Content uploaded by Steven M. Radil
Author content
All content in this area was uploaded by Steven M. Radil
Content may be subject to copyright.
150
Eurasian Geography and Economics, 2009, 50, No. 2, pp. 150–171. DOI: 10.2747/1539-7216.50.2.150
Copyright © 2009 by Bellwether Publishing, Ltd. All rights reserved.
Terrorism and Counter-Terrorism: Situating
al-Qaeda and the Global War on Terror within
Geopolitical Trends and Structures
Colin Flint and Steven M. Radil1
Abstract: Two North American political geographers situate contemporary terrorism in the
world within broad historical trends and geopolitical structures. They employ Rapoport’s
“four waves of terrorism” to illustrate the changing geography of terrorism (from an intra-
state to an international phenomenon) and place it within the context of broad historical shifts
in modes of warfare that envisages terrorism as a form of war stemming from imperialism and
state-building. The authors broaden the structural setting of terrorism to include geo-
economics and the concept of relative deprivation, using empirical analysis to argue that con-
temporary terrorism is primarily a feature of the semi-periphery of the world-economy
(middle-income countries). Specific characteristics of the Global War on Terror are inter-
preted by relating processes of imperialism and state-building to a situation in which the U.S.
is facing geopolitical challenges and a possible decline in global-power status. Journal of
Economic Literature, Classification Numbers: I390, O100, Y900. 3 figures, 1 table, 52 refer-
ences. Key words: terrorism, al-Qaeda, Global War on Terror, political geography, world-
systems theory.
INTRODUCTION
he rise of terrorism as a topic of interest for geographers was a direct function of the ter-
rorist attack of September 11, 2001. Prior to that event geographers had generally
neglected the topic. Human geographers began by struggling for conceptual frameworks
(Flint, 2003a, 2003b; Mathewson and Steinberg, 2003; Murphy, 2003), but the initial push
was for the discipline to respond to the new situation in a manner to promote its position as a
science with policy relevance (Cutter et al., 2003). However, the policy-minded approach
was quickly subsumed in an overwhelming, and highly critical, focus by geographers upon
the counter-terrorist reaction of the Bush Administration. The geography literature still
reflects this imbalance between critical discourses aimed at the U.S.-declared and -led Global
War on Terror (GWOT) and analysis, with potential policy applications, of terrorist acts
themselves (see for example Elden, 2007; Gregory, 2004; Hannah, 2006).
Although it has been noted that the history of terrorism is quite long, the systematic
study of terrorism is a relatively new phenomenon (Laquer, 2001). Beginning in the late
1960s, scholarly work on terrorism reached a critical mass, creating a topical field of study
1Respectively, Associate Professor and Research Associate, Department of Geography, University of Illinois at
Urbana-Champaign, 607 S. Matthews Avenue, Urbana, IL 61801 (flint@illinois.edu). We thank Roger Sambrook
for assistance with the terrorism data used in this paper.
T
Flint.fm Page 150 Thursday, February 19, 2009 3:18 PM
FLINT AND RADIL 151
that drew from a variety of social science perspectives (Schmidt and Jongman, 1988). Stock-
taking exercises are common in terrorism studies and a notable point of agreement among
many is that the field of terrorism studies has not advanced very far in the last few decades
(cf. Schmidt and Jongman, 1988; Silke, 2001; Victoroff, 2007).
The main themes of the critiques of terrorism studies are largely either conceptual or
methodological. Some have argued that the problem lies with conceptual confusion about
terrorism itself and point out that definitional (and therefore, analytic) clarity on an obviously
socially constructed phenomenon is difficult to achieve (cf. Cooper, 2001; Turk 2004).
Boyns and Ballard (2004, p. 6) summarize this point: “political violence represents a social
problem, but exactly what that problem is and how to understand it . . . escapes the contem-
porary discussion of terrorism.” Others have found the problem to be more one of method.
Silke (2001, p. 2) noted that the terrorism literature was largely qualitative and descriptive in
nature and argued that the field of terrorism was “hamstrung by an obvious inability to make
accurate predictions of future events” due to the dearth of statistical analysis at any scale.
Although both arguments represent significant critiques of terrorism research, clearly one
argument flows from the other. Issues of method are important but only once the challenge of
conceptual clarity is met.
Terrorism remains a hotly contested term and long-standing debates over what consti-
tutes terrorism continue in the literature. As Boyns and Ballard (2004) have noted, many of
the simplest definitions used by various scholars and governmental agencies are power-laden
attempts to delegitimize and stigmatize opponents with the terrorist label. For instance, fol-
lowing the 2001 USA PATRIOT Act legislation in the United States, the legal definitions of
terrorism have been stretched to encompass property damage, and the FBI has argued since
that environmental sabotage represents a significant “domestic terrorist” threat in the U.S.
(cf. Amster, 2006; FBI, 2005). Critical definitions that emphasize the embedded power rela-
tions behind the definitional act are helpful (cf. Chomsky, 2000), but often remain fixated on
the behavior of states and fail to account for the diverse and changing nature of the phenom-
enon (Boyn and Ballard, 2004).
Perhaps the most important feature of terrorism is that it is an act of politics. This state-
ment distinguishes terrorism from similar acts that are motivated by purely criminal inten-
tions, such as extortion. Moreover, the motivations of terrorists have been defined as
“political altruism,” meaning that terrorists are motivated by a common ideology that their
behavior is for the benefit of a “greater good” (Hoffman, 1999, p. 43). Although the idea of
“political altruism” helps explain the ideology of terrorist movements, and the psychology of
terrorists, it is only a partial explanation of the reasons underlying terrorism. More precisely,
terrorists are motivated by “political geographic altruism” in the sense that their perceived
mission is to create new political spaces that they deem to be more just than current political
geographic arrangements. The most obvious cases are terrorist groups seeking national inde-
pendence from imperial powers or multi-national states.
By keeping the definition of terrorism simple, and emphasizing the use of political vio-
lence, we approach the phenomena in two separate ways. First, terrorism is identified as a
form of warfare, and hence connected to the main causes of war; imperialism, counter-
imperialism, and state building. Although the form of terrorist violence differs from what
social scientists have typically identified as war (notably the emphasis upon civilian targets
and relatively small numbers of casualties), the underlying causes and goals are the same.
Second, terrorism is a form of politics motivated by political and economic inequities. Using
the concept of “relative deprivation” and situating states within the hierarchy of wealth and
life opportunities in the capitalist world-economy, we provide empirical evidence to support
Flint.fm Page 151 Thursday, February 19, 2009 3:18 PM
152 EURASIAN GEOGRAPHY AND ECONOMICS
the notion that a combination of disparities of wealth and political restructuring are the chief
catalysts of terrorism. Obviously, if terrorism is political, then so is counter-terrorism. We
conclude the paper by situating the GWOT in the broad historical trends of war, with the
added factor of challenges to the U.S. as global power.
CHANGING GEOGRAPHIES OF TERRORISM
By taking the idea of “political geographic altruism” as a starting point, it is possible to
reinterpret existing historical understanding of the changing forms of terrorism as political
strategies within particular macro-political geography settings. This approach is useful in
emphasizing that terrorism has been a constant feature of politics, though we will discuss just
the modern period. However, its greatest utility lies in situating terrorism within broader
political geographic trends that help us to understand the motivations behind different phases
of terrorism, why the form of terrorism has changed over time, and the ways in which the for-
mation of new political geographic entities helped ameliorate the force or energy of the dif-
ferent phases.
Historical phases of terrorism have been identified as “waves” in Rapoport’s (2004)
classification of modern terrorism. The first wave of terrorism began in the 1880s and ended
at the beginning of World War I in 1914. Mobilized around the rallying cry of “anarchism,”
its aim was to spark revolutionary and whole-scale political change in a context of the politi-
cal reforms of the Russian Tsar. The political geographic context was one of an established
state, Russia, facing economic and political change that threatened the established aristo-
cratic system. Simply, the terrorists were not aiming to carve out a new national political
space, but change the way in which Russia was governed. Despite the existence of monarchi-
cal regimes and limited forms of democracy in other parts of Europe, the spread of “anar-
chist” terrorism was limited. The political geography of this first wave was defined by the
overarching understanding of the state as the focus or container of politics. Hence, the goal of
achieving control of the state apparatus and the national basis of terrorist groups.
The first wave of terrorism ended with the beginning of World War I, a conflict sparked
by a terrorist act motivated by the driving force of the second wave. The assassination of the
Austrian Archduke Franz Ferdinand in Sarajevo by a Serbian illustrated the growing political
force of nationalism: the belief in national self-determination, or national communities as
independent and sovereign states, as a just and natural geographic ordering of politics.
