Jihad As Grand Strategy

i
Jihad as Grand Strategy

ii
iii

Islamist Militancy, National
Security, and the Pakistani State
S. Paul Kapur
1
iv
1
Oxford University Press is a department of the University of Oxford. It furthers
the University’s objective of excellence in research, scholarship, and education
by publishing worldwide. Oxford is a registered trade mark of Oxford University
Press in the UK and certain other countries.
Published in the United States of America by Oxford University Press

198 Madison Avenue, New York, NY 10016, United States of America.
© Oxford University Press 2017
All rights reserved. No part of this publication may be reproduced, stored in

a retrieval system, or transmitted, in any form or by any means, without the
prior permission in writing of Oxford University Press, or as expressly permitted
by law, by license, or under terms agreed with the appropriate reproduction
rights organization. Inquiries concerning reproduction outside the scope of the
above should be sent to the Rights Department, Oxford University Press, at the
address above.
You must not circulate this work in any other form

and you must impose this same condition on any acquirer.
CIP data is on file at the Library of Congress

ISBN 978–0 –19–9 76852–3
9 8 7 6 5 4 3 2 1
Printed by Sheridan Books, Inc., United States of America
v
CON T EN T S
1. Introduction 1
2 . The Logic of a Militant Proxy Strategy 13
3. Partition and an Emerging Strategy 32
4. Pakistan’s Militant Strategy Evolves 51
5. Kashmir and Afghanistan Reprise 81
6. Jihad as Grand Strategy: An Assessment 111
7. The Future: Can Pakistan Abandon Jihad? 127
Notes 143
Index 173
vi
1
CHAP T ER 1
Introduction
T errorism’s ascendance as one of the world’s leading strategic dan-

gers has been a central development of the post–Cold War security
environment. Its effects have included a fundamental reorientation of the
United States’ grand strategy, de-emphasizing long-standing postures of
deterrence and containment in favor of prevention and preemption; the
launch of a global “war on terror”; and the initiation of wars in Afghanistan
and Iraq.1 Scholars and analysts have generated a voluminous literature
attempting to identify demographic, economic, psychological, ideo-
logical, strategic, and other patterns in terrorist violence.2 Although the
nature and prevalence of such patterns are a matter of vigorous debate,
one recurring theme concerning terrorism is strikingly clear: A dispro-
portionate amount of it has been linked to Islamist militants3 based in
Pakistan.4
For example, members of the team that attacked the United States on
September 11, 2001, received training from senior al-Qaeda operative
Khalid Shaikh Mohammed in Karachi. He later wired them funds to sup-
port their preparations for the attacks. Khalid Shaikh Mohammed was
subsequently arrested in a predawn raid on his safe house in Rawalpindi,
the twin city of the Pakistani capital of Islamabad and headquarters of
the Pakistan Army. 5 Al-Qaeda leader Osama bin Laden, the most hunted
terrorist in history, was discovered and killed in a Pakistani garrison town
crawling with security personnel, in the shadow of Pakistan’s military
academy. He apparently had been there, living in a large, custom-built
home, for over five years.6 The leader of the group that carried out the
July 2005 London bomb attacks had undergone paramilitary training
in Pakistan. In addition, the group received bomb-making instructions
over the telephone from a caller in Rawalpindi.7 And the perpetrators of
the Mumbai attacks in November 2008 were members of the Pakistani
2
( 2 ) Jihad as Grand Strateg y
militant group Lashkar-e-Taiba. They trained for their mission under

the tutelage of active and retired Pakistani army and intelligence officers
in Pakistan and Pakistani Kashmir, departed by sea for Mumbai from
Karachi, and were directed in real time during the operation by handlers
in Pakistan.8
Acts like these are important not only because they kill significant
numbers of people around the world but also because they undermine
international stability, instigating conflict that can ultimately result in far
more death and destruction than the acts of terror themselves. The 9/11
attacks drove the United States to launch two major wars, which are esti-
mated to have cost the United States between $4 trillion and $6 trillion9
and resulted in well over one hundred thousand deaths.10 At present,
Pakistan-linked militant attacks threaten to incite a large-scale Indo-
Pakistani conflict. Given India’s and Pakistan’s possession of nuclear
weapons, the outcome of such a confrontation could be catastrophic. The
detonation of a single 50-k iloton weapon over New Delhi or Islamabad
would likely kill more than five hundred thousand people and result in
more than one million casualties.11 The consequences of Pakistan-based
militancy thus reach far beyond particular acts of violence or even South
Asian regional politics—they severely damage stability and security
across the globe. Not surprisingly, Pakistan is widely viewed, in the words
of the Economist, as “the world’s most dangerous place.”12 A survey of
leading international relations experts by the journal Foreign Policy tied
Pakistan with Iran as the top security challenge facing the United States.
Indeed, more than twice the number of respondents identified Pakistan as
the United States’ leading security threat as chose China.13
This book seeks to understand the connection between Pakistan and
Islamist militancy. Where did the link between the Pakistani state and
Islamist militants come from? How did it evolve over time? What have
been its effects on Pakistani strategic interests? How has it affected
regional and international security? The book addresses these questions
by departing from standard approaches to militancy in Pakistan, which
treat it either as an inexorable sociological phenomenon, as a historical
accident, or as an obviously self-destructive policy devised by short-
sighted generals and self-interested politicians. Instead, the book explores
the Pakistani state’s use of militancy as a rational strategy, designed to
generate security for itself in the face of an extremely demanding domes-
tic and international environment. It argues that Pakistan’s use of Islamist
militancy is the latest incarnation of a sophisticated asymmetric warfare
campaign,14 deliberately developed and prosecuted since the Pakistani
state’s founding. From 1947 to the present day, Pakistan15 has used reli-
giously motivated nonstate actors as strategic tools to confront stronger
3
I n t r o d u c t i o n ( 3 )
adversaries and shape its strategic environment without the costs and
risks of direct combat, and to help promote internal cohesion to compen-
sate for Pakistan’s weak domestic political foundations.16
THE LITERATURE
The Pakistan–militant nexus has come under intense public scrutiny in

recent years. Most discussions of the problem, however, are of limited
utility in understanding its true nature. The popular press often charac-
terizes the Pakistan–militant relationship as the result simply of Pakistani
evildoing or paranoid bumbling, without identifying the strategic factors
that originally created the connection and continue to drive it today.17
More careful journalistic reporting on Pakistani militancy is often lim-
ited to discrete aspects of the problem, such as the agenda or leadership of
a given militant group, or the unfolding of a particular terrorist operation.
Even when broader in scope, such work is mainly descriptive and does not
explain the deeper causal forces underlying Pakistan’s militancy problem
and their impact on current Pakistani security policy.18
The scholarly literature on Islamist militancy in Pakistan, though less
extensive than popular commentary and reporting, is growing apace.
This work provides far more context and analysis than do discussions in
the popular press and offers valuable insights into the Pakistan–militant
problem. It consists of four main camps. They focus on in-depth empirical
documentation, sociopolitical variables, politico-military and organiza-
tional strategies, and broad historical narratives.
The first camp seeks to provide a careful empirical record of the
Pakistan–militant connection. Authors in this camp offer extremely
detailed discussions of Pakistan’s historical involvement with Islamist
militant groups, including fine-g rained accounts of Pakistani leaders’
decision-making processes and development of ongoing militant strate-
gies and relationships with the militants.19 Some authors in the empirical
camp focus their attention on particular militant organizations, such as
Lashkar-e-Taiba, or on the use of militants in specific conflicts, such as
Pakistan’s use of the Taliban in Afghanistan.20
The second camp in the scholarly literature emphasizes the role of
societal and political variables in causing Pakistan’s militancy problem. It
describes the processes of Islamization and militarization that have taken
place within Pakistan in recent decades and explains how they have cre-
ated an environment in which militancy could thrive. Works in this camp
focus on a wide range of specific causes, including the military dominance
of Pakistani society, the lack of experienced political leadership following
4
independence, the impact of the Iranian revolution and the anti–Soviet-

Afghan war, and the policies of Zia-u l-Haq.21
A third category of scholarship does not seek to link Pakistan-based
militancy to broad sociopolitical trends. Rather, it characterizes the phe-
nomenon as a politico-military and organizational tool that Pakistani
leaders have used to promote their personal and bureaucratic interests.
Advancing the Kashmir cause through support for jihad enables politi-
cal and military leaders to corner significant financial resources and to
bolster their legitimacy by burnishing their Islamic bona fides. It also
allows the army to advance its organizational agenda by maximizing its
share of state resources and dominating national policymaking. The army
also uses militancy in an effort to honor its ideological commitments,
protecting Pakistan’s Islamic identity by challenging India’s dominant
position in the region. This competitive behavior is enabled by Pakistan’s
generous strategic endowments, especially its important geographic posi-
tion, which reduces the incentives for Pakistani leaders to devise rational
domestic and foreign policies and produce the public goods needed to
undergird a modern, prosperous state.22
A final group of scholars discusses Pakistan-based militancy in the
context of broad narratives, which trace the arc of Pakistani political and
social history from independence to the present. Militancy is not the pri-
mary focus of these works; it is one subject among the many issues that
they address. These issues include the institutional history of the Pakistan
Army, in which emergent praetorianism facilitated the Pakistan Army’s
adoption of militancy as a central strategic tool; the troubled United
States–Pakistan relationship, which, by providing Pakistan with exten-
sive US military and development aid, has enabled Pakistan to support
militancy; and the regime of Zia-ul-Haq, whose Islamization policies
promoted the religious extremism that undergirds the Pakistan–militant
connection.23
Each scholarly camp makes important contributions to our under-
standing of Pakistan’s relationship with Islamist militancy. For exam-
ple, largely empirical, descriptive work offers more detailed and reliable
information than is readily available elsewhere. By examining the general
history of Pakistan’s relationship with militancy, the specific character-
istics of particular jihadist groups, and the use of militants in particular
conflicts, such work can significantly improve the quality of more con-
ceptually driven analyses of the Pakistan–militant connection. Such
studies are also especially useful from a policy perspective. For exam-
ple, detailed exposition of the characteristics of an important organiza-
tion like Lashkar-e-Taiba can improve policymakers’ understandings of
5
its motives, strengths, and weaknesses, and help to devise techniques of

more effectively combating it.
The sociopolitical camp in the literature explains the complex pro-
cesses by which the Islamization and militarization of Pakistani society
actually occurred. It also shows that, in the absence of these develop-
ments, Pakistan’s militant problem would probably never have emerged,
and almost certainly would not have reached its current level of urgency.
This strand of the literature thus demonstrates that societal trends can be
understood as a deep or permissive cause of Pakistan’s current difficulties
with Islamist militancy.
The politico-military and organizational strategy camp of the litera-
ture demonstrates that Pakistan’s use of militancy has been an extremely
important tool both in Pakistan’s broad security competition with India
and in its efforts to undo the territorial status quo in Kashmir. The strat-
egy has significantly attrited Indian military and economic resources and
has enabled Pakistan to keep open the possibility of redrawing Kashmiri
territorial boundaries. Supporting militancy also has played a significant
role in advancing the interests of individual Pakistani political figures. For
example, leaders as diverse as Zia-ul-Haq, Benazir Bhutto, Nawaz Sharif,
and Pervez Musharraf have bolstered their domestic political credentials
by publicly supporting the militants, whom they have referred to as “free-
dom fighters” capable of waging “one thousand years of war” with India.24
Instrumental military and political calculations such as these explain a
good deal of Pakistan’s close relationship with militancy. In addition, by
enabling it to continue to prosecute conflict with India, a militant strat-
egy has provided the Pakistan Army with a bureaucratic and ideological
raison d’être and ensured its status as the state’s most important institu-
tion. Pakistan’s geo-strategic endowments, meanwhile, insulate Pakistani
leaders from the need to rethink this approach. Consequently, they face
little pressure to devise new policies that de-emphasize competition with
India, provide their population with basic public goods, and enable their
country to move beyond militancy.
Finally, broad historical overviews, even if they do not focus specifi-
cally on militancy, put the problem in a larger chronological and politi-
cal context. In doing so, they can help to clarify militancy’s relationship
to important characteristics of the Pakistani state and turning points in
its trajectory. They also make important basic points about the militancy
problem, such as its links with the Pakistan Army, its inadvertent exacerba-
tion by outside benefactors such as the United States, and its close connec-
tion to the policies of General Zia. These points can significantly improve
the general reader’s understanding of the Pakistan–militant nexus.
6
Prior scholarship thus has made important contributions to our under-

standing of militancy in Pakistan. The existing approaches nonetheless
fall short in a number of ways. For example, the descriptive literature pro-
vides an important source of empirical evidence on militancy in Pakistan,
but it offers little accompanying analysis. Thus, it leaves a host of questions
regarding Pakistan’s connection with Islamist militancy unanswered.
How should we interpret the information about the Pakistan–militant
nexus that the descriptive literature provides? What was the underly-
ing logic of Pakistan’s use of militancy? How effectively did the strategy
advance Pakistani goals? What generalizable propositions about states’
strategic use of militancy can we derive from the Pakistan case? Because
they seek primarily to provide rich factual description, empirical works
generally do not attempt to answer such questions.
The sociopolitical camp of the literature often fails to capture the deliber-
ateness with which Pakistani leaders created and exploited the connections
between Islamist militants and Pakistan’s military and security services.
These works tend to imply that the Pakistan–militant connection was the
unfortunate result of inexorable social and political forces, rather than the
product of a carefully formulated and executed Pakistani security policy.25
In truth, Pakistani political and military leaders have not been idle specta-
tors to the growth of their country’s relationship with Islamist militancy;
they have actively nurtured and worked with the militants, in a variety of
operational contexts, since the founding of the Pakistani state. Indeed, the
sociopolitical forces to which this literature refers were often unleashed
by the Pakistani leadership’s efforts. A study of Pakistan’s problem with
Islamist militancy must convey the deliberate nature of this relationship and
the logic underlying it, as well as evaluate its impact on Pakistani security.
The politico-military and organizational strategy camp of the litera-
ture leaves a number of questions unanswered. Work in this camp char-
acterizing Pakistan’s use of militancy primarily as a tactic to support
the Kashmir insurgency or promote the interests of particular political
leaders cannot explain the longevity, geographical breadth, and political
diversity that characterize the Pakistan–militant nexus. Why is a tactic
designed to support the ongoing Kashmir insurgency actually decades
older than that uprising? Why is Pakistan closely involved with militants
not just in Kashmir, but also in other states such as Afghanistan? And if
the promotion of militancy was a tactic that particular Pakistani leaders
used to advance their own domestic political interests, why has Pakistan’s
use of militancy been consistent across a wide range of military and civil-
ian officials? This strand of the literature does not provide satisfactory
answers to these important questions. It also focuses primarily on the
7
effects that Pakistani state support has had on the interests and capabili-
ties of militant groups. This book, by contrast, is concerned mainly with
the impact that supporting militancy has had on the strategic interests of
the Pakistani state.
Work in the politico-military and organizational camp that focuses on
the bureaucratic interests and commitments of the Pakistan Army leaves
open the question of their source. From where do these interests and com-
mitments come? In fact, the army’s attachment to militancy is rooted in
a source even deeper than its own bureaucratic proclivities—namely, the
founding logic of the Pakistani state, which provides the ideological basis
for the military’s bureaucratic commitments.
Work in the politico-military and organizational camp stressing the
perverse impact of Pakistan’s strategic assets may overplay structure and
underplay the importance of preferences in driving Pakistani security
behavior. Why do other countries with resource endowments similar to
Pakistan’s not engage in similarly pathological security behaviors? The
difference in behavior results primarily from divergent leadership prefer-
ences, quite apart from structural similarities.
Finally, broad studies of Pakistani political history, which address
a diverse spectrum of issues in addition to the Pakistan–militant con-
nection, often provide overly simplified discussions of Pakistan’s use of
Islamist militancy. As a result, they can be misleading. For example, these
studies generally trace the roots of Pakistan’s militancy problem to the
Islamization of Pakistani society under General Zia. This Islamization
process, however, began earlier, under Prime Minister Zulfikar Ali
Bhutto. Bhutto attempted to deflect public anger over the failure of his
economic programs through pious public gestures, such as declaring the
minority Ahmadi sect to be non-Muslim and banning the consumption
of alcohol.26 These were the first steps in the Islamization policy that Zia
later adopted. Moreover, a number of other factors underlay Pakistan’s
use of Islamist militants, such as the lack of a coherent national found-
ing narrative and material weakness relative to India. Finally, Pakistan
did not adopt its militant strategy during the Zia era; the Pakistanis had
been using Islamist militants as strategic tools since achieving indepen-
dence, long before Zia’s emergence.27 Other works in the historical camp
avoid these shortcomings but pay little specific attention to the subject of
militancy.
Despite its many strengths, then, the current literature on Pakistan’s
connection with Islamist militancy falls short in a number of areas.
Although I draw upon it, this book differs from existing scholarship in
important ways.
8
The Argument
Pakistan’s support for Islamist militancy has not been a mere accident or
a short-term military or political tactic. It is a deliberate, long-r unning
policy as old as the Pakistani state. Indeed, supporting jihad has con-
stituted nothing less than a central pillar of Pakistani grand strategy. 28
Grand strategy is a state’s theory of how to produce national security. It
identifies the goals that the state should seek in the world and also speci-
fies the military instruments that it should use to achieve them.29
Pakistan has three main grand strategic tools: nuclear weapons, con-
ventional forces, and militant proxies. Nuclear weapons have played a
defensive role for Pakistan, deterring large-scale Indian attacks against the
Pakistani homeland. During the 1999 Kargil war and a 2001–02 milita-
rized standoff, for example, nuclear danger led the Indians to rule out any
offensive that could have threatened Pakistan with catastrophic defeat.
Nuclear weapons have thus helped to guarantee Pakistan’s survival even
in the face of confrontation with a militarily stronger India. They have not
by themselves enabled Pakistan offensively to alter the territorial or politi-
cal status quo in South Asia, however. 30
Pakistani conventional forces have served a combination of pur-
poses. In some cases, such as the 1947 and 1965 wars, they have joined
conflicts against India that militants had already launched. In one
instance, the 1971 Bangladesh conflict, they began and fought a war
essentially on their own. In other cases, such as the Kashmir insurgency
and the Afghan conflicts, they have avoided direct involvement, leav-
ing the fighting up to the militants. Although conventional military
forces have occasionally engaged in offensive action against India, since
the Bangladesh war their main purpose has been to provide Pakistan
with a robust defense against any Indian conventional attack. Prior to
Bangladesh, Pakistani leaders believed that their forces were inherently
superior to the Indians and would inevitably defeat them on the battle-
field, much as the subcontinent’s Muslim invaders had done to its Hindu
inhabitants centuries earlier. At the very least, a small, Muslim Pakistan
would be able to fight a larger Hindu India to a draw, as it had in 1947
and 1965. The 1971 war, which saw India vivisect its adversary and cre-
ate Bangladesh out of East Pakistan, disabused the Pakistanis of this
notion. They realized that, in the future, a direct conventional military
confrontation with India could have catastrophic consequences. Since
then, the Pakistanis have avoided such fights and used their conven-
tional forces in a primarily defensive role. 31
Militant forces, by contrast, have served as Pakistan’s primary offen-
sive tool. They have started conflicts in which conventional forces have
9
subsequently participated, such as the 1947 and 1965 wars. In addition,

Pakistan has used them to wage the Kashmir insurgency and shape the
Afghan security environment largely unassisted. In doing so, the militants
have enabled Pakistan to pursue its most cherished security goals: attrit-
ing Indian military resources, ejecting India from the disputed territory
of Kashmir, and gaining strategic depth through the installation of a
friendly regime in Afghanistan.
Pakistan is certainly not the only country to have used nonstate actors
as proxies. Many states have supported terrorists or militants to promote
their strategic agendas. Iran, for example, has long employed the terror
group Hizbollah to advance its interests in Lebanon. The Soviet Union
supported the operations of leftist organizations such as the Red Brigades
and Black September against Israel and Western Europe during the 1970s.
And the United States supplied the Afghan mujahideen with extensive
financial and military support in their battle against the Soviet Union
during the 1980s.32 The Pakistan case is unique, however, in the impor-
tance of the role that supporting militancy plays in its national security
policy. Support for militants has not simply been one among many tools
of Pakistani statecraft. Rather, the use of Islamist militants has been a pri-
mary component of Pakistani grand strategy.
How effective has Pakistan’s use of militants been? Has it helped
Pakistan to achieve its strategic goals? This question has not received
in-depth examination, 33 but most commentary on the issue is extremely
negative. Scholars and commentators characterize Pakistani policy as the
product of chronic misjudgment and careless decision making—“strate-
gic myopia,” in the words of one analysis. 34 As another puts it, Pakistani
leaders have “lacked imagination and vision” and been guilty of “a near-
total reliance on tactical opportunism.” As a result, Pakistan’s national
security policies, including its strategic use of Islamist militants, “[have]
not made Pakistan more secure.”35
To be fair, there is considerable truth in this negative characterization
of Pakistan’s militant strategy. Perhaps the strongest evidence of the inef-
fectiveness of Pakistani policy can be found in the current map of the
Indian subcontinent. Over six decades after independence and Pakistan’s
initial use of Islamist militancy in Kashmir, territorial boundaries in the
region remain largely unchanged, with India continuing to govern Jammu
and Kashmir State. Thus, Pakistan’s militant policy has failed to achieve
its most important goal. Meanwhile, support for militancy has taken a sig-
nificant human and financial toll on Pakistan, as well as contributed to
ongoing Indo-Pakistani tension, which could potentially lead to another
regional war. Thus, at first glance, the costs of Pakistan’s strategic use of
Islamist militancy seem to far outweigh its benefits.
10
Nonetheless, the reality of Pakistan’s militant policy is considerably

more complex than many scholars suggest. It has not been wholly disas-
trous. Rather, it has achieved a number of important domestic and inter-
national successes. The strategy has helped to promote internal political
cohesion, giving Pakistan a raison d’être in the absence of a coherent
founding narrative. By steadily attriting Indian military and financial
resources, it has also played a role in redressing Pakistani material weak-
ness vis-à-v is India. In addition, the strategy has enabled Pakistan to
continue to undermine Indian control of Kashmir and ensure that the
disputed territory remains a subject of international attention. Finally,
Pakistan’s militant strategy has enabled it to shape the strategic environ-
ment in Afghanistan, helping to install a friendly government on its criti-
cal western border.
Recently, however, these successes of Pakistan’s militant strategy have
given way to serious problems. The militant organizations that Pakistan
nurtured over the decades are increasingly exceeding its control. They
have begun behaving in ways that seriously undermine Pakistani inter-
ests, such as attacking Pakistani political leaders and security personnel,
adopting political agendas that far exceed those of their Pakistani spon-
sors, and launching operations that threaten to drag Pakistan into cata-
strophic conflicts. In addition, Pakistan’s militant strategy has diverted
scarce resources from pressing domestic projects, impeding the country’s
internal development. Finally, Pakistan’s militant campaign has led India
to undertake military improvements, such as a significant conventional
arms build-up and the formulation of an aggressive conventional military
doctrine. These developments seriously undermine Pakistani interests,
threatening to leave it less politically cohesive and externally secure than
it was before.
Pakistan thus suffers from a “jihad paradox.” Political and mate-
rial weakness originally made Pakistan’s militant policy attractive and
useful. Now, however, that same weakness makes Pakistan’s support
for militancy extremely dangerous. Thus, despite its past benefits, the
strategy has outlived its utility, and Pakistan will have to abandon it
to avoid catastrophe. Other weak states, which may also be tempted to
use nonstate actors as strategic tools, should take the Pakistani case as
a cautionary lesson. The political and material shortcomings that could
make such a strategy appealing also increase the likelihood that they
will lose control of their proxies, face painful developmental trade-offs,
and provoke stronger adversaries to adopt policies that further threaten
their security.
11
The Plan of the Book
The book proceeds as follows: Chapter 2 discusses the general logic of

using nonstate actors as strategic tools. I show that the use of nonstate
proxies enables relatively weak sponsor states to challenge stronger adver-
saries and shape the international strategic environment in ways that
would be too expensive and risky to attempt with conventional military
forces. It does so by lowering battlefield costs for the relatively weak state,
which does not have to commit its own soldiers to the fight; creating oper-
ational advantages by impeding potential target-state military responses
to weak-state provocations; and creating bargaining advantages for a weak
sponsor state, enabling the sponsor to demand a higher price for ending its
militant campaign than it would be able to extract if it were fighting alone.
Chapter 2’s next section explains that although a militant strategy offers
weak sponsors a number of potential advantages, it also involves serious
downside risks. Militant proxies can prove difficult for a weak sponsor to
control. This can enable the militants to provoke the sponsor’s adversaries
without authorization, triggering unwanted conflict; lead them to work
directly against sponsors; and create a competitive security environment,
forcing overinvestment in defense and hurting development. These prob-
lems with proxy strategies are likely to be particularly severe for weak
states and can ultimately outweigh such strategies’ advantages.
Chapter 2’s final section illustrates militant proxy logic by briefly consid-
ering historical evidence from South Asia stretching from 1947 to the present
day. It shows that, over the decades, Pakistan has consistently exploited the
cost, operational, and bargaining advantages inherent in its militant strategy.
With the passage of time, however, Pakistan’s strategy also has given rise to
damaging control problems and developmental opportunity costs.
Chapter 3 traces the origins of Pakistan’s militant strategy, showing
that it emerged in the wake of the partition of British India, out of the
new Pakistani state’s acute political and material weakness. To ameliorate
these problems, Pakistani leaders sought to seize the disputed territory of
Kashmir from India, without facing India in a direct military confronta-
tion. Pakistani leaders settled on a strategy using local militants to battle the
Maharaja of Kashmir—and Indian forces sent to rescue him—for control
of the territory. Although the Pakistanis’ militant strategy did not enable
them to capture Kashmir in 1947–48, Pakistani leaders did not view it as a
complete failure and believed that it could be successful in the future. The
strategy thus became a central component of Pakistani security policy, its
sophistication and importance increasing with each subsequent conflict.
12
Chapter 4 examines the evolution of Pakistan’s militant strategy from

the aftermath of the first Kashmir war through the 1965 Kashmir war,
the Bangladesh war, and the anti–Soviet-A fghan conflict of the 1980s.
It shows that Pakistan’s use of Islamist militants became more extensive
with each of these conflicts, progressing from the use of rag-tag local
militias following independence to the management of a complex inter-
national effort to support anti-Soviet mujahideen in Afghanistan during
the 1980s. The Pakistanis were eventually able to employ the resources
and expertise that they acquired over these years to support yet another
round of jihad in Kashmir, which has lasted from the late 1980s to the
present day.
Chapter 5 discusses Pakistan’s current use of its militant strategy in
Kashmir and Afghanistan. It shows how the strategy helped Pakistan to trig-
ger the Kashmir insurgency and to influence its subsequent character and
trajectory. It also shows how, in Afghanistan, Pakistan’s strategy made pos-
sible the rise of the Taliban, affording the Pakistanis a friendly regime and
increased strategic depth on their western border.
Chapter 6 evaluates the impact of Pakistan’s militant strategy on Pakistani
security interests. It shows that although the strategy is generally derided as
an unmitigated disaster, it has actually achieved important domestic and
international successes that have allowed it to strengthen its political foun-
dations and to continually challenge India and the South Asian status quo.
Recently, however, the militant strategy has given rise to control and devel-
opment problems that threaten to make Pakistan even less secure that it was
originally. What was once a useful strategy has thus become extremely dan-
gerous, threatening the very survival of the Pakistani state.
Chapter 7 discusses possible solutions to the problems associated with
Pakistan’s militant strategy. The chapter argues that Pakistan will truly have
to abandon militancy, ending support for all jihadist groups and taking con-
crete steps to crush those operating on its territory. This will be extremely
difficult because of Pakistan’s founding narrative, which necessarily makes
it an oppositional state, dedicated to combating India and revising territorial
boundaries in South Asia. Therefore, Pakistan will be able to renounce mili-
tancy only if it fundamentally transforms its national purpose and identity,
becoming a state that no longer defines itself in terms of opposition to India
and the current territorial status quo. In the absence of such a transforma-
tion, there is little reason to expect that Pakistan will ever renounce its strat-
egy of jihad.
13
CHAP T ER 2
The Logic of a Militant

