Full Terms & Conditions of access and use can be found at
http://www.tandfonline.com/action/journalInformation?journalCode=fsst20
Security Studies
ISSN: 0963-6412 (Print) 1556-1852 (Online) Journal homepage: http://www.tandfonline.com/loi/fsst20
“Explaining Suicide Terrorism: A Review Essay”
Martha Crenshaw
To cite this article: Martha Crenshaw (2007) “Explaining Suicide Terrorism: A Review Essay”,
Security Studies, 16:1, 133-162, DOI: 10.1080/09636410701304580
To link to this article: https://doi.org/10.1080/09636410701304580
Published online: 17 Apr 2007.
Submit your article to this journal
Article views: 5048
Citing articles: 39 View citing articles
Security Studies 16, no. 1 (January–March 2007): 133–162
Copyright © Taylor & Francis Group, LLC
ISSN: 0963-6412 print / 1556-1852 online
DOI: 10.1080/09636410701304580
“Explaining Suicide Terrorism: A Review Essay”
MARTHA CRENSHAW
Mia Bloom, Dying to Kill: The Allure of Suicide Terror (New York: Columbia
University Press, 2005).
Joyce M. Davis, Martyrs: Innocence, Vengeance and Despair in the Middle
East (New York: Palgrave Macmillan, 2003).
Diego Gambetta, ed., Making Sense of Suicide Missions (New York: Oxford
University Press, 2005).
Mohammed M. Hafez, Manufacturing Human Bombs: The Making of Palestinian
Suicide Bombers (Washington, D.C.: United States Institute of Peace
Press, 2006).
Raphael Israeli, Islamikaze: Manifestations of Islamic Martyrology (London:
Frank Cass, 2003).
Farhad Khosrokhavar, Suicide Bombers: Allah’s New Martyrs, translated from
the French by David Macey (London: Pluto Press, 2005).
Anne Marie Oliver and Paul F. Steinberg, The Road to Martyrs’ Square: A
Journey into the World of the Suicide Bomber (New York: Oxford University
Press, 2005).
Robert A. Pape, Dying to Win: The Strategic Logic of Suicide Terrorism (New
York: Random House, 2005).
Ami Pedahzur, Suicide Terrorism (Cambridge: Polity Press, 2005).
Ami Pedahzur, ed., Root Causes of Suicide Terrorism: The Globalization of
Martyrdom (London and New York: Routledge, 2006).
Christoph Reuter, My Life is a Weapon: A Modern History of Suicide Bombing,
translated from the German by Helena Ragg-Kirkby (Princeton: Princeton
University Press, 2004).
Shaul Shay, The Shahids: Islam and Suicide Attacks (New Brunswick: Transaction
Publishers, 2004).
Martha Crenshaw is the Colin and Nancy Campbell Professor of Global Issues & Democratic
Thought and Professor of Government at Wesleyan University, Middletown, CT. She is
a Lead Investigator with the National Center for the Study of Terrorism and the Response to
Terrorism (NC-START), funded by the Department of Homeland Security at the University of
Maryland.
I thank Richard Boyd and the past and present editors of Security Studies for their helpful
comments.
133
134 M. Crenshaw
Barbara Victor, Army of Roses: Inside the World of Palestinian Women Suicide
Bombers (Emmaus Pa.: Rodale [distributed by St. Martin’s Press] 2003).
THE PROBLEM OF “SUICIDE TERRORISM”
The current trend toward suicide bombings began in Lebanon in the early
1980s. The practice soon spread to civil conflicts in Sri Lanka, the Kurdish
areas of Turkey, and Chechnya. Palestinian attacks on Israeli civilians in
the 1990s and during the Al Aqsa intifada further highlighted the threat.
Al Qaeda’s adoption of the tactic brought a transnational dimension. Interest
in the phenomenon then surged after the shock of the 2001 attacks,
which involved an unprecedented number of both perpetrators and casualties.
Since then, suicide bombings have expanded in number and geographical
range, reaching extraordinary levels in the Iraq War and spreading
around the world to countries such as Saudi Arabia, Morocco, Tunisia, Kenya,
Indonesia, Turkey, Pakistan, India, Afghanistan, Egypt, Jordan, Bangladesh,
and Britain.1
This review covers thirteen of the books published on the subject since
2002. Three analyze the Palestinian case and four others focus on Islamist
violence. The other six, including two edited collections, intend to be comprehensive.
This review also refers to a few selected publications that discuss
the arguments presented in the sources listed above. It aims to give
readers a glimpse of the content of the different volumes as well as offer a
critique.
These works make important contributions, although explanations are
still at an early and uneven stage. The concept remains imprecise, the facts
are not well established, and neither explanations nor policy recommendations
distinguish sufficiently between suicide and other terrorist or insurgent
attacks or account for variations within the phenomenon. Specifications of
what is to be explained vary by author. Findings are often based on incompatible
datasets, and references to cases or examples do not always fit the stated
definition of the concept. Contradiction, ambiguity, and error are particularly
consequential because the overall number of suicide attacks is quite small
compared to total numbers of attacks on similar targets using other means.
Inclusion or exclusion of a few events can thus shape the conclusions that
are drawn.
In addition, many accounts are being overtaken by events. Suicide attacks
in Iraq outnumber all other campaigns and challenge some of the
1 They also resumed in Sri Lanka in 2006. Nevertheless, few analysts would agree with Shay’s depiction
of modern suicide terror as “a strategic threat against the security and stability of Western society,
and perhaps against the security of human society as a whole.” See Shay, The Shahids, xiii.
Explaining Suicide Terrorism: A Review Essay 135
explanations offered by general studies.2
For example, Mohammed Hafez
lists 443 suicide attacks in Iraq from 22 March 2003 to 20 February 2006.3
The Brookings Iraq Index as of 24 January 2007 reports 1,188 multiple fatality
bombings, including at least 403 suicide bombings.4
In comparison,
fewer than 200 incidents were associated with the next most consequential
campaign, Palestinian violence against Israel from 1993 to 2006.5
The review is organized as follows. I first ask what it is that the authors
are trying to explain and then compare their answers to three questions:
why sponsoring organizations would see suicide attacks as effective, why
a community would support them, and why individuals would engage in
them. Last, I review the resulting policy recommendations.
THE CONCEPT: WHAT IS SUICIDE TERRORISM?
Research is plagued by the lack of a common definition of the concept. As a
start, is the unit of analysis a special form of terrorism or a politically neutral
suicide mission or attack?6
Some authors avoid the term terrorism altogether,
adopt the terrorism label without defining it, or expand the range of applicability
of the definition well beyond terrorism so as to include, for example,
the Japanese kamikazes. Among those who resist the term terrorism, some
focus on the subjective interpretations of the perpetrators and the idea of
martyrdom in specific cultures and thus avoid the term suicide as well, because
both labels are objectionable to those who practice or condone such
2 Of the works that discuss Al Qaeda and Iraq, note the time periods they cover: Khosrokhavar,
Suicide Bombers and Reuter, My Life is a Weapon end in 2002; Pape, Dying to Win, Israeli, Islamikaze
and Shay, The Shahids conclude at the end of 2003; Gambetta, ed., Making Sense ends in summer of
2004; Bloom, Dying to Kill about January 2005; and Pedahzur, Suicide Terrorism and Pedahzur, ed., Root
Causes through June 2005.
3 Hafez, “Suicide Terrorism in Iraq: A Preliminary Assessment of the Quantitative Data and Documentary
Evidence,” Studies in Conflict and Terrorism 29 (2006): 1–27. An extended analysis of Iraq will
be published as a monograph by the United States Institute of Peace in 2007.
4 See www.brookings.edu/iraqindex, esp. 19. The report notes an American military estimate that
96 percent of suicide bombers are not Iraqis. Brookings ceased to track car bombs in the fall of 2006
because of data inadequacies.
5 From October 2000 to February 2005, during the Al Aqsa Intifada, approximately 116 attacks occurred
with an overall 1993–2005 total of 142 attacks. See Hafez, Manufacturing Human Bombs, 19–20.
According to the National Memorial Institute for the Prevention of Terrorism (MIPT) Terrorism Knowledge
Base (TKB), Palestinian Islamic Jihad was responsible for 10 suicide attacks from February to June 2006,
leading to a total of 152 suicide attacks by Palestinian groups against Israel. Robert J. Brym and Bader
Araj count 20 attacks between 1993 and 1997, and 138 from 2000 to July 2005. They use Arabic press
accounts as well as the standard sources. They list 832 total deaths from 1993–2005. See Robert J. Brym
and Bader Araj, in “Suicide Bombing as Strategy and Interaction: The Case of the Second Intifada,” Social
Forces 84, no. 4 (June 2006): 1969–86. On the other hand, Shay concludes that over 900 Israelis were
killed between 2000 and 2003 alone. See Shaul Shay, The Shahids, 62.
6 Assaf Moghadam also raises some of these issues. He settles for suicide attack. See Assaf Moghadam,
“Defining Suicide Terrorism,” in Pedahzur, ed., Root Causes, 13–24. In Gambetta et al., Making Sense, they
also emphasize the lack of uniformity in definitions of the phenomenon.
136 M. Crenshaw
violence. Others use the term terrorism to refer exclusively to attacks on civilian,
not military, targets. Sometimes the reader has to tease out a definition
by implication.
Diego Gambetta and contributors stick largely to suicide missions (SM).
However, other social scientists (for example, Mia Bloom, Hafez, Robert
Pape, and Ami Pedahzur) refer explicitly to suicide terrorism.7
Shaul Shay,
a historian who heads the Israeli Defense Forces Department of History,
uses the terms suicide attack and suicide terrorism interchangeably. Historian
Raphael Israeli is adamant in insisting that the term terrorism be
used to describe violence associated with Islam; he implies that relatively
neutral terms such as suicide bomber are too weak and insufficiently
condemnatory.8
The English translation of Khosrokhavar’s Les Nouveaux Martyrs d’Allah
adds the title Suicide Bombers, relegating Allah’s New Martyrs to a subtitle,
and a prefatory Note on Terminology explains that terrorism and jihadism will
mean the same thing. His focus is on the concept and practice of offensive
martyrdom in Islam.
Anne Marie Oliver and Paul Steinberg, who document the culture that
supports Hamas, refer to suicide bombings and to murder-suicides. They do
not use the term terrorism at all; the only reference in the index is to a picture
of a poster displaying the late Palestinian cleric Abdullah Azzam, whose text
reads: “If preparation is terrorism, then we are terrorists, and if defense of
the land is extremism, then we are extremists, and if jihad against enemies
is fundamentalism, then we are fundamentalists.”9
Barbara Victor and Joyce Davis, respectively, use interviews and biographical
vignettes to present the subjective side of martyrdom in the Palestinian
and Islamic contexts. Victor’s scope is further narrowed to women
martyrs.