The second wave of terrorism (approximately 1920–1960) was dominated by political
challenges to imperialism, or decolonization, and the belief, subsequently cemented in the
ideology of the United Nations, in nation-states as a just form of political geographic organi-
zation. Particularly in situations where colonial powers were reluctant to leave, terrorism was
sometimes pronounced a necessary and just tactic. In one way, the geography of this wave
was similar to that of the first, as the nation-state was the target and motivation for terrorists.
However, the first and second waves differed in a significant way. In the second wave, the
agenda of terrorist groups within particular countries was framed within a global dynamic of
political change. Hence, terrorist groups were able to perceive and construct a shared sense
of victimhood and a belief in the justness of their cause that was at the scope of humanity
rather than the nationally based and class-specific terrorism of the first wave.
The third wave of terrorism was also dominated by nationalist groups who argued that
the project of decolonization was incomplete and/or had brought into being states that gave a
state to one nation at the expense of another. For example, Irish nationalists believed that the
British Empire was still extended into Ireland while the Palestine Liberation Organization’s
Flint.fm Page 152 Thursday, February 19, 2009 3:18 PM
FLINT AND RADIL 153
cause reflected its failure to claim a national space as a result of the construction of Israel.
However, the geography of nationalist-based terrorism in the third wave was very different
from the second wave. The internationalization of terrorist activity became prominent in the
third wave, despite the nationalist motivations. Internationalization took two main forms.
Terrorists consciously internationalized their acts, such as the PLO’s hijacking of airplanes or
its killing of Israeli athletes at the 1972 Munich Olympic Games, to achieve greater visibility
in an attempt to put their cause on the agenda of foreign policy. By internationalizing their
behavior, terrorism became a matter for international relations or statecraft, and not some-
thing to be left to the domestic politics of the country in question.
In the third wave, the internationalization of terrorism also intensified through growing
cooperation between, ostensibly, nationally oriented terrorist groups. From this time it
becomes fruitful to think in terms of terrorist networks. Although it is arguable that the
Serbian-organized group of terrorists involved in the assassination of Archduke Franz
Ferdinand was a manifestation of transnational terrorism, the period of the third wave intro-
duced cross-border cooperation as the norm for terrorists. Networks facilitated the training,
armament, and financing of terrorists. However, this geography of networks was catalyzed
by the role of some states in sponsoring terrorist activities (Laqueur, 1987). The geographies
of networks and sovereign states intersects here, a theme we will return to when discussing
the activities of al-Qaeda and the Global War on Terror.
The intersection of terrorism with a legitimate political ideology (national self-
determination), and the prominence of state sponsorship, within the context of Cold War
competition, was to make terrorism “almost respectable” (Laqueur, 1987, p. 269) in the sense
that a sufficient majority of states blocked effective UN coordination of counter-terrorist
actions. The context of the Cold War played another role in defining the form of the third
wave of terrorism. An ideological element was added with the emergence of terrorist groups
motivated by Marxist ideology. For example, the Baader-Meinhof gang in Germany, the Red
Brigades in Italy, the Weather Underground in the U.S., and the Naxalist movement in India
(Joseph and Srivasatava, 2008) were all motivated by left-wing ideology.
The geography that emerged from the third wave of terrorism may be seen, in hindsight,
as an important transitional moment. The intersection of networks and sovereign territorial
spaces came to the fore, an issue that is at the center of the difficulties facing contemporary
counter-terrorism. Moreover, though it was relatively easy to see the bulk of terrorist activity
in the third wave as a matter of nationalist groups seeking international cooperation and a
global stage, the appearance of ideologically motivated terrorism began to blur any clear dis-
tinction between local, or national, issues and global reach and practice.
The fourth wave of terrorism (1990s–present) brought some of the emerging issues from
the third wave to a head. But, and surprisingly to analysts and policymakers, it was not polit-
ical ideology that was the driving force but religious belief. Rapoport identifies the fourth
wave of terrorism as a period of religious terrorism, though terrorism motivated by national-
ism is far from gone. If Rapoport is correct, a very new geography of terrorism has been cre-
ated in the fourth wave. The goals and beliefs of religious terrorists are more than the
intersection of networks and sovereign spaces. They transcend the state and seemingly, at
least rhetorically, reduce the importance of the state, or control of any territorial entity, as the
goal of terrorist activity.
Religious terrorism has become a global phenomenon, a violent expression of the funda-
mentalist interpretation of the Christian, Jewish, Muslim, Sikh, and Buddhist religions
(Juergensmeyer, 200). Juergensmeyer uses the term “cosmic war” to describe the beliefs of
contemporary religious terrorists: a war of good against evil, with God, or another form of
Flint.fm Page 153 Thursday, February 19, 2009 3:18 PM
154 EURASIAN GEOGRAPHY AND ECONOMICS
supreme being, the sole adjudicator. Terrorists believe themselves to be the foot soldiers of
their deity’s will. According to Juergensmeyer, religious terrorism has drastically changed
the geography of conflict. It has removed the state from the equation, as the struggle is now
for people’s souls and not the control of the state, seen as a secular institution.
The key geographical differences in the move to religious terrorism are a purported
reduction in the centrality of the state, as both venue and goal of terrorist activity. From the
Russian anarchists of the 1890s to the nationalist terrorists of the 1990s the organs of the
state were targeted either directly (government officials, police officers, etc.) or indirectly by
attacks on civilians aimed at pressuring governments to address the terrorists group’s cause.
The essence of terrorist causes were either the construction of a new state (Palestine, for
example), redrawing the boundaries of an existing state (e.g., the Irish Republican Army), or
challenging the existing political-economic system (e.g. the Baader-Meinhof gang). It is
because of an apparent rejection of the state as a relevant political actor that leads
Juergensmeyer to claim that acts of religious terrorism are more likely to intend mass civilian
casualties. He argues that there is no longer a need for terrorists to commit acts that negotiate
the delicate balance between provoking the state into concessions without so alienating pub-
lic opinion that negotiation is politically impossible for state politicians.
However, a more sober evaluation of religious terrorism is required. The argument that
the state is no longer a relevant arena seems quite dubious. First, equating the perpetrators of
the September 11, 2001 attacks with Taliban support resulted in the invasion of Afghanistan
as the first act of the Global War on Terror. The use of sovereign state territory by ostensibly
religious groups has been the focus of the GWOT. Second, U.S. allegations of state sponsor-
ship of terrorism aimed at, for example, Syria and Iran regarding Hezbollah echo Laquer’s
(1987) evaluation of terrorism in the third wave. Third, the aims of religious terrorists can
still be interpreted as a desire to control the state apparatus. For example, the goal of
Christian fundamentalists in the U.S. targeting doctors and clinics performing abortions is to
change secular legislation. The Taliban emerged as a proxy agent for the United States in the
Cold War battle with the Soviet Union, whose goal was to “liberate” the sovereign territory
of Afghanistan from colonial oppression. Although the politics of the Taliban entails the
application of laws derived from a particular interpretation of Islam, the outcome is still the
control of the state apparatus. A final example of the blending of religious and nationalist
ideals is the attacks in Mumbai in November 2008, in which the politics of Kashmiri nation-
alism, Muslim fundamentalism, inter-state conflict between India and Pakistan, and alleged
transnational links to al-Qaeda were all relevant.
Though Rapoport’s waves are a useful framework for understanding the manner in
which terrorism has changed, the weakness of the approach is the separation of terrorism
from broader political and economic processes. Terrorism is isolated from possible causal
linkages with political-economic dynamics. In the next section we will attempt to link terror-
ism with broader changes in the nature of war before, in the following section, making an
empirical case for situating terrorism in the global structure of the capitalist world-economy.
TERRORISM AS AN ACT OF WAR
Broad classifications of war, such as Holsti’s (1991), have defined the causes of war in
terms of territorial disputes, ideological conflict, nation-state creation, economic conflicts,
and human sympathy (ethnic or religious affiliation). Holsti noted that over the course of the
19th and 20th century “relatively abstract issues—self-determination, principles of political
philosophy, and ideology and sympathy for kin—have become increasingly important as
Flint.fm Page 154 Thursday, February 19, 2009 3:18 PM
FLINT AND RADIL 155
sources of war, while concrete issues such as territory and wealth have declined.” Two points
are relevant. First, though they start from very different positions, Holsti and Rapoport draw
the same conclusions. Conflict in general, and terrorism in particular, have become increas-
ingly focused upon issues of group identity. Second, the use of the term “abstract issues” is
telling. The social science tradition has been focused on economic forces and political insti-
tutions as both the cause of social processes (such as war and terrorism) as well as what
needs to be explained. An overarching understanding of humans as “rational actors” from a
Western theoretical perspective (either rational choice or Marxist) led to the marginalization
of things that matter to the bulk of the human population. Group affinity, or communal
belonging, has always been central to the human existence. It may be hard to explain, espe-
cially within a quantitative modeling framework, but that does not make it less “concrete”
than, say, economic growth or the formation of states.
