Abstract
Suicide attacks continue to plague a multitude of conflict zones. However, the scholarly literature on the phenomenon has yet to produce a theory that explains why militant organizations at different stages of development, facing dissimilar enemies, and situated in unique conflict environments adopt suicide attacks. Moreover, the suicide-attack phenomenon now presents an intriguing puzzle. While most militant organizations fail to achieve their core political ends or ‘outcome goals’, organizations that employ suicide attacks are even less likely to succeed. Still, organizations have adopted suicide attacks at increasing rates. Given their ineffectiveness in precipitating outcome-goal success, why do organizations continue to adopt suicide attacks? Like all organizations, militant organizations share two common aims: (1) to survive and (2) to achieve outcome goals. As martyrdom operations often fulfill the ideological or cultural expectations of certain identity groups, representative organizations may adopt suicide attacks to expand constituent support or enhance status within a particular political landscape. Further, organizations can conduct suicide attacks to signal ideological solidarity with fellow militant organizations. This article analyzes an original large-N dataset of militant organizations alongside an original database of over 5,000 suicide attacks, and evaluates numerous case examples. The results demonstrate that militant organizations across regions and over time have adopted suicide attacks in order to gain supporters, promote organizational longevity, and boost or preserve status.
On 28 May 1985, the Palestinian National Salvation Front (PNSF) deployed four young Palestinian women aged 17–19 to carry out ‘martyrdom operations’/suicide attacks 1 against Amal forces in the Shatila area of Beirut, Lebanon. 2 Detonating explosive belts strapped to their bodies, the bombers killed numerous Amal militiamen. Earlier that month, the ‘War of the Camps’ had erupted as a sidebar within Lebanon’s ongoing civil war. At the center of the fight, the Shi’a-Muslim organization Amal, which sought control over large sections of Beirut, laid siege to areas inhabited by Palestinians loyal to Fatah leader Yasser Arafat. The umbrella group PNSF – comprising the Popular Front for the Liberation of Palestine (PFLP), Popular Front for the Liberation of Palestine-General Command, as-Sa’iqa, and others – had emerged months earlier as a coalition of anti-Arafat Palestinian organizations. Since its inception, the PNSF had kept a working relationship with Amal (Sayigh, 1997). Thus, the PNSF’s intervention on behalf of besieged rival Palestinians doubly surprised Amal. Artillery attacks made up most of the PNSF’s intervening support for its rival. But, once Amal boxed in pro-Arafat Palestinian fighters, the PNSF dispatched suicide bombers. While bringing tactical advantages against unsuspecting Amal militiamen, the PNSF’s four suicide bombers sent a clear message aimed at Lebanon’s Palestinian population: the PNSF ‘first and foremost represents all Palestinians and will come to the aid of even rival Palestinian fighters’. 3 Importantly, the PNSF’s move exemplifies a prime incentive that often drives the organizational adoption of suicide attacks – the opportunity to rally old and new supporters alike and improve the employing organization’s political status among its larger identity group.
From October 1980 to September 2015, 123 militant organizations carried out a total of 5,305 suicide attacks, killing upwards of 40,000 people. 4 Making the modus operandi a central component of contemporary political violence, militant organizations have adopted suicide attacks at higher rates with each passing decade. In attempts to explain the tactic’s diffusion, some claim suicide attacks work as a useful instrument of coercion and political bargaining (Pape, 2005; Kydd & Walter, 2006; Pape & Feldman, 2010). In contrast, others emphasize the tactic’s ineffectiveness in helping organizations achieve their core political ends or ‘outcome goals’ (Abrahms, 2006, 2010, 2012; Moghadam, 2006; Cronin, 2009; Acosta, 2014). Ultimately, key studies have produced contradictory or seemingly unrelated empirical findings, leaving the study of the suicide-attack phenomenon devoid of a general explanatory paradigm. In this article, I put forth and test a theory that explains the use of suicide attacks as a source of militant longevity and sociopolitical relevance.
Analyzing an original database containing the universe of suicide attacks alongside an original dataset that tracks the connections of 310 militant organizations between the years of 1980 and 2013, I find evidence that organizations adopt suicide attacks to strengthen their support structures, extend their life spans, and boost or preserve status. Two main points frame the theory. First, suicide attacks directly enhance the likelihood of organizational survival by contributing to mobilizing or maintaining constituents and gaining external supporters. Second, among some types of militants, adopting suicide attacks precipitates the establishment of interorganizational connections or ‘network ties’ between militant organizations. Network ties provide access to new resource stocks and can expand an organization’s constituency, 5 and as a result further increase survivability.
While carrying out martyrdom operations usually represents an end-unto-itself for individual suicide bombers (Berko & Erez, 2005; Hafez, 2006, 2007a; Lankford, 2013), for militant organizations, suicide attacks have largely become a means to securing status. Organizations regularly adopt suicide attacks to appease constituent populations and appeal to a broader network of militant organizations. Connections to ‘high-status’ organizations may heighten the popularity of militant organizations, reinforcing base support and generating new supporters and resources. In order to establish specific network ties, many militant organizations attempt to mimic high-status actors – a process that has gradually involved using suicide attacks. Over the years, conducting suicide attacks has brought certain organizations high status in the form of numerous interorganizational ties and augmented constituent support, including Hezbollah, Hamas, al-Qaeda, al-Tawhid wal-Jihad, and Tehrik-e Taliban, among others. In turn, with high-status organizations gaining popularity and capability from employing suicide attacks, ‘lower-status’ organizations have increasingly adopted the modus operandi.