Christoph Reuter, also a journalist, defines his topic as modern suicide
bombing, although he includes chapters on the Assassins of medieval Islam
and the Japanese kamikaze, the historical precedents that most studies cite.
He criticizes the indiscriminate labeling of groups as terrorist. His interest, too,
is in the lived experiences of individuals and societies rather than theoretical
explanation.10
Gambetta’s edited collection compares political suicide that kills others
to that which does not (for example, self-immolation) and also asks why
7 Hafez, Manufacturing Human Bombs, 4. It is hard to escape the term. For example, although Hafez
states that he will use the terms “suicide bomber” or “human bomb” because they are not highly charged
normative terms like “suicide terrorist,” he then devotes a chapter to explanations of suicide terrorism.
8 Israeli, Islamikaze, 5.
9 Oliver and Steinberg, The Road to Martyrs’ Square, fig. 12. Azzam, who was assassinated in 1989,
is also notable as the mentor of Osama Bin Laden.
10 Reuter, My Life is a Weapon, 10. Reuter’s caution makes it ironic that Walter Laqueur, in his
endorsement of the book, praises it as an account of suicide terrorism.
Explaining Suicide Terrorism: A Review Essay 137
some groups refrain from suicide tactics.11
However, the authors who define
their subject as suicide terrorism are typically less sensitive to the need to
compare suicide attacks to other forms of terrorism or political violence. Pape
is the most explicit, although the facts he uses to support his argument do not
always fit the requirements of his definition. He defines the phenomenon as
the most aggressive form of terrorism, compared to demonstrative or destructive
terrorism, and proposes that its purpose is to kill the maximum number
of people from the opposing community.12
Pedahzur similarly says that such
attacks are intended to create an atmosphere of terror and to harm as many
people as possible, in most cases civilians.13
However, suicide attacks have been used to assassinate individuals and
to strike specific military targets and killing civilians does not distinguish
suicide attacks from other forms of violence.14
In several well-known cases,
suicide attackers did not target civilians indiscriminately or did so rarely.
Nevertheless, three such cases are included in both Pape’s and Pedahzur’s
studies.15
First, Hezbollah, the originator of the tactic in 1980s Lebanon, targeted
embassies (American and Israeli), military headquarters (American, French,
and Israeli), military convoys (Israeli), and on one occasion the Jewish Cultural
Center in Buenos Aires. Some of these attacks were extremely lethal,
but overall most were against military targets. Shay describes Hezbollah attacks
on Israeli targets in Lebanon as follows: “With the exception of the
1983 attack against the IDF [Israeli Defense Forces] headquarters in Tyre, all
of the attacks were directed against IDF convoys.”16
Second, the Liberation Tigers of Tamil Eelam (LTTE) in Sri Lanka adapted
the tactic from Hezbollah as a precise tool for assassinating government
officials and political rivals. The 1991 assassination of former Indian Prime
Minister Rajiv Gandhi was probably the LTTE’s most famous effort.17
Like
Hezbollah, the LTTE primarily attacked military targets, often to stop or thwart
Sri Lankan army offensives against LTTE-held territory or to damage the Sri
Lankan Navy.
11 See, respectively, Michael Biggs, “Dying Without Killing: Self-Immolations, 1963–2002,” in
Gambetta, Making Sense, 173–208, and Stathis Kalyvas and Ignacio Sanchez-Cuenca, “Killing Without
Dying: The Absence of Suicide Missions,” in Gambetta, Making Sense, 209–32.
12 Pape, Dying to Win, 9–10.
13 Pedahzur, Suicide Terrorism, 10–11.
14 Acknowledged by Pedahzur and Arie Perliger in “Introduction: Characteristics of Suicide Attacks,”
in Pedahzur, ed., Root Causes, 1–12.
15 Jeff Goodwin also makes this point about Pape. See Jeff Goodwin, “What Do We Really Know
About (Suicide) Terrorism?” Sociological Forum 21, 2 (June 2006): 315–30. His review also considers the
Bloom and Gambetta volumes.
16 Shay, The Shahids, 43.
17 Pape, Dying to Win, 227. Pape says that this was the first use of a belt bomb, but Shay lists one
in Lebanon in 1987.
138 M. Crenshaw
Hopgood contends that in Sri Lanka “there are no clear examples of
civilians being directly targeted by SMs [suicide missions]. . ..”18
“In none of
these examples was damage to civilians, or the spreading of ’terror,’ a principal
motivation.”19
Reuter agrees that the LTTE aims not to terrorize the enemy
population, but to strike the nerve center of the state.20
This does not mean,
however, that the LTTE has clean hands. Even if Hopgood argues that civilians
were not the intended target and that the LTTE could have attacked restaurants
and shops to much greater destructive effect, the LTTE’s suicide bombing of
the Central Bank in downtown Colombo, in January 1996, killed around 90
people and injured over a thousand. The January 1998 suicide truck bomb attack
on the Temple of the Tooth in Kandy targeted a major Buddhist shrine.21
Third, the PKK (Kurdistan Workers Party) in 1996 and again in 1998–
99 targeted Turkish government buildings and military assets, killing around
twenty people all told.22
The PKK killed thousands of Kurdish civilians, so
precision and restraint were not the organization’s hallmarks.
Readers might assume that suicide attacks would necessarily involve the
death of the perpetrator—and the dominant image is of the bomber who
dies in a single explosion along with his or her victims—but that is not
necessarily the case. On the one hand, Bloom, Hafez, and Shay think so. As
Bloom states it, the perpetrator’s death “is the precondition for the success
of the attack.”23
However, Pedahzur requires only that the odds of returning
alive are “close to zero,” and Pape specifies that “the attacker does not expect
to survive the mission” and “often employs a method of attack. . .that requires
his or her death in order to succeed.”24
Such a definition could include, for
example, the assassination of Indira Gandhi by her body guards, who were
then immediately killed by other security forces. These definitions also raise
the question of how we can know what a perpetrator expects.
More refinements further expand the scope of the concept. Gambetta
relaxes his initial assumption that the attacker’s death is a requirement to
18 Stephen Hopgood, “Tamil Tigers, 1987–2002,” in Gambetta, ed., Making Sense, 55.
19 Ibid., 43–76, esp. 59. The record of the LTTE in Sri Lanka is also disputed. Hopgood puts the
number of LTTE suicide attacks as “probably somewhere between 100 and 200. We simply cannot be more
accurate than that.” See Ibid., 44. Pedahzur lists more than 170 attacks between 1987 and 2003. Thus, as
Hopgood charges, Pape’s count of 75 is far too low. See Pedahzur, Suicide Terrorism, 88–89.
20 Reuter, My Life is a Weapon, 161.
21 According to the MIPT Terrorism Knowledge Base, eleven people were killed, including three or
four attackers, and 25 people were injured. In July 1996, two bombs on a commuter train killed at least
70 and wounded several hundred, although these were not suicide attacks. See New York Times, 25 July
1996.
22 Pape, Dying to Win, 162, 257–58. He cites 22 people killed in 14 or 15 attacks. Pedahzur, Suicide
Terrorism, 89. He describes 15 successful and 7 failed operations.
23 Bloom, Dying to Kill, 76. She defines her subject as suicide bombings, although she does not
restrict her scope to explosives and her examples in the introduction are eclectic. Citing the Thugs in
colonial India is particularly curious; they strangled unsuspecting fellow travelers and took few personal
risks. They certainly were not suicide attackers and they probably were not even terrorists.
24 Pedahzur, Suicide Terrorism, 8. Pape, Dying to Win, 10. Italics mine.
Explaining Suicide Terrorism: A Review Essay 139
include missions that were possible without the perpetrator’s death.25
Gambetta,
Pape, and Pedahzur include no-escape attacks, but Gambetta excludes
high-risk missions.26
Israeli implicitly includes no-escape incidents and is
silent about high risk. The rest simply do not tell us.
Another question is whether unsuccessful attempts are included. Both
Pape and Pedahzur include them, and Shay, although he says that he will
not, does sometimes include what he calls work accidents. Victor specifically
includes failed and foiled attempts. In fact, she suggests that since 2000, Palestinian
women were involved in 250 cases of which only four succeeded.27
Hafez notes that in 2004 Israel thwarted 74 percent of planned attacks and
arrested 365 militants.28
Excluding unsuccessful missions limits the numbers
of incidents and perpetrators. It also results in the omission of a number of
chilling plots, such as the August 2006 plan to bomb multiple airliners over
the Atlantic. One reason for excluding failures, of course, is a lack of information.
Researchers may not be aware of them because governments resist
disclosing sensitive information about their capacity to disrupt operations.
On the other hand, knowledge of intentions at the group or individual level
is incomplete without them.
Yet another point of contention is whether or not the suicides must be
acts of free will, which is also something that is not always easy to know.
To Gambetta, the perpetrators must be neither deceived nor blackmailed.
Coercion cannot be part of the process.29
Both Shay and Hafez name willingness
to die as the key element of their definitions. Bloom refers vaguely
to “a deliberate state of awareness.”30
Victor and Davis include cases where
bombers were duped or manipulated. Victor believes that Palestinian women
bombers were subjected to intense social pressures that pushed them in the
direction of martyrdom. Similarly, Farhad Khosrokhavar finds that in Iran the
25 Gambetta, ed., Making Sense, vi–vii.
26 It is interesting that contemporary veneration of martyrdom in jihad got its start with no-escape or
high risk attacks on military targets during the Iran-Iraq war, 1980–1988, when Iran employed human wave
attacks to clear mine fields and charge Iraqi battle formations. Some of the tens of thousands of untrained
youngsters, exhorted to go into battle wearing their shrouds and with keys to paradise around their necks,
saw themselves as martyrs, but others thought of their actions as high risk with a correspondingly high
potential pay-off in terms of socio-economic status. See Khosrokhavar, Suicide Bombers.
27 Victor, Army of Roses, 239. She cites an interview with Liat Pearl, a spokesperson for the Israeli
Border Police.
28 Hafez, Manufacturing Human Bombs, 72. He quotes an Israeli General Security Services report.
Many analyses of the characteristics of suicide attackers are based on interviews with those who failed,
so excluding these attempts would severely limit this line of research.
29 New York Times, 26 October 2005, A12. The bombings of the Palestine and Sheraton hotels in
central Baghdad on 24 October 2005, involved three car bombs. According to a hotel guard who saw
them, one was a cement mixer driven by a man whose hands were handcuffed to the steering wheel. In
Reuter, My Life is a Weapon, 165, he describes the case of a teenaged PKK recruit who blew herself up in
1996, after another young woman in the PKK refused to volunteer for a suicide attack and was shot dead
in front of her.
