Compassionate Communalism: Welfare and Sectarianism in Lebanon

By Melani Cammett

In Lebanon, religious parties such as Hezbollah play a critical role in providing health care, food, poverty relief, and other social welfare services alongside or in the absence of government efforts. Some parties distribute goods and services broadly, even to members of other parties or other faiths, while others allocate services more narrowly to their own base. In Compassionate Communalism, Melani Cammett analyzes the political logics of sectarianism through the lens of social welfare. On the basis of years of research into the varying welfare distribution strategies of Christian, Shia Muslim, and Sunni Muslim political parties in Lebanon, Cammett shows how and why sectarian groups deploy welfare benefits for such varied goals as attracting marginal voters, solidifying intraconfessional support, mobilizing mass support, and supporting militia fighters.

Cammett then extends her arguments with novel evidence from the Sadrist movement in post-Saddam Iraq and the Bharatiya Janata Party in contemporary India, other places where religious and ethnic organizations provide welfare as part of their efforts to build political support. Nonstate welfare performs a critical function in the absence of capable state institutions, Cammett finds, but it comes at a price: creating or deepening social divisions, sustaining rival visions of the polity, or introducing new levels of social inequality.

Compassionate Communalism is informed by Cammett’s use of many methods of data collection and analysis, including Geographic Information Systems (GIS) analysis of the location of hospitals and of religious communities; a large national survey of Lebanese citizens regarding access to social welfare; standardized open-ended interviews with representatives from political parties, religious charities, NGOs, and government ministries, as well as local academics and journalists; large-scale proxy interviewing of welfare beneficiaries conducted by trained Lebanese graduate students matched with coreligionist respondents; archival research; and field visits to schools, hospitals, clinics, and other social assistance programs as well as political party offices throughout the country.

• Language: English
• Publisher: Cornell University Press
• Release date: Apr 17, 2014
• ISBN: 9780801470318

Book preview

Compassionate

Communalism

WELFARE AND

SECTARIANISM IN

LEBANON

Melani Cammett

Cornell University Press

ITHACA AND LONDON

For the people

who shared their stories—

may their future be

peaceful and secure.

Contents

List of Figures, Maps, and Tables

Acknowledgments

Introduction

1. Welfare and Sectarianism in Plural Societies

2. Political Sectarianism and the Residual Welfare Regime in Lebanon

3. Political Mobilization Strategies and In-Group Competition among Sectarian Parties

4. The Political Geography of Welfare and Sectarianism

5. Political Loyalty and Access to Welfare

6. Sectarian Parties and Distributional Politics

7. Welfare and Identity Politics beyond Lebanon

Conclusion: The Consequences of Welfare Provision by Identity-Based Organizations

Appendixes:

A. List of Elite Interview Respondents and Provider Questionnaire

B. List of Nonelite Interview Respondents and Questionnaire

C. National Survey Questions

Notes

References

Figures, Maps, and Tables

FIGURES

1.1 Patterns of community and individual-level distribution of welfare goods

1.2 Partisan commitments of risk and time for different forms of political participation

2.1 Lebanese schools by religious affiliation, 1920 and 1978

2.2 Students enrolled in public and private primary and secondary schools, selected years

2.3 Primary and secondary schools in Lebanon, 1974–2006

2.4 Hospitals by affiliation (2008 estimates)

2.5 Clinics and dispensaries by affiliation (2008 estimates)

4.1 A woman sells Hezbollah and Lebanese national flags, as well as posters of Sayyed Hassan Nasrallah, Hezbollah leader, in a southern suburb of Beirut, August 23, 2006

4.2 Amal Movement poster in southern suburbs of Beirut in 2009

4.3 Poster of Saad Hariri in Beirut during the 2009 elections with photos below of Rafiq Hariri and two Future Movement supporters killed during the clashes and their aftermath in May 2008

4.4 Lebanese Forces poster of Samir Geagea in East Beirut in 2005

4.5 Taxis contracted by the Armenian Tashnaq Party waiting to shuttle voters to polls across Lebanon during the June 2009 national elections

4.6 Demographic spread of major religious communities in Lebanon

4.7 Propensity to target mixed or out-group communities (measured by fractionalization) by institutional type

4.8 Propensity to target in-group communities (measured by percentage of co-religionists) by institutional type

4.9 Distribution of households by socioeconomic status (SES) and propensity of institutional types to target different communities by SES in Beirut and Mount Lebanon provinces

6.1 Food box distributed during Ramadan by the Beirut Association of Social Development, a charitable organization linked to the Future Movement