Proxy Strategy
P akistan’s use of militancy is widely dismissed as being a foolish and

even irrational policy. Why, then, might a militant proxy strategy be
attractive to a state like Pakistan? What are its potential costs and ben-
efits? Is there a logic to the use of nonstate proxy forces?1 If so, Pakistan’s
militant policy might be more reasonable than many have suggested.2
Scholars and analysts have extensively examined nonstate actors’ own
motivations for engaging in violent behavior.3 They have also explored the
impact of relationships with sponsor states on nonstate actor interests.
They have focused less attention on the effects that nonstate proxy strate-
gies have on the national interests of state sponsors, however.4 The reason
is that international relations (IR) scholarship has been concerned pri-
marily with conflict between nation-states. For example, major IR theo-
ries such as realism and liberalism focus primarily on interstate relations.
Formal methodologies such as game theory typically model strategic
interactions between states. Common data sets have generally recorded
state-to-state disputes. 5 And scholarly discussions of proxy warfare have
often defined the phenomenon in state-centric terms, ignoring the pos-
sibility that nonstate entities could serve as proxies for sponsor countries.6
In the next section, I explore the effects that a militant proxy strategy
may have on the interests of state sponsors. I do so by examining the ways
in which such a strategy might operate in the case of a relatively weak
state using militants to challenge a stronger adversary.7 I do not claim that
every state in this position will experience all of the effects that I discuss.
Rather, I seek to identify potential advantages and disadvantages that
sponsor states may experience if they pursue a militant proxy strategy.
14
This can help to explain more clearly the Pakistani case, as well as illumi-
nate the incentives that other states in a similar position may face in the
future. As I explain, a proxy strategy can offer a sponsor attractive cost,
military, and bargaining benefits. It can also give rise to control and devel-
opmental problems, however, which over the long term make it extremely
costly and dangerous.
THE BENEFITS OF A MILITANT PROX Y STRATEGY
The first benefit of militant proxy strategies is straightforward: They are

relatively inexpensive. This is the case for two reasons. First, fielding
militant proxies is cheaper than fielding a conventional military force.
Militants will require some investment in recruitment and training, as
well as ongoing material and financial assistance. These requirements,
however, are likely to be considerably less extensive, sophisticated, and
costly than those for conventional military forces.8 For example, mili-
tants are unlikely to need expensive equipment like armor, ships, and
aircraft. Nor will they need the same level of personal support, such as sal-
ary, healthcare, and social benefits, as regular service members. Second,
by sending proxies into battle rather than a conventional army, the state
avoids risking its own personnel in a confrontation with its stronger
adversary. This increases the likelihood that the sponsor state’s military
will remain intact and undamaged, even if the operations that it supports
are defeated.
The relative cheapness of a militant proxy strategy could make it
appealing to a wide range of states; all else being equal, states are likely
to prefer policies that enable them to pursue their security interests at the
lowest possible cost. The cheapness of such a strategy would be especially
attractive, however, to a relatively weak state lacking in military capabili-
ties in comparison with its adversaries. By enabling it to field forces at a
low initial cost, and without risk of its own military subsequently suffer-
ing damage or defeat, the strategy could enable a weak state to engage
stronger opponents than it otherwise would be able to face.9
The ability to field forces inexpensively is important, but it is not the
only benefit that a state can realize by employing militant proxies. The
second main advantage of such a strategy is its ability to increase the diffi-
culty of the military responses open to an adversary. Although this benefit
could have broad appeal, it would again be especially attractive to weak
states, potentially allowing them to engage powerful opponents that they
would otherwise be unable to fight.10
15
Lo gi c o f a M ili ta n t P r o x y S t r at e g y ( 15 )
There are two main avenues of military response open to a strong

state facing a challenge from a weaker adversary: denial and punish-
ment. A denial-based response would seek to prevent the weak state from
achieving its aims by defeating its military operations. A punishment-
based response, by contrast, would not seek to defeat the weak state’s mili-
tary efforts. Rather, this response would impose costs on the weak state by
damaging targets that it values.11
Both of these strong-state options hold significant risks for a weaker
challenger.12 A weak state facing a denial campaign would probably lose
any battlefield confrontation with its stronger enemy. At the very least,
such a loss would result in the failure of the weak state’s military efforts.
The confrontation could also escalate, however, leading to a broader con-
flict extending beyond the confines of the weak state’s initial military
operation, perhaps even reaching its homeland. Material costs could
include lost personnel and equipment, civilian casualties, and damage to
or loss of home territory. Political costs could include significant harm
to the weak state’s leaders, whom domestic constituencies could vote out
or overthrow as punishment for perceived incompetence.13 If the strong
state seems likely to be able to wage a highly effective denial campaign, a
weak adversary contemplating an attack could decide against launching
it. In this case, the strong state would have achieved deterrence by denial.
A denial campaign, then, could protect a strong state against weak-state
attack either by thwarting the operation in the field or by deterring the
weak state from even trying to launch it.14
If a strong state decided to respond to the weak challenger’s provoca-
tions with a punishment campaign, the weak state would probably be
hard-pressed to stop it. A punishment campaign would result in poten-
tially extensive harm to the weak state’s homeland or other territories or
populations that its leadership values. Material costs could include dam-
age to infrastructure, natural resources, and cultural or historic sites, as
well as civilian casualties. As with denial, costs could also include signifi-
cant political harm to weak-state leaders.15 If a strong state seemed likely
to be able to wage a highly damaging punishment campaign, it could
achieve deterrence by punishment, convincing a weak state not to attack
because the costs of strong-state retaliation would outweigh the potential
benefits of weak-state battlefield success.
Both denial and punishment constitute powerful tools for a strong
state facing the possibility of weak-state attack. A weak state contemplat-
ing aggression against a stronger adversary will therefore need to devise a
means of undermining the strong state’s denial and punishment capabili-
ties. A militant proxy strategy can help a weak state to do exactly that.
16
On the denial side, weak-state use of militant proxies can make strong-
state defense considerably harder. A strong state could defeat an opera-
tion by the weak state’s proxy forces and achieve denial in three basic
ways. First, assuming that the weak state’s proxy forces were outside of
the strong state, the strong state could prevent the proxies from enter-
ing its territory. Second, if the weak state’s proxies were already inside
the strong state, the strong state could find the militants before they
launched their operation. Third, if the weak state’s proxies were already
in the strong state and the strong state was unable to find them before
they launched their operation, it could defeat the proxies’ attack once it
was already underway.
Unfortunately for strong states, each one of these denial options faces
formidable challenges when pitted against a militant proxy strategy. First,
preventing small groups of incognito militants from crossing interna-
tional boundaries and entering a strong state is an enormous task. The
only way for strong states wholly to stop them from entering their ter-
ritory is through flawless patrolling of extensive borders and coastlines,
and policing of multiple ports of entry. This is a virtually impossible chal-
lenge, even for the most capable of states. Indeed, large, powerful coun-
tries, which are likely to have relatively long borders and coastlines and
numerous ports of entry, may find this problem to be especially difficult.16
If militant forces do manage to evade detection at borders or ports and
infiltrate the strong state’s territory, preventing them from launching their
attack will be extremely difficult. Thwarting the militants will require the
strong state to locate and apprehend them before they can strike. The mili-
tants’ ability to operate in small groups that can blend into the local pop-
ulation, however, will make such detection and apprehension difficult,
particularly in the high-density urban centers in which they are likely to
launch their attacks.17 As a result, the strong state may be unable to find
and arrest the militants before they strike, even if it has some foreknowl-
edge of their plans and is actively searching for them. In this case, the
strong state’s last denial option would be to defeat the militants’ operation
by thwarting an attack once it was already underway.
Defeating an attack in progress is also likely to prove to be problem-
atic for the strong state. The strong state will probably have limited, if any,
warning of the attack, while the militants will have surprise and initia-
tive on their side. In addition, the attackers may not seek to escape, killing
themselves or fighting to the death instead. This can enable them to inflict
significant damage on their targets, thus making their operation costly to
the strong state even if they are quickly eliminated.18 The use of militant
proxy forces thus makes all three possible strong-state methods of denial
extremely difficult.
17
It is important to note that militant proxies do not undermine strong-

state denial efforts solely at the operational level. They can also make
denial politically difficult. For example, if the weak-state militant cam-
paign takes the form of a long-r unning insurgency, the defender will need
to hunt down fighters in its territory and thwart their operations. This can
be challenging if the insurgents are, or appear to be, native to the disputed
area. Such apparent indigeneity can give their efforts an air of legitimacy,
which would be absent if the weak state simply attacked the strong state
with its own, “outside” conventional military forces. This sense of legiti-
macy can make strong-state defensive measures seem undemocratic and
oppressive. This, in turn, can undermine support for strong-state counter-
insurgency efforts in the disputed area, among the strong state’s broader
population, and in the international community.19
We also should note that even if a weak sponsor initially fails in its
attempt to attack a stronger adversary through the use of militant proxies,
the weak state can keep trying. Given the limited number of personnel
and rudimentary materiel necessary to launch an attack, the cost of such
attempts is low; if one mission fails, another can be undertaken. Strong-
state denial capabilities will have to be perfect to prevent the weak state
from eventually succeeding. Over an extended period of time, such per-
fection is highly unlikely.20
This is not to argue that the aforementioned tasks will be easy for mili-
tant proxy forces to accomplish. The barriers to infiltration of a target
state, avoidance of detection once there, and successful execution of an
attack are formidable. Nonetheless, a weak state will be far more likely to
be able to accomplish these tasks with militants than with conventional
military forces. Conventional forces would have little chance even of pen-
etrating the strong state’s border, quite apart from moving to locations
deep within the strong-state homeland and launching successful attacks
against high-value targets there. Militant proxy forces stand a much better
chance of succeeding in these efforts, especially if they are able to attempt
them repeatedly.
In addition to undermining strong-state denial operations, a militant
proxy strategy can hinder strong-state efforts to punish a weaker chal-
lenger. The strategy can do so by creating uncertainty as to who is attack-
ing the strong state. Proxy-force sponsors generally do not announce their
relationship with militants and, in fact, often deny it.21 The militants,
for their part, do not wear uniforms and are unlikely to have any pub-
licly acknowledged connection to their weak-state patron. Consequently,
a strong state facing a proxy strategy may be unable to determine with
certainty the identity of its adversary.22 Even if the strong state is able
to identify the militant organizations or individual fighters carrying out
18
the attacks, it may still be unsure as to who is facilitating their activity by

means of financial, military, or other support.23
Knowledge of attacker or sponsor identity would be essential if the
strong state wished to respond to weak-state attacks with a punishment
campaign. For punishment to succeed, the strong state would have to
damage targets that the militants or their weak-state sponsor valued.24
Without knowing their identities, the strong state would be unable
to do so.
In addition to these operational issues, a strong-state punishment
campaign would require knowledge of an attacker’s identity for politi-
cal reasons. The strong state would have to justify to both domestic
and international audiences its infliction of harm on weak-state targets.
This could be a difficult task, since the strong state’s actions would not
be taken strictly in self-defense and would strike targets that were not
directly responsible for the attacks that the strong state suffered. To suc-
ceed, the strong state would need to make a convincing case that the weak
state facilitated the attacks against it. If it failed to do so, the strong state’s
actions could appear to be unwarranted. This could increase opposition
to the strong state in any disputed territories, undermine public support
for a punishment campaign within the strong state itself, and create inter-
national opposition to the strong state’s policy.
A strong-state punishment campaign thus requires clear knowledge of
the identity of attacking parties and their supporters. Without such knowl-
edge, the strong state may find itself unable to identify appropriate targets
or to justify its behavior to domestic and international audiences. A weak-
state militant proxy strategy, by obscuring this information, undermines
a key defensive option available to stronger adversaries.
In addition to the cost savings and military benefits outlined pre-
viously, a militant proxy strategy can offer a weak sponsor state a third
advantage. The use of militant proxies can enhance the sponsor’s bargain-
ing leverage during negotiations to end a conflict. This can potentially
enable the sponsor to secure a more favorable settlement with a strong
opponent than it could get if it were fighting alone.
Wars are settled when the antagonists strike a bargain resulting in
payoffs that each prefers to continued conflict. 25 A sponsor state uses
a militant proxy strategy to force its adversary to the negotiating table,
where the sponsor seeks payment in return for ending its militant cam-
paign. Proxies can be useful to the sponsor state in this bargaining pro-
cess because they can enable it to demand an especially high payment
from its adversary. The sponsor can demand such a payoff premium
because of the control problem associated with its use of third-party
19
proxies. Specifically, the state can claim that reining in its proxies will
be difficult and costly, requiring an especially large compensatory
payment.
This control issue is rooted in what is known as a principal–agent prob-
lem. A principal–agent problem arises where one party employs another
to carry out an assigned task. The hiring party, or principal, employs the
agent because the agent is better able to do the work in question, for rea-
sons such as cost, legitimacy, deniability, or expertise. After having been
hired, however, the agent may prove unwilling or unable to do the princi-
pal’s bidding. This failure may occur because, from the outset, the princi-
pal and agent’s interests were misaligned, or because the agent lacked the
capacity to carry out the assigned task. Alternatively, this failure could
result from the principal’s inability to devise mechanisms to ensure that
the agent is behaving properly while under its employ.26 Regardless of the
failure’s precise source, principal–agent relationships can be extremely
costly for the sponsor. Misbehaving agents can consume the principal’s
resources without delivering promised results, or even drag the principal
into an unwanted war.
Although scholars typically focus on principal–agent problems’ costs
to sponsor states, a sponsor’s lack of control over its proxies can actually
help it to negotiate a favorable settlement to a conflict. Limited control
increases the difficulty of the sponsor’s task of reining in proxy forces.
The sponsor can therefore demand higher payment from its adversary in
return for calling off the proxies than it could if it had firm control over
them.27 Even if the sponsor is able to exercise relatively good control over
its proxies, it can still probably demand a higher settlement price than it
could if it were fighting alone, in which case the task of reining in a third
party would not exist.28 In addition to helping a weak state initiate and
prosecute a conflict with a stronger adversary, then, the use of militant
proxies can help a weak state to end a conflict with a powerful opponent
on relatively favorable terms.
The use of militant proxies thus can hold significant cost, military,
and bargaining benefits for a weak state seeking to challenge a stron-
ger adversary. Such a strategy does not, however, offer a sponsor state
unmitigated advantages; the use of militant proxies can also subject a
sponsor to important costs. These include principal–agent problems,
costly trade-offs between security and development, and antagonism of
stronger adversaries. Next, I discuss each of these costs in turn. All of
them are heightened by the characteristic that is most likely to make
a militant strategy attractive to a state sponsor—t he weakness of the
sponsor state.
20
THE COSTS OF A MILITANT PROX Y STRATEGY
The benefits of a militant proxy strategy are likely to be particularly

attractive to weak states, which are unable to challenge stronger adver-
saries directly. Such weakness, however, can lead to a number of difficul-
ties for sponsor states. First, sponsor-state weakness can make proxies
difficult to control. This is the downside of the principal–agent relation-
ship that I discussed earlier. A degree of difficulty in controlling militant
proxy forces can create bargaining advantages for a sponsor. A sponsor
that experiences too much difficulty in controlling its proxies, however,
risks having them genuinely exceed its grasp. Adversaries may be unwill-
ing to negotiate with such a sponsor, as it may prove unable to call off
its proxies even if it wishes to do so.29 In addition, militant proxies may
take actions that directly harm a sponsor that is unable to control them.
For example, militants could adopt agendas more ambitious than those
of their sponsors. This could lead them to threaten, attack, or otherwise
challenge the sponsors’ stronger adversaries without approval.30 Such
aggressive behavior could provoke adversaries to undertake retaliatory
measures that threaten to damage the sponsor state and force it to divert
scarce resources away from domestic priorities to the defense sector. 31 If
a sponsor attempts to solve control problems by re-establishing authority
over errant proxies, the proxies could turn against it, attacking domestic
targets and even contesting control over portions of the sponsor state’s
territory.
Ironically, the more successful the sponsor state’s militant campaign
has been, the more likely these problems are to emerge. A successful cam-
paign is likely to be characterized by an able militant force and a signifi-
cantly damaged adversary. These characteristics increase the likelihood
that the militants will be able to exceed the sponsor’s control and that
the adversary will be highly motivated to inflict retaliatory harm on the
sponsor.
Maintaining control over militant proxies does not ensure that the
sponsor state will be able entirely to avoid these types of problems. Any
successful militant campaign, regardless of control issues, will antagonize
the sponsor state’s adversary, potentially leading it to respond militarily.
This could present the sponsor with challenges ranging from arms races
and painful resource trade-offs to outright conflict with the adversary.
Once again, the more successful the sponsor state’s campaign is, the
more motivated the adversary will be to respond in some manner, and
thus the more likely these problems are to emerge. The most dangerous
militant proxy campaign to a sponsor state, then, may also be the most
successful one.
21
The logic of a militant proxy strategy thus involves a complex mix of

potential benefits and challenges for a sponsor state. The strategy offers
a sponsor the possibility of significant cost, operational, and bargaining
benefits but also threatens to create serious control, resource, and exter-
nal security problems—problems that are exacerbated by the state weak-
ness that is likely to make a militant strategy attractive to a sponsor in the
first place. Next, I briefly discuss these benefits and problems of a militant
proxy strategy in the context of the Pakistani case, identifying some spe-
cific ways in which they have helped Pakistan to pursue its national secu-
rity goals, as well ways in which they have undermined Pakistani strategic
interests.
THE BENEFITS AND COSTS OF PAKISTAN’S

MILITANT STRATEGY
Pakistan’s use of Islamist militants has consistently exploited the cost,

operational, and bargaining advantages outlined previously. Let us first
examine the cost benefits of Pakistan’s militant strategy.
Cost: Employing a militant proxy strategy is cheaper for Pakistan than
fielding regular forces. This advantage is difficult to quantify. Pakistan
does not publicize the price of supporting the militant groups, and we do
not know how much it would cost Pakistan to launch conventional mili-
tary operations designed to achieve the same goals as their militant cam-
paign. Nonetheless, available evidence strongly suggests that a militant
proxy strategy is the cheaper of the two options.
Consider the cost to Pakistan of supporting the group Lashkar-e-Taiba
(LeT). LeT is the Pakistanis’ most dependable ally in Kashmir.32 The
group has an estimated annual budget of $50 million, of which $5 mil-
lion is reportedly dedicated to military operations. 33 LeT’s specific mil-
itary budget does not include a number of expenses associated with its
broad militant campaign, such as weapons and ammunition. And its
overall budget includes many costs not directly related to militancy, such
as preaching and education. 34 The amount of money needed to support
Lashkar-e-Taiba’s militant activities, then, is probably more than $5 mil-
lion and less than $50 million. Pakistan does not need to provide all of
this funding, as the group receives monies from a wide variety of sources,
including other foreign governments, an international network of charita-
ble organizations, legitimate business operations, and a range of criminal
activity. 35
Let us assume, however, that LeT’s militant activities cost the full
$50 million and that Pakistan is the group’s sole funder. In addition,
22
since Pakistan supports multiple militant organizations, let us assume

that Pakistan underwrites five Lashkar-e-Taiba equivalents annually. If
this is the case, the cost to Pakistan of fielding militant proxy forces is
approximately $250 million per year. 36 This is a tiny fraction—less than
4 percent—of Pakistan’s $6.3 billion defense budget. Yet the militants
surely account for far more than 4 percent of the value of Pakistan’s overall
strategic effort. The militants, it would appear, do offer Pakistan a strate-
gic bargain in comparison with conventional military forces.
Pakistan’s proxy strategy is relatively cheap not just in financial terms,
but in human and territorial terms as well; it has enabled Pakistan to
avoid risking its own forces and homeland in battle. Regular Pakistani
forces’ level of participation in operations against India has varied across
conflicts. In some cases, such as the first two Kashmir wars, militants
launched the conflict’s opening campaign and subsequently were joined
by the regular Pakistani military, which then assumed the primary com-
bat role. In other instances, such as the current Kashmir insurgency and
a series of conflicts in Afghanistan, the militants have worked primarily
on their own, without the regular Pakistan military’s active participation
in combat operations. 37 In all cases, the Pakistanis have been able to avoid
paying the full battlefield price that these operations would have entailed
if they had been executed only using regular Pakistani forces. As history
has shown, that price can be extremely high. During the 1971 Bangladesh
war, the only conflict in which Pakistan primarily utilized conventional
military forces, India vivisected Pakistan, cutting the country’s eastern
from its western wing, taking ninety-five thousand prisoners, and creat-
ing the state of Bangladesh. Not surprisingly, the Bangladesh conflict was
the last time that Pakistan engaged India in a full-fledged conventional
war. Since then, the Pakistanis have generally fought their battles with
India through the less risky strategy of using Islamist militants in place of
their own soldiers.
Military Responses: Pakistan’s use of militant proxies has made its
operations against targets in Kashmir and India proper extremely dif-
ficult to defeat. India has employed a host of denial measures, including
the stationing of several hundred thousand security forces in Kashmir,
many of which specialize in counterinsurgency warfare; the mainte-
nance of physical barriers along the Line of Control; and the deploy-
ment of domestic security forces ranging from local police forces to
federal antihostage units. 38 None of these measures have succeeded in
stopping militant forces from infiltrating Indian territory and launching
violent attacks, which have included bombings, assassinations, kidnap-
pings, and spectacular operations such as the November 2008 assault
on Mumbai.
23
Indian denial efforts also have had difficulty defeating militant opera-
tions once the operations have already begun. For example, during the
2008 assault on Mumbai, ten Lashkar-e-Taiba operatives fought Indian
police, paramilitary, and military hostage–rescue forces to a standstill
while killing over 160 people and wreaking havoc across the city. The
Indians did not succeed in defeating the militants until three days after
the attacks began.39
India’s inability to defeat the militant campaign has resulted not only
from operational difficulties but also from a political problem—the mili-
tant forces’ appearance of legitimacy. Militant proxies give the struggle
against Indian control of Kashmir the aura of a genuine independence
movement, which it would lack if the Pakistanis simply attempted to take
the territory with their own military forces. This has undermined Indian
defensive efforts, which have often appeared heavy-handed. This heavy-
handedness, in turn, has subjected the Indians to intense international
criticism and further stoked discontent in Kashmir.40
On the punishment side, over the course of four wars and decades of
hostility, India has never attempted to inflict significant harm on non-
combat-related Pakistani targets. Indian forbearance has had numerous
sources. Some of them are rooted in domestic politics. Partition and the
subsequent decades of Indo-Pakistani tension cannot change the fact that
India and Pakistan originally were a single country, which shared com-
mon geography, history, cultural, and colonial experiences. Moreover,
the two states continue to share important cultural ties. For example,
ethnic Punjabi culture, or “Punjabiyat,” spans the Indo-Pakistani divide
and creates links between populations on both sides of the border.41 As a
result, Indians and Pakistanis in this region may have more in common
with each other than they do with domestic populations in other parts
of their own countries. This shared past and present has domestic politi-
cal salience and, even in the face of ongoing antagonisms, may reduce the
willingness of either side to inflict large-scale harm on the other.42
From a more practical perspective, India, though sometimes character-
ized as a Hindu country, is home to approximately 180 million Muslims,
more than any other state except Pakistan and Indonesia.43 Indian leaders
are wary of taking actions that could inflame them. Muslim anger could
increase sympathy for and cooperation with Islamist terrorists plan-
ning to attack targets within India, or even lead to large-scale domestic
unrest.44 A punishment campaign against targets in Pakistan is precisely
the type of policy that could have this effect, and thus carries considerable
domestic political and security risks.
Military calculations also have played a role in India’s failure to
launch punishment campaigns against Pakistan in retaliation for past
24
provocations. From an operational standpoint, the Indians have not

always had good punishment options available to employ against the
Pakistanis. For example, although India has long possessed a significant
conventional military advantage over Pakistan, it has generally not been
sufficient to allow the Indians to defeat the Pakistanis in a quick rout.
Three of four Indo-Pakistani wars ended in a stalemate, with the two
sides restoring the status quo ante once the conflicts had ended. Detailed
estimates in recent decades have indicated that while India would have
won a conventional military contest with Pakistan, it could not have done
so quickly, and would have required weeks of attrition warfare to achieve
victory.45 Thus, a punishment campaign that involved large-scale war
against Pakistan would have proven costly for the Indians, despite their
conventional superiority.
Strictly speaking, India would not have had to defeat Pakistan in a con-
ventional war to punish it. For instance, India could have inflicted pain on
Pakistan by bombing or shelling Pakistani cities, or blockading Pakistani
ports. Such a strategy would have faced several problems, however. First,
Pakistan would not have allowed India to take such steps at will. Bombing
or blockading Pakistan would have involved a costly fight with defending
forces. Second, Pakistan could have retaliated against India in kind, and
the Indians could not have been sure of their ability wholly to stop the
Pakistani attacks. Third, India’s punishment campaign against Pakistan
might not have remained limited; it could have escalated into a large-scale
Indo-Pakistani war, resulting in a bloody, protracted slog.
Pakistani conventional capabilities are not the only military impedi-
ment that India has faced in devising strategies for punishing Pakistan.
Since the late 1980s, Pakistan has possessed at least a latent nuclear weap-
ons capability.46 Any large-scale Indian attack against high-value Pakistani
targets thus would have risked triggering nuclear retaliation against India.
This has not wholly prevented the Indians from fighting with Pakistan. For
example, India waged a high-intensity limited war against the Pakistanis
at Kargil in 1999, less than one year after India’s and Pakistan’s nuclear
tests. The Kargil conflict demonstrated to the Indians that considerable
room existed to fight Pakistan under the nuclear umbrella.47
Yet the Indians simultaneously recognized that nuclear weapons
limited their retaliatory options against Pakistan. While the Pakistanis
would probably not employ nuclear weapons in response to limited
Indian attacks, they might well do so if India behaved in a more provoca-
tive manner. Thus, Indian leaders, even as they recognized their ability
to wage limited war against Pakistan, ruled out the possibility of launch-
ing large-scale attacks or other highly threatening operations against the
Pakistanis.48
25
India could have attempted to avoid these problems by designing

retaliatory measures that, while damaging to Pakistan, did not inflict
large-scale harm on the country. For example, India could have decided
completely to avoid damaging Pakistan proper and instead struck targets
such as terrorist training camps inside Pakistani Kashmir. This approach
would have been considerably less provocative than damaging the
Pakistani homeland but would have harmed a territory and population
that Pakistan values, and might also have hindered Pakistan’s ability to
wage its militant campaign against India.49
Despite its apparent advantages, however, this limited approach might
not have succeeded in altering Pakistani behavior. It would have failed
to impose truly significant costs on the Pakistanis in retaliation for their
militant proxy campaign. Militant training camps are generally located
in remote areas of Pakistani Kashmir, contain little infrastructure, and
hold few high-value human targets. Thus, attacking a handful of such sites
would probably have inflicted only minimal damage on people or terri-
tory of concern to the Pakistanis and done little to hinder the Pakistanis’
ability to wage their militant campaign. As a result, the Pakistanis would
have been unlikely to decide that Indian punishment had made their
proxy war strategy prohibitively costly.
Shared history and culture, a lack of conventional military options, and
nuclear dangers all limit the punishment options available to India as it
seeks to combat Pakistan’s militant strategy. Despite these problems, the
Indians have not ruled out the use of a punishment campaign against the
Pakistanis, and in fact have seriously considered launching one on a num-
ber of occasions. When they have come close to doing so, however, the
Indians have ultimately decided against it. Their unwillingness to resort
to punishment has resulted largely from one of the main benefits of a mili-
tant proxy strategy—uncertainty as to attacker identity.
The identity and composition of militant groups seeking to oust
India from Kashmir have long been common knowledge, as has the fact
that they have received extensive financial, logistical, and military sup-
port from within the Pakistani security establishment. 50 Nonetheless,
Pakistan’s use of proxies has created a significant degree of opacity around
its strategy, obscuring important details regarding Pakistani activities.
For example, it has not always been clear which Pakistani organizations
and personnel were supporting the militant campaign. Was it the army?
Was it the intelligence service? Was it factions within one or both of those
organizations operating without the approval of central authorities? Were
individual Pakistanis who worked with the militants active-duty military
and intelligence officers assigned to that task by their superiors? Or were
they retired or other nonofficial personnel offering freelance support for
26
the anti-Indian jihad? The lack of clear answers to questions such as these
has afforded the Pakistanis a measure of deniability that, in turn, has often
complicated India’s response to the militant campaign. 51
India probably came closest to launching punishment operations
against Pakistan in the wake of the December 2001 militant attack on
the Indian parliament. The Indian government quickly identified Jaish-e-
Mohammed and Lashkar-e-Taiba as the organizations behind the opera-
tion. The Indians subsequently deployed approximately five hundred
thousand troops along the Indo-Pakistani border and threatened to attack
Pakistan unless Islamabad ceased its support for militancy and handed
over to India a list of fugitives implicated in anti-Indian terrorist activity.
Pakistan undertook a reciprocal deployment and soon the two countries
were locked in a standoff that threatened to escalate into large-scale war. 52
Before conflict could erupt, however, Pakistani leaders managed to
defuse the crisis by distancing themselves from the militant groups. They
claimed that although the attackers belonged to organizations operat-
ing within Pakistan, the Pakistani government did not actually support
their activities and would take concrete steps to prevent further attacks
in the future. These arguments convinced American officials attempting
to mediate the crisis that the Pakistan government and military were suf-
ficiently distinct from the militant organizations that they should not be
punished for the militants’ activities. The United States, therefore, encour-
aged India to exercise restraint and refrain from attacking Pakistan,
promising that the Pakistani government would rein in the militants and
prevent events such as the parliament attack from occurring in the future.
This combination of pressure and assurance from the Americans played
an important role in convincing Indian officials to stand down and refrain
from attacking Pakistan. 53 Thus, in the instance where India was perhaps
closest to punishing Pakistan for its involvement with militancy, opacity
regarding the Pakistani government’s actual level of complicity helped
to prevent the Indians from doing so. This uncertainty, in turn, resulted
directly from the nature of Pakistan’s militant proxy strategy. 54
Bargaining: Pakistan has used the challenges inherent in its third-
party relationship with the militants as a means of justifying high payoffs
for bringing them to heel. For example, when Indian leaders press them to
rein in Islamist groups, the Pakistanis state that they cannot control the
militants, who will not stand down until the Kashmir dispute has been
resolved in a just manner. 55 Implicit in this argument is the claim that
Pakistan’s lack of control over its proxies necessitates a high Indian pay-
off to settle the conflict—namely, Indian retreat from Kashmir. Similarly,
Pakistan extracted considerable financial and political support from the
United States in return for its participation in the war on terror following
27
the attacks of September 11, 2001. This high level of payment was in large
part predicated on the difficulty of reining in Pakistan’s erstwhile Taliban
allies, as well as other associated groups. 56
Downsides: Although Pakistan has benefited from the advantages dis-
cussed earlier, its use of militant proxies has also resulted in significant
costs. For example, the militants have adopted far more ambitious goals
than those of the Pakistani government. Lashkar-e-Taiba, for example,
hopes not just to free Kashmir or to claim it for Pakistan, but to conquer
India proper. Significantly, the group’s ambitions do not end on the sub-
continent; LeT views its South Asian struggle as only part of a larger,
global jihad. The group is unconcerned with the impact of such a maxi-
malist agenda on Pakistani interests and is undeterred by the prospect of
opposition from Islamabad. 57
In addition, militant groups have begun to challenge the central gov-
ernment for sovereignty over Pakistani territory. Groups such as the
Tehrik-e-Taliban, for instance, have seized control of large sections of
South Waziristan. There they have repudiated Islamabad’s writ, imposed
an extreme interpretation of sharia law, and supported attacks on govern-
ment and coalition targets in Afghanistan.58
Also, militant organizations have refused to subordinate their inter-
ests to Pakistan’s broader strategic imperatives. For example, following
events such as September 11 and a December 2001 terrorist attack on the
Indian parliament, Pakistan was forced to bow to international pressure
and scale back its support for militancy. The militants viewed this as an act
of betrayal and retaliated violently against Pakistan government targets,
including President Musharraf himself. 59
Pakistan’s militant strategy has also resulted in significant opportunity
costs. Continual support for jihad directly consumes resources and, more
broadly, creates a hostile security environment that forces prioritization
of the military over other sectors, thereby impeding Pakistani internal
development. Pakistan’s education sector offers one of the most urgent
examples of this problem, with only 62 percent of primary school–aged
children and 30 percent of secondary school–aged children actually
enrolled in school.60
Finally, Pakistan’s militant proxy strategy is leading the Indians to
develop new offensive military capabilities. For instance, the Indians
are increasing the speed with which their conventional forces can attack
Pakistan. They hope that this will enable them to retaliate for provoca-
tions before the Pakistanis can ready their defenses or the international
community can convince Indian leaders to stay their hand.61 Pakistan’s
asymmetric warfare strategy, by driving aggressive Indian military inno-
vation, thus threatens to trigger the very disaster that it was designed to
28
avoid: a direct Indo-Pakistani military confrontation. Significantly, given

the two countries’ nuclear capabilities, such a confrontation could result
not simply in a Pakistani conventional military defeat; it could escalate
beyond the conventional level and lead to a potentially catastrophic
nuclear exchange.62
Pakistan is thus caught in a jihad paradox. State weakness has made the
strategic use of Islamist militants attractive to Pakistan. Over the decades,
the strategy has brought Pakistan a number of important benefits that
would otherwise probably have been unattainable. But jihad has outlived
its utility; the very weakness that originally made the strategy useful now
renders Pakistan vulnerable to a host of dangerous pathologies that result
directly from it. If Pakistan fails to abandon militancy and escape its jihad
paradox, the result could be disaster.
POTENTIAL COUNTERARGUMENTS
Here, I consider a number of potential counterarguments to the claims

that I have made. Although these objections raise important issues, they
do not fundamentally undermine my argument.
One possible objection is that my argument is, at root, not really about
militant proxies. Instead, it is about secrecy on the part of the sponsor
state. The operational benefits that the sponsor receives by using militant
proxies are the same as those that it would get through the covert use of
special operations forces.
It is true that secrecy is one important benefit of using militant proxy
forces. It is not, however, the only benefit, and special operations forces
would, in many instances, not be as desirable for a sponsor state as the
use of militant proxies. This is the case for three reasons. First, special
operations forces, as part of the sponsor state’s military, can be more eas-
ily linked back to the sponsor. In the event that special-forces operators
are killed or captured, it would be more difficult for the sponsor to main-
tain deniability than it would be if the personnel in question were mili-
tants who had never officially been part of the sponsor’s military. Second,
a militant strategy enables sponsor states to use personnel indigenous
to the contested area. They are likely to appear more legitimate to local
populations and to international audiences than would members of the
sponsor’s military, who may be ethnically and linguistically different
from the locals. This can significantly aid the sponsor’s efforts, particu-
larly if it is attempting to stoke an ethnically based revolt or insurgency.
Third, militants are likely to be cheaper than special operations forces.
29
They require less training and equipment, and using them does not put
the sponsor state’s own soldiers at risk. Enhanced deniability and legiti-
macy, as well as cost savings, then, can give militant proxies advantages
that special operations forces lack. It is not surprising that the Pakistanis
have used militants more extensively than their own special operations
forces.
A second potential objection to my argument is that it rests on an
overly mechanistic framework. Not all states, a skeptic might claim, will
enjoy the benefits, or bear the costs, that I identify in my discussion of
militant proxy logic. I agree that outcomes in these cases are not fixed and
will vary across time and place. My purpose in exploring militant proxy
logic is not to make ironclad predictions. Rather, I seek to lay bare some
of the core strategic advantages and disadvantages associated with a mili-
tant proxy strategy, to explain more clearly behavior and outcomes in the
Pakistani case, and to suggest possibilities regarding the future behavior
of the other states in a strategic position similar to that of Pakistan.
Actual outcomes in particular cases will depend to a significant degree
on factors such as political preferences, risk acceptance, history and cul-
ture, the conventional military environment, and the presence or absence
of nuclear weapons. Nonetheless, all things being equal, a militant proxy
strategy is likely to hold at least some appeal for weak, dissatisfied states
because of the potential benefits that I identified. And this type of strategy
is likely to be dangerous to such states because of the potential costs that
I discuss. To highlight these possibilities is not to predict that they will
necessarily come to pass.
A third potential objection to my argument is that it overstates the
importance of the role that militants have played in Indo-Pakistani con-
flicts. According to this objection, India’s and Pakistan’s conventional mili-
tary forces, in direct confrontation with one another, ultimately decided
the outcome of the wars of 1947, 1965, and 1971, as well as the 1999 con-
flict at Kargil. Thus, in this view, the significance of the contribution to
Pakistani military efforts by militant proxies has generally been limited.
It is true that in 1947 and 1965 the Pakistanis did not use militants
exclusively, and that both wars eventually evolved into conventional
conflicts between the Pakistani and Indian armies. Nonetheless, in both
cases, militant proxy forces were essential to Pakistani war plans. Indeed,
when they decided to launch these wars, Pakistani planners envisioned
either winning entirely with militants, without a significant combat role
for the regular Pakistan army, as in 1947, or prevailing primarily with
militants and using conventional forces largely in a holding and mop-up
capacity, as in 1965.
30
During the Kargil conflict, offensive forces were not members of