30 Bloom, Dying to Kill, 76.
140 M. Crenshaw
prototypical martyrs, nominally volunteers, were “systematically manipulated
by the state.”31
The case of the train bombs in Spain in March 2004 illustrates some of the
problems of classification. The bombings caused 191 deaths, making them
one of the most destructive terrorist attacks since 9/11 and they involved the
coordination of multiple bombs by a relatively large conspiratorial group. The
initial bombings were conventional, since the bombs were left on trains. Most
of the group members, however, later blew themselves up when cornered by
the police. The group may have been planning a second attack that would
have involved their deaths. Were the Madrid bombings and/or the subsequent
self-destruction of the perpetrators acts of “suicide terrorism”?32
Readers might think that these distinctions are arcane or trivial, but they
matter not only for analytical clarity and consistency and data collection but
also for the policies of state actors. For example, Shay describes the support
provided to Hamas and Hezbollah by Iraq, Iran, and Syria.33
Saddam
Hussein’s regime consistently supported Palestinian suicide attacks, and after
2001 “Iraq clearly differentiated between the categories—which was not customary
among other Arab nations that transferred funds to the families of the
shahids—to distinguish between the financial grant transferred to the family
of a ’regular’ martyr and that bestowed upon a martyr who died during a
suicide mission.”34
Iraqi authorities apparently also made a fine distinction
between suicide and no-escape attacks.35
Saudi Arabia has also supported
radical Islamic causes and provided financial assistance to Hamas and the
families of martyrs, but they apparently do not discriminate between suicides
and other martyrs who die in battle.
EXPLANATIONS
Both Gambetta and chapter author Ricolfi reject the idea of a general explanation
for suicide attacks.36
Considering how difficult it is to define the
concept, modesty may be the wisest course of action. It is possible to discern
the outlines of a preliminary framework of analysis, although answers
31 Khosrokhavar, Suicide Bombers, 79.
32 For details and a positive answer, see Rogelio Alonso and Fernando Reinares, “Maghreb Immigrants
Becoming Suicide Terrorists: A Case Study on Religious Radicalization Processes in Spain,” in Pedahzur,
ed., Root Causes, 179–97. Yoram Schweitzer, in “Al-Qaeda and the Global Epidemic of Suicide Attacks,”
in Pedahzur, ed., Root Causes, Ibid., 132–51, 144–45, also says yes, due to their plans. Pedahzur, however,
does not include the bombings in his list.
33 Shay, The Shahids, 141–87. He often does not distinguish between suicide and non-suicide attacks,
instead referring generally to all terrorism.
34 Ibid., 156.
35 Ibid., 157–58.
36 Luca Ricolfi, “Palestinians, 1981–2003,” in Gambetta, ed., Making Sense, 77–129, esp. 104–5, 259.
Explaining Suicide Terrorism: A Review Essay 141
to central questions will vary. Not surprisingly, most accounts focus on the
interaction of individuals, organizations, and societies.37
There is an emerging consensus that suicide attacks are instrumental or
strategic from the perspective of a sponsoring organization that represents the
weaker party in an asymmetrical conflict. They serve the political interests of
identifiable actors, most of whom are non-states opposing well-armed states.
The method is mechanically simple and tactically efficient and it possesses
high-symbolic value as well as versatility. The suicide bomber can gain access
to well-guarded targets, kill a lot of people, terrify the enemy, and signal
resolve and dedication to a cause. Common wisdom holds that such a strategy
cannot be deterred because of the perpetrator’s willingness to die.
Presumably suicide attacks also mobilize sympathetic constituencies and
attract recruits and financial support. The death of the perpetrator is thought
to legitimize the action. These influences are reciprocal; organizations use
suicide attacks to generate support, but at the same time they respond to
popular demand. Religion can motivate both support and participation, because
of its emphasis on redemption and martyrdom, but it is not required.38
When present, it is often mixed with nationalism and communal solidarity.
There is no single profile of the individual suicide attacker, even within
the same context. A range of emotions, including pride, anger, rage, frustration,
humiliation, shame, hopelessness, and despair, can be powerful driving
forces, as is the desire for revenge or personal glory and honor. Suicide attacks
can demonstrate power and overcome feelings of helplessness: They
convey the message “we may be materially weak, but we are more powerful
because we do not fear death.” Loyalty to a group, leader, or comrades and
family further strengthens individual commitment.
EFFECTIVENESS
Suicide attacks are presumed to pay off for the political organizations that
use them in two possible ways: by coercing an adversary and/or by giving
an organization an advantage over its rivals in terms of support from constituencies.
Many authors also think that the method is a last resort, used
when other means have failed.
37 See Israeli, Islamikaze. Israeli’s lengthy disquisition in Islamikaze is an exception. He blames
Islamic ideology and society as a whole. His book is basically a condemnation of the Muslim world.
On frameworks of causation, see Assaf Moghadam, “The Roots of Suicide Terrorism: A Multi-Causal
Approach,” in Pedahzur, ed., Root Causes, 81–107.
38 The LTTE is usually cited as the preeminent secular case, but Leonard Weinberg reminds readers
that the Viet Cong also practiced suicide attacks, although they were a relatively minor tactic. See Leonard
Weinberg, “Suicide Terrorism for Secular Causes,” in Pedahzur, ed., Root Causes, 108–21. In Kalyvas and
Sanchez-Cuenca, “Killing Without Dying,” in Gambetta, Making Sense, they also cite anarchist attacks in
Russia in the early-twentieth century.
142 M. Crenshaw
Coercing the Adversary
Pape makes a strong claim that campaigns of suicide terrorism are successful
in compelling democracies to withdraw from the military occupation of a
territory that terrorists consider their own.39
“Every group mounting a suicide
campaign over the past two decades has had as a major objective. . .coercing
a foreign state that has military forces in what the terrorists see as their
homeland to take those forces out.”40
“The target state of every modern suicide
campaign has been a democracy,” because democracies are vulnerable
to coercion.41
Presumably, suicide attacks magnify “the coercive effects of
punishment”: they are more destructive than other types of terrorism, they
effectively signal future attacks, and they deliberately violate norms.42
This argument relies on a rather arbitrary determination of what constitutes
a campaign, which is especially significant because Pape excludes
isolated attacks.43
It is possible that incidents that would be inconvenient to
explain are thus omitted. For example, although a campaign can consist of
only one incident,44
the 1995 suicide truck bombing of the Egyptian Embassy
in Islamabad is treated as isolated45
despite the fact that Egyptian Islamic Jihad
(EIJ, later merged with Al Qaeda) organized a series of attacks against
the Egyptian government in the 1990s, one involving an unsuccessful suicide
attack against a former interior minister. Furthermore, Pape lumps different
groups together in a common campaign, such as Iraqi rebels and does not
distinguish between Al Qaeda and local affiliates or start-ups.
Another ambiguous case is the Armed Islamic Group (Groupe Islamique
Arm´ee, or GIA) in Algeria. If the GIA practiced suicide terrorism it would
undercut Pape’s thesis, since the Algerian government was neither democratic
nor foreign. The record, however, is in dispute. Reuter says the GIA never
perpetrated a suicide mission, and the Memorial Institute for the Prevention
39 Pape, Dying to Win, 4–38. Note that only six democracies are included in the analysis. Goodwin
also directly challenges Pape’s thesis, which has received a great deal of attention. He notes that it is
based on two claims that cannot be shown to be true: that all campaigns of suicide terrorism are aimed
at ending foreign occupation and that all such occupations provoke suicide terrorism. See also Assaf
Moghadam, “Suicide Terrorism, Occupation, and the Globalization of Martyrdom: A Critique of Dying to
Win,” Studies in Conflict & Terrorism 29 (2006): 707–29.
40 Pape, Dying to Win, 21.
41 Ibid., 45.
42 Ibid., 28–29.
43 Campaigns are described as organized and coherent, but not further specified. See a similar critique
of Pape and Bloom in the review by Clark McCauley, “The Politics of Suicide Terrorism,” The Middle East
Journal 59, no. 4 (Autumn 2005): 663–66. Brym and Araj comment that Pape’s description of the entire
second intifada as one campaign lumps too much together. See Brym and Bader Araj, in “Suicide Bombing
as Strategy and Interaction: The Case of the Second Intifada,” Social Forces 84, no. 4 (June 2006): 1976.
44 Pape, Dying to Win, 44, 100, 257. For example, the Sikhs versus India, although Pape admits that
this is a borderline case. See ibid., 154 ff. Pape lists two Hamas attacks as a campaign in 1994, four in
1996, and three in 1997. Overall Hezbollah is credited with three separate campaigns, the LTTE also with
three, the PKK with two, and Hamas and Hamas/PIJ with five.
45 Ibid., 264.
Explaining Suicide Terrorism: A Review Essay 143
of Terrorism (MIPT) Terrorism Knowledge Base does not list any missions.46
However, the GIA certainly tried. In 1994, the group threatened to crash a
hijacked airliner in Paris, but a French commando team seized the plane
while it was on the ground in Marseille. Furthermore, Pedahzur argues that
the GIA organized at least two suicide attacks in 1995.47
Shay lists four attacks
between 1994 and 1998, including the hijacking.48
Kalyvas and SanchezCuenca
assert that although the GIA did not hesitate to massacre civilians, they
only engaged in one suicide attack. Pape also lists one attack as an isolated
incident and not a campaign.49
Gambetta says that the GIA committed one
attack and then renounced the practice.50
Did suicide campaigns actually pay? Pape qualifies his argument by saying
that suicide terrorism is successful only when occupying powers have
“limited or modest” goals as opposed to those “central to their wealth or se-
curity.”51
He concludes that six of thirteen completed campaigns from 1983
to 2001 resulted in “no change” in the foreign occupation (four other campaigns
were considered ongoing).52
However, he argues that “even a 50
percent success rate is remarkable” since, in general, coercion only works
in international politics a third of the time.53
His conclusion that Sri Lanka
would grant Tamil autonomy in 2001 was premature, so by his own measure
the success rate is under 50 percent.54
Even if we accept (1) the idea of a
campaign as the unit of analysis, (2) the coding of the outcomes of completed
campaigns, and (3) the calculation of advantageous cost-benefit ratios
as something under fifty-fifty, the odds do not seem encouraging, especially
considering the risks.
