Kurds want to keep their autonomy from Damascus. For their part, Sunni Arab rebels do not want the regime to resume control in their territories as long as Assad is in power—they want to keep the country intact, but only on the condition that they become the new powerbrokers. he Kurds have championed an alternative scenario by calling for a federal Syria, but federalism in a non-homogeneous country seems farfetched and has little support among other sects. If the international community wishes to restore Syria’s territorial integrity, it will need to foster greater decentralization. Transferring power to the municipalities based on sectarian distribution could ease tensions somewhat, and this solution might be more acceptable to the regime and its external supporters than federalism. Tehran formally opposes federalism in Syria because it could spur similar appeals in Iran, where Persians make up a dwindling 60% majority of the population2 and have a low fertility rate3 compared to ethnic minorities. Likewise, Turkey has no desire to fuel federalist demands among its sizable Kurdish population. Yet all parties may need to resign themselves to at least some degree of unoficial federalism in Syria, since the army lacks the means to reconquer the entire country even with Iran and Russia’s support. Some territories should therefore be given informal autonomy so that the S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r  ighting can cease, with one caveat— such arrangements would only apply to peripheral regions and the northeast because the regime will not accept anything less than full central authority in the populous western zone between Damascus and Aleppo. It is diicult to predict Syria’s future over the next ten years—the question of whether it will explode along sectarian lines or reunify itself depends strongly on global and regional geopolitical parameters. But the outcome also depends on the central government’s capacity to build a new social and territorial pact with the population. he war has widened and multiplied the country’s cleavages, and years of heavy ighting and mass population displacement have deeply inscribed these fractures onto Syrian territory. Even if the country retains its territorial integrity and current administrative divisions, the wartime situation of decentralized governance over sharply divided sectarian regions is likely here to stay. FOur SCENArIOS he war has put a spotlight on Syria’s two main cleavages: sectarian and economic. he irst type—which includes religious, ethnic, and tribal issures—is the most prominent, since Syrians are killing each other mainly on the basis of these identities. he second type corresponds to class diferences and urban/ rural issures, which partially overlap: in the cities, residents who have rural origins tend to be poor and in conlict with the original urban dwellers, who generally belong to the middle and upper classes. Yet these economic cleavages are less lethal than the sectarian cleavages. It is not social or class conlict that has motivated so many Syrians to join militias, though poverty and the hope of social advancement are strong drivers. Similarly, political conlicts between Arab nationalists, Baathists, communists, liberals, and so on are marginal. Even in the politically stratiied Kurdish community, opposition to the PYD is not powerful enough to trigger internecine war. In projecting how sectarian cleavages might afect Syria’s mid-term future, it is useful to divide the country into four territorial categories: minority zones (i.e., areas heavily populated by Alawites, Christians, Ismailis, Druze, Shia, or Kurds), the Sunni Arab tribal zones in the east, the Sunni Arab rural zones in the west (a category that includes the small Turkmen minority for convenience’s sake), and the large Sunni Arab majority cities (Damascus, Aleppo, Homs, and Hama). Of course one could create other categories, especially within the Kurdish community; Syria is a mosaic, and each of its communities is a mosaic within a mosaic. Yet the nature of this study calls for simplifying (some might say caricaturing) the country’s  95 sectarian categories in order to develop comprehensible scenarios for its future. Before laying out these scenarios, it is important to clarify their scope. his study does not delve very far into the regional and international geopolitics that will be essential to Syria’s fate. many other books and articles exist on those subjects, and there is little need to synthesize them here. moreover, geopolitical conditions may still evolve before the war’s conclusion, which is the catalyst for the scenarios presented below. his study does not venture to give a deinite timeframe for peace in Syria, since it is diicult to estimate how long the winning side will need to wear the others down, and how long international actors will take to come to an agreement. Given these parameters and the sectarian trends examined in the previous chapters, the war is most likely to result in one of four outcomes: 1. he regime wins and Syria remains territorially united. 2. he rebels topple the regime and Syria remains territorially united. 3. he regime stays in power but Syria is partitioned. 4. he rebels topple the regime but Syria is partitioned. he scenarios in which the regime holds out do not assume that Bashar al-Assad himself remains in power. And the rebel victory scenarios assume that only a co96  alition dominated by al-Qaeda ailiate Jabhat Fatah al-Sham (under one name or another) is capable of taking Damascus. Other scenarios are theoretically possible, such as the moderate rebels experiencing a revival and rising up to challenge the radical Islamists. Yet this seems unlikely given the current realities on the ground—if the regime is going to be defeated, the radicals will be the ones to do it. Similarly, there is little point in elaborating a scenario in which foreign powers intervene to help secular-democratic forces overthrow the regime, guarantee the country’s territorial unity, and safeguard minority rights. his outcome would of course be ideal, but one must remain realistic. he most useful aspect of this exercise is to deine how Syria’s internal or external borders would look in each scenario, depending on the camp that winds up occupying Damascus and essentially winning the conlict. A simpliied visual rendering of the four sectarian territorial categories described above is perhaps the best way of illustrating each scenario; the irst map shows how these four zones are currently arrayed. Each scenario also outlines the type of governance that would prevail (centralism or federalism; direct or indirect government), the demographic consequences (ethnic cleansing, refugee lows), the prospects for stability, and the main social, political, economic, and security problems that might arise. A separate section is devoted to briely assessing the geopolitical consequences of these outcomes, particularly regarding the stability of neighboring countries, the solidity of the Iranian axis, and Russia’s future role in the region. scenario 1: syria remains united, the regime holds damascus In this scenario, the regime does not have adequate military resources to reconquer all of Syria, so it negotiates agreements with local powerbrokers outside its zone of control. he north remains under the PYD’s dominion, and the Kurds agree to a strategic and economic alliance with Damascus in order to counter the Turkish threat. In the Euphrates Valley, newly liberated from the Islamic State by the Syrian Democratic Forces (SDF), the regime governs indirectly through local Sunni Arab notables. In western Syria, the regime continues to impose direct rule, but with many new exceptions; for instance, it rewards Jabal al-Druze for its loyalty by giving it more autonomy, and extends similar privileges to former Sunni Arab rebel zones that rallied to Assad during the war. For its part, Turkey maintains an area of inluence in northern Syria, including the Turkmen areas, and another safe zone is created on the Jordanian border. he political system remains formally centralized, but the country is more visibly divided between areas of direct and indirect administration. meanwhile, in the former rebel province of Idlib, around half the population goes into exile abroad, unwilling to ight on or accept continued Baath/Alawite rule. After this last wave of emigration, ethnic cleansing is limited. Yet the majority of Syrian refugees do not return home once the ighting stops, especially those who resided in opposition strongholds such as Qusayr, Homs, or certain districts of Damascus and Aleppo. Apart from potential safe zones on the borders, Syria remains territorially united, allowing for rapid rebuilding of infrastructure and the economy. he country’s stability remains fragile, however, and only successful economic reconstruction and expanded sectarian power-sharing can preserve it in the long term. his in turn would require the regime to open a wider political space for Sunni Arabs and Kurds. As for Syria’s foreign relations, the country becomes a Russo-Iranian protectorate. scenario 2: syria remains united, the rebels take damascus After marching into Damascus, rebel forces led by Jabhat Fatah al-Sham impose a centralized political system with a Sunni Arab, Islamist character. As they seize sectarian strongholds throughout western Syria, most of the Kurds and  FABrICE BAL ANChE 49. SYRIA'S ETHNIC AND SOCIAL STRUCTURE Aleppo Turkey Qamishli Raqqa Latakia Hasaka Deir al-Zour Hama MEDITERRANEAN Tartus Palmyra Homs Iraq Lebanon Damascus Suwayda scenario 3: syria is partitioned, the regime holds damascus Deraa Israel Jordan Rural Sunni Arab City Urban Sunni Arab Primary road Minorities Secondary road Sunni Arab tribes Desert area S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r  other minorities lee the country, along with numerous secular Arabs. many of the millions of Sunni Arab refugees who led Syria during the war decide to return home, believing they will be permitted to seize the property of those who leave postwar due to large-scale ethnic and economic cleansing. he country’s reconstruction is facilitated by Gulf investors eager to strengthen a friendly regime that has managed to break the pro-Iranian axis. Russia is ousted from Syria along with Iran. And now that the country is homogeneous in ethnoreligious terms, it has a better chance of becoming stable. Yet external conlict is likely to arise with Lebanon, where a large portion of Syria’s minorities seek refuge. Partly as a result of this, Lebanon’s own civil war resumes. his scenario is similar to the irst one, but the regime is now weaker, and the internal boundaries between the zones of direct and indirect regime control instead become external borders separating discrete territories. Although the southwest remains under the regime’s rule, Aleppo and Hama fall into rebel hands, Syrian Kurdistan becomes independent, and the Euphrates Valley falls under tribal council control and is cut of from Damascus and Aleppo. A safe zone is formed at the Jordanian border, with foreign ground troops supporting various opposition factions. Ethnic cleansing leads to population exchanges between the diferent zones, and some wartime refugees return to Syria if their ethnoreligious status corresponds with that of the local authorities. he new borders complicate the reconstruction efort and threaten to spark new conlicts. In the northeast, clashes arise between Kurds and Arabs over access rights to water from the Euphrates. In the central steppes, Sunni Arab tribes compete with the regime for control of key hydrocarbon reserves. Iran and Russia remain in Syria but are weakened by the challenge of propping up the regime indeinitely amid economic impoverishment, potent competition from numerous warlords,4 and the continually looming threat of further secessions. scenario 4: syria is partitioned, the rebels take damascus Although rebel forces seize Damascus, they lack the means to oust minorities with well-deined, defensible sectarian strongholds in other parts of the country. Accordingly, Kurdistan breaks away from Syria, along with an “Alawistan” and possibly a “Druzistan,” the latter protected by Jordan. his leads to extensive ethnic cleansing, with the rebels pushing minorities out of the Damascus  97 50. UNITY WITH REGIME IN DAMASCUS SAFE ZONE Turkey SAFE ZONE 51. UNITY WITH REBELLION IN DAMASCUS Qamishli Aleppo Latakia Deir al-Zour Latakia Tartus SHAER GAS FIELD Deir al-Zour Hama OIL SHAER GAS FIELD EUPHRATES OIL Palmyra Homs OIL Hasaka Raqqa OIL Hama Qamishli Aleppo Turkey Hasaka Raqqa EUPHRATES WATER OIL EUPHRATES OIL Homs Tartus MEDITERRANEAN MEDITERRANEAN Palmyra Iraq Lebanon Damascus Suwayda CLEAVAGES Lebanon Damascus Suwayda Jordan SAFE ZONE Israel Iraq Israel Deraa NEW ORGANIZATION GEOGRAPHY Rural Sunni Arab Internal border City Urban Sunni Arab External border Primary road Minorities Conflict Secondary road Sunni Arab tribes Ethnic cleansing Desert Deraa CLEAVAGES Jordan NEW ORGANIZATION GEOGRAPHY Rural Sunni Arab Internal border City Urban Sunni Arab External border Primary road Minorities Conflict Secondary road Sunni Arab tribes Ethnic cleansing Desert Social cleansing 98   FABrICE BAL ANChE 52. PARTITION WITH REGIME IN DAMASCUS Turkey Aleppo EUPHRATES WATER Qamishli 53. PARTITION WITH REBELLION IN DAMASCUS OIL Hasaka Raqqa EUPHRATES WATER Turkey Qamishli OIL Hasaka Raqqa Aleppo SEA ACCESS SEA ACCESS Deir al-Zour ORONTES WATER Hama SHAER GAS FIELD Tartus Deir al-Zour ORONTES WATER EUPHRATES OIL Palmyra Damascus Suwayda EUPHRATES OIL Palmyra Tartus Lebanon Homs Iraq Lebanon Damascus Suwayda Deraa SAFE ZONE CLEAVAGES SHAER GAS FIELD SEA ACCESS Iraq Israel OIL Hama Homs MEDITERRANEAN Latakia OIL MEDITERRANEAN Latakia Deraa Jordan Jordan NEW ORGANIZATION Israel GEOGRAPHY Rural Sunni Arab Internal border City Urban Sunni Arab External border Primary road Minorities Conflict Secondary road Sunni Arab tribes Ethnic cleansing Desert CLEAVAGES NEW ORGANIZATION GEOGRAPHY Rural Sunni Arab Internal border City Urban Sunni Arab External border Primary road Minorities Conflict Secondary road Sunni Arab tribes Ethnic cleansing Desert Social cleansing Social cleansing S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r   99 area and the breakaway statelets likely expelling Sunni Arabs. Some wartime refugees return to Syria, while many new emigres likely go into exile. he political system in Sunni Arab Syria is formally centralized, with direct governance in the west and indirect governance for the Euphrates Valley, as local tribes demand to keep control of their hydrocarbons. Competition likely arises between Aleppo and Damascus