militant groups. Nor were they members of the regular Pakistan Army,
however. Rather, Pakistani fighters came from a paramilitary unit called
the Northern Light Infantry, which was composed primarily of person-
nel from the Kashmir region. Pakistani forces at Kargil therefore held a
type of hybrid status, and the conflict should not strictly be considered an
example of conventional Pakistani war fighting.
Although Bangladesh was primarily a conventional war, it was the
first and last such conflict that Pakistan fought against India. After 1971,
Pakistan never again engaged India directly in a conventional military
conflict. Thus, as I have argued earlier, the Bangladesh war is best viewed
as the exception that proved the importance of Pakistan’s militant proxy
strategy.63
Finally, it is important to note that during the Kashmir insurgency,
which has lasted nearly three decades and cost India more blood and
treasure than all of its other wars combined, Pakistan has used militants
almost exclusively; the Pakistanis have not employed regular military
forces in a combat role in Kashmir. Not only has the Kashmir conflict
been exceptionally costly up to now, but also it could trigger a wider Indo-
Pakistani war in the future. To focus solely on the role of Pakistan’s mili-
tant strategy in more standard Indo-Pakistani wars, then, misses what is
perhaps the most important example of its use.
The potential objections considered previously make a number of
important points. In the end, however, they do not significantly under-
mine my arguments. Secrecy is an important part of the appeal of a mili-
tant proxy strategy, but it does not fully explain such a strategy’s utility.
An abstract assessment of the costs and benefits of using militant proxies
need not be rigidly predictive, but it can alert us to incentives that may
help to explain past behavior and to anticipate future outcomes. Finally,
although Pakistani conventional forces have played an important role in
numerous Indo-Pakistani conflicts, the use of militants has nonetheless
been central to Pakistan’s ongoing efforts to challenge India and alter the
status quo in South Asia.
CONCLUSION
In this chapter, I explored the question of why states might wish to pur-
sue their security interests through the use of nonstate proxies. I argued
that such a strategy can provide a weak sponsor state such as Pakistan
three main benefits. First, it can offer a sponsor a relatively inexpensive
means of fielding forces and engaging a stronger adversary. Second, it can
31
undermine the adversary’s ability to respond to weak-state provocations,

either with denial or with punishment campaigns. Third, it can create
bargaining advantages for a sponsor, potentially enabling it to demand a
higher payoff for settling a dispute than it might have received if it were
fighting alone.
I then showed that, despite these possible benefits, a militant proxy
strategy can pose a number of significant risks for a sponsor state. First,
principal–agent problems can undermine the sponsor’s control over
its proxy forces. Second, the sponsor can suffer opportunity costs as an
increasingly competitive international security environment forces it to
divert resources away from domestic priorities. Third, stronger adver-
saries may respond to sponsor-state provocations by developing highly
threatening retaliatory capabilities, rendering adversaries better able to
harm sponsors than they were previously. Ironically, the very characteris-
tic of a sponsor that can make a militant proxy strategy attractive—state
weakness—can also give rise to these problems and make the strategy
especially costly. For Pakistan, this has resulted in a jihad paradox—jihad
has offered Pakistan one of the few viable means of aggressively pursu-
ing its national security goals, but the strategy has simultaneously put
Pakistan in grave danger. I illustrated this discussion by offering brief
examples of militant proxy logic at work with historical evidence from the
Pakistani case.
In c hapter 3, I return entirely to the Pakistani case, tracing the politi-
cal and material roots of Pakistan’s militant strategy during the early
days of independence. I show that in the wake of the partition, Pakistan
suffered from material and domestic political weakness so severe that it
threatened the new state’s viability. Pakistani leaders decided that they
could ameliorate these problems by emphasizing Pakistan’s Islamic iden-
tity. This project, in turn, could be furthered if Pakistan were able to seize
the disputed territory of Kashmir from India. Although confronting India
over Kashmir would be dangerous, the Pakistanis devised a strategy that
would minimize the costs and risks of an Indo-Pakistani conflict—the
use of Islamist militants as a proxy force to seize Kashmir.
32
CHAP T ER 3
Partition and an
Emerging Strategy
I n chapter 2, I stepped back from the Pakistani case to explore the gen-
eral logic of a proxy strategy. I argued that a state sponsor could realize
cost, operational, and bargaining benefits by employing militant proxy
forces against an adversary. These benefits can make the use of a militant
proxy strategy highly attractive to a sponsor, particularly if it is relatively
weak and unlikely to prevail in a direct military confrontation. I then
explained that a militant proxy strategy can subject sponsor states to sig-
nificant control problems, developmental opportunity costs, and external
security threats. The likelihood of suffering from these problems is espe-
cially high for weak states, which may lack the ability to protect them-
selves against them. As I showed briefly at the end of chapter 2, Pakistan
has profited from all three types of benefits throughout its history, and
also fallen prey to all three types of dangers.
In this chapter, I return my focus entirely to the Pakistani case and
begin to explore its historical evidence in detail. Pakistan used nonstate
proxies as strategic tools, and benefited from the advantages that I out-
lined in c hapter 2, from its earliest days of nationhood. To ameliorate
severe material and political weaknesses, Pakistani leaders decided to
emphasize the country’s Islamic identity, hoping that religion’s broad
appeal would help to hold the country together. Wresting control of the
Muslim-majority territory of Kashmir from India could contribute to this
effort. Given Pakistan’s acute lack of resources, however, doing so would be
difficult and dangerous. Pakistani leaders therefore devised a strategy of
fomenting unrest within Kashmir and then using local militants to attack
the territory. This, they hoped, would enable them to acquire Kashmir
without the cost and risk of direct military intervention. Although the
33
Pa r t i t i o n a n d a n Em e r gi n g S t r at e g y ( 33 )
Pakistanis were not able to seize the territory entirely, they did succeed
in taking roughly one-third of it, fighting the Maharaja of Kashmir and
the Indian forces sent to support him to a standstill. This ensured that the
use of militant proxies became a central pillar of Pakistani security policy,
with the strategy’s importance and sophistication increasing with every
subsequent conflict.
PARTITION AND THE SOURCES OF PAKISTAN’S

MILITANT PROX Y STRATEGY
The primary motivation for Pakistan’s initial resort to a militant proxy

strategy was a desire to redress the country’s material and political weak-
ness. Pakistan emerged from the ashes of India’s 1947 partition, which
divided British India between a now-independent India and the entirely
new state of Pakistan. The fledgling state immediately found itself in an
extremely vulnerable position on a number of fronts.1 First, prepartition
map making had left Pakistan with a territorial allotment that was not only
far smaller than India but also difficult to defend. One thousand miles of
Indian territory separated East and West Pakistan, forcing the Pakistanis
to divide their military assets and to contend with unwieldy lines of com-
munication between the country’s two wings.2 West Pakistan, moreover,
was physically narrow and lacked strategic depth. Two of its major cit-
ies, Islamabad and Lahore, were located within one hundred miles of the
Indo-Pakistani border. The country was thus potentially susceptible to
an Indian armored thrust. The Indo-Pakistani territorial division also
left much of the choicest agricultural land, as well as the majority of the
subcontinent’s industrial capacity, on the Indian side of the border. In
addition, the partition’s mass violence and population transfers had rav-
aged significant portions of Pakistani territory and forced it to absorb
large numbers of refugees streaming in from India. This created serious
humanitarian challenges, including resettlement difficulties and public
health crises. 3
These territorial problems were exacerbated by Pakistan’s severe lack
of economic, military, and bureaucratic resources. Prior to independence,
Indian and Pakistani leaders had agreed that Pakistan would receive
17.5 percent of British India’s financial assets and 30 percent of its military
resources. Even if Pakistan had received this official allotment on time
and in full, it would have been severely disadvantaged in comparison to
India. After partition, however, the Indian government stalled, repeatedly
delaying the transfer of funds and materiel to Pakistan that had been spec-
ified in their agreement. The Indians did not deliver Pakistan’s allotment
34
of cash balances until January 1948, after intense pressure, including a

fasting campaign from Mahatma Gandhi. Pakistan’s share of military
equipment arrived both late and incomplete, with the Indians supplying
approximately twenty-three thousand out of one hundred sixty thousand
tons of ordnance that the Pakistanis had been allotted.4
Indian unwillingness to deliver promised resources, coupled with the
lingering effects of the mass violence and upheaval of partition, encour-
aged Pakistani leaders to believe that the Indians had never accepted
Pakistan’s creation and had tried to destroy it before it could become a
viable state. This, in turn, greatly heightened Pakistani leaders’ sense of
insecurity, their feelings of antagonism toward India, and their desire to
ensure that Pakistan acquired sufficient strength to never again find itself
in such a vulnerable position. Mohammed Ali Jinnah referred to Indian
stalling and shortchanging of Pakistan as “efforts … to strangle the new
[Pakistani] State at its very birth,” achieving “what the fire and sword of
the enemy could not” during the violence of partition. “But,” he contin-
ued, “not only has Pakistan survived … it has emerged stronger, more
chastened and better equipped than ever.”5
Finally, in addition to this lack of financial and military resources,
Pakistan was hobbled by a dearth of bureaucratic capacity. India had
inherited the British Raj’s administrative system, with its hub at New
Delhi, and regional superstructure radiating outward through the rest
of the country. Much of Pakistan, by contrast, was located in what the
British had considered frontier territory, organized primarily for exter-
nal defense rather than for effective domestic administration. As a result,
the Pakistanis were forced to construct a central government largely from
scratch, with officials lacking even such rudiments as desks and chairs.
This significantly impeded Pakistan’s ability to coordinate national, pro-
vincial, and local affairs and address the severe humanitarian and security
challenges that it faced in the weeks and months following partition.6
In addition to territorial, financial, military, and bureaucratic short-
comings, Pakistan also suffered from severe political weakness. This
weakness resulted primarily from Pakistan’s lack of a solid intellectual
foundation, tension between Pakistan’s creation and the economic and
political interests of many South Asian Muslims, and ambiguity regard-
ing the role that religion was to play in Pakistan. Together these problems
resulted in a lack of clarity regarding the purpose, and even the necessity,
of the new Pakistani state.
The case for the creation of Pakistan was based on what was known
as the “two-nation” theory. The theory maintained that India’s Muslim
and Hindu communities constituted two separate national groups, each
of which possessed unique traditions, histories, cultures, and religions.
35
Indian Muslims needed their own autonomous homeland to practice

their religion, traditions, and culture free from Hindu domination. Such
domination was inevitable in a unified India, given the subcontinent’s
demographics, in which Muslims composed but a small minority in a vast
Hindu population. The only practical solution to this problem, propo-
nents of the two-nation theory believed, was political independence for
India’s Muslim community.7
The British had historically treated India’s Hindu and Muslim com-
munities as different political entities. For example, the 1909 Indian
Councils Act had created separate Hindu and Muslim electorates in
response to Indian National Congress demands for increased Indian
self-government, and Muslim pressure for institutional protection from
a large Hindu electoral majority. The British also hoped that such elec-
toral separation would help to prevent Hindus and Muslims from form-
ing a united front in opposition to the Raj. Indian Muslims such as the
reformer and philosopher Syed Ahmad Khan had previously laid the
foundations of a distinct Muslim cultural and political identity through
their writings and activism in the late nineteenth and early twentieth cen-
turies. A sense of Muslim community ethos, distinct from the primarily
Hindu Indian National Congress, had initially begun to develop at Syed’s
Muhammedan Anglo-Oriental College in Aligarh during this period.8
Despite such growing communal distinctions, however, Muslim lead-
ers had not sought to achieve formal political separation from the rest
of India. Instead, they had worked in a generally unified fashion with
Hindus to extract concessions from the British, increase the degree of
Indian self-government, and move the country forward toward indepen-
dence. This was true even of Jinnah, who eventually assumed leadership
of the All India Muslim League and became the driving force behind
the Pakistan movement. Jinnah had begun his career with the Indian
National Congress and served as one of the independence movement’s
foremost spokespeople for Hindu–Muslim unity. Even after joining the
Muslim League, he remained a passionate advocate for communal har-
mony, maintaining that good relations between Hindus and Muslims
would be essential to successful Indian self-government.9
The goal of creating a separate political entity on the basis of the two-
nation theory emerged with the poet Muhammed Iqbal’s call in 1930 for
a Muslim state consisting of Punjab, the North-West Frontier Province,
Sind, and Balochistan. Such a “consolidated Northwest Indian Muslim
state,” providing “the fullest national autonomy,” would afford “the Indian
Muslim … full and free development on the lines of his own culture and
tradition in his own Indian homelands.” This was, in Iqbal’s view, “the
final destiny of the Muslims, at least of Northwest India.”10
36
Although Iqbal’s vision received some support from students and intel-
lectuals during the 1930s, it did not become a viable political project until
the Muslim League called in 1940 for the creation of two states out of
the old British India. Jinnah and the league then made the cause a cen-
tral plank of their electoral platform, arguing that Congress was a Hindu
institution that, by its very nature, would fail to protect Muslim interests
in a unified independent India. The manner in which many Hindu leaders
framed their arguments justifying the Indian National Congress Party’s
preeminent role in the independence movement, and in a post-British
India, lent Jinnah and the Muslim League’s position a significant measure
of credence. These leaders insisted that a wholly secular Congress should
be sole heir to the British Raj, as it was the only political organization that
could represent all Indians regardless of their communal differences or
preferences. Historically, India’s physical geography had only rarely coin-
cided with its political architecture. The British had attempted to per-
manently create such a situation, developing centralized political power
rooted in a single, unified sovereignty that extended its reach through-
out the entire country. The Indian National Congress internalized this
approach, adopting the British goal of creating what Ayesha Jalal calls
“a composite nationalism based on an indivisible central authority.”11
The rhetoric of leading Hindu figures emphasized both the exclusiv-
ity of Congress’s claim to govern independent India and its insistence on
the creation of a nonsectarian, all-India national identity. As Mahatma
Gandhi argued, “Congress alone claims to represent the whole of India,
all interests. It is no communal organization; it is a determined enemy of
communalism in any shape or form.” Nehru similarly maintained, “There
are only two forces in the country, the Congress and the [British] govern-
ment… . It is Congress alone that is capable of fighting the government.
The opponents of Congress are bound with each other by a community of
interests.”12
It was not unreasonable for an organization attempting to consolidate
power in a geographically, linguistically, and religiously diverse coun-
try after centuries of colonial domination to adopt such an approach.
Insisting that it was the exclusive heir to the Raj enabled Congress to
neutralize potential separatist movements and to bring India’s quasi-
independent princely states within its control. In addition, its claim to be
the sole legitimate representative institution in an India riven by religion,
ethnicity, and caste necessarily implied a secular orientation for the new
Indian state.13 Despite this internal logic, however, Congress’s approach
also seemed to suggest that groups whose members’ primary identity
was not anchored in the abstraction of a homogenous, secular nation-
state might be marginalized—particularly if they also were members of
37
a minority religious or cultural community. Thus, ironically, Congress’s

all-encompassing secularism, by appearing to leave limited room for reli-
gious or cultural distinction in the new India, may have fed Muslim fears
of discrimination and helped to advance the Muslim League’s two-nation
project. Indeed, campaigning on an anti-Congress, two-nation platform
during the 1945–4 6 provincial and general elections, the Muslim League
was able to win all of the Muslim seats in the central legislative assembly,
and almost 75 percent of all Muslim votes in provincial assembly races
throughout India.14
Despite the apparent utility of the two-nation project as a campaign
issue, many of its practical requirements remained unclear. For example,
it was not obvious that, to enjoy broad political autonomy from India,
a new Muslim state would have to be wholly separate and independent
from a larger Indian Union. In principle, an autonomous Muslim nation
could have existed within an Indian Union under a special power-sharing
agreement.15 In his early articulation of the two-nation theory, Iqbal had
called for such an arrangement. Iqbal envisioned the creation of a loose
federal system consisting of a “Central Federal State” as well as multiple
“self-governing states.” Under this system, “residuary powers” would “be
left entirely to self-governing states, the Central Federal State exercising
only those powers which are expressly vested in it by the free consent of
federal states.”16
It is possible that Jinnah had a similar arrangement in mind, rather than
the establishment of an entirely separate nation-state, when he initially
became a proponent of the two-nation theory and took up the Pakistan
project.17 The political, legal, military, economic, and humanitarian impli-
cations of such a federal arrangement, which would not have required a
full-scale partition like the one that eventually occurred in 1947, would
have been very different from the implications of establishing an entirely
separate and independent Pakistani nation-state. As the two-nation proj-
ect progressed rapidly from intellectual debating point to mass political
movement, however, there was little opportunity to analyze the differ-
ences between competing models of autonomy or to articulate these dif-
ferences to the public. As a result, it was not clear that the ultimate result
of this process, a wholly independent Pakistani state, was in fact juridi-
cally necessary.
In addition to this juridical ambiguity, the case for Pakistan faced
other challenges within the Muslim community. Perhaps most impor-
tant, it was not obvious that the creation of a new Pakistani state would
advance Indian Muslims’ economic and political interests. In some cases,
the protections of the new state would be unneeded. In the North-West
Frontier, for example, Pashtuns enjoyed a significant numerical majority
38
and did not fear Hindu domination in an independent India. In other

cases, the creation of a new state would actually be harmful to Muslim
interests. Muslims in the Punjab, for example, would lose the government
ministries that they had come to dominate, as well as highly productive
agricultural areas. Those in Bengal would lose the political and economic
hub of Calcutta. The fiercely independent tribes of the Northwest and
Balochistan would be forced to submit to the central authority of the
Pakistani state. Muslims from other regions, who remained behind in
independent India, would be left without the support of their Pakistani
brethren. Finally, India would retain the Raj’s industrial infrastructure,
which was centered around Bombay, Ahmedabad, Calcutta, and West
Bengal and Bihar. This would leave Pakistan, which possessed 23 percent
of the Indian subcontinent’s landmass and 18 percent of its population,
just 10 percent of its industrial base. Thus, as Ayesha Jalal argues, “The
most striking fact about Pakistan is how it failed to satisfy the interests
of the very Muslims who are supposed to have demanded its creation.”18
Given these juridical, economic, and political problems, the need
for Pakistani statehood was not universally obvious to British India’s
Muslims, whose support would be essential to the success of the Pakistan
project. The Muslim League therefore offered only vague public justifi-
cations for the creation of Pakistan. In Jinnah’s words, Pakistan would
provide a “State in which we could live and breathe as free men and which
we could develop according to our own lights and culture and where prin-
ciples of Islamic social justice could find free play.”19 Such language held
widely different meanings for different audiences. To some, it meant that
Pakistan would be an Islamic state, governed according to religious prin-
ciples. Much of Jinnah’s own rhetoric supported such an interpretation.
For example, by excoriating the Indian National Congress as an instru-
ment of Hindu interests, he implied that the new Pakistani state not only
would have a religious foundation but also would be directly opposed to
“Hindu” India.20
To others, the Muslim League’s arguments for Pakistan indicated that
although the new state was to serve as a Muslim homeland, it would do
so in a pluralistic and wholly secular manner. Jinnah’s own behavior and
rhetoric often suggested as much. At a personal level, Jinnah was famously
irreligious; he drank alcohol, ate pork, and generally avoided religious
observances. Rhetorically, Jinnah advanced what appeared to be a thor-
oughly secular vision for the new country. As he put it to Pakistan’s con-
stituent assembly in August 1947:
You are free; you are free to go to your temples. You are free to go to your mosques or
any other places of worship in this State of Pakistan. You may belong to any religion
39
or caste or creed… . We are starting in the days when there is no discrimination, no

distinction between one community and another… . We are starting with this fun-
damental principle that we are all … equal citizens of one State. 21
Such language strongly suggested to many observers that, although Jinnah

sought to create a home for South Asian Muslims, he did not intend for
it to be strictly Islamic in character. This was probably a more accurate
assessment of Jinnah’s desires and intentions than the more religious
interpretation mentioned earlier. As Ian Talbot argues, “Jinnah’s aim and
that of the professional elite who controlled the [Muslim] League was to
wrest a state in which Muslim economic, political and cultural interest
could be safeguarded, but not to create an Islamic state.”22 Nonetheless,
Jinnah and other Muslim League leaders deliberately employed rhetoric
that could be seen as supporting both religious and secular interpreta-
tions of the founding principles of the new Pakistan.
This lack of clarity regarding Pakistan’s founding principles increased
the new state’s appeal to a diverse range of constituents. Such wide appeal
was essential to the Pakistan movement’s success, given its origins as a
top-down project championed by political and intellectual elites, rather
than as a bottom-up, grassroots phenomenon driven by broad-based pop-
ular pressure. But this religious ambiguity, combined with the juridical
problems and political and economic costs discussed earlier, also contrib-
uted to a crisis of identity, and deep uncertainty as to Pakistan’s reason
for existence. What was Pakistan’s purpose? Why was its establishment
necessary? What national goals would it pursue in the years and decades
to come? No immediate answers to these questions were evident.23 But
such answers would be essential if Pakistan were to achieve the political
unity that it would need in the difficult domestic conditions and hostile
international environment that it faced.
In the wake of independence, Pakistani leaders attempted to address
the new country’s material and political challenges. In doing so, they were
determined to ensure not simply that Pakistan survived, but also that it
achieved some measure of equality with India and did not settle for infe-
rior status vis-à-v is its larger neighbor. They understood, of course, that
structural impediments such as size, population, and natural resources
would always prevent Pakistan from directly matching Indian capabili-
ties. Nonetheless, they sought to achieve with India a broad, normative
equality, which T. V. Paul calls “civilizational parity.” This was a higher
standard, requiring a far more complex and competitive approach than
mere survival would have.24
The Pakistanis adopted a number of policies to achieve their ends.
For example, they joined international financial institutions such as the
40
International Monetary Fund and the World Bank, as well as military alli-
ances such as the Southeast Asia Treaty Organization (SEATO) and the
Central Treaty Organization (CENTO). They established close bilateral
security relationships with the United States and China. Pakistani leaders
also spoke directly to the people, encouraging them, cajoling them, and
attempting to define for them the purpose of their new national project.25
These measures played important roles in helping to ensure Pakistan’s
viability in the wake of partition. Despite their significance, however,
Pakistani leaders viewed still another strategy, which was purely mili-
tary in nature, as especially important and promising. It was a strategy
that could address the country’s material and ideological problems while
also speaking to a deep sense of injustice that many Pakistanis had felt at
their treatment by India during the partition process. This was the pos-
sibility of acquiring the territory of Kashmir. Kashmir, which was located
directly between the new Pakistani and Indian states, was significant for a
number of reasons. At the material level, Kashmir could provide Pakistan
with desperately needed strategic depth. Forces positioned there could
potentially threaten key areas of India in the event of conflict. In addition,
the region contained important water resources.26
Even more significantly, however, the acquisition of Kashmir could
strengthen Pakistan’s tenuous political foundations. Pakistani leaders had
decided that, if their new country were to survive, they would have to put
an end to the ambiguity surrounding its fundamental purpose. To do so,
they would need to ensure that Pakistan did not follow a wholly secular
path and become simply a pluralistic homeland for South Asian Muslims.
Rather, the new country would have to become a state based on a concept
meaningful to the majority of ordinary Pakistanis, regardless of their eth-
nic, economic, or geographical interests or backgrounds—it would have to
become a state based on Islam. As George Cunningham, governor general
of the North-West Frontier Province, put it to Pakistani Prime Minister
Liaquat Ali Khan, for the Pakistan project to succeed, Pakistan and Islam
must be “really synonymous.”27 The army, which quickly emerged as
Pakistan’s preeminent institution, in particular sought to use Islam as a
means of promoting national unity. Not coincidentally, the army’s use of
religion also justified its own leading position within Pakistan, since it
offered the country’s only means of defense against “Hindu” India.28
Ensuring that Pakistan became synonymous with Islam would have
important implications for Pakistani state building, helping to make it a
fundamentally revisionist project, opposed to the territorial status quo
in South Asia. Iqbal had held that Islam did not recognize the primacy
of Westphalian, sovereignty-based territorial arrangements such as the
modern nation-state, which it viewed as inferior to supranational religious
41
institutions. Indeed, he had believed that the nation-state was nothing

less than a tool of British colonialism, designed to facilitate Western intel-
lectual hegemony over Muslims. Nationhood, in this view, could prop-
erly exist only in the context of a religious community, or millat. Thus, all
South Asian Muslims constituted a single nation of the faithful, united by
the teachings of the Prophet and the Koran.29 In this conception, inter-
national boundaries were less important than supranational communal
identity.
This strain of thought informed Pakistani leaders as they developed
their new country’s strategy, and it remained relevant decades later. For
example, in the preface to Brigadier S. K. Malik’s The Quranic Concept of
War, a 1979 treatise on Islam and warfare, former Pakistani ambassador
to India Allah Bukhsh K. Brohi argued that Islam “does not subscribe
to the concept of the territorial state.” The reason is that a true Muslim
community, the “Umma of Mohammed,” is “incapable of being realised
within the framework of territorial states.” Even “Muslim” nation-states
can therefore be viewed as possessing only “interim” status since they
eventually are to be incorporated into a Muslim “commonwealth.” In
the meantime, these states must “acquire strength and stability,” thereby
enabling them to “prepare the ground on which a unified state of Islam
can appear on the historical scene.” According to Brohi, the sovereign
rights of existing states must not stand in the way of this project, since “in
Islam … no nation is sovereign… . God alone is the only sovereign in
whom all authority vests.”30
This view did not, of course, lead Pakistani leaders to deny the legiti-
macy of their own state-building project. It did, however, justify Pakistani
efforts to undo existing regional territorial boundaries, which separated
Muslims from one another and prevented the establishment of a larger
religious community. Indeed, in this view, a Pakistan that took its Islamic
identity seriously could not help but oppose the territorial status quo, for
this was the only way for a unified Muslim political entity to emerge in the
region. 31 As Olivier Roy puts it, “Pakistan is first and foremost an ideologi-
cal concept: The Muslim state on the Indian subcontinent. It therefore
has a duty to defend and represent all Muslims in the region, and its ter-
ritorial limits are … immaterial.”32
The most objectionable aspect of territorial status quo in South Asia was
the plight of Kashmir, which, despite being a Muslim-majority territory,
was legally part of the Indian Union. Correcting this injustice and estab-
lishing Pakistani control over Kashmir would bring Muslims together
into a larger political community, thereby honoring the commitments
inherent in Pakistan’s Islamic identity, while also supporting the two-
nation theory’s claim that South Asian statehood should be determined
42
on the basis of religion. To allow Kashmir to remain part of India, by con-

trast, would suggest that a Muslim-majority territory need not be part of
a Muslim political entity, but could instead exist within a Hindu-majority
Indian state. This would undermine the principles inherent in Pakistan’s
Islamic identity and directly contradict the arguments that Jinnah and
the Muslim League had made when promoting the Pakistan project prior
to independence. Thus, by strengthening the country’s ties to Islam and
to religion-based nation building, Kashmir offered Pakistan an opposi-
tional external solution to its potentially fatal problem of internal political
cohesion. As Farzana Shaikh puts it, Kashmir created the possibility that
foreign policy could serve “as a vital compensation for [Pakistan’s] lack of
a clearly defined sense of nationhood.”33
Acquiring Kashmir would not be easy, however. Historically, Kashmir
had been one of India’s more than five hundred princely states. Individual
treaties determined the precise terms of these states’ relations with the Raj.
In general, they enjoyed considerable autonomy, deferring to the British
only on matters such as defense and foreign affairs. With independence,
the former princely states had to decide whether to join India or Pakistan.
Although they were in principle free to opt for one country or the other,
their choice was constrained by two ground rules. First, they had to honor
their countries’ religious demographics, with Hindu-majority territories
acceding to India, and Muslim-majority territories acceding to Pakistan.
Second, they had to bow to geographical realities; a state located deep in
Indian territory could not join Pakistan, and a state deep in Pakistani ter-
ritory could not join India, regardless of the state’s demography.34
Maharaja Hari Singh, who governed Kashmir, remained undecided as
to whether to join the territory to India or to Pakistan. His choice was
complicated by two factors. First, although the majority of Kashmir’s
population was Muslim, Hari Singh was himself Hindu. Second, Kashmir
was located directly between India and Pakistan. Thus, it was not obvi-
ous, either from a demographic or from a geographical standpoint, which
country Kashmir should join after partition.35 As a result, the maharaja
was unsure what to do and delayed making a decision regarding the dis-
position of his state.
Hari Singh’s dithering left Kashmir without the protection of a state
patron, and thus vulnerable to attack. This vulnerability created an oppor-
tunity for Pakistan. It was, however, an opportunity fraught with risk. If
Pakistan attempted to seize Kashmir, it could potentially find itself fight-
ing not only the maharaja’s forces but also the Indian military, as Indian
leaders were acutely interested in the territory’s fate. Nehru in particular,
whose family originally hailed from Kashmir, felt a strong personal attach-
ment to the region. Substantial evidence suggests that Indian leaders may
43
have harbored designs to incorporate Kashmir into the Indian Union

even before the outbreak of hostilities.36
Given Pakistan’s material weakness, becoming enmeshed in a direct
conflict with India over Kashmir would be extremely risky. If they were
to move against the territory, then, the Pakistanis would need a means
of offsetting this danger. They found such a means in the form of a mili-
tary strategy that would minimize the direct use of their armed forces in
Kashmir, thus avoiding confrontation with defending forces and lowering
the risk of large-scale Pakistani losses.
PAKISTAN’S MILITANT STRATEGY