6.2 Rafiq Hariri Government Hospital with a photo of Hariri on the façade

6.3 Photos of Pierre and Amin Gemmayel in the pharmacy of the Kataeb Party clinic in Mar Mikhael, East Beirut in 2007

6.4 Voter turnout by sect in the 1992 and 1996 Lebanese national elections

MAPS

2.1 Territories controlled by militias in Lebanon (approx. 1983–1989)

4.1 Governorates, districts, and administrative zones in Lebanon

4.2 Neighborhoods in West Beirut

7.1 Sadrist offices in Baghdad (post-2007)

7.2 Sadrist offices in Iraq (approx. 2009)

TABLES

1.1 Political mobilization strategies, intrasect competition, and the distribution of social benefits by sectarian parties

1.2 Indicators of state-centric and extra-state political mobilization strategies

1.3 Typology of domestic providers in plural societies

1.4 Patterns of distributing social benefits of selected sectarian parties

2.1 Coverage and benefits of public and private insurance schemes

4.1 Descriptive statistics of the dependent variables (zone level)

4.2 Descriptive statistics of the independent variables (zone level)

4.3 Presence of party-based welfare agencies in community (SUR)

5.1 Measures of the dependent variable, Welfare, and its component forms of social assistance

5.2 Coding and distribution of the independent variable PAI and its component forms of political activity

5.3 Measures of sectarian identity and partisanship

5.4 Political activity index by sectarian identity and partisan affiliation

5.5 Relationship between partisan and sectarian identities

5.6 Measures of main demographic control variables

5.7 Measures of piety, religious participation, and religious fractionalization

5.8 Mean socioeconomic status by sectarian identity and partisan affiliation

5.9 OLS regressions for the variable welfare

6.1 Selected communities with Future Movement and Hezbollah welfare agencies

6.2 Turnout rates by governorate in post-war national elections in Lebanon, 1992–2009

6.3a Vote share of Hezbollah by sect of registered voters in the 2005 and 2009 national elections (%)

6.3b Vote share of the Amal Movement by sect of registered voters in the 2005 and 2009 national elections (%)

6.3c Vote share of the Future Movement by sect of registered voters in the 2005 and 2009 national elections (%)

Acknowledgments

The list of people who have inspired and helped me while I researched and wrote this book is long. First and foremost, I am grateful to the Lebanese people who agreed to share their experiences with me and with the members of my research team. Many of their stories are heart-wrenching, exposing the conditions of insecurity in which they must navigate their lives. This book is dedicated to them.

In writing this book, I have benefited from the critical engagement of a large community of scholars, friends, and family members. Special thanks go to Pauline Jones Luong and Ellen Lust, both of whom read the entire manuscript and provided in-depth commentary. Maria Angelica Bautista and Wendy Pearlman provided extensive feedback on specific chapters, and Marsha Pripstein Posusney, who is sorely missed, helped me to hash out some basic ideas early on. As ever, I am grateful for the input and reality checks from Julie Lynch and Lauren Morris MacLean.

At various stages of the project, I received valuable feedback from Lisa Anderson, Lisa Blaydes, Dawn Brancati, Giovanni Capoccia, Kanchan Chandra, Yen-Ting Chen, Janine Clark, Dan Corstange, Bryan Daves, Lara Deeb, Mila Dragojevic, Grzegorz Ekiert, Tulia Falletti, Carol Hakim, Nahomi Ichino, Amaney Jamal, Jennifer Lawless, Evan Lieberman, Rick Locke, James Mahoney, Eddy Malesky, Tarek Masoud, Feryaz Ocakli, Roger Owen, Betsy Palluck, Liz Perry, Hugh Roberts, James Robinson, Jonah Schulhofer-Wohl, Dan Slater, Judith Tendler, Tariq Thachil, Lily Tsai, John Waterbury, Lisa Wedeen, and Steven Wilkinson. Many of my colleagues at Brown University were also generous with their feedback, including Engin Akarli, Peter Andreas, Linda Cook, Andrew Foster, Shirene Hamdy, Patrick Heller, Sukriti Issar, John Logan, Rose McDermott, Elias Muhanna, Dietrich Rueschemeyer, Richard Snyder, Ian Straughn, Rebecca Weitz-Shapiro, Ashu Varshney, and Alan Zuckerman. Seminar participants at the American University of Beirut, University of California–Berkeley, Brown, University of Chicago, University of Guelph, Harvard University, McGill University, Massachusetts Institute of Technology, Northwestern University, New York University, George Washington University Project on Middle East Political Science (POMEPS), Princeton University, University of Toronto, World Bank, Yale University, and elsewhere provided useful feedback on drafts and ideas at various stages. I particularly appreciate the support and insightful comments of Jorge Dominguez and the intellectual home provided by Larry Winnie and Kathleen Hoover at the Harvard Academy.