as well, echoing their longstanding rivalry during the Baath era. As in scenario 3, the new borders complicate reconstruction. Although the new Sunni Arab regime may still beneit from Gulf inancial assistance, it is now cut of from access to the sea due to the breakaway Alawite statelet on the coast. And depending on its relations with the Kurds and the Arab tribes, it may also be deprived of wheat from Jazira and hydrocarbons from the east. hese uncertainties could lead to new conlicts over borders and resources. Finally, Russian and Iranian inluence in Syria are weakened. Alawistan becomes a Russian-Iranian protectorate, while moscow potentially shields Kurdistan from Turkish threats. ThE wEST’S SECTArIAN BLINd SPOT? In the nineteenth century, France and Great Britain took a clearly sectarian approach to the middle East. Fast100  forward to the Syria crisis, however, and Western countries no longer seem to attach great importance to sectarianism, preferring to draw political parallels between the war and the European Revolutions of 1848 or the fall of communism in the Eastern Bloc. Such comparisons tend to leave out the fact that both of these historical antecedents had a communitarian dimension; when communism fell, for example, non-homogeneous countries such as Czechoslovakia, Yugoslavia, and the Soviet Union were divided along ethnoreligious lines. It is therefore signiicant that Russia was the irst country to highlight the sectarian aspects of the Syria war at a time when most Western European analysts were ignoring or denying them. As early as 2011, Russian media published ethnoreligious maps of the conlict.5 his may help explain the relative success of Russian policy in Syria—moscow immediately understood the conlict’s most powerful dynamic. Since early in the revolt, many Syrian religious minorities believed they had no choice but to support the Assad regime in order to prevent an increasingly radicalized Sunni Arab opposition from prevailing. he regime and Iran organized these minorities into self-defense groups within the framework of the “People’s Army,” later renamed the National Defense Forces. Even the Kurds, who did not make common cause with the regime, quickly moved away from an opposition that was divided between Arab nationalists and Islamists, all of whom rejected Kurdish autonomy. As for the country’s majority Sunni Arabs, they are hardly a monolith, but that is due in no small part to the regime’s decades-long efort to divide and disempower them. For half a century, Damascus did everything it could to create strife between the “noble” and “common” tribes in the Euphrates Valley, between the Sunni bourgeoisie and popular classes, between devout and secular Sunnis, between Suis and Salafists, and so on. Likewise, tribal rivalries have helped divide the rebels throughout the war, making it impossible to unite the Free Syrian Army. Some Sunni Arabs even joined regime militias, including civil servants who feared the rebellion and city dwellers who sought to protect their neighborhoods from looters. Despite their greater numbers, then, Syrian Sunnis remain far too fractured to represent a credible political alternative to the regime. he refusal to take this past and present sectarianism into consideration is one of the main reasons behind the failure of Western policy in Syria and, consequently, the continuation of the Assad regime. At irst, Western governments may have had good reason for this approach, since adopting an openly sectarian policy toward the conlict could have raised doubts about Syria’s territorial unity—an apparent redline for the United States and other countries. Yet after a civil war of this magnitude, Western observers should be clear-eyed enough to realize that no general reconciliation is possible. millions of Syrian refugees will be unable to return home because of their support for the opposition, and if the regime reconquers areas still under rebel control, a new forced exodus will occur. Even if the regime’s central leadership expresses a sincere desire for amnesty (hardly likely), there is no guarantee this leniency would be respected by the sectarian militias and other actors who wield varying degrees of independent authority at the local level. he international players competing in Syria will therefore be compelled to rely on factions for whom sectarian concerns are a primary motivation. he Kurds want their own autonomous area in the north. he Alawites and other religious minorities want to retain their privileged place in the halls of power and ensure they can remain in their homeland. he various Sunni Arab tribes are more focused on material interests than politics—they seem to realize that taking power in Damascus is farfetched at this point, but they still want more power at the local level, more land, more jobs, better access to water, hydrocarbon sharing, and so on. If certain domestic or foreign actors aim  FABrICE BAL ANChE 54. POTENTIAL IRANIAN ROUTES TO THE MEDITERRANEAN, August 2017 N W E Tigris River Euphrates River TURKEY PKK Qamishli Kobane S Manbij Afrin Hasaka Aleppo Idlib Latakia Tal Afar Sinjar Mosul Raqqa Erbil KDP al-Tabqa SYRIA Hama CYPRUS Tartus Homs Kirkuk Deir al-Zour Sulaymaniyah IRAN PUK Sukhna Palmyra Abu Kamal LEBANON Damascus al-Tanf Ramadi Baghdad Golan Heights Mediterranean Sea Karbala IRAQ Najaf ISRAEL Amara JORDAN Basra SAUDI ARABIA EGYPT Persian Gulf KUWAIT IRANIAN ROUTES AREAS OF CONTROL IN SYRIA 200 km AREAS OF CONTROL IN IRAQ Potential Syrian army and militias Rebels Iraqi army and militias Islamic State Existing SDF (Kurdish-led) Islamic State KRG (KDP and PUK) PKK S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r  0 PKK-KDP Major city Locality  101 55. RUSSIA’S STRATEGY IN SYRIA N W Tigris River Euphrates River TURKEY SUPPORT TO ASSAD Protected city or area E Qamishli Russian military base S Afrin Kobane Manbij HASAKA Control of desert Rebel supply route closure Mediterranean ALEPPO IDLIB RAQQA CONDOMINIUM WITH IRAN Hmeimim Air Base LATAKIA Boundary Iranian/Russian zones Alawite region Strong military cooperation w/ Iran HAMA New military base KURDISH STRATEGY DEIR AL-ZOUR Recognition of Rojava TARTUS Palmyra Abu Kamal Afrin-Kobane junction HOMS LEBANON Deconfliction with Assad IRAQ DAMASCUS COUNTER-REBEL STRATEGY Reconquest of Euphrates region QUNEITRA Sunni rebel safe zone Golan Heights DERAA to facilitate these demands, they should understand that the tribes are not for sale, merely for rent. moreover, in each of these cases, the empowered actors are armed forces rather than political groups, and the cohesion of these combatants is essentially based on sectarian solidarity. his fact may be engendering a paradigm shift in how foreign governments approach Syria, as realpolitik 102  GEOGRAPHY SUWAYDA ISRAEL JORDAN compels them to consider sectarian issues when courting or confronting local armed groups. On the regime side, Russia has been pushing Assad to reorganize the national army by integrating the National Defense Forces militias. he regime has long presented the army as a model of national unity, but in reality the elite units are clearly dominated by Alawites. 0 100 km On August 2017, the “Desert Hawks” militia—a force founded by Ayman Jaber, a businessman close to Assad—was disbanded and its ighters were integrated into the Syrian marines and the army’s 5th Corps.6 moscow has encouraged such moves,7 based on the belief that integrating pro-regime militias will make the regular army a national rallying point. he Kremlin also hopes Capital Province border Province center International border Other locality to use its inluence over the army as leverage with the regime at a time when Iranian-backed militias are still ubiquitous in western Syria. he United States seems to have adopted a similar strategy in the east, attempting to transform the SDF into a melting pot of Kurdish and Arab ighters. Washington hopes that this force will prove capable of taking the entire  FABrICE BAL ANChE 56. RUSSIA’S STRATEGY IN THE LEVANT: ENERGY AND TURKISH SIEGE SOU TH S TREA M Sebastopol Black Sea KAZAKHSTAN RUSSIA South Ossetia Abkhazia GEORGIA Tbilisi Istanbul Batouni ARMENIA Ankara Yerevan Gyumri NABUCCO TURKEY C BT TURKMENISTAN Baku AZERBAIJAN Caspian Sea PKK Ceyhan Aleppo CYPRUS Mediterranean Sea Hmeimim Tartus ROJAVA LEBANON SYRIA IRAN Baghdad IRAQ ISRAEL JORDAN N Tehran Palmyra Damascus W KRG Mosul Basra EGYPT KUWAIT E r Pe Gu n 400 km sia SAUDI ARABIA S 0 Red Sea POLICY RUSSIAN SIEGE AROUND TURKEY lf ENERGY STRATEGY Strategic alliance w/ Russia Strategic alliance w/ Russia Block transport from GCC Tactical alliance w/ Russia Russian base Hydrocarbon road Pro-USA Support to Kurds Potential hydrocarbon road Potential Shia uprising S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r  Euphrates Valley back from the Islamic State, stabilizing the area, preventing the Assad regime from returning there, and impeding Iran’s attempts to create a land corridor to the mediterranean— all while maintaining a pro-American posture. For this plan to work, however, U.S. oicials would need to make sure that the Kurdish forces in charge of the SDF do not behave as an occupying power, instead allowing Arab tribes to play the main role in liberating the valley. his may be a tall order given the numerous reports of inexperienced Arab units playing second iddle to Kurdish forces in the battle for Raqqa.8 In the end, the United States may need to establish a semi-permanent presence on the ground in eastern Syria in order to bolster its local allies—a thorny prospect for many reasons, including Turkey’s strong feelings on the Kurdish issue and Russia’s apparent unwillingness to treat the Euphrates River as a line of deconliction.9 ChOOSINg BETwEEN FLAwEd OPTIONS All of the available scenarios for Syria’s near-term future present serious dificulties and dangers, and the viability of any given approach depends on the interests of the foreign power attempting to carry it out. If the main goal is to break the pro-Iranian axis and block Russian inluence in the  103 middle East, a rebel victory must be prioritized. Yet this runs the risk of partitioning Syria, making the country a base of radical Islam, and creating millions more refugees. Can one realistically expect a group like Jabhat Fatah al-Sham to evolve toward tolerance and democracy? Alternatively, if the goal is to stabilize the middle East and prevent millions of Syrian, Lebanese, and even Jordanian refugees from overwhelming the European Union, then the existing regime in Damascus must be maintained, with or without Assad at its head. Although this could produce a stalemate of sorts and allow some refugees to return home, it would probably require partitioning the country, and the geopolitical and moral consequences of backing a regime with such a brutal track record could be considerable. he prospects seem equally fraught when considering what these scenarios might mean for the future governance of Syria. Partitioning the country could solve some of its sectarian problems, but not all of them, and this approach would surely create new divisions between and within sects. he fragmentation would also present each sectarian zone with problems (and, likely, conlicts) related to sea access, narrowing economic markets, sharing of river waters, and allocation of oil resources. Even worse, ethnic cleansing would likely be maximal in this scenario. Al104  57. SECTARIAN DISTRIBUTION IN LEVANT AND POTENTIAL STATELETS Tigris River Euphrates River TURKEY N W Syrian Kurdistan Turkmenistan Sunni Idlib Idlib Iraqi Kurdistan S Erbil Mosul Raqqa Aleppo E Sulaymaniyah Sunni Arab Tribes Latakia Deir al-Zour Hama Tripoli Kirkuk Sunni Arab Palmyra Homs IRAN SYRIA Beirut Damascus LEBANON Baghdad Deraa ISRAEL Karbala Hillah IRAQ Sunni Deraa Amara Najaf Diwaniyah Amman Basra JORDAN EGYPT SAUDI ARABIA Aqaba 0 KUWAIT 200 km SUNNI MAJORITY ETHNIC MINORITY RELIGIOUS MINORITY Sedentary Kurd Alawite Ismaili Official borders Nomadic Turkmen Druze Christian De facto future internal borders Yezidi Shia Twelver Main city  FABrICE BAL ANChE Regional power Local power Client country TURKEY Kurdistan Regional confrontation Civil War Hezbollah Islamic State U.S. military base SYRIA Russian military base IRAN Arab-Israeli conflict Hydrocarbon resources IRAQ Suez Canal Military conflict JORDAN Baluchistan Sinai Alarm Shia Revolt Strategic position Destabilization EGYPT Hormuz SAUDI ARABIA Russian influence UAE Royal Succession OMAN al-Qaeda 58. THE NEW THIRTY YEARS WAR Houthi Rebellion YEMEN Piracy Bab al-Mandab S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r  0 ternative scenarios do not project much better. Keeping Syria intact while establishing some form of federalism would still lead to partition sooner or later given the country’s sectarian heterogeneity. Decentralization at the provincial level would not be much different, since the regime’s administrative boundaries generally align with Syria’s sectarian boundaries. True protection of minorities and progressive evolution toward democracy are only possible within the framework of a system that is decentralized to the lowest level, the municipality. Yet one should not make the mistake of idealizing municipal democracy as a goal given its keen vulnerability to clientelism and rampant corruption. Ideally an intermediate scenario between these extremes will materialize, but the opportunities for such solutions are dwindling. To seize the least unpalatable options before they disappear, the international community must act quickly—and with eyes fully open to the sectarian dynamics that will drive the rest of the war and its aftermath. 400 km  105 the (de)legitimation of the Western intervention in Syria], Orient XXI, May 11, 2015, http://orientxxi.info/magazine/ la-carte-outil-de-de-legitimation-de-lintervention-occidentale-en-syrie,0899. Notes 1. 2. 3. 4. 5. Farzin Nadimi, “Iran’s Military Options Against Kurdish Independence,” PolicyWatch 2861 (Washington Institute for Near East Policy, September 22, 2017), http://www.washingtoninstitute.org/ policy-analysis/view/irans-military-options-against-kurdish-independence. 6. Leith Fadel, “Pro-government ‘Desert Hawks’ disbanded after three years of service,” al-Masdar, August 2, 2017, https://www.almasdarnews.com/ article/pro-government-desert-hawksdisbanded-three-years-service. 7. Youssef Courbarge and Emmanuel Todd, Le rendez-vous des civilisations [The convergence of civilizations] (Paris: Editions du Seuil, 2007). Author interview with a Syrian official, June 2017. 