This article continues with four sections. The first section reviews the literature and descriptive statistics, presents the theory, and formulates hypotheses concerning the relationship between suicide attacks and organizational longevity/sociopolitical relevance. The second section details the research design and evaluates empirical tests. The third section discusses explanations and the final section concludes.
Revisiting the organizational adoption of suicide attacks
Warranting a revisiting of the phenomenon, significant discrepancies exist between the literature on suicide attacks and newly collected data. Descriptive statistics reveal that various targeting patterns have emerged over the three and a half decades of the contemporary suicide-attack phenomenon (see Table I). 6 In the early 1980s, militant organizations launched suicide attacks mostly against foreign combatants. In the late 1980s, organizations carried out suicide attacks chiefly against domestic combatants. During the 1990s, the primary targeting profile shifted to domestic civilians, and since the mid-2000s the main targeting profile has reverted back to domestic combatants. As the primary target type of suicide attacks has shifted multiple times, it has rendered any generalizable targeting profile unsupported empirically. A single targeting profile – in the sense that Pape (2005) and Berman & Laitin (2005) assert – does not exist. 7 Whereas Pape argues that organizations adopt suicide attacks because they assist in expelling foreign occupiers and reaching goals of national self-determination, an overwhelming majority of suicide attacks have struck domestic targets (see Table II). 8 Berman & Laitin posit a theory based on the argument that nearly all suicide attacks target ‘victims whose religion was different from the attackers’ (2005: 9). However, internal conflict in the Islamic world has characterized the suicide-attack phenomenon over the last decade (see Figure 1), and most suicide attacks have involved an Islamist organization directly striking a Muslim target. Shi’a Muslims in Iraq and Pakistan, Kurdish paramilitary forces in Syria and Iraq, and domestic security forces in Afghanistan, Algeria, Egypt, Iraq, Libya, Mali, Pakistan, Somalia, Syria, Yemen, and elsewhere make for the most frequent target types. New data evince that changes in the employers of suicide attacks have resulted in concomitant shifts in the general profile of targets.
Different general targeting profiles
Location type of employers and targets of suicide attacks (1980–September 2015)
Muslim and non-Muslim targets of suicide attacks
Accordingly, any general explanatory paradigm will necessarily focus on the perpetrators of suicide attacks and explain why diverse organizations conduct suicide attacks against dissimilar target types, yet in the pursuit of similar organizational aims. 9 The dynamics of political violence and the disadvantages militant organizations encounter in asymmetric warfare provide a window into the decisionmaking process that leads different types of organizations – in unique settings – to adopt suicide attacks. Turning to general organizational aims offers a way forward in understanding why a diverse set of militant organizations adopt suicide attacks.
Common militant aims and the ‘fashionability’ of suicide attacks
The suicide-attack phenomenon presents an interesting puzzle. While most militant organizations do not successfully achieve their outcome goals (DeRouen & Sobek, 2004; Cronin, 2009), 10 organizations that employ suicide attacks are even less likely to succeed (Abrahms, 2010; Acosta, 2014). Still, over the years, organizations continue to adopt the modus operandi.
Although some earlier studies on suicide attacks claimed that the tactic’s ‘success’ in coercing targets ‘caused’ its diffusion (Pape, 2005; Kydd & Walter, 2006), others have highlighted the inability of suicide attacks to coerce targeted populations in any politically meaningful way (Abrahms, 2006, 2010; Moghadam, 2006). Moreover, recent large-N empirical analyses outright debunk Pape’s thesis (Horowitz, 2010; Abrahms, 2012; Acosta & Childs, 2013) 11 and show definitively that suicide attacks embolden target resolve and consequently diminish the likelihood of employing organizations achieving their outcome goals (Acosta, 2014).
Indeed, organizations that conduct suicide attacks typically fail in achieving their political ends. Despite achieving some short-term victories (e.g. Hezbollah’s expulsion of multinational forces from Lebanon or al-Qaeda’s attacks on 11 September 2001 [9/11]), suicide attacks have brought employing organizations few outcome-goal successes.
12
While Hezbollah, Hamas, and the Islamic State have achieved limited success in gaining control over parts of coveted territory, none of the three have set up an Islamic theocracy over all of their respective desired territories. Further, as Acosta (2014: 141) notes, organizations known distinctly for their use of suicide attacks have a noticeably dismal record of achieving outcome goals. Despite its ability to adapt and survive, al-Qaeda has not established anything reminiscent of the Caliphate. Partiya Karkerên Kurdistan has yet to exact a Kurdish homeland from Turkey. Imarate Kavkaz and precursor organizations Special Purpose Islamic Regiment and Riyadh as-Saliheyn Martyrs’ Brigade have failed to coerce Russia into withdrawing from its Caucasian territories. Boko Haram remains far from turning Nigeria into an Islamist state, and Tehrik-e Taliban has more than fallen short in its struggle to overthrow the secular Pakistani government.