AND THE FIRST KASHMIR WAR
Pakistani strategy followed the broad contours of Colonel Akbar Khan’s

secret plan for “Armed Revolt Inside Kashmir.” Khan served as direc-
tor of weapons and equipment at the Pakistan Army general headquar-
ters in Rawalpindi. It was a post that afforded him considerable insight
into Pakistani logistical capabilities and enhanced his ability to supply
an insurgent force with necessary arms and materiel. Khan had initially
been approached by a senior Muslim League leader, Mian Iftikharuddin,
and asked to devise a strategy by which Pakistan could unofficially help
Kashmiri Muslims to prevent India from acquiring Kashmir. He received
little guidance other than this. As Khan put it, “the object to be achieved
by the contemplated action was clear, namely to get Kashmir’s accession
to Pakistan.” Otherwise, Iftikharuddin’s request was “entirely vague.” He
said only “that some money could be spent, though he did not know how
much. Further, he said that any action by us was to be of an unofficial
nature, and no Pakistani troops or officers were to take part in it.”37 Based
on these very broad instructions, Khan quickly devised a plan to trigger
“armed revolt” in Kashmir by “strengthening the Kashmiris themselves
internally” and by “[preventing] the arrival of … military assistance
from India into Kashmir.” He envisioned diverting four thousand rifles
from the Punjab police along with condemned ammunition to poten-
tial Kashmiri rebels, and blocking the land and air routes from India to
Kashmir on the Kathua-Jammu road and at Srinagar airport. 38
This plan was modified after further consultation with Iftikharuddin
and other Pakistani leaders. The eventual Pakistani strategy consisted
of two main stages. The Pakistanis would seek first to destabilize the
Kashmiri government by capitalizing on brewing unrest in the Poonch
region of Kashmir. Maharaja Hari Singh was deeply unpopular in
Poonch. His policies had eroded Poonchi autonomy from the Kashmiri
44
government, imposed punitively high taxes on the region, and sought

to disarm the population in a manner that appeared primarily to target
Muslims. 39 The Pakistanis hoped to help local rebels to transform simmer-
ing discontent with the maharaja’s rule into a full-blown revolt. Second,
the Pakistanis planned to assist militias of several thousand tribesmen in
launching an external attack on Kashmir. The tribesmen were composed
mainly of Mahsuds and Waziris from the North-West Frontier Province,
though others hailed from far-off locations such as Balochistan and the
Afghan–Pakistan border region. The Pakistanis hoped that this combi-
nation of internal and external pressure would overthrow the maharaja
and help to ensure Kashmir’s accession to Pakistan. Significantly, the plan
would enable Pakistan to avoid direct confrontation with enemy forces, as
its military would play only supporting roles. For example, the Pakistanis
would arm the fighters attacking Kashmir; operate their radio network;
supply them with food, clothing, and other materiel; cut regional road and
rail links; and prevent the provision of such essentials as food and gasoline
to forces defending Kashmir.40
The Pakistanis guarded their preparations closely. The need for secrecy
was particularly acute because of the dual nature of Pakistan’s military
command structure. Although Pakistan was now an independent coun-
try, its army still included British personnel, who were assigned to assist
the Pakistani military with the transition to statehood. Had the British
gotten wind of Pakistani plans, they would likely have put a stop to them,
or at least have alerted their compatriots in the Indian Army, which had
a similar transitional command structure. Khan therefore employed only
a small group of military and civilian collaborators, including Brigadier
Sher Khan, an intelligence officer who had written an assessment of the
likelihood of popular revolt in Kashmir; Colonel Tommy Masud, a cavalry
officer who helped to collect and store ammunition; and Air Commodore
M. K. Janjua and Rawalpindi Commissioner Khwaja Abdul Rahim, who
provided further logistical support ranging from the provision of cloth-
ing and rations to the procurement of volunteers. Khan also had the
approval and assistance of key Pakistani officials, including Pakistani
Prime Minister Liaquat Ali Khan, who was present at meetings during
which the plan was discussed in detail. Thus, despite its secrecy, Colonel
Khan’s operation had the imprimatur of Pakistan’s political leadership.41
This may well have included the tacit approval of Jinnah himself.42
The initial phases of the operation went largely as the Pakistanis had
foreseen. Internal uprisings, aided by Pakistani weapons and supplies,
as well as by some tribal infiltration from Pakistani territory across the
Jhelum River, succeeded in loosening the maharaja’s grip on power during
45
the first weeks of October 1947.43 Even more effective, however, was the
tribal militias’ external attack on Kashmir, which began on October 22.
Invading forces quickly captured Muzaffarabad and pushed toward the
Kashmiri capital of Srinagar. Although the tribesmen plundered exten-
sively, they were not driven exclusively by the quest for booty; religion
also played an important role in motivating them. The militants consis-
tently referred to their attack on Kashmir as “jihad” and stated that they
were liberating the territory from Hindus for their Muslim brethren.44
The Pakistani government sought to promote this view, asking Muslim
clerics to issue fatwas declaring that the tribesmen’s invasion was in fact a
bona fide jihad. Colonel Khan himself adopted the nom de guerre General
Tariq, hearkening back to Tariq bin Ziyad, the Muslim conqueror of Spain
who gave his name to Gibraltar. According to Andrew Whitehead, the
militants’ religious motives played a central role in driving their efforts
and were “even more important [to them] than evicting Hari Singh from
his throne.”45
In the face of this tribal invasion, a frightened Hari Singh appealed
for Indian military assistance in repelling the intruders. Lord Louis
Mountbatten, the British viceroy, and Indian Prime Minister Jawaharlal
Nehru agreed to the maharaja’s request for help. They stipulated, how-
ever, that in return for India’s intervention, Kashmir must accede to the
Indian Union. They also insisted that, upon the cessation of hostilities in
the region, the Kashmiri people would ratify the accession through a pleb-
iscite. Hari Singh agreed to these terms, signing an instrument of acces-
sion on October 26, 1947.46
Indian forces were airlifted into Kashmir immediately thereafter and
soon intercepted the intruders, thwarting their advance on Srinagar.
Pakistan supported the militants against the Indians but kept its assis-
tance unofficial until spring 1948, when the army formally took charge
of the war effort.47 The conflict proved to be indecisive. India succeeded
in defending critical areas such as Srinagar and the Valley of Kashmir.
But it was not able to expel intruding forces altogether or to compel
Pakistan to cease its operations within the territory, despite a large-
scale military commitment, threats to invade Pakistan proper, and an
extensive diplomatic campaign, which included energetic lobbying
against Pakistan in international fora such as the United Nations. After
approximately one year, the conflict ground to a stalemate, ending with
a UN-sponsored ceasefire on January 1, 1949. The ceasefire left one-
third of the territory in Pakistani hands and two-t hirds under Indian
control, essentially the same division of territory as had existed at the
beginning of hostilities.48
46
EVALUATING PAKISTANI STRATEGY
Did Pakistan’s use of militant proxies during the first Kashmir war
enable it to benefit from the strategy’s three advantages that I discussed
in c hapter 2? In this section, I argue that Pakistan probably did realize
all three types of advantages during the war, though to widely varying
degrees.
The first potential benefit of a militant proxy strategy is reduced cost.
The Pakistanis almost certainly realized significant cost benefits by using
rebels and militants, rather than their own conventional military forces,
to launch the first Kashmir war. The initial phase of the Pakistani plan,
which involved supporting an uprising in Poonch against the maha-
raja, did not require the Pakistanis to field forces of any kind. Instead,
the Pakistanis simply supplied the Kashmiris with a modest number of
rudimentary firearms to assist them in carrying out their rebellion. The
Pakistani plan’s second phase did require Pakistan to provide invading
tribesmen with more extensive organizational and logistical support
than they had previously given to the rebels in Poonch. It is impossible to
know exactly how the cost of this support compared to the likely cost of
using Pakistan Army forces to launch an attack on Kashmir. But given the
indirect nature of the assistance, which did not include Pakistani involve-
ment in combat operations, it is clear that even the second phase of the
Pakistani plan was relatively cheap. The low cost of the indirect approach
was extremely important, as it enabled the Pakistanis to husband their
scarce military and financial resources.
It is true that, although Pakistan exclusively used rebels and mili-
tants in the opening phases of the conflict, the Pakistani military did not
remain on the sidelines of the first Kashmir war indefinitely. They became
directly involved in the fighting after Indian forces intervened in Kashmir
and successfully blunted the tribesmen’s advance on Srinagar. During
this phase of the conflict, Pakistan did not benefit from a proxy strat-
egy and paid the full price of direct military confrontation with India.
Nonetheless, the Pakistanis’ initial use of proxies probably made the war
less costly overall than it otherwise would have been, as it avoided direct
Pakistani military involvement in the fighting for at least some period of
time. Had the Pakistanis not used proxies during the opening phase of
hostilities and instead fought directly with the Indians throughout the
conflict, the first Kashmir war would have been even more costly for the
Pakistanis than it actually was.
The second potential benefit of a militant proxy strategy is limitation of
the military options open to adversary states. Did Pakistan’s use of rebels
and militants limit the military responses that India was able to employ in
47
response to the invasion of Kashmir? The answer to this question is mixed.

Pakistan’s proxy strategy does not appear seriously to have undermined
Indian efforts to defend Kashmir and deny intruding forces a battlefield
victory. Although the Indians were not able wholly to eject the Pakistanis
and their allies from the territory, they did succeed in defending critical
parts of it, including Srinagar and the Valley of Kashmir. Tribal forces did
not generally employ guerilla tactics, which would have made them diffi-
cult to fix and destroy. Rather, they openly marched on Kashmir, enabling
the Indians to identify and engage them. It does not appear that this was
any more difficult than fighting a conventional army would have been for
the Indians. Indeed, given the tribal militias’ rudimentary capabilities, it
was probably easier.
If Pakistan’s use of proxies did not derail Indian denial operations, did
the strategy help to limit India’s ability to punish Pakistan for attacking
Kashmir? There is a strong possibility that this was the case. Indian leaders
had publicly threatened to retaliate against Pakistan for its provocations
in Kashmir by launching a cross-border invasion not only into Kashmir
but also into Pakistan proper.49 In the end, however, India did not make
good on its threat to invade. Why did it fail to do so?
Opposition from Great Britain played a major role in convincing the
Indians not to attack Pakistan. The British wished to prevent the outbreak
of a wider war that could harm Pakistan, as well as to achieve a settlement
on Kashmir favorable to Pakistani interests. They were motivated pri-
marily by concerns regarding the Kashmir conflict’s possible impact on
British equities in the Middle East. Following World War II, Great Britain
had relinquished its League of Nations mandate for Palestine, placing
the Arab–Israeli dispute before the United Nations. The United Nations
had then passed a measure calling for the partition of Palestine into Arab
and Jewish states. This, in turn, triggered the eruption of Arab–Israeli
violence. British leaders realized that their policies had deeply angered
Palestinian Arabs. They now worried that if they were perceived as sup-
porting India over Pakistan in the Kashmir dispute, they would further
inflame Arab opinion, badly damaging their position in the Middle East
and throughout the larger Muslim world. 50
The British thus sought to influence the outcome of events in Kashmir
in a manner supportive of the Pakistanis. To this end, they strongly dis-
couraged the Indians from widening the war and attacking Pakistan.
Mountbatten forcefully warned Nehru of the potential costs of a larger
conflict and pushed him to refer the dispute to the United Nations. British
Prime Minister Clement Atlee went so far as to suggest that an Indian
decision to attack Pakistan and expand the Kashmir war would violate
international law. 51
48
Given the global pressures that they faced, the British would probably
have argued against Indian expansion of the Kashmir war regardless of
who was responsible for starting it. The Indians, for their part, viewed
Pakistan as the aggressor in Kashmir despite the militants’ role in trig-
gering the conflict. Nonetheless, British arguments might well have been
less persuasive, particularly to an international audience, if the Pakistan
Army had unambiguously begun the war by attacking Kashmir and not
appeared simply to be joining a war that was already in progress. If this
had been the case, the British arguments for forbearance would have
seemed far weaker both to the Indians and to outside observers, and the
British would probably have had more difficulty than they did in pres-
suring the Indians to refrain from attacking Pakistan. This counterfac-
tual discussion is, of course, speculative. It does suggest, however, that
a degree of ambiguity regarding the outbreak of the first Kashmir war,
which resulted from the militants’ role in starting the conflict, may have
helped to insulate Pakistan from an Indian punishment campaign dur-
ing the conflict.
The third potential benefit of a militant proxy strategy is the sponsor
state’s ability to leverage principal–agent problems to improve its bargain-
ing position with the adversary. Did a lack of full control over the mili-
tants enable the Pakistanis to reach an especially favorable settlement to
the first Kashmir war? The suspension of hostilities between India and
Pakistan, which took place on January 1, 1949, simply stopped both sides
in place. The resulting ceasefire line was determined by the location of
Indian and Pakistani forces at the time, and the Pakistanis did not benefit
from any negotiations regarding their ability to influence their militant
allies. The Pakistanis did, however, use their lack of control over the mili-
tants to help them remain in Kashmir long after the ceasefire, indefinitely
retaining the territory that the militants had seized during their invasion.
The Pakistanis were able to do so despite the fact that the ceasefire line had
been understood by all parties to be temporary, and despite the promulga-
tion of a United Nations Security Council resolution stating that Pakistan
and the militants must withdraw entirely from Kashmir. 52
The Pakistanis justified their continued presence by maintaining
that they did not formally control the territory that they held after
the ceasefire, or the people who lived in it. They were simply keeping
custody of the territory until a permanent disposition of Kashmir was
reached. Such a disposition could occur only by direct appeal to the
Kashmiri people, through a plebiscite. Thus, the Pakistanis were pro-
tecting Kashmiri interests by remaining in the captured territory, but
they were not in charge of the militants and other Kashmiris located
49
there and could not force them to withdraw or otherwise cooperate with
India if the Indians would not meet their demands. As Robert Wirsing
argues, this “calculated ambiguity” in Pakistan’s relationship with the
territory and the militants significantly advanced Pakistani interests. It
“strengthened Pakistan’s case for a plebiscite to settle the question of a
permanent boundary between India and Pakistan; and it meant that,
in the meantime, Pakistan would have at its disposal … an ostensibly
independent entity that could act as its political and even military sur-
rogate in Jammu and Kashmir.”53 This state of affairs quickly became
the status quo in the region and has endured to the present day. Thus,
the Pakistanis’ limited control over their proxies did not affect the
favorability of the ceasefire agreement that ended the first Kashmir war,
but it did help them to create an advantageous territorial and political
arrangement in Kashmir after the conflict had ended.
CONCLUSION
In this chapter I showed that Pakistan emerged from partition in a perilous

condition, suffering from severe material and political weakness. These
material and political factors led to uncertainty regarding not only the
country’s viability but also its very reason for existence. Pakistani leaders
decided that the acquisition of the princely state of Kashmir could play an
important role in helping them to ameliorate both sets of shortcomings.
Given their tenuous material circumstances, however, a direct confronta-
tion between Pakistani and Indian forces for control of Kashmir would
have been exceedingly dangerous. They therefore devised a strategy that
employed rebels and militants to seize the territory, thereby limiting the
need for Pakistani military forces to intervene directly. Although the
Pakistanis did not succeed in taking Kashmir completely, they did man-
age to acquire roughly one-third of the territory. The Pakistanis benefited,
to varying degrees, from the three types of advantages associated with a
militant proxy strategy: They were able to launch the conflict at a rela-
tively low cost, may have constrained Indian retaliatory military options,
and used their limited control of their proxies to justify a continued pres-
ence in Kashmir.
Despite its inconclusive ending, the first Kashmir war had two impor-
tant results. First, it demonstrated that nonstate actors could enable
Pakistan to challenge India in a manner that limited the prospect of direct
military confrontation and catastrophic Pakistani defeat. The use of mili-
tants therefore constituted Pakistan’s only realistic means of attempting
50
to revise territorial boundaries in Kashmir. Second, the war enhanced

Kashmir’s importance to Pakistan, extending the dispute well past the
time of partition and transforming it into a contest of national resolve
with India. 54 Thus, even after the war had officially ended, Pakistani mili-
tary leaders sought to continue their militant campaign in Kashmir. I turn
to this subject in c hapter 4.
51
CHAP T ER 4
Pakistan’s Militant
Strategy Evolves
P akistan’s militant strategy during the first Kashmir war did not enable
it to capture the region entirely. The strategy was sufficiently success-
ful, however, that it became a significant component of Pakistani security
policy, growing in importance and sophistication with every subsequent
conflict. I turn to several of those subsequent conflicts in this chapter.
Specifically, I examine the evolution of Pakistan’s militant strategy
from the aftermath of the first Kashmir war through the 1965 Kashmir
war, the 1971 Bangladesh war, and the Afghan conflict of the 1980s. In
the 1965 Kashmir war, Pakistan moved from its previous use of informal
local militias to the employment of a well-trained and organized force of
religiously motivated guerrillas. The 1965 war did not succeed in achiev-
ing Pakistan’s broad strategic goal of capturing Kashmir. Operationally,
however, it was not a complete failure, and in fact it achieved a number
of successes. Moreover, the operation’s shortcomings could be attributed,
in large part, to poor planning and political misjudgments, rather than to
a flawed strategic concept. Thus, Pakistani planners remained sanguine
regarding their eventual ability to achieve strategic success through the
use of militant proxies.
During the Bangladesh war, Pakistan primarily used conventional
military forces against India. Nonetheless, the conflict increased the
importance of Pakistan’s militant strategy, for the Pakistanis’ crushing
loss clearly demonstrated the danger of engaging India in a conventional
military confrontation. It also led Pakistan to buttress its state-building
efforts by further promoting the country’s Islamic identity. The resulting
Islamization of Pakistan increased the importance of its security compe-
tition with India, particularly in Kashmir. These military and political
52
developments combined to strengthen the rationale for Pakistan’s adher-

ence to its militant proxy strategy.
Pakistani leaders consequently viewed the Soviet Union’s 1979 inva-
sion of Afghanistan not just as a danger, but also as an opportunity. They
hoped to protect Pakistan against Soviet aggression while improving
their ability to pursue long-standing strategic interests elsewhere in the
region. A militant proxy strategy could enable Pakistan to achieve both
of these ends. By exploiting the strategy’s operational benefits, Pakistan
could keep the Soviets at bay without undue cost or risk. And by acting as
the sole conduit for international assistance to anti-Soviet militant orga-
nizations, Pakistan could amass resources that could be redirected to the
pursuit of its primary national security goal: undermining Indian rule in
the territory of Kashmir.
THE 1965 KASHMIR WAR
The conclusion of the first Kashmir war in no way signaled the end of
Pakistan’s interest in Kashmir. The political and material incentives for
Pakistan to acquire the territory remained as strong as they had been in
the wake of partition. In addition, the Pakistanis’ efforts during the war
had been far from a complete failure. Pakistan had managed to seize one-
third of Kashmir and seemed likely to be able to keep its forces there
indefinitely. The results of the first Kashmir war had thus suggested that,
despite Pakistan’s relative weakness, the acquisition of Kashmir might be
within its reach—provided it employed an appropriate strategy that could
insulate it from the full cost and danger of direct military confrontation
with India.
Colonel Khan, who had masterminded Pakistan’s use of tribal mili-
tias in 1947, publicly championed the continuation of a militant-based
approach, arguing against efforts to seize Kashmir with Pakistani conven-
tional forces. Khan’s case for continuing to pursue a militant strategy in
Kashmir rested on two foundations. First, he believed that India, despite
Prime Minister Nehru’s promise and United Nations Security Council
Resolution 47, would not voluntarily hold a plebiscite to ratify its control
of Kashmir. India would agree to a plebiscite only if international pres-
sure became sufficiently severe that it had no other choice. This would
occur only if a crisis erupted in Kashmir, bringing the problem to the fore-
front of the world agenda. If such a crisis resulted in an Indian attack on
Pakistan, Khan believed, the likelihood of overwhelming international
pressure, including the possibility of direct involvement by outside par-
ties, would be even greater. Second, Khan was convinced that full-scale
53
Pa k is ta n ’ s M ili ta n t S t r at e g y E v o lv e s ( 53 )
Pakistani military efforts would not be necessary to create instability in

Kashmir and draw significant international attention to the dispute. This
goal could be achieved simply by using locally raised militants—what
Khan called a “people’s militia”—to conduct sabotage and foment unrest
in the region. According to Khan, several hundred fighters armed with
rudimentary weapons would be sufficient for the task.1
Khan managed to secure 1 million rupees from the Pakistani govern-
ment to support these operations. Bombings and explosions in Kashmir
during the 1950s suggest that his plans may have been carried out, though
little direct evidence exists in this regard. Even if the violence was the
work of Khan’s militias, it lacked the scale to loosen India’s grip on the
region. Such an outcome could not occur without a larger offensive effort,
combined with broad political and strategic changes in the region. By
1965, it appeared to Pakistani leaders that such a combination of factors
might be in the offing.2
In 1964, President Ayub Khan had established a Kashmir cell to pro-
vide ongoing assessments of developments in the region. This was a high-
level group, which included personnel such as the foreign secretary, the
defense secretary, the Intelligence Bureau director, and the army’s chief
of general staff and director general of military operations. The group
believed that, if supported by Pakistani military forces, Muslims in the
Kashmir Valley were likely to revolt spontaneously against Indian rule. 3
The cell’s belief was based primarily on recent unrest in the valley,
which stemmed from a controversy known as the Hazbatral Affair. In late
1963, a hair believed to be from the beard of the Prophet Muhammad
had gone missing from a Srinagar shrine. Kashmiri Muslims considered
the hair to be a relic of great religious significance, and by early 1964 the
region was in an uproar. Violence flared across Kashmir and soon took
a communal turn, forcing large numbers of Hindu refugees to leave the
valley for the relative safety of Jammu. A worried Prime Minister Nehru
responded by engineering the release from prison of the Kashmiri nation-
alist leader Sheikh Abdullah, who had been incarcerated for eleven years
on conspiracy charges. Relations between Kashmir and the Indian state
had been badly damaged, however. The Pakistanis sensed this and cal-
culated that Kashmiris would now be ready to revolt against Indian rule,
particularly if they had military support in doing so. Subsequent contro-
versy over legislative fiats resulting in the expansion of Indian federal
power in Kashmir in 1964 and 1965, which resulted in the 1965 rearrest
of Sheikh Abdullah, reinforced this belief among Pakistani leaders.4
In addition to these issues, the Pakistanis had other reasons to believe
that the time might be right for another attempt to seize Kashmir. First,
India appeared to be politically off balance. The Indian prime minister,
54
Jawaharlal Nehru, who had led India since independence, had died in
1964, following several years of declining political authority. This cre-
ated a vacuum of leadership that the Pakistanis might be able to exploit.
Second, India had responded to an early 1965 Pakistani military probe
in a disputed border area known as the Rann of Kutch in a limited man-
ner, choosing not to escalate the conflict and achieving only a return to
the status quo ante. The Indians had preferred a reversion to the status
quo, rather than escalation, largely because they believed that little was at
stake in the desolate border region. Their restraint, however, convinced
Pakistani leaders that India lacked the will for a fight with Pakistan.
Third, Pakistan had grown close to China in recent years. The Pakistanis
believed that India, fearing Chinese intervention, would limit any retalia-
tion against Pakistan for provocations in Kashmir. Fourth, the Pakistanis
feared that an Indian military build-up, launched in the wake of its disas-
trous 1962 war with China, would soon make Pakistani action against
Kashmir impossible. Finally, by the mid-1960s, the Pakistanis had con-
siderable experience with militant proxy warfare. They had used militants
extensively in the first Kashmir war, had the benefit of Colonel Khan’s
years of plans and activities during the 1950s, and had formally studied
guerrilla warfare at US military schools. 5 Thus, the likelihood of a suc-
cessful militant campaign in Kashmir seemed higher than it had been
previously.
Given the apparently propitious political and strategic environment,
Pakistani President Ayub Khan directed the army to develop options
for offensive action against Indian Kashmir. Central to the army’s plan-
ning process was the Kashmir cell’s assumption, subsequently supported
by the Inter-Services Intelligence Agency, that an attack would trigger
a Kashmiri revolt against Indian rule. There were sharp differences of
opinion over India’s likely reaction to such a move. A number of military
assessments, for example, had predicted that India would attack Pakistan
if the Pakistanis moved against Kashmir. This was the view that the direc-
tor of military operations conveyed through army general headquarters
to the government. Senior leaders in the foreign ministry, by contrast,
such as Foreign Minister Zulfikar Ali Bhutto, were convinced that while
the Indians might launch a military response in Kashmir, they would not
retaliate against Pakistan proper in the event of an attack. This group advo-
cated swift and decisive action, unlike the more cautious military. Both
camps, however, believed that the predicted uprising within the Kashmiri
population would happen if the Pakistanis intervened in Kashmir.6
President Ayub, after some initial indecision, became convinced that
the benefits of offensive action outweighed the risks and decided to move
against Indian Kashmir. The army was instructed to devise an operation
55
that avoided overly antagonizing the Indians, however, as this could

lead to large-scale war. Consequently, “Do not provoke, do not escalate”
became the military’s watchword during the planning process. Militant
forces provided the most useful tool to enable the Pakistanis to achieve
their diverse aims. Militants could agitate the Kashmiris and trigger
the expected rebellion, while minimizing evidence of direct Pakistani
involvement in the operation, thereby lowering the likelihood of full-scale
Indian retaliation.7
The initial Pakistani offensive, code-named Gibraltar, relied almost
entirely on militant forces. It did so in a manner that was considerably
more sophisticated than Pakistan’s use of militants in 1947. For example,
the guerillas of Operation Gibraltar were far abler than the motley array of
tribal forces that Pakistan had hastily deployed during the first Kashmir
conflict. Unlike the tribesmen, the Gibraltar fighters were local; they
were drawn primarily from Razakar militias, which had been raised in
Pakistani Kashmir in 1962, and were supplemented with personnel from
Mujahid militias, who served primarily as porters. The Gibraltar fight-
ers were also better organized, led, and trained than the tribal militias of
1947. The roughly thirty thousand militants were divided into approxi-
mately ten forces, each of which consisted of five companies. A Pakistani
officer led each of the forces. Pakistani officers, junior-commissioned offi-
cers, and noncommissioned officers occupied key leadership positions in
the companies as well and were supported by other specialized personnel
from units such as the Special Service Group. In addition to basic mili-
tary training at various sites in Pakistani Kashmir, the fighters received
six weeks of instruction in guerrilla warfare techniques, with emphasis on
skills such as unarmed combat, laying ambushes, destroying bridges and
other lines of communication, and conducting raids. They were equipped
with radios, rockets, machine guns, and explosives. The ten forces divided
twenty regions of Kashmir between them, with each force assigned
between one and three areas of responsibility. The operation came under
the overall command of the Pakistan Army’s 12th Division and Major
General Akhtar Malik.8
In addition to this superior selection, training, leadership, and orga-
nization of personnel, Pakistan’s plans were better grounded in Islamic
religious tropes in 1965 than they had been during the 1947 war. The
operation’s code name commemorated Gibraltar’s eighth-century con-
quest by the Muslim general Tariq bin Ziyad. In addition, the names
of the invasion forces, such as Tariq, Qasim, Ghaznavi, Murtaza, Jacob,
Salahuddin, Khalid, and Babar, celebrated prominent Muslim military
heroes. Thus, from its inception, the operation was explicitly religious
in tone.9
56
Despite these strengths, the Pakistani plan suffered from a number

of important shortcomings. For example, the Pakistanis had only two
months to train the guerillas who would infiltrate into Indian Kashmir.
The equipment that Pakistan provided them was better suited to a con-
ventional military than to guerilla forces. The Pakistanis did not consult
leaders in Indian Kashmir to assess their likely reaction to a militant
infiltration. Finally, the Pakistanis lacked the optimal number of troops
needed to hold off the Indians in case they responded to the guerilla infil-
tration with an attack.10 Nonetheless, Pakistan’s plan constituted a con-
siderable improvement over its operations during the first Kashmir war,
and Pakistani leaders remained confident that it would have the desired
effects within Kashmir.
The Pakistanis began moving Gibraltar forces into Kashmir over a
four-day period in late July 1965. The infiltrators planned to join crowds
celebrating a religious festival on August 8 and then to proceed to Srinagar
on August 9, as part of a group commemorating the arrest of the Kashmiri
leader Sheikh Abdullah. They would then seize the airfield and radio sta-
tion, establish a temporary government, and call upon Pakistan for assis-
tance. Meanwhile, the Kashmiri population, emboldened by the presence
of the infiltrators and by a radio-borne propaganda campaign run from
Pakistan, would rise up spontaneously in rebellion against their Indian
overlords. The Pakistan Army would then step in with a supplementary
conventional military operation, code-named Operation Grand Slam.
Grand Slam would seek to cut Kashmir off from India proper through the
capture of the towns of Chamb, Akhnur, and Rajauri, thereby enabling
the Pakistanis to complete their annexation of the territory without
Indian interference.11
The initial phases of Operation Gibraltar were successful. Approxi
mately fifteen hundred fighters infiltrated Indian Kashmir during the first
two weeks of August 1965. A second group of roughly six thousand men
followed them in the third week of August.12 Once these forces arrived
in Indian Kashmir, however, the Pakistanis discovered that the opera-
tion was not going to unfold as they had planned. The central problem
was that the expected Kashmiri uprising did not occur. Despite signifi-
cant Kashmiri discontent with Indian rule, most Kashmiris were not yet
prepared to seek violent political change. Not only did the Kashmiris fail
to rebel against the Indians, but also they alerted Indian authorities to
the infiltration and turned Pakistani fighters over to them. In response,
the Indian army sealed the Line of Control, closing the segments through
which the intruders had passed, and attacked a number of posts in Pakistani
Kashmir. Pakistan, having lost the element of surprise, was now unable
to move all of its militant forces into Kashmir. Approximately fifty-five
57
hundred more fighters waiting to enter Indian Kashmir remained stuck

on the Pakistani side of the territory. In the absence of Kashmiri coop-
eration, those militants who did manage to infiltrate into Indian Kashmir
were of limited use.13
With Gibraltar’s impending failure, Pakistani field commanders
sought to launch Operation Grand Slam quickly, to relieve pressure on
their troops defending Kashmir. They were not able to secure permission
from higher headquarters to do so, however, until September 1, 1965.
Then, after a series of skirmishes along the Ceasefire Line, Pakistani
forces attacked Indian territory in the Bhimbar-Chhamb area of Southern
Kashmir. The Pakistanis subsequently advanced quickly toward Akhnur.
India responded with large-scale horizontal escalation on September
6, driving forces toward Lahore and Sialkot in Pakistan proper, leading
Pakistan to abandon Akhnur. India’s advance on Lahore stalled when
its forces reached the irrigation canal just outside the city. A number of
inconclusive battles between Indian and Pakistani forces followed. By
mid-September, the war had ground to a stalemate, and the adversaries
accepted a UN ceasefire resolution soon thereafter. By the third week of
September, the 1965 conflict was over. Under the postwar settlement,
known as the Tashkent Agreement, the adversaries agreed to return to
the status quo ante.14
EVALUATING PAKISTAN’S 1965 MILITANT