The scramble to understand “Islam” or “fundamentalism” was more than just a contem-
porary exercise in Orientalism (Said, 1978); it was a reflection of social science’s failure in
understanding processes that motivate human beings. However, if communal affiliation,
including religious belief, has always been a feature of human societies it does not explain
why we have witnessed the emergence of terrorism motivated by religious beliefs only
recently. Another classification of war is helpful in addressing this issue. Johnston et al.
(1987) used world-systems theory to classify warfare as different expressions of the expan-
sion of the capitalist world-economy. Warfare was seen as a feature of the expansion of the
capitalist system across the globe through Western exploration, imperialism, and colonial-
ism. In parallel, war has also been a manifestation of state-building, often in the wake of
colonial powers ending direct rule, sparking civil wars in which the lines are often drawn
with regard to ethnic and religious affiliation (Tilly, 1990). O’Loughlin and van der Wusten
(1993) noted that the second half of the 20th century had seen a dramatic rise in such resis-
tance violence compared to the predominance of imperial-colonial wars between 1890 and
1945. This trend has been confirmed in updated analysis by Harbom et al. (2008) to cement
the current dominant wisdom that war has become an intra-state rather than an inter-state
concern.
Terrorism was not included in O’Loughlin and van der Wusten’s (1993) and Johnston et
al.’s (1987) classifications and the interpretation of the trends for a simple empirical reason:
The social science definitions of war require a large number of battle deaths, rather than the
relatively smaller number of civilian casualties that usually signify terrorist campaigns.
However, the causes of terrorism match the dominant causes of war, as well as the trends in
the form of warfare. The second and third waves of terrorism were motivated by resistance to
colonial powers in the wake of the failure of colonial projects. O’Loughlin and van der
Wusten claimed that no wars (within the rigid social science definition) were identified that
met the definition of state resistance, weaker (peripheral) states attacking stronger (core)
states. However, as O’Loughlin and van der Wusten acknowledge, this is precisely the moti-
vation for state-sponsored terrorism.
Terrorism is a form of warfare that has changed with the changing structure of the mod-
ern social-system. One constant feature has been the violence related to the construction of
nation-states. This has involved two issues: the imperfect match between communal identity
and political representation, and the rise of the industrial political class and its claim for
political representation. These two processes merged in the 19th century ideology of popular
nationalism, which gave “the masses,” to coin Mosse’s (1975) term, a sense that they
deserved a seat at the political table because they shared membership of the national group
with the ruling classes. Imperfections in defining the extent of the national community and
Flint.fm Page 155 Thursday, February 19, 2009 3:18 PM
156 EURASIAN GEOGRAPHY AND ECONOMICS
gaining representation and benefits have blended material grievances with the quest for more
meaningful affiliation.
The second issue is resistance to the spread of Western imperialism, most evident in
Rapoport’s second wave of terrorism. A combination of resistance to Western colonialism
and nation-state building peaked in this post-war period, but the legacies remain as newly
independent states have faced conflicts over the boundaries of states and how they should
ruled. For example, terrorism in India and Pakistan is a function of disputes over the territo-
rial extent of these states2 as well as the form of governance within the two countries (Islamic
fundamentalist terrorism, Naxalism, and Hindu extremists all fit the latter).
The fundamental conclusion from a combined interpretation of Rapoport’s waves of ter-
rorism and the broad trends in warfare that Holsti (1991) and O’Loughlin and van der
Wusten (1993) identify is that terrorism is another form of warfare and should be seen as a
manifestation of the broad political-economic changes of human society. However, the par-
ticularity of terrorism as a form of war should not be discounted. Without creating a romantic
view of terrorists, the struggles they claim to be the vanguard of are those that pit the margin-
alized against the strong. They intend to mobilize subordinate groups against the dominant,
whether that be classes (anarchist vs. aristocrats), nations (Basques vs. the Spanish state), or
religions (fundamentalists against a perceived hostile secularity or competing proselytizing
religion). It is not the weaker political position that dictates the resort to terrorist acts but the
imbalanced resources of violence: The ideological terrorist uses kidnap and shooting against
the state security apparatus, while al-Qaeda mobilizes suicide bombers against the might of
the U.S. military and its allies.
Conceptualizing terrorism as a reaction to marginalization requires consideration of eco-
nomic as well as political processes. Though the ultimate goal is to provide a framework that
combines political and economic factors, we move from thinking of terrorism as a form of
warfare within processes of imperialism and state-building to a macro-analysis that situates
terrorism within the hierarchy of well-being and disparate life-chances that is the reality of
the capitalist world-economy.
TERRORISM AS MANIFESTATION OF RELATIVE DEPRIVATION
In the definition of terrorism an important distinction is made between violence by the
state as a strategy of control and violence against the state as a strategy of resistance (e.g.,
Post, 2001). This distinction has been an important one to theory construction as it recog-
nizes a divergence in motivation behind each form of violence. Following the “relative depri-
vation” theoretical literature from the early 1970s, economic conditions have long been
considered prime suspects behind all kinds of anti-state political violence, including terror-
ism (van der Wusten, 2005). The theory of relative deprivation, introduced in political scien-
tist Ted Gurr’s (1971) seminal work Why Men Rebel, locates the source of this kind of
political violence in the discrepancy between economic expectations and capabilities. In
other words, frustration over the juxtaposition of high economic expectations and low eco-
nomic opportunities are likely to result in collective political violence. Understanding terror-
ism as a response to economic powerlessness or marginalization is quite common in
terrorism literature (cf. Boyns and Ballard, 2004; Turk, 2004). However, a number of recent
studies have challenged the thesis that poverty generates anti-state terrorism at the scale of
2Acts such as those in November 2008 are connected to the Jammu and Kashmir conflict.
Flint.fm Page 156 Thursday, February 19, 2009 3:18 PM
FLINT AND RADIL 157
the state (Bravo and Dias, 2006; Piazza, 2006; Krueger and Laitin, 2008) while at the scale of
the individual, a recent study argues that among Palestinians, those that were more well-off
were also more likely to resort to terrorism against Israelis (Krueger and Malečková, 2002).
The policymaking establishment seems relatively untroubled by such findings, and
questions about the links between economic development and terrorism appear largely set-
tled in the policy community. For example, a recent United Nations report asserts that “[eco-
nomic] development has to be the first line of defense for a collective security system that
takes prevention seriously. Combating poverty will not only save millions of lives but also
strengthen States’ capacity to combat terrorism . . . Development makes everyone more
secure” (UN, 2004, p. 2). The U.S. National Security Strategy concurs that poverty plays a
significant role in terrorism. Accordingly, the Bush administration security policy identifies
the spread of “development, free markets, and free trade to every corner of the world,” as a
centerpiece of its counter-terrorism efforts (Bush, 2002). Clearly, anti-state terrorism is
understood by these examples as part of a larger group of ills largely caused by low levels of
economic development and is therefore expected to have a negative relationship to develop-
ment: as development increases, terrorism is expected to decrease.
This certainly testifies to the assumption of the existence of a negative relationship
between economic development and anti-state terrorism but also to the presumed form of the
relationship, which is both universal and linear. The universalist assumptions are unambigu-
ous: references to increasing development in “every corner of the world” and development
making “everyone” more secure leave little doubt that the presumed logic behind develop-
ment and anti-state terrorism is thought to apply equally to every region, state, place, or con-
text. This uniformity also suggests linearity, where anti-state terrorism will be expected to
diminish in each specific instance as development measures increase, no matter the develop-
mental starting point. This context-free argument has little traction for those who expect a
theory of anti-state terrorism to not only address questions of why but questions of where as
well. A systemic approach may prove useful in this regard and a few voices have made such
an argument. For example, one of the few common elements between the findings of a panel
of the U.S. National Research council on terrorism (Smelser and Mitchell, 2002) and
Baudrillard's, 2002) interpretation of the attacks of 9/11 is that the phenomenon only makes
sense in reference to global economic patterns.
Bergesen and Lizardo (2004) take such a perspective and use world-systems theory to
argue that terrorism is likely a feature of the semi-periphery of the capitalist world-economy.