The fact, for example, that Hamas claimed that Israel withdrew because
of their attacks does not mean that it is true or even that Hamas leaders believed
that it was true.55
Jon Elster notes that Pape’s argument that suicide
attacks in 1994–95 caused Israel to speed up its withdrawal from the occupied
territories is not persuasive because it “rests exclusively on ambiguous
statements by the then Prime Minister Rabin and self-serving statements by
Hamas spokesmen,” although he finds it not implausible.56
However, Brym
and Araj disagree with Pape entirely: “During the second intifada, the results,
46 Reuter, My Life is a Weapon, 10.
47 Pedahzur, Suicide Terrorism, 105.
48 Shay, The Shahids, 94–96.
49 Pape, Dying to Win, Appendix.
50 Gambetta, ed., Making Sense, 289.
51 Pape, Dying to Win, 75.
52 Ibid., 40.
53 Ibid., 65.
54 Moghadam, “Defining Suicide Terrorism,” in Pedahzur, ed., Root Causes. He finds that the success
rate is significantly lower, 24 percent. He disagrees with Pape’s judgment that Hamas gained important
concessions from Israel.
55 Pape, Dying to Win, 66–73.
56 John Elster, “Motivations and Beliefs in Suicide Missions,” in Gambetta, ed., Making Sense, 249.
144 M. Crenshaw
objectives and precipitants of suicide bombing reveal little of the strategic
logic that, according to Pape, lies at its core.”57
They point instead to revenge
and retaliation in response to specific Israeli actions. In their view,
suicide tactics had high costs (arrests, assassinations of leaders, collateral
damage, destruction of houses, roadblocks, and checkpoints) that were not
offset by Israeli concessions.
Further disputing the contention that suicide attacks are effective instruments
of coercion against democracies, Pedahzur and Arie Perliger argue that
actually more than one-third of suicide attacks from 1982 to June 2005 were
against undemocratic regimes. Examining the five countries that suffered
more than 80 percent of all suicide attacks worldwide, they conclude that
most campaigns are directed at weak democracies.58
Moreover, the proposition
that only democratic foreign occupiers are the target of campaigns
of suicide attacks is particularly questionable with regard to post 9/11 Islamist
violence. For example, Scott Atran comments that “Pape’s basic data,
correlations, and conclusions about the causes of terrorism are problematic,
outdated in the wake of the September 11 attacks and sometimes deeply mis-
leading.”59
Reviewing events in Indonesia, the Philippines, Morocco, Saudi
Arabia, and Turkey, Pedahzur concludes that “for the most part, suicide actions
performed under the flag of the jihad and attributed to Al-Qaeda were,
in effect, local initiatives stemming from Islamic organizational interests operating
within the borders of a given country and whose aspirations principally
amounted to a change of rule in that same country.”60
Another contradictory
piece of evidence is that in 2005, radicals in Bangladesh seeking to establish
an Islamic state organized a series of suicide attacks against the government,
targeting particularly the judicial system.
It is impossible to say whether discrepancies in findings are due to differences
in definitions, data collection, or key variables (for example, campaigns
versus number of incidents). Pedahzur and Pape helpfully provide a
list of events in appendices to their volumes. Without examining each incident,
one cannot reconcile Pedahzur’s worldwide database from March 1977
to February 2004, containing 418 incidents, with Pape’s data from January
1980 to December 2003, listing 315 attacks.61
More research thus needs to be
done to resolve this empirical question and to specify a causal mechanism by
57 Brym and Araj, in “Suicide Bombing,” Social Forces, 1982.
58 Pedahzur and Perliger in “Introduction,” in Pedahzur, ed., Root Causes. However, Gambetta accepts
Pape’s claim that suicide attacks are only used against democracies. See Gambetta, ed., Making Sense,
265.
59 Scott Atran, “The Moral Logic and Growth of Suicide Terrorism,” The Washington Quarterly 29,
no. 2 (Spring 2006): 130. See Moghadam, “Defining Suicide Terrorism,” in Pedahzur, ed., Root Causes.
60 Pedahzur, Suicide Terrorism, 109.
61 Pape, Dying to Win, 3, 4. Pape claims that his database at the Chicago Project on Suicide Terrorism
(University of Chicago) is “the first complete universe of suicide terrorist attacks worldwide” and “the most
comprehensive and reliable survey. . .now available.” Pedahzur’s database is held at the National Security
Studies Centre at the University of Haifa. Neither is yet available online although Pedahzur’s will be
Explaining Suicide Terrorism: A Review Essay 145
which suicide terrorism is effective as coercion against democracies, if this is
indeed the case. Moreover, if it does not target democracies exclusively, has
it been successful against nondemocracies or weak democracies?
Gaining an Edge Over the Competition
Bloom agrees that resorting to suicide terrorism is meant ultimately to end foreign
occupation and secure autonomy or independence. However, she thinks
that suicide terrorism is also, or even more, attractive because it enhances
an organization’s prestige and gives it an advantage in intra-movement competition
by attracting recruits, publicity, and money.62
Effectiveness, then, is
dependent on mobilizing constituencies in order to dominate the competition
in a local power struggle. Atran tends to support this interpretation,
as do Gupta and Mundra.63
Elster notes generally that game-theoretic and
econometric analyses of the intentions of the organizers of suicide attacks
are “doomed to fail” not only because the organizers may not be as rational
as theory requires, but also because “game theory has very little to
say about the equilibrium outcome of strategic interaction among more than
two actors.”64
A comprehensive model would have to take into account not
only divisions within the parties to the conflict—government and dissident
organizations—but also foreign supporters of both sides.
Bloom’s argument is probably most persuasive regarding the Palestinian
case. Victor, for example, provides anecdotal support for it. Ricolfi, however,
concludes that each of the three Palestinian campaigns that he identifies
had different causes and that internal politics were dominant only during
the post-2000 Al Aqsa intifada wave, when secular and religious organizations
competed for control over the Palestinian insurgency.65
The recourse to
accessible shortly through the University of Texas. In the meantime, he will provide the data in electronic
form to researchers on request. Several authors (for example, Ricolfi, “Palestinians,” in Gambetta, Making
Sense, 309, Hopgood, “Palestinians,” in Gambetta, Making Sense, and Moghadam, “Defining Suicide
Terrorism,” in Pedahzur, ed., Suicide Terrorism, 15 find that Pape underestimates numbers of attacks. For
example, Pape lists 110 suicide attacks by Palestinians vs. Israelis from April 1994 to December 2003,
whereas Brym and Araj, who use Arabic press accounts, count 158 attacks between 1993 and July 2005.
Because of the twin difficulties of measuring what it is to be explained and finding accurate and comprehensive
data, Ricolfi created a new integrated database. In his view, the main databases are incomplete
and differentially selective. They include the Institute for Counter Terrorism (ICT) in Israel, the Terrorism
Research Center, and the MIPT. The Terrorism Research Center’s website www.terrorism.com states that it
was founded in 1996 and maintains its database as a public service. It is located somewhere in northern
Virginia. It does not list the names of its Board of Directors. Ricolfi discusses his methodology in great
detail in an appendix to the chapter.
62 Bloom, Dying to Kill, 3, 16–17, 76–77.
63 Dipak Gupta and Kusum Mundra, “Suicide Bombing as a Strategic Weapon: An Empirical Investigation
of Hamas and Islamic Jihad,” Terrorism and Political Violence 17, no. 4 (2005): 573–98. Pape,
however, disagrees.
64 Elster, “Motivations and Beliefs,” in Gambetta, ed., Making Sense, 257–58.
65 Ricolfi, “Palestinians,” in Gambetta, ed., Making Sense, 94, 96, 98, 104.
146 M. Crenshaw
suicide missions was cyclical. He also points out, quite rightly, that Palestinian
factions cooperate as well as compete, a point that Pape also makes.66
Moreover, Bloom’s analysis of Sri Lanka, one of her three major case
studies (in addition to Palestinians and the PKK), is contradictory. She indicates
that when the LTTE began its suicide attacks in 1987, the other Tamil
militant groups had largely been destroyed and she accepts the conclusion
that the systematic elimination of rivals culminated in 1986, a point that Reuter
confirms and Pape also supports.67
In a footnote, she says that the process
of competition began then.68
If her interpretation of organizational strategy
is correct, the suicide attacks should have occurred between 1983 and 1987,
when the LTTE faced the most competition. In fact, whereas Pedahzur lists the
first LTTE attack as 1987,69
Pape says that the actual campaign only began in
1990, which places it after the withdrawal of Indian peacekeeping forces and
well after the elimination of rivals.70
Hopgood agrees that it is not entirely
sure that the attacker intended to die or that the LTTE had settled on the tactic
of suicide attack by 1987.71
(One problem complicating research is that the
LTTE rarely comments or accepts responsibility. The government, on the other
hand, tends to over attribute.)
Furthermore, Al Qaeda’s actions are hard to explain in terms of competition
with rivals. Yoram Schweitzer is more convincing in arguing that
Al Qaeda’s ultimate goal is to pressure the West into changing its policies
toward the Muslim world, and to that end the organization deliberately promotes
suicide terrorism in order to become a role model for Islamic militancy
worldwide.72
Sacrificing one’s life has deliberately been made into the exclusive
symbol of global jihad, intended to inspire imitation in affiliate groups
and thus augment organizational resources. It furthers transnational cooperation,
coordination, and unity of purpose, not local competition.
If suicide attacks are effective in either or both of these ways, why
are they infrequent? Kalyvas and Sanchez-Cuenca note that suicide attacks
can be counterproductive if they kill civilians indiscriminately. They suggest
that international audiences regard suicide bombings as “particularly repug-
nant.”73
They conclude that “the trend in suicide bombing that has been
observed in the early years of the twenty-first century may be more of a
temporary aberration than a sign of things to come.”74
Hafez contributes a
66 Ibid., 100.
67 Reuter, My Life is a Weapon, 159. Pape, Dying to Win, 139–40.
68 Bloom, Dying to Kill, 217.
69 Pedahzur, Suicide Terrorism, 73, 242.
70 Pape, Dying to Win, 254, Appendix I.
71 Hopgood, “Tamil Tigers, 1987–2002,” in Gambetta, ed., Making Sense, 50.
72 Schweitzer has also constructed his own database at the Jaffe Center for Strategic Studies at Tel
Aviv University. Atran, too, has a separate database.
73 Kalyvas and Sanchez-Cuenca, “Killing Without Dying,” in Gambetta, Making Sense, 218.
74 Ibid., 232.
Explaining Suicide Terrorism: A Review Essay 147
practical discussion of restraint during the first Palestinian intifada: it was difficult
to acquire weapons; it was easy to attack Israeli military forces because
they were omnipresent; Palestinian casualty rates were moderate; the public
lacked access to news coverage; and the PLO provided effective leadership.75
To generalize, lack of resources for high levels of destruction combined with
ample opportunity for low-level and low-risk resistance, an absence of inflammatory
rhetoric, and control over the use of violence by local authorities
inhibited suicide terrorism.