Comparably, Sri Lanka eliminated the Liberation Tigers of Tamil Eelam (LTTE), ending its long and bloody military effort, which it waged mainly with suicide attacks. Jaish-e Mohammad and Lashkar-e Taiba stand today as far from breaking off Kashmir from India as they did 20 years ago. Nevertheless, over the decades, as it has become more evident that using the tactic decreases the likelihood of outcome-goal achievement, organizations have adopted suicide attacks at higher rates (see Figure 2). 13
Continued adoption of suicide attacks
Given the ineffectiveness of suicide attacks to bring about outcome-goal success, why do organizations continue to adopt the modus operandi? Like all organizations, militant organizations share two common aims: (1) to survive and (2) to achieve the goals that spawned and justify their existence (Crenshaw, 1988; Acosta, 2014). For militant organizations, achieving political ends depends on surviving and becoming or staying relevant long enough to reach desired outcomes (Cronin, 2009). Across organizational typologies, whether in the worlds of business, nonprofit, or revolution, survival is a ‘prerequisite for [organizational] success’ (Suárez & Utterback, 1995: 415). 14 In some instances, organizations pursue survival at the cost of efficiency or potentially reaching a more successful outcome (Zegart, 1999). In these respects, survival is the default mode of organizational behavior (Barnard, 1938; Weber, 1947; Krell, 2000). Ultimately, the adoption of suicide attacks strengthens an organization’s connection to the preferences of supporters and prospective supporters and as a result increases organizational status and thus durability.
The literature on organizational durability demonstrates that organizations can increase their likelihood of longevity in various manners. Adapting to new trends proves essential for organizational survival (Krell, 2000; Montuori, 2000). Gimeno et al. (1997) show that supporters of a firm will sustain it even when it ‘underperforms’, so long as the firm meets certain symbolic expectations. These findings underscore the importance of an organization’s relationship with its supporters (whether in the context of constituents and members or investors, employees, and customers). In waging militant campaigns, organizations necessarily remain popular within their constituency or risk becoming defunct (Mao, 1937/2000). 15
In numerous ways, conducting suicide attacks aids organizations in endeavors to maintain or improve popularity and survive. The extreme act of self-annihilating martyrdom signals a willingness to ‘do all it takes’ for a collective cause – an important message to send in any contentious setting where organizations ask supporters to put their own livelihoods at risk. As such, Hoffman & McCormick contend that organizations carry out suicide attacks as ‘a strategy of political mobilization [… which attempts to] stimulate popular activism, bolster the group’s base of support, jump-start the mobilization process, and help achieve a self-sustaining rate of organizational growth’ (2004: 246). The strategy’s mechanism hinges on the adoption of suicide attacks strengthening the link between organizational action and the people’s expectations of its representative (Pedahzur, 2005). As martyrdom operations often fulfill the ideological or cultural preferences of constituents or potential supporters (Hafez, 2006, 2007a; Acosta, 2010; Braun & Genkin, 2014), utilizing suicide attacks can elevate the sociopolitical status of an employer (Pedahzur & Perliger, 2006) or keep or make an organization relevant within a specific political landscape (Hoffman & McCormick, 2004; Moghadam, 2007). Among constituencies that attach social value to acts of self-annihilating martyrdom, deploying suicide bombers enhances an organization’s political support (Bloom, 2005; Khosrokhavar, 2005), and this is the case despite the inability of suicide attacks to precipitate outcome-goal success (Acosta, 2014). For example, Bloom observes: ‘Support for suicide [attacks] works against the stated [outcome] goals [of Palestinian society …] Yet, the majority of Palestinians support the continuation [… of] martyrdom operations regardless of Israeli retaliatory policies’ (2005: 20). The prestige that some identity groups allot to martyrdom acts serves to motivate organizations seeking to increase constituent support to conduct suicide attacks. Further, survival and popularity regularly depend on an actor’s ability to adjust, adapt, or brand itself in the zeitgeist of the era. Organizations operating in a particular environment prior to the popular emergence of suicide attacks might imperatively adopt the tactic to preserve popularity once the modus operandi receives widespread support (Bloom, 2005). In this sense, jumping on the suicide-attack bandwagon could mean the difference between survival and oblivion or popularity and irrelevance.
H1: Adopting suicide attacks increases an organization’s likelihood of survival.
The greater arena of political violence also affects a militant organization’s popularity. Organizations situate within a global network of militant organizations (see Figure 3). Much like social networks channel resources and mobilize individuals to participate in politics (Lake & Huckfeldt, 1998), relationships within the network of militant organizations boost the resources, mobilization, and general activity of individual militant organizations (Asal & Rethemeyer, 2008; Acosta & Childs, 2013; Horowitz & Potter, 2014).
Network of militant organizations (1980–2013)
Network ties, or connections between militant organizations, play a central role in organizational efforts to survive and augment capacity to institute political change (Karmon, 2005; Acosta, 2014; Phillips, 2014). Ties create channels from which resources may disperse, as well as increase the potential for expanding or uniting previously unassociated constituencies or identity groups (Hegghammer, 2010). 16 As network ties provide organizations with multiple benefits, one can conceptualize (and quantify) organizational status in terms of an organization’s network centrality (i.e. its number of network ties). 17
Among militant organizations, a type of political violence ‘fashion’ persists, evolves, and works to confer status on organizations. Simmel observes, ‘fashion can to all appearances and in abstracto absorb any chosen content: any given form of clothing, of art, of conduct, of opinion may become fashionable’ (1904: 352). Different eras of political violence have witnessed distinct fashions. Circa the end of the 19th century, militants frequently framed acts of violence as a pursuit of anarchism and preferred the tactics and targeting profiles of assassinating politicians and using dynamite to bomb civilian sites. The first half of the 20th century saw the dawn of militant nationalist organizations such as the Haganah and Irgun in Mandatory Palestine and the Front de Libération Nationale in Algeria. The 1960s and 1970s brought about the ‘international revolutionary’ and ‘urban guerrilla’, made popular by organizations like Rote Armee Fraktion, Revolutionäre Zellen, the Japanese Red Army, and PFLP. As their complementary modus operandi of choice, such leftist organizations preferred airline hijackings and hostage taking (Hoffman, 2006). With the advent of political Islamist organizations over the last few decades, the contemporary suicide attack has increasingly found favor with militant organizations (Moghadam, 2009).
