Correspondence: What Makes Terrorists Tick 189
Instead of seeking to determine whether the strategic model or the social/psychological paradigm is correct—they are both right in the sense that some organizations
act according to the expectations of the strategic model and others because of social/
psychological expectations—scholars should be asking why some organizations seem
to be more strategic and others less so or not at all.
An initial explanation can be derived from the works of Barry Collins and Harold
Guetzkov, as well as Martha Crenshaw.6 These scholars observed that, as movements
go deeper underground, the “interpersonal” rewards of group membership become
more important than the strategic goals of their political mission. One would assume
that as long as organizations are able to grow more politically salient, they will maintain or cultivate a strategy to achieve deªnable political goals. Conversely, failure,
setback, or prolonged marginality will make organizations less strategic, as their immediate social environment overwhelms them. The seeds of a competing argument to that
proposed by Abrahms may have emerged. Abrahms argues that it is the quest for affective ties that drives terrorist organizations, whereas this alternative explanation is
based on the premise that it is the competitive environment between terrorists and government and the relative success achieved by the terrorists over time that predicts how
strategic or nonstrategic they may become. This is only one of many potential alternative explanations, however. What is important is the need to take up the challenge to
explain the puzzle of why some terrorist organizations seem to be strategic, others not,
and still others seem to be both over different time periods. One can then tease out the
policy implications of the ªndings for conºict resolution or, short of that, for successful
Abrahms has made a valuable contribution to explaining why some terrorist organizations are not necessarily motivated by political goals, and many more at some time
during their life cycles. The puzzle of why this is true of some organizations most of the
time, or most organizations some of the time, remains a challenge that students of terrorism should endeavor to meet.
Ramat Gan, Israel
To the Editors (Paul Staniland writes):
In “What Terrorists Really Want: Terrorist Motives and Counterterrorism Strategy,”1
Max Abrahms argues that terrorists are “social solidarity maximizers” (p. 101), rather
6. Barry E. Collins and Harold Guetzkow, The Social Psychology of Group Processes for Decision
Making (New York: John Wiley and Sons, 1964); and Martha Crenshaw, “An Organizational Approach to the Analysis of Political Terrorism,” Orbis, Vol. 29, No. 3 (Fall 1985), pp. 465–489. These
works are analyzed in McCormick, “Terrorist Decision Making,” p. 489.
1. Max Abrahms, “What Terrorists Really Want: Terrorist Motives and Counterterrorism Strategy,” International Security, Vol. 32, No. 4 (Spring 2008), pp. 78–105. Further references to this article
appear parenthetically in the text.
International Security 33:4 190
than strategic political actors, meaning that individuals join terrorist groups “to develop strong affective ties with other terrorists” (p. 96). As a result, terrorist groups
“routinely engage in actions to perpetuate and justify their existence, even when these
undermine their ofªcial political agendas” (pp. 101–102). Abrahms argues that the
seven predictions of the natural systems model ªnd support in “the preponderance of
theoretical and empirical evidence” (p. 103).
Abrahms makes a valuable contribution to scholars’ understanding of terrorism. His
argument, however, has signiªcant theoretical and empirical shortcomings. I focus on
three issues: the diverse motivations of individual participants, armed group elites’ use
of organizations to shape the actions of foot soldiers, and the behavior of terrorist organizations in Northern Ireland and Sri Lanka, two cases that Abrahms repeatedly refers
to in support of his claims.
the multiple logics of participation in militancy
Research on civil wars suggests that it is impossible to offer a single answer to the question, “What do terrorists want?”2 This work shows that there is no dominant logic of
participation in militancy. In a study of combatants in Sierra Leone’s civil war,
Macartan Humphreys and Jeremy Weinstein ªnd that “different logics of participation
may coexist in a single war.”3 Social sanction, grievances, and selective incentives all
provided motivation to join. Stathis Kalyvas and Ana Arjona’s survey of former combatants in the Colombian civil war reveals that revenge, ideology, and material rewards
were the dominant motivations for participation.4
Qualitative ªeldwork makes similar points. Lucian Pye explores the varied reasons
people joined communist guerrilla groups in Malaya.5 Elisabeth Wood highlights multiple mechanisms that encourage participation, particularly oppressive state policy and
the pleasure of agency.6 Weinstein’s work on insurgent organization suggests at least
two types of participants—investors and consumers.7 Roger Petersen’s study of insurgent mobilization reveals several mechanisms that encourage participation, including
community norms, rational calculations, psychological hopes, and a lack of other
2. It is possible that insurgent groups differ from purely terrorist groups; indeed Abrahms draws
such a distinction on page 83 of his article. Still, he repeatedly references insurgents such as the
Tamil Tigers; Hezbollah; PKK; Hamas; Afghan, Iraqi, and Chechen guerrillas; and FARC in support of his argument. Thus there does not appear to be an empirically relevant distinction. There
are few major organizations that do not kill both civilians and agents of the state.