The guidance and engagement of friends and colleagues in the Middle East were invaluable in helping me to grasp the complexities of Lebanon and informed my broader intellectual development. I am especially grateful to Abdul-Rahim Abu-Husayn, Marlin Dick, Marwan Khawaja, and Fawwaz Traboulsi for their insightful comments and for facilitating the field research process. At various stages of my research, I benefited from input from As ad Abu-Khalil, Mirvat Abu-Khalil, Walid Ammar, Sami Atallah, Munir Bashshur, Hassan Charif, Shaz Faramarzi, Rita Giacaman, Charles Harb, Mona Harb, Judith Harik, Waleed Hazbun, Samer Jabbour, Ray Jureidini, Tahar Labib, Kamel Muhanna, Fadi Riachi, Bassel Salloukh, Nisreen Salti, Michelle Woodward, Huda Zurayk, and Rami Zurayk. I regret that I cannot give copies of this book to Salim Nasr and Kamal Salibi, each of whom influenced my thinking in important ways, and to Anthony Shadid, who was so generous with his time, networks in the region, and intellectual engagement. I am also grateful to the government officials, party representatives, and welfare-agency staff members who agreed to be interviewed for this book, even though some were reluctant or wary of my intentions.

I owe a huge debt of gratitude to the incredible team of researchers who worked with me on this project between 2007 and 2009, including Tamar Boladian, Salwa Maalouf, Lamia Moghnie, Jamil Oueini, Zina Sawaf, and Dalia Mikdashi. I hope that they gained as much from the experience as I learned from them. Lina Mikdashi helped me to get the lay of the land early on in the research process and became a good friend and colleague. The research assistance of Ghenwa Hayek during my initial visits to Lebanon was critical, and together we shared hilarious experiences during field research. I am also grateful to Ali Abboud, Tarek Abu-Husayn and Abdul Rahman Chamseddine, who carried out valuable research in Lebanon, and to Hamid Yassin, who collected data in Iraq for the project. It was a pleasure to work with Rabih Haber and the team at Statistics Lebanon, and I am especially grateful to Ibrahim Khoury for his personal insights and side-splitting commentary on the subject of my research. My research assistants at Brown never failed to remind me how lucky I am to work with such intellectually creative and brilliant students. The contributions and critical engagement of Nathan Einstein, Jeb Koogler, Mari Miyoshi, and Christian Sorenson were exceptional. I also appreciate Nawal Traish’s careful editorial work and the background research provided by Haydar Taygun. Funding from the Smith Richardson Foundation, U.S. Institute of Peace, Harvard Academy, and Brown University supported the data collection for this project. It has been a privilege to work with Roger Haydon and the entire editorial and production team at Cornell University Press.

Field research was all the more fun thanks to friends in Lebanon. I have shared many laughs and dinners with Marti Farha and Janmarie Muhanna—including our memorable dinner on July 11, 2006, when we proved that our predictive powers are not very reliable. I am also grateful to Marti and Janmarie for introducing me to Shaz Faramarzi, who is a dear friend and a true source of inspiration. Betty Anderson, Michaelle Browers, Waleed Hazbun, Adib Rahhal, and Michelle Woodward also made my time in Lebanon all the more enjoyable.

Families are the foundations of welfare regimes, and I am immensely grateful for my own family support system. The commitment to social justice of my parents, Sandi Cooper and John Cammett, has been a lifelong influence and ultimately helped me to find my passion in my work. I regret that my father died before I finished the book. Connie and George Manioudakis (aka Yia Yia and Papou) also played an indispensable part in the production of the book by helping to take care of the children whenever I left for extended research trips. The commitment and sense of adventure of Helena Barros Mackie makes everything seem possible. I am fortunate to have the support, love, and examples of Anni, Marci, Mena, Theo Howie and Thea Myrnie, and the rest of the extended Cammett and Manioudakis clans. It has been a joy to watch Alex, Lena, and, later, Nikos grow up while I was researching and writing this book. My love and admiration for them and for Angelo have helped me to understand the hopes and fears of families in less secure places.

A Note on Arabic Transliteration

Throughout the book, I use a modified version of the International Journal of Middle East Studies (IJMES) transliteration system guidelines while retaining conventional transliterations of Lebanese, Iraqi, and Palestinian place names and figures.

