SIGNIFICANCE OF THIS ANALYSIS

Analyzing the supply lines rather than the frontlines reorients discussions of rebellion away from initial mobilization and toward the processes and relationships that sustain it over time (Petersen Reference Petersen2001). Concentrating on understudied “back-end” processes of armed rebellion clarifies the role that they play in everyday militant decision making and long-term survival, in particular because these processes strongly influence strategic choices and affect military effectiveness (Van Creveld Reference Van Creveld1977). Problematizing the diverse roles that “behind the scenes” actors play within rebel organizations challenges current scholarly understandings of participation in political violence by underscoring the simple fact that sustained, violent rebellion requires some militants to keep their hands (or records) clean. This seemingly paradoxical point provides an opportunity for scholars to move beyond the standing assumption that everyone participates for the same reasons; instead, maintaining a rebel organization requires many different types of militants, who not only mobilize along different trajectories (Viterna Reference Viterna2006) and accept varying levels of risk (Petersen Reference Petersen2001) but also serve in diverse roles. Rather than relying on imprecise terminology such as “participation” or “fighting” to describe all rebel mobilization trajectories, analyzing logistics and support networks highlights the different ways in which people apply their skills, backgrounds, and connections to their understandings of what it means to “resist,” “struggle,” or “rebel.”

Moreover, co-analyzing militant and quotidian social network structures in the context of rebel logistics provides direct insight into the social processes of civil war; that is, “the transformation of social actors, structures, norms, and practices at the local level—that sometimes leave profound social changes in their wake” (Wood Reference Wood2008, 540). Although many scholars have identified social network structure as being a fundamental component of high-risk mobilization and organized violence (Barkey and Van Rossem Reference Barkey and Van Rossem1997; Centola and Macy Reference Centola and Macy2007; Gould Reference Gould1993, Reference Gould1995, Reference Gould1996; Marwell, Oliver, and Prahl Reference Marwell, Oliver and Prahl1988; McAdam Reference McAdam1986; Petersen Reference Petersen2001; Siegel Reference Siegel2009; Wickham-Crowley Reference Wickham-Crowley1992), few researchers, as Jeff Goodwin (Reference Goodwin1997, 53) notes, have paid attention to the influence that rebels’ quotidian kinship, romantic, and friendship ties have within militant organizations.Footnote ² This perspective is even more important because militant organizations are often active, strategic agents in wartime social change; they may, for example, regulate dating and marriage, restructure health care, or remake land tenure agreements (Mampilly Reference Mampilly2011; Peteet Reference Peteet1991; Pool Reference Pool2001; Wood Reference Wood2003).

Studying clandestine supply networks also highlights the frequently gendered nature of high-risk mobilization and collective violence and opens new avenues for inquiry into this phenomenon. Women have been centrally involved in logistics and support work in, for example, Irish (Alison Reference Alison2004a; Reference Alison2004b; Ryan Reference Ryan1999), Algerian (Amrane-Minne Reference Amrane-Minne1992, Reference Amrane-Minne1999, Reference Amrane-Minne2004), Salvadoran (Viterna Reference Viterna2006; Wood Reference Wood2003), Eritrean (Bernal Reference Bernal2001; Pool Reference Pool2001, chap. 3; Wilson Reference Wilson1991), Zimbabwean (Lyons Reference Lyons2004), Sri Lankan (Alison Reference Alison2003), and Sierra Leonean (Coulter Reference Coulter2009, 135–36) rebel organizations, as well as serving as ground troops. Evidence indicates that the same pattern may hold for recent uprisings in Libya and Syria (“Heroes of the Tripoli Underground” 2011; Holmes Reference Holmes2012; “The Country Formerly Known as Syria” 2013). Female militants are disproportionately represented in logistics and support work relative to their overall participation in militant organizations; for example, only 3.1% of the roughly 10,949 Algerian women who fought the French served as ground troops (2,000 women joined the maquis, or rural guerrilla rebellion, with many serving as nurses), whereas approximately two-thirds of women in the Algerian civil resistance “were either in charge of refuges or supplies” (Amrane-Minne Reference Amrane-Minne1999, 62–65). In 1979, 30% of the Eritrean People's Liberation Front's (EPLF) members were female, but women represented 35% of the administrative staff, 55% of the health care staff, and only 13% of ground troops (this number officially rose to 23% by 1989) (Pool Reference Pool2001, 95–96). Louise Ryan argues that Irish women's “invisibility”—due to their participation specifically in covert operations and clandestine logistical work during the 1916 uprising—“may have contributed to their omission from accounts of the war” (Ryan Reference Ryan1999, 264); her observation suggests that women are most likely to participate in the kind of revolutionary work that is generally hard to observe from outside and less likely to be recorded due to its secretive nature. This tendency contributes both to a systematic underestimation of logistic sub-units’ importance and a biased representation of female rebels.