8. Fritz Schaap, “Assad’s Control Erodes as Warlords Gain Upper Hand,” Spiegel Online, March 8, 2017, http://www. spiegel.de/international/world/assadpower-slips-in-syria-as-warlords-growmore-powerful-a-1137475-2.html. See, e.g., John Davison, “Arab Fighters Struggle to Assert Role in Raqqa Assault,” Reuters, August 16, 2017, https:// www.reuters.com/article/us-mideastcrisis-syria-raqqa-idUSKCN1AW1L4. 9. Andrew J. Tabler, “Russia Crosses the Euphrates: Implications,” Policy Alert (Washington Institute for Near East Policy, September 18, 2017), http://www. washingtoninstitute.org/policy-analysis/view/russia-crosses-the-euphratesimplications. According to the CIA World Factbook, https://www.cia.gov/library/publications/the-world-factbook/geos/ir.html. Lucile Housseau, “La carte, outil de (dé)légitimation de l’intervention occidentale en Syrie” [The map, tool for 106   FABrICE BAL ANChE part ii ThE OrIgINS OF SyrIA’S SECTArIANISM 6. A dI V I S I V E E VOLu T ION I N PrE-CIVIL-wAr SyrIA, community structures were ubiquitous—members of any given community were bound together by endogamous marriage, preferential hiring, solidarity against outsiders in all matters, allegiance to local leaders, and so forth. Each individual’s identity therefore moved in two dimensions: through state territory and community territory. he irst dimension was geographically limited by the state’s borders. he second was geographically noncontiguous and often consisted of a network of smaller territories. Depending on the community to which they belonged, some individuals also had a transnational identity. herefore, political actors were inluenced by community structures as well as state structures. As individuals, they depended on their community for personal support; they also had to respect its structures for the sake of political expediency. he communitarian organization of Syrian society has long played a role in establishing its interior borders, so analyzing the distribution of these communities throughout the state and within the regime that oversees them is crucial. he administrative divisions, housing distribution, road networks, and other tangible state structures all relect the power relations between different territorial actors. Consequently, they give us information on the way these actors operate, and more generally on the state’s system of territorial demarcation. Among other things, this information conirms that sectarianism is omnipresent in Syria, and that the economic development of diferent community territories has been closely linked to both cross-sectarian relations and each sect’s relative weight on the scales of power. he current civil war is part of a violent rebalancing of these scales, one that began long before the 2011 uprising, so examining the various state policies and socioeconomic developments that produced this seismic shift is instructive. EFFECTS OF ArABIZATION Northern Syria’s Turkmen and Kurdish communities are both Sunni, but they tend to assert their ethnic rather than religious identity. his includes using their native tongues at home, a practice reinforced in recent years by Turkishand Kurdish-language satellite television programming. In contrast, some communities were so thoroughly Arabized by successive regimes that their ancestral memories are now limited to placenames and family names. For example, Jabal alAkrad is geographically part of the Alawite mountains and lies in the heart of the Assad regime’s coastal stronghold, but it was populated by Kurdish tribes brought there by Sultan Baibars when he took the region back from the Crusaders in the thirteenth century. he population has not spoken Kurdish for centuries and does not sympathize with the Kurdish nationalist movement. Kurds from the Rukn al-Din neighborhood of Damascus were Arabized as  109  well; newer generations have cut ties with their ancestral villages in Jazira and Kurgh Dagh, while local youths joined anti-Assad rebel groups after 2011 instead of the Kurdish nationalist groups that formed in the north. Yet ethnic identity is not necessarily the strongest community cement, and it often dissolves more easily than religious identity. For example, many Kurds who moved from Rukn al-Din to informal Damascus neighborhoods such as mashrou Doumar still use the Kurdish language and defend their Kurdish heritage. more broadly, until the war broke out, ethnicity did not represent a social code that sharply distinguished Kurds from other Syrians. Yet as a response to discrimination suffered under Arab nationalism, Kurdish identity is now quite strong. he Syrian version of Arab nationalism sought to eliminate Kurdish and Turkmen identities for ideological and geopolitical reasons. Because both groups are situated along large stretches of the border with Turkey, they theoretically posed a threat to Syria’s territorial integrity, unlike more dispersed, less numerous non-Arab minorities such as Armenians and Circassians. he Arabization policy launched in the 1950s was subsequently intensiied by the Baath Party. Beginning in 1963, the regime created a northern border belt of Arab villages west of Qamishli to break up Kurdish territorial contiguity. Authorities also prohibited the use of the Turkish and Kurdish languages, stripped numerous Kurds of their Syrian citizenship (the “bedoon” described in previous chapters), and economically abandoned many Kurdish and Turkmen areas, spurring locals to move southward to cities where the regime believed Arabization would occur faster. Decades of such treatment eventually provoked a Kurdish revolt in 2004, but that only resulted in a brutal crackdown followed by empty promises on development and other grievances. It was not until the 2011 uprising that the regime inally yielded somewhat, restoring citizenship to the bedoon and pragmatically accepting certain aspects of Kurdish nationalism (though as noted in chapter 3, many bedoon have yet to receive their citizenship papers). Unlike the Kurds, who implicitly or directly challenged the regime’s attempts to enforce Syrian national unity, the much smaller Armenian community acquiesced and, in the process, avoid110  ed Arabization. he majority of them came from Turkey after the 1915 genocide and settled in Aleppo, later spreading to Damascus, Homs, Latakia, and a lone rural enclave around Kasab. Once in Syria, Armenians continued to use their own language and practice strong endogamy. During the nationalization and Arabization of the educational system in the 1970s, Haiz al-Assad gave them permission to keep using their language in their own schools, which would become semi-private. Such privileges were the regime’s way of ensuring Armenian support; indeed, this community has been one of the most loyal to Bashar al-Assad during the uprising. In Aleppo, the Armenian party Tashnag organized pro-Assad demonstrations as early as march 2011, and notable Armenians immediately declared their support for the regime—a striking development given their traditional quietism on political matters. ThE OTTOMAN LEgACy Under the Ottoman millet system, Jews and various Christian denominations had autonomy, while most muslims were considered part of a single umma (community). heoretically, there was no diference between Sunnis and Shia in this system, but in practice some groups were ostracized. Sects that were deemed heterodox, such as Alawites, Druze, and Ismailis, were not consid- ered muslims, so they were usually isolated in the mountains or used as agricultural laborers and bufers against Bedouin attacks on the plains. During the nineteenth and twentieth centuries, however, these heterodox sects were more or less recognized as muslims for political reasons. Ismailis were placed under Ottoman protection during the mid-nineteenth century to escape persecution from Alawites in the coastal mountains, giving them the possibility of integrating into the umma. he Alawite-Ismaili conlict was not a religious war, but a broader sectarian conlict motivated by demographic pressures that had pushed the Alawites to take land from the outnumbered Ismailis. By 1919, Alawites had launched a revolt rooted in their disputes with Ismailis. It was led by Sheikh Saleh alAli, the irst revolutionary of modern Syria, who took advantage of the Ottoman Empire’s collapse to attack Ismaili lands around his tribal village of Sheikh Badr. When the French army intervened to end the ighting, it caused a widespread revolt in the southern Alawite mountains.1 Decades later, in 2004, tensions between Alawites and Ismailis in al-Qadmus would provoke a similar confrontation, prompting local Ismailis to ask for help from their brethren in masyaf and Salamiya. Authorities eventually calmed their constituents, but the conlict has remained latent ever since. In 1932, Alawites and Druze were formally recognized as muslims through a fatwa issued by the mufti of Jerusalem, Hajj Amin al-Husseini. His aim was to unify Arabs against the Zionist threat in Palestine and Franco-British colonization, as well as stem the growing wave of Alawites converting to Christianity. At the time, Jesuits were reportedly beginning to convert some Alawite villages of southern Jabal al-Ansariyya. Because Alawites were not considered muslims prior to the fatwa, the missionaries were permitted to proselytize in their communities under an Ottoman “Capitulation” treaty covering conversions outside the umma. his led to an outcry among Sunni and Arab nationalists, spurring the mufti’s fatwa. Each of these historical cases shows how one Syrian community’s recognition or mistrust of another depends more on the political state of play than on religious dogma. And for individual members of these communities, exempting oneself from this sectarian wagon-circling is often impossible. In theory, communities are composed of believers, practitioners, and non-believers, but the last group is largely silent in the middle East. In Syria, “atheists” are not recognized: everyone possesses a religion by birth, marriage, or conversion, and eschewing one altogether is not a practical option. his obligation is both administrative and social— declaring oneself an atheist typically leads to marginalization or ostracism.  FABrICE BAL ANChE N TURKEY W E S iterr anea n Se a Aleppo Raqqa Kasab Latakia Med Jabal al-Ansariyya Hama Deir al-Zour Salamiya Homs Tripoli Beirut Mount Lebanon LEBANON Saida Damascus Jabal Amil IRAQ Haifa Mount Carmel Jabal al-Druze 59. SECTARIAN DISTRIBUTION IN THE LEVANT AT END OF OTTOMAN EMPIRE Irbid ISRAEL Jaffa Amman SUNNI MAJORITY Jerusalem Bethlehem Sedentary Madaba Nomadic Karak RELIGIOUS MINORITIES JORDAN Alawite Druze SAUDI ARABIA Shia Ismaili Maan Christian EGYPT Aqaba S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r  0 100 km Contemporary border  111 Iskenderun 60. CENTERS AND PERIPHERAL ZONES IN THE LEVANT AT END OF OTTOMAN EMPIRE Antakya Aleppo Latakia Hama ALAWITE MEDITERRANEAN SEA Homs Central city (Sunni and Christian) ALAWITE Center (communities integrated into power) Tripoli BEDOUIN Bedouin (repelled periphery) MARONITE CIRCASSIAN DRUZE Saida Refuge mountain (forgotten periphery) Damascus Beirut Frontline (dominated periphery) SHIA DRUZE Haifa Tel Aviv Jaffa Jewish settlement Contemporary border MARONITE DRUZE 112  ISMAELI Coastline Jerusalem CIRCASSIAN  FABrICE BAL ANChE 61. OLD CITY OF DAMASCUS AT END OF OTTOMAN EMPIRE Souk Harir (silk market) Citadel BAB TOUMA Umayyad Mosque CHRISTIAN AREA Souk Hamidyeh MUSLIM AREA Saint John of Damascus Souk Midhat Bacha Patriarchate Greek-Catholic Souk al-Souf (wool market) 0 S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r  JEWISH AREA Jewish School 200 m Wall BAB SHARQI Mosque Church Synagogue  113 Sandjak of Iskenderun 62. THE ALAWITE STATE 1920–1936 To Antakia To Aleppo KASAB (detached from Turkey in 1942) JISR AL-SHUGHOUR BASSIT and AKRAD (detached from Iskenderun in1924) MAIN RELIGION KANSABA Sunni AL-HAFFAH MAARAT AL-NUMAN MZERAA Christian LATAKIA (state capital) QALAAT AL-MADIQ Ismaili Mediterranean Sea JABLEH Alawite State of Syria BANIYAS International border AL-QADMUS HAMA AL-RAWDA MASYAF AL-SAWDA Province border KHAWABI DRAYKISH TARTUS SAFITA ARWAD Alawite state MASHTA AL-HELU Non-Alawite population TAL KALAKH (detached from Lebanon in 1921) HOMS Province/statelet capital QOUBAIYAT District center TRIPOLI 0 20 km State of Great Lebanon To Baalbek To Damascus Locality Road 114   FABrICE BAL ANChE ThE TrIBE: ThE BASIC uNIT OF SECTArIANISM Although ethnicity and religion are important factors in Syrian sectarianism, clan and tribal ailiations are its most basic elements. he clan is actually a subset of the tribe, but for the sake of simplicity, this chapter often treats them as synonymous. Philippe DrozVincent provided a succinct description of the modern tribal phenomenon in the middle East: he word “tribe” is...more stable than the reality it is supposed to describe. he “Bedouin” world has been destroyed, through sedentarization, and by the state powers. But the tribal culture and its references to blood and ancestors, a genealogical hierarchic, the use of power inside family groups persist...he tribes even found in some states, such as Iraq in 1990, a new strength. Distinguishing diffuse social tribalism within a society that inds ways to survive in diicult conditions (“ruralization” of city neighborhoods by rural migration, import of the tribal ethos in solidarity networks), of political tribalism in power struggles between family networks deployed around leaders. he “reinvented” or rediscovered tribalism is therefore a symptom of the State (nation) weakening.2 Tribalism is more prevalent in Syria than in most middle East states. It is not limited to remote regions such as Jazira; one can also ind it in the coastal Alawite mountains and the southern Jabal al-Druze region. Because Syrian society is not nomadic, Alawite tribes are based on territorial ties rather than blood ties; references to a common ancestor exist, but territory is the primary deining element. In a traditional Alawite village, the sheikh gives the tribe its name and serves as guarantor of village unity. Land is the means by which the community survives and is sustained, while the extended family defends the land politically. As for the Kurds, the importance of tribalism decreased temporarily after 2004, when the regime allowed political parties to proliferate following the local uprising and crackdown. At the time, each major Kurdish tribe created its own party. Interestingly, the two most prominent factions—the Kurdistan Democratic Party of Syria (KDPS) and the now-hegemonic Democratic Union Party (PYD)—lack a tribal ailiation and difer ideologically. he PYD adheres to a marxist ideology while the KDP is liberal, producing a notable diference in recruitment: the PYD attracts workingclass Kurds while the KDP attracts the urban middle