The semi-periphery is defined as a mixture of high- and low-end economic processes, and
the most dynamic zone of the world-economy where states are competing to gain core status
(Flint and Taylor, 2007, p. 18). It is within the semi-periphery that social struggles are most
intense and the role of the state as an interventionist organ of economic development is most
pronounced (Flint and Taylor, 2007, pp. 19–22). Hence, there are expectations that relative
deprivation will be most evident in semi-peripheral countries and that there will be resistance
to an active state apparatus. As such it represents a testable position of the kind sought after
by Silke (2001) and others in terrorism studies but also represents an opportunity to reflect on
policy approaches built around universal and linear assumptions about combating terrorism
through economic development.
The evidence for relationships between development and terrorism can be evaluated
using cross-sectional data, with states as the unit of analysis. Position in the world-economic
system is estimated using the United Nations Human Development Index and classifications
are analyzed against country-level domestic terrorism incident data from 1998 to 2006.
Employing a nonparametric methodology, we find evidence that there are statistically signif-
Flint.fm Page 157 Thursday, February 19, 2009 3:18 PM
158 EURASIAN GEOGRAPHY AND ECONOMICS
icant relationships between development and terrorism, but that they are not linear and not
uniformly negative. Instead, the form predicted by Bergesen and Lizardo (2004) was evident:
the highest levels of anti-state terrorism occurred among countries that occupy the middle
ranks of UN development measures, while countries at either end of the developmental spec-
trum experienced far lower levels of anti-state terrorism.
Terrorism event data were drawn from the RAND-MIPT Terrorism Incident Database
(RAND-MIPT/START, 2008) covering a period from 1998 to 2006.3 This dataset relied on
media reports in an attempt to create a global database of numerous different types of terror-
ism. This analysis makes use of one of the dataset’s typologies, the distinction between
“international” and “domestic” terrorism. For the database, international terrorism consists of
incidents that begin in one state and end in another, or that involve a group from one state or
region attacking in another state.4 These kinds of incidents may be seen as originating in the
periphery or semi-periphery zones but aimed against states in the core zone. Conversely
domestic terrorism is categorized by the database as incidents that originate and end in the
same state and most often represent local issues and grievances. This emphasis on localized
and less publicized incident data is new in terrorism studies, where most data-driven analysis
has traditionally relied on small databases of well-known international incidents (cf. Schmidt
and Jongman, 1988; Silke, 2001). The database’s domestic terrorism thus may be seen as
violence that originates in the same economic zone in which it is aimed. While both interna-
tional and domestic terrorism may fit the kind of violence to which the theory of relative dep-
rivation and Bergesen and Lizardo’s system-level framework refer—anti-state violence in
response to economic marginalization—the small number of international events (just over 8
percent of the total in the database) were excluded to provide a clearer look at the dominant
trends at the state scale. This approach yielded 22,069 incidents responsible for 63,349 inju-
ries and 32,142 fatalities across 114 different states or territories. Incidents were aggregated
by state by year and rates per 100,000 persons were calculated using annual population esti-
mates.5
Developmental data for each country in the international system was drawn from the
United Nations Human Development Index (HDI; see UNDP, 2007). The HDI is a composite
index produced annually by the United Nations Development Programme since 1990. This
composite index is based on the assumption that single measures of development, such as
gross national or domestic product per capita, fail to capture all aspects of development. As
such, human economic development should be measured by a larger range of socio-
economic indicators. The HDI is built around the basic tenets that a long, healthy life,
education, and a decent standard of living are the essential components of human economic
3At the time of the original data collection and analysis, the incident data used for this study (through Decem-
ber 2006) were made available to individual researchers by the National Memorial Institute for the Prevention of
Terrorism (MIPT). As of April 2008 elements from this database (through 2004) were transferred to the National
Consortium for the Study of Terrorism and Responses to Terrorism (START) at the University of Maryland and
merged into its Global Terrorism Database. For more information, see http://www.mipt.org/TKB.asp and http://
www.start.umd.edu/.
4The numerous aircraft hijackings in the 1960s, 70s, and 80s discussed earlier are an example.
5It is worth noting that a significant number of states in the international system had no recorded incidents.
States with no presence in the RAND-MIPT/START database may indeed reflect the absence of terrorism or may
simply reflect the absence of reporting. Zero-observation states were included in the analysis, but the relatively large
number of states without recorded terrorism does raise concerns about the reliability of the reporting process in some
countries. This issue is discussed in greater detail later.
Flint.fm Page 158 Thursday, February 19, 2009 3:18 PM
FLINT AND RADIL 159
development and measures development through the use of three factors: longevity, knowl-
edge and relative purchasing power.
The HDI assigns each country a score from 0.000 to 1.000 with higher scores represent-
ing higher levels of development. The HDI also classifies countries into categories of high,
medium, and low development; values above 0.800 are categorized as high development,
values between 0.800 and 0.500 are categorized as medium, and values less than 0.500 are
categorized as low development. We make use of this classification scheme to approximate
the three economic zones of the world-systems framework. The high-development countries
are understood to meaningfully approximate and comprise the geographic extent of the core
economic zone at the scale of the state. Similarly, medium- and low-development countries
are understood to approximate the semi-periphery and periphery zones, respectively. While
there are certainly other ways to estimate which states comprise each zone in the world-
systems literature (cf. Snyder and Kick, 1979), the HDI does provide a convenient summary
index for many factors that influence development, including economic deprivation.6 Addi-
tionally, it performs well in terms of cross-country and inter-temporal availability and, like
many composite indices, is useful for the ability to “simplify complex measurement con-
structs, to focus attention and to catch the eye” (Booysen, 2002, p. 146).
Preliminary exploratory analyses suggested that terrorism was present to some degree in
all development categories (Fig. 1). It was also clear that terrorism was not well represented
in geographic regions with the lowest levels of economic development, especially Africa.
Instead, terrorism appeared more frequently in areas of the world with higher levels of devel-
opment. Indeed of the 20 states with the highest incident rates for which there are corre-
sponding HDI values, none are in the UN’s low-development category. The variability
present in the geographic distribution of the incident data distribution lent little support to
assumptions that terrorism was universally related to development and suggested that what-
ever relationships might exist, they were unlikely to be linear in form.
The incident data also displayed a great deal of range, with a low of 1 incident over
nine years occurring in 20 different states to an extreme value of 6,795 incidents over nine
years occurring in Iraq (Fig. 2). Despite this large range, 93 out of the 114 states with
recorded terrorism had annual incident frequencies well below 0.1 per 100,000 persons.
The existence of significant outliers—especially the high incidences of terrorism recorded
in Iraq, the West Bank/Gaza Strip, and Israel—indicated that a mean incident value for ter-
rorism would not represent the typical value for terrorist events. These outliers led to the
selection of nonparametric statistical tests that do not rely on a priori distribution assump-
tions and that are flexible and robust in the presence of large outliers. Incidents over the
time period of the database were aggregated by country. Population estimates for each
country were used to calculate incident, injury, and fatality rates per 100,000 persons for
each year. Rates were selected for this analysis over raw incident counts to control for pop-
ulation variation as a source of cross-county incident variation. Average annual incident
rates and average annual HDI scores were calculated and ranked, with the highest rates
receiving the largest rank.
Obvious linear relationships between terrorism rates and developmental measures were
not present either when compared year by year or when averaged across the entire time
6We also recognize that it is improper to simply equate states and clusters of states with core, periphery, and
semi-periphery zones. The terms core and periphery refer to processes, not geographic entities. Here we are using a
common shorthand terminology whereby middle-income countries are identified as states in which there is a rela-
tively even mixture of core and peripheral processes.
Flint.fm Page 159 Thursday, February 19, 2009 3:18 PM
160 EURASIAN GEOGRAPHY AND ECONOMICS
period. Several tests were performed to confirm this observation, including two nonparamet-
ric tests for association: Spearman’s rank-order correlation (Spearman’s p) and Kendall’s
Fig. 1. Map of mean terrorism incidents per 100,000 persons and mean HDI categories (average
of HDI values reported in annual volumes of the Human Development Report). Sources: UNDP
(1998–2006), CIA (2008), and RAND-MIPT/START (2008).
Fig. 2. Frequency distribution of terrorism incidents, 1998–2006. Source: RAND-MIPT/START
(2008).