The Issue of Timing
Is suicide terrorism a weapon of last resort? Pape says that it is not the
province of “tiny bands of ordinary terrorists” but of national liberation movements
that have failed at guerrilla warfare.76
Bloom asserts that “non-state
actors tend to resort to atrocities in the second iteration (or more) of conflict
after the other strategies have failed to yield the desired results, and when
faced with a hurting stalemate.”77
However, Gambetta counters that the last
resort hypothesis does not apply universally.78
In some cases, such as the Chechen resistance, suicide terrorism did
appear to be a desperate last measure.79
Similarly, Hafez argues that Palestinian
suicide terrorism followed the low-intensity violence of the first in-
tifada.80
Note, however, that the point at which Palestinians began suicide
attacks is disputed. It could be that suicide attacks began in 1993 (according
to Pedahzur, Shay, and Hafez) or it could be later, in April of 1994 (Pape).81
Pedahzur finds that both the LTTE and PKK “committed suicide attacks
in situations where they felt powerless or were pushed into a corner, most
often as a result of protracted military pressure.”82
However, Hopgood sees
the LTTE’s actions as a strategic decision—a way to reach difficult targets and
a response to the government’s military offensives—and not a last resort.
75 Hafez, Manufacturing Human Bombs, 53–57.
76 Pape, Dying to Win, 93.
77 Bloom, Dying to Kill, 89. With regard to last resort, she distinguishes between on-battlefield and
off-battlefield use; in the latter, suicide attacks are ends in themselves rather than a last resort tactic of
desperation. This premise requires further explanation.
78 Gambetta, ed., Making Sense, 261.
79 See Kalyvas and Sanchez-Cuenca, “Killing Without Dying,” in Gambetta, Making Sense, 218. They
suggest that the timing is evidence of recognition of the method’s counterproductiveness, not its effectiveness.
Suicide attacks were a sign that the Chechens were losing the war.
80 Hafez, Manufacturing Human Bombs, 53.
81 April 1993 according to Pedahzur and Shay. See Pedahzur, Suicide Terrorism, 54–55 and Shay,
The Shahids, 52. Hafez lists the first attack as September 1993. See Hafez, Manufacturing Human Bombs,
79. Shay says that before April 1993 there were several unsuccessful attempts, which he does not include.
See Shay, The Shahids, 88. Pedahzur notes that Hamas publicly advocated their use as early as 1989. See
Pedahzur, Suicide Terrorism, 55. Ricolfi notes some sporadic attacks earlier. See Luca Ricolfi, “Palestinians,
1981–2003,” in Gambetta, ed., Making Sense, 11, 77–129.
82 Ibid., 96.
148 M. Crenshaw
There are other well-known cases where suicide attacks did not seem
to be a last ditch measure after alternatives failed. For example, it is most
likely that Hezbollah launched its campaign with suicide attacks in 1983, a
year after the Israeli invasion, although, yet again, the facts are unclear. Many
accounts (for example, Pedahzur, Pape, and Ricolfi) list the first suicide attack
in Lebanon as a car bombing of the Iraqi Embassy in 1981 (although
Pape says that Hezbollah’s first attack was in 1982). Yet Nasra Hassan, who
has researched the case for a forthcoming book, finds the evidence still inconclusive
as to whether it was a suicide attack at all.83
Ricolfi also says that
the Israeli military headquarters in Beirut was bombed in 1982, killing 75 soldiers
and 15 Palestinian prisoners.84
Shay recounts: “The first suicide attack
perpetrated by the Hizballah against an Israeli target was on November 4,
1983 when a car bomb driven by a suicide terrorist exploded near the IDF
headquarters in Tyre.”85
It killed 28 soldiers and 33 detainees.86
Pedahzur
agrees with Shay.87
The authors’ discussion of the onset of Al Qaeda violence is also extremely
problematic. In 1995 and 1996, the bombings of two American military
installations in Saudi Arabia made a large impact on American policy.
Bloom, Pape, and Pedahzur count both as suicide attacks. Pedahzur blames
Al Qaeda for the first attack; Bloom blames them for the second;88
and Pape
attributes both to them. If contemporary press reports and government documents
are right, neither was a suicide attack.89
Only the first might plausibly
be linked to Al Qaeda. The second, the Khobar Towers bombing, was
the work of Saudi Hezbollah and connected to Iran and to Hezbollah in
Lebanon. It proved a major impediment to Bill Clinton’s administration’s effort
to improve relations with Iran. It is most likely that the suicide bombings
of the embassies in Kenya and Tanzania in 1998 initiated Al Qaeda’s offensive
against American targets.
In Iraq, suicide tactics and insurgency have been simultaneous, not sequential.
Suicide attacks are also on the rise in Afghanistan, associated with
the resurgence of the Taliban. In 2005, there were 27, and in 2006, 139 such
incidents.90
83 Nasra Hassan, personal communication.
84 Ricolfi, “Palestinians,” in Gambetta, ed., Making Sense, 86–87.
85 Shay, The Shahids, 41.
86 Ibid., 40.
87 Pedhahzur, Suicide Terrorism, 48.
88 Bloom, Dying to Kill, 166.
89 Elaine Sciolino, New York Times, 14 November 1995. She reported that Clinton administration
officials said that the 1995 bomb, placed in a van, was detonated by remote control. Douglas Jehl, New
York Times, 16 November 1995. He quotes the American Ambassador as saying that the device was set
off by a timer. Steven Erlanger, New York Times, 27 June 1996. He quotes eye witnesses who said that the
driver of the truck that carried the bomb in the Khobar Towers attack fled in an accompanying car. The
U.S. government’s annual report, Patterns of Global Terrorism, does not refer to either as a suicide attack.
90 New York Times, 17 January 2007.
Explaining Suicide Terrorism: A Review Essay 149
SOCIAL SUPPORT
Suicide attacks are not generally supposed to occur in the absence of public
support for them. This assumption guides the U.S. war of ideas, which stresses
public diplomacy as a component of the global war on terrorism. However,
Kalyvas and Sanchez-Cuenca argue that an organization’s sensitivity to public
opinion varies. They propose that suicide terrorism occurs when there is
minimal or maximal popular support, thus where there is dependency or
indifference, not in intermediate cases.
A key question is how religion affects support. The burning issue today,
of course, is the relationship between Islam and suicidal violence, framed
as martyrdom. Other than Israeli and Shay, the authors reject the idea that
religion alone is a necessary or sufficient cause for suicide terrorism.
In Lebanon, for example, suicide attacks were not the exclusive province
of Hezbollah. Secular political parties were equally engaged. Reuter describes
1980s Lebanon: “Religion, patriotism, and the willingness to sacrifice oneself
can easily blend together when the challenge is to resist a foreign occupier
of a different faith.”91
Pape argues that it is not religious doctrine per se,
but religious difference between occupier and occupied.92
“The taproot of
suicide terrorism is nationalism,” but it is inflamed by religious difference.93
In such zero-sum conflicts, occupied populations support suicide terrorism
because the consequences of a successful conquest by a foreign invader
would be too sweeping. It is easy to demonize the enemy, killing enemy
civilians is morally acceptable, and suicide can be framed as martyrdom.94
Later Pape acknowledges that religious difference is not a “hard, necessary
condition for suicide terrorism.”95
In its absence, terrorists will still be
tempted because the technique works, but the campaigns will apparently
be short, as in the case of the PKK versus Turkey. Pedahzur disagrees: the
PKK’s restraint was due not to religious affinity with Turks but to the PKK’s
ideological remoteness from its Kurdish constituency.96
Turkish army offensives
made the PKK unpopular. Bloom thinks that lenient Turkish policies
moderated the suicide campaign.97
Atran finds, however, that Salafi ideology is independently important
to suicide terrorism.98
He contends that Pape underestimates the influence
of Salafism (in part because his analysis concluded in 2003, when both its
91 Reuter, My Life is a Weapon, 60.
92 Pape, Dying to Win, 22.
93 Ibid., 79–80, 88–92.
94 Ibid., 88–92.
95 Ibid., 167.
96 Pedahzur, Suicide Terrorism, 96.
97 Bloom, Dying to Kill, 92.
98 Atran, “The Moral Logic and Growth of Suicide Terrorism,” 32–34.
150 M. Crenshaw
appeal and the incidence of suicide attacks were growing rapidly). Atran also
questions Pape’s focus on Salafist ideology in the countries where terrorist
groups might have originated (such as Morocco for the 2004 bombers in
Spain), rather than its influence in diasporas.
Interpretation of the role of religion in the Sri Lankan case is also disputed.
Hopgood contends that “religion is not a feature which can explain
the emergence of SMs in the Tamil case.”99
“Given the absence of any powerful
transcending political or religious ideology, and the ubiquitous statements
of attachment only to the cause of national liberation, the best explanation
for the adoption of SMs at the collective level is as a tactic in a wider military
strategy for victory in a war of uneven force.”100
Pedahzur and Reuter
agree with Hopgood.101
Pape attributes the LTTE’s suicide terrorism to “fear
of religious persecution” and contends that “the most prominent factor driving
Tamil community support for individual self-sacrifice is fear of Buddhist
extremism.”102
This is despite the fact that “the LTTE is a secular group that
disavows Tamil religious motivations as a driving force behind national re-
sistance.”103
Pape argues, for example, that the LTTE’s restraint toward Indian
forces in Sri Lanka was due to a shared religion, although he admits that
the assassination of Rajiv Gandhi was an exception.104
Contemporary sectarian
violence within the Muslim world, primarily in Iraq but also in Pakistan,
raises more difficult questions about the role of religion. Are Sunni and Shia
considered different religions? Would the Salafist takfiri strain of condemning
fellow Sunnis as apostate qualify as a religious distinction? Why have Sunni
factions used suicide attacks in Iraq when Shi’ite militias have not? Furthermore,
if nationalist resistance to foreign occupation by a democracy is the
cause of suicide terrorism, why is it that Al Qaeda in Mesopotamia and Ansar
al-Sunnah are the main actors in Iraq? Why are most of the suicide bombers in
Iraq foreign? Why is the tactic directed primarily against Shia civilians rather
than the occupying military forces?