class. Yet the political future of the Kurdish region is uncertain because the war and associated quest for autonomy are spurring some parts of the community to move back toward tribal and ideological solidarity. A similar war- S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r  time reversion to tribalism can be seen in other parts of Syria as well. PrIMAry LINkS ArE PArAMOuNT When analyzing the connections between individuals, anthropologists often distinguish between primary links (family, sect, tribe, etc.) and secondary links (political ailiation, class, etc.). In Syria, primary links are strengthened by a system of patronage that forces citizens to use sectarian networks to access resources. Economic diiculties often prevent individuals from leaving their families and communities—they need primary solidarity networks to survive. he potential for “emancipation” from one’s sect is stronger when an individual is higher in the social hierarchy, but lower-status members will nevertheless solicit such individuals for privileges, essentially pulling them back into their sectarian identities. he regime’s core Alawite constituency is the most glaring example of these phenomena, since the Assad family relies primarily on Alawites to control the country. At the same time, it does not stop them from expanding their patronage network to include non-Alawites that belong to the second or third circle of power. In peacetime, a sect is a social network to be mobilized when one needs to ind a job, get preferable treatment in dealing with the bureau- cracy, and so forth. his communitybased patronage system has endured for centuries.3 Even in times of peace, those who denounce this system and turn away from their sect risk marginalization; if they do so during wartime, they risk their very survival. MINOrITIES ANd NATIONALISM In the muslim world, the idea of nation is often conceptualized in three diferent ways: as the community of religious believers (umma), as Arab nationalism (al-watan al-Arabiya), and as the nationalism of individual states (qawmiyya). Scholar Xavier de Planhol best deined these fundamental constructions in 1993, describing their ambiguity in the Arab world and explaining why they had not supplanted communitarianism (taiiyya).4 He concluded that one should have no illusions about the weak integrative capacity of supra-community identities, including Arabism, and that further regional fragmentation along sectarian lines should be expected. In 1984, Laurent and Annie Chabry had ofered a similar criticism of the Arabism on which contemporary national constructions are based, arguing that it had attempted to “blur and erase” essential aspects of ethnic and minority identity in a manner that ultimately proved more divisive to Syrian society.5  115 In addition, the concept of umma competes with the notion of the Arab homeland (watan). he term watan itself is ambiguous because it can signify both “country” and “the Arab world.” Within Syria’s Baath Party, the contest between Arab nationalists (who focused on defending the “Arab Nation”) and regionalists (who sought to defend the Syrian homeland) ended in the latter’s victory toward the end of 1960s. he Arab nationalists were mainly Sunni, while the regionalists primarily consisted of minorities (Alawites, Christians, Druze, and Ismailis). Haiz al-Assad was the irst prominent Syrian igure to promote and implement a political ideology based on building a unique Syrian identity, even if the oicial Baath Party line remained focused on Arab nationalism. he failure of the pan-Arab project became evident in 1961, when Syria unilaterally dissolved its union with Egypt after the population rejected being subsumed as an Egyptian colony. To be sure, Assad oicially maintained the dogma of Arab unity, speaking of the Syrian Baath Party as a regional subcommand of the higher “Arab National” command. Yet this was a iction intended to unite the Arab nationalists and regionalists while the regime consolidated power. Syria’s Political Life Has Been Locked Since 1963 Following the 1963 coup, political competition was completely suppressed, 116  with the Baath becoming the country’s only party until 1971. Assad eventually softened his rule by authorizing some “brother” parties to form, integrating them into the wider National Progressive Front. hanks to this allegiance, these parties gained parliamentary and ministerial positions. Decades later, as the 2011 uprising unfolded, the regime passed a new law authorizing the creation of parties outside the National Progressive Front. According to the new constitution adopted after a February 2012 referendum, the Baath are no longer the country’s oicial ruling party; the president can be elected from a diferent faction. In practice, however, nothing has changed. he revised constitution also stipulates that any new party must be represented in all of Syria and eschew religious or ethnic identity. Oicially, the purpose of this provision is to prevent the rise of sectarian or regional parties that might call the country’s unity into question. Yet the ulterior motive is to exclude the muslim Brotherhood; membership in this openly Sunni Arab, panIslamist organization is still punishable by death. Regardless of the avowed Baath effort to prevent sectarian politicking, the communitarianism that deines Syrian society is naturally relected in the ruling party’s composition and policies. he founding Baath project of building a “Syrian Arab Nation” where sectari- anism and tribalism would cease to exist never materialized. he leaders who believed in this utopia were swept away by realists and opportunists like Haiz al-Assad. Although the Baath recruited from all ethnicities and religions, Assad ensured that his Alawite constituency remained ascendant. Syria’s other political parties also tend to favor speciic minorities, even if their “universalist” names and rhetoric might indicate otherwise. he Syrian Social Nationalist Party (SSNP) exempliies this practice. Founded in Lebanon in 1932, the party is now headquartered in the Christian district of Hasaka in northeast Syria, where it has become a central part of the city’s Christian community. Prior to the Assad era, the SSNP served as an opposition party to the Baath. Both factions were dominated by non-Sunnis, but they pursued difering sectarian strategies. minorities in the Baath Party tended to promote Arab unity in order to defend themselves against the Sunni political Islam of the muslim Brotherhood. In contrast, the SSNP initially promoted the creation of “Greater Syria,” an entity that would have merged Lebanon, Syria, Jordan, Palestine, and Cyprus in a manner that allowed minority groups to outnumber Sunnis (though this quest was abandoned by the 1950s). Another SSNP defensive strategy was to break with transnational ideologies by using the term qawmiyya in its name rather than watan, emphasizing Syrian rather than Arab nationalism. he Syrian government outlawed the SSNP in 1955, and the party was not reauthorized until 2005. By then, the Baath and SSNP were no longer ideological antagonists, since the Baath had become a de facto “regionalist” party over the years. he phenomenon of communities forming political parties and alliances for sectarian rather than ideological reasons is hardly unique to Syria. In Lebanon, for example, maronite Christian leader michel Aoun allied with the Shia parties Hezbollah and Amal mainly to ward of perceived threats from the Sunnis—much like the Assad regime has done in Syria. meanwhile, leaders of the Sunni-majority Gulf countries ind it inconceivable and unacceptable that non-Sunnis are ruling any muslim nations. In their view, all positions of high authority should be held by Sunnis. A similar mindset prevails in Iran, where all high oices must be held by Shia. he ethnic factor has also played a role in the formation of Syrian political parties. Syrian Kurds, uncomfortable with the Arab nationalist character of the Baath, preferred to join the Syrian Communist Party (SCP), which was permitted to exist as a member of the National Progressive Front. In 1991, amid the fall of the Soviet Union, the party imploded in a dispute between pro-Soviets and Syrian nationalists, resulting in two separate communist parties that both  FABrICE BAL ANChE claimed the name SCP. While the split was publicly characterized as a political disagreement, in reality it resulted from a sectarian cleavage—as described in chapter 3, Kurds dominated one of the new splinter parties and Christians dominated the other. Since their 2004 revolt in Qamishli, Kurds have created additional political parties that the regime unoicially tolerates, each of them representing a speciic tribe rather than an ideology (apart from the PYD). How the Alawites Became a Client Haiz al-Assad’s early success permitted the ascension of his clan (the Kalbyeh) within Alawite society, to the detriment of other historically dominant Alawite clans and tribes such as the Haddadin and Khayatin. Yet ensuring Kalbyeh supremacy was not his primary objective; it mostly resulted from granting his relatives key positions of power. Within the bureaucracy, army, and public sector, factors such as familial ties, tribal proximity, and even personal friendship with the president soon supplanted personal merit as a mechanism of social advancement. Assad also had other levers for achieving and maintaining power. He created multiple alliances with other communities by exploiting state institutions and mobilizing traditional structures such as tribes and sects, giving him a broad base of support among the population. Even so, Assad was careful to place Alawites in the top positions most crucial to the regime’s grip on power. A comprehensive analysis of the main igures in the armed forces and internal security agencies between 1970 and 1997 shows that 61.3% of them were Alawite, compared to 35.5% Sunni.6 By 1992, seven out of the army’s nine divisions were led by Alawites, compared to two out of ive in 1973. he majority of these Alawite elites came from the hinterland of Jableh, the Assad clan’s stronghold. Given their geographical and familial proximity to the president, they constituted the center of the “ruling asabiyya,” one of the politicized solidarity structures described in the introduction to this study. Assad placed them in key strategic positions such as director of the mukhabarat and commander of the paramilitary Defense Companies. Although other seemingly crucial posts were given to Sunnis (e.g., defense minister and chief of staf), they proved to be mere igureheads—the real decisions were made by their Alawite deputies. he prominence of Alawites in the armed forces and security services was also evident at lower ranks, though exact statistics on this matter are unavailable. Today, military camps around Damascus, Homs, and Hama are populated exclusively with Alawite families, and in the villages of the Jabal al-Ansariyya region, it is rare for a family not to have at least one member in the military. his is the result of a process that began un- S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r  der the French mandate, during which Alawites were encouraged to enlist en masse. Adhering to the old adage that colonial powers prefer to use minority troops, France chose to integrate Alawites, Druze, Christians, Ismailis, and Circassians into its Army of the Levant rather than Sunni Arabs, whom the French regarded as too receptive to nationalist sentiments. By 1945, Alawites represented a third of this transnational army’s forces in Syria and Lebanon.7 Following the 1963 coup, the number of oicers from these minorities increased drastically, as the new leaders promoted members of their family, clan, or region in order to reinforce their power. Half of the 700 Sunni oicers who were dismissed after the coup were replaced by Alawites.8 he abortive November 1965 putsch by Sunni oicers,9 the elimination of Druze oicers in 1966–1968,10 and various removals of Ismaili oicers all reinforced the army’s Alawite character. By the time of the 2011 uprising, Sunnis had become more important in the rank and ile due to the Alawites’ relative demographic weakness.11 But the oicer corps were still majority Alawite, which helped hold them together in the early years of the war despite rampant Sunni desertions. Using Alawite Sheikhs to Assert Authority Traditional social support structures, particularly religion, have been essen- tial tools of the Assad regime for decades. Devoid of democratic legitimacy, and with the smokescreen of Arab nationalism fading, Haiz al-Assad (and, later, his son) turned to religion with greater frequency as a means of inluencing the population.12 he preponderant weight of religion in Syrian society left the regime with little choice—in January 1973, Assad’s attempt to enact a constitution that did not stipulate sharia as the source of law sparked riots.13 He was forced to reinstate the charter’s sharia roots, including an article mandating that the head of state has to be muslim. Likewise, during the economic liberalization period that began in 1991, the wealth of the new bourgeoisie spread but poverty grew, leaving a society that had hitherto been protected by socialist measures (e.g., food subsidies, public employment, rent control) with few mechanisms of support. he regime soon found itself relying on religion more than ever to avoid social pressures. Every Friday, the president would go to a mosque to pray, and this event was broadcast live on television. Additionally, leading representatives of each sect were invited to the presidential palace during every religious celebration. he regime has also used religion to strengthen cohesion among Alawites, usually by evoking the past persecution they sufered at the hands of Sunnis. his goes hand in hand with the tac 117 tic of stoking fear about the bloodshed Alawites would experience if they lost power. memories of the 1979 massacre of Alawite cadets at the Aleppo military school and the wave of muslim Brotherhood attacks against Alawite igures around the same time help reinforce the sect’s unity and its faithfulness to Assad. As a popular Alawite saying goes, “If you support Assad, you support yourself.”14 Another reason why the regime had to invoke religion to mobilize Alawites is because it could not rely on the tribal leaders who served as the traditional political elites, namely the Kinj15 and Abbas16 families. During his ascent to power, Haiz al-Assad was an Alawite military leader from the rural bourgeoisie, not a member of an important Alawite family. When he and others in the Alawite military class sought membership in the National Progressive Front, they were essentially rejecting Alawite feudal lords. hese lords had established a political alliance with Sunni and Christian oligarchs from the cities, with whom they shared an interest in dominating the peasant class and preventing the social ascension of the rural bourgeoisie. he elites were also mutually wary of agrarian reforms that hit large Alawite, Sunni, and Christian landowners in equal measure. In the 1990s, Assad’s ruling asabiyya tried to regain the support of former Alawite feudal lords who had lost many of their lands but still maintained symbolic inluence on the community. 