‘High-status’ organizations, that is, highly connected organizations within the global network of militant organizations, have greater sway in affecting which modi operandi become ‘fashionable’. Reflecting on Simmel’s insights, Gilbert (2008: 35) recognizes: some [political] actors are of higher status than others and […] all actors are motivated to try to gain status by imitating high-status actors […] For example, in a collectivity driven by fashion, all actors will want to be as fashionable as they can, which means adopting the clothing styles, musical tastes, or whatever of those whom they perceive to be of the highest status.
Thus, emulating and making connections to high-status organizations by taking on the fashionable trends of political violence may help an organization accumulate more network ties – elevating its status among fellow militant organizations.
Therefore, in addition to heightening popularity within its constituency (Bloom, 2005), lower-status organizations can adopt suicide attacks to look like and catch the attention of higher-status organizations that promote martyrdom operations, potentially opening up channels of external support. Giving weight to the adage that ‘birds of a feather flock together’, research shows that similarity in attributes and ideological congruence beget the formation of network ties between actors (Siverson & Emmons, 1991; Currarini, Jackson & Pin, 2009; Maoz, 2012), as actors are more willing to divert resources from their own cache to aid like actors if they share a sociopolitical end (Acosta & Childs, 2013). As such, organizations can employ suicide attacks to signal solidarity with a larger cause and exhibit alignment with like-minded militants, improving the odds of making network connections, receiving outside support, and moving from low to high status.
Al-Tawhid wal-Jihad’s adoption of suicide attacks and subsequent rise exemplifies this process. From its establishment in 1999 by Abu Musab al-Zarqawi to the Iraq War’s onset, al-Tawhid wal-Jihad conducted numerous conventional attacks yet held low status – failing to attract many supporters or interorganizational ties. In June 2003, al-Tawhid wal-Jihad began carrying out suicide attacks across Iraq, quickly placing al-Zarqawi and his organization at the forefront of a jihadi insurgency against United States-led coalition forces (Rosen, 2006). In the following years, al-Tawhid wal-Jihad carried out 90 suicide attacks, using the tactic to express commitment to the international jihadi-Salafi cause. At the time, no other militant organization operating in Iraq demonstrated a willingness to deploy mass suicide bombers. As a result, al-Tawhid wal-Jihad gained Osama bin Laden’s recognition and earned an official alliance with al-Qaeda in 2004, at which point the organization morphed into al-Qaeda in Iraq (AQI). Employing suicide attacks facilitated al-Tawhid wal-Jihad’s evolution from an insignificant band of fighters to a durable al-Qaeda satellite and leader within the jihadi-Salafi movement (Acosta, 2014). AQI went on to form a variety of network ties in Iraq and across the Middle East and Africa (Hafez, 2007b), ultimately expanding and transforming into the Islamic State of Iraq (ISI), later the Islamic State of Iraq and the Levant (ISIL), and currently the Islamic State. The organization’s regional prestige afforded by the high profile of its suicide attacks made AQI/ISI an attractive ally to Iraqi organizations such as Jaish Rijal al-Tariqa al-Naqshbandia and Ansar al-Islam – connections that helped AQI/ISI survive aggressive counter-insurgency programs like the 2007 US troop surge.
H2: Adopting suicide attacks increases an organization’s likelihood of making network ties.
Research design and empirical results
This study analyzes an original dataset of 310 militant organizations operative sometime between the years 1980 and 2013. 18 Accordingly, militant organizations mark the unit of analysis. The dataset contains the universe of organizations that carried out suicide attacks during the time period as well as a sample of militant organizations that refrained from employing the tactic. 19 In constructing the dataset, I randomly selected non-suicide attack employing organizations from a pool of organizations listed in Stanford’s Mapping Militants Project 20 and START’s Global Terrorism Database (GTD). 21
Testing organizational durability and status
Testing the hypotheses necessitates a series of cross-sectional statistical analyses. First, I run a Cox proportional hazards model or survival analysis to test H1 and assess the effects of suicide attacks on the durations of militant organizations. 22 Second, I conduct an OLS analysis to test H2 and evaluate the relationship between the organizational use of suicide attacks and network ties. 23
In the survival analysis, the Duration of militant organizations is the dependent variable. 24 In the OLS analysis, the number of Network ties an organization has made to fellow militant organizations is the dependent variable. 25 I test these two dependent variables for various reasons. Following previous scholarship, I treat an organization’s duration or life span as the chief proxy for organizational durability (Carter, 2012; Phillips, 2014). The choice to analyze network ties as a dependent variable is twofold: (1) network ties likely extend organizational duration (Acosta, 2014; Phillips, 2014) and (2) theoretically, network ties represent an appropriate proxy for identifying and quantifying an organization’s status among fellow militant organizations. In coding the dataset’s variables, I referred to a multitude of sources, including: GTD, the National Counterterrorism Center’s ‘Terrorist Groups’ and ‘Terrorist Profiles’, 26 the ‘Violent Extremism Knowledge Base’ of the Institute for the Study of Violent Groups, 27 the Memorial Institute for the Prevention of Terrorism’s ‘Terrorist Organization Profiles’, 28 Vice President’s Task Force on Combating Terrorism (1988), and Boot (2013).