Introduction

In June 2007, Hamza Shahrour, a twenty-four-year-old Lebanese man, died of heart failure in Beirut. Hamza’s death might have been prevented had he received timely medical attention, but the hospital where his family first took him refused to admit him. In Lebanon, examples abound of low-income patients who are turned away from hospital emergency rooms because they cannot cover the costs of treatment, and the Lebanese media periodically feature stories about patients who die in ambulances outside private hospitals that have refused to accept them on financial grounds (Al-Nahar Staff 1998; Balaa 2005). In the case of Hamza Shahrour, however, it was not poverty but sectarian identity that allegedly compelled the hospital staff to refuse to treat him. A Shi i Muslim, Hamza was taken to the Rafiq Al-Hariri Hospital, which is officially public but which at various times has been controlled by the Future Movement, a predominantly Sunni Muslim political party and an important force in Lebanese politics. After his death, Hamza’s mother lamented, I wish my son had been a Sunni. Maybe he would be sitting next to me now instead of dying, having been turned away from the Hariri hospital (IRIN News 2008). This account of Hamza’s treatment suggests that the Future Movement allocates social benefits along sectarian lines. But this claim is surprising in light of the history of the Future Movement. For years, the organization was seen as relatively open to all Lebanese, regardless of sect, even though its founder was a prominent Sunni leader. Thus, the interpretation by Hamza’s mother suggests that Lebanese citizens view the organization as sectarian, despite its history of cross-sectarian generosity.

The Future Movement is hardly the only political party in Lebanon accused of discrimination along sectarian lines in recent years. Doctors from the Rasoul al- Azam Hospital, a hospital in the southern suburbs of Beirut run by the Shi i Muslim party Hezbollah, admit that Hezbollah members and their families receive priority treatment (IRIN News 2008). Although their own welfare institutions are currently far less developed than those of their Sunni and Shi a counterparts, Christian political parties use connections with religious charities and other provider organizations to ensure that their supporters receive preferential access to social services. Christian leaders with bases of regional support, such as Nayla Moawad or Suleiman Franjieh, both of whom come from important political families in North Lebanon, run welfare networks that are widely perceived by Lebanese citizens to favor their own supporters. Long-standing Christian political parties, such as the Lebanese Forces and Kataeb, whose welfare programs were largely dismantled after the end of the fifteen-year Lebanese civil war in 1990, are resuscitating and building their social service wings at present. Political leaders of these parties openly acknowledge that they must reward their supporters with services as they rebuild party institutions; We know we need to help our supporters, especially now that we are constituting ourselves into a real political party, explained one Lebanese Forces official.¹

Social welfare, then, not only concerns the ways in which people meet their basic social needs; in Lebanon and in other countries in the Middle East, South and Southeast Asia, and other regions of the Global South, where public welfare functions are underdeveloped and identity-based organizations provide social protection, it is a lens through which to study sectarian and ethnic politics. At its core, sectarianism refers to processes of constructing and maintaining the boundaries of a religious community, demarcating who belongs and who is excluded. Social welfare, too, entails processes of inclusion and exclusion, shaping both the constitution and experiences of membership in a political community. At the national level, for example, access to social services and benefits is at the heart of contests over citizenship, a status that determines one’s rights and obligations and their entailments within the polity. Similarly, who benefits from the provision of social services by sectarian organizations—whether based on formal or informal eligibility criteria—effectively constitutes membership in these groups. Through the direct provision of social services or through indirect access to benefits provided by other public and private organizations, these organizations aim to build support, consolidate their control over territory and people, and present themselves as protectors and guarantors of well-being. A focus on the relationship between provider and beneficiary exposes the kinds of linkages—material and immaterial—that sectarian organizations² construct with ordinary people, enabling identity-based groups to lock in their control over social and political life.

In light of standard expectations of sectarian politics in academic and journalistic accounts, the story of Hamza’s death told by his mother is tragic but not surprising; sectarian or ethnic groups generally favor their own—in access to social services, jobs, the distribution of patronage, or other forms of resource allocation—especially when resources are scarce and a larger, cross-cutting sense of solidarity is absent or underdeveloped (Alesina et al. 2003; Easterly and Levine 1997; Habyarimana et al. 2007; Lieberman 2003; Tsai 2007). Far more puzzling, then, are instances when sectarian organizations purposively serve people from other sects. Sunnis and Christians attest that they receive medical care, financial assistance, and even educational scholarships from Hezbollah institutions,³ and Hezbollah emphasizes that it welcomed Christian business owners and residents in neighborhoods located in the southern suburbs of Beirut, such as Shiyah and Baabda, after the civil war (Harik 2004). Likewise, Shi a Lebanese report that they benefit from similar services provided by Sunni institutions linked to political parties and religious charities.⁴ Even their harshest critics in Lebanon attest that sectarian parties make deliberate efforts to serve members of out-groups.⁵