118  For example, the Kinj family still led the Haddadin confederation, so the regime returned some of its lands in the Jableh plains17 and gave some of its members preferential access to high-proile state jobs (e.g., one was placed in charge of the prominent Sham Hotel in Latakia). Even so, Assad apparently felt compelled to cultivate additional means of inluence within his sect, a quest that led him to exploit Alawite sheikhs. Traditionally, sheiks have been secondary characters in the Alawite community; they subordinate themselves to tribal leaders and generally stay away from politics. At the same time, they are recognized by the community for their honesty, probity, and religious knowledge. hey also perform secular functions such as professor, judge, and farmer, resulting in a modest lifestyle that contributes to their “sheikh-ly” aura—contrary to members of the ruling asabiyya, who live luxuriously thanks to rampant, open corruption. Sheikhs mostly serve as guardians of Alawite doctrine and transmitters of the religion’s secrets, which puts them in charge of initiation. From Assad’s perspective, this was their most important role, since membership in the ranks of the initiated is an important distinction in the Alawite community, one that confers authority and respect while generating solidarity among initiates. In return, initiates must obey their sheikh as a spiritual father of sorts. Assad there- fore made sure to appear alongside his sheikh during various religious ceremonies. his individual had little inluence on the president, at least in policy terms; rather, Assad used him and other sheikhs to airm his religious authority over the community. His son would later copy this system of coopting the sheikhs; today, most Alawite military oicers belong to the circle of the initiated. In addition, the elder Assad became the honorary president of the “Council of the Alawite Sect” (majlis al-mili),18 an informal body composed of eighteen sheikhs from diferent tribes and tasked with deining the direction of the Alawite community. During the Brotherhood revolt of 1979–1982, he relied on this council to unify Alawites against the perceived Sunni Arab threat.19 After this period of disorder, the council became entirely subservient to Assad and limited itself to a local role of providing consultation to families.20 In its place, Haiz’s brother Jamil formed a separate association that fostered Alawite religious practice by equipping mosques, building and maintaining Alawite shrines, inancing mecca pilgrimages for sheikhs, and related activities. In more recent years, Bashar al-Assad has been less invested in supporting religious Alawites. He rarely visits his ancestral home of Qardaha, preferring to spend his time developing relationships with businessmen—a practice criticized by the sheikhs. Feeding on Alawite Poverty It is important to remember that political power in Syria does not belong to Alawites in general, only to certain Alawites. Not all Alawites support the regime simply because the president hails from their sect; opposition to Assad exists at both the clan and political level. he community certainly rallied behind the regime during the Brotherhood revolt, but they did so mainly because they were a favorite target of insurgents, not so much out of Alawite solidarity. Indeed, this dynamic—mobilizing Alawites by playing on fears of Sunni revenge if the ruling asabiyya loses power—has come to deine many of the Assad family’s eforts to maintain sectarian solidarity over the years. hese fears did not always exist, however, and they abated at the turn of the century as memories of the Brotherhood revolt faded. he 2011 uprising reactivated them, of course, but in previous years the regime had to rely on patronage relationships to maintain solidarity among poorer Alawites who held various grievances against the ruling elite. he Baath regime, like most other Arab regimes, tried to reduce poverty, alternately presenting it as a consequence of the colonial period or the preceding bourgeois-feudal regimes.21 Direct and indirect rents helped it inance development programs (including health and education), regulate the prices of every FABrICE BAL ANChE day goods and services, and drastically increase the size of the bureaucracy.22 he economic prosperity that Syria experienced between 1973 and the mid1980s also facilitated the reduction of inherited economic inequalities and the promotion of social mobility. Yet this mobility was fragile because it depended on public monies distributed by the state. When Saudi Arabia stopped making large donations to Syria in the 1980s, it contributed to a crisis in the welfare state and impoverished many public employees. hereafter, these employees had only two solutions to improve their situation: demand bribes or seek a second, more lucrative, job in the private sector. Not every position lent itself to procuring bribes, however, and many Alawites found it diicult to ind second jobs in their overpopulated mountain region with limited agriculture. hus, the assistance that Haiz alAssad preferentially ofered to such Alawites allowed him to cultivate an inner core of trusted servants and a true praetorian guard—one whose devotion was rooted in their material dependence on him. his spurred historian Raymond Hinnebusch to observe that after the Baath revolution, Syrian peasants went “from the service of the large landowners to the intelligence services.”23 hanks to its inluence on the state, Assad’s asabiyya was able to take advantage of land distribution during the agrarian reform era, monopolizing the allocation of agricultural loans, public sector hiring, road construction, irrigation, electriication, and so forth. Each project was inaugurated with fanfare and aixed with a placard thanking the president for his generosity. Not all of the Alawite community developed a client relationship with the regime: some refused due to political opposition or because they had other means of material autonomy. For example, many Alawite businessmen from Tartus spent years abroad and never developed patronage links with Assad; rather, their economic interests place them closer to the city’s Sunni/Christian bourgeoisie. As a result, they tend to sufer the same bureaucratic problems as others outside the ruling asabiyya—they do not get special treatment simply because they are Alawites. ThE STATE EXACErBATES SECTArIANISM In the middle East, the state is not independent of communities but closely linked to this form of societal and political organization. In Syria, the government oicially sought to dissolve individual communities in a broader Syrian Arab identity, but such rhetoric was at odds with the fact that sectarianism was omnipresent in the state’s very structure for decades. From daily administration to the higher-level func- S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r  tioning of institutions, nearly every aspect of governance fell back to sectarian membership. more than seventy-ive years ago, scholar Jacques Weulersse noted that any individual hired as a civil servant in the middle East automatically became a representative of his community within the government, and his observation still holds true in Syria today.24 Administrative Limits Create Sectarian Borders A state’s use of administrative divisions demonstrates its true policies toward sectarianism. hese policies can be used to reinforce traditional structures or destroy them. Actors within the state are not neutral, and patronage operates at its full potential because administrative promotions create a locus of power and result in job creation, service delivery, and other beneits. In 1945, the Syrian state inherited administrative divisions created by the Ottoman Empire, and France did not try to modify them during the mandate period. he French occupation was therefore only a digression—its impact on the organization of Syrian territory was minimal compared to Algeria, which experienced 130 years of colonial politics.25 When the French mandate divided the Levant, it demarcated the limits of sectarian space. he province of Suwayda, announced at the beginning of the independent Syrian state, stemmed from Jabal al-Druze, a Druze state created by the French.26 he Shia Beqa Valley was removed from Damascus province so it could be integrated into Lebanon; the Lebanon-Syria border separated a Christian-majority state and a Shia-majority region from a Sunnimajority territory, exceptions notwithstanding. he creation of the Alawite State in the 1920s also deined the Lebanon-Syria border by separating the Sunnis of Akkar from local Alawites, again with some exceptions. he Nahr al-Kabir River demarcated sectarian communities as well.27 Upon achieving independence, the states did not alter these divisions. he Alawite State became Latakia province. At the time, two non-Alawite districts (Tal Kalakh and masyaf ) were attached to this province to dilute its Alawite identity, per the policy of the Sunni Arab government. Yet these districts were later detached from Latakia in 1952. And in 1970, following Assad’s coup, the regime further divided Latakia by creating Tartus province—this time to strengthen the Alawite region by creating a second Alawite “capital.” Oicially, this administrative upgrade was justiied with “objective” reasons such as population distribution, territorial planning, remoteness from the capital, and so forth. Yet there was an obvious contradiction at work because these rationales were not used to promote other cities to provincial capi 119 tals—one of many regime methods of keeping Sunni Arabs and Kurds marginalized. For example, the manbij and al-Bab areas of Aleppo province could easily have been made into their own provinces at any point since the 1980s, but this would have meant devolving power from Aleppo city and creating new centers of Sunni power in an area already hostile to the regime. Similarly, the regime’s reticence to develop Qamishli into a provincial capital clearly indicated its desire to avoid reinforcing Kurdish nationalism in the northeast. he Baath regime also came to prefer smaller administrative divisions because state patronage worked best at that level. Important members of society could redistribute public monies through their personal networks more eiciently within these smaller frameworks than on the wider provincial level. he small divisions were also representative of existing local structures: for example, the new subdistricts established in the coastal region corresponded to the territories of Alawite clans and rural Christian communities28; in Jabal al-Druze, they were based on longstanding boundaries established by Druze clans29; and in Jazira, they were based on tribal boundaries.30 his system is well adapted to rural territories but not cities, especially large metropolises with mixed sectarian and tribal populations. 120  63. ADMINISTRATIVE DIVISIONS IN SYRIA HASAKA IDLIB 1970 RAQQA 1960 ALEPPO LATAKIA HAMA DEIR AL-ZOUR TARTUS 1970 HOMS RURAL DAMASCUS 1970 DAMASCUS 1960 province border New province QUNEITRA Complete Partial 1964 DERAA PRE-1960 POST-1960 SUWAYDA Province center created 0 100 km District center created  FABrICE BAL ANChE he Dual Strategic Role of Highway Networks he construction of Syria’s road and highway network has long obeyed a strategy of national construction. By concentrating roads on the coast to the detriment of the hinterland, the state reinforced its centralism. Accordingly, adjusting road plans to beneit a particular sectarian interest is not easy; roads are planned according to the interests of the state (which wants to control the entire country) and the economic bourgeoisie (whose commercial imperatives often transcend sectarian considerations). At the same time, the large-scale drawing of road networks and the priority given to road construction projects is clearly inluenced by sectarianism.31 he Aleppo-Latakia highway is illustrative. Its construction was interrupted in 1982 during the Brotherhood revolt and not restarted until 1995. he minister of communication gave inancial reasons to justify the delay, but the move was widely understood as a means of punishing Aleppo’s bourgeoisie for supporting the Brotherhood. he city’s Sunni Arab business class stood to beneit from being linked to the country’s principal port; Haiz al-Assad also feared that the highway’s construction would strengthen a sectarian axis between Latakia (whose population was half Sunni) and the northern Sunni areas of Jabal al-Ansariyya, Jabal al-Akrad, and Aleppo. Based on developments in the current civil war, those fears appear to have been well founded: between 2012 and 2015, rebels took the Sunni territories of Jabal al-Turkman and Jabal al-Akrad, then used them as launching points to threaten Latakia, the Alawite stronghold that was expected to be a refuge for the regime if Damascus fell.32 he Aleppo-Latakia highway gives this territory strategic importance for Sunnis in the interior; if Alawites were to secede, conditions would be suitable for the formation of a “Latakia corridor” similar to the Danzig corridor in World War II.33 From the beginning of the civil war, this corridor was the subject of intense ighting34 as rebels tried to gain access to the sea. While road construction was suspended between Latakia and Aleppo, the minister of communication announced that a highway would be built between Jableh and Hama, with a stop in Tal Salhab, a small Alawite town on al-Ghab plain. Once completed in 1993, the new highway had a steep grade that made it diicult for trucks to access. his problem was compounded by the fact that sections of the road between the coastal highways and the Aleppo-Damascus highway never materialized. In fact, the highway’s true function was to connect coastal Alawites with their al-Ghab brethren, reinforce the region’s Alawite territorial unity, and give the regime a route for sending the army to Hama if the Damascus-Aleppo highway was ever S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r  cut of by insurrection—the oicial economic goal for this