3. Macartan Humphreys and Jeremy M. Weinstein, “Who Fights? The Determinants of Participation in Civil War,” American Journal of Political Science, Vol. 52, No. 2 (April 2008), p. 437.
4. Stathis Kalyvas and Ana Arjona, “Preliminary Results of a Survey of Demobilized Combatants
in Colombia,” unpublished paper, Yale University, 2006.
5. Lucian Pye, Guerrilla Communism in Malaya: Its Social and Political Meaning (Princeton, N.J.:
Princeton University Press, 1956).
6. Elisabeth Jean Wood, Insurgent Collective Action and Civil War in El Salvador (Cambridge: Cambridge University Press, 2003).
7. Jeremy M. Weinstein, Inside Rebellion: The Politics of Insurgent Violence (Cambridge: Cambridge
University Press, 2007).
8. Roger D. Petersen, Resistance and Rebellion: Lessons from Eastern Europe (Cambridge: Cambridge
University Press, 2001).
Correspondence: What Makes Terrorists Tick 191
elite goals and organizational control
Abrahms assumes that individuals’ motives for joining a group determine the goals
and actions of the organization. The problem with this assumption is that insurgent
elites often have political goals distinct from foot soldiers’ varied motivations. Elites
use organizational mechanisms to bring cadre behavior in line with leadership aims.
For instance, recruits joined the Provisional Irish Republican Army (PIRA) for many
reasons (though only after the political crisis of the late 1960s). PIRA’s ranks included
thugs, intellectuals, ordinary people, criminals, Marxists, Catholics, and dozens of
other types of individuals.9 But throughout the conºict, the PIRA leadership pursued
the political goal of unifying Ireland by calibrating offensives, strategizing about how
to use the Sinn Féin political party, negotiating with other parties and the British government, and ªnally agreeing to a political settlement and disbandment. The leadership
used coercion, material rewards, persuasion, and social pressures to largely keep its
cadres in line, regardless of their original motivations for joining.10
Foot soldiers similarly end up in the Liberation Tigers of Tamil Eelam (LTTE, or
Tamil Tigers) through numerous mechanisms, but the core leadership has resolutely
pursued the goal of an independent Eelam despite heterogeneous cadre-level motivations. Some foot soldiers were abducted as children; others fervently believe in the need
for a Tamil state; others are in thrall to the personality cult of LTTE leader Velupillai
Prabhakaran; others need a job; still others joined after seeing their mothers raped by
soldiers.11 The Tigers’ leadership has harnessed these diverse motivations toward a
consistent separatist goal.
Abrahms emphasizes organization theory, but he does not deal with the vast literature on the control, indoctrination, and incentives that leaders use to create coordinated
outcomes from disparate individuals.12 The Tigers and PIRA are fundamentally political, not affective. In the next section I show that both groups’ leaders have used their organizational control to pursue strategic and political goals.13
how do armed groups behave?
Below I brieºy survey the accuracy of Abrahms’s seven predictions of group behavior. I
ªnd little support for his argument in Northern Ireland and Sri Lanka, where I have
done detailed ªeld research on armed groups. The failure of the natural systems model
to illuminate these cases casts doubt on its broader accuracy. I focus on the Provisional
IRA and LTTE but also discuss the other major groups in each war.
9. Eamon Collins provides an inside picture of the range of personalities in the PIRA. See Collins
with Mick McGovern, Killing Rage (London: Granta, 1998).
10. Rogelio Alonso highlights “internal repression” as central to PIRA operations. See Alonso, The
IRA and Armed Struggle (London: Routledge, 2007), especially pp. 3–4.
11. A knowledgeable aid worker in Sri Lanka listed six major motivations for individuals to join
the Tigers, in addition to forced conscription. Interview by author, Colombo, March 2008.
12. A pioneering study is Nathan Leites and Charles Wolf Jr., Rebellion and Authority: An Analytical
Essay on Insurgent Conºicts (Chicago: Markham, 1970).