This book takes these apparently anomalous cases of cross-sectarian welfare provision as a starting point for studying the broader phenomenon of sectarianism: How does sectarianism affect the efforts of ordinary people to meet their basic needs? This broad concern points to more targeted questions about the behavior of sectarian parties: Why do some sectarian providers distribute welfare goods broadly, even to out-group members, while others concentrate service provision within their associated communities? Similarly, why do sectarian organizations purposively cater to members of out-groups to a greater or lesser degree in different time periods or geographical areas? The book also poses a second set of questions about what it means for organizations to serve their own communities: Even if sectarian organizations primarily serve in-group members, are all treated equally or are some favored over others? Given resource limitations, sectarian organizations are compelled to distribute welfare goods unevenly among different categories of in-group members.

The logic of welfare outreach cannot be reduced to a single factor, nor are charitable and political motivations mutually exclusive, as I emphasize in the next chapter. In focusing on the political dimensions of the provision of social services, however, I argue that two key factors shape how a sectarian or ethnic party distributes welfare goods: (1) whether the party engages in a state-centric or extra-state political strategy and (2) whether it faces competition from other parties claiming to represent the same community (intrasect competition).

The first factor—the type of political strategy that parties prioritize—shapes both whether organizations target out-group members with welfare benefits and the degree to which they favor core versus passive supporters and the politically uncommitted. When parties pursue a state-centric strategy, or opt to work through formal state institutions to seek national power, they are more likely to serve members of other religious communities and to target more passive supporters and even those with no record of support for the party. When they engage in an extra-state strategy, which might include protests, riots, or even militia politics, they favor core supporters, who tend to be in-group members.

The second factor, intrasect competition, is most applicable to political systems premised on power sharing, in which political life is effectively structured around ethnic, religious, or other social identities. If a party has achieved dominance within its group, it is more likely to distribute welfare goods inclusively, perhaps even to out-group members. Conversely, when a party faces competition from other parties within its sect, it tends to focus services on in-group members and, under some conditions, to a narrow group of hard-core activists. Thus, who benefits from the welfare activities of a given sectarian party is shaped by the type of political strategy it prioritizes and whether it faces competition from in-group rivals.

Focusing on Lebanon, in this book I compare the welfare distribution strategies of Christian, Shi a Muslim, and Sunni Muslim political parties, with background comparisons to other ostensibly identity-based political groups in Iraq and India. In Lebanon, a quintessential case of a weak state⁶ in which power-sharing arrangements enshrine the political salience of religion and ethnicity, welfare is a terrain of political contestation. The distribution of welfare goods varies across different parties, however: the Sunni Muslim Future Movement has generally offered services relatively broadly, even locating some health clinics in Christian neighborhoods, whereas the Shi i Muslim Hezbollah provides services mainly in Shi a areas, although it welcomes members of other sects in its welfare institutions and has recently expressed interest in branching out beyond its core areas of operation. Christian political parties tend to focus social assistance efforts in heavily Christian communities. The degree to which these parties face serious competition within their respective sects and the types of politics they prioritize help to explain the variation in the propensities of these organizations to serve out-group communities or to reach out beyond their core base of supporters. In all cases, service provision is used not only to address pressing social needs but also to build political support. In the allocation of basic health care, educational services, food, and other forms of material assistance—the main sectors on which I focus—service providers linked to Lebanese political parties make choices about whom to reward, attract, or exclude.

Unfortunately, the discretionary allocation of welfare goods by political parties and movements across identity-based lines is hardly unique to Lebanon. Sectarianism and other forms of identity politics have witnessed resurgences in recent decades across the Global South; ostensibly ethnic or sectarian political parties and organizations are important providers of basic services in many countries in the Middle East, Asia, and Latin America.⁷ In Iraq, Palestine, India, Sri Lanka, Pakistan, Indonesia, and elsewhere in the developing world, politicized religious or ethnic organizations have highly developed social service programs, and evidence suggests that some favor in-group members in distributing or facilitating access to social welfare, or at least employ social benefits to attract support (Cockburn 2008; Flanigan 2006; Hefner 2005; Levinson 2008; Shadid 2002; Thachil 2009). When states fail to provide universal access for citizens to basic public goods and social services, welfare can become a terrain of political contestation, providing the opportunity for such organizations to establish or expand the distribution of social benefits.