costly highway was just a pretext. he war has only reinforced the importance of this strategic axis, which constitutes a vital artery for the defense of Alawites in al-Ghab. Secondary Road Networks Privatized by Sects Beyond major highways, Syria’s secondary road network relects the country’s social networks and their relationship with the state. he construction of such roads is not a natural step in Syria’s infrastructure development; instead, it results from tough negotiations between local populations and the state. Access to road networks is not a right but a beneit granted by the president in return for political allegiance. In the Jabal al-Alawiyya area, for instance, the merchedites35 towns were not accessible by paved road until the community clearly submitted to the Assad clan’s patronage at the beginning of the 1990s. And Khawabi, an isolated Sunni town in the Tartus hinterland, did not receive access to the road network until ten years after surrounding Alawite towns did. Hama province likewise illustrates the regime’s road-building calculus. he local road network is clearly organized to connect small cities with the countryside directly, allowing them to bypass the provincial capital, an insular, strictly conservative Sunni Arab city that served as a muslim Brotherhood stronghold for years. Alawite, Christian, and Ismaili populations in the province had already developed their own service centers in masyaf, Tal Salhab, maharda, and Salamiya to avoid going to Hama. In addition to linking these areas, the regime also wanted to further punish Hama for its 1982 uprising. In contrast, the road network in Homs province is centralized around the capital, transforming the area into an attractive cross-sectarian metropolis spread over a wide area.36 ECONOMIC ANd dEMOgrAPhIC TrANSITION CrEATES SOCIAL CrISIS Since 1991, when Haiz al-Assad enacted Investment Law no. 10, Syria has been trying to transition from a Baath-imposed command economy to a free-market system. his process has been slowed by the regime’s fear of a popular uprising and its omnipresent, entrenched bureaucracy. But the Assad family also realized that some liberalization was needed to meet the demands of entrepreneurs and integrate the hundreds of thousands of young people entering the job market, especially once the welfare state was no longer able to provide secure employment to the masses. At the same time, Syria has been undergoing a period of critical demographic transition. After half a century of  121 122  64. BAATHIST PLANNING 1963–1990: DEVELOPMENT OF THE PERIPHERY Turkey ALEPPO EU PHR ATES VALLEY RAQQA HASAKA IDLIB LATAKIA DEIR AL-ZOUR MEDITERRANEAN SEA natural growth in which the population doubled every twenty years, the birthrate inally slowed down after the turn of the millennium. Generations born during the demographic peak were still entering a saturated job market while expecting greater purchasing power than previous generations, but the number of positions created in the private sector had yet to compensate for losses in agriculture and industry.37 After Bashar al-Assad came to power, Syria was more open to foreign investment, with Gulf companies investing heavily in real estate on the coast and in the capital. But this created few jobs and much frustration—unlike in other emerging markets, Assad’s approach increased the wage gap and spurred economic growth that beneited a minority without encouraging wider social development. his new economic context challenged the geographical organization inherited from the Baath period, which had favored the peripheries from which new political employees hailed.38 Bashar’s generation of Damascus-born elites was uninterested in the peripheries and small cities where the Baath Party had its social base—the state was no longer willing or able to inance the past politics of territorial planning. Yet the regime seemed to forget that these were also the politics of security. For example, Assad abandoned the promised plan to develop northeast Syria after the 2004 Kurdish revolt, and this neglect JABLEH G HAB VALLEY HAMA BANIAS TARTUS COAS T HOMS Iraq ECONOMIC DEVELOPMENT PROGRAM GHAB Lebanon Dam irrigation State industry DAMASCUS Port and oil terminal Raw materials route URBAN HIERARCHY SUWAYDA QUNEITRA Capital Israel DERAA Jordan Province center Principal road  FABrICE BAL ANChE helped create the conditions for Rojava’s potential breakaway a decade later. Difering Growth Rates 65. PERIPHERY CLEAVAGE SINCE THE 1990s QAMISHLI Turkey ALEPPO RAQQA HASAKA Latakia DEIR AL-ZOUR MEDITERRANEAN SEA HAMA Tartus HOMS Iraq Center Lebanon Periphery Uninhabited area Capital DAMASCUS Secondary city Regional center Secondary city SUWAYDA Principal road DERAA Israel Jordan Secondary road International airport International port S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r  In the 1950s, Syria began a period of demographic transition. High birthrates and decreasing mortality due to medical progress generated fast population growth. By the early 1980s, Syria had one of the top ten fertility rates in the world with more than eight children per woman. Syria’s age pyramid indicated a decreasing birthrate in the mid-1990s, but it increased again later on, probably due to the re-Islamization of society. he birthrate fell again after 2004, but life expectancy increased, so the population continued to grow. Yet these demographic trends were uneven in Syria’s diferent communities. he fertility rate of religious minorities declined sharply over the past thirty years, but the Sunni Arab rate remained high, as shown by the difering population growth in Alawite and Druze provinces (Latakia, Tartus, Suwayda) vs. Sunni Arab provinces (e.g., Raqqa, Deir al-Zour, Idlib). he decline among Alawites was due in large part to the regime’s decades-long efort to improve educational opportunities for its core constituency, relecting the common correlation between higher levels of education and lower birthrates. After the 1963 coup, Alawites who held positions of responsibility in the bureaucracy routinely gave state jobs to  123 66. SYRIAN POPULATION 2010 AGE 80 + 75 – 79 70 – 74 65 – 69 60 – 64 55 – 59 50 – 54 45 – 49 40 – 44 35 – 39 30 – 34 25 – 29 20 – 24 15 – 19 10 – 14 5–9 0–4 1.4 1.2 1 0.8 0.6 0.4 0.2 0 (in millions) their coreligionists from neighboring villages. Yet chronically low levels of education on the Alawite coast eventually forced the state to draw from a pool of recruits who held no more than a certiicate of basic high school education and were only capable of performing the lowest-level civil service and security jobs. To consolidate his control over the state and avoid schisms between his co-religionists at the top and bottom of the bureaucracy, Haiz al-Assad recognized that he would need a steady low of faithful and suiciently skilled Alawites to ill every stratum of the state apparatus. Accordingly, he extended state education services to what were traditionally isolated and marginalized minority communities.39 124  0 0.2 0.4 0.6 0.8 1 1.2 1.4 Source: U.S. Census Bureau International Data Base As Alawite recruits swelled the rank and ile of the state apparatus, low-level posts became scarce, and competition for public employment became ierce. Political connections proved invaluable to many job candidates, giving Alawites a clear advantage due to regime favoritism.40 By 2004, more than 80% of employed Alawites worked in the military, state-led industry, or public sector.41 he proliferation of primary and secondary schools in the Alawite coastal region fostered such dominance.42 In 1960, around 60% of Latakia residents age ifteen and over were illiterate, corresponding with the national average at the time.43 By 2004, however, illiteracy had fallen to 10.7% in Latakia and 12.4% in Tartus,44 signiicantly lower than the national average of 17.5%. Furthermore, the gender gap in Alawite literacy rates disappeared entirely. As more and more women took up educational opportunities over the past two decades, many sought careers and put of getting married until they inished their studies, usually after age twentyive. hus, Alawite fertility rates steadily declined—within two generations, the average number of children per family dropped from eight to two.45 Birthrates in the Alawite provinces of Latakia and Tartus are now much lower than in the Sunni-dominated interior. Charting Syrian demographic growth in 1970–1981, the period in which Assad’s development policies were most efective, tells a diferent story from the last census period (1994–2004), when these policies were all but abandoned and the Alawite community had achieved its demographic transition. In the 1970s, the Alawite region’s growth rate was comparable to the national average, but by the 1990s it had noticeably declined, mirroring the pattern in other minority regions such as Suwayda, Shahba, Salkhad, and Salamiya. In Sunni-majority districts such as Hama and Abu Kamal, demographic growth exceeded 3% per year by 2004, but the igure was less than 2% in coastal Alawite districts such as Tal Kalakh and masyaf. he coastal Alawite region still enjoys a positive net migra FABrICE BAL ANChE tion rate (for reasons elaborated below); its lower rate of demographic growth can be attributed to a lower Alawite birthrate than in Sunni-majority regions, where net migration levels are negative. he aging populations of mountainous districts in Latakia and Tartus have experienced the most signiicant reduction, with young adults from Jabal al-Alawiyya locking to coastal towns and Damascus to take advantage of greater employment opportunities beginning in the 1960s. For their part, the coastal districts have been able to keep their growth rate between 2 and 3% thanks to the dense concentration of the labor force there. he growth rate in Kurdish-majority provinces has been weak as well, but this has much to do with their severe underdevelopment, which spurred many residents to migrate from areas such as Afrin, Kobane, Hasaka, and Qamishli to Aleppo and Damascus. Impoverished and underdeveloped Sunni-majority territories, such as the high plateaus of Qalamoun, witnessed equally strong migration to Damascus but still maintained their high birthrates. Consequently, the growth rate in the Sunni plains south of Damascus did not fall below 3%, and the same goes for the vast Sunni crescent stretching from the Iraqi border to the foothills of Jabal al-Ansariyya via Deir al-Zour, Raqqa, Idlib, and Aleppo. It is no coincidence that these “neglected but still growing” zones are where the current rebellion was most irmly rooted. When the State Disengaged he 2011 revolt began in the Houran, an agricultural region between Damascus and the Jordanian border. At the time, foreign journalists portrayed the Houran as a hotbed of perpetual opposition toward the Assad regime, but that was inaccurate. Up until the 1990s, locals actually supported the regime because they were still beneiting from the agrarian reforms it launched in the 1960s. Previously, the Houran breadbasket had been dominated by large landowners and Damascus merchants who monopolized agricultural proits, but the reforms redistributed a great deal of land to smaller farmers, agricultural banks, and other groups. he state also created crop-buying organizations to further break farmers’ dependence on large landowners who often doubled as agricultural products traders. he livelihood of local farmers improved, and the regime provided them with better public services to ensure their continued support. multiple villages grew into cities thanks to these public investments, most notably the small border town of Deraa, which became a regional service center. he Houran’s population increased rapidly due to high fertility rates, from 180,000 in 196046 to 900,000 in 2015.47 And local residents were appointed to many executive positions within the regime, including Farouq al-Shara, who served as foreign minister and vice president.48 S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r  Bashar’s succession in 2000 came at a time when third-generation beneiciaries of Syria’s agrarian reform were coming of age and discovering that the state’s eforts were failing in many respects. By this point, farms in the Houran and elsewhere had been subdivided and were insuicient to feed a family, despite the transition from cereal to fruit growing and market gardening. Overexploitation of aquifers had led the ministry of Agriculture to ban new water wells and shut down many illegal ones. hese measures were one of the primary reasons for local discontent because they deprived the rural population of their main source of income. Frustrations were exacerbated whenever exceptions were made for farmers close to the regime. he decrease in farm income in turn afected urban economies that were not diversiied and hence unable to absorb the rural underemployment. Although this combination of neglect and resentment afected much of Syria’s countryside, it was felt most acutely in the Houran because of the region’s proximity to the previously generous leadership in Damascus. he water problem greatly afected sectarian dynamics in other areas as well.49 In 2005, the regime launched a plan to modernize irrigation techniques and reduce agricultural water consumption, but the program’s mechanisms were questionable because they tended to favor large landowners and eliminate jobs. moreover, it did not solve the un- derlying problem of resource management, which requires massive public investment in rehabilitating the irrigation network, treating waste water, building new retention dams, establishing restrictive tarif policies in irrigated areas, and so forth. For example, the situation in al-Ghab plain, formerly a model of agricultural development,50 had become critical—the low of the Orontes River had dramatically decreased due to massive withdrawals upstream in Homs and Hama. hose responsible for managing the area’s water ignored the proliferation of illegal wells because they knew that state irrigation systems could no longer supply local needs. And just as in the Houran, oicials were tolerant of abuses and illegal activities by Alawites who were close to the regime. more illegal wells were permitted to spring up in the southern and western parts of al-Ghab, where Alawites live, than in the northern and eastern sections, where Sunni Arabs are the majority. he fact that the director of al-Ghab’s irrigation system is a member of the Assad family is probably no coincidence.51 In the northeast, the rural water situation came to resemble something out of John Steinbeck’s Grapes of Wrath. As described in chapter 3, domestic overconsumption in Syria and heavy withdrawals