STRATEGY
Many of the themes that I discussed earlier were evident in Pakistan’s

use of Islamist militants during the 1965 war. First, the war was part of
a trend in which Pakistan, over time, used militants in an increasingly
sophisticated manner. Despite numerous problems, Pakistan’s employ-
ment of militants in 1965 was far better planned and executed than it had
been in 1947. Pakistani officers had been formally studying insurgency-
related curricula in US military schools. The militants themselves were
more carefully selected, received superior training and equipment, and
were better led than the tribesmen who had composed the invasion force
during the first Kashmir war. In addition, Pakistan more fully grounded
its militant strategy in religious themes than it had in 1947.
Second, Pakistan’s militant strategy realized a number of operational
advantages. These advantages were mainly in the area of denial.15 The
Pakistanis thwarted India’s efforts to prevent penetration of its territory.
Pakistan was able to infiltrate a sizeable force into Indian Kashmir before
triggering an Indian conventional military response.16
58
In this sense, the 1965 war was more successful for Pakistan than was
the 1948 conflict. During the first Kashmir war, the attacking militants
were intercepted by Indian conventional military forces and were turned
back before they could reach their destination of Srinagar. In 1965, by
contrast, a sizeable militant force reached its destination within Indian
Kashmir before India reacted. Indeed, the Indians were surprised to
discover that such a large group of militants had entered their territory
without their knowledge. As the Indian Ministry of Defence’s history of
the war puts it, “The invasion of Gibraltar Force was one of the best kept
secrets of Pakistan.” Although Indian leaders had expected some militant
activity in Kashmir, “no one had visualized a well-coordinated operation
on such a massive scale. Even on 2 August 1965, when a high-level confer-
ence was being held at Srinagar to review the security arrangements along
the Cease-fi re Line, India had no inkling of the guerrilla invasion knock-
ing at her doors.”17
In addition, the infiltrators experienced some tactical successes once
they were inside Kashmir, destroying a number of military posts, as well
as tying down and inflicting casualties on Indian forces. Thus, in addi-
tion to failing to prevent infiltration, the Indians were unable to prevent
the infiltrators from launching operations once they were in Indian terri-
tory. It is worth noting that neither of these Pakistani successes required
Pakistan to pay the costs of a direct conventional fight against India;
they occurred while the conflict remained at the subconventional level.
Full-scale Indo-Pakistani conventional war erupted only after Pakistan
had launched Operation Grand Slam by attacking Chamb with regular
Pakistan Army forces. This conventional conflict also was less costly to
Pakistan than it would have been without the preceding militant cam-
paign, for the militants succeeded in absorbing the energies of four Indian
divisions in Kashmir during the month of September. This had a sig-
nificant impact on the outcome of the conventional fight. As the Indian
Ministry of Defence’s history of the war acknowledges, “Had these divi-
sions been available to fight the Pakistanis in Punjab, the result of the
Indo-Pak War of 1965 would have been different.18
Of course, despite their two initial denial failures, the Indians did suc-
ceed in defeating the Pakistani operation on Indian territory after the
operation had begun. As a result, Gibraltar did not achieve its broad strate-
gic objective of wresting Kashmir from India. To a large degree, however,
this failure resulted from political misjudgment and negligent planning,
rather than any problem inherent in the use of militants. For example, the
Pakistanis simply assumed that the Kashmiris would rise up against India
when they were given the opportunity to do so. Senior Pakistani military
and political decision makers do not appear seriously to have questioned
59
these beliefs. As a result, the Pakistanis did not bother to take basic pre-
paratory steps, such as making contact with local Kashmiri leaders prior
to launching Gibraltar. The Pakistanis were then surprised when the
Kashmiris not only refused to rebel but also turned the Gibraltar infiltra-
tors over to Indian authorities.19
In addition, the Pakistani officers leading the Gibraltar forces lacked
essential local skills and knowledge. For example, Pakistani officers did
not speak Kashmiri. This significantly impeded their effectiveness, given
that their mission was not simply to prevail on a conventional battlefield,
but rather to foment a rebellion inside Kashmir. 20 Such a mission would
require sophisticated communication skills—skills the Pakistani mili-
tary leadership had failed to ensure that they possessed.21
Finally, much of the Pakistani security establishment, including senior
Pakistan Army officers, was not informed of Operation Gibraltar until
after its launch. As a result, the military establishment did not take full
responsibility for the effort. They tended to view it, rather, as a hybrid
operation, born of an unsavory alliance between the foreign minister,
Zulfikar Ali Bhutto, and the leadership of the 12th Division. This impeded
planning for Gibraltar, as well as the provision of support for the opera-
tion once it was underway.22
I do not seek in any way to downplay Operation Gibraltar’s ultimate
failure. My point is simply that, although the project did not realize its
overall aims, it enabled the Pakistanis to tell themselves a hopeful story
about their militant proxy strategy. The strategy had achieved a number of
operational successes. In addition, Gibraltar’s shortcomings had resulted
largely from flawed planning and execution, which in turn stemmed from
political misjudgments; there was little evidence that these problems were
the result of a proxy strategy per se. Indeed, the outcome of Operation
Gibraltar suggested that, with a bit more foresight, the use of militant
proxies might enable Pakistan to achieve strategic success in the future.
This Pakistani hope was soon to be reinforced by a bitter lesson—when
the Pakistanis deviated from their militant strategy, as they did during the
Bangladesh war, the result was unmitigated disaster.
THE BANGLADESH WAR
The Bangladesh war occurred a mere six years after the 1965 conflict.
Although it was primarily a conventional military confrontation, 23 it had
profound implications for Pakistan’s militant strategy. The war’s cata-
strophic outcome exacerbated the material and political weakness that
had originally given rise to Pakistan’s militant policy and demonstrated
60
the danger of engaging India directly in combat. Thus, the conventional

nature of the Bangladesh war actually increased Pakistan’s future reliance
on Islamist militancy as a strategic asset.
The Bangladesh war erupted out of what was initially an internal
Pakistani political dispute. This dispute was rooted in long-standing ani-
mosity between predominantly Bengali East Pakistan and predominantly
Punjabi West Pakistan. Simply put, East Pakistanis viewed themselves as
the victims of domination and marginalization at the hands of their West
Pakistani brethren. The West Pakistanis, they believed, had created politi-
cal, legal, and economic structures so favorable to West Pakistan that
they composed a system of de facto internal colonialism. Specific East
Pakistani grievances included the establishment of Urdu as Pakistan’s
national language; underrepresentation of East Pakistanis in national
institutions such as the army, civil service, and legislature; and constitu-
tional arrangements that favored West Pakistani electoral interests. West
Pakistanis, for their part, tended to view their eastern compatriots as sym-
pathetic toward India, uninterested in Pakistani nationalist causes such
as Kashmir, and amenable to forcible rule by West Pakistan.24
In 1966, following the conclusion of the second Kashmir war, Sheikh
Mujibur Rahman, head of an East Pakistani political party called the
Awami League, promulgated a set of six demands. They included consti-
tutional reform that would increase East Pakistani electoral power, the
right to establish an East Pakistani currency, the right to formulate sepa-
rate East Pakistani fiscal policies, and the right to establish East Pakistani
paramilitary or militia forces.25 Although billed merely as steps toward
East Pakistani autonomy within a larger Pakistani federal framework,
many West Pakistanis saw Sheikh Mujib’s demands as bordering on seces-
sionism. Nonetheless, the Awami League won a plurality in Pakistan’s
first national elections in 1970, securing 38 percent of the overall vote and
defeating the second-place Pakistan People’s Party (PPP) by 18 percent-
age points.26
Given these results, the Awami League should have had the right to
form a national government. PPP leader Zulfikar Ali Bhutto, however,
sought to prevent it from doing so. He argued that the Awami League’s
mandate was confined to East Pakistan and that the PPP was the legiti-
mate representative of West Pakistanis. The league, Bhutto maintained,
would have to enter a power-sharing arrangement with the Pakistan
People’s Party if it wished to govern at the national level. A long series of
negotiations involving Bhutto, Rahman, and Pakistani President General
Yahya Khan ensued but failed to resolve the impasse. Bhutto remained
adamant that the Pakistan People’s Party must be allowed to form a gov-
erning coalition with the Awami League. Rahman, by contrast, held that
61
given the league’s electoral victory, it did not need to share power with the
PPP. When, in early March, the Pakistani government decided to post-
pone the opening session of the National Assembly in Dhaka, large-scale
rioting erupted in East Pakistan.27
Yahya Khan ordered the Pakistan Army to quell the uprising. It did so
in brutal fashion, massacring tens of thousands of East Pakistanis. The
military operation, called Searchlight, particularly targeted teachers,
journalists, medical doctors, other members of the intellectual and pro-
fessional classes, and students. In addition to regular military forces, the
Pakistanis also used nonstate groups such as the al Shams and al Badr
brigades to assist in these efforts.28
The US consul general in Dhaka, Archer Blood, and the US ambas-
sador to India, Kenneth Keating, were so horrified by the violence that
they labeled it “selective genocide” in cables to Secretary of State William
Rodgers. Blood went on to describe the violence as a “reign of terror by
the Pak military” to which “we here in Dacca are mute and horrified wit-
nesses.”29 Soon, refugees seeking to escape the carnage began to flow into
India. By late spring, roughly ten million East Pakistanis had crossed the
border. This created a severe humanitarian challenge as India struggled
to provide appropriate care to the refugees, many of whom were sick,
wounded, and malnourished. The influx also created political difficul-
ties for India, for although most of the refugees were Hindus, whom the
Pakistani Army had deliberately targeted and ejected from East Pakistan,
large numbers of Muslims had fled the violence as well. Their presence
changed the demographic balance of Indian border states such as West
Bengal. It also triggered severe economic dislocations in these areas, lead-
ing to tensions between the refugees and local Indian populations. 30
Unable to absorb the refugee flow, Indian leaders decided to resolve
the situation by force, splitting East Pakistan off from the rest of the coun-
try. In addition to resolving the ongoing humanitarian crisis, severing
Pakistan’s eastern wing would provide India with strategic benefits. First,
it would end India’s two-f ront war problem. If East Pakistan became an
independent state, India would no longer automatically have to contend
with an enemy on both its eastern and western flanks every time it fought
with Pakistan. Second, vivisecting Pakistan would enhance India’s over-
all regional status, firmly establishing it as the premier power in South
Asia. 31
India pursued both subconventional and conventional military strat-
egies against the Pakistanis. At the subconventional level, the Indians
helped to organize and train East Pakistani refugees, creating a guerrilla
force known as the Mukti Bahini. The Mukti Bahini engaged in a range
of activities, such as conducting hit-and-r un attacks on Pakistani forces,
62
collecting intelligence, and sabotaging targets in East Pakistan. The

guerrillas thus constituted an important force-multiplier for India. They
attrited Pakistani resources and wore down Pakistani forces in advance
of India’s conventional military attack on East Pakistan, and supported
Indian operations during the attack. They also lent Indian efforts to pro-
mote East Pakistani independence an air of legitimacy. Despite their util-
ity, however, the Mukti Bahini would not be decisive in achieving Indian
strategic goals. The guerillas were lightly armed, generally operated in
small groups, and varied widely in their suitability for military training
and small-unit operations. In addition, they were riven by internal politi-
cal and social differences. The Indians did not wish simply to harass the
Pakistanis, inflicting modest costs on them as the civil war and refugee
crisis slowly played themselves out. Rather, the Indians were seeking
maximalist goals: to defeat the Pakistan army, to sever East from West
Pakistan, and to bring a quick end to the refugee problem. The surest way
of achieving these ends would be through the large-scale use of conven-
tional military power.32
Such an approach would be particularly promising because India
enjoyed a significant conventional military advantage over Pakistan.
Overall, the Indian Army was roughly twice the size of Pakistan’s, the
Indian Air Force had about double the number of combat aircraft as did
Pakistan, and the Indian Navy possessed approximately three times as
many warships as the Pakistan Navy. Although this overall balance mat-
tered, the ratio of forces in the eastern sector was particularly significant.
Here the Indians enjoyed an even more marked advantage. India pos-
sessed three corps against the Pakistanis’ one, eight infantry and moun-
tain divisions against the Pakistanis’ four, twenty-three infantry brigades
against the Pakistanis’ thirteen, seventy-t wo infantry battalions against
the Pakistanis’ thirty-four, and one parachute brigade against none on
the Pakistani side. These forces benefited not just from numbers, but also
from superior mobility, control of the air and sea, solid logistics, and a
friendly local population in East Pakistan. Pakistani troops, by contrast,
were isolated and poorly supplied, faced a hostile population, and were
badly fatigued after months of suppressing the East Pakistani rebellion.
Although unconventional forces such as the Mukti Bahini could play an
important role in supporting Indian military efforts, then, it was India’s
advantage in conventional forces that would prove to be decisive. Indian
commanders understood the implications of this lopsided conventional
balance and contemplated the impending conflict with considerable
confidence. Pakistani military leaders, for their part, recognized that the
conventional imbalance would likely present them with an insurmount-
able handicap and hoped to avoid a direct fight with the Indians. As the
63
Pakistan Army’s chief of staff, General Abdul Hamid Khan, said to a State
Department official in August 1971, Pakistani forces were “far inferior to
India’s numerically and logistically.” Therefore “no senior Pak military
commander would want to take on war with India.”33
Senior Indian military officers had informed their government shortly
after the onset of the crisis that any conventional military attack on East
Pakistan would have to wait until the monsoon rains had ended in the fall
of 1971. In the meantime, they consolidated their forces, many of which
were dispersed on operations in other parts of the country, and finalized
their war plans. The Indians prepared to launch a blitzkrieg against East
Pakistan. The operation would rapidly drive armor, mechanized infantry,
and heliborne forces, backed by close air support, deep into East Pakistani
territory on three fronts, without pausing for set-piece battles along the
way. If successful, this deep penetration would cripple Pakistani command
and control, bringing Pakistani military operations to a halt and enabling
the Indians to achieve a quick victory. The operation would thus resemble
tank warfare in Europe or the Middle East far more than the militant-
centric conflict of recent decades in Kashmir. 34 While the military was
completing these preparatory tasks, Indian political leaders traveled the
world to make their case against Pakistan and garner international sup-
port for the upcoming attack. This included a three-week tour in October
and November 1971 by Prime Minister Gandhi, in which she visited the
United Kingdom, France, West Germany, and the United States. The trip
culminated in an acrimonious meeting in Washington with President
Nixon, during which the two leaders essentially talked past one another.
Nixon sternly warned the prime minister against launching a potentially
disastrous war against Pakistan, while Gandhi decried Pakistani atroci-
ties, emphasized the enormous costs of the refugee crisis, and stated that
East Pakistani independence was inevitable. 35
The Indians’ primary motive for undertaking this round of interna-
tional diplomacy was concern regarding the appearance of the upcom-
ing military campaign. They worried that attacking East Pakistan would
make them seem to be belligerent in the eyes of the international commu-
nity, and they sought to minimize this problem by consulting with world
leaders first. As it happened, however, the Indians had little to fear. On
December 3, 1971, Pakistan launched preemptive strikes against Indian
air bases in Rajasthan, Kashmir, and the Punjab, providing the Indians
with a clear justification for commencing large-scale offensive operations
against East Pakistan. The Pakistani strikes were largely ineffective and
managed only to incapacitate the airfield at Amritsar and destroy a radar
site. India responded quickly, striking air installations in West Pakistan
and attacking the port of Karachi. Indian air operations ultimately
64
succeeded in destroying nearly half of Pakistani aircraft, while the Indian

navy kept Pakistan’s fleet largely confined to port for the rest of the con-
flict. 36 The Pakistani attack thus had little military impact. It made the
Indians’ diplomatic position easier, however, by giving them a justifica-
tion for launching their offensive against East Pakistan.
The Indian army commenced its assault on East Pakistan, code-named
Operation Jackpot, on December 5. The Indians’ objective was the East
Pakistani capital of Dhaka, and they drove toward it from four main
directions—2 Corps from the West, 33 Corps from the Northwest, 4
Corps from the East, and the 101 Communications Zone from the North.
The Indians emphasized speed and penetration, bypassing large enemy
concentrations, encircling and attacking opposing formations from the
rear, cutting lines of communication, and sowing fear and disorienta-
tion among the Pakistanis. Pakistani resistance quickly crumbled and by
December 8 Indian forces from the North and East were on the outskirts
of Dhaka. 37
Recognizing the gravity of their situation, East Pakistani Governor
A. M. Malik along with his military aide, General Rao Farman Ali Khan,
and the commander of the Pakistan Army’s Eastern Command, General
Amir Abdullah Khan Niazi, attempted to avoid complete disaster by
offering India a conditional ceasefire. The ceasefire would have allowed
the creation of an elected civilian government in East Pakistan, but with-
out the surrender of Pakistani forces, who would have been given safe
passage to West Pakistan. The offer was both rejected by the Indians
and condemned as illegitimate by Pakistani President Yahya Khan. By
December 14, however, even Yahya recognized that the continuation of
hostilities was pointless. As he wrote to Niazi, “You have now reached a
stage where further resistance is no longer humanly possible, nor will it
serve any useful purpose… . You should now take all necessary measures
to stop the fighting.” On December 16 in Dhaka, Niazi surrendered to
Indian Lieutenant General Jagjit Singh Aurora. India declared a unilateral
ceasefire the next day, bringing the Bangladesh war to a close. 38
EVALUATING THE BANGLADESH WAR
The Bangladesh war was primarily a conventional conflict, particu-

larly from Pakistan’s perspective. Unlike in previous Indo-Pakistani
conflicts, Pakistan did not use militants extensively against India dur-
ing the war. India, for its part, did employ militant forces against the
Pakistanis, but only in a supporting role. The war was ultimately decided
not by Mukti Bahini guerilla operations, but rather by rapid, large-scale
65
thrusts of conventional military forces deep into East Pakistani territory.

Nonetheless, despite its largely conventional nature, the Bangladesh war
increased the importance of Pakistan’s militant proxy strategy.
The war provided concrete evidence of what could happen in the event
of a full-scale conventional confrontation between India and Pakistan. In
previous conflicts, Pakistan had been able to fight India to a standstill,
primarily because it had prevented the Indians from fully exploiting their
conventional military advantage. Militant proxy forces had helped to do
this by enabling Pakistan to delay direct combat against India, by thwart-
ing Indian denial efforts, by absorbing losses in place of regular Pakistani
forces, and by consuming Indian resources. Thus, although these ear-
lier conflicts had eventually evolved into conventional military contests
between India and Pakistan, the Pakistanis had not been forced to pay the
costs of full-scale conventional confrontation with the Indians.
Although these costs could presumably have been catastrophic, no
concrete proof of the magnitude of the danger had previously existed; the
risks had been entirely hypothetical. Now, however, the potential dan-
gers of full-scale conventional Indo-Pakistani conflict had become abun-
dantly clear. In such a war, Pakistan could be torn asunder. Moreover, if
such a catastrophe were to recur, it would necessarily happen in the West
Pakistani heartland, for with the creation of independent Bangladesh,
no future war could be fought in Pakistan’s far-off former East Wing.
These potential costs of conventional Indo-Pakistani conflict, which the
Bangladesh war made clear, rendered the use of militant proxies even
more essential to Pakistan than it had been previously.
The Bangladesh war also increased the importance of Pakistan’s mili-
tant strategy by heightening Pakistan’s long-standing insecurities. It did
so at both the material and the ideological levels. At the material level, the
war’s outcome was nothing short of disastrous. In the space of approxi-
mately two weeks, the Indians had split Pakistan in two, creating the new
state of Bangladesh. In addition to this, the Indians had seized approxi-
mately 5,000 square kilometers of Pakistani territory and taken roughly
ninety thousand Pakistani prisoners. These losses severely weakened the
Pakistan military. They also limited Pakistan’s strategic options. With
its eastern wing removed, Pakistan would no longer be able to employ a
two-f ront strategy against India, which could have forced the Indians to
divide their resources in the event of an Indo-Pakistani conflict. Instead,
if war occurred, India could now concentrate all of its firepower against
the west. Finally, at the end of the Bangladesh war, the Pakistanis found
themselves without reliable allies. Pakistani leaders had hoped that in the
event of hostilities with India they would receive international help, either
from the United States or from China, which the Pakistanis considered an
66
all-weather friend. When such assistance was not immediately forthcom-

ing, Pakistani leaders continued to expect that it would arrive, right up
to the end of the war. But help never came. Both the United States and
China had more urgent priorities than saving Pakistan from defeat at the
hands of the Indians. The Bangladesh war thus made clear that, despite
Pakistan’s careful cultivation of relationships with the United States and
China, the Pakistanis would be able to rely only on themselves to provide
for their future security. 39
Such loss and isolation left Pakistan even more vulnerable to India
than it had been previously.40 Despite this increased vulnerability, how-
ever, Pakistani leaders did not change their core strategic priorities in
the wake of the Bangladesh conflict. As President Bhutto quickly made
clear, Pakistan remained committed not simply to defending itself against
India, but also to continuing to pursue a revisionist agenda in Kashmir. In
June and July 1972, Bhutto finalized the terms ending the Bangladesh war
during meetings with Prime Minister Gandhi at the Indian hill station
of Simla. As he left Pakistan for the meetings, Bhutto stated that, follow-
ing the Bangladesh conflict, Pakistan would have to decide its position
regarding “the right of self-determination of the people of Kashmir. Do
we still uphold this principle or do we, after three wars, need some kind of
settlement?” The answer, in Bhutto’s view, was clear. “As far as the ques-
tion of Kashmir is concerned, there can be no compromise on the issue of
the right of self-determination because this right belongs to the people of
Kashmir.”41
Bhutto and Gandhi’s meetings produced a document known as the
Simla Agreement. In addition to beginning the process of normalizing
Indo-Pakistani diplomatic relations, the agreement committed the two
countries to respecting the existing territorial division of Kashmir, to
resolving their differences bilaterally, and to refraining from the threat
or use of force in their dealings with each other.42 Upon his return to
Pakistan, however, Bhutto maintained that the Simla Agreement’s lan-
guage left open the possibility that Pakistan could continue to pursue its
interests in Kashmiri self-determination in multilateral fora such as the
United Nations. In addition, Bhutto made plain that should such diplo-
matic efforts fail, the resort to war remained a real possibility for Pakistan.
Appealing to the United Nations, Bhutto said, had never helped Pakistan
to further the Kashmir cause. Moreover, he maintained that although
“the Simla agreement goes to the extent of saying that the two sides will
refrain from the threat or use of force … to refrain from the threat of force
or use of force is not a no-war pact.” If “people think that this is a no-war
pact, they are sadly mistaken. The phraseology of a no-war pact is entirely
different from the phrase ‘refrain from the threat or use of force.’ ”43 As
67
Bhutto told the Pakistani National Assembly, if “tomorrow the people of

Kashmir start a freedom movement … we will be with them… . This
is an eternal position.”44 Thus, Stanley Wolpert explains, Bhutto believed
the outcome of the Bangladesh war to be “irrelevant as far as Kashmir
was concerned … no matter what was written on the piece of paper he
may have signed” at Simla. Despite Pakistan’s apparent commitments to
peaceful dispute resolution and respect for the status quo, Bhutto “never
doubted that Pakistan would rise again some day to reclaim Kashmir.”45
The Bangladesh war, then, left Pakistan acutely vulnerable at the material
level but nonetheless committed to continuing to undermine existing ter-
ritorial arrangements in Kashmir. As a result, Pakistan found itself in even
greater need than before of a strategy that would enable it to challenge
India without risking catastrophic defeat.
At the ideological level, the Bangladesh conflict badly undermined the
two-nation theory that had justified Pakistan’s founding. The theory held
that a separate Muslim political entity was necessary if Muslims were to
enjoy their religious and cultural rights in a Hindu-majority South Asia.
In this separate entity, Muslims would live together, bound by a common
religion and heritage, despite divergent characteristics such as ethnicity
or language. In practice, however, such a coherent Muslim political entity
had not been viable. A common religious heritage had failed to unite East
and West Pakistanis in the face of diverse political and economic inter-
ests that resulted from ethnic and linguistic differences. Indeed, these
differences had torn East and West Pakistani Muslims apart in horrific
fashion.46
Given this failure, Pakistani leaders faced a choice. They could have
decided that religion had proven to be an unreliable source of political
unity and thus sought to de-emphasize it in their public discussions and
interpretations of Pakistan’s founding narrative. Alternatively, they could
have decided that religion was in fact capable of generating political unity
for Pakistan, but that past leaders had not properly deployed it as a state-
building tool. Consequently, to compensate for prior failings that had
culminated in the fiasco of the Bangladesh war, religion would have to
play an increased role in Pakistan’s state-building project in the future.
Pakistani leaders chose the latter course of action.
Pakistan therefore pursued a process of “Islamicizing” the central
elements of its society. Although this process is usually associated with
General Zia-ul-Haq, it actually had begun earlier, under Zulfikar Ali
Bhutto. Bhutto had employed pious measures, such as the banning of
alcohol, in an effort to insulate himself from criticism of his unsuc-
cessful economic policies. General Zia then took the policy in a much
more aggressive direction, primarily in an effort to generate unity and
68
strengthen Pakistan’s shaky political foundations after Bangladesh. It

included the creation of shariat courts, the appointment of a Muslim con-
sultative assembly, the implementation of charity taxes, the introduction
of punishments based on the Quran and Sunnah, and the expansion of the
madrassa system of Islamic education. Schoolbooks were edited to con-
form to religious teachings and civil servants were required to perform
daily prayers and were graded on religious knowledge and piety.47
The Pakistan Army was the most powerful and important institution
in Pakistan and therefore became a primary object of these Islamizing
efforts.48 General Zia believed that “the professional soldier in a Muslim
army, pursuing the goals of a Muslim state, CANNOT become ‘profes-
sional’ if in all his activities he does not take on ‘the colour of Allah.’ ”49
Zia’s efforts to Islamicize the military were not limited to the promotion
of religion within the ranks. He also wished to ensure that civilians sup-
ported the military’s religious orientation. “The non-military citizen of
a Muslim state,” Zia wrote, “must likewise be aware of the kind of sol-
ider that his country must produce and the ONLY pattern of war that
his country’s armed forces may wage.”50 Islamic teaching thus acquired
an important role in military education. The army’s Command and Staff
College, for example, established a Directorate of Religious Instruction.
Officers’ piety became a factor in their career prospects. Islamic teachings
were included in promotion exams, and officers deemed insufficiently
religious often failed to advance. 51
The effect of these post-Bangladesh ideological developments was to
increase Pakistan’s reliance on militancy, for they strengthened Pakistan’s
identity as a Muslim state founded in opposition to “Hindu” India. The
strengthening of Pakistan’s oppositional Muslim identity, in turn,
increased the importance to Pakistan of a favorable disposition of the
Kashmir dispute. As a result, a military strategy that could enable Pakistan
to seize Kashmir without directly confronting India and risking a cata-
strophic, Bangladesh-scale defeat—namely, the use of militant proxies—
assumed even greater significance. As Lawrence Ziring puts it, in the
wake of the Bangladesh war, Kashmir would become nothing less than “a
fetish of national identity for Pakistan,” with “Pakistan’s raison d’être …
intertwined with the jihad to liberate it from Indian non-believers.”52
It is important to note that Pakistan’s loss of the Bangladesh war also
created strong incentives for it to develop a wholly separate military capa-
bility to insulate it from India’s conventional superiority. Prime Minister
Bhutto had long believed that Pakistan should acquire nuclear weapons,
declaring in 1966 that Pakistanis would “make the bomb” even if they had
to “eat grass.”53 Now, in the wake of its catastrophic loss to India, Pakistan’s
need for a nuclear capability was especially clear; Pakistan had to be
69
able to ensure that a calamity like Bangladesh would never occur again.
Nuclear weapons, with their vast destructive potential, would enable it to
do exactly that, preventing India from leveraging the full extent of its con-
ventional military advantage and potentially threatening Pakistan with
catastrophic defeat. Pakistan thus launched a full-fledged nuclear weap-
ons program in 1972, in response to the debacle of the Bangladesh war. 54
Significantly, nuclear weapons would not only provide Pakistan with
a robust defense but also enhance its offensive capability. Pakistan could
now vigorously challenge the South Asian status quo, insulated from the
dangers of conventional retaliation by the threat of a Pakistani nuclear
response. This would ensure that, despite the military vulnerability that
the Bangladesh war had laid bare, Pakistan could pursue its most cher-
ished national goals, even more aggressively than before, in relative
safety. 55
A symbiotic relationship thus developed between nuclear weapons
and militancy in Pakistani security policy. After Bangladesh, Pakistan
would exclusively employ militant proxies in its efforts to undermine
the regional status quo. Nuclear weapons insulated Pakistan against the
possibility of all-out Indian retaliation in response to its militant provo-
cations. This insulation, in turn, enabled Pakistan to employ militancy
and challenge the status quo even more aggressively than it otherwise
would have. Pakistan’s militant strategy, for its part, enhanced nuclear
weapons’ relevance, giving them a purpose beyond simply preserving the
status quo, and enabling them to play an important role in underwriting
Pakistan’s offensive efforts. As India found, such offensive efforts, when
backed by nuclear deterrence, were very difficult to combat.
Despite being a primarily conventional conflict, then, the Bangladesh
war played an extremely important role in the evolution of Pakistan’s mili-
tant proxy strategy. Bangladesh was the exception that proved the rule
regarding Pakistan’s use of militancy. By confronting India in a conven-
tional conflict, Pakistan fell prey to the danger that had long haunted it
in its confrontations with India—that India would leverage its conven-
tional military superiority to destroy or dismember Pakistan. Bangladesh
showed that this danger was real and not merely hypothetical. In doing
so, it reinforced the need to adhere to a militant proxy strategy and avoid
direct conventional military confrontations with India. By revealing the
weakness of the two-nation theory and undermining Pakistan’s politi-
cal foundations, Bangladesh also made the Indo-Pakistani dispute over
Kashmir even more important to Pakistan than it had been previously.
This, in turn, made the militant strategy that could enable Pakistan
to achieve success in Kashmir more significant as well. In addition,
Bangladesh spurred Pakistan’s nuclear weapons program, which made
70
militant-based offensive action safer. Pakistan could thus pursue its revi-
sionist agenda through the use of militant proxies more energetically than
it could have in a nonnuclear environment. Nuclear weapons and mili-
tancy therefore became deeply intertwined. Together, they would make
up two of the three main legs of Pakistani grand strategy.
The Pakistanis would not repeat the mistake of Bangladesh; Pakistan
never again confronted India in a direct conventional military conflict.
Rather, the Pakistanis would apply the lessons that they learned during
the Bangladesh war in future conflicts. The next opportunity that pre-
sented itself would be the Soviet Union’s invasion of Afghanistan.
THE AFGHAN WAR
As the Pakistanis’ militant campaign to acquire Kashmir became

more important in the years following the Bangladesh war, their abil-
ity to prosecute it increased. The reason was a turn of events seemingly
unrelated to Kashmir: the Soviet Union’s invasion of Afghanistan. In
April 1978, the nationalist President Mohammad Daoud had been over-
thrown in a communist coup known as the Saur Revolution. A rebellion
erupted that summer in reaction to the new Khalq regime’s agenda of
radical reforms, which undermined the traditional clan-based structure
of Afghan society. By early 1979, antigovernment violence had spread
throughout the country. The Soviet Union, which had signed a friend-
ship treaty with the new government in December 1978, attempted to
support it through the provision of weapons and military advisers and
by managing its leadership. These efforts, however, proved to be ineffec-
tive. In December 1979, to salvage the rapidly worsening situation, the
Soviet Union invaded Afghanistan outright. By early 1980, the Soviets
had installed a new government under the nominal leadership of Babrak
Karmal, taken control of Afghanistan’s cities, and brought over eighty
thousand troops into the country. 56 Afghanistan, it seemed, was on its
way to becoming a puppet state of the Soviet Union.
These developments posed a considerable danger to Pakistan.
Specific Pakistani concerns included the potential resurrection of
Pakistan’s long-standing border dispute with Afghanistan; possible
economic, social, and humanitarian crises created by large numbers
of refugees flowing across the border from Afghanistan into Pakistan;
and the risk of direct military attack—perhaps on two fronts at once,
given the Soviet Union’s historically troubled relations with Pakistan
and its close relationship with India, which could have facilitated Indo-
Soviet military coordination. 57 These dangers were particularly acute
71
because of Pakistan’s relatively poor strategic situation at the time of the

Soviet invasion. The United States had recently levied sanctions against
Pakistan to punish it for pursuing a nuclear weapons program. United
States–Pakistan relations had also been damaged by the Pakistani
authorities’ slow reaction to a mob attack against the US Embassy in
Islamabad in November 1979. In addition, Pakistan had suffered broad
condemnation and isolation in response to its execution of former prime
minister Zulfikar Ali Bhutto. Pakistani military equipment had not
recently been upgraded and had consequently become badly outdated.
Finally, Pakistan’s economy was in the midst of a severe slump. 58 The
Soviet invasion of Afghanistan thus occurred at an especially inauspi-
cious time for Pakistan.
The Pakistanis responded to these dangers in two main ways. First, they
attempted to mitigate the effects of the refugee influx that had followed
the Soviet invasion. By 1987, over three million Afghans fleeing the war
had settled in Pakistani territory. This was the largest refugee population
in the world. The refugees brought with them disease, economic competi-
tion, and downward wage pressure for Pakistani workers and tradesmen;
strained infrastructure; and the exacerbation of political and religious
fault lines within Pakistan. Despite these problems, the Pakistanis contin-
ued to welcome Afghan refugees onto their soil. The Pakistanis offset the
costs of their hospitality through support from private, governmental, and
multilateral relief agencies, which they enthusiastically invited into the
country. These agencies, particularly the United Nations Commissioner
for Refugees and the World Food Programme, shouldered the lion’s share
of the economic burden of the refugee crisis. The Pakistanis also repeat-
edly insisted on the refugees’ right to return to Afghanistan as soon as
circumstances allowed. These policies not only helped Pakistan to deal
with the challenges of the refugees but also drew attention to the wider
Afghan war and ensured that the international community was politically
and economically invested in it. 59
The Pakistanis’ second major response to the Soviet invasion came
at the military level. The Bangladesh war had shown what could happen
when they engaged a significantly stronger adversary in a conventional
military conflict. Thus, there was no question of Pakistan directly con-
fronting the Soviet Union in Afghanistan. Rather, the Pakistanis decided
to return to their old strategy of challenging a stronger opponent through
the use of militant proxies. The main difference between past conflicts
and the Afghan war was that, in Afghanistan, the Pakistanis did not need
to raise their own proxy army. Afghan resistance forces, or mujahideen,
already existed. The Pakistanis needed primarily to assist these forces in
their efforts to fight the Soviets.
72
Such involvement in Afghanistan would not be an entirely new

endeavor for the Pakistanis. They had done similar things in the past to
protect their long-standing interests in the country, which included for-
malizing the Pakistan–A fghanistan border, acquiring strategic depth,
gaining access to central Asia, and minimizing India’s regional presence.60
Pakistan had pursued these interests through a range of policies including
alliance with the United States, appeals to international law, maintenance
of British administrative practices along the Pakistan–A fghanistan bor-
der, and coercive trade measures aimed at Kabul. The Pakistanis had also
employed a militant campaign against Afghanistan, training approxi-
mately five thousand Afghan Islamists at camps inside Pakistan from
1973 to 1977 and then sending them home to destabilize the govern-
ment.61 These Pakistani efforts increased following Afghanistan’s 1978
Marxist coup. Indeed, it appears that Pakistan was training Afghan anti-
government forces in the months just before the Soviet invasion. Thus,
as Robert G. Wirsing argues, Pakistan’s seemingly new support for the
Afghan resistance in fact represented “more of an augmentation than a
radical departure from Pakistan’s policy toward Afghanistan in the pre-
ceding decade.62
Even if support for the mujahideen was an augmentation of an existing
approach rather than a conceptually new policy, Pakistan’s latest strategy
in Afghanistan was a significant increase over anything that had gone
before. The Pakistanis took a number of important steps to aid the resis-
tance, including allowing mujahideen freely to cross the Afghan–Pakistan
border, providing fighters with sanctuary inside Pakistan, and permitting
the resistance to recruit new personnel within Pakistani refugee camps.
In addition, Pakistan served as the middleman between international
aid donors and the mujahideen, insisting that any military support for
the resistance flow through its Inter-Services Intelligence agency (ISI).
By 1987, when US aid to Afghanistan peaked at over $650 million,
approximately twenty thousand fighters per year were receiving train-
ing at seven Pakistani camps, with more being trained in Afghanistan.
Financial assistance from Saudi Arabia and other Arab states roughly
matched US contributions. The US Central Intelligence Agency (CIA),
the Saudi General Intelligence Presidency, and the ISI oversaw the influx
of money. The Pakistanis also funneled tens of thousands of tons per year
of arms to the mujahideen. The CIA used Saudi and US funds to purchase
Chinese, Egyptian, Polish, Israeli, and other weapons and materiel, which
were shipped to Pakistan. There the ISI took over, trucking the arms to
mujahidin-controlled depots along the Afghan border. The CIA paid for
these transportation expenses, along with other necessary supplies for
the mujahidin, such as food and clothing. The Pakistanis ensured that the
73
most hardline Islamist groups, such as Gulbuddin Hekmatyar’s Hizb-i-