Flint.fm Page 160 Thursday, February 19, 2009 3:18 PM
FLINT AND RADIL 161
rank-order correlation (Kendall’s τ). Both Spearman’s p and Kendall’s τ measure associa-
tions between ranked data, and the calculated coefficients of both statistics range from -1.0
and 1.0. For both statistics, a positive correlation indicates that the ranks of both variables
increase together while a negative correlation indicates that as the rank of one variable
increases, the other decreases.7
As shown by Table 1, year-to-year correlations of incident rates with developmental
scores revealed no significant associations, although correlations of average incident rates
and developmental scores for all years did indicate a significant but weakly positive relation-
ship instead of the expected negative (Spearman’s p = 0.157, p < 0.05). For thoroughness, the
same analysis was performed after excluding states with no recorded terrorism and no signif-
icant associations were identified. The correlational test results seem to contradict the nega-
tive linear relationship presumed by many policymakers, but were consistent with the
observations gleaned from the map. To further explore the issue, a second analytical
approach was implemented. A Kruskal-Wallis non-parametric one-way analysis of variance
test was selected to determine if there were statistically significant differences between inci-
dent rates between the three developmental categorizations. This test is an alternative to
parametric ANOVA used when the assumption of normality or equality of variances is not
met. Like the Spearman’s p and Kendall’s τ tests, Kruskall-Wallis uses the ranks of the data
to calculate a statistic that tests if the observations in each category are drawn from the same
population. A significant result suggests observations are drawn from different populations,
which is interpreted here as suggesting different causal mechanisms at work in different eco-
nomic zones of the world system.
States were grouped based on their average developmental scores using the same high,
medium, and low categorizations as the UN, and the Kruskal-Wallis statistic was calculated
for average incident rates. This test revealed some patterns undetected by the correlational
tests. The Kruskal-Wallis test detected significant differences among the groups for incidents
(p <0.05). When the data are viewed in this fashion, a distinct form takes shape that strongly
evokes the theoretical position of Bergesen and Lizardo (2004). As seen in Figure 3, the mid-
dle development groups had the highest overall mean incident rates. This finding was also
robust against changes in the number of categorizations or to using different breakpoints.
Several different categorizations were tested using different breakpoints and the pattern held
every time: states in the middle of the development data had statistically significant higher
Tabl e 1 . Spearman’s p Correlation Coefficients for Incident Rates and Developmental
Scores, 1998–2006
Year Incident rates Year Incident rates
1998 0.163 2003 0.144
1999 0.540 2004 0.001
2000 0.055 2005 0.008
2001 -0.035 2006 0.098
2002 0.088 Meana0.157*
aFor 1998–2006. Asterisk (*) denotes statistically significant correlation (p < 0.05).
Source: Compiled by the authors from UNDP, 2007: CIA, 2008; and RAND-MIPT/START, 2008.
7For simplicity, only results for the Spearman’s p tests are presented, as the results for both tests are similar and
lead to the same conclusions.
Flint.fm Page 161 Thursday, February 19, 2009 3:18 PM
162 EURASIAN GEOGRAPHY AND ECONOMICS
levels of terrorism. This adds to the correlational findings and suggests that while relation-
ships can be demonstrated at the scale of the state, they are not linear in form. Instead they
take an inverted-U shape for the variables considered here.
Our results lead us to reject the common assertion that the anti-state version of terrorism
negatively and universally correlates with measures of development in a linear fashion. Cor-
relation analyses show no clear linear relationship between terrorism and development from
1998 to 2006, and although average rates for the entire data set were found to associate with
average developmental scores, this association was positive instead of the negative relation-
ship assumed by conventional policy thinking. However, by grouping countries in a way that
approximates the economic zones of world-systems theory, statistically significant differ-
ences were observed that help to clarify the form of the relationship: states in the middle
development category experienced the most incidents, injuries, and fatalities. A curvilinear
inverted-U shape is clearly suggested by these findings. This may be construed as some evi-
dence for Bergesen and Lizardo’s argument that one should expect to find anti-state terrorism
more commonly in the semi-periphery economic zones.
This last finding is particularly important, as it has clear policy implications. In the case
of the least developed countries, this suggests that increases in development might actually
lead to an increase in terrorism. In other words, efforts to restructure the economies of the
lowest ranked states may in fact contribute to the very phenomenon the action is meant to
guard against. On the other end of the spectrum, states in the middle of the developmental
pack may benefit from increasing development by seeing a reduction in terrorism. However,
as demonstrated by the incident rankings, the drop in terrorism from middle- to high-
development categories is slight, especially compared to the large jump in incident rankings
from the low to the middle categories. Viewed through the lens of anti-state terrorism, the
move from low development to medium seems to engender more violence than the move
from medium to high seems to do away with.
As previously discussed, the theoretical story that accounts for most of the interest in the
linkages between development and terrorism has origins in the relative deprivation literature.
Fig. 3. Mean terrorism incident rate per 100,000 persons by mean HDI categories. Sources:
UNDP (2007), CIA (2008), and RAND-MIPT/START (2008).
Flint.fm Page 162 Thursday, February 19, 2009 3:18 PM
FLINT AND RADIL 163
Our findings resonate with this literature, or at least condition the expectations to some
degree by suggesting that the kind of injustice and frustration assumed to inspire acts of ter-
rorism at an individual level may be somehow felt more strongly by citizens of states in the
middle of the developmental pack. Additionally, our findings parallel those by Callaway and
Harrelson-Stephens (2006). The authors anticipate the inverted U-shape demonstrated here
when they state that “citizens at both extremes of subsistence are less likely to engage in ter-
rorist activity.” However, as Callaway and Harrelson-Stephens (2006) acknowledged, a
clearer theoretical narrative that could account for this pattern at the scale of the state has yet
to emerge. More plainly, the theory of relative deprivation may suggest a generic rationale
for anti-state terroristic violence, but it does not expressly suggest where and when one
should expect to find such terrorism, especially at the scale of the state. For such an under-
standing, the systemic, world-systems theory approach of Bergesen and Lizardo (2004)
seems useful.8
The correspondence of the pattern of terrorism with the expectations of world-systems
theory provides further support for the need to consider contemporary terrorism not as some
new and unique geopolitical threat but as manifestation of the constant features of the mod-
ern world. To the long-term processes of colonialism and state building can be added the per-
manent (world-systems theory would say “necessary”) disparities of wealth of the capitalist
world-economy, and the social tensions that result from the politics of trying to achieve eco-
nomic success, or “development,” within the world-economy’s structural constraints. If the
importance of long-standing structures in the causes of terrorism is acknowledged, then
longer term perspectives on the geopolitics of counter-terrorism should also prove useful. It
is to this concern we now turn.
8A few cautionary statements are in order regarding these findings. Quantitative analyses of terrorism are
somewhat rare for a variety of reasons, including a relative lack of large and robust publicly available data sets and
the subjective nature of the phenomenon. While the RAND-MIPT/START data allow researchers to undertake large-
sample, cross-sectional quantitative analyses, like all large data sets, it is likely to be somewhat noisy. Most of the
potential noise is probably tied to the collection method. Media-driven terrorism data has received some attention in
the research literature, and problems in the data used for this study could stem from a number of sources; biases by
either reporting agencies or data set analysts, such as a differential focus on highly developed states, non-state
groups, or large-scale attacks, may act to filter out some incidents. Because of the subjective and contested nature of
terrorism, there is also a clear possibility of including activities in the data that reasonably might not be construed as
terrorism. The sum effect of these issues (and others not considered here) might serve to amplify the presence of ter-
rorist incidents in some states and dampen them in others. These are common challenges and concerns to all
engaged in quantitative studies of terrorism and deserve explicit acknowledgment as circumstances that can condi-
tion findings.
Two additional issues bear mentioning: the possibility of endogeneity and/or dependence in the data. On the
first issue, there may be endogenous relationships between terrorism and development—the attacks of 9/11, the Bali
bombings, and the ongoing insurgency in Iraq are all examples of where terrorism had a subsequent negative eco-
nomic impact on states. In most cases, individual terrorist incidents do not seem to have a significant impact on HDI
scores. However, in some circumstances large scale and/or long term insurgencies such as Iraq, Afghanistan and the
Palestinian Territories have obvious deleterious effects on development over time. The risk of this issue seriously
affecting the results is somewhat lessened both by using data aggregated over a number of years and by using meth-
ods that reduce the influence of large outliers. On the second issue, independence is an assumption unlikely to be
met in any study that uses the state as the unit of analysis. Developmental data used for this study does show signif-
icant measures of spatial autocorrelation, both globally and locally. However this may be due to the high level of
data aggregation necessary whenever the state is the unit of analysis. The terrorism data shows far less global spatial
autocorrelation and local measures vary strongly by region. Again the level of aggregation necessary for this study
makes the potential for dependence difficult to bypass but important to acknowledge. Future research in this area
should strive to evaluate and address these issues as much as possible by using disaggregated data.
Flint.fm Page 163 Thursday, February 19, 2009 3:18 PM
164 EURASIAN GEOGRAPHY AND ECONOMICS
THE GEOGRAPHY OF THE GLOBAL WAR ON TERROR
Terrorism has been a feature of global politics since the beginning of the 20th century,
and its changing features are a function of global processes and structures related to warfare,
imperialism, state-building, and the structure of the world-economy. However, as terrorism
has become globalized, counter-terrorism has also evolved into a global geopolitical practice.