across the border in Turkey created a structural deicit, and the problem was exacerbated by several years of drought that began in 2006. he Khabur River  125 would have gone dry if the regime had not supplied it with pumps that tap into groundwater. As a result, cultivated areas in Hasaka province decreased by 25% between 1995 and 2008.52 he plan to modernize irrigation techniques fell far behind schedule there due to various bureaucratic obstacles.53 hanks to a 2008 agreement with Turkey, the regime proposed a huge project to transfer water from the Tigris River to the Khabur in order to prevent desertiication in Hasaka. he project would cost $2 billion, however, and would need ten years of work54 if Syria obtains international funding to carry it out. As for neighboring Jazira, the province’s agricultural crisis in the decade before the war contributed to discontent elsewhere in the country because several hundred thousand locals left the region to resettle in slums in Aleppo and Damascus. he state’s disengagement from large swaths of the countryside also had wider consequences for the many rural towns that had become small cities due to years of exceptional population growth. By 2011, many of these cities were openly rebelling because of neglect by the state, which had frozen administrative promotions and therefore the creation of public jobs. During the 1970s, the regime had encouraged the growth and health of small rural towns, providing public jobs and facilities that contributed to local economic dynamism and gave residents ample avenues of employment.55 he 126  67. DEMOGRAPHIC GROWTH IN SYRIA BY DISTRICT 1970–1981 KURDISH KURDISH KURDISH Hasaka Aleppo Idlib Raqqa Latakia Deir al-Zour ALAWITE Tartus Hama ISMAILI ALAWITE Homs Palmyra Damascus % AVG ANNUAL GROWTH 0–1 Golan Heights 1–2 2–3 Suwayda 3–4 Deraa DRUZE 0 100 km KURD Dominant minority Province border District border 4–5 Uninhabited area 5–6 Syrian Golan Source: Population census 1970 and 1981, Central Bureau of Statistics, Damascus  FABrICE BAL ANChE 68. DEMOGRAPHIC GROWTH IN SYRIA BY DISTRICT 1994–2004 KURDISH KURDISH KURDISH Hasaka Aleppo Idlib Raqqa Latakia Deir al-Zour ALAWITE Hama Tartus ISMAILI ALAWITE Palmyra Homs Damascus % AVG ANNUAL GROWTH 0–1 Golan Heights 1–2 2–3 Suwayda Deraa 3–4 DRUZE 0 100 km KURD Dominant minority Province border District border 4–5 Uninhabited area 5–6 Syrian Golan Source: Population census 1970 and 1981, Central Bureau of Statistics, Damascus S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r  number of districts (mantiqa) increased from forty to seventy between 1960 and 1994. Since then, however, no new provinces (muhafaza) have been established, even as the population increased from 14 to 22 million people before the war.56 Likewise, the number of district centers with more than 20,000 inhabitants (the oicial minimum for designation as a city in Syria) almost doubled,57 but the administrative status of these new rural cities was not formally changed. Public funds and facilities in Syria are distributed based on this categorization system, so these cities have been underfunded and underequipped for years, with fewer job openings in the public sector. his lack of public support has not been ofset by private sector development, mainly because the modernization of Syria’s transport networks reduces the commercial attraction of these small communities compared to larger cities. medium-size district capitals with more than 40,000 residents (e.g., Jisr al-Shughour, maarat al-Numan, mayadin, Rastan) are experiencing similar declines in favor of major cities for the same reasons. Notwithstanding the war, Syrian territory today echoes the period preceding the Baath era, as the ruling authorities focus their attention and largesse on the large cities along the Damascus-Aleppo axis with extensions to the coast. Tartus and Latakia have become de facto parts of this regime heartland thanks to their port infrastructure. In the northeast, the  127 Raqqa, Deir al-Zour, Hasaka, and Qamishli regions remain intermediate hubs that still bring local resources toward the center despite their contested status during the war. he mutation of Syria’s territory has resulted in a socio-spatial crisis that the regime did not anticipate and may yet be unaware of. 69. ILLITERACY IN SYRIA BY PROVINCE 2004 N W E S KURDISH 128  Hasaka Aleppo Wealth Gap Leads to Urban Crisis As discussed in previous chapters, socioeconomic tension between the country’s center and peripheries was an important driver of the war, and this urban/ rural cleavage was experienced in miniature within Syria’s major cities. In Latakia, for example, the revolt began in the poorest neighborhood, Ramel alFilistini, but stalled closer to the central squares because the middle and upper classes did not sympathize with the protestors—a clear indicator of social gaps materializing as territorial boundaries.58 Damascus is the best example of this phenomenon. Beginning in march 2011, the capital’s outlying areas suffered endemic revolts, and the army blockaded the town of Duma in the sprawling Ghouta suburb. he unrest was very much a function of Ghouta’s sharp contrast with the rest of Damascus on multiple levels: economics, class status, demography, urban planning, and so forth. By the time of the 2004 census, Ghouta was home to over 100,000 inhabitants, with a high population growth rate of 3.5%. most KURDISH KURDISH Idlib Raqqa Latakia ALAWITE Hama Tartus Deir al-Zour ISMAILI ALAWITE Homs Damascus % ILLITERACY Golan Heights > 30 20–30 Suwayda 15–20 Deraa ALAWITE Minority-dominant district Province border 10–15 DRUZE 0 100 km <10 Desert area Source: 1994 and 2004 population census, Central Bureau of Statistics, Damascus  FABrICE BAL ANChE 70. INTERNAL MIGRATION IN SYRIA, 1990 –2010 TURKEY JAZIR A HASAKA ALEPPO RAQQA IDLIB LATAKIA HAMA DEIR AL-ZOUR TARTUS HOMS LEBANON IRAQ DAMASCUS Alawite region with low fertility rate QUNEITRA Retired Alawite military and civil servant migration Golan Heights DERAA SUWAYDA Weak Aleppo attraction area Robust Damascus attraction area Flow of migrants N W JORDAN Daily civilian commute routes E Provincial/international border SAUDI ARABIA S 0 100 km S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r   129 of its buildings were one- or two-story ediices constructed quickly and illegally to meet the housing demand at a time of limited state resources. Indeed, the town has spread like an oil stain on farmland now submerged in concrete. Government planners typically take a long time to decide on urban development in the Ghouta, by which time facts on the ground have usually changed. his leaves the population with little choice but to build illegally and then negotiate access to urban services and retroactive legalization with public authorities, all in return for money and political allegiance.59 Oicially, urban growth near the capital was supposed to be limited to the northern wasteland beyond Jabal Qasioun in order to preserve the fertile lands of the Ghouta. But for various reasons—including lack of knowledge about how urbanization should be carried out, lack of access to housing, and corruption—over 80% of urban expansion occurred in the Ghouta instead. he northern districts attracted the middle classes who work in the public sector, have a regular salary, and as such are eligible for housing assistance. Today, these modern northern neighborhoods are mostly populated by oicials who originally hailed from other provinces. Indeed, the distribution of public and private employment perfectly illustrates the split between the capital’s northwestern and southeastern districts—that is, between the 130  new Damascus, which is loyal to the regime, and the Ghouta, which rose up violently. As for the center of Damascus, it was largely insulated from the uprising. he bourgeoisie presumably supported the regime because they believed it would ensure them social stability and, by extension, prosperity. Pro-Assad protests were organized with the support of leading businessmen, who gave their employees days of to attend demonstrations and inanced the associated festivities, such as a large July 2011 concert in Umayyad Square with singer George Wassouf.60 In their view, Assad has been trying to facilitate Syria’s business environment— he opened the door to Gulf investment (which was behind many real estate projects in greater Damascus), removed exchange controls, and allowed private banks. As a result, Syrian economic growth was recovering strongly up until the war. Yet this growth directly beneitted a wealthy, loyalist minority rather than the population as a whole. And while these wealthy individuals funded charities that sought to reduce the social cost of economic transition, their eforts were insuicient given the magnitude of Syria’s social degradation. In 2007, a UN Development Programme report noted that poverty afected 33% of Syrians, with 12.3% considered extremely poor. As Syrian economist Nabil Sukkar put it on the eve of the rebellion, “the wealth gap widens in Syria.”61 he regime’s “XI Plan” (2007–2011) made poverty reduction and job creation its major objectives, but the plan was unfeasible. In 2007, Assad called tourism “the new oil of Syria” and declared that it would create two million jobs by 2020.62 Clearly, the regime was either blind to the country’s realities or had run out of viable ideas to solve the unemployment problem. he resultant social situation generated tremendous frustration among young people, which helps explain the power of the protests that erupted on the outskirts of major cities. In the words of journalist Caroline Donati, “What will this generation that cannot aford consumer goods in the shop windows become? It is ready to radicalize.”63 his social crisis, she concluded, “will be expressed in sectarian terms.” SuPrA-SECTArIAN NATIONAL CONSTruCTION hAS FAILEd In social and economic terms, the Syrian revolt is comparable to what happened in Tunisia and Egypt, but it is also very diferent because of its sectarian character. most researchers do not want to use a sectarian lens when examining the conlict, preferring “neutral” social explanations. Syrian opposition leaders are uncomfortable with it as well, regarding sectarianism as an ar- chaic method of organizing society that is vulnerable to Western exploitation and imperialist control, in line with the theories of Edward Said and other intellectuals.64 Ironically, Assad uses the same argument in many of his speeches, claiming that Syria is the victim of a conspiracy by foreign actors who use sectarianism to achieve their ends— an astounding bit of hypocrisy from a regime that relies on the loyalty of an Alawite minority and does all it can to divide its Sunni citizens. more than anything else, the war has revealed the deep cracks in the system of power that Assad inherited from his father, and his inability to reform it. hus far, the regime has survived by ensuring that its security forces are dominated by elite Alawite units,65 and by cultivating support among minorities (despite Kurdish reluctance on that front). It has also sought to fracture Sunni Arabs along territorial, tribal, ethnic, and social lines, while simultaneously channeling their resentment into ideological byways such as Baath nationalism, the struggle for occupied Palestine, and resistance to foreign “imperialism.” Bashar’s marriage to a Sunni and his integration of the Sunni bourgeoisie into the ruling asabiyya were part and parcel of this strategy. Despite all these eforts to neutralize and coopt Sunni dissent, however, the fact remains that the majority of the population considers the regime Alawite  FABrICE BAL ANChE and therefore illegitimate. In addition to religious objections, they also oppose Assad’s geopolitics, particularly his alliance with Shia Iran. In their view, Syria is more beholden to the Gulf oil monarchies for its economic development than to Iran. hey believe that continuing the alliance with Tehran can only bring international sanctions to Syria while angering potential patrons in Saudi Arabia. In the years leading up to the war, the Sunni Arab community became more aware of its socioeconomic strength as the private sector grew and sectarian charitable associations began providing services that the welfare state could not.66 When the regime proved inadequate in the new suburban districts and neglected rural areas, Sunni imams and leaders of the bourgeoisie reconstituted their clientelist networks—a development that will make them increasingly diicult to control and divide even if Assad is able to survive the war. Perhaps most important, Syria’s demographic trends are irmly against the regime, and no amount of mass wartime displacements will change that fact. Over the years, high population growth accelerated the failure of its development policy, which sought to placate the Sunni majority through agrarian reform and job creation but was incapable of absorbing the massive number of unemployed youths. moreover, its favoritism toward minorities had the unintended efect of reducing their demographic weight. Since the 1980s, Alawites have undergone rapid demographic transition due to radical changes in their community, with the regime’s largesse essentially transforming them from a poor, illiterate peasant class to an educated corps of civil servants. heir fertility rate has quickly declined in the process, while Sunni birthrates remain twice as high.67 hus, even if political change does not come to Syria in the short term via foreign intervention, economic crisis, or insurrection, it will still arrive at some point due to the simple, stark math of sectarian demography. NOTES 1. Jacques Weulersse, Le pays des Alaouites [The country of the Alawites] (Institut Français de Damas, 1940). 2. Philippe Droz-Vincent, Moyen-Orient: pouvoirs autoritaires, sociétés bloquées [Middle East: authoritarian powers, locked societies] (Paris: Presse Universitaire de France, 2004). 11. Fabrice Balanche, “Go to Damascus, My Son: Alawi Demographic Shift under Baath Party Rule,” in Michael Kerr and Craig Larkin, The Alawis of Syria: War, Faith and Politics in the Levant (Oxford University Press, 2015), pp. 79–106. 12. Bertrand Badie explained that for many Muslims, only the power of Allah is legitimate, and he does not delegate this power to men. It is nevertheless necessary that men submit to a temporal power in order to avoid anarchy. From this perspective, “a bad or even unjust order is better than anarchy,” provided it is at least compliant with sharia. See Badie’s book Les deux etats [The two states] (Paris: Fayard, 1987), p. 334. 