RESEARCH DESIGN

Studying the emergence of Palestinian clandestine supply networks in 1980s Lebanon provides a unique opportunity to evaluate the complex mechanisms through which rebels create, join, and participate in logistics units. Analyzing this single, multifaceted case allows me to exploit three factors: the preexisting overlap between Palestinian militant groups and quotidian kinship, marriage, and friendship ties (Peteet Reference Peteet1991; R. Sayigh Reference Sayigh1979, Reference Sayigh1998); the similar institutionalization of Palestinian communities across Lebanon before 1982 (Rubenberg Reference Rubenberg1983; Team International Engineering and Management Consultants 1981); and variation in the wartime social context across time (see Fig. 1). Specifically, because the Israeli occupation effectively partitioned Beirut from Saida and Sour between 1982 and 1985, and due to local distinctions in violent tactics between 1985–88, I am able to nest regional substudies within the broader case.Footnote ³ Leveraging sources of variation within a single case allows me both to parse how local contexts produced different network interactions across similar communities and to endogenize the effects of regionally specific wartime violence on organizational development (George and Bennett Reference George and Bennett2005). Moreover, this approach controls for such factors as groups’ ideologies, ethnic makeup, social bases, and training regimes. By examining a single case, I thus privilege the capture of micro- and meso-level dynamics in order to theoretically inform future research, rather than seeking to test hypotheses via a cross-national design (George and Bennett Reference George and Bennett2005, 19–22; Gerring Reference Gerring2004).

FIGURE 1.

An ethno-historical approach to studying militant organizations has several advantages in comparison to other methodologies. In contrast to survey-based research, such as, for example, Macartan Humphreys and Jeremy Weinstein's (Reference Humphreys and Weinstein2006) study of militants in Sierra Leone, ethnographic methods allow researchers to progressively build trust with their informants—thereby gaining access to insider perspectives, experiences, and meaning-making practices rather than taking for granted external categories of participation or behavior (Bayard de Volo and Schatz Reference Bayard de Volo and Schatz2004; Gerring Reference Gerring2001; Pachirat Reference Pachirat and Schatz2009, 143–144; Wedeen Reference Wedeen and Schatz2009, Reference Wedeen2010). Trust and analytic insight are crucial to accessing and situating restricted information that is unlikely to be gathered in a one-shot interaction and that is not captured in documents or other accessible historical material. Second, the “present-day” relationships and knowledge that ethnographers develop allow them to probe interlocutors’ past experiences for points of continuity and rupture. Immersion in an organization permits researchers to observe its members, spaces, and everyday practices and to probe the ways in which that organization and the roles within it change over time (Yanow Reference Yanow2012, 3). Ethnographic approaches are thus particularly suited to capturing the structures, dynamics, and practices of organizational divisions that were never “put to paper.” Third, ethnography gives researchers the context necessary to identify and analyze what Lee Ann Fujii terms “meta-data”—the “informants’ spoken and unspoken thoughts and feelings which they do not always articulate in their stories or interview responses, but which emerge in other ways”—rather than taking interlocutors’ responses at face value or as given “truth” (Fujii Reference Fujii2010, 231). When dealing with potentially sensitive topics, it thus has a marked advantage over work based solely on formal interviews. Fourth, an ethnographic perspective, particularly when informed by social network theory, avoids making the individualist assumptions that have dominated recent political science scholarship (Wedeen Reference Wedeen and Schatz2009, 81; Reference Wedeen2010, 260). Rather than adopting the premise of maximizing actors driven by known, ordered preferences and “divorced, for the sake of general propositions, from actual historical processes” (Wedeen Reference Wedeen2010, 260), these perspectives treat actors as inherently socially embedded (Granovetter Reference Granovetter1985; Wedeen Reference Wedeen2002) and allow researchers to problematize the relationship between individual agency, social context, and organizational structure.