Even in countries where access to power is not as contingent on ethnicity or religion and where public welfare functions are more developed, political parties and movements use the provision of social services as a means of building support. In Turkey, for example, reports indicate that the ruling Islamist political party Adalet ve Kalkınma Partisi (AKP; Justice and Development Party) used the distribution of white goods, or basic household appliances, to drum up support during the 2009 elections (ANF News Agency Staff 2008; Demir 2009; Kuwait Times Staff 2009; Milliyet Staff 2007). In Middle Eastern countries such as pre-1992 Algeria, Morocco, Egypt, and Jordan—countries not often described as divided because the ethnic or religious identities of citizens are not overtly politicized—Islamist organizations run their own social welfare programs (Clark 2004; Wickham 2002; Harrigan and El-Said 2009; Wiktorowicz 2001). Declining or underdeveloped public welfare functions provide the space—and, indeed, the need—for nonstate actors such as international or domestic nongovernmental organizations (NGOs), religious charities, and even political parties and movements to provide basic social services. In such contexts, local political organizations—whether linked to ethnoreligious parties or not—can use service provision to gain political support as much as to fulfill commitments to communal or religious principles of social justice.

The question of who benefits from nonstate service provision by sectarian parties highlights an important and under-recognized political factor in shaping public well-being and mediating access to social services. In analyzing the determinants of health, for example, the public health literature has traditionally privileged individual demographic factors such as age, education, gender, and socioeconomic status (Hummer, Rogers, and Eberstein 1998; Marmor, Barer, and Evans 1994; Wagstaff 2002). Structural influences, including broader social and political determinants of health, have only recently attracted more attention (Marmot and Wilkinson 2005). Where the state is virtually absent and political organizations allocate welfare on a discretionary basis, political considerations complicate the question of how individuals and households assure their basic social needs and gain access to welfare services. As the case of Hamza Shahrour sadly attests, sociopolitical factors can make access to social assistance a matter of life and death.

Apart from the restrictions on access to basic services that confront individual citizens and their families, the provision of social services by political organizations may also have detrimental effects on a macro level. The creation of multiple welfare networks, each linked to a different political organization, leads to fragmented welfare regimes. Although it is true that nonstate organizations are most likely to launch welfare operations where state programs are undeveloped in the first place, the emergence and consolidation of nonstate providers encourages further fragmentation of the welfare system and hinders any longer-term efforts to construct national welfare regimes. In the health sector, this is likely to produce inefficiencies in the overall public health regime and to create or exacerbate existing inequalities in access to medical care. Inefficiencies and inequalities may also arise in the educational sector when multiple nonstate organizations launch their own school systems without regard for regional disparities in the distribution of educational facilities.

Control over schooling by ethnic or sectarian parties also has important implications for socialization and the long-term prospects for fostering a commitment to a national political community. In divided societies, the stakes are particularly acute because such groups may promote distinct understandings of national history and hinder the construction of a coherent national identity in future generations (Anderson 1991; Doumato and Starrett 2006; Freedman et al. 2004; Levy 2004; Kaplan 2006; Podeh 2000; Weber 1976). In turn, the fragmentation of political identities and, hence, allegiances can undercut the provision of public goods in the long term, a self-reinforcing process (Habyarimana et al. 2007; Miguel 2004). If a feeling of solidarity based on a shared identity is critical for cooperation and joint action to pursue common goals, then the fragmentation of identities within a national territory may inhibit efforts to establish broader, more universalistic public welfare functions.

To be sure, NGOs can play a key role in providing basic public goods and empowering citizens to meet their basic needs (Brinckerhoff 1999; Brown 1998; Salamon 1995). Governments in many developing countries simply lack the capacity to provide for their own populations, whether due to a lack of material resources, administrative deficiencies, or corruption. This is especially true in post-conflict, divided societies such as Lebanon, where the experience of war has depleted public resources and undercut state institutions, including national welfare regimes, from the level that they existed prior to the outbreak of conflict. In these contexts, NGOs—including local ethnic, sectarian, and other political actors—play a key role in providing social protection. Nonstate organizations may alleviate pockets of need, but if they differentiate along religious or political lines in distributing basic services, they do not necessarily contribute to the broader public good and, at the extreme, can lead to avoidable tragedies such as the death of twenty-four-year-old Hamza in Beirut. The use of ethnic or religious identity as a criterion for access to medical care is among the most egregious violations of the Hippocratic Oath and creates new inequalities, further entrenches existing inequalities, and, at the extreme, can even strain social cohesion.