13. Between 1923 and 1968, the IRA was a peripheral and largely inactive organization. The LTTE
was formed in 1972, the other Tamil militant groups between 1975 and 1980; major combat did not
occur until 1983. It is hard to understand how the quest for affective ties explains this over-time
variation, unless there were no socially alienated single males in these societies prior to the late
1960s. This seems unlikely. Abrahms is trying to explain a variable with a constant.
International Security 33:4 192
Abrahms’s ªrst prediction is that armed groups “prolong their existence by relying
on a strategy that hardens target governments from making policy concessions”
(p. 102). By contrast, the LTTE has repeatedly brought the government of Sri Lanka to
the negotiating table. It won some concessions from the government in 1989–90 and
2002, though neither set of concessions went far enough for the Tigers. The PIRA extracted concessions from the British and then disbanded. There is little evidence that either group intentionally stopped the government from making concessions that it
believed would advance its strategic goal. Conºicts endured because of ºuctuating
power and distrust between the state and its foe.
This is also true of the other armed groups. In Northern Ireland, the Ofªcial IRA
(OIRA) unilaterally declared a cease-ªre in 1972 despite the absence of concessions, and
the Ulster Volunteer Force (UVF) signed on to the 1998 Good Friday agreement. In Sri
Lanka the other major Tamil militant groups—Eelam People’s Revolutionary Liberation Front (EPRLF), Tamil Eelam Liberation Organization (TELO), Eelam Revolutionary
Organization of Students (EROS), and People’s Liberation Organization of Tamil Eelam
(PLOT)—all agreed to the 1987 Indo-Lanka accord.14
Abrahms’s second prediction is that armed groups “ensure their continued viability
by resisting opportunities to peacefully participate in the democratic process” (p. 102).
Cross-national research casts doubt on this claim.15 Northern Ireland’s history is similarly disconªrming. The PIRA’s Sinn Féin political wing has engaged in the democratic
process and now shares power in the Northern Ireland Assembly. The OIRA entered
mainstream politics as the Workers’ Party. The UVF declared a cease-ªre in 1994 and
put forward the Progressive Unionist Party as its electoral face. In Sri Lanka, EPRLF,
PLOT, and TELO all agreed to lay down their arms and entered the democratic mainstream via the 1987 Indo-Lanka accord. Even after rearming during 1987–90 to protect
themselves, EPRLF and TELO are now unarmed parties that regularly contest elections.
The armed parties keep their weapons to protect themselves from the Tigers, who have
deemed them political traitors.16
Abrahms’s third prediction is that armed groups “avoid disbanding by reºexively
rejecting negotiated settlements that offer signiªcant policy concessions” (p. 102). In
Northern Ireland, the PIRA and the OIRA have disbanded, and the UVF is in the process of doing so. The Irish National Liberation Army (INLA) and Ulster Defence Association (UDA) remain intact, but have become fractious drug dealers, not seekers of
In Sri Lanka, EPRLF, TELO, and a faction of EROS have disbanded as combatants.
Groups that have not disbanded have plausible strategic rationales. The LTTE does not
trust a Sinhalese-dominated government to deliver on the terms of a settlement, much
14. M.R. Narayan Swamy, Tigers of Lanka: From Boys to Guerrillas (Delhi: Konark, 1994), pp. 240–
15. Abrahms asserts, “No peace process has transformed a major terrorist organization into a
completely nonviolent political party” (p. 86). A study of 648 terrorist groups between 1968 and
2006, however, reveals that “a transition to the political process is the most common way in which
terrorist groups ended (43 percent).” Seth G. Jones and Martin C. Libicki, How Terrorist Groups End:
Lessons for Countering al Qa’ida (Washington, D.C.: RAND, 2008), p. xiii.
16. There are three pro-state Tamil armed groups that contest elections: the PLOT, the Eelam People’s Democratic Party (EPDP), and Tamil Makkal Viduthalaip Puligal (TMVP).
Correspondence: What Makes Terrorists Tick 193
less grant independence. PLOT, EPDP, and TMVP remain armed to protect themselves
and reap rewards from extortion.17
Abrahms’s fourth prediction is that armed groups “guarantee their survival by espousing a litany of protean political goals that can never be fully satisªed” (p. 102). Although protean political goals certainly are espoused by some groups around the
world, in Northern Ireland and Sri Lanka, organizations have advanced clear separatist
goals. The IRA wanted the uniªcation of Ireland—an unlikely goal, but given the
history of 1916–22, the isolation of Northern Ireland from the British mainstream, and
demographic trends, not unthinkable.18 The OIRA was willing to settle for democratization and popular participation, and the PIRA showed it would settle for power sharing. The Protestant paramilitaries wanted to maintain union with the United
Kingdom.19 In Sri Lanka, even the hardest-line Tamil groups have sought a straightforward goal—an independent Tamil homeland. As shown in 1987 and since, several
Tamil groups have been willing to settle for power devolution.