This article is based on 19 months of fieldwork carried out in the cities of Beirut, Saida, and Sour between 2007 and 2012.Footnote ⁴ I conducted interviews with 106 current and former members of Palestinian militant organizations, social and humanitarian aid workers, and long-time United Nations Relief and Works Agency for Palestine Refugees in the Near East (UNRWA) employees.Footnote ⁵ Of these interviews, 24 were oral histories with current or former militants who had personal involvement with clandestine operations during the 1980s; however, many other interviewees discussed underground activities. Additionally, my participant observation among current and former members of militant organizations included thousands of informal conversations and interactions that I recorded in my field notes. To account for local and national-level processes, including regional variations in the belligerent parties and the violent tactics used between 1982 and 1988, I interviewed people who worked locally in Beirut, Saida, and Sour, in addition to those who worked across regions. This approach allowed me to separately examine the nascent communications and supply networks under Lebanese government rule in West Beirut from 1982–84 and the Israeli occupation of Saida and Sour from 1982–85, and then to evaluate these separate systems’ consolidation into a national supply and communications apparatus during the War of the Camps (1985–88).

The reconstruction of this network, which has been only peripherally mentioned in existing studies, illustrates the value of ethnographic methods in sensitive research. My repeated interactions with interlocutors yielded longitudinal, egocentric network data for key clandestine actors who operated in each camp, each region, and across regions. Interested in the range of positions available, the practices that accompanied them, and the reasons that people remembered for participating in them, I catalogued each interviewee's organizational and social roles over time and the activities of any close family and friends. Rather than treating rebellion as a dichotomous variable by listing people as “fighters” or “nonfighters,” I sought to tease out the ways in which people vacillated between fighting and logistics jobs over time and how clandestine organizations evolved around and shaped their members’ careers. Combined with relevant historical texts (such as official documents, battlefront video footage, and party publications), fighters’ and aid workers’ memoirs, surveys (published by political factions, aid organizations, and the United Nations), and Arabic newspaper reports, this data allowed me to evaluate interactions between formal organizational hierarchies and quotidian systems of affiliation. Secondary case and theoretical literature provided further perspective. Taken together with sustained ethnographic work, these data reveal a rich picture of mass mobilization, organizational evolution, and social change.

THEORIZING CLANDESTINE MOBILIZATION

Why and how do individuals join clandestine logistics networks? Participation in underground political activity under occupation or during civil war is not only profoundly risky but it also often lacks the public status rewards that are associated with high-risk collective action. Former participants in clandestine operations frequently noted that most of their family and friends were unaware of their activities. Some interviewees hid their activities for safety reasons—Palestinian activists’ family members were frequently arrested or abused if they could not be found (R. Sayigh Reference Sayigh1979, 151).Footnote ⁶ Others worried that their parents would try to stop them. In contrast, individuals who openly participated in military operations were publicly revered as heroes or martyrs. For example, many interlocutors remembered “Dalal's operation” in 1978; women in particular saw the female fighter as an inspiration.Footnote ⁷ This evidence suggests that more nuanced factors than status rewards and public pressure played a large role in mobilizing militants into logistical and support roles.

Roger Gould argues that formal organizations are critical to expanding the scale at which group identities are considered salient, but that “their impact on social interaction tends to go unnoticed” (Gould Reference Gould1995, 21–22). That is to say, membership in formal organizations feeds back into individuals’ relationships, shaping their social worlds. In this world, A meets B outside of her daily social interactions (e.g., family, friends, neighbors) because of the ideological and organizational ties that place them in each other's spheres for the first time. B becomes part of A's regular circle; perhaps they get to know each other's families and friends, establishing an entirely new set of social network ties. Thus “an unintended consequence of formal organizations, in other words, is the creation of social ties that encourage the recognition of commonalities on a scale considerably broader than would be expected on the basis of social networks alone” (Gould Reference Gould1995, 22). However, Gould never notes how diverse affiliations within families or friend groups may change the ways in which relatives or friends engage politically within the context of external organizations, nor how these relationships may have larger consequences for the organizations in which they are members.