Explanatory variables and controls
The key explanatory variables for the survival analysis are whether an organization Conducts suicide attacks and an organization’s number of Network ties. The main explanatory variable for the OLS analysis is Conducts suicide attacks. Alternative explanatory variables and controls include: organization Age, the organization size variables of 1,000-plus members and 10,000-plus members, the organization ideology variables of Political Islam and Leftist, the nationalist goal variable of Self-determination, an organization’s number of State sponsors, the number of Safe havens an organization has utilized, the number of Attacks an organization has carried out, the number of Kills an organization has inflicted, the primary enemy development variables of GDP per capita and Polity, 29 whether an organization has held Hegemonic status over its rivals, 30 and whether an organization signifies a Unified front, that is, an alliance of all groups seeking a particular outcome goal. To enhance the dataset’s reliability, I cross-referenced data entries with two or more independent sources. For further details on the coding and sourcing of individual variables, see the Codebook and Table D in the online appendix.
Statistical analyses
Table III reports the survival analysis results. I present hazard ratios, where results below 1 denote an increase in organizational survivability and results above 1 mark a reduction in survivability. Model 1 shows that employing suicide attacks increases the likelihood of militant organizations surviving. Network ties also boost militant survival. The unified front variable, which indicates the absence of interorganizational rivalries, has no effect. Leftist and nationalist outlooks increase survivability. Reaching 1,000-plus members and having more members than rival organizations (i.e. being hegemonic) bolster survivability as well. The significance of these latter two variables highlights the importance of mobilization and constituent representation in lengthening the durations of militant organizations. With a 49% jump in the survival rate of organizations that conduct suicide attacks, the difference in survivability between organizations that carry out suicide attacks and those that do not paints a stark picture (see Figure 4). Model 1 supports H1 that the use of suicide attacks directly, and indirectly through the accumulation of network ties, extends the life spans of militant organizations.
Cox proportional hazards regression results
Hazard ratios with robust standard errors in parentheses. **p < 0.01, *p < 0.05, † p < 0.1.
The effects of conducting suicide attacks on organizational survival
Table IV displays the OLS results. Model 2 shows that leftist ideologies, access to safe havens, number of inflicted kills, wealthier primary enemies, hegemony, and the use of suicide attacks are all positively associated with network ties. With some expectation, organizations that represent a unified front have a negative relationship with ties. Notably, safe havens and suicide attacks have the clearest associations with network ties. Model 2 supports H2 that adopting suicide attacks increases the likelihood of forming network ties. 31
OLS regression results
Coefficients with robust standard errors in parentheses. **p < 0.01, *p < 0.05, † p < 0.1.
Explanations
I now contextualize the results by contrasting my theory’s explanatory capability with common alternative explanatory frameworks found in the literature. First, assessing directionality, I address the nexus between the adoption of suicide attacks, the establishment of network ties, and organizational longevity/sociopolitical relevance. Second, countering expectations from outbidding theory, I review the role of organizational bandwagoning in the use of suicide attacks and illustrate why some organizations may choose to adopt the tactic late in their organizational development.
The relationship between suicide attacks and network ties
The relationship between the organizational adoption of suicide attacks and network ties runs both ways. As the statistical analyses indicate, the two variables have a high degree of association. The empirical findings allude to the central role that high-status organizations like Hezbollah, Hamas, and al-Qaeda have played in perpetuating suicide attacks. The results also support contentions that even high-status organizations have influenced one another in their initial adoption of suicide attacks – for example, Hezbollah inspiring the LTTE (Hoffman & McCormick, 2004) and Hamas and Palestinian Islamic Jihad (PIJ) (Atran, 2003), the latter two which influenced al-Qaeda (Acosta, 2010), and in turn inspired al-Tawhid wal-Jihad and various others (Rosen, 2006). Some previous studies speculate on a causal direction of the relationship between suicide attacks and network ties, suggesting that ties ‘cause’ the use of suicide attacks (Atran, 2003; Horowitz, 2010). 32 However, surveying the universe of observations reveals that in some instances particular network connections lead organizations to adopt suicide attacks, but in most cases organizations adopt suicide attacks in order to establish new ties. (See Figure B in the online appendix for annual visualizations of ties and the adoption of suicide attacks within the global network of militant organizations.) Next, I review some case examples that further demonstrate that organizations tend to adopt suicide attacks then form numerous network connections rather than vice versa.