Islami, received the lion’s share of the assistance. Thus, US efforts to com-
bat the Soviets in Afghanistan during the 1980s inadvertently supported
Islamists whom the US would have to fight post-9/11 as part of the war on
terror.63
The Pakistanis went to considerable lengths to conceal the extent
of their involvement with the mujahideen. For example, they publicly
denied providing the resistance with anything beyond moral and politi-
cal support. Pakistan also sought to control mujahideen activities by
managing their arms supply and involving Pakistani advisers closely in
the Afghans’ military planning and operations. This enabled Pakistan to
keep the conflict at a relatively low boil, to reduce the likelihood of being
directly implicated in the anti-Soviet resistance, and thereby to minimize
the danger of Soviet retaliation.64
Through these policies, General Zia turned what had initially been a
significant danger into an invaluable opportunity. Success did not come
immediately, however. Pakistan’s strategy got off to a slow start, as the
mujahideen were initially unsuccessful in their efforts to fight the Soviets.
The Afghans had little formal military training and were politically dis-
united. Their tactics were characterized primarily by poorly organized
direct offensives, often involving large troop formations. They proved to
be easy prey for Soviet forces’ overwhelming firepower. The Soviets, mean-
while, expanded their reach across Afghanistan, methodically establish-
ing bases in every province of the country. Although this strategy would
not enable them to win quickly, it seemed likely to result in victory over
the long term. Thus, as Larry Goodson writes, by three years after their
invasion, Soviet forces had “[grown] in strength while the mujahideen
were weak and ineffective… . The Soviets had penetrated every depart-
ment of the government, had reorganized their armed forces to meet the
Afghan situation … and appeared to be comfortable winning the war
gradually.”65
Despite these successes, however, the Soviets’ hold over Afghanistan
was limited. They primarily controlled roads, cities, and military
bases; Afghanistan’s vast countryside remained mostly beyond their
grasp. Also, with time, the mujahideen became more effective. They
managed to ameliorate, if not eliminate, their political divisions; they
made better use of small-u nit tactics; and the quality of their arms
and equipment improved. In response, the Soviets intensified their
counterinsurgency efforts, using air power to bombard areas that sup-
ported opposition forces. The goal of such attacks was not simply to
force these regions into compliance. Rather, it was to destroy them
completely through “rubbleization” and depopulation to eradicate the
74
popular support network upon which the rebels relied. The Soviets
also employed airborne and special operations forces to disrupt muja-
hideen supply lines and engage the rebels in areas inaccessible to con-
ventional forces. 66
The mujahideen lacked the organization and firepower to combat these
tactics. The Soviets therefore managed to retain the upper hand in the
Afghan conflict through the mid-1980s. Despite their advantage, how-
ever, the Soviets were unable to win. The countryside remained beyond
their reach. There the rebels exploited their local knowledge and identity,
launching raids and ambushes and then blending back into the popula-
tion or disappearing into mountain redoubts. They particularly targeted
Soviet lines of communication, impeding the movement of crucial sup-
plies and forcing maneuver elements to provide security for logistics oper-
ations. In addition, the Afghan people were completely opposed to the
Soviets’ presence and supportive of the resistance. Finally, international
opinion was strongly against the Soviet Union’s position in Afghanistan.
As a result, after years of inconclusive conflict, the Soviets were becoming
war-weary; the morale of their forces, as well as the political will of their
government and home population, was eroding. The mujahideen, for their
part, did not need to defeat the Soviet Union on the battlefield to prevail
in the overall conflict. They needed only to remain in the fight and avoid
being eradicated.67
This basic dynamic continued until, in 1986, the mujahideen acquired
the tool that turned the tide of the war—the shoulder-fi red US Stinger
antiaircraft missile. There had been considerable debate in the United
States and in Pakistan regarding the prudence of supplying the mujahi-
deen with the Stinger, given its potential for use against civilian aircraft.
The devastating effects of Soviet airpower on Afghan resistance forces
ultimately led US and Pakistani officials to make the Stinger available to
them. Even after doing so, however, Pakistani trainers imposed tight con-
trol over the missiles’ distribution and use.68
The Stinger finally gave the resistance a viable defense against Soviet air
capabilities. It quickly proved its utility, destroying nearly three hundred
Soviet aircraft in 1987 alone. This badly undermined the effectiveness of
Soviet air operations. For example, helicopter gunships could no longer
conduct close-range bombardment at will, while bombers were forced to
release their loads at higher altitudes, thereby reducing accuracy. Given
the centrality of airpower to Soviet counterinsurgency efforts, this sig-
nificantly reduced the Soviets’ ability to combat the mujahideen. In addi-
tion, by eroding the aura of invincibility that the Soviets’ air capabilities
had previously given them, these developments damaged Soviet morale
even further. The mujahideen, for their part, grew more confident in their
75
abilities and significantly increased the pace and intensity of anti-Soviet

operations.69
The Soviet Union now faced rising losses, an invigorated enemy, and
growing domestic political opposition to the Afghan conflict.70 Mikhail
Gorbachev, who had taken office in 1985 hoping to salvage the war, con-
cluded that the situation had become hopeless and decided to end it. By
1987, Soviet offensive operations had essentially ceased. In February
1988, referring to the war as one of the world’s “bleeding wounds, which
can result in gangrenous growth on the body of mankind,” Gorbachev
announced that Soviet withdrawal from Afghanistan would commence
in May of that year.71 By February 1989, the last remaining Soviet troops
had left the country.72
The Pakistanis had not initially imagined that their militant strat-
egy would succeed in ejecting Soviet forces from Afghanistan. By the
war’s end, however, it was clear that the strategy had played a critical
role in doing just that. As Ziring argues, when the last Soviet soldier left
Afghanistan in February 1989, “any doubt that Pakistan had had a role
in handing Moscow a stunning politico-strategic reversal was finally van-
quished.” “To the extent that [Pakistan’s strategy] motivated the ultimate
Soviet decision to exit from Afghanistan, it was hugely successful.”73
The Soviet Union’s withdrawal left chaos in its wake. Afghanistan now
faced years of civil conflict, as warlords overthrew the Soviet puppet gov-
ernment of Mohammad Najibullah and then fought among themselves
for control of the country. Over the coming decade, Pakistan would
capitalize on this situation to increase its influence in Afghanistan. First,
however, the Pakistanis would exploit an even more important strategic
opening that had resulted directly from their successful management of
the Afghan conflict.
From the beginning, the Pakistanis had viewed the Afghan war not
simply as a danger, but also as an opportunity—an opportunity once
again to pursue their militant strategy in an effort to wrest Kashmir from
India. General Zia believed that if Pakistan could serve as the primary
conduit for international aid to the Afghan resistance, it could inflate
the cost of its support for the war effort and then divert the profits from
US reimbursements to Kashmiri rebels. He maintained that the United
States, preoccupied with its goal of damaging the Soviet Union, would
ignore the Pakistani scheme. The Afghan conflict would thus serve as a
“smokescreen” behind which Pakistan could wage a renewed militant
campaign in Kashmir. Indeed, Zia reportedly referred in private to the
war in Afghanistan as “the Kashmir jihad.” 74
This was an important part of the reason that Pakistan insisted on
serving as the exclusive link between international aid donors and the
76
mujahideen. By controlling money and materiel intended for the resis-

tance while being subjected to only minimal oversight, the Pakistanis
were able to divert considerable resources—one estimate puts the
amount at 30 percent or more—away from the war to other uses.75
Some of this diversion took the form of ordinary corruption and simply
enriched individual Pakistani personnel. Diverted resources were also
redirected in support of Pakistan’s national-level efforts, however, and
helped to advance its goal of promoting the Kashmir jihad. In addition,
as General Zia had predicted, Pakistan’s role as the only conduit to the
mujahideen enabled him to charge the United States and other sponsors
such as Saudi Arabia the highest possible price for Pakistani services.
Zia eventually secured from the United States an aid package of $3.2
billion, a deal to purchase F-16 fighter aircraft, and promises to reduce
pressure on Pakistan over its nuclear program and its human rights
record.76 This money, materiel, and political breathing space could be
used to support Pakistan’s agenda in Kashmir and to defend Pakistan
against India.
Even as they supported US efforts in Afghanistan, then, the Pakistanis
were setting the stage for a new, expanded Kashmir campaign. Pakistan
used its newfound resources to employ militants even more extensively
than it had previously. In the mid-1980s, the Pakistanis struck an agree-
ment with the Jammu and Kashmir Liberation Front (JKLF), under
which they would provide military and financial support for the JKLF
to fight India in Kashmir. The Pakistanis made clear that unlike in the
past, when their troops fought along with the militants, they would not
commit their own forces to battle; the militants would have to face the
Indians on their own. The militants would, however, “have the support of
the Pakistani military and the ISI… . [T]‌he ISI would pay the bill … and
stand behind them with other kinds of support.” 77 Collaboration between
Pakistan and the JKLF became extensive, with the Pakistanis performing
such services as publishing JKLF propaganda materials and assisting with
recruitment, in addition to providing the JKLF with military and finan-
cial backing. Meanwhile, the JKLF established roughly three hundred
sleeper cells in Indian Kashmir and waited for instructions from front
leaders in Pakistan.78
By the time Soviet troops began withdrawing from Afghanistan
in 1988, the Pakistanis had thus significantly bolstered their ability
to fight a Kashmiri proxy war. And, just as Zia had predicted, a pre-
occupied United States had turned a blind eye toward their activi-
ties. As long as the Pakistanis were helping to oust the Soviets from
Afghanistan, US leaders were not concerned with such peripheral
matters as Kashmir.79
77
EVALUATING THE AFGHAN WAR
At first blush, Pakistani support for the mujahideen during the Afghan
war appears to have been a new strategy, devised in reaction to the Soviet
Union’s unexpected and highly threatening invasion of Afghanistan. It is
better understood, however, as a reprise of Pakistan’s traditional approach
to dealing with its security challenges. Most broadly, Pakistan’s Afghan
policy was a return to the strategic use of militancy after the catastrophe
of pursuing a conventional military strategy against a stronger oppo-
nent during the Bangladesh war. More specifically, within Afghanistan,
Pakistan’s conduct bore many similarities to its policy in years past, which
had included training and arming Islamist fighters to combat the Afghan
government. Pakistan’s new policy did differ from its past approach in
scale, far exceeding the levels of earlier efforts. Nonetheless, the new pol-
icy was an augmentation of, rather than a qualitative departure from, its
earlier approach. Finally, Pakistan’s strategy during the Afghanistan war
enabled it to return to its oldest strategic theme, the use of militants to
undermine Indian rule in Kashmir. Indeed, thanks to its Afghan expe-
rience, Pakistan would now be able to support the Kashmir jihad with
significantly increased resources. Pakistan’s policy during the war in
Afghanistan thus ensured that the sophistication and efficacy of its mili-
tant proxy strategy would continue to grow.
Operationally, the strategy enjoyed many of the advantages that
I discussed in c hapter 2. First, it was inexpensive. The strategy did not
require the Pakistanis to use their own personnel. Nor did the Pakistanis
have to raise the resistance forces themselves, as the Afghan mujahideen
already existed. Pakistan had only to train and equip an extant force.
In addition, Pakistan did not bear the costs even of these efforts, since
foreign donors like the United States and Saudi Arabia footed the bill
for training and equipment and then paid the Pakistanis a premium in
return for their services. It is of course true that the millions of Afghan
refugees who fled into Pakistan were costly to maintain. The presence
of the refugees did not, however, result from Pakistan’s support for the
mujahideen. It resulted, rather, from the Soviet invasion of Afghanistan
and the ensuing violence that wracked the country. Also, international
organizations covered much of the monetary cost of the refugees. From
a financial perspective, then, Pakistan’s use of militants in Afghanistan
was especially cheap.
Second, Pakistan’s Afghan strategy made denial operations difficult
for the Soviets. The Pakistanis were able to infiltrate mujahideen in and
out of Afghanistan with relative ease. They could not have accomplished
this with their own military, which the Soviets would have detected and
78
repulsed. Inside of Afghanistan, the militants were difficult to fix and

destroy; they blended into the population and terrain of the countryside,
which the Soviets never succeeded in controlling. This was particularly
true as the rebels’ tactics and equipment improved over the course of the
war. The Soviets’ only real recourse was to rubbleize and depopulate the
countryside, in an effort to eradicate the mujahideen’s support networks.
Even such extreme tactics failed to defeat the rebels, however. In addition,
they ensured that the Afghan people would remain firmly opposed to the
Soviet presence in their country.
Third, fighting the Soviets with nonstate militants, rather than with their
own conventional or even special operations forces, helped the Pakistanis
to maintain deniability regarding their involvement in the Afghan war.
This, in turn, enabled the Pakistanis to avoid giving the Soviets a clear jus-
tification for launching punishment attacks against them. Pakistan’s sup-
port for the mujahideen was an open secret; the Soviets were well aware
that they were training, equipping, and providing sanctuary to the rebels.
Nonetheless, the fact was that the forces actually fighting the Soviets were
a collection of nonsate actors—they were not members of the Pakistani
military. As a result, large-scale attacks on Pakistan would have been dif-
ficult for the Soviets publicly to justify. Not surprisingly, the Soviets never
undertook such operations. They did seek to put pressure on Pakistan
during the later years of the war through border violations in the form of
air and artillery strikes, as well as terrorist attacks on Pakistani soil by the
Afghan intelligence service. These terrorist operations may have included
failed attempts on Zia’s life. Despite their deep frustration with Pakistani
support for the mujahideen, however, the Soviets avoided more extensive
attacks on Pakistan.80
The Pakistanis’ militant strategy, then, not only helped to hasten the
Soviets’ exit from Afghanistan but also gave Pakistan significant addi-
tional resources to devote to the Kashmir struggle—a ll at low human
and financial cost and while insulated from the danger of Soviet punish-
ment. After the debacle of Bangladesh, it was a considerable success. The
Pakistanis could now turn their attention back to the East and the prob-
lem of Kashmir.
CONCLUSION
This chapter traced the development of Pakistan’s militant proxy strat-

egy from the 1965 Indo- Pakistani war to the anti- Soviet struggle
in Afghanistan during the 1980s. I showed that Pakistan’s use of a
79
well-trained and organized force of religiously motivated guerrillas dur-

ing the 1965 war represented a considerable advancement from its origi-
nal employment of informal local militias in 1948. Although the 1965 war
did not attain its main strategic goal of capturing Kashmir, it was not a
complete failure. It benefited from several of the operational advantages
that I identified in c hapter 2 and achieved a number of tactical successes.
In addition, because the Pakistanis’ shortcomings resulted to a significant
degree from planning errors, rather than from an inherently flawed oper-
ational concept, Pakistani strategists could remain optimistic regarding
the potential of their militant strategy.
In the next section, I showed that Pakistan’s use of conventional mili-
tary forces during the 1971 Bangladesh war actually increased the signifi-
cance of Pakistan’s militant strategy. The war demonstrated the danger
of confronting India directly in a conventional conflict while increasing
the importance of the security competition and dispute over Kashmir
with India to Pakistani state building. Because it could enable Pakistan to
continue to challenge India while avoiding catastrophic defeat, Pakistan’s
militant strategy assumed an unprecedented level of national importance.
Pakistan’s development of a nuclear weapons capability, which also was
motivated to a significant degree by the bitter experience of Bangladesh,
would help to make its militant strategy particularly effective and
destabilizing.
Pakistani leaders consequently treated the subsequent conflict in
Afghanistan not just as a danger, but also as an opportunity. By provid-
ing assistance to the mujahideen, the Pakistanis were able once again to
benefit from the operational advantages of a militant proxy strategy. This
enabled them to help eject Soviet forces from Afghanistan while avoiding
a catastrophic confrontation. Simultaneously, the Pakistanis were able to
divert considerable military and economic resources to their own pur-
poses. These resources, in turn, enhanced the Pakistanis’ ability to pursue
their primary national security goal of supporting jihad and undermining
Indian rule Kashmir.
In chapter 5, I examine Pakistan’s renewed efforts in Kashmir in detail.
I show that although the Pakistanis did not create the Kashmir insur-
gency, they helped to trigger it and then carefully managed it by manipu-
lating their support for a range of militant organizations. In so doing,
the Pakistanis exerted a tremendous impact on the rebellion’s character
and trajectory. Later in the chapter, I discuss Pakistan’s continued use
of militancy in Afghanistan following the Soviet Union’s 1989 with-
drawal. By facilitating the rise of the Taliban during the 1990s, Pakistan
ensured that it had a friendly regime and increased strategic depth on
80
its western border, as well as improved access to the natural resources

of Central Asia. Pakistani support for the Taliban endured even after it
joined the US-led war on terror in 2001. Pakistan thus received consider-
able international aid to fight the Taliban even as it continued to assist
the Taliban, thereby keeping open the possibility that it could someday
return to power.
81
CHAP T ER 5
Kashmir and Afghanistan Reprise
I n the decades following the first Kashmir war, in a series of conflicts

ranging from the 1965 Kashmir war, to the Bangladesh conflict, to the
anti-Soviet struggle in Afghanistan, Pakistan had acquired experience
and resources that increased the importance and sophistication of its mil-
itant proxy strategy. By the late 1980s, it was ready to bring its improved
capabilities to bear on Kashmir once more, in a renewed effort to wrest
the territory from Indian control. Later, in the 1990s, it also returned its
attentions to Afghanistan, utilizing its proxy strategy to promote its inter-
ests in the chaos of post-Soviet Afghanistan.
In this chapter, I closely examine these campaigns. I first explain the
Pakistanis’ role in helping to launch and manage the Kashmir insurgency.
I show that, in doing so, they transformed a grassroots nationalist uprising
into an organized rebellion seeking Kashmiri accession to Pakistan. This
enabled Pakistan to impose significant political, financial, and military
costs on India while avoiding direct Indo-Pakistani conflict. In the second
section, I discuss Pakistan’s continued use of militancy in Afghanistan.
By supporting the rise of the Taliban during the 1990s, Pakistan was able
to acquire a friendly neighbor and improved strategic depth to the West,
greater access to Central Asian natural resources, and a training ground
for anti-Indian militants, without the cost and risk of direct military inter-
vention. Pakistan continued to support the Taliban even after joining the
US-led “war on terror” following the attacks of September 11, 2001. In
both cases, Pakistani strategy benefited from the operational advantages
I discussed in chapter 2, making both denial and punishment operations
difficult for defenders.
These successes demonstrated the potential effectiveness of Pakistan’s
militant strategy when executed in a favorable political environment. As
I explain in the chapter’s final section, however, Pakistani strategy began
82
to develop serious problems during this period, despite its successes.

These problems were related to an ongoing challenge for Pakistan’s mili-
tant proxy strategy—ensuring adequate convergence between the inter-
ests of Pakistan and those of its proxies.
THE KASHMIR INSURGENCY
Pakistan had launched the 1965 war in the belief that Kashmiris were vio-
lently opposed to Indian rule and would rise up to support invading mili-
tant forces. The Pakistanis were right that Kashmiris were not happy with
India, but they were wrong in their belief that the Kashmiris were ready
to fight to free themselves from it. The Pakistanis had deployed their strat-
egy at the wrong time, when the structural conditions necessary for suc-
cess did not yet exist.
The right time came decades later, during the late 1980s. By this
point, sufficient resentment had built up among the Kashmiri popula-
tion to lead them to violent rebellion. This resentment resulted from
the simultaneous political mobilization of Kashmiris and the decay
of regional governmental institutions. Kashmiris became more aware
of the world around them through improved access to education and
media, but also lost autonomy as India tightened its grip on the terri-
tory. This left them without legitimate avenues for political participation
even as they became more politically interested and aware. Blame for
the situation lay primarily with the Indian government, which had over
the decades steadily encroached on Kashmir’s traditional autonomy
within the federal union through a combination of legislative fiat and
electoral malfeasance.1
The tipping point in this process was the Jammu and Kashmir state
assembly elections of April 1987. In November 1986, Kashmiri National
Conference leader Farooq Abdullah signed an accord with Rajiv Gandhi,
which enabled Farooq to return to the chief ministry of Jammu and
Kashmir. Gandhi’s government had unceremoniously dismissed Farooq
just two years earlier, in 1984. The accord damaged Farooq’s legitimacy,
giving him the appearance of an opportunist. More broadly, it sug-
gested the existence of an unprincipled alliance between the National
Conference and the Congress(I) party in New Delhi. This hurt both par-
ties, though in different ways. The National Conference could no longer
be seen as a truly nationalist organization, protecting Kashmiri interests
against Indian encroachment. The Congress(I), for its part, could no lon-
ger be seen as a buffer against ethnic Kashmiri domination in Jammu and
Kashmir, particularly in the Jammu region of the state.2
83
K a s h mi r a n d A f g h a n is ta n R e p r is e ( 83 )
The Kashmiri and Indian governments’ credibility thus was already

damaged when the April 1987 state elections occurred. It was hurt even
further by their conduct during the elections, which involved voter
intimidation, ballot-box stuffing, and the arrest of opposition workers.
The Congress(I)–National Conference alliance won in a landslide, tak-
ing sixty out of seventy-six seats in its contest with the opposition, the
Muslim United Front (MUF). But the victory was hollow, given the alli-
ance’s misbehavior during the campaign. That was a turning point for the
Kashmiri people. As Sumit Ganguly writes, “The conduct and outcome of
[the 1987] election closed the last possible venue for legitimate dissent in
Kashmir.”3 This created a situation in which Kashmiris were finally ready
to challenge Indian rule with violence and increasingly to support Islamist
political groups opposed to both the Kashmiri and Indian governments.
The Pakistanis recognized this changed environment and sensed that the
time was ripe to exploit the Kashmiris’ discontent and changing political
preferences.
Some scholars have noted that, even without the Congress(I) and
National Conference parties’ electoral malfeasance, the MUF was
unlikely to have won a majority in the 1987 elections. Despite widespread
unhappiness with both of the entrenched parties, the opposition simply
lacked sufficient popular support to defeat the Congress(I)–National
Conference alliance. Thus, while the Indian and Kashmiri government’s
campaign irregularities were real and undoubtedly had some impact on
the final vote tallies, they did not constitute outright theft of the election.
In this view, much of the Kashmiri population’s subsequent willingness to
support violent political change, and its increased sympathy for Islamist
political agendas, can be attributed to frustration with the limits of the
Islamist opposition’s electoral viability, rather than the damage to the
Congress(I) and National Conference parties that resulted from their
electoral misbehavior.4
Both sets of arguments are important to understanding the changed
political environment in Kashmir. Even if the Congress(I) and National
Conference did not, strictly speaking, steal the 1987 elections, their con-
duct during the campaign further undermined their already-damaged
credibility, alienating the Kashmiri population to the point that it was
willing to countenance violent change and increasingly to support
Islamist sociopolitical goals. Members of the Islamist opposition, for their
part, became convinced by their loss at the polls that they were unlikely
to take power through the mechanisms of electoral politics and would
have to resort to violence to realize their agenda. Together, these two fac-
tors ensured that the popular mood turned decisively against the status
quo, and that violent change and increasingly Islamist politics lay on the
84
horizon for Kashmir. As Praveen Swami writes, “for the first time since
1947 … Jammu and Kashmir did have a genuine mass constituency for
the Islamists, hostile both to the National Conference and to New Delhi.
Where earlier phases of the jihad had failed precisely because of the
absence of such a constituency … the conditions now seemed right to
make another attempt.”5
Conditions also were right to resume support for the Kashmiri jihad
for reasons external to Kashmir, such as the conclusion of the Afghan war.
General Zia had viewed participation in the war not simply as a means of
protecting Pakistan from the dangers of the Soviet Union’s presence in
Afghanistan. He also viewed it as an integral part of the Kashmir struggle
and ensured that the Pakistanis used the war effort to amass financial and
military resources that would improve their ability to fight in Kashmir.
Now, after years of supporting the Afghan mujahideen and helping to
drive the Soviet Union out of Afghanistan, the Pakistanis were ready to
put those resources to use. The Afghan war had also given the Pakistanis
valuable experience in the tasks necessary to manage an insurgency, which
could be applied to similar projects elsewhere. This augmented opera-
tional experience and influx of material resources significantly improved
Pakistani asymmetric warfare capabilities.6
The combination of political change within Kashmir and improved
Pakistani capabilities meant that, when the Pakistanis began working
closely with the Jammu and Kashmir Liberation Front (JKLF) in the
mid-1980s to lay the groundwork for revolt, structural conditions were
finally favorable for a successful outcome. The results of Pakistan’s efforts
were therefore far more effective than they had been in 1965.
These efforts were critical to the outbreak of the insurgency. Scholars
often point out that Pakistan did not cause the discontent underlying the
Kashmiri uprising. This, of course, is true. Popular discontent in Kashmir
resulted largely from chronic mismanagement and malfeasance on the
part of the Indian central government, as well as the Kashmiri National
Conference. It was not a Pakistani creation. The Pakistanis actively capi-
talized on Kashmiri discontent, however, and played a crucial role in
transforming spontaneous, decentralized opposition to Indian rule into a
full-fledged insurgency, dedicated to promoting an Islamist sociopolitical
agenda and violently joining Kashmir to Pakistan.
For example, the insurgency’s initial round of violence occurred only
after the Pakistanis coerced the JKLF into launching a series of attacks in
Kashmir. The front had been delaying the start of its campaign to carry out
more extensive preparations. It wanted to train additional personnel and
establish more cells inside of Indian Kashmir. The Pakistanis, however,
sensed that the moment was ripe to launch the insurgency and believed
85
that the JKLF had sufficient resources at its disposal to do so. About ten
thousand fighters reportedly had already been trained and were ready for
use. The Pakistanis wanted to ensure that they did not miss a long-awaited
opportunity and were impatient to realize a return on their investment of
time, money, and military resources. Thus, during 1988, the Inter-Services
Intelligence agency repeatedly pushed the JKLF to launch attacks in
Kashmir, at one point threatening to betray JKLF leaders to the Indian
security services if it did not act soon. Under Pakistani pressure, the JKLF
finally gave in and orchestrated two successful bombings in Srinagar in
July 1988.7 This was the insurgency’s opening salvo.
The initial phases of the insurgency were not primarily designed to
achieve military effects. Their aim, rather, was symbolic, seeking to dem-
onstrate in dramatic fashion India’s loss of political authority in Kashmir.
Thus, the rebels staged relatively few attacks on Indian security forces.
Instead, they assassinated judges, intelligence officers, and National
Conference leaders; prevented voters from participating in elections;
and staged high-profile kidnappings, such as the abduction of an Indian
Congress party leader’s daughter, who was freed only after the Kashmiri
government released several JKLF fighters being held on terrorism
charges. Meanwhile, the Pakistanis increased the intensity of their opera-
tions. By 1990, several hundred Kashmiris per month were crossing into
Pakistani territory for military training and then returning to Kashmir to
fight the Indian occupation. Violent clashes between militants and secu-
rity forces soon became a daily occurrence.8
The insurgency quickly became far more than a mere uprising, or even
a campaign of symbolic violence—it became a genuine war for control of
Jammu and Kashmir. Unlike in the past, this war was not limited to short
periods of state-to-state conflict between India and Pakistan. Rather,
thanks to General Zia’s careful preparations, the struggle was now an
ongoing jihad, which continued even in peacetime and had no end in sight.
Realizing this, Indian leaders devoted significant resources to combatting
the rebellion. The Indian government dissolved the Kashmiri state assem-
bly, placed Jammu and Kashmir under governor’s rule, and, in January
1990, appointed the well-k nown civil servant Jagmohan Malhotra as
governor.
Jagmohan’s appointment is indicative of the seriousness with which the
Indians took the Kashmir problem. He had served successfully as lieuten-
ant governor of a number of states and territories, including Delhi and
Goa, and as governor of Jammu and Kashmir during the 1980s. In addi-
tion, Jagmohan was a firm believer in the utility of brute force. His mandate
was clear: to crush the rebellion decisively and restore New Delhi’s writ
in Kashmir. As he explains, “Our first and foremost objective was to …
86
assert the authority of the state … no matter what the costs, no matter what
the sacrifices. Our resolve, our will, had to be made clear… . It had …
to be conveyed to all concerned, in no uncertain terms, that … no soft
underbelly of the state would be offered to punch or fool with.”9
Despite the seriousness with which the Indians took the Kashmir prob-
lem, they had great difficulty stopping the Pakistanis’ militant campaign.
They could not prevent the movement of fighters across the porous border
separating Indian from Pakistani Kashmir. And once the militants had
entered Indian Kashmir, it was difficult to apprehend them in advance of
their attacks. During 1990, approximately 1,000 civilians and 130 mem-
bers of the security forces died in roughly 4,000 instances of militant
violence.10
The Indians responded with draconian measures. They deployed hun-
dreds of thousands of security forces to Kashmir. The government used
the 1987 Terrorist and Disruptive Activities (Prevention) Act (TADA) to
hold terrorism suspects for up to one year without filing formal charges
against them; to try them in camera, utilizing secret witnesses; to treat
them as guilty until proven innocent in certain cases, such as those involv-
ing explosives or firearms; and to allow them to make appeals only to the
Supreme Court of India, within thirty days of a judgment.11 Numerous
Indian government officials, including the members of the Supreme
Court, publicly recognized TADA’s propensity for abuse.12 Indian secu-
rity forces also employed heavy-handed tactics such as cordon-and-search
operations, which resulted in the widespread destruction of property and
harm to civilians; routinely engaged in extrajudicial killings; and regu-
larly employed torture to extract confessions and intelligence from sus-
pected militants.13 Whatever their tactical utility may have been, these
measures deepened India’s strategic problem by further alienating the
Kashmiri population. They also badly tarnished India’s international rep-
utation as a democratic, law-abiding state dedicated to the promotion of
human rights.14
Finally, after multiple attempts, the Pakistanis had managed to devise
a winning militant strategy in Kashmir, which extracted significant mili-
tary, economic, and human costs from India while avoiding the dangers of
direct conventional conflict. The Pakistanis quickly discovered, however,
that the militant campaign would not run itself. It required close super-
vision if it was to remain successful over the long term. They therefore
spent the subsequent decades carefully managing the insurgency, using
their military, financial, and political resources to determine its charac-
ter and trajectory. The biggest challenge that Pakistan faced was ensuring
proper alignment between Pakistani national interests and the interests
and capabilities of its militant proxies. In this vein, the Pakistanis’ first
87
major step after triggering the outbreak of the Kashmir rebellion was to
undermine the position of their primary allies, the JKLF.
Pakistan’s problem with the JKLF was threefold. First, the front’s
ideology was at least notionally secular, emphasizing the importance of
democratic rights such as sovereignty and self-determination, rather than
the pursuit of religious goals. It is not clear how deep the group’s secular-
ism actually ran. As Praveen Swami has noted, despite its rhetoric, the
front never sought to curb religion-based violence in Kashmir, such as
attacks on liquor stores and cinemas or assaults against unveiled women.
In fact, the JKLF was implicated in a series of assassinations of members
of Kashmir’s Hindu Pandit community that triggered the Pandits’ exodus
from the Kashmir Valley. In addition, some of the most oppressive poli-
cies ever imposed in Kashmir, particularly with regard to women’s rights,
were instituted while the front was ascendant.15
These are important points that illustrate the complex relationship
between the JKLF’s stated ideology and the reality of its political prac-
tice. Even if the front tolerated or engaged in religiously motivated vio-
lent behavior, the fact remains that its stated organizational goals were
secular. This put the JKLF fundamentally at odds with its Pakistani
patrons.
Pakistan’s second problem with the JKLF was that it was a genuinely
nationalist organization, dedicated to achieving Kashmiri independence
rather than joining Kashmir to Pakistan. According to its official litera-
ture, the “Kashmir issue is not a territorial dispute between India and
Pakistan but concerns the unfettered right of self-determination of the
people of Jammu Kashmir State. Jammu Kashmir State or any part of it is
not a constitutional or integral part of India, Pakistan, or any other coun-
try.” Their objective is the “re-unification and complete independence of
Jammu Kashmir State.”16 These views ran counter to Pakistan’s policy of
joining Kashmir to Pakistan.
Pakistan’s third problem with the front was its lack of operational capa-
bilities. The JKLF’s cadre, despite their nationalist zeal, did not consist of
hardened fighters. Although the front was certainly willing to engage in
extensive violence, it therefore did not prove to be as militarily effective as
the Pakistanis had hoped.17
The Pakistanis tried to push the JKLF in directions more compatible
with their interests. For example, they asked the front to stop calling for
Kashmiri sovereignty in its public statements. They also requested that an
ISI representative be allowed to observe the workings of the JKLF Central
Committee. The front, however, refused to entertain these suggestions.
The JKLF’s failure to cooperate with their patrons led the Pakistanis to
conclude that their differences with the front were irreconcilable.18
88
The Pakistanis therefore decided to replace the JKLF with more use-
ful partners. They cut off funding for the front and, during the early
1990s, supported a variety of other militant groups. Eventually, they
settled primarily on the Hizb-u l-Mujahideen (HM). HM based its
agenda on that of its political patron, the Jamaat-e-Islami (JI), which
promoted Islamist sociopolitical goals and advocated the accession of
Kashmir to Pakistan. Pakistan had engineered the alliance between the
JI and HM during the 1989–9 0 timeframe, to ensure ideological com-
patibility with its new proxy. The relationship between the two groups
became extremely close—so close that HM allowed the Jamaat to
nominate one of its members to become leader of HM. In Arif Jamal’s
words, these developments “virtually turned [HM] into a subsidiary of
Jamat-i-Islami.19
HM used the newfound resources that it acquired through its relation-
ship with Pakistan to wage a full-blown war against the JKLF. It publicly
accused the front of atheism and of indifference to religious goals and
concerns. In addition, HM fighters assassinated, kidnapped, and tortured
front personnel. This campaign, combined with vigorous targeting by
Indian security forces, devastated the front and enabled HM to emerge
as the dominant militant group in Kashmir, a status that it maintained
through the rest of the 1990s.20 During this period, violence in Kashmir
increased significantly. For example, the number of militant attacks
climbed from about 3,700 in 1991 to more than 5,800 in 1995. Civilian
fatalities rose from approximately 900 to 1,200, and security-force deaths
jumped from roughly 170 to nearly 240 during this period.21
Although the insurgency at times appeared to be wholly beyond Indian
control, the Indians did manage to stem its tide by the late 1990s. They
used a number of techniques in their efforts to defeat the militants. For
example, they sowed discord within militant groups and turned fighters
against their own comrades, cooperated with relatively moderate militant
factions and individuals, and attrited militant forces through detention
and outright killing.22 One of India’s biggest advantages was a recurring
Pakistani problem that had nothing to do with Indian policy: Pakistan’s
selection of its militant allies. Although the Pakistanis had finally found
the right political environment in which to pursue Kashmiri jihad, they
had not yet identified the optimal militant groups to carry it out. The
Pakistanis’ decision to abandon the secular nationalist JKLF and sup-
port the Islamist, pro-Pakistan HM had paid handsome dividends.
Nonetheless, the overlap of interests and goals between Pakistan and
HM was far from perfect. Like the JKLF before it, HM diverged from
Pakistan’s agenda in important ways and ultimately proved to be less than
an ideal partner for Pakistan.
89
The main problem with HM was that the group was composed primar-
ily of Kashmiris, who had brethren in Indian Jammu and Kashmir and
who cared deeply about the territory’s welfare. HM members thus worried
about the consequences of their militant activities and were less inclined
to engage in extremely violent actions against the Indian authorities than
the Pakistanis wished. In addition, in important instances, HM leaders
were willing to strike compromises directly with the Indian government.
Perhaps the most dramatic example of this propensity for compromise
was legendary HM commander Abdul Majid Dar. Dar was, according to
a senior Indian counterinsurgency force commander, a hardened mili-
tant and “a real fighter,” unlike many of the native Kashmiri militants.
Nonetheless, Dar ultimately decided to renounce violence, proclaiming
an HM ceasefire in 2000 and encouraging Kashmiri separatist groups to
resolve their differences with India in an amicable fashion. Given Dar’s
highly influential status, numerous HM field commanders followed his
example. Although the HM ceasefire eventually fell through, the damage
had been done; Dar’s behavior seriously undermined Pakistani leaders’
faith in the group. Dar was among the most aggressive and highly moti-
vated members of HM. If he could not be trusted to wage an all-out battle
with the Indians, the Pakistanis reasoned, neither could any other mem-
bers of the group. As a result, the Pakistanis became far less supportive of
the HM than they had been previously. This confluence of developments
led to infighting with the HM, which damaged its effectiveness.23
Given the combination of increasingly sophisticated Indian counterin-
surgency tactics and rifts between Pakistan and its militant allies, insur-
gency was on the wane in Kashmir by the late 1990s. Incidents of militant
violence declined to fewer than three thousand, and civilian deaths
dipped below one thousand in 1998.24 The Pakistanis, however, were
not simply going to abandon their militant campaign and accept defeat
in response to these setbacks. Their problems resulted in large measure
from improper selection of their militant proxies. The Pakistanis needed
a militant force that not only was dedicated to Pakistan’s Islamist agenda
and its goal of joining Kashmir to Pakistan but also was willing to employ
extreme violence in the pursuit of these ends, able to pose a significant
tactical challenge to Indian security forces, and utterly unwilling to com-
promise with its adversaries. Organizations without this combination of
characteristics would be unable to wage jihad effectively. Groups that pos-
sessed these characteristics, however, could be highly effective, giving the
insurgency the boost it needed and ensuring that it continued well into
the future.
The Pakistanis therefore began to recruit a new class of militant groups
as their partners in the Kashmir jihad. Like HM, these organizations
90
promoted Islamist sociopolitical agendas and sought Kashmir’s acces-