In a similar vein, Mark Granovetter argues that individuals bound by strong ties should be part of overlapping networks; that is, individuals who have long-term, intense, intimate, reciprocal relationships should have similar relationships with each others’ friends (Granovetter Reference Granovetter1973, 1361). Thus, if A and B meet through an organization and form a friendship, they should eventually start becoming friends with each other's friends as well. This claim is founded on the concept of structural balance: The central idea is that “friends” evaluate others in similar ways (i.e., they both “like” or “dislike” the same people). Thus, if two people, A and B, have positive feelings about each other, and person B has positive feelings about person C, then person A will also know and have positive feelings about person C (Wasserman and Faust Reference Wasserman and Faust1994, 221). However, because these networks tend to produce extensive overlap in relationships, weaker ties—thought of as “acquaintances”—are necessary to diffuse information across larger networks of individuals (Centola and Macy Reference Centola and Macy2007; Granovetter Reference Granovetter1973, 1365–1366). Otherwise, all information would remain within overlapping cliques bound by strong ties, unable to jump between tight-knit groups. In line with this theory, Granovetter contends that, “except under unlikely conditions, no strong tie is a bridge” (emphasis in original) (Granovetter Reference Granovetter1973, 1364).

I argue that clandestine supply networks emerge from formal militant hierarchies because strong, quotidian ties serve bridging functions. The configuration of the overlap between rebel organizational structures and militants’ quotidian relationships—rather than the replication of ties across these networks alone—produces these brokering positions. For example, A and B, the two close female friends, may work in a particular organizational sub-unit, whereas C—A's husband—works in another division. Certain organizational structures may thus allow for strong ties—in particular kinship, marriage, and community ties—to act as bridges between organizational subdivisions. Particularly in socially segregated societies, Granovetter's bridging rule may consequently be violated. For example, in a more gender-stratified society, B may not be acquainted with C because she does not socialize with married men; thus, A is a bridge between B and C. When this triad is present within a rebel organization, the bridge between A and C allows for the secure transmission of information, money, military orders, or even weapons across an organizational divide, reaching B and others who share unit membership with A. Strong ties in bridging positions allow for trust between otherwise isolated cliques of individuals, a critical building block of underground networks.

Quotidian relational bridges may become active as violent conflict progresses, rather than being preemptively designated for organizational tasks. In other words, when A and C are initially recruited, the tie between them may only represent a latent source of structural power, rather than a deliberate effort to provide a communication redundancy. As a militant organization seeks to adapt to new circumstances or to manage crises, bridges such as the one between A and C offer means through which to resolve information or supply problems by bypassing inflexible organizational hierarchies (Burt Reference Burt2005). The activation of A and C's quotidian relationship as an organizational asset introduces a new mode of participation for A and C, altering their organizational roles. In other words, these informal, latent social structures present alternate pathways of mobilization within rebel organizations. Yet the changing nature of their organizational relationship will also feed back into their “private” relationship, driving social change (Padgett and McLean Reference Padgett and McLean2006; Padgett and Powell Reference Padgett and Powell2012).