3. Jacques Weulersse, Le pays des Alaouites [The country of the Alawites] (Institut Français de Damas, 1940). Also see his book Paysans de Syrie et du Proche Orient [Peasants of Syria and the Middle East] (Paris: Gallimard, 1946). 4. Xavier de Planhol, Les Nations du Prophète [Nations of the Prophet] (Paris: Fayard, 1993), p. 24. 5. Laurent Chabry and Annie Chabry, Politique et minorités au Proche-Orient: les raisons d’une explosion [Politics and minorities in the Middle East: the reasons for an explosion] (Paris: Maisonneuve et Larose, 1984), p. 358. 13. Martin Kramer, “Syria’s Alawis and Shi’ism,” in Shi’ism, Resistance, and Revolution, ed. Martin Kramer (Boulder, CO: Westview, 1987), p. 248. 6. Hanna Batatu, Syria’s Peasantry, the Descendants of Its Lesser Rural Notables, and Their Politics (Princeton University Press, 1999), p. 219. 7. Nikolaos van Dam, The Struggle for Power in Syria: Politics and Society Under Asad and the Ba’th Party (London: I. B. Tauris, 1996), p. 4. 15. Under the French Mandate, the Kinj d’Ain Shqaq family from the plains of Jableh was recognized as the representative of the Haddadin tribal confederation. Ibrahim al-Kinj was president of the territory known as the “Alawite State” in the 1930s. 8. Ibid., p. 32. 9. Michel Seurat wrote about an additional putsch attempt in 1973, spearheaded by a group of Nasserite and pro-Iraqi Baath officers, But even in this case, the overwhelming majority of those involved were Sunni. See his book L’Etat de Barbarie (Paris: Le Seuil, 2012), p. 136. 10. An entire chapter of Nikolaos van Dam’s 1996 book The Struggle for Power in Syria is dedicated to this purS E C TA r I A N I S M I N S y r I A’ S C I V I L wA r  ging of Druze officers, which essentially led to the dismissal of the entire Druze community from power. 14. Michel Seurat, L’Etat de Barbarie (Paris: Le Seuil, 1989), p. 94. 16. The Abbas de Tlii family from northwest Akkar represented the Khayatin confederation under the French Mandate. 17. Eberhard Kienle, “Entre jama’a et classe: le pouvoir politique en Syrie contemporaine” [Between “jama’a” and social class: political power in contemporary Syria], Revue du monde musulman et de la Méditerranée 59, no. 1 (1991), p. 219, http://www.persee.fr/doc/remmm_0997-1327_1991_ num_59_1_2681.  131 18. Alain Chouet, “L’espace tribal Alaouite à l’épreuve du pouvoir ,” Maghreb-Machrek 147 (1995), p. 96. 19. For example, Michel Seurat offered an account of the council’s August 1980 meeting in Qardaha, in which Assad exhorted the entire Alawite hierarchy to stand together as a bloc to overcome the crisis, even if that meant cutting off economic activity with the Sunni bourgeoisie. See his book L’Etat de Barbarie (Paris: Le Seuil, 1989), p. 88. 20. Alain Chouet, “L’espace tribal Alaouite à l’épreuve du pouvoir ,” Maghreb-Machrek 147 (1995), p. 114. 21. The tactic of blaming the colonial experience was also used in Algeria, while blaming bourgeois-feudal predecessors was common in Egypt and Iraq. The phrase “bourgeois-feudal” may seem contradictory, but it suitably describes the historical alliance between Syria’s commercial bourgeoisie and great landowners. 22. Direct rent is the product of hydrocarbon exports. Indirect rent is aid given by oil-rich Arab countries or remittances from immigrants in these countries. Syria benefited from indirect rent between 1973 and 1987 as Arab countries with greater oil resources paid subsidies to support Damascus in its fight against Israel. See Blandine Destremau, “Pauvres et pauvreté en Afrique du Nord Moyen-Orient: essai de balisage d’une problématique de recherche en sciences sociale” [The poor and poverty in North Africa and the Middle East: an attempt to shed light on a research problem in the social sciences], Les Cahiers D’Urbama 13 (1997), p. 25. es] (Institut Français de Damas, 1940), p. 123. 25. Marc Côte, L’Algérie ou l’espace retourné [Algeria, or the returned area] (Paris: Flammarion, 1989). 26. Fabrice Balanche, Atlas du ProcheOrient Arabe [Atlas of the Arab Middle East] (Paris: Presses de l’Université Paris-Sorbonne, 2011), pp. 32–33. 27. Ibid., pp. 28–29. 28. Ibid., p. 147. 29. Cyril Roussel, Les Druzes de Syrie: Territoire et mobilité [The Druze of Syria: territory and mobility] (Beirut: Institut Français du Proche-Orient, 2011), http://www.ifporient.org/node/1024. 30. Myriam Ababsa, Raqqa: territoires et pratiques sociales d’une ville Syrienne [Raqqa: territories and social practices of a Syrian city] (Beirut: Institut Français du Proche-Orient, 2009). 31. Fabrice Balanche, “La prise en compte du facteur communautaire dans l’analyse de l’espace syrien” [Taking the community factor into account in spatial analysis], Géographie et Culture, May 2005, pp. 5–22, https://halshs.archivesouvertes.fr/file/index/docid/61242/ filename/La_prise_en_compte_du_facteur_communautaire_dans_l_analyse_ spatiale_article.pdf. (Washington Institute for Near East Policy, September 23, 2016), http://www. washingtoninstitute.org/policy-analysis/view/latakia-is-assads-achilles-heel. 43. According to the Syrian government’s 1960 census. 35. The merchedites are an Alawite subgroup distinct from the rest of the community because of their belief in the divine essence of Sleiman Merched, a prominent tribal leader during the French Mandate whom they regard as their prophet and God. 44. According to the 2004 census. 36. Fabrice Balanche, “Transports et espace Syrien” [Transportation and the Syrian space], Annales de Géographie 630 (March–April 2003), pp. 146–166, https://hal.archives-ouvertes.fr/file/index/docid/61252/filename/Transports_ et_espace_syrien.pdf. 47. Estimate based on Syria’s 2010 census. 37. Samir Aita, “Labour Market Policies and Institutions, with a Focus on Inclusion, Equal Opportunity and the Informal Economy: The Case of Syria,” Etude pour l’Organisation Internationale du Travail, 2010, http://www. economistes-arabes.org/Cercle_des_ economistes_arabes/Samir_Aita_files/ Ilo-aita-summary.pdf. 38. Alasdair Duncan Drysdale, “Center and Periphery in Syria: A Political Geography Study,” PhD dissertation, University of Michigan, 1977. 39. Ibid., p. 199. 32. Fabrice Balanche, “Il feudo degli Alaouti” [The Alawite feud], LIMES, March 2013, http://www.limesonline.com/cartaceo/il-feudo-degli-alauiti?prv=true. 40. Fabrice Balanche, La region Alaouite et le pouvoir Syrien [The Alawite area and the Syrian power] (Paris: Karthala, 2006), p. 164. 41. Ibid., p. 165. See also Noujoud Allouche, “La péri-urbanisation de Lattaquié” [The peri-urbanization of Latakia], PhD dissertation, Josef Fourier University, 2009, p. 239. 23. Raymond Hinnebusch, Peasant and Bureaucracy in Ba’thist Syria (Boulder, CO: Westview, 1989), p. 325. 33. Fabrice Balanche, “Syrie: la Montagne des Kurdes, prochaine zone martyre de la révolte?” [Syria: Kurds Mountain, site of the revolt’s next massacre?], Libération, June 12, 2012, http://www. liberation.fr/planete/2012/06/12/syriela-montagne-des-kurdes-prochainezone-martyre-de-la-revolte_825808. 24. Jacques Weulersse, Le pays des Alaouites [The country of the Alawit- 34. Fabrice Balanche, “Latakia Is Assad’s Achilles Heel,” PolicyWatch 2489 132  Study,” PhD dissertation, University of Michigan, 1977. 42. A quarter of all secondary schools opened in Syria between 1963 and 1970 were located in this region. See Alasdair Duncan Drysdale, “Center and Periphery in Syria: A Political Geography 45. Youssef Courbage and Emmanuel Todd, Le rendez-vous des civilisations [The rendezvous of civilizations] (Paris: Seuil, 2007), p. 56. 46. According to the 1960 census. 48. Bashar al-Assad put Shara in charge of the national dialogue committee in 2011 to make sure the opposition would accept the program of “reforms” the regime was offering. The decision to appoint a politician from the Houran was no accident. Shara was later put under surveillance due to concerns that he would defect. 49. Fabrice Balanche, “La pénurie d’eau en Syrie: compromis géopolitiques et tensions internes” [Water scarcity in Syria: geopolitical compromises and internal tensions], Maghreb-Machrek, September 2008. https://www.academia. edu/769789/La_p%C3%A9nurie_deau_ en_Syrie_Water_scarcity_in_Syria. 50. Francois Metral, “Le monde rural Syrien à l’ère des réformes (1958–1978)” [The Syrian rural world in the age of reform], in André Raymond, ed., La Syrie d’Aujourd’hui (Paris: National Center for Scientific Research, 1980), pp. 297–328, http://books.openedition. org/iremam/744. 51. As the author discovered during a research trip to Hama in January 2011. 52. Statistic from the French Ministry of Agriculture, January 2011. 53. For example, most farmers in Hasaka did not own the land they cultivated, so they could not secure loans to purchase irrigation equipment. The regime half-heartedly sought to address  FABrICE BAL ANChE this problem when the uprising broke out in 2011, with little success. com/pulse/business/2012/01/boom-tocrash.html. 54. Author interview with Syria’s director of irrigation, Hasaka, January 2011. 63. Caroline Donati, L’exception Syrienne: Entre modernisation et résistance [The Syrian exception: between modernization and resistance] (Paris: La Decouverte, 2009), p. 269. 55. Fabrice Balanche, “Le Djebel Ansaryeh: une montagne assistée” [Jabal al-Ansariyya: an assisted mountain], Montagnes Méditerranéennes 14 (2001), pp. 183–192, https://halshs.archivesouvertes.fr/halshs-00009578. 56. This estimate is an extrapolation from Syria’s 2004 census data; for more information, see the notes on methodology in other parts of this study. 57. In the 1994 census, 67 towns had more than 20,000 residents; by 2004, the figure was 125. 58. Alain Reynaud, Société, espace et justice: inégalités régionales et justice socio-spatiale [Society, space and justice: regional inequalities and socio-spatial justice] (Paris: Presses Universitaires de France, 1981). 59. Fabrice Balanche, “L’habitat illégal dans l’agglomération de Damas et les carences de l’etat” [Illegal settlements in the agglomeration of Damascus and the shortcomings of the state], Revue de Géographique de l’Est 49, no. 4 (2009), http://rge.revues.org/1980. 60. Wassouf is a singer from Kafroun, a Christian village in the southern part of Jabal al-Ansariyya. During the Damascus show he called for national unity and performed a song supporting Assad. He has lived abroad since 1969, so his gesture was important for the regime. See footage of the performance at https://www.youtube.com/ watch?v=uG27BGQtzmY. 64. In his 1978 book Orientalism, Said denounced communitarianism as a Western invention intended to weaken Arabs. 65. For example, the all-important 4th Armored Division is led by the president’s brother and wholly staffed by Alawites. The regime’s other armored units and its air force are Alawite as well. This arrangement helps offset the fact that most of the army’s rank and file are Sunni, and that many Alawite youths have been avoiding military service in recent years due to improvements in their standard of living. 66. Thomas Pierret and Kjetil Selvik, “Limits of ‘Authoritarian Upgrading’ in Syria: Private Welfare, Islamic Charities, and the Rise of Zayd Movement,” International Journal of Middle East Studies 41, no. 4 (November 2009), https://www.cambridge. org/core/journals/international-journal-of-middle-east-studies/article/ div-classtitlelimits-of-authoritarian-upgrading-in-syria-private-welfare-islamic-charities-and-the-rise-of-the-zaydmovementdiv/6126FC225642D3203CE980ACAA33C6C0 67. Youssef Courbage and Emmanuel Todd, Le rendez-vous des civilisations [The rendezvous of civilizations] (Paris: Seuil, 2007). 61. Roueida Mabardi, “Syria’s economic reforms see wealth gap widen,” Daily Star, April 28, 2010. 62. Jihad Yazigi, “Syrian Tourism Industry: From Boom to Bust,” Al-Monitor, January 24, 2012, http://www.al-monitor. S E C TA r I A N I S M I N S y r I A’ S C I V I L wA r   133 THE WASHINGTON INSTITUTE FOR NEAR EAST POLICY BOARD OF DIRECTORS BOARD OF ADVISORS Chairman Martin J. Gross Gen. John R. Allen, USMC Birch Evans Bayh III Howard L. Berman Eliot Cohen Henry A. Kissinger Joseph Lieberman Edward Luttwak Michael Mandelbaum Robert C. McFarlane Martin Peretz Richard Perle Condoleezza Rice James G. Roche George P. Shultz R. James Woolsey Mortimer Zuckerman President Shelly Kassen Chairman Emeritus Howard P. Berkowitz Founding President, Chairman Emerita Barbi Weinberg Senior Vice Presidents Bernard Leventhal Peter Lowy James Schreiber Vice Presidents Benjamin Breslauer Walter P. Stern Vice President Emeritus Charles Adler Secretary Richard Borow Treasurer Susan Wagner Board Members Jay Bernstein Anthony Beyer Robert Fromer Michael Gelman Roger Hertog, emeritus Barbara Kay Bruce Lane Moses Libitzky Daniel Mintz Lief Rosenblatt Zachary Schreiber John Shapiro Merryl Tisch Diane Troderman Gary Wexler In Memoriam Richard S. Abramson, president Fred S. Lafer, chairman emeritus Michael Stein, chairman emeritus Fred Schwartz, board member EXECUTIVE STAFF Executive Director Robert Satloff Managing Director Michael Singh Counselor Dennis Ross Director of Research Patrick Clawson Director of Publications Mary Kalbach Horan Director of Communications Jeff Rubin National Director of Development Dan Heckelman Chief Financial Officer Laura Hannah Operations Manager Rebecca Erdman AB Ou T T h E Au T hOr F ABRICE BALANCHE, an associate professor and research director at Université Lyon 2, as well as director of the Research Group on the Mediterranean and the Middle East (GREMMO), was a visiting fellow at The Washington Institute from 2015 to 2017. His primary fields of research are the political geography of the Arab world and the geopolitics of the Middle East, with particular focus on Lebanon and Syria, where he has spent 10 years since first engaging in fieldwork in 1990. Dr. Balanche is frequently consulted as an expert on Middle East development issues and the Syrian crisis. As head of the Institut Français du Proche-Orient (IFPO) Urban Observatory in Beirut between 2003 and 2007, he carried out expert work on the issue of water and the environment in the Middle East. His much-acclaimed maps of the Syria conflict created during his Washington Institute tenure have been republished widely. A prolific writer, he is the author of Atlas of the Arab Near East (2012, in French, English, and Arabic), and the book version of his thesis, The Alawite Region and Syrian Power (2006, in French). Balanche holds masters and doctoral degrees in geography from the Université de Tours, an accreditation to supervise research from Université Lyon 2, and an undergraduate degree in history and geography from Université Besançon. He is currently a visiting fellow at the Hoover Institution. THE WASHINGTON INSTITUTE FOR NEAR EAST POLICY W W W. WA S H I N G T O N I N S T I T U T E . O R G