sion to Pakistan. Unlike HM, however, they consisted mainly of non-
Kashmiris, who were less concerned with the insurgency’s impact on the
people of Kashmir. Thus, these mehmaan mujahideen, or guest fighters,
were not amenable to any form of compromise with India and did not hes-
itate to employ extreme violence in pursuit of their ends. They also were
willing to target Muslims who ran afoul of their philosophy.25
By shifting their support to foreign mujahideen, the Pakistanis were
accepting a significant trade-off. For the Pakistanis to ensure that the
insurgency enjoyed the appearance of a genuine Kashmiri movement,
they needed to employ fighters of native Kashmiri origin. As noted ear-
lier, however, the Kashmiris’ proclivity for combat left much to be desired;
they shied away from extremely violent tactics and were prone to com-
promise with the Indians. Outside fighters, by contrast, lacked the legiti-
macy of indigenous Kashmiris. Their goals for Kashmir, however, were
more closely aligned with those of the Pakistanis. In addition, they were
willing to employ extreme violence and were capable of doing so effec-
tively, many of their fighters having gained extensive combat experience
in Afghanistan. Moreover, their lack of personal ties to Kashmir insulated
them from Indian reprisals against their families, as well as inducements
to cooperate with the government. Finally, the inclusion of foreign fight-
ers could give the Kashmir struggle an international flavor, which could
help to attract world attention to the Kashmiris’ cause. The Pakistanis
decided that these benefits would be worth the possible legitimacy costs
of supporting non-K ashmiri militant organizations.26
The Pakistanis afforded these outside groups, which included Lashkar-
e-Taiba (LeT), Jaish-e-Mohammed, al-Jehad, al Omar, Harkat-u l-A nsar,
and the Ikhwanul Muslimeen, extensive financial, logistical, and mili-
tary support.27 Lashkar-e-Taiba soon emerged as the Pakistanis’ favorite
militant organization. LeT arrived on the scene as a small group, without
an extensive base of support in Pakistan. ISI officials therefore believed
that LeT would be especially dependent upon Pakistani assistance and,
consequently, relatively easy to control. The Pakistanis also believed that
LeT’s agenda for Kashmir and South Asia was compatible with their own
strategic interests. Wresting Kashmir from Indian control and joining it
to Pakistan was Lashkar’s central mission. Beyond this, the group viewed
the Kashmir jihad as part of a wider struggle against Hindu India; once
Kashmir had been recovered for the Muslims, the territory could serve
as a stepping stone to restoring Islamic control over the rest of the sub-
continent. This larger, pan–South Asian agenda ensured that LeT would
fight India to the end and would not seek compromise with New Delhi in
the manner of HM. Finally, the Pakistanis did not fear that LeT would
91
turn against them in the future. LeT’s doctrine held that it was wrong to
wage jihad against Muslim rulers, even if those rulers were misguided.
LeT’s leaders also believed that fighting against the Pakistan government
would distract them from their main task of freeing Kashmir from Indian
oppression. The Pakistanis were thus confident that LeT’s ideology and
practical concerns would ensure that the group would not bite the hand
that fed it.28
The Pakistanis therefore worked with LeT more closely, and sup-
ported them more extensively, than other militant organizations. Indeed,
LeT’s military capabilities were largely a product of Pakistan Army and
ISI efforts to construct a force for use against India from the ground up.
Specific forms of aid included organizational, strategic, and financial sup-
port; military training; the supply of weapons, communications gear,
and other materiel; assistance infiltrating and exfiltrating Kashmir and
India proper; and protection on Pakistani soil. Numerous army and ISI
personnel also contributed their expertise by joining LeT following their
retirement.29
In addition to material support from Pakistan, LeT benefited from
tactical innovations that significantly increased the effectiveness of their
operations. One such innovation was the fedayeen attack. 30 In fedayeen
operations, heavily armed assailants launched high-r isk assaults against
prominent military and civilian objectives. During these operations, the
militants did not seek immediate death, but rather sought to fight for as
long as possible, to maximize damage to their targets. When and if the
militants died, they fell at the hands of the enemy. These tactics enabled
the militants to push violence in Kashmir to a new level. This heightened
violence, they hoped, would intimidate Indian security forces and rein-
vigorate the flagging insurgency. In addition, fedayeen tactics avoided the
Muslim injunction against suicide, which some religious commentators
had raised as an objection to martyrdom operations. 31
The results of Pakistan’s shift to LeT and the mehmaan mujahideen were
impressive. Fatalities among Indian forces rose to their highest-ever level
in the Kashmir insurgency. In 2000, over 480 security personnel died in
militant violence. By 2001, that figure had increased to 613. Significantly,
civilian deaths during this period were lower than they had been during
the mid-1990s. Approximately 1,400 civilians had died in militant vio-
lence during 1996, as compared to about 940 in 2000 and 1,100 in 2001. 32
These figures highlight an interesting paradox. The new, “foreign” militant
groups like LeT were less concerned with the welfare of Kashmiris, and
more willing to deploy extreme violence, than indigenous organizations
such as HM. Nonetheless, the mehmaan fighters apparently inflicted less
harm than indigenous militants on Kashmir’s civilian population, even as
92
they did greater damage to Indian security forces. This may have resulted
in part from the new groups’ tactical skill, which enabled them to better
discriminate between targets than the Kashmiri organizations. This, in
turn, had possible implications for how the insurgency was viewed among
the Kashmiri population and could at least help mitigate legitimacy prob-
lems arising from the new groups’ outsider status.
In addition to their Kashmiri operations, the new militant groups took
jihad into the Indian homeland. In December 2000, two LeT militants
entered Delhi’s Mughal-era Red Fort, which was being used to house
army personnel. They killed two Indian soldiers and a civilian sentry prior
to escaping. 33 In December 2001, five militants attacked India’s national
legislature in New Delhi while it was in session. Although the attackers
managed to enter the parliament compound, security forces were able to
repulse them before they could gain access to the parliament house itself.
Had the militants succeeded in doing so, the result could have been a mas-
sacre of India’s legislators. 34
Such attacks made an important statement, demonstrating that the
militants’ capabilities did not end at the borders of Kashmir, but could
reach the very heart of Indian power. The Red Fort operation was particu-
larly rich with symbolism, since this had been the seat of Mughal rule in
India before the arrival of the British Raj. The attack on the old fortress
suggested that history might repeat itself, and that the Muslims could rule
India once again. The new organizations capable of such audacious opera-
tions quickly assumed the role of Pakistan’s primary proxy forces. 35
LeT was the Pakistanis’ favorite of the emerging mehmaan mujahideen
groups during the mid-1990s. By the end of the decade, it had solidified
its leading position, and LeT’s dominance continued in the months and
years following the attacks of September 11, 2001. 36 This was the case for a
number of reasons. First, LeT continued to be the most trusted of the mil-
itant organizations. The Pakistanis were confident that, for all of its feroc-
ity, it would not attack them, even if it disagreed with their policies. The
Pakistanis could not say the same thing of the other groups. For example,
Jaish-e-Mohammed had turned on Pakistan in a number of instances,
including plots to assassinate President Musharraf. Second, LeT’s tacti-
cal abilities were unmatched by the other groups. By the late 1990s, they
were widely viewed as the most capable of the militant organizations at
large in Kashmir, even by members of the Indian security forces. Third,
despite some personality-driven internal conflict, LeT was generally not
riven by ideological or policy disputes. Thus, they resisted splintering in
the manner of other militant organizations such as Jaish-e-Mohammed,
Harakat-u l-Mujahideen, and Lashkar-e-Jhangvi. Finally, LeT’s opera-
tions were audacious, drawing attention to their cause, embarrassing
93
the Indians, and providing encouragement to militants in South Asia

and beyond. 37 Perhaps the most spectacular of these operations was the
assault on Mumbai in 2008. The Mumbai attack encapsulated the par-
ticular strengths of LeT, as well as the broader advantages of Pakistan’s
militant strategy.
The operation was executed by ten militants whom LeT had trained
in small-u nit tactics, weapons handling, physical fitness, and seaman-
ship in a number of locations in Pakistan and Pakistani Kashmir. The
recruits also received extensive religious indoctrination. Their train-
ers and handlers were mostly retired members of the Pakistani Army
and ISI. The group traveled in a Pakistani boat from Karachi into the
Arabian Sea, where they seized an Indian fishing trawler. They sailed
the trawler to a point just off the Indian coast, where they disembarked
and loaded into an inflatable boat for the rest of the journey to shore.
Once they reached the Mumbai coast, the group split into five teams of
two. Four teams walked into the street and took cabs to their targets.
These were the Chhatrapati Shivaji railway station, the Taj Mahal Hotel,
the Jewish cultural center known as Chabad House, and the Leopold
Restaurant and Cafe. A fifth team took the boat to the Trident-Oberoi
Hotel. 38
The militants succeeded in inflicting considerable carnage across the
city. At the railway station, police personnel armed with 1950s-era bolt-
action rifles cowered as the attackers spent nearly ninety minutes pump-
ing machine-g un bullets into crowds of commuters. When the chief of
the Mumbai police’s antiterrorism unit arrived on the scene, the militants
shot him dead and commandeered his car. At the Leopold Cafe, the mili-
tants fired and threw grenades at customers. At Chabad House, they mur-
dered the rabbi and his wife who managed the property, along with two
guests. In the hotels, the attackers set fires, killed guests indiscriminately,
and took hostages. It was nearly three days before Indian security forces,
some of whom had to be called from as far away as Delhi, finally managed
to kill or capture all of the attackers. By this time over 160 people were
dead, and some of Mumbai’s most prominent landmarks were smoking
ruins. 39
The Indians did not retaliate against Pakistan in reaction to the Mumbai
assaults, or to the long litany of other attacks on Kashmir and India proper
by Pakistan’s militant proxies. They came closest to doing so following
the attack on the Indian parliament in December 2001. After the attack,
which the Indians blamed on Lashkar-e-Taiba and Jaish-e-Mohammed,
India launched Operation Parakram, deploying five hundred thousand
troops along the Line of Control and the international border. India
demanded that Pakistan surrender twenty criminals allegedly hiding in
94
Pakistan, forswear its support for terrorism, close militant training camps
in Pakistani territory, and prevent militant infiltration from Pakistan
into Jammu and Kashmir. In the event of Pakistani noncompliance, the
Indians planned to strike militant camps and seize territory in Pakistani
Kashmir. Pakistan reacted with similar force deployments, resulting in
a standoff between approximately one million troops along the Line of
Control and the international border, where they remained in place for
the following months.40 The crisis escalated further when LeT militants
killed thirty-t wo people at an Indian army camp at Kaluchak in Jammu in
May 2002. Indian leaders responded by expanding their planned retalia-
tion. They now envisioned attacking Pakistan with three strike corps and
capturing territory in the Thar Desert. Despite their costly ten-month
military deployment and extensive retaliatory plans, the Indians ulti-
mately decided not to attack Pakistan, and they withdrew their forces in
October 2002.41
The Indians did, however, apply significant diplomatic pressure on
Pakistan to rein in groups like LeT. The Pakistanis at times seemed as if
they were prepared to cooperate. For example, in January 2002, President
Musharraf outlawed Lashkar-e-Taiba and Jaish-e-Mohammed and pub-
licly promised to prevent Pakistani territory from being used to foment
terrorism in Kashmir. The Pakistanis also convinced US officials that
they were seriously considering the handover to India of non-Pakistanis
on India’s twenty-person extradition list.42
In the end, however, the Pakistanis took few, if any, concrete steps to
honor these pledges. Indeed, they often behaved in ways that were directly
opposed to their publicly stated goals. For example, the ISI helped LeT
to insulate itself from Pakistan’s official crackdown, through a range of
measures including name changes, the relocation of financial assets, and
the calibration of its activities to an intensity level that was not overly
provocative but nonetheless demonstrated the insurgency’s continued
viability, as well as Pakistan’s ongoing commitment to sustaining it. As
a result, Lashkar was able to remain active despite Musharraf ’s putative
ban in January 2002, and just a few months later it was implicated in the
massacre at the Indian Army camp at Kaluchak.43 To defuse the military
standoff and to prevent potential Indian punitive action in the wake of the
Kaluchak attack, President Musharraf promised US Deputy Secretary of
State Richard Armitage to end permanently all militant infiltration into
Indian Kashmir. Armitage’s communication of Musharraf ’s pledge to
Indian leaders helped to convince them that the Pakistanis had changed
their minds and were now serious about reining in their proxy forces. This
belief played an important role in the Indians’ eventual decision not to
attack Pakistan in 2002.44
95
In the following months and years, however, it became clear that the
Pakistanis were not going to end militant infiltration into Indian Kashmir.
Infiltration rates did slow, but they nonetheless remained significant. For
example, the Home Ministry estimated that over twenty-four hundred
militants entered Jammu and Kashmir in 2001, with over fifteen hundred
entering in 2002 and over thirteen hundred entering in 2003.45 In addi-
tion, the Indian government directly implicated Lashkar in numerous
attacks within India. Pakistan either ignored or denied the charges.46 The
ISI did close or move militant training camps in Kashmir and restrict the
activities of a number of groups. Lashkar-e-Taiba, however, was allowed
to continue operating openly. Within Pakistan, it regularly held large ral-
lies at its headquarters at Muridke on the outskirts of Lahore. Officially,
Muridke is home to Jamaat-ud-Dawa (JuD), a charitable organization
dedicated to such projects as preaching, promoting religious education,
assisting widows and orphans, and engaging in disaster assistance. The
sprawling Muridke complex houses schools, farming tracts, a mosque,
shops, and medical facilities. As Stephen Tankel puts it, “Muridke’s pal-
pable sense of ambition in terms of scale is unrivalled in the world of
Pakistani jihadi organizations.”47 Although JuD leaders have vehemently
denied any connection between their organization and Lashkar-e-Taiba,
the United Nations identified Jamaat-ud-Dawa as an LeT front and out-
lawed the JuD in 2008. This hampered some of JuD’s later relief opera-
tions, forcing it to assume different names, such as Falah-e-Insaniyat.48
Muridke, however, continued to remain open, even after officially hav-
ing been taken over by the Punjab government in the wake of the 2008
attacks on Mumbai.49
Although overwhelming evidence tied LeT to the 2008 Mumbai
attacks, Pakistani authorities continued to decline to move decisively
against the group. 50 The Pakistanis initially denied that any connection
existed between Pakistan and the Mumbai assault. Indian authorities sub-
sequently offered them incontrovertible proof of such a link, from sources
that included the contact lists and call logs of the attackers’ cell phones,
recorded telephone conversations between Pakistan-based handlers and
the militants during the attacks, and the confession of the one surviving
member of the Mumbai assault team.
Faced with this evidence, the Pakistanis finally acknowledged the
Pakistan–Mumbai connection and placed Hafiz Mohammad Saeed,
founder of LeT and leader of Jamaat-ud-Dawa, under house arrest. He
was released for want of evidence, however. Soon thereafter, a Lahore
court dismissed terrorism charges against Saeed. 51 At present Saeed
lives openly in Lahore, in a compound that includes a house, office, and
mosque. He is protected by personal security forces and by the Pakistani
96
police. Despite his long-standing association with LeT, as well as a US

offer of $10 million for information leading to his arrest and conviction
on terrorism charges, Saeed is essentially a free man. “I move about like
an ordinary person,” he said. “That’s my style.”52 JuD’s Muridke head-
quarters continues to serve as a hub for fundraising, education, charity
work, and religious indoctrination. 53 And although it has avoided high-
profile operations in recent years, LeT remains active. Indeed, it is widely
viewed as the most important militant group in South Asia. 54 Despite
publicly disavowing militancy, then, the Pakistani government has con-
sistently refused to abandon its militant allies. It has thereby ensured
that, more than six decades after partition, the territorial division of
India remains violently contested.
SHAPING POST-S OVIET AFGHANISTAN
Pakistan’s management of the Afghan anti-Soviet jihad during the 1980s

had been designed in no small part to advance its Kashmir agenda.
Nonetheless, the Pakistanis viewed Afghanistan as being important inde-
pendent of Kashmir. They therefore began working actively to shape the
Afghan strategic environment in the wake of the Soviet Union’s defeat.
The Pakistanis employed essentially the same strategy as they did in
Kashmir, using Islamist militants to promote their interests without sub-
jecting Pakistani forces to the costs and risks of direct conflict.
Following the Soviet Union’s withdrawal from Afghanistan in 1989,
the country slipped into competition for power between various warlords
and ethnic factions. The Pakistanis were to become deeply involved in
these events. Initially, however, Pakistani leaders evinced only a pass-
ing interest in the outcome of Afghanistan’s power struggle. Their atten-
tion was largely consumed by events in Kashmir, where the anti-Indian
insurgency had recently erupted. The Pakistanis wished to gain as much
advantage as possible from the Kashmir uprising and were content to
leave Afghanistan largely to its own devices in the meantime. In addition,
the Pakistanis believed that there was probably not a great deal at stake for
them in the Afghan situation, as any of its likely outcomes would be favor-
able to Pakistani interests. Most of the warlords jockeying for control had
connections with and an affinity for Pakistan, given Pakistan’s support for
the resistance during the anti-Soviet struggle, as well as the fact that many
of the Afghan leaders had property and family in Pakistan. The Pakistanis
therefore sought to maintain good relationships with a number of Afghan
leaders, including Gulbuddin Hekmatyar, Abdul Rashid Dostum, and
Ismael Khan. 55
97
The Pakistanis were not friendly with all of the Afghan warlords,
however. One significant exception was the Tajik military commander
Ahmed Shah Massoud. Massoud had received intermittent support from
Pakistan during the anti-Soviet conflict. The Pakistanis had come to
believe that he was not truly dedicated to fighting the Soviets. For exam-
ple, Massoud had never cut off the Soviets’ main supply line in his native
Panjshir Valley. Even more egregiously, in 1983, Massoud had struck a
truce with Soviet forces, agreeing to allow them to operate unmolested
in the South Panjshir in return for a promise not to attack his forces else-
where in the valley. Finally, Massoud had long been a fierce opponent of
Gulbuddin Hekmatyar, who was a Pakistani favorite. Years after the anti-
Soviet war had ended, the Pakistanis still harbored suspicions regarding
Massoud’s motives and loyalties. Consequently, Massoud’s relations with
Pakistan remained fraught. This, in turn, narrowed the range of possible
allies for Pakistan in Afghanistan. 56
The spectrum of potential Afghan allies was important to the Pakistanis
because, despite their current preoccupation with Kashmir, their interests
in Afghanistan were too strong for them to remain agnostic regarding the
outcome of the Afghan power struggle. Pakistani leaders wanted to estab-
lish a reliable route through Afghanistan to the markets and resources of
Central Asia. They also needed to repatriate large numbers of Afghan ref-
ugees still in Pakistan after the end of the anti-Soviet war. In addition, the
Pakistanis hoped that a friendly Afghanistan would provide them with
increased strategic depth. An Afghan government well disposed toward
Pakistan could limit India’s presence in the country, reducing the likeli-
hood of Pakistan’s encirclement by Indian forces on both its western and
its eastern frontiers. In addition, a cooperative Afghanistan could make
its territory available to the Pakistanis in case they needed to fall back in
the West during a war with India. Finally, an Afghan government sympa-
thetic to Pakistan could permit the training of militants in Afghanistan
to wage the Kashmir jihad. Given the strength of these interests, it was
only a matter of time before the Pakistanis turned their attention back to
Afghanistan and attempted to affect political outcomes there. In doing so,
they hoped to help install a regime that would ensure the country’s stabil-
ity and that would be as congenial as possible to Pakistan. 57 Therefore,
the question of available Afghan allies was an essential one for Pakistani
leaders.
Despite efforts to maintain good relations with a number of Afghan
warlords, the Pakistanis preferred Gulbuddin Hekmatyar. This had been
the case since the 1980s when, impressed with Hekmatyar’s discipline,
ruthlessness, and battlefield successes, the Pakistanis had considered
him their most reliable ally in the anti-Soviet jihad. Now, years after the
98
Soviet withdrawal, they continued to hold that view. The Pakistanis also
believed that, as an ethnic Pashtun and the product of Pakistani military
training during the early 1970s, he would serve as a pliant proxy in the
future. Subsequent events demonstrated to the Pakistanis that their faith
was misplaced, however. Despite his past successes, Hekmatyar proved to
be a less-than-effective leader during the post-Soviet era. He was widely
unpopular with his fellow Pashtuns, suffered a number of important mili-
tary failures, and was unable to deliver the Pakistanis their long-standing
goal in Afghanistan—a stable country with a government friendly to
Pakistan and its interests. By 1994, it became clear that the Pakistanis
would have to look elsewhere for an Afghan proxy. 58 They eventually
decided on the Taliban.
The Pakistanis found the Taliban appealing for a number of reasons.
First, the Taliban were ethnic Pashtuns. Pashtuns make up only about
15 percent of Pakistan’s population, but they are heavily concentrated
in the country’s Northwest, along the Pakistan–A fghanistan border.
This had long been a problem for Pakistan, as Pashtun nationalists liv-
ing in Pakistani territory had sought either to establish an independent
region or to join the area to Afghanistan. Governments of Afghanistan,
where Pashtuns account for 40 percent of the population, had tradition-
ally refused to recognize the Durand Line that demarcated the Afghan–
Pakistan border. Instead, they argued that Pashtuns in the area should
be able to choose either independence or union with Afghanistan. The
Pakistanis expected that a friendly Pashtun government in Kabul would
be sensitive to their interests in these matters, recognizing the Durand
Line as the legitimate Afghan–Pakistan border and curbing potentially
destabilizing Pashtun separatism. In addition, the Pakistanis simply
trusted the Pashtuns more than other Afghan ethnic groups. Pashtuns
made up approximately 20 percent of the Pakistan Army and had a strong
lobby in the military and ISI. The Pakistanis thus became convinced that
only the Pashtuns would promote their interests in Afghanistan. 59
The Taliban’s second major advantage was that Benazir Bhutto’s gov-
ernment already had a relationship with them. In 1993, the political
party Jamaat Ulema-e-Islam (JUI) had aligned with Bhutto’s Pakistan
People’s Party (PPP). The JUI then developed a close association with
General Naseerullah Babar, Bhutto’s minister of the interior. Also, one of
the group’s principal leaders, Maulana Fazlur Rahman, became chair of
the Standing Committee on Foreign Affairs of the Pakistani parliament.
During the Soviet occupation of Afghanistan, the JUI had set up a net-
work of madrassas, or religious schools, in Pashtun areas of the North-
West Frontier Province (NWFP) and Balochistan. The Taliban, which
literally means “students,” emerged largely from these madrassas in 1994.
99
Thus, from their inception, the Taliban enjoyed close ties to the Pakistan
government.60
Finally, the Pakistanis were impressed by the Taliban’s success on the
battlefield. Other characteristics aside, they needed an ally that could
defeat competing forces and impose its will on the country. Indeed, mili-
tary ineffectiveness is what led the Pakistanis to abandon Gulbuddin
Hekmatyar, despite his numerous attractions. In October 1994, the
Taliban seized the border town of Spin Boldak, defeating Hekmatyar’s
forces. Then in November they captured Kandahar, Afghanistan’s second-
largest city. Pakistani leaders were surprised at the speed and extent of
their victory. General Babar now referred to them as “our boys.”61 The
Pakistanis were further impressed when the Taliban captured Herat in
late 1995, giving them control of western Afghanistan and demonstrating
that they could operate effectively beyond the Pashtun South. It was now
clear that the Taliban would be both a potent military force at the national
level and a potential long-term ally for Pakistan.
As with Kashmiri militant groups, the Pakistanis gave the Taliban
extensive assistance. For example, General Babar created the Afghan
Trade and Development Cell in the interior ministry to provide the
Taliban with logistical support, Pakistan’s Public Works Department
and Water Development Authority repaired roads and supplied elec-
tricity in Kandahar, Frontier Corps paramilitary forces helped to con-
struct communications networks for Taliban commanders, and Pakistan
International Airlines and the Civil Aviation Authority assisted with the
repair of Kandahar airport and Taliban military aircraft. Other Pakistani
assistance included the recruitment and training of Taliban personnel,
intelligence and combat advisory support, and direct military action
such as cross-border artillery fire in conjunction with Taliban operations.
Pakistan also provided the Taliban with crucial diplomatic support and
in 1997 was the first state to recognize it as the legitimate government of
Afghanistan. This wide-ranging assistance was essential to the Taliban’s
success. Their ascendance to power, which was complete by 2000, could
not have occurred without Pakistani backing.62
Despite their debt to Pakistan, the Taliban exhibited considerable inde-
pendence upon taking power, diverging from Islamabad’s preferences in
important areas. For example, as noted earlier, the Pakistanis had hoped
that the Taliban would honor their wishes regarding the Afghan–Pakistan
border dispute. The Taliban was intransigent on this issue, however,
refusing to formalize the border in accordance with Pakistani desires.
Nonetheless, the Taliban cooperated with the Pakistanis on a number of
other important matters, including the establishment of militant training
camps in Afghanistan.
100
During the early 1990s, the United States had nearly classified Pakistan
as a state sponsor of terrorism. The Pakistanis had therefore come under
intense pressure to demonstrate that they were taking steps to reduce
anti-Indian militancy. To create the appearance of compliance, they
moved militant training camps out of Pakistani territory and into eastern
Afghanistan, where the Taliban allowed them to operate. In Pakistan’s
view, the Taliban’s support for the Kashmir struggle outweighed its non-
compliance on other Afghan-related matters such as the Durand Line,
and in fact made it an indispensable ally. Despite Afghanistan’s strate-
gic import, it was second for the Pakistanis in significance to Kashmir.
Indeed, the Pakistanis saw the creation of a favorable strategic environ-
ment in Afghanistan as a means of achieving its goals in Kashmir, much as
General Zia had done during the anti-Soviet war. As Ahmed Rashid puts
it, Kashmir “became the prime mover behind Pakistan’s Afghan policy
and its support to the Taliban.” Pakistan would “deny [the Taliban] noth-
ing, as long as they provided bases for Kashmiri and Pakistani militants.”63
The Taliban thus served as what Larry Goodson calls a “proxy army in
Afghanistan,” helping Pakistan to install a friendly government in Kabul
that increased the country’s stability and promoted Pakistan’s central stra-
tegic interests.64 Whatever the Taliban’s other problems, the Pakistanis
therefore considered them allies who were deserving of extensive ongo-
ing assistance. As a senior Pakistani diplomat characterized his country’s
position regarding the Taliban, “we will support whoever can bring stabil-
ity to Afghanistan. If they are angels, nothing like it. And if they are devils,
we don’t mind.”65
Pakistan joined the US-led coalition to combat global terrorism follow-
ing the September 11, 2001, attacks on New York and Washington. Pakistani
president Pervez Musharraf realized that participation in this effort would
likely require the Pakistanis to help the United States fight against their
Taliban allies. In the wake of 9/11, the United States had offered the
Pakistanis a stark choice: stand with the United States and obtain the ben-
efits of its support, or stand against it and suffer the consequences of its anger.
Upon reflection, Musharraf decided that Pakistan’s only option was to coop-
erate with the United States. The Taliban was an important Pakistani ally,
but it was not worth incurring the wrath of the United States to protect it. As
Musharraf writes, “The ultimate question that confronted me was whether it
was in our national interest to destroy ourselves for the Taliban. Were they
worth committing suicide over? The answer was a resounding no.”66
Although Pakistan appeared to face an “either-or” choice in Afghanistan
after September 11, 2001, its options proved not to be as binary as they
initially seemed. The Pakistanis ultimately managed to have it both ways
in Afghanistan; they joined the US-led antiterror coalition but avoided
101
wholly abandoning their erstwhile allies. In truth, turning completely on