On the one hand, this is manifest in international cooperation among sovereign states through
organizations such as INTERPOL. On the other, the agenda and form of counter-terrorism is
a matter of power politics, and is one way in which powerful states attempt to further their
goals in competition with one another. The United States, as the most powerful (or hege-
monic) country in the world, has defined its current foreign policy agenda under the guise of
a global counter-terrorist project. Though stimulated by the terrorist attacks of September 11,
2001 and their specific targeting of the U.S., the Global War on Terror has been defined as a
hegemonic project in the interests of the whole world (Flint and Falah, 2004). In this final
section, we will discuss how the global trends in terrorism we have discussed previously
have been manipulated to justify a specific foreign policy agenda for the United States, and
how this is best interpreted within the historical processes of violent challenges to U.S.
authority.
Since 2001, in the wake of the September 11 attacks, discussions of terrorism have been
dominated by the actions of al-Qaeda and the counter-terrorist project of the United States,
the Global War on Terror (GWOT). This is unfortunate for a number of reasons. First, it dis-
tracts from the actual geography of terrorism by promoting the U.S. as primary victim when
the analysis in the previous section shows this to be untrue. Second, it also confuses our
understanding of the causes and goals of contemporary terrorism. The now familiar “with us
or against us” plea of President George W. Bush has been discussed as a means of corralling
states to comply with U.S. foreign policy imperatives. It has another implication. Most acts
of terrorism have been interpreted through the twin prisms of U.S. foreign policy institutions
and the media as connected to an Islamic fundamentalist challenge to the United States.
Other acts of terrorism, such as recent ones in China, are either ignored or tied to Islam to
make sure the dominant trope of Muslim fundamentalist–inspired terrorism is not chal-
lenged. The broad understanding of terrorism has become one that is restricted to a single
component of Rapoport’s fourth wave. Terrorism is mainly understood as solely motivated
by religion, and it is the actions of Islamic fundamentalists that are placed in the spotlight.
Such an interpretation is a representation of broader trends by the U.S. in order to justify its
contemporary foreign policy agenda.
This is not to deny the reasoning behind this focus. The United States, Great Britain,
India, and Spain have all been victims of terrorist attacks that have understandably focused
the attention of security agencies, the media, and public. However, the challenge to the U.S.,
and more generally its Western allies, is only one interpretation of contemporary terrorism.
On the one hand, many ongoing terrorist campaigns remain rooted in the goals and practices
of Rapoport’s third wave. They are matters of ethno-nationalism rooted in processes of state-
building and post-colonialism: the conflicts in Algeria and Sri Lanka, for example. An alter-
native interpretation is that the motivations of resistance to state formation, and the growing
importance of “human sympathy” in global war trends, are the dominant causes of conflict,
with terrorism just one manifestation of such violence. These broader structures underlying
terrorism are not addressed by the U.S., as the GWOT agenda is abstracted from these trends
and is a narrow agenda around U.S. concerns.
Flint.fm Page 164 Thursday, February 19, 2009 3:18 PM
FLINT AND RADIL 165
By framing terrorism within the narrow and politicized agenda of the GWOT, we have
lost the big picture of global conflict trends that are the intersection of colonialism and post-
colonial trends, and the violence inherent in the imperfect and always incomplete construc-
tions of nation-states. However, the current situation is not simply more of the same. There
are other cross-cutting processes that do make the form of contemporary terrorism different
from previous ones, and especially challenging for counter-terrorism activities.
The processes of internationalization identified by Rapoport in the third wave were just
one manifestation of increasing social, political, economic, and cultural integration. The pro-
cess intensified and increased across all spheres of society to fundamentally alter the way
human society is organized, commonly referred to as globalization. Globalizations is a more
appropriate term, as all social interactions, including war (Grenfell and James, 2009) have
become globalized in different but related ways.
The globalization of armed conflict, including terrorism, is connected to another signifi-
cant process, the political role of the United States. The power of the U.S. is being chal-
lenged, or has been significantly depleted, in a number of ways. Its post-war economic
prowess has suffered from the economic strength of competitors (notably the European
Union, Japan, and now China). Its financial leadership had been in question since the col-
lapse of the Bretton Woods agreement in 1975, and increasingly so with the financial crisis of
2008. However, its cultural power remains strong; Bollywood has yet to mount a global chal-
lenge to Hollywood’s supremacy. Most importantly for the focus of this essay, the U.S. is far
and away the strongest military power in the world in terms of the hardware and personnel it
can mobilize as well as its global reach.
However, military resources that can be mobilized are not a good indication of the abil-
ity of the U.S. to achieve its goals. Rather, its ability to effect change through relations with
other geopolitical entities is a better measure of U.S. power in foreign policy, including mili-
tary operations. Here the story of overwhelming power is harder to write. The military suc-
cesses of the U.S. since the Vietnam War have been either minuscule or incomplete. The
invasion of Grenada in 1983 was a baby-step for the U.S. military trying to find its way in the
wake of the Vietnam humiliation: a success, but a geopolitical blip. The same could be said
of the invasion of Panama. Furthermore, both of these actions were relatively close to home.
The baby-steps became a more aggressive stride with air attacks on Libya, but a bloody
humiliation in Lebanon. The first Gulf War demonstrated the ability to reach across the globe
and achieve limited military objectives, but not the will and commitment to turn them into
political victories. The targeting of the U.S. by Islamic fundamentalist terrorists, or the emer-
gence of al-Qaeda, became evident in visible attacks such as those on the Word Trade Center
in 1993 and the USS Cole in 2000. The increased confidence of the military to project its
power in the face of a territorial inter-state dispute (the first Gulf war) was accompanied by
caution and consternation as to how to counteract a globalizing network of terrorists. Cruise
missile attacks upon the Sudan and Afghanistan were critiqued for being both wimpy and
counter-productive, in light of whether the “factory” in Sudan was a legitimate military tar-
get.
Overall, the geopolitical context leading up to the terrorist attacks of 9/11 and the subse-
quent declared GWOT was defined by a weakening of U.S. economic power. In addition, the
military was resource rich but still wary of its ability to project power across the globe,
though feeling the necessity to demonstrate that it still had that capability. Conversely, al-
Qaeda was gaining force through its opposition to the U.S. presence in southwest Asia and its
support for Israel. For al-Qaeda the struggle was framed within the legacies of colonialism,
and especially neo-colonial struggles for control of states that were, in their eyes, the corrupt
Flint.fm Page 165 Thursday, February 19, 2009 3:18 PM
166 EURASIAN GEOGRAPHY AND ECONOMICS
products of Western colonialism. The goal was the expulsion of the United States from the
Middle East in order to facilitate revolutionary change within existing states. In other words,
the broad framework used by O’Loughlin and van der Wusten (1993) is still very relevant:
colonialism and its legacies intersect with competition to control the state apparatus. These
processes are coupled with the recognition, by both Holsti and Rapoport, of the increasing
causal strength of “human sympathy.”
Globalization, a mixed pattern of U.S. power capabilities, and a terrorist threat that is
simultaneously fighting a “cosmic war” over issues of territory and sovereignty make for a
complex geopolitical situation. The global network of al-Qaeda is indeed a noteworthy geo-
graphic feature. However, its ability to recruit in, for example, Great Britain while maintain-
ing a fortress-like strength in the tribal areas of Pakistan, demonstrates that the success of the
network requires that it be grounded in places with unique histories and attributes.
Looking at the situation from the perspective of the United States, the geopolitical
conundrum is how to counteract a foe that is organized in a global network while political
globalization is far from complete and the legal basis for international politics remains the
Westphalian system of sovereign states. In geographic language, how do territorial states
counter a hostile global network? The simple, perhaps cynical, answer is through the exercise
of a new form of imperialism, the GWOT being its manifestation (Gregory, 2004).
The initial campaigns of the GWOT were exercises in traditional military power: the use
of massive military might to invade and occupy sovereign territories against their will. The
initial success of the military actions in Afghanistan and then Iraq were followed by transpar-
ent attempts to put the state under the control of interests sympathetic to the U.S. In
Afghanistan, President Hamid Karzai was generally a success story. In Iraq, the clumsy
attempted coronation of Ahmed Chalabi was an embarrassing failure. The subsequent devel-
opment of both occupations has been a combination of military force to repress insurgencies
and an attempt to control the political process to the benefit of the occupying force. Both
strategies have met with limited success in both countries.
In addition, the surprising foreign policy success of the United States in gaining Pakistan
President Musharref’s commitment to join the U.S. anti-terror coalition was soon tarnished.