Abrahms’s ªfth prediction is that armed groups “avert organization-threatening reprisals by conducting anonymous attacks” (p. 102). In Northern Ireland, the PIRA,
OIRA, and INLA often claimed responsibility for their attacks. When they did not, it
was because they did not want to lose support or be condemned internationally for an
operation that went wrong. The INLA took credit for killing Member of Parliament
Airey Neave, but not for the “Darkley massacre.” The PIRA tried to distance itself from
the Enniskillen bombing, but took credit for bombing Thiepval Barracks.20 The
Protestant paramilitaries claimed responsibility when it suited their purposes. In Sri
Lanka in the 1980s, the competition between Tamil groups encouraged them to take
credit for attacks. Since the LTTE gained primacy, it has publicized and claimed credit
for many of its suicide attacks.
Abrahms’s sixth prediction is that armed groups “annihilate ideologically identical
terrorist organizations that compete for members” (p. 102). It is unclear what behavior
this would refer to in either case. The PIRA was ideologically different from the OIRA,
which it split from in 1969. The Irish National Liberation Army differed from the OIRA
leadership, resulting in a break in 1974.21 The PIRA espoused a hard-line Irish nationalism, in contrast to the mass mobilizing, violence-averse focus of the OIRA or the far-left
vision of the INLA.22 These ideological differences help to account for why the PIRA
and the INLA broke away from the OIRA and why the OIRA stopped ªghting. Despite
feuds between these groups, there was no “annihilation.”
17. Human Rights Watch, “Sri Lanka: Political Killings during the Ceaseªre,” Human Rights
News (New York: Human Rights Watch, August 7, 2003), http://hrw.org/backgrounder/asia/
18. For an analysis, see M.L.R. Smith, Fighting for Ireland? The Military Strategy of the Irish Republican Movement (London: Routledge, 1995).
19. Steve Bruce, The Red Hand: Protestant Paramilitaries in Northern Ireland (Oxford: Oxford University Press, 1992).
20. The name used to take credit for PIRA attacks was P. O’Neill. After the disastrous Enniskillen
bombing (which killed eleven civilians), the PIRA leadership blamed local mistakes by the
21. Jack Holland and Henry McDonald, INLA: Deadly Divisions (Dublin: Torc, 1994).
22. Peter Taylor, The Provos: The IRA and Sinn Fein (London: Bloomsbury, 1998); and Richard English, Armed Struggle: The History of the IRA (Oxford: Oxford University Press, 2003).
International Security 33:4 194
Abrahms’s discussion of Sri Lanka is empirically suspect. First, he incorrectly claims
that the Tamil Tigers “did not target the Sinhalese government in the mid-1980s”
(p. 90). The LTTE’s initial killings in the 1970s targeted pro-government political
ªgures—the mayor of Jaffna in 1975, two policemen in 1977, and two police inspectors
and a member of Parliament in 1978. The LTTE’s June 1983 ambush of a convoy in
Jaffna, which killed thirteen soldiers, triggered a major escalation of violence between the Tigers and the state. The LTTE did make a bid for dominance (primarily
between April 1986 and March 1987) by targeting other groups, but it was simultaneously involved in antigovernment insurgency throughout northeastern Sri Lanka.
Second, there were important ideological differences among Tamil groups. The LTTE’s
closest ideological match was TELO, but it was an Indian proxy army. The LTTE, by
contrast, distrusted the Indian state. PLOT broke away from the LTTE in part because
of PLOT supporters’ left-wing vision of Sinhalese-Tamil cooperation in seeking to forge
an island-wide revolution.23 EPRLF and EROS espoused leftist ideologies and emphasized lower-caste mobilization. These differences mattered in the negotiation and
(failed) implementation of the Indo-Lanka accord of 1987: EPRLF and TELO supported
it, the LTTE opposed it, and PLOT tried to remain neutral.24 The remnants of PLOT
and parts of EPRLF and EROS now support the Sri Lankan government; another faction of EROS was integrated into the Tigers.
Abrahms’s seventh prediction is that armed groups “refuse to split up after the
armed struggle has proven politically unsuccessful for decades or its political rationale
has become moot” (p. 102). Contrary to Abrahms’s emphasis on social solidarity, most
of these organizations have suffered a vari ...
Purchase answer to see full