ORGANIZATIONAL FRAGMENTATION AFTER THE 1982 INVASION

On June 6, 1982, the PLO's organizational apparatus, headquartered in Beirut since its 1971 expulsion from Jordan, experienced a massive shock when the Israeli Defense Forces (IDF) invaded Lebanon. Prompted by the fringe group Fateh-Revolutionary Command's June 3, 1982, attack on the Israeli ambassador to the United Kingdom, the IDF bombed PLO offices in Beirut. In retaliation, the PLO launched missiles from South Lebanon toward civilian settlements in northern Israel, triggering a long-planned invasion by the IDF aimed at destroying the PLO (Y. Sayigh Reference Sayigh1997, 505–21; Schiff and Ya'ari Reference Schiff and Ya'ari1984, chaps. 1–6). The Israeli government had initially anticipated clearing a limited border zone to prevent further PLO missile attacks and guerrilla infiltrations, but instead directed the IDF to continue north to Beirut as the Israeli Air Force bombarded it (Schiff and Ya'ari Reference Schiff and Ya'ari1984, chaps. 7–11). Throughout the summer, Beirut remained under siege as the PLO and leftist Lebanese militias battled the IDF in the southern suburbs. Following negotiations, the PLO's international bureaucracy and guerrilla fighters from multiple parties (14,398 personnel and soldiers total) evacuated Lebanon at the end of August (Y. Sayigh Reference Sayigh1997, 537). International peacekeeping troops—dubbed the multinational forces (MNF)–deployed in Beirut and its southern suburbs to monitor the ceasefire, evacuation, and IDF withdrawal.

In the fall of 1982, hundreds of thousands of Palestinian civilians and low-level party cadres who remained in Lebanon were still reeling from the invasion, which destroyed Palestinian refugee camps in South Lebanon, left many Beirut neighborhoods in ruin, and fragmented Palestinian resistance organizations.Footnote ⁸ The IDF occupied South Lebanon, including refugee camps in Saida and Sour; more than ten thousand Palestinian and Lebanese men and many women from this region between the ages of 15 and 60 were interned at facilities such as the Ansar prison camp outside Nabatiyya or, if they were deemed high status, were sent to detention facilities in Israel (Abourahme Reference Abourahme2010; Denkner Reference Denkner1983; Giannou Reference Giannou1982; Khalidi Reference Khalidi1985; Khalili Reference Khalili and Lentin2008; Schiff and Ya'ari Reference Schiff and Ya'ari1984, chap. 8; Yermiya Reference Yermiya1984). Men from South Lebanon who were not immediately arrested fled to the Shouf Mountains, to the Beqa'a Valley, or to Trablous. This series of events deprived most Palestinian families in South Lebanon of their male household members, leaving them both homeless and lacking income to rebuild.

Assassinations and denunciations defined the post-evacuation environment in Saida, making political organizing risky. By October 1982, there were clear signs of targeted killings by Christian militias in ‘Ayn al-Hilwa camp.Footnote ⁹ The IDF also created a militia dubbed the “Palestinian National Guard” that policed the camps in South Lebanon and denounced political activists to the Israelis. Assassination and attacks on political figures in ‘Ayn al-Hilwa and Saida became increasingly common.Footnote ¹⁰ Yousef, who alternately fought with Fateh, the PFLP, and the Democratic Front for the Liberation of Palestine (DFLP), noted that “the Lebanese Forces [in Saida] killed a lot of people, they used to kidnap people, would later throw him dead on the road. . . . It was a way to show they could do it.”Footnote ¹¹ Almost everyone whom I asked about Saida during this time could remember people they knew being assassinated; one former fighter summarized, “There was a lot of violence because of revenge.”Footnote ¹² Mahmud, a former guerrilla fighter from Sour, illustrated the pervasiveness of collaboration by asking, “Do you know about those manāyīk [fuckers] in the black hoods on the beach, the ones who told the Israelis who were fidā’īyīn [guerrillas]? Those guys were traitors, and they became the leadership in Ansar.”Footnote ¹³ His question also elucidates why people became suspicious of released prisoners, wondering if they had supplied information to the Israelis that caused them to be freed when others were not. The Israelis’ and the Christian militias’ extensive recruitment of collaborators created an environment of deep suspicion and mistrust.