In an interesting geopolitical twist, the ability of the U.S. to dominate Pakistan’s political-
military state was undermined by a combination of democratic forces and armed resistance to
the state apparatus. Pakistani civil society opposed the alliance with the United States and
actions in the name of the GWOT. The northern Federally Administered Tribal Areas (FATA)
region became a base for Islamic fundamentalists pushed out of Afghanistan. A strange neo-
colonial balance is now in effect. The democratically elected Pakistani government tries to
distance itself from the dominance of the United States, but at the same time acquiesces to
U.S. attacks upon alleged terrorist/insurgent camps within its sovereign borders.
The geography of the conflict has become increasingly complicated. The terrorist net-
work is glocalized, transcending borders but grounded on the specificities of places. The
counter-terrorist aims of the GWOT were ostensibly targeted toward a network but immedi-
ately became the invasion and occupation of sovereign spaces, the removal of existing lead-
ers, and a reconstruction of the governmental apparatus. These acts were resisted and the
GWOT morphed so that it is more accurately a GSWOI (Geographically Specific War on
Insurgents).
The territorialization of the GWOT was evident in the focus upon “terrorist safe
havens” and the related geopolitical representations of “failed states” and “rogue states.”
These political spaces were constructed within an orientalist framework of otherness and
barbarity that was deemed to require violent occupation and control (Gregory, 2004; Flint,
Flint.fm Page 166 Thursday, February 19, 2009 3:18 PM
FLINT AND RADIL 167
2007). The flip-side of the coin was the need for the counter-terrorist state to construct its
own spaces within which the norms of the rule of law could be violated. The spaces of
exception in Guantanamo and the mysterious facilities to which people were “rendered”
and subsequently tortured required the apparent “unpacking” of the territory of Western
states fighting the GWOT into parcels that contained the processes of the democratic norm
and those spaces in which a democratic country could act like an imperial despot (Hannah,
2006). The attempt, one that ultimately failed, to fragment the U.S. in this way (into legal
and extra-legal zones, or spaces of exception “outside” the norm) was mirrored by the
attempt to essentialize other states and cultures. Hence Afghanistan needed to be con-
quered in its entirety to save the civilized, democratic, and moral and just U.S. (that was
represented as separate and untarnished from its acts of torture and brutality in
Guantanamo or Abu Ghraib).
Although the broad geopolitical processes of colonialism and nation-state building are
the foundations of al-Qaeda, the political geography of terrorism has changed dramatically
because of the processes of U.S. decline and globalization. Networks interacting with sov-
ereign spaces led to occupation and invasion, which in turn led to neo-colonial state build-
ing, and consequently led to terrorism morphing into insurgency. The GWOT required
geopolitical representations of orientalized “bad” places and the attempt to fragment the
political space of the U.S. so an illusion of democratic normality could survive the dark
practices of the GWOT. The latter part of this has failed with the impending closure of
Guantanamo. Ironically, the apparent termination of the most visible “space of exception,”
Guantanamo, means that the geopolitical actions of the U.S. will be based upon a more
open legal terrain, requiring clarity in the geopolitical vision and practice of an Obama-led
GWOT.
The specifics of the GWOT are best understood as manifestations of broader trends.
The nature of the terrorist threat is organized through transnational networks that gain their
meaning and power by being grounded in particular localities. Terrorist goals and motiva-
tions are largely framed through religious rhetoric. Hence, the terrorism targeted by the
GWOT displays the tendencies towards internationalization and religious motivations
identified by Rapoport that have developed over the past 100 years or so. Also, the territo-
rial aims of terrorists remain and these can be understood within the constant features of
modern warfare: imperialism and state-building. In addition, the material realities of rela-
tive deprivation, a constant feature of the capitalist world-economy, provide a structural
setting for feelings of injustice. These structural trends intersect with the trajectory of U.S.
hegemony and the way that the al-Qaeda network–based challenge to the U.S. has become
a territorialized project to assert military and political power across the globe—a project
that raises more questions about U.S. failings than it does acclamations of its assumed
legitimacy.
CONCLUSION
The goal of this paper has been to frame discussions of contemporary terrorism within
broader historical trends and macro-geopolitical structures. The result is an understanding
of terrorism as simply another manifestation of warfare, one with a different set of tactics
than violence that is legitimated under the heading “war.” Long-term processes of state
formation and imperialism are the underlying causes of terrorism, with terrorists claiming
they are resisting the domination of their territory by rich and, mostly, Western powers
and/or fighting for a more just political congruence of communal identity and state appara-
Flint.fm Page 167 Thursday, February 19, 2009 3:18 PM
168 EURASIAN GEOGRAPHY AND ECONOMICS
tus. The trend toward conflicts motivated by “human sympathy” has been reflected in the
changing nature of terrorism, but the root causes of contemporary terrorism can still be
traced back to the processes of imperialism and state building. Economic marginalization
is also a factor, and terrorism can be situated in the dynamic and competitive zone of the
world-economy in which the experience of relative deprivation is coupled with state inter-
vention aimed at creating economic winners and losers within a state in the name of
“development strategies.”
The broad framework for understanding terrorism that we offer has implications for
analysis and policy. With regard to policy, development strategies cannot be seen as a simple
panacea, but must be mixed with attempts to provide means to alleviate social tensions that
result from promoting certain economic sectors over others in development strategies. In
terms of analysis, the posited relationship between broad trends in warfare and terrorism sug-
gest opportunities for analyses that connect incidences of terrorism with the occurrence and
processes of inter- and intra-state warfare.
In the context of the transition from the administration of President George W. Bush to
that of Barack Obama, there is a sense of optimism that the GWOT will either be wound
down, or the tactics fundamentally altered. However, the concentration upon broad trends
calls for a less optimistic note, as the incoming administration may only have limited possi-
bilities to counteract current trends. The challenge to the United States is unlikely to fade
given the combination of perceptions that it is still the global power involving itself in poli-
tics across the globe but has weakened and is, therefore, open to challenge. The morphing of
the GWOT into messy anti-insurgency campaigns means that withdrawal is likely to be frus-
trating and complicated. Are withdrawals from Iraq and Afghanistan deemed possible? Will
matters get worse if withdrawal is complete? And would withdrawal be seen as U.S. military
failure, meaning that the baby-steps of Grenada and Panama did not alter the trend of U.S.
inability to project military force across the globe? The expectation from our analysis is that
the GWOT is likely to connect to other forms of warfare, and so the situation is not likely to
improve over the near term.
At the time of writing, the first foreign policy initiative of President-elect Obama’s
administration has seemingly been defined: a massive increase in troop deployment in
Afghanistan. It appears unlikely that NATO allies or other countries will provide support.
The U.S. thus looks set to continue a policy of attempting to control sovereign territory to
counteract what is portrayed as a transnational and religiously motivated terrorist threat. His-
tory does not provide grounds for optimism that this particular military campaign in
Afghanistan by an outside power will be successful. It is difficult to draw any other conclu-
sion than the next few years will see another failed extra-territorial mission by the United
States, one that will be costly in terms of money, personnel, and international support. The
GWOT will be continued, but it is likely to accelerate, rather than arrest, the decline in U.S.
power and will not make any serious inroads into negating a terrorist threat.
REFERENCES
Amster, Randall, “Perspectives on Ecoterrorism,” Contemporary Justice Review, 9, 3:287–301, 2006.
Baudrillard, Jean, “L'esprit du terrorisme,” The South Atlantic Quarterly, 101, 2:403–415, 2002.
Bergesen, Albert J. and Omar Lizardo, “International Terrorism and the World-System,” Sociologi-
cal Theory, 22, 1:38–52, 2004.
Booysen, Frederik, “An Overview and Evaluation of Composite Indices of Development,” Social
Indicators Research, 59, 2:115–151, 2002.
Flint.fm Page 168 Thursday, February 19, 2009 3:18 PM
FLINT AND RADIL 169
Boyns, David and James D. Ballard, “Developing a Sociological Theory for the Empirical Under-
standing of Terrorism,” The American Sociologist, 35, 2:5–25, 2004.
Bravo, Ana Bela Santos and Carlos Manuel Mendes Dias, “An Empirical Analysis of Terrorism:
Deprivation, Islamism, and Geopolitical Factors,” Defence and Peace Economics, 17, 4:329–341,
2006.
Bush, George W., “The National Security Strategy of the United States of America,” 2002 [http://
www.whitehouse.gov/nsc/nss.pdf].
Callaway, Rhonda L. and Julie Harrelson-Stephens, “Toward a Theory of Terrorism: Human Secu-
rity as a Determinant of Terrorism,” Studies in Conflict and Terrorism, 29, 7:679–702, 2006.
Chomsky, Noam, Rogue States: The Rule of Force in World Affairs. Boston, MA: South End Press,
2000.