Another complication for understanding social support for suicide attacks
is the question of what foreign occupation means. Bloom argues that
an occupier’s specific strategy, particularly whether or not the use of force
causes excessive civilian casualties, helps determine whether suicide attacks
resonate with a given population.105
Pape is much more sweeping: it is the
fact of occupation, not the degree of severity measured by civilian deaths.106
99 Hopgood, “Tamil Tigers, 1987–2002,” in Gambetta, ed., Making Sense, 76.
100 Ibid., 65–66.
101 Pedahzur, Suicide Terrorism, 24. Reuter, My Life is a Weapon, 13.
102 Pape, Dying to Win, 140, 146.
103 Ibid., 149.
104 Ibid., 151, 153–54.
105 Bloom, Dying to Kill, 82, 91.
106 Pape, Dying to Win, 69.
Explaining Suicide Terrorism: A Review Essay 151
At one point, he specifies the presence of “heavy combat troops,”107
yet
later he contends that troops need not be physically present: American occupation
can be either a military presence or “an explicit or widely understood
security guarantee that could be implemented using its [American]
forces in an adjacent country.”108
He excludes advisers and joint training. In
his view, Saudi Arabia, Oman, the United Arab Emirates, Bahrain, Kuwait,
Qatar, Turkey, Afghanistan, and Uzbekistan qualify as territory occupied by
the United States, but Jordan, Yemen, Pakistan, Turkmenistan, and Tajikistan
do not. He categorizes attacks on U.S. allies in the Iraq War as resistance to
U.S. occupation. In fact, all attacks associated even loosely with Al Qaeda or
with jihadism are defined as targeted against the United States (for example,
attacks on tourist targets by Jemaah al Islamiya in Indonesia).
Atran is highly critical: “It is quite a stretch to identify the common thread
as a secular struggle over foreign occupation of a homeland, unless ‘secular’
covers transcendent ideologies, ‘foreign occupation’ includes tourism, and
‘homeland’ expands to at least three continents.”109
Goodwin also disagrees,
arguing that the issue is not military occupation but U.S. support for unpopular
local governments.110
Neither theory explains cases such as Bangladesh in
2005, which involved local opposition to local authority in the interest of
instituting a different form of government.
Bloom suggests that social support depends not just on the nature of the
occupation but also on “how the tactic [of suicide attacks] is used, against
whom, and for what purpose,” particularly when “hatred for the other side
is very high.”111
When is hatred high? Bloom says that “the explanation is
somewhat endogenous to the cases and results from a variety of personal,
economic, structural and organizational issues” such as the “interplay of domestic
politics and external factors like the ongoing conflict, a ’hurting stalemate’
or the counter-terror strategies employed by the opposing side.” “The
success of the strategy. . .will depend on the existing domestic political backdrop,”
which “explains both how suicide terror becomes popular in some
cases and why it is rejected or repudiated in others.”112
She adds that suicide
bombing spreads in countries where the population is receptive to terrorists
targeting civilians, but this hypothesis cannot explain suicide attacks that
do not target civilians or why non-suicide attacks on civilians would not be
107 Ibid., 79.
108 Ibid., 108–9.
109 Atran, “The Moral Logic,” The Washington Quarterly 29, no. 2 (Spring 2006): 134.
110 Goodwin, “What Do We Really,” Sociological Forum 21, no. 2 (June 2006): 322.
111 Bloom, Dying to Kill, 81.
112 Bloom, Dying to Kill, 90. In the Palestinian case, she refers to deadlock when the peace process
appeared to be stalled. See ibid., 25, 26, 28. For Sharon’s tactics of attacking civilians, see ibid., 26–27.
For the unpopularity and weakness of the Palestinian Authority and Arafat, see ibid., 34. For economic
hardship, see ibid., 28. For lack of success with other tactics, see ibid., 27. For an emphasis on martyrdom
that captured the Palestinian imagination, see ibid., 29.
152 M. Crenshaw
equally gratifying and thus equally contagious.113
Furthermore, as Kalyvas
and Sanchez-Cuenca emphasize, suicide attacks can occur where there is
little to no popular support.
Pedahzur is more inclined to attribute support to deliberate cultivation
by an organization rather than self-generated demand: “Social support for
suicide terrorism. . .is very rarely a grass-roots phenomenon. In fact, it is a
highly calculated top-down phenomenon.”114
Nevertheless, he thinks that
communities that have experienced a long and painful conflict with a more
powerful enemy eventually react to perceived inferiority and the failure of
other efforts by supporting suicide attacks.115
It is not so much religion as
attitudes acceptant of death, stemming from hopelessness after long suffer-
ing.116
He stresses that political organizations give populations an outlet for
expressing feelings of deprivation, injustice, despair and hostility. An enemy’s
brutal response reinforces these feelings.
Hamas deliberately framed suicide attacks in terms of a culture of martyrdom
that was previously unfamiliar to Palestinian society. Hafez finds that
Palestinians came to venerate martyrdom because of a “confluence of perceived
threats and a sense of victimization.”117
The public sought both revenge
and empowerment in response to harsh Israeli actions. He also emphasizes
that while Islam presented a cultural opportunity to frame suicide
attacks as martyrdom, the failure of the secular Palestinian Authority (PA) and
religious leaders to counter this framing gave Hamas and Palestinian Islamic
Jihad the political opening they sought. Like Khosrokhavar, Hafez is sharply
critical of the PA’s failure to halt or even to condemn attacks on Israeli civilians.
Oliver and Steinberg present a vivid picture of “the world of the suicide
bomber” in Palestine.118
Their study of Hamas is memoir, rather than political
analysis, focused on the scripts, rhetoric, texts, documents, posters, martyr
cards, graffiti, video statements, and films that institutionalized and inspired
a culture of martyrdom. They discuss a barrage of media, permeated by
references to historical myths and narratives. They describe rituals and images
as well as discourse (poems, songs, and speeches). They see striking fear in
the enemy as an end in itself, not a path to a political objective.
113 Ibid., 192.
114 Pedahzur, Suicide Terrorism, 159.
115 Ibid., 31, 159.
116 Ibid., 180. Some authors (for example, Pape) also note high rates of suicide in Sri Lanka, indicating
a societal tolerance of death as a solution to problems.
117 Hafez, Manufacturing Human Bombs, 61.
118 Not all reactions to the work have been positive. See Lori A. Allen, Review of “Dying to Win:
The Strategic Logic of Suicide Terrorism” and “Dying to Kill: The Allure of Suicide Terror” and “The Road
to Martyrs’ Square: A Journey into the World of the Suicide Bomber,” Journal of Palestine Studies 35,
no. 2 (2006): 110–13. Reviewing this book, Allen calls it “an extended anti-Palestinian rant” that portrays
“Palestinians as backwards, dirty, murderous and sadistic, offering no account of the history or presence
of Israeli occupation.”
Explaining Suicide Terrorism: A Review Essay 153
Social support might be due to a variety of factors: religion, a mixture of
religion and nationalism, foreign occupation in general (defined narrowly or
broadly), specific practices of opposing governments (for example, excessive
brutality and civilian casualties), deliberate cultivation by political organizations,
the failure of other organizations to effectively counter the tactic, or
long experience of suffering and deprivation (which could be related to the
length and severity of the conflict and perhaps to the failure of alternatives).
We do not know how much weight to accord each factor or how we might
measure them.
INDIVIDUAL PARTICIPATION
Very few of those exposed to even the most intense pressures become
suicide attackers. Schweitzer estimates that there were a total of 1,300
individual suicide terrorists between 1983 and July 2005.119
Why would an
individual choose to die killing others for a political cause? Such behavior is
exceptional rather than routine. Small numbers of recruits may be one of the
most important explanations for the comparatively low frequency of suicide
attacks, despite their notoriety. Analysis of individual motivation leads to
comparisons in other contexts; for example, with the Japanese kamikaze,
who left behind extensive correspondence, and with individuals who have
practiced self-immolation, or simply with those who have died heroic deaths
in battle.120
The search for a profile of the suicide attacker seems to be as
hopeless as the search for a typical terrorist. Organizations channel diverse
personal motivations.
Clearly the act is not just about dying and killing. The expectation of
gaining status and respect as a martyr for the cause is important, so that individual
action is linked to anticipation of both popular approval and collective
political success. Gambetta, Pedahzur, Pape, and others refer to altruism. Sacrifice
for the cause is both personally redemptive and a mark of honor, a way
of becoming a hero and part of an exalted elite, as much as a way of seeking
death. It involves an aspiration to live on after death and to give lasting
meaning to an otherwise insignificant or disappointing life. In some cases,
the choice to become a martyr is followed by elaborate rituals that reinforce
commitment and prevent backsliding.
Pedahzur finds the individual level too complex to explain in psychological
terms. Elster also concludes that attackers may not act out of a stable
set of motivations based on beliefs and that if they do, we may not be able
119 Schweitzer, in “Al-Qaeda and the Global Epidemic,” in Pedahzur, ed., Root Causes, 134.
120 See Peter Hill, “Kamikaze, 1953–5” and Michael Biggs, “Dying Without Killing: Self-Immolations,
1963–2002” in Gambetta, ed., Making Sense and Reuter’s chapter “Bushido Replaces Allahu akbar: The
Japanese Kamikaze,” in Reuter, My Life is a Weapon. It is curious that none of the authors reviewed here
discusses the IRA or Turkish hunger strikers.
154 M. Crenshaw
to identify them.121
Motivations are highly context-dependent, and critical
information is lacking. Pedahzur categorizes suicide attackers as either members
of an organization or social network or volunteers for a specific mission.
The first type is motivated by commitment to the group and its cause, and
the second is driven by a personal or family crisis.122
Isolation, indoctrination,
and peer pressure also figure in decisions.123
Suicide attackers do not
have to be highly religious or ideological or to have been active members
of an organization. Palestinian suicide bombers of the 1990s were likely to
be individuals devoted to the organization and its goals, whereas after 2000
individuals volunteered because of the acute political crisis.124
After 2002,
many attacks were local initiatives, not directed by the leadership, due to
intense Israeli pressure.125
Only in Sri Lanka was there a professional cadre of Black Tigers trained
and prepared for suicide or other high-risk missions, according to Pedahzur,
who says that “the majority of terrorist organizations do not recruit candidates
specifically for suicide missions.”126
As a chosen elite, they are carefully
screened. In his view, mental preparation is key in this case.127
As in all
group situations, socialization and peer pressure are relevant. Hopgood also
describes the careful process by which Black Tigers are formed.
Hafez offers propositions drawn from the Palestinian case, which are potentially
applicable elsewhere.128
Religious redemption is linked to identity
and supported by the following: the concept of jihad as individual obligation,
selective references to religious texts, historical narratives based on the
life of Muhammad, euphemistic labeling (suicide becomes martyrdom), and
the use of ritual and ceremony. The martyr’s videotaped statements stress
redemption, both personal and collective, the necessity of martyrdom, and
reward in the afterlife. Hafez stresses that nationalism is also critical, interwoven
with religion and community solidarity; the individual intends his or
her act to arouse the consciousness of the people.