Established in 1957, Hezb al-Da’awa al-Islamiyya (HDI) in its first few decades remained fairly inconsequential in its bid to replace the Iraqi state with a Shi’a-dominated republic. By the 1980s, facing the powerful and secular Ba’athi regime, HDI sought to enlist new supporters and allies in its uphill battle. On 15 December 1981, HDI attacked the Iraqi embassy in Beirut, Lebanon using a suicide bomber – a tactic that the Iranian Bassidj had revived for the contemporary era a year earlier. 33 With its suicide attack, HDI aimed to draw greater attention to the plight of Iraq’s Shi’a (Parker & Salman, 2013). In the following years, HDI continued its limited use of suicide attacks for the purpose of gaining new allies and supporters. By the mid-1980s, HDI had employed suicide attacks to make alliances with other Shi’a organizations, such as Lebanese Hezbollah. In addition, HDI’s use of suicide attacks helped to sustain its relationship with the Islamic Republic of Iran as it reaffirmed commitment to Ayatollah Khomeini’s preferred concept of self-annihilating martyrdom. 34
Founded in 1988, al-Qaeda (Central) refrained from using martyrdom operations in its first attacks. 35 Not until 1998 with attacks on US embassies in Kenya and Tanzania did al-Qaeda employ suicide attacks. Al-Qaeda’s belated adoption came after five years of Islamist-Palestinian organizations Hamas and PIJ making the suicide bomber/self-annihilating martyr popular in the Sunni world (Acosta, 2010). The mid-1990s had seen widespread support for Palestinian suicide bombers across the Sunni world. Only after al-Qaeda recognized the potential of martyrdom operations as a mobilizing force and alliance-building mechanism did bin Laden sanction their use (Acosta, 2010). Subsequent to adopting suicide attacks in 1998, al-Qaeda reached a new level of prestige within the larger sphere of militant organizations – forming multiple new organizational connections. A similar development on a much greater magnitude occurred after al-Qaeda’s 9/11 attacks. Centrally, adopting suicide attacks allowed al-Qaeda ‘to inspire [segments within] the wider Muslim Community and to show the way to other Islamist movements’ (Gunaratna, 2005: 43). Employing suicide attacks aided al-Qaeda in making connections to smaller, localized organizations and redirecting their ire to international foes (Moghadam, 2009).
Egypt’s Ansar Beit al-Maqdis (ABM) depicts a similar relationship between the adoption of suicide attacks and the formation of network ties. Established amid Egypt’s 2011 anti-Mubarak Revolution by Bedouin-jihadis in the Sinai, ABM increased its violent activity after the overthrow of the democratically elected Egyptian President (and Muslim Brotherhood official) Muhammad Morsi. In September 2013, in an effort to heighten its profile, ABM started conducting suicide attacks. Speaking to prospective domestic supporters, ABM has claimed that some of its attacks are ‘in revenge for the killing of protestors [by security forces]’. 36 Yet, in a message taking responsibility for a December 2013 suicide attack, ABM leadership stated that the use of suicide attacks signifies a call on fellow mujahedin (holy warriors) to assist the fight against ‘Egyptian apostates and tyrants’. 37 By mid-2014, ABM’s suicide attacks had garnered the Islamic State’s attention – leading to the establishment of a link with the organization, which began supplying ABM with material support. 38 In November 2014, ABM’s leaders swore a bayah (loyalty oath) to Caliph Ibrahim ibn Awad (aka Abu Bakr al-Baghdadi, leader of the Islamic State). 39 Soon after, al-Baghdadi brought ABM into the Islamic State’s fold, rebranding the organization Wilayat Sinai (Sinai Provence) of the Islamic State. 40
In a minority of cases, organizations form ties with an organization that conducts suicide attacks and then adopt the tactic. For example, both Hamas and PIJ made connections with Hezbollah and subsequently employed suicide attacks. In late 1992, as a response to the seemingly endless first intifada (uprising), Israel’s Rabin government deported 418 Hamas and PIJ members to southern Lebanon (Atran, 2003). Under the auspices of Hezbollah and the Iranian Pasdaran, some of the deportees learned functional and ideological elements of conducting suicide attacks and established enduring connections with their hosts (Hoffman, 2006). By the end of 1993, the Rabin government had allowed most of the deportees to return to the disputed territories. By no coincidence, Hamas and PIJ attacked Israel with a combined nine suicide bombers during 1993 and went on to carry out upwards of 200 suicide attacks over the next 15 years. Throughout the 1990s and early 2000s, a commitment to martyrdom operations, more than any other tactical or ideological component, symbolized the alliance between Islamist-Palestinian organizations (Hafez, 2006). 41
Under both circumstances – when ties precipitate the use of suicide attacks and when the use of suicide attacks leads to the establishment of new ties – employing organizations display an effort to maintain or boost popularity and status. Organizational popularity and prestige among constituents, and thus support, do not rise or fall in isolation. Constituents usually have multiple options to choose from when deciding how to have their preferences and interests represented. Organizations may connect to high-status organizations in hope that some of their prestige will rub off – this is the status-by-association strategy. Organizations can also attempt to mimic high-status actors by adopting their preferred mode of operation in efforts to form an alliance – this is the adopt-the-fashionable-trend-to-get-status strategy. As discussed below, organizations that adopt suicide attacks to make ties and organizations that adopt suicide attacks after making specific ties typically use the modus operandi for the same purpose: to elevate organizational status within a certain constituency, identity group, or network.
Bandwagoning, cooperation, and joint suicide attacks
I now focus the discussion by applying the bandwagoning framework to an analysis of direct operational cooperation. Studies of political violence analyze bandwagoning in terms of competition and rivalry between militant organizations (Crenshaw, 1981; Kydd & Walter, 2006). Notably, Bloom (2005) asserts that rival militant organizations use suicide attacks to compete for popularity in societies where martyrdom operations generate high degrees of support. In contrast, this study’s empirical analysis suggests a cooperative spirit between organizations that adopt suicide attacks – even among some organizations with explicit and lasting rivalries.