CIA (U.S. Central Intelligence Agency), The 2008 World Factbook. Washington, DC: U.S. Central
Intelligence Agency, 2008 [https://www.cia.gov/library/publications/the-world-factbook/].
Cooper, H. H. A., “Terrorism: The Problem of Definition Revisited,” American Behavioral Scientist,
44, 6:881–893, 2001.
Cutter, Susan L., D. B. Richardson, and T. J. Wilbanks, eds., The Geographical Dimensions of Ter-
rorism. New York, NY: Routledge, 2003.
Elden, Stuart, “Terror and Territory,” Antipode, 39, 5:821–845, 2007.
FBI (Federal Bureau of Investigation), “Congressional Testimony: Addressing the Threat of Animal
Rights Extremism and Eco-Terrorism,” 2005 [http://www.fbi.gov/congress/congress05/
lewis051805.htm].
Flint, Colin, “Geographies of Inclusion/Exclusion,” in Susan L. Cutter, Douglas B. Richardson, and
Thomas J. Wilbanks, eds., The Geographical Dimensions of Terrorism. New York, NY: Routledge,
2003a, 53–58.
Flint, Colin, “Terrorism and Counterterrorism: Geographic Research Questions and Agendas,” The
Professional Geographer, 55, 2:161–169, 2003b.
Flint, Colin, “Netwar, the Modern Geopolitical Imagination, and the Death of the Civilian,” in M.
Innes, ed., Denial of Sanctuary: Understanding Terrorist Safe Havens. Westport, CT: Praeger
Security International, 2007, 34–48.
Flint, Colin and Ghazi-Walid Falah, “How the United States Justified Its War on Terrorism: Prime
Morality and the Construction of a ‘Just War’,” Third World Quarterly, 25, 8:1379–1399, 2004.
Flint, Colin and Peter J. Taylor, Political Geography: World-Economy, Nation-State, and Locality,
fifth ed. Harlow, UK: Pearson Education Limited, 2007.
Gregory, Derek, The Colonial Present: Afghanistan, Palestine, Iraq. Oxford, UK: Blackwell, 2004.
Grenfell, Damian and Paul James, “Debating Insecurity in A Globalizing World,” in D. Grenfell and
P. James, eds., Rethinking Insecurity, War, and Violence: Beyond Savage Globalization?
Abingdon, UK and New York, NY: Routledge, 2009, in press.
Gurr, Ted R., Why Men Rebel. Princeton, NJ: Princeton University Press, 1971.
Hannah, Matthew, “Torture and the Ticking Bomb: The War on Terrorism as a Geographical Imagina-
tion of Power/Knowledge,” Annals of the Association of American Geographers, 96, 3:622–640,
2006.
Harbom, Lotta, Erik Melander, and Peter Wallensteen, “Dyadic Dimensions of Armed Conflict,
1946–2007,” Journal of Peace Research, 45, 5:697–710, 2008.
Hoffman, Bruce, Inside Terrorism: From the Iran Hostage Crisis to the World Trade Center Bombing.
New York, NY: Columbia University Press, 1999.
Holsti, Kalevi J., Peace and War: Armed Conflicts and International Order, 1648–1989. New York,
NY: Cambridge University Press, 1991.
Joseph, Mallika and Devyani Srivasatava, “Left Extremism: The Naxal Conflict in India,” in D.
Suba Chandran and P. R. Chari, eds., Armed Conflict in South Asia 2008: Growing Violence. New
Delhi, India and London, UK: Routledge: 2008.
Johnston, Ron J., John O’Loughlin, and Peter J. Taylor, “The Geography of Violence and Pre-
mature Death," in Rumo Väyrynen, Dieter Senghaas, and Christain Schmidt, eds., The Quest
Flint.fm Page 169 Thursday, February 19, 2009 3:18 PM
170 EURASIAN GEOGRAPHY AND ECONOMICS
for Peace. Beverly Hills, CA: Sage and the International Social Science Council, 1987, 241–
259.
Juergensmeyer, Mark, Terror in the Mind of God: The Global Rise of Religious Violence. Berkeley,
CA and Los Angeles, CA: University of California Press, 2003.
Krueger, Alan B. and David D. Laitin, “Kto Kogo? A Cross-Country Study of the Origins and Tar-
gets of Terrorism,” in P. Keefer and N. Loayza, eds., Terrorism, Economic Development, and
Political Openness. New York, NY: Cambridge University Press, 2008, 149–173.
Krueger, Alan B., and Jitka Malečková, “Does Poverty Cause Terrorism?: The Economics and Edu-
cation of Suicide Bombers,” The New Republic, 226:27–33, 2002.
Laquer, Walter, The Age of Terrorism. Boston, MA: Little, Brown, 1987.
Laquer, Walter, A History of Terrorism. New Brunswick, NJ: Transaction Books, 2001.
Mathewson, Kent and Michael Steinberg, “Drug Production, Commerce, and Terrorism,” in Susan L.
Cutter, Douglas B. Richardson, and Thomas J. Wilbanks, eds., The Geographical Dimensions of
Terrorism. New York, NY: Routledge, 2003, 59–66.
Mosse, George L., The Nationalization of the Masses: Political Symbolism and Mass Movements in
Germany from the Napoleonic Wars Through the Third Reich. New York, NY: Howard Fertig,
1975.
Murphy, Alexander B., “The Space of Terror,” in Susan L. Cutter, Douglas B. Richardson and
Thomas J. Wilbanks, eds., The Geographical Dimensions of Terrorism. New York, NY: Routledge,
2003, 47–52.
O’Loughlin, John and Herman van der Wusten, “Political Geography of War and Peace,” in Peter J.
Taylor, ed., Political Geography of the Twentieth Century: A Global Analysis. London, UK:
Belhaven, 1993, 63–114.
Piazza, James A., “Rooted in Poverty?: Terrorism, Poor Economic Development, and Social Cleav-
ages,” Terrorism and Political Violence, 18, 1:159–177, 2006.
Post, Jerrold, “Congressional Testimony: The Mind of the Terrorist: Individual and Group Psychology
of Terrorist Behavior,” 2001 [http://crab.rutgers.edu/~goertzel/PostTerrorism.htm].
RAND-MIPT/START, “RAND-MIPT Terrorism Incident Database/START Global Terrorism Data-
base,” 2008 [http://www.start.umd.edu/].
Rapoport, David C., “The Four Waves of Modern Terrorism,” in Audrey Cronin and J. Ludes, eds.,
Attacking Terrorism: Elements of a Grand Strategy. Washington, DC: Georgetown University
Press, 2004, 46–73.
Said, Edward, Orientalism. New York, NY: Vintage Books, 1978.
Schmid, Alex P. and Albert J. Jongman, Political Terrorism: A New Guide to Actors, Authors, Con-
cepts, Data Bases, Theories and Literature. New Brunswick, NJ: Transaction Books, 1988.
Silke, Andrew, “The Devil You Know: Continuing Problems with Research on Terrorism,” Terrorism
and Political Violence, 13, 2:1–14, 2001.
Smelser, Neil J. and Faith Mitchell, eds., Terrorism: Perspectives from the Behavioral and Social Sci-
ences. Washington, DC: The National Academies Press, 2002.
Synder, David and Edward L. Kick, “Structural Position in the World System and Economic Growth,
1955–1970: A Multiple-Network Analysis of Transactional Interactions,” American Journal of
Sociology, 84, 5:1096–1126, 1979.
Tilly, Charles, Coercion, Capital, and the European States, AD 990–1990. Cambridge, MA:
Blackwell, 1990.
Turk, A u s t i n T., “Sociology of Terrorism,” in Annual Review of Sociology, 30, 2004, 271–286.
UN (United Nations), A More Secure World: Our Shared Responsibility: Executive Summary. New
York, NY: United Nations Press, 2004.
UNDP (United Nations Development Programme), Annual volumes of the Human Development
Report, 1998–2006 [http://hdr.undp.org/eu].
van der Wusten, Herman, “Violence, Development, and Political Order,” in Colin Flint, ed., The
Geography of War and Peace: From Death Camps to Diplomats. Oxford, UK and New York, NY:
Oxford University Press, 2005, 61–84.
Flint.fm Page 170 Thursday, February 19, 2009 3:18 PM
FLINT AND RADIL 171
Victoroff, Jeff, “Managing Terror: The Devilish Traverse from a Theory to a Plan,” in Jeff Victoroff,
ed., Tangled Roots: Social and Psychological Factors in the Genesis of Terrorism. Amsterdam,
The Netherlands: IOS Press, 2007, 1–19.
Flint.fm Page 171 Thursday, February 19, 2009 3:18 PM