Khosrokhavar agrees: “Martyrdom in Iran, Algeria and Palestine obeys
an internal logic born of the frustrated ambition to have a nation whose existence
has been denied.”129
Religion can frame loss of dignity as a sin, and
then offer redemption through martyrdom.130
Like Pedahzur, he emphasizes
121 Elster, “Motivations and Beliefs,” in Gambetta, ed., Making Sense, 256–57.
122 Pedahzur, Suicide Terrorism, 125–51.
123 Ibid., 181.
124 Ibid., 153–54.
125 Ibid., 170.
126 Ibid., 170.
127 Ibid., 171.
128 See also Hafez, “Dying to be Martyrs: The Symbolic Dimension of Suicide Terrorism,” in Pedahzur,
ed., Root Causes, 54–80.
129 Khosrohkavar, Suicide Bombers, 109.
130 Ibid., 133.
Explaining Suicide Terrorism: A Review Essay 155
loss of dignity and personal or family crisis.131
In Palestine, he notes that
the degradation of everyday life becomes intolerable. Constant Israeli military
presence, boredom, anxiety, and the impossibility of leading a normal
life combine: “The explosive mixture of prison, exile, living on the run,
the instability of life and the constant need to keep moving to avoid being
arrested. . .give rise to an extremism that can even overcome fear.”132
The alternative
to humiliation is utopia,133
and Oliver and Steinberg describe some
intended bombers as ecstatic about their fate. He adds to the mix an element
of defiance—suicide attacks show Israelis that they are vulnerable.
Thus, in explaining Palestinian actions, social scientists seem not to have
gone far beyond what journalist Barbara Victor calls the “fatal cocktail”: religious
doctrine that promises eternal life, deprivation that offers no hope,
nationalism, and the hardships of living under a military occupation.134
Khosrokhavar also addresses the more disturbing case of Al Qaeda or
Islamist inspired suicide outside of conflict zones. Here the experiential intensity
of the Palestinian, Chechen, or Sri Lankan cases is absent. Discrimination,
while real, is milder and the grievance felt represents a more abstract and
mediated form of identification. Why, for example, would British citizens
who have never been to Iraq or Palestine blow themselves up on London
buses? Why would foreigners travel to Iraq or Afghanistan to become suicide
bombers? Globalization, the tensions of diaspora life, and crises in Muslim
societies in the Middle East and the former Soviet Union contribute to identification
with a “transnational neo-umma” or community of believers that
is imaginary and virtual.135
Desperate young men are motivated by feelings
of humiliation, which may be “humiliation by proxy” generated by media
coverage of conflicts involving Muslims, particularly in Palestine. They share
hatred of perceived Western arrogance, immorality, and hostility toward the
Islamic world. As in Palestine, the act of martyrdom is also an act of defiance
in the face of an enemy’s vast superiority. Gambetta suggests that the
Islamic tradition of aggressive martyrdom helps to rationalize suicide attacks
and remove constraints that might otherwise operate.136
Holmes argues that non-religious motives may have driven the nineteen
perpetrators of the 9/11 attacks. In Mohammed Atta’s behavior, Holmes sees
a possible mix of personal frustration, political protest, and religious convic-
tion.137
Marc Sageman also argues that the 9/11 leaders were not particularly
131 Ibid., 113.
132 Ibid., 116.
133 Ibid., 119.
134 Victor, Army of Roses, 117–18.
135 Khosrokhavar, Suicide Bombers, 149–223, esp. chap. 3. Some of his findings are based on interviews
in French prisons. The chapter focuses particularly on Britain and France, with some attention to
the U.S. and to converts to Islam.
136 Gambetta, ed., Making Sense, 293–95.
137 Stephen Holmes, “Al Qaeda,” in Gambetta, ed., Making Sense, 140.
156 M. Crenshaw
religious when they became radicalized.138
Overall, only 13 percent of the
394 jihadists that he analyzed attended madrassas. Along with Khosrokhavar,
Sageman sees the global Salafi jihad as a diaspora phenomenon derived
from “loneliness, alienation, marginalization, underemployment, and exclusion
from the highest status in the new or original society.”139
Individuals
drift to mosques for companionship and then develop a collective religious
identity. Friendship and kinship bonds predate ideological commitment. Personal
resentment is translated via Islamist doctrine into hatred of society and
devotion to the group. Then, as Schweitzer noted, martyrdom becomes the
ultimate test of personal conviction.
Are women a special case? Victor argues that Palestinian women suicide
bombers suffered personal problems that made their lives unbearable.
Marginalized in Palestinian society and denied appropriate social roles, they
were then recruited by male relatives who pushed them to redeem the family
name. A cynical leadership exploited their despair. Bloom agrees that
Palestinian women were empowered by their participation.140
Reuter says the opposite about the LTTE, which is sixty percent women in
its special commando units.141
He thinks the explanation is practical; the few
men of fighting age are needed for combat operations. Women are better at
concealing bombs than fighting. He accepts that female PKK bombers might
have sought liberation from a conservative society.142
Bloom refers to women suicide bombers in the Middle East, Sri Lanka,
Turkey, Chechnya, and, oddly, Colombia.143
In general, she sees men as
motivated by religion or nationalism, but, like Victor, she attributes women’s
behavior to personal reasons.144
Bloom comments that it is striking that “so
many of these women have been raped or sexually abused in the previous
conflict either by the representatives of the state or by the insurgents them-
selves.”145
The evidence presented in support of this charge is weak. It consists
of a rumor reported by a senior advisor to Russian President Putin and
undocumented charges of rape in Sri Lanka.146
The other cases—Palestine,
Lebanon, Turkey—are not substantiated.
138 Marc Sageman, “Islam and Al Qaeda,” in Pedahzur, ed., Root Causes, 122–31, esp. 126.
139 Ibid., 128.
140 Bloom, Dying to Kill, 147.
141 Reuter, My Life is a Weapon, 160.
142 Ibid., 164–65.
143 Bloom, Dying to Kill, 143. As evidence of this assertion she cites a New York Times article that
makes no mention of Colombia.
144 Ibid., 145.
145 Ibid., 143.
146 Ibid., 158–64. Bloom claims that the LTTE killed two potential interviewees while they were en
route to talk to her. See Ibid., 236. Much other correspondence and interviews that she cites are anonymous.
However, Christian Caryl accepts as fact the assertion that Rajiv Gandhi’s assassin had been gangraped
by Sri Lankan soldiers. See Caryl, “Why They Do It,” The New York Review of Books, September 22,
2005. Pape also refers to women suicide bombers in Sri Lanka as reportedly or believed to be victims of
rape. See Pape, Dying to Win, 226, 230. Sexual violence has been prevalent in the Sri Lankan conflict,
Explaining Suicide Terrorism: A Review Essay 157
Speckhard and Ahkmedova dispute both Bloom’s and Victor’s argu-
ments.147
After conducting systematic interviews in Chechnya, they conclude
that “the importance of traumatic loss and avenging deaths of family
members. . .likely plays the greatest role in women’s decisions to become
’martyrs.’”148
Motivations for men and women are the same. Despite often
lurid Russian press reports, Chechyen women did not act in order to redeem
their honor after being raped, nor were they coerced. Speckhard also interviewed
the same Palestinian families that Victor did and found no support for
the claim that women became suicide bombers because they were unable to
fulfill expected social roles or because they had been disgraced.
Individuals are motivated differently. There is no single pattern. The
organization that recruits and directs the suicide bomber remains the most
important agent.
POLICY PRESCRIPTIONS
Considering the level of disagreement about the nature of the problem, it
would be unreasonable to expect a clear consensus about what to do. To
Pape the answer is simple and entirely consistent with his diagnosis of the
problem: offshore balancing. The United States should withdraw from foreign
military occupations, particularly of countries with different religions
(presumably including abandoning security guarantees). However, if weak
oppositions launch suicide attacks because they are effective in compelling
foreign powers to retreat, surely withdrawing rewards them and provides incentives
for future use. Pape says that it does not matter that Al Qaeda’s leadership
would interpret withdrawal from Iraq as appeasement, but Bin Laden
frequently refers to Lebanon and Somalia (and the Soviet withdrawal from
Afghanistan) as encouraging precedents for his anti-American campaign.149
Most analysts think that a major goal of Al Qaeda in Iraq is to create another
base to replace Afghanistan. This solution would simply leave Iraqi
victims to their fate, since there is no reason to think that an American withdrawal,
especially a precipitate one, would end sectarian violence. Hafez
disagrees with the withdrawal option in the Palestinian case. He thinks that
the Palestinian public learned the wrong lessons from Israel’s withdrawal
from south Lebanon in 2000 (and so, too, did Hezbollah) and concludes
but I can find no concrete evidence of a link with suicide bombings. See www.tamilnation.org. A similar
conclusion can be found in Arjuna Gunawardena, “Female Black Tigers: A Different Breed of Cat?” in
Yoram Schweitzer, ed., Female Suicide Bombers: Dying for Equality? (Tel Aviv: Tel Aviv University, Jaffee
Center for Strategic Studies, August 2006).
147 Anne Speckhard and Khapta Ahkmedova, “The Making of a Martyr: Chechen Suicide Terrorism,”
Studies in Conflict & Terrorism 29, no. 5 (2006): 429–92, esp. 468–77. On the other hand, the essays on
Palestinian women in Schweitzer, ed., Female Suicide Bombers tend to support Victor’s interpretation.
148 Speckhard and Ahkmedova, “The Making of a Martyr,” Studies in Conflict, 468.
149 Pape, Dying to Win, 249.
158 M. Crenshaw
that granting major concessions is a mistake. Furthermore, what are we to
make of the fact that suicide attacks have ended in Chechnya although Russia
has not withdrawn? In any case, the United States is not likely to abandon
its commitments to Saudi Arabia, Oman, the United Arab Emirates, Bahrain,
Kuwait, Qatar, Turkey, Afghanistan, and Uzbekistan, as Pape’s prescription
would require.
The authors mostly agree that governments are not defenseless against
the tactic. They can protect their home territory. Pape considers Israel’s abandoning
the Gaza Strip and building physical barriers to be an “optimum strategy”
and he recommends accordingly that the United States augment border
and immigration controls. Hafez also thinks that governments can erect better
defenses, even though they are costly, citing Israel’s impressive successes in
foiling suicide attacks. Hafez thinks Israel’s construction of the wall should be
halted because it is a provocation. Pedahzur and Shay agree that much can
be done physically to impede suicide tactics. Shay especially recommends
border protection and the establishment of buffer zones. Davis, however,
warns of the potentially negative effect of defensive measures, as she cites
the alienation produced by overzealous Federal Bureau of Investigation tactics
toward American Muslims.