Along these lines, the suicide-attack phenomenon involves two types of cases: (1) ‘newer’ organizations aiming to increase support and resources and (2) ‘older’ organizations attempting to stay relevant, ‘keep up with the times’, and retain their sources of sustainability. 42 Employing suicide attacks might breathe new life into an old and dying organization that finds itself eclipsed by a newer organization that rose to power on the popularity of its martyrs/suicide bombers. Alternatively, a newer organization can ‘make a name for itself’ by carrying out suicide attacks. 43 Older and newer organizations normally affect the regional fashionability of suicide attacks differently, with the latter often setting the fad and the former trying to readapt in order to maintain popularity.
Adopting suicide attacks tends to help newer militants gain a distinctive branding and attain greater support among a certain constituency. As previously noted, al-Tawhid wal-Jihad’s use of martyrdom operations led to its heightened popularity within the international jihadi-Salafi movement. As with al-Tawhid wal-Jihad, Somalia’s Harakat al-Shabaab al-Mujahedin and Nigeria’s Boko Haram failed to make names for themselves until they adopted suicide attacks and received the support of al-Qaeda and its regional branches. Similarly, once the burgeoning Jabhat an-Nusrah li-Ahl ash-Sham (JNAS) conducted numerous suicide attacks against the Assad regime in late 2011 and early 2012, Syria’s Civil War quickly became a focal point for international jihadi organizations like ISI/ISIL and the predominately Chechen Jaish al-Muhajerin wal-Ansar. 44 Such organizations employ suicide attacks as a way to stand out and acquire status and with it new resources and the mobilization of additional followers and supporters.
Many older organizations have adopted suicide attacks with the aim of preserving status in the face of the trendy phenomenon sweeping across their area of operation. Older organizations jumping on the bandwagon of martyrdom operations reflects efforts to survive and prevent a loss of status amid a changing zeitgeist. As Krell contends, ‘a failure to adjust interaction [to changes] in the external environment [threatens organizational survival]’ (2000: 9). When suicide attacks reach a degree of popularity among an identity group or constituency, it becomes difficult for organizations that claim to represent the same identity group to remain relevant and sometimes survive without adopting popularized martyrdom operations. Learning and demonstrating the new trade becomes paramount for organizational survival (Montuori, 2000). Suicide attacks can therefore function as a blood oath between organizations, new and old, and high and low status. This kind of arrangement, which takes place today in Pakistan, took place in Iraq in the mid-2000s, in Lebanon in the mid-1980s, and among Palestinian organizations in the early 2000s. The strategy is not so much an example of organizational ‘outbidding’ wherein organizations compete for popular support, like Bloom (2005) argues, as it is an organization’s articulation of ideological solidarity and cooperation. Employing martyrdom operations under these circumstances may keep an organization relevant in the eyes of its constituency and larger collective identity group, and strengthen ties and partnerships with other organizations that use suicide attacks.
Ultimately, cooperation and solidarity rather than competition have defined the relationships between most older and newer organizations using suicide attacks contemporaneously. Forming physical ties, alone, exhibits a collaborative outlook between organizations, and co-sponsoring or conducting joint operations exposes a willingness to conspire and have their martyrs kill and die together. For example, during the mid-1980s in Lebanon, as Hezbollah’s international infamy and domestic clout steadily rose due to the attention received from its suicide attacks on US, French, Israeli, and South Lebanon Army targets, other organizations began adopting the tactic. Older organizations such as the Lebanese Communist Party (LCP) (established in 1924), the Syrian Social Nationalist Party (SSNP) (established in 1932), Syrian Ba’ath Organization (established in 1946), al-Mourabitoun (established in 1957), Arab Socialist Union (established in 1973), and Amal (established in 1975) began conducting suicide attacks to stay relevant amid the buzz surrounding the martyrdom operations of the mysterious and much newer Hezbollah (established in 1982 as the Islamic Jihad Organization). Underscoring the cooperative spirit of suicide-attack bandwagoning, various organizations launched joint suicide attacks in Lebanon during the mid-1980s. Jammoul incorporated a variety of groups under its umbrella, including the LCP and the SSNP, which collectively carried out suicide attacks. And, once Hezbollah formally announced itself to the public in 1985, the Shi’a organization and Jammoul began co-sponsoring martyrdom operations. 45
As discussed above, Palestinian organizations have likewise cooperated in conducting suicide attacks more than they have competed. Hamas and PIJ almost immediately started cosponsoring suicide attacks in 1993 when they both began using the tactic. As suicide attacks bolstered the popularity of Hamas and PIJ, the much older Fatah started deploying suicide bombers. In the early 2000s, rather than compete with other organizations, Fatah frequently co-sponsored attacks with PIJ, the Democratic Front for the Liberation of Palestine (DFLP), and Popular Resistance Committees, as well as with its rivals PFLP and Hamas. Founded in 1958, Fatah had always sponsored feda’i (self-sacrificer) attacks, or high-risk operations, and energetically praised the generic shahid (martyr), but the organization did not begin employing the istishhadi (self-annihilating martyr) in full-fledged suicide attacks until 2000 amid the onslaught of the second intifada. At that point in time, producing martyrs had become the benchmark for political support within Palestinian society (Hafez, 2006). Consequently, Fatah’s political survival depended on its ability to overtly demonstrate ideological solidarity with Islamist-Palestinian organizations like Hamas and PIJ through the use of suicide attacks (Zelkovitz, 2008).