In Beirut, in contrast, the Lebanese government and Christian militias quickly deployed large-scale disarmament campaigns against the Palestinian population, detaining hundreds of Palestinian men throughout the autumn of 1982 (Picard Reference Picard2002, 127; Y. Sayigh Reference Sayigh1997, 555; Schiff and Ya'ari Reference Schiff and Ya'ari1984, chap. 13); the risk of physical harm confined many Palestinian men to their homes and immediate neighborhoods.Footnote ¹⁴ The September 1982 massacre of thousands of civilians at the Sabra and Shatila camps in South Beirut following President-elect Bachir Gemayel's assassination only reinforced feelings of insecurity (Ang Reference Ang1989; Fisk Reference Fisk2002, chap. 11; Picard Reference Picard2002, 126; Y. Sayigh Reference Sayigh1997, 539).Footnote ¹⁵ After the massacre, Palestinians living in other Beirut neighborhoods feared that the violence would spread. Aisha, a Fateh cadre from Burj al-Barajna, recalled that people came to the camp saying that a massacre was happening, that Israel was killing people, and that the IDF would come into Burj al-Barajna next.Footnote ¹⁶ For Palestinian refugees remaining in Beirut, the events of autumn 1982 echoed the fall 1976 massacres at Tel al-Za'atar and Jisr al-Pasha refugee camps in East Beirut and convinced them that they had to find ways to protect the camps.Footnote ¹⁷

A FAMILY AFFAIR: WOMEN AS ORGANIZATIONAL BROKERS

Given the fragmentation that Palestinian organizations experienced and the environments of deep insecurity in Beirut and South Lebanon, how did Palestinian rebel organizations rebuild? The emergence of clandestine logistics networks reflected a fundamental change in the ways that the PLO and guerrilla organizations operated. Before the 1982 invasion, quotidian social connections played a central role in mobilization (Peteet Reference Peteet1991, chap. 4; R. Sayigh Reference Sayigh1979, chap. 4, 1998); when asked how they were recruited, Fateh members, in particular, tended to note family connections, whereas PFLP and DFLP members more often cited friendships. However, quotidian ties were not systematically used for critical “formal business”; recruits’ assignment into different subdivisions depended on many factors, including interest, skill, perceived social appropriateness, and need. Although some women participated in combat units, they were largely placed in administrative offices, social services (e.g., kindergartens, literacy programs, health services), or support units such as information divisions (Kawar Reference Kawar1996, 70–73; Peteet Reference Peteet1991, chap. 5).Footnote ¹⁸ Post-invasion, this underlying latticework of quotidian affiliations provided the scaffolding onto which militants informally remapped military units’ communications, finance, intelligence, and other operations, thereby changing the content of militants’ quotidian relationships. This rearrangement allowed the PLO and several of the guerrilla organizations to continue operations after extreme disruption. But under what conditions do individuals enter these specific, task-differentiated networks, and how do their roles evolve? The following subsections address this question in detail, specifically focusing on regional differences in mobilization related to the nature of wartime violence and repression.

THE WAR OF THE CAMPS: ROUTINIZING CLANDESTINE ACTIVITY AND INVOKING INTERORGANIZATIONAL TIES

Fateh and the PLO's expanding presence in Lebanon threatened a volatile combination of Lebanese militias, the Syrian military, and Palestinian militias that were part of the Syrian-allied, anti-Arafat (and thus anti-PLO and anti-Fateh) opposition Palestinian National Salvation Front (PNSF).Footnote ⁴² The Syrian government saw Fateh's resurgence as an attempt to pull Lebanon back into war, whereas its ally, the Lebanese Shi'a militia Amal, worried about renewed Palestinian-Israeli hostilities in South Lebanon again taking their toll on civilians (Y. Sayigh Reference Sayigh1997, 582). With Syrian support, Amal attacked Sabra and Shatila on May 19, 1985, starting the three-year War of the Camps. Throughout the following years, Shatila, Burj al-Barajna, and, later, Rashidiyya sustained heavy mortar and rocket shelling and infantry attacks, and they withstood sieges lasting up to six months (Giannou Reference Giannou1990, 248–49). During sieges, Amal, bolstered by the Lebanese Armed Forces’ predominantly Shi'a Sixth Brigade, prohibited the entry of food and medical supplies, forcing thousands to the brink of starvation (Ang Reference Ang1989; Giannou Reference Giannou1990). Even when Amal periodically lifted the siege, the militia and its allies prohibited Palestinians from bringing fuel, batteries, building materials, or weaponry into the camps. Starting in 1986, battles between Palestinian guerrillas and Amal took place in the hills surrounding Magdousha above ‘Ayn al-Hilwa, a camp on the outskirts of Saida.