1

Welfare and Sectarianism in Plural Societies

Sectarianism has emerged with renewed vigor in the past two decades across the Middle East, sub-Saharan Africa, South and Southeast Asia, and central Europe. Ostensibly perpetrated in the name of religion, headline-grabbing acts of violence have been ascribed to an allegedly enduring Sunni-Shi‘a divide in Islam in Iraq, Pakistan, and other predominantly Muslim countries (Nasr 2007, 60–62). Recurrent tensions between Muslims and Hindus in India have led to periodic outbursts of violence, leaving many dead and injured in their wake (Brass 1997; Varshney 2002; Wilkinson 2004). Similarly, riots between Muslims and Christians have led to frequent bloodshed in Nigeria and Indonesia (International Crisis Group [ICG] 2008a; Scacco 2010). In response to the resurgence of identity-based conflict, research on ethnic and sectarian politics has burgeoned. Many studies analyze the causes of violence waged in the guise of ethnic or religious differences (Brass 1997; Horowitz 2000; Varshney 2002; Wilkinson 2004; Young 1976). Others explore the electoral behavior of ethnic parties and the impact of institutional design on the propensity for ethnic conflict (Birnir 2007; Chandra 2004; Horowitz 2000; Lijphart 1977; Posner 2004a; Reilly 2002).

This book adds a new dimension to the study of identity politics by focusing on how sectarian groups establish and reinforce their control over everyday social and political life. Where ethnic and sectarian organizations, among other nonstate actors, are key suppliers of social services, welfare provision offers a valuable lens for analyzing identity politics. The provision of public goods and social services is especially politicized when these groups are important vehicles for popular representation and mobilization (Banfield and Wilson 1963; Cammett 2011; Kitschelt 2000; Kitschelt and Wilkinson 2007). Political organizations with sectarian orientations either directly provide or broker access to social services in India, Indonesia, Iraq, Lebanon, Pakistan, and Palestine, to cite a few examples. Social welfare, then, concerns not just the ways in which people meet their basic social needs; where public welfare functions are underdeveloped and religious or ethnic organizations provide social protection, the provision of social services both constitutes and reproduces the politics of sectarianism.

The dynamics of the provision of social welfare by sectarian organizations reveals how sectarianism operates on the ground. In particular, the relationship between providers and beneficiaries exposes the multifaceted connections that these groups construct with ordinary people. Social welfare involves an obvious material exchange in which the beneficiary receives assistance to meet his or her family’s basic needs. The immaterial dimensions of the relationship are less obvious but equally, if not more, important. Providing services and meeting basic needs are acts of community-building because they signal who is a member of a protected group. The provision of social welfare also brings a sense of security and psychological comfort that is especially valuable to beneficiaries of more limited means, who, by definition, lead more precarious lives. This is even truer in polities where states fail to provide basic social safety nets.

What, then, can social welfare reveal about the ways that sectarian groups establish linkages with populations? It is generally assumed that religious or ethnic groups merely serve their own, excluding those affiliated with other communities; however, this blanket assumption should be questioned. We know little about how ethnic or religious providers decide which beneficiaries to target or even which services to offer. Why would ostensibly identity-based groups engage in public goods provision at all if they are virtually assured of the support of in-group members, as the ethnic politics literature implies (Chandra 2004; Fearon and Laitin 1996; Kasara 2007)?

Charitable concerns and visions of social justice undoubtedly compel these groups to provide social assistance, but political motivations are also important. How do sectarian organizations distribute welfare goods? Under what conditions do they reach out beyond their own communities, and when do they concentrate on in-group members? Within their own communities, why do some providers target only hard-core supporters, whereas others offer services more inclusively to marginal supporters or even nonsupporters? Why does the same organization reach out across religious lines more actively in some historical moments than in others?

In this chapter, I present an analytical framework to explain the politics of the provision of social services by sectarian organizations. I contend that two main political factors influence the allocation of social benefits. First, the types of political strategies adopted by sectarian organizations shape whether they disburse benefits in an inclusive or exclusive manner. In some contexts, sectarian organizations opt for a state-centric strategy, in which they choose to participate in the formal institutions of the state, including electoral contests. Under other conditions, however, sectarian groups opt for an extra-state strategy, in which they work outside formal state institutions and challenge state authority, often through militancy. Sectarian organizations that adopt a state-centric strategy are more likely to cultivate linkages with and, hence, distribute social benefits to members of other religious communities and to have a broader range of in-group members beyond core activists. Those that opt for a militant, extra-state strategy are more likely to funnel benefits to the most committed supporters, who tend to be in-group members. Second, in plural societies, the degree of competition for political representation of the sect affects the distribution of benefits. Competition from co-religionist organizations compels sectarian groups to prioritize in-group members. In this chapter, I elaborate the logic of these arguments in more detail, and in the remainder of the book, I apply this framework to analyze the social welfare activities of sectarian parties in Lebanon and comparable parties in Iraq and India.