This bandwagoning process, as announced through the deployment of martyrs/suicide bombers, has occurred in other central geographies of suicide attacks. During the mid-2000s, the Islamic Army of Iraq – mainly consisting of former Ba’ath party members – began carrying out suicide attacks to gain a foothold in Iraq’s new domestic political milieu that the jihadi and largely foreign al-Tawhid wal-Jihad had reshaped. During the late 2000s in Afghanistan, a comparable development transpired with older organizations jumping on the suicide-attack bandwagon. Sending martyrs on suicide attacks rejuvenated the Taliban’s domestic standing, forcing other militants in Afghanistan to attempt to ‘keep up with the times’. Hekmatyar’s Hizb-e Islami (established in 1977) and the Haqqani Network (established in 1973) began conducting suicide attacks in 2007 and 2008 respectively. Other older organizations that did not begin launching suicide attacks until the phenomenon swept across their area of operation include Tehrik-e Nafaz-e Shariat-e Mohammadi, Harakat-ul Mujahedin, Ansar-ul Islam, and Lashkar-e Islam in Pakistan, Jundallah, Islamic Jihad of Yemen, Jamaat-ul Mujahedin in Bangladesh, Turkestan Islamic Party, Egyptian Islamic Jihad, and al-Qaeda.
Cooperation in conducting suicide attacks highlights the survival motive of newer and declining organizations’ efforts to copy the pathways of higher-status organizations that have attained prestige by carrying out martyrdom operations. To test cooperation directly, I run two additional survival models that evaluate the variables Conducts joint suicide attacks and Conducts joint conventional attacks. Referring back to Table III, Model 3 shows that organizations that conduct joint suicide attacks increase their likelihood of survival by 78%. And, organizations that carry out joint suicide attacks boost their survivability more than organizations that only use the modus operandi in solo operations. Model 4 reveals that conducting joint suicide attacks increases organizational survival considerably more than carrying out joint conventional attacks.
Bandwagoning in efforts to survive or elevate status explicates the adoption of suicide attacks by numerous organizations that had existed for many years, if not decades, before deciding to utilize the tactic. Bandwagoning indicates an organizational awareness of the necessities to obtain or preserve amicable ties to more popular organizations and bend to the greater will of the constituency. Certainly older organizations in survival mode facing the rise of a newer organization that employs a trendy tactic have had to reinvent themselves by ‘dressing’ in the new fashion. In the end, the ‘purpose of [any] fashion is to facilitate differentiation of “types” in the process of social interaction’ (Pesendorfer, 1995: 772). Thus, mimicking the appearance and behavior of popular or high-status organizations, including through collaboration, tells the constituency at hand that an older organization is still ‘one of them’. Organizations like Fatah, DFLP, Hizb-e Islami, and the Haqqani Network – all of which had notable militant reputations yet whose track records lacked suicide attacks – exemplify such reinventions. When an organization galvanizes societal support by instituting and popularizing an ideological concept such as self-annihilating martyrdom, other organizations quickly realize that clearly demonstrating solidarity or at least semblance may mean the difference between popularity and irrelevance, or worse – the difference between surviving and becoming defunct.
Conclusion
Contributing to the understanding of organizational durability, this study illustrates that insights from the literatures on the longevity of enterprising organizations and firms and the relationship between fashion and status apply to organizations involved in militant and revolutionary campaigns. Centrally, an organization’s ability to adapt and/or market itself within an era’s zeitgeist is essential to establishing, preserving, or enhancing constituent and external support. Organizational popularity and survival depend on a consistent stream of support, which the use of symbolic acts such as suicide attacks can generate.
In particular, the organizational use of suicide attacks – over time and across regions – has functioned as an instrument of differentiation or political fashion. As martyrdom operations have become emblematic of the contemporary era of political violence and fulfill the ideological expectations of specific identity groups, representative organizations have adopted suicide attacks to maintain or gain popularity or simply survive. Low-status, usually newer, organizations conduct suicide attacks to ‘make a name for themselves’, acquire status and supporters, and increase resources. Older organizations, years or decades after their founding, adopt suicide attacks in order to ‘keep up with the times’, jump on the bandwagon of an emerging popular sentiment, and prevent a loss of status and support.
The theory and complementary empirics put forth in this study explain that organizations at different stages of development, facing dissimilar enemies, and situated in unique conflict environments tend to adopt suicide attacks for the most basic of organizational aims – to ensure longevity and sociopolitical relevance. While suicide attacks have remained a generally ineffective instrument of coercion and political bargaining, militant organizations have continued to adopt the modus operandi and at higher rates with each passing decade. No political entity can effectively pursue its outcome goals without first securing its capability to persist, building up its capacity, and enlisting enough supporters to carry out the fight. Acting within these traditional mechanisms of political activity, organizations have turned to suicide attacks to better represent their constituencies and supporters and signal ideological alignment with external militants and identity groups. As such, militant organizations that conduct suicide attacks live longer than militant organizations that do not use suicide attacks.
Importantly, improving status is not the only reason organizations employ suicide attacks. However, status marks the leading strategic rationale for the organizational adoption of suicide attacks. Organizations like Tehrik-e Taliban, the Islamic State, and others often launch suicide attacks with the simple objective of killing ‘infidel’ enemies or punishing ‘heretics’. Yet, even such brute-force applications regularly serve the strategic purpose of representing the wishes of constituents and thereby work to bolster status and support.
Footnotes
Replication data
Acknowledgements
I thank Ronit Ben-Shoham for assistance in data collection and Kristen Ramos, AE Shencopp, Steve Childs, Craig Whiteside, Lawrence Rubin, and anonymous reviewers for helpful comments.
Notes
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