Changing popular attitudes to reduce support for suicide attacks is another
solution that is often proposed. Such a response would involve political
processes to resolve conflicts and ameliorate the social and economic conditions
that produce anger, humiliation, and despair (see Bloom, Hafez, and
Pedahzur). In particular, the Palestinian conflict is a key source of grievance
among Muslims worldwide and a resolution would have a moderating effect
extending well beyond Palestine. Hafez thinks that ending the PalestinianIsraeli
conflict would erode the cultural underpinnings of suicide attacks
there. He argues that the international community must be committed to the
peace process and condition assistance to both parties on their working for
a negotiated settlement. In his view, the United States above all should stay
out of cultural debates and avoid calls for educational reform in the Muslim
world. At the same time, Palestinian political and religious authorities should
make every effort to delegitimize suicide attacks. Khosrokhavar agrees that it
is up to Muslim thinkers to challenge radical Islam and that revitalization of
thought and discourse requires democratic openings. “The current obstacles
to the democratic opening up of Muslim societies are the conflicts in the
Middle East, Kashmir, and Chechnya, the presence of the nepotistic or even
corrupt ruling classes in their midst that are monopolizing power and repressing
the democratic demands of the emerging new classes.”150
Schweitzer also
argues that the answer to Al Qaeda is to provide an ideological alternative,
a moderate and pragmatic interpretation of Islam, which cannot be imposed
from the outside.
150 Khosrokhavar, Suicide Bombers, 228–29.
Explaining Suicide Terrorism: A Review Essay 159
Davis blames American foreign policies, referring to support for Israel, a
military presence in Saudi Arabia, alliances with corrupt and despotic leaders
in the Middle East, earlier economic sanctions against Iraq, and exploitation
of oil resources.151
She criticizes American policy for its inflammatory rhetoric,
inept efforts to win over Muslim public opinion, and inadequate attention to
intelligence needs.
Even if it could be accomplished, would a reduction in popular support
end suicide attacks? Unfortunately we have far too many counter-examples
to be able to assume that organizations cannot or will not act without public
approval. If change does occur, it will be gradual and suicide attacks or the
potential for them will be with us for some time.
In the immediate, what should governments do? With the exceptions of
Shay and Israeli, the authors considered here are generally critical of offensive
military responses to the challenge of suicide attacks, whether at home or
abroad. Gambetta, for example, calls for “astute policing” as opposed to
war.152
Davis faults the United States for over-reliance on military means, as
does Pedahzur, who also criticizes Israel, Russia, Sri Lanka, and Turkey.
Once governments are committed to a military response, as these countries
are, how should they manage the conduct of operations? Shay thinks that
militaries can ensure their safety reasonably well, so the issue is protecting
civilians by disrupting or deterring attacks. Bloom and Pedahzur stress the
need to avoid civilian deaths or collateral damage. Hafez criticizes Israel’s
excessive use of military force in the first months of the Al Aqsa intifada.
Pedahzur finds that targeted assassinations are particularly counterproductive
and empirical research by Kaplan, Mintz, Mishal, and Samban supports
his conclusion.153
Their analysis suggests that preventive arrests reduced suicide
bombings inside Israel after March 2002, while targeted killings only
increased the terror stock, that is, they sparked recruitment. Interestingly
enough, they also find that killing specific terror suspects caused more of
an increase in recruits than killing civilians. On the other hand, Gupta and
Mundra conclude that acts of political provocation best predicted future suicide
attacks and they warn that Palestinians interpreted the building of the
wall as provocation rather than self-defense.154
It is worth recalling in this
context that the death of Zarqawi in 2006 did not reduce suicide terrorism in
Iraq.
Whether or not to negotiate with the organizers of attacks (as opposed
to isolating them from their base) is a controversial question. Hafez calls for
creating disincentives for organizers and offering alternatives to violence as
151 Davis, Martyrs, chaps. 8 and 9.
152 Gambetta, ed., Making Sense, 299.
153 Edward H. Kaplan, Alex Mintz, Shaul Mishal, and Claudio Samban, “What Happened to Suicide
Bombings in Israel? Insights from a Terror-Stock Model,” Studies in Conflict & Terrorism 28, no. 3 (2005):
225–36.
154 Gupta and Mundra, “Suicide Bombing,” Terrorism and Political Violence, 591.
160 M. Crenshaw
well as options for exit. He thinks, for example, that Israel was mistaken to
insist on unconditional surrender when Hamas offered cease-fires in 2001
and 2003: “Such a victory is not possible against fragmented and decentralized
militant groups living among a sympathetic population with national,
regional, and international support.”155
On the other hand, Shay concludes
that a negotiated solution is only possible with secular organizations. They
are willing to adopt alternative methods, which Islamic organizations are not.
He calls for “an ongoing and uncompromising battle, while employing every
available defensive and offensive means.”156
CONCLUSIONS
These studies have not produced a consensus or even clearly demarcated
schools of thought, but they amply demonstrate that the process involves
complicated and contingent social, psychological, and political interactions.
There is no longer any need to introduce an analysis of suicide attack by
explaining to the uninitiated that it is not rooted in psychopathology or fanaticism
or indeed in any single cause such as deprivation, religious belief,
or frustration. It is an adaptable and controllable tactic. It has an instrumental
value for an organization. Despite impressions of ubiquity, its popularity
and effectiveness are limited. Very few individuals actually become suicide
bombers. Based on these studies, my estimate is that the total number of
suicide attacks worldwide since the 1980s is under 1,500, although figures
from Iraq increase by the day. The number of terrorist incidents in the MIPT
Terrorism Knowledge Base is over 20,000 (and this total does not include
domestic incidents before 1998).
Even though the numbers are low, data remain problematic. There are
multiple incompatible datasets rather than a common standard, and not all are
accessible to other researchers. References to the facts can be inconsistent and
contradictory. Assertions are not always supported by evidence. Researchers
need better information and more coordination in order to understand the
phenomenon, and to have confidence in their findings.
Even with perfect information, however, we still need to develop better
explanations of cause and effect. Two changes of approach might help. First,
most accounts do not systematically compare suicide to non-suicide attacks
committed by the same organization.157
Many of the reasons and motivations
proposed for suicide attacks apply equally to violence that does not require
the death of the perpetrator. None of the groups employing suicide tactics
155 Hafez, Manufacturing Human Bombs, 71.
156 Shay, The Shahids, 221.
157 For a start on this task, see Bruce Hoffman and Gordon H. McCormick, “Terrorism, Signaling,
and Suicide Attack,” Studies in Conflict & Terrorism, 27, no. 4 (2004): 243–81.
Explaining Suicide Terrorism: A Review Essay 161
uses them exclusively. Compared to other methods, is a suicide attack necessarily
more effective than a non-suicide attack in satisfying popular demands
for revenge or competing with rivals? Why would suicide bombings be more
resonant than other types of bombings, rocket attacks, shootings, beheadings,
kidnappings, torture, or snipers? The urban bombing campaign of the
Front de Lib´eration Nationale (FLN) during the Algerian war did not involve
suicide attacks, but it was effective in provoking the French and inspiring
Algerians. Why would a state be more sensitive to the cause of casualties
than to the volume? The assertion that suicide tactics are a superior demonstration
of resolve and elicit pity and admiration as well as shock and horror
is not implausible, but it is not supported by comparative empirical research.
Here analyses need to do more to account for the role of the news media
(especially television, the internet, and videos) as an amplifier of reactions.
Undoubtedly publicity is an incentive.
In terms of analyzing audience reactions, most attention has focused on
constituencies, not adversaries. The issue is not just how opposition organizations
frame the message of suicide tactics, but how those who are the object
of coercion frame them and thus receive them. Why are audiences swayed by
people who will simultaneously die and kill for a political cause? The public,
the policy community, scholars, and journalists seem much less affected by
the two acts if they are undertaken separately. What is the source of our
fascination? Why does the combined mode of violence seem distinctive and
unique, making it apparently more effective than other forms in signaling
superior motivation and spreading fear? Why should we be astonished by or
afraid of adversaries who claim to demonstrate that they do not fear death?
Surely the answer has to do as much with our own predispositions as with
those of the community that produced the suicide bomber.
Comparing suicide to non-suicide attacks would also contribute to explaining
timing and sequencing. We do not yet understand onset and duration.
If suicide tactics are effective in coercing the adversary and outbidding
rivals, why are they thought to be a last resort? Why not the first resort?
Why are some campaigns brief (PKK) and others prolonged (LTTE)? The tactic
cannot be explained independently of its strategic context, meaning its
place in the group’s overall conception of ends and means as well as its role
in government-challenger interactions in a given environment. We know little
of the details of group decision making and planning.158
Suicide tactics
cover a continuum: they can be an integral part of insurgency or civil war
(Iraq, Afghanistan, Sri Lanka) or isolated attacks on the inhabitants of the
capitals of major powers (Britain and Spain) or tourist centers in Asia or the
158 One effort in this respect is Alex Mintz, J. Tyson Chatagnier, and David J. Brul´e, “Being bin
Laden: An Applied Decision Analysis Procedure for Analyzing and Predicting Terrorists [sic] Decisions,” in
Pedahzur, ed., Root Causes, 152–78. They conclude that bin Laden’s decisions are determined by concern
for his personal prestige, honor, and standing among the public.
162 M. Crenshaw
Middle East. As it stands, analyses are inconsistent in linking the sequencing
of attacks to specific actions by governments or changes in public attitudes
and there is little empirical evidence for the relationships that are frequently
suggested (for example, that suicide attacks diminish when the public disapproves,
competition among groups decreases, or occupiers withdraw from a
disputed territory).
A second analytical problem is over-aggregation. The tactic is usually
treated as though it were a single unified method of violence. All types of suicide
attacks are merged together, despite their serving different instrumental
purposes: destroying military targets, assassinating public figures (both opponents
and collaborators), killing enemy civilians, or massacring co-religionists
in factional struggles. The users range from tiny groups to large and complex
organizations such as Hamas. Is their common feature, the death of the
perpetrator (more or less willing or intended, depending on the definition),
sufficient to produce conceptual coherence? What explains variations within
the phenomenon? For instance, why do some groups target civilians and others
military assets or individual officials? Why should the manner of violence
matter more than the target or the purpose? It is by no means clear that all
varieties of suicide attack would have the same origins and outcomes.
Perhaps governments do not need a specific political response to suicide
tactics. They could concentrate simply on guarding against all manner of acquiring,
transporting and detonating explosive devices broadly construed; a
task that will vary according to where attacks are expected to occur. Suicide
attacks are no longer surprising, even in Western democracies. At the most
basic, they increase the opportunities for attacking an adversary and governments
will find it hard to prevent all opportunities. The response should
be attentive to the place of suicide tactics in an organization’s overall strategy
and employ a broad conception of the threat. Governments also need
to be alert to substitution effects. Preventing suicide attacks could stimulate
adaptation and innovation, not the abandonment of violence. Governments
should also look to the future, in that the purpose and instigation of suicide
attacks might change. Today’s threat stems from the powerful association
between jihadist beliefs and suicide tactics, but radical Islamists do not own
the method.