CONCEPTS, DEFINITIONS, AND ASSUMPTIONS

Before I develop my core arguments, it is essential to clarify some key concepts and assumptions embedded in the central claims, including my understandings of sectarian identity and organizations, the components of social welfare, motivations for the provision of social welfare by sectarian parties, and the impact of the distribution of social benefits on citizens.

Sectarian Identity and Sectarian Political Parties

The central focus on sectarian organizations in this book and the contested meanings of identity in the literature on ethnic politics call for working definitions of these terms. Religion, sect, and other social identities can be viewed as types of ethnic identity if they are based on descent-based attributes, or inherited characteristics that seem to have little scope for voluntary adoption or rejection (Chandra 2006, 398).¹ Yet this should not imply that the categories of in-group and out-group, or co-religionist and non-co-religionist, are constant. Even in places such as Lebanon, where political representation and the rights and obligations of citizenship are structured along sectarian lines, the content and political salience of sectarian groups vary across time and space. For example, the Sunnis and Shi a were not mutually antagonistic in Lebanese political life until regional tensions heightened along ostensibly Sunni-Shi a lines in the past decade. Similarly, the same political party may deploy more or less sectarian rhetoric in different time periods or before various audiences. Thus, ascribed identity is endogenous to political institutions and the larger political context (Chandra 2001; Lieberman and Singh 2012).

In the short term, however, identities may appear rigid, particularly when they are activated, or categories in which an individual actually professes membership or to which she is assigned membership by others (Chandra 2012, 9).² In contexts where ethnicity or religion is institutionalized in politics or conflict is waged in the name of such categories, communal identity can resonate politically and take on real meaning for individuals and public life. As a result, phrases such as sectarian group and sectarian party appear throughout the book. Although they seem to imply fixed classifications, they are not expressions of immutable identities, nor should they suggest that individuals choose to be or should be reduced to these groupings. Rather, they reflect categories that are viewed as locally relevant and that structure political life in a particular time and context. Given my focus on Lebanon, where sectarian rather than ethnic identity is more politically significant, I use the term sectarian rather than ethnic, although I expect the arguments to be relevant to contexts where ostensibly ethnic rather than religious cleavages are politically salient.

My understanding of the term sectarian party takes Kanchan Chandra’s definition of an ethnic party as a starting point. In defining ethnic parties, Chandra (2004, 3) emphasizes the messages and symbols that parties employ in cultivating their ethnic credentials rather than the social identities of their constituents (Horowitz 2000). When we classify a party as ethnic, focusing exclusively on its social base can be misleading. Followers of ethnic parties are generally members of the corresponding ethnic group, but not all parties whose leadership and membership share a common ethnic identity can be categorized as ethnic. Furthermore, even when parties contain large concentrations of members from particular ethnic or sectarian groups, these identities do not necessarily motivate individuals to join or participate in them. For example, from independence until the 1980s, communist and other leftist parties had strong followings in Middle Eastern countries (Anderson 2005). In countries with significant Shi a populations, such as Iraq and Lebanon, many supporters of leftist parties were Shi a, in part because the Shi a tended to be more economically marginalized than other groups. Christians were also important in the leadership and membership of these parties. Yet it would be a mistake to characterize leftist parties as Christian or Shi a parties (or even quasi-Christian or quasi-Shi a parties) because their leadership explicitly rejected sectarian identity as an organizing principle in favor of themes related to social justice, equality, and nationalism (Jabar 2003; Traboulsi 2007).³

Furthermore, the label sectarian should not imply a fixed or uniform set of behaviors. Instead, the status of a party as sectarian should be seen as fluid because the same party can employ sectarian language and symbols to varying degrees in distinct historical moments or with different audiences and constituencies. The case of the Lebanese Future Movement, whose membership is largely composed of Sunni Muslims,⁴ illustrates this point. Sunnis have always made up the most important membership base of the party, but its leaders’ emphasis on the Sunni credentials of the organization has varied in different periods and contexts. Although some point to the efforts by the party to serve a national, cross-sectarian audience, even party operatives admit that Future Movement leaders have increasingly emphasized the sectarian identity of the party through subtle messages and appeals since the assassination of Rafiq al-Hariri, founder and former leader of the party, in 2005 (Abu-Khalil 2008; Shadid 2005a).

The changing salience of sectarian identity as a mobilizing tool for the leadership of the Future Movement and other Lebanese parties raises a related conceptual issue—the meaning of sectarianism. Drawing on constructivist