Clandestine networks in Beirut, Saida, and Sour evolved to encompass a broader, more sophisticated set of tasks during the War of the Camps. In response to Amal's attacks on the Beirut camps, the PLO and Fateh headquartered their infiltration and financial operations in Saida. This move created two tiers of smugglers within the clandestine apparatus: women who worked between Saida and either Sour or Beirut (and thus had to cross from relatively safe areas into combat zones) and women who worked locally within Beirut and, later, Sour (who had to cross from besieged camps into territory controlled by Amal). Within these zones, militants’ roles frequently changed. For example, Zahra, the Fateh guerrilla who worked in a Shatila kindergarten, smuggled Fateh salaries into the camp during the War of the Camps. However, her affiliation was public knowledge; in 1986 she was denounced by members of a PNSF organization and imprisoned by Syrian intelligence. On her release, Syrian-allied groups set up a checkpoint under her apartment in West Beirut, forcing her to sleep outside her family home and to curtail her activities substantially. Rather than return to clandestine work, Zahra then began working in a humanitarian aid division of the GUPW, providing food and clothing to civilians.

From the beginning of the War of the Camps in May 1985, violence against the Beirut camps assumed an explicitly collective character. As such, it recalled recent campaigns against Palestinians in Beirut and reactivated the community ties that undergirded emergent militias in the region. The decentralized local militias that female couriers linked from 1982–85 evolved into shared military fronts in Shatila and Burj al-Barajna; Chris Giannou notes that, in Shatila, “Each political organization had an allotted area of the battle-front periphery as its military responsibility”(Giannou Reference Giannou1990, 33).Footnote ⁴³ Because of the restrictions on men's movement, women became responsible for transferring all supplies into the two camps. Aisha noted the difference between women's and men's roles in the defense of the camps during this period, emphasizing that, because of the stationary fronts, smuggling and logistics roles became combat positions: “Women died going to get food and water. . .women came face to face with Amal and the Syrians. They were searched, threatened, beaten. . .they were raped and killed while the men hid and shot at [Amal].”Footnote ⁴⁴ Rawan, also a former member of the logistics network, noted that she had to move under fire through the camp to deliver food, ammunition, and sandbags to the fronts while male fighters hunkered in fortified positions.Footnote ⁴⁵

Similar practices evolved in Sour during the siege of 1986. In spite of the PLO-PNSF rivalry, Tala pointed out that women in her politically divided family still smuggled money and food from Saida to Sour, using kinship networks in multiple militant organizations as cover for their travel and political contacts. When Amal besieged Rashidiyya in 1986, people in Burj al-Shemali organized secret nautical supply routes into the camp using fishing dinghies and small military boats.Footnote ⁴⁶ Tala, her husband, and her brothers-in-law described how increasing attacks on Rashidiyya and al-Buss, and thus the need for relief missions, inspired the creation of “[military] operations rooms” in the southern camps and the construction of underground emergency rooms to treat those injured during smuggling assignments.Footnote ⁴⁷ Clandestine operations thus acquired geographic footprints within the camps that were not under siege, representing some routinization and institutionalization of their activities.

The GUPW played an increasingly central role in coordinating clandestine logistics and support activities (Giannou Reference Giannou1990, 42–49; Khalili Reference Khalili2007, chap. 7): It was the hub through which female money smugglers connected to families in the camps and, in some cases, to the male military leadership. Yousef noted that by 1985 or 1986 “there was a woman's office that managed all women's activities,” including both smuggling and other support work. By the Battle of Magdousha in 1986, the Women's Union in ‘Ayn al-Hilwa was also helping coordinate support and medical services for fighters. Munadileh noted that serving in these functions “brought women together. There were people from all of the organizations together. It [the GUPW] shared the revolutionary work with the men.”Footnote ⁴⁸ As the Lebanese and Syrians became aware of this system, the PLO also began using other organizations such as the General Union of Palestinian Students and various humanitarian associations—frequently staffed by former female smugglers—to funnel money into Lebanon.Footnote ⁴⁹