the Taliban was never a serious option for Pakistan, regardless of US pres-
sure or of Pakistan’s purported role in the global war on terror. Pakistan
had in no way altered its Afghan agenda in the wake of 9/11. As before, the
Pakistanis desired a friendly regime in Kabul to facilitate their acquisi-
tion of strategic depth, access to Central Asia, training of militants, and
favorable resolution of border issues. The Pakistanis still believed that the
Taliban fit this description better than any of Afghanistan’s other con-
tenders for power. Therefore, the Pakistanis continued to protect and sup-
port the Taliban, even as they joined the US-led war on terror.
For example, the Pakistanis helped to prevent the United States from
eradicating the Taliban immediately following September 11, 2001. In
November 2001, thousands of Taliban forces were trapped in the north-
eastern city of Kunduz, along with hundreds of Pakistan army and intel-
ligence personnel. President Musharraf convinced President Bush to
permit the evacuation of the Pakistani personnel, arguing that allow-
ing them to be captured or killed would seriously damage Musharraf ’s
domestic political standing. In an effort to protect Musharraf, Bush
agreed. The Pakistanis did not limit themselves to evacuating their own
personnel, however. They also evacuated Taliban commanders and fight-
ers in a series of airlifts from Kunduz into Pakistan. As Ahmed Rashid
puts it, “What was sold as a minor extraction” of Pakistani security per-
sonnel “turned into a major air bridge” for the Taliban. This helped to
ensure that, despite the US onslaught against them, the Taliban would
live to fight another day. Also, it demonstrated to the Pakistanis that
they could hoodwink the United States, purportedly supporting the US
campaign against the Taliban while continuing to aid and protect their
Taliban allies.67 Not surprisingly, this is exactly the approach that they
adopted moving forward; the Pakistanis have continued to provide the
Taliban with extensive financial, logistical, and intelligence support to
the present day.68
The Taliban was not the only militant group that the Pakistanis utilized
to promote their interests in Afghanistan. They also supported associ-
ated organizations such as the Haqqani network. The network’s founder,
Jallaludin Haqqani, had risen to prominence as a mujahideen commander
during the war against the Soviet Union. Haqqani had helped to recruit
and train Osama bin Laden and remained close to al-Qaeda in subsequent
years. During this period, he also forged close ties to Pakistan and the
ISI. He served as minister of tribal affairs in the Taliban government and
then relocated to Pakistan’s tribal regions after the Taliban’s overthrow
in 2001. From there he has directed a militant network dedicated to
fighting coalition forces in Afghanistan. They have been implicated in a
102
number of high-profile operations in Afghanistan, including attacks on

the Indian and US embassies in Kabul. The Haqqanis also have supported
the Taliban in critical ways, such as assisting in the Taliban’s capture of
Kabul in September 1996. In addition, they provide Taliban fighters shel-
ter from coalition forces on the Pakistani side of the Afghan–Pakistan
border. Although they are allies, the Haqqanis are not part of the Taliban;
they remain an independent organization.69
The Haqqani network promotes Pakistani interests in a number of
ways. For example, it helps to increase Pakistan’s strategic depth in
Afghanistan, to augment Pakistani military capabilities in Afghan and
Pakistani tribal areas, and to facilitate Pakistan’s efforts to manage its
diplomatic relations with the tribal regions. In return, the Haqqanis are
allowed sanctuary in Pakistani territory; receive operational support,
including materiel, assistance with logistics, and training from the ISI;
and are permitted to pursue their business interests without interference
from the Pakistani authorities. Thus, as Brown and Rassler argue, “the
Haqqani network has historically functioned as a proxy through which
elements of the Pakistani state have pursued their strategic interests and
sought military and political influence inside the FATA and Afghanistan.”
Today it “remains a strategic, state asset.70
Pakistan, then, has provided the Taliban, as well as associated groups
like the Haqqanis, with a wide range of critical assistance, which has
enabled them to survive over a decade of US-led war intended to defeat
them. Meanwhile, Pakistan has accepted billions of dollars in US aid
in return for its purported cooperation in the war effort.71 Pakistani
support for the Taliban and Haqqanis has been so extensive that Joint
Chiefs of Staff Chairman Admiral Michael Mullen publicly charac-
terized them as “proxies of the government of Pakistan” that were
“hampering efforts to improve security in Afghanistan” and “spoiling
possibilities for broader reconciliation.” 72 General Joe Dunford, com-
mander of the International Security and Assistance Force and of US
forces in Afghanistan, described the Haqqani network as “the greatest
risk” to coalition forces and “perhaps the greatest risk to the [Afghan]
campaign.” 73
Pakistan’s Afghan strategy has thus closely mirrored its strategy in
Kashmir, using militant proxies to promote its security interests while
avoiding direct military intervention. Indeed, the strategies were so simi-
lar, and so deeply intertwined, that they could actually be understood as
two parts of the same policy. Although Pakistan had a number of impor-
tant interests in Afghanistan, the most important were those that affected
its standing Kashmir. And it pursued these interests in essentially the
same way in both places.
103
EVALUATING PAKISTAN’S STRATEGY
Pakistani strategy during the Kashmir insurgency and in post-Soviet

Afghanistan highlighted many of the themes that I have discussed
throughout the book. First, during this time period, Pakistan’s militant
strategy continued its trend of becoming both more sophisticated and
more important to Pakistani security policy. This increase in sophistica-
tion and importance was particularly evident in Kashmir. The Pakistanis’
1965 effort to capture Kashmir had been a significant improvement over
Pakistan’s use of militants during the first Kashmir war, utilizing fight-
ers who were better organized, trained, and equipped than the tribesmen
of 1947, and making better use of religious tropes to motivate them. The
Kashmir insurgency of the 1980s and ’90s was a further improvement
over the 1965 conflict. Where the Pakistanis had spent months prepar-
ing for Operation Gibraltar, they spent years laying and maintaining
the groundwork for the Kashmir insurgency. They amassed resources
diverted from the Afghan war, gained expertise in running an insurgency
from the Afghan conflict, monitored and managed insurgent groups to
ensure their willingness and ability to promote Pakistani interests, and
helped the militants to create domestic infrastructure that made them
largely self-sustaining. This was far more extensive than anything the
Pakistanis had done to support militancy in the past.
In Afghanistan, Pakistani efforts were somewhat more modest. They
did not build the Taliban “from the ground up” in the manner of the
Kashmiri militant groups. In addition, Pakistan’s efforts in Afghanistan
during the 1980s had already been sophisticated and important, contrib-
uting to the defeat and withdrawal from the country of a superpower.
Nonetheless, the quality and significance of Pakistan’s efforts were high
in the post-Soviet era. The Pakistanis made the Taliban’s rise to power
possible by providing them with extensive financial, military, and politi-
cal support. In doing so, they used their militant strategy not just to defeat
an enemy in Afghanistan, but also to help install a friendly government in
that country. Support for the Taliban thus became the main thrust of the
Pakistanis’ Afghan policy, which, in turn, became a central pillar of their
overall security policy; it provided opportunities and mitigated dangers
on Pakistan’s western flank and also helped to advance Pakistani interests
in Kashmir.
Second, these cases demonstrated that the use of militants, under
the correct structural conditions, could be successful. The failure of
Pakistan’s militant strategy in Kashmir in decades past resulted to a
significant degree from the fact that the strategy was employed at the
wrong time and Kashmiris were not prepared to take up arms against
104
the Indian government. This was the result of political miscalculation,

rather than an inherent problem with the use of militants. When condi-
tions within Kashmir were ripe, and the Kashmiri population was suf-
ficiently disenchanted with India to support violent political change,
a militant strategy proved not only to be viable but also to be highly
effective.
By “effective,” I do not mean that support for the insurgency enabled the
Pakistanis to achieve their foremost strategic goal of wresting Jammu and
Kashmir from Indian control and joining it to Pakistan. Pakistani strat-
egy of course did not do that. The strategy did, however, allow Pakistan to
achieve a number of other important ends, such as significantly attriting
Indian military and economic resources; tarnishing India’s international
image as a liberal, law-abiding state; and keeping the Kashmir dispute
on the agenda of the world community—a ll without fighting a major
war against India. These were not trivial achievements, and they must be
taken into account when assessing the effectiveness of Pakistani strategy.
In Afghanistan, the Pakistanis had not previously experienced fail-
ure. Their use of the mujahideen during the 1980s to eject the Soviet
Union from the country had been highly successful. Their subsequent
success in supporting the Taliban lends further credence to my argu-
ment that Pakistan’s previous setbacks in Kashmir did not result from
flaws inherent to a militant proxy strategy. They resulted, rather, from
problems specific to the political context in which the Pakistanis were
operating. In addition, the Taliban’s victory and assumption of power
in Afghanistan held the promise of significant benefits for Pakistan. It
created opportunities such as the potential for easier access to Central
Asia and mitigated dangers such as the presence of unfriendly states on
both its western and eastern borders. And by bringing a friendly regime
to power in Afghanistan, the Taliban’s ascendance enabled Pakistan to
pursue more effectively its long-standing strategic goals in Kashmir.
Third, Pakistan’s strategy benefited significantly from the operational
advantages of using militants. India had great difficulty achieving denial
against militant forces, which were able to penetrate Indian territory with
relative ease. In Kashmir, hundreds of Indian Kashmiris per month could
enter Pakistani territory, receive military training, and then return to
Jammu and Kashmir to attack government targets. This was crucial to the
rebellion’s ability to gain momentum during its early days, and it did not
change much in the coming years. During 2001, for example, after well
over a decade of energetic counterinsurgency efforts, the Indian home
ministry estimated that more than twenty-four hundred militants had
infiltrated into Jammu and Kashmir. This was a rate of better than two
hundred per month.74 The militants also were easily able to enter Indian
105
territory elsewhere. The Mumbai attackers, for example, simply sailed a

boat into Mumbai harbor, walked ashore, and took taxis into the city.
After penetrating Indian territory, the militants blended easily into the
local population. Consequently, they were difficult to identify and appre-
hend prior to commencing their attacks. In Kashmir, this led the Indians
to resort to draconian tactics such as forceful interrogations and cordon-
and-search operations. This damaged the Indians’ relations with the pop-
ulation and further undermined their position in Kashmir, but it did not
stop militant operations. In 2001, for example, militants were responsible
for over forty-five hundred violent incidents in Indian Kashmir, a rate of
over twelve incidents per day.75
Once attacks were underway, the Indians had difficulty defeating
them. The militants had surprise on their side and were small in num-
ber, and thus could not easily be fixed and engaged by Indian defenders.
This enabled the militants to inflict significant damage on their targets
before finally being killed or captured. The Mumbai attacks dramatically
demonstrated this problem. It took the full spectrum of Indian security
forces, ranging from city police to federal special operations units, three
days to defeat ten attackers. During this time the militants killed over
160 people, caused extensive damage to some of Mumbai’s most iconic
locations, and showcased their cause before a rapt international audience.
Less dramatic, but nonetheless extremely damaging, was the militants’
effect on Kashmir, where they were able to kill an average of several peo-
ple per day. The militants’ operational advantages, then, resulted in the
Indians’ failure at all three of a defender’s denial tasks: keeping the mili-
tants from entering Indian territory, finding the militants after they had
entered Indian territory, and defeating ongoing attacks before they could
inflict serious harm on their targets.76
In Afghanistan, the operational environment differed from that in India.
During its initial phase of support for the Taliban, prior to September 11,
2001, Pakistan did not need to infiltrate an outside force into Afghanistan
to pursue its militant strategy. Pakistan’s proxy force was already in the
country and engaged in a multiparty civil war. Pakistan needed only to
support its ongoing operations. After Pakistan joined the war on terror,
however, the operational environment in Afghanistan became more simi-
lar to that of India. In addition to continuing support for Taliban elements
already in Afghanistan, Pakistan hosted groups such as the Haqqani
network on Pakistani territory. From there, Pakistan aided their transit
in and out of Afghanistan to launch attacks against Afghan targets and
coalition forces. This resembled the situation in Kashmir, where Pakistan
helped an outside militant force to infiltrate and launch attacks on the tar-
get territory.
106
As in Kashmir, this strategy made all three basic denial tasks extremely
challenging for defenders in Afghanistan. It was difficult to prevent mili-
tants from penetrating Afghan territory, as they could transit from sanc-
tuaries inside Pakistan across the porous Afghan–Pakistan border in
small groups with relative ease. This enabled them not only to launch
attacks but also to escape back into Pakistan to regroup, recover, and pre-
pare for further operations. This greatly increased the difficulty of defeat-
ing the militants and posed a major challenge to the coalition mission.
As a December 2012 Defense Department report stated, “the availability
of sanctuary inside of Pakistan enables key elements of the insurgency
to remain potent and threatening, including the Afghan Taliban and the
Haqqani Taliban Network.” This constitutes “the most critical opera-
tional threat to the International Security Assistance Force (ISAF) cam-
paign in Afghanistan.”77
Once they were in Afghanistan, militants were difficult to detect and
apprehend, as they blended into the local population, along with Taliban
elements that remained in the country. This gave them the opportunity
to prepare and launch attacks, which they did at the rate of hundreds and
often thousands of incidents per month. For example, the number of insur-
gent attacks during 2010 ranged from approximately 1,250 in January to
nearly 4,500 in August. 78
The militants exploited the element of surprise, knowledge of terrain,
and local support and intelligence networks to ensure that their attacks,
once launched, were difficult to defeat before they inflicted damage
on their targets. Over 1,800 military personnel from the United States
alone died in combat with the militants, and over 19,500 were wounded
between the launch of Operation Enduring Freedom (OEF) in late 2001
and early 2014. Coalition partners lost over 1,000 personnel from OEF’s
launch through 2012. The Afghan National Army and Directorate of
Security lost over 2,300 personnel between 2007 and mid-2012. Senior
commanders viewed this militant violence, and its associated losses, as a
major threat to the success of the ISAF mission. The difficulty of achiev-
ing denial against Pakistan’s militant strategy thus had severe strategic
ramifications for Afghanistan.79
The Pakistanis suffered almost no direct losses or damage as a result of
their militant campaigns in either India or Afghanistan. The Pakistanis did
not use their own forces to launch attacks. Therefore, neither Indian nor
coalition denial efforts directly engaged the Pakistan military. This, in turn,
meant that any attacker losses were restricted primarily to the militants.
In addition, neither Indian nor coalition forces in Afghanistan launched
punitive attacks against the Pakistanis to punish them for their behavior.
107
Why did they not do so? Indian forbearance had multiple sources. As
noted in c hapter 2, to some extent it resulted from India’s and Pakistan’s
shared history and culture, domestic political dangers to India inherent
in attacking a Muslim country, India’s lack of good conventional military
options against Pakistan, and concerns regarding the dangers of nuclear
escalation in the event of an Indo-Pakistani conflict. Despite these imped-
iments, however, India did at times seriously contemplate undertaking
punishment operations against Pakistan. Its decision not to do so turned,
to a significant degree, on the deniability inherent in Pakistan’s militant
proxy strategy.
For example, during the crisis following the 2001 parliament attack,
India planned to launch punitive strikes on Pakistani Kashmir and large-
scale incursions into Pakistan proper. India ultimately refrained from doing
so because Pakistani leaders made a convincing case that they opposed
militant violence and would take steps to limit or end it. Implicit in the
Pakistanis’ case was the claim that the militants were an entity separate
from the Pakistan government, whose interests and actions were distinct
from Islamabad’s. As President Musharraf put it during a televised address
in January 2002, the government viewed the militants as “religious extrem-
ists” who were subverting “Pakistan and Islam.” They were not instru-
ments of Pakistani policy. “We condemn the terrorist acts of September
11, October 1 and December 13,” Musharraf announced. “Anyone found
involved in any terrorist act would be dealt with sternly. Pakistan rejects
and condemns terrorism in all its forms and manifestations. Pakistan will
not allow its territory to be used for any terrorist activity anywhere in the
world. No individual, organization or party will be allowed to break the law
of the land.”80
If, as Musharraf claimed, the militants were in fact separate from and
at odds with the Pakistani state, it implied that the Pakistan government
was not truly responsible for the militants’ attack on the Indian parlia-
ment. This, in turn, suggested that Pakistan should not be punished for
the militants’ behavior, since the government, whatever its past, was not
behind current militant violence such as the parliament attack and would
work to prevent such violence in the future.
The Indians were not naïve regarding Pakistan’s use of militancy; they
fully understood the Pakistanis’ militant proxy strategy and its role in
creating an environment in which events like the parliament attack were
likely to occur. Still, Pakistan’s ability to distance itself from the militants
and their actions in the parliament case mattered. Had Pakistan been
unable to do so, it is unlikely that the United States would have believed
its promises to rein in the militants and urged the Indians to exercise
108
restraint. As I noted earlier, this US pressure played an important role in

convincing India not to attack Pakistan.
In Afghanistan, there is no evidence that the coalition ever seriously
considered punishing Pakistan. This was likely due, at least in part, to the
fact that Pakistan was a putative member of the coalition seeking to de-
Talibanize Afghanistan. For coalition forces to have attacked Pakistan
would therefore have been politically difficult, if not impossible. It also
would have severely hampered coalition operations in Afghanistan, which
relied on Pakistani support for a range of activities including the move-
ment of supplies into the country. This, however, once again points out the
advantages of Pakistani deniability. Pakistan was able to join the coali-
tion, and remain a member, because it could pretend to support coalition
aims and deny working against them. This was a position made possible
by its use of militant proxies, rather than conventional military forces.
Despite these successes in Kashmir and Afghanistan, the news for
Pakistan’s militant strategy was not all good; during this period, problems
began to emerge. The main issue was proper alignment of the militants’
interests and capabilities with the goals of their Pakistani sponsors. This
adverse-selection issue was a significant part of the story of the Kashmir
jihad. The Pakistanis were forced constantly to monitor and evaluate
their proxies, and in a number of cases to switch less useful allies for more
helpful ones. For example, as noted, the Jammu and Kashmir Liberation
Front was helpful in starting the insurgency, but its secular-nationalist
ideology and lack of military capability quickly made it obsolete, forc-
ing the Pakistanis to undermine it and to seek new allies. The Hizb-u l-
Mujahideen shared the Pakistanis’ religious goals and desire to merge
Kashmir with Pakistan; however, it was insufficiently ruthless and was
willing to compromise with India at important junctures.
The agendas of the mehmaan mujahideen and especially Lashkar-
e-Taiba aligned most closely with those of Pakistan. LeT promoted an
Islamist sociopolitical agenda, sought to join Kashmir with Pakistan, was
willing to employ extreme violence in pursuit of its goals, would under no
circumstances compromise with India, and was unwilling to turn directly
against the Pakistani state. LeT therefore was highly effective at advanc-
ing Pakistani interests. Shepherding LeT to this point, however, was a
difficult process, involving costly monitoring and the management of vio-
lent transitions with former allies. In addition, even LeT required careful
Pakistani management. Although its interests were more closely aligned
with Pakistan than were the interests of other militant groups, the overlap
was not perfect. For example, though LeT did not turn against Pakistan,
many of its personnel were deeply unhappy about Pakistani support for
US policies. In addition, LeT launched operations that, while spectacular,
109
created serious problems for Pakistan with India and the broader interna-
tional community. Finally, the group worked with militant organizations
whose interests diverged sharply from those of Pakistan. At times, this
amounted to de facto efforts to undermine Pakistani policies. All of these
issues would give rise to serious principal–agent problems, and badly
damage Pakistani interests, in the future.
The Pakistanis also faced adverse-selection issues with their allies in
Afghanistan. Gulbuddin Hekmatyar’s Hizb-i-Islami, for example, had
been the Pakistanis’ favorite Afghan militant organization during the
anti-Soviet war. In the post-Soviet period, however, Hekmatyar proved to
be politically and militarily ineffective. Such problems would continue for
the Pakistanis in Afghanistan. For example, the Taliban was uncoopera-
tive in areas where its interests did not align with Pakistan’s, including the
harboring of allies repugnant to the international community. This forced
the Pakistanis to assist the United States in its ouster of the Taliban. As
a result, much of the Pakistanis’ effort in facilitating the Taliban’s rise
to power was wasted. In addition, many militants in Afghanistan and
Pakistan began to view the Pakistani government as an enemy and turned
violently against it. This problem has become one of Pakistan’s most sig-
nificant security challenges, potentially threatening the survival of the
Pakistani state.
CONCLUSION
In this chapter I examined Pakistan’s support for the Kashmir insurgency,

and its efforts to shape post-Soviet Afghanistan, from the late 1980s for-
ward. I showed that, in Kashmir, Pakistan’s strategy enabled it to impose
significant military, financial, and political costs on India, and to ensure
that Kashmir’s future remained seriously contested, while avoiding the
dangers of direct Indo-Pakistani conflict. In Afghanistan, by facilitating
the Taliban’s ascension to power, Pakistan was able to gain a friendly west-
ern neighbor, improve its strategic depth, increase its access to Central
Asia, and acquire a training ground for anti-Indian militants, without the
dangers of direct military intervention in Afghanistan. Pakistan contin-
ued to support the Taliban and related organizations even after joining
the US-led coalition dedicated to defeating these groups. These successes
in Kashmir and Afghanistan showed that, in the right political environ-
ment, Pakistan’s militant strategy could be highly effective.
Despite its successes during this period, Pakistan’s militant policy also
began to develop problems, resulting primarily from a divergence of inter-
ests between Pakistan and its proxies. In c hapter 6, I examine these and
110
other emergent downsides of Pakistan’s strategy. I show that the strategy’s

costs and risks have become so severe that they could result in catastro-
phe, in the form of either Islamabad’s loss of sovereignty over Pakistan to
homegrown enemies, crippling domestic development failures, or a major
war with India. Thus, what once was a useful strategy has now become a
major liability for Pakistan. The Pakistanis must abandon it if they are to
avoid disaster.
111
CHAP T ER 6

An Assessment
S ince the end of the British Raj and the founding of the Pakistani state
in 1947, Pakistan has sought to promote its security interests through
the use of Islamist militants. Today, militant proxies are one of the pri-
mary tools, along with conventional and nuclear forces, that Pakistan uses
to produce security for itself. Jihad has become a central component of
Pakistani grand strategy.
How successful has Pakistan’s use of militant proxies actually been?
Has the strategy made Pakistan more secure? Or has it in fact undermined
Pakistani security interests? In this chapter I assess the benefits and the
costs of Pakistan’s militant proxy strategy. In the first section, I argue that,
contrary to many analyses, Pakistan’s use of Islamist militants has not
been an unmitigated disaster. In truth, the strategy has yielded a num-
ber of benefits over the decades. These include the promotion of domestic
political cohesion within Pakistan, partial compensation for Pakistan’s
material imbalance with India, the ability to continue challenging Indian
control of Kashmir, and the protection of Pakistani interests in the larger
South Asia region, particularly in Afghanistan. To recognize these advan-
tages is in no way to endorse Pakistan’s strategy, or to argue that it has
been an optimal means of addressing Pakistan’s myriad geo-strategic
challenges. It is simply to recognize that the truth regarding Pakistan’s
use of militancy is more complicated than it sometimes appears.
In the chapter’s second section, I explore the costs of Pakistan’s mili-
tant strategy. I argue that it has given rise to a number of serious problems
and pathologies, including loss of control of its militant proxies, damaging
developmental trade-offs, and increased tension with India. Some of these
112
problems are so severe that they have begun to endanger the continued
survival of the Pakistani state. Thus, despite its past successes, Pakistan’s
militant strategy has outlived its utility, and Pakistani leaders must move
decisively to abandon it.
THE BENEFITS OF PAKISTAN’S MILITANT

STRATEGY
From its inception, one of Pakistan’s most serious problems was its lack
of a coherent founding narrative. The creation of Pakistan was initially
an elite project; it originated not at the grassroots level, but rather out of
discussion and debate within a community of poets, scholars, and politi-
cal entrepreneurs. As a result, the Pakistan project was reflective mainly
of their interests and aspirations. Those ordinary Muslims who were to
live in the new Pakistani state had much to lose economically, politically,
and socially by leaving India and were often highly ambivalent about the
prospect of doing so. This gave rise to fundamental questions regard-
ing the Pakistan project that remained salient even after partition and
the founding of the Pakistani state: What was the purpose of Pakistan
and why did it need to exist? Without clear answers to these questions,
it would be difficult for the new Pakistani state to generate the domestic
political cohesion necessary for survival. Pakistan’s militant strategy has
helped to ensure that, despite its weak domestic political foundations, the
country has in fact had a plausible reason to exist. This, in turn, has helped
Pakistan to promote internal political unity.
The strategy has done so by offering a practical means of operational-
izing Pakistan’s Muslim identity. Given Pakistan’s lack of firm political
foundations, its early leaders decided to adopt an approach to state build-
ing that emphasized Islam. Although a more inclusive, secular Pakistani
nationalism could have been preferable to religion as an organizing
principle, it was not viewed as a realistic option. As Olivier Roy argues,
secularism created the danger that once Pakistan had been founded and
“the Muslims were all together on the same territory,” they would be left
without an adversary. As a result, disparate groups would revert back to
“an ethnic or linguistic category,” thereby fragmenting the country. Thus,
Roy maintains, “the promotion of a secular Pakistani nationalism proved
impossible.”1 Pakistan would need to become a Muslim state; it could
not simply serve as a secular state for Muslims. This religious foundation
would help to generate critical domestic political cohesion by ensuring
that all Pakistanis, regardless of interest, history, or background, had a
common normative point around which they could rally.2
113
A n Ass e ssm e n t ( 113 )
To a significant degree, Pakistan’s religious identity emphasized not

what the country wanted to create or become, but rather what it sought to
oppose. Pakistani leaders believed that such a negative identity, focused
on opposition to so-called Hindu India, would most effectively unify the
country. As Jean-Luc Racine argues, “nothing is more likely to strengthen
the feeling of Muslim identity, or to reinforce nationalist sentiments, than
fear of the other and the spectre of an Indian threat.”3 Pakistan therefore
became not just a Muslim state that sought to organize its political and
social life according to the requirements of Islam. It became a state that
existed in direct contradistinction to India.
Such an oppositional Pakistan would need continually to confront
its Hindu adversary to effectuate its national identity. Pakistan could
do so by rejecting the territorial status quo on the Indian subcontinent
and seeking to “liberate” Muslim-majority Jammu and Kashmir from
Indian rule. As Racine points out, this project “conjures up the prospect
of a secessionist move by a Muslim population under India’s control” and
thus is “extremely useful” in promoting Pakistan’s oppositional identity.
It also, more generally, gives Pakistanis a sense of participation in a larger
collective effort. By enabling Pakistan to struggle in a visible and dramatic
manner against a regional status quo that is seen as favoring India, it cre-
ates a common national project in which all Pakistanis share. Pakistanis’
sense of participation in this joint project has been real and widespread.
The use of militant proxies has facilitated Pakistan’s pursuit of these
goals, allowing it to confront India repeatedly over Kashmir without suf-
fering catastrophic defeat, despite Pakistan’s significant material inferior-
ity vis-à-v is India. Pakistan’s militant strategy has thus played a role in
creating what Christophe Jaffrelot describes as a “strong nationalism.”
Strong nationalism, in turn, has helped Pakistan to surmount the host of
severe ideological and material challenges that it has encountered since
independence. In the face of repeated war and crisis, many have doubted
that Pakistan could endure; experts’ predictions of doom were frequent
and widespread.4 Yet, in every case, Pakistan has managed to survive. As
Ian Talbot writes, reports of its demise “have been greatly exaggerated.”5
This has resulted, in no small measure, from the high degree of loyalty
that Pakistanis have demonstrated to the state.6 This loyalty rests, to a
significant extent, on shared opposition to India. As a senior Pakistani
military officer put it, “Pakistanis may disagree on everything else. The
nation is beset by ethnic divisions, religious differences, and political dis-
agreements. But if there is one thing on which one can always generate
agreement, it is opposition to India. Every Pakistani will agree on it. This
is the glue that we can count on.” 7 By making this opposition possible,
Pakistan’s militant proxy strategy has helped the country to cohere.
114
To recognize the domestic political logic of Pakistan’s militant strategy

is in no way to endorse its use in Pakistan’s state-building project, which
has been characterized by major shortcomings. For example, the strategy
has not offered Pakistanis any aspirational vision for the future or means of
material self-improvement. Given Pakistan’s failure to provide its citizens
with basic economic, educational, and security-related public goods, both
are desperately needed. Nor has the strategy prevented the occurrence
of severe social fragmentation, including ethnic strife that led to the loss
of East Pakistan, and ongoing violence between different Muslim sects
within the country. Indeed, government efforts to impose religious uni-
formity on Pakistan, to which the militant strategy is closely linked, have
directly promoted both ethnic and sectarian strife.8 Nonetheless, the fact
remains that Pakistan’s strategy also has helped Pakistan to define itself,
promoting an oppositional Muslim identity and sense of common cause
that have provided a basis for national unity in the face of tremendous
social diversity. Thus, the domestic political impact of Pakistan’s strategy
is more complicated than it might initially seem. The strategy’s impact on
state building has obviously been far from ideal. Yet it has yielded some
benefits, despite its serious failures and shortcomings.
The second major benefit of Pakistan’s militant strategy has been its
ability to help Pakistan to compensate for its material imbalance vis-à-v is
India. At its birth, the new Pakistani state suffered from significant mate-
rial disadvantages relative to India. Geographically, economically, and
militarily, Pakistan was by far the weaker party in the Indo-Pakistani rela-
tionship. Since then, these material inequalities have endured, and today
Pakistan remains considerably weaker than India. Nonetheless, Pakistan’s
militant strategy has helped to mitigate the effects of this imbalance, tak-
ing a significant toll on India over the decades by attriting its resources,
tying down its assets, and damaging it politically.
The extent of the damage that Pakistan has wrought is evident from the
costs to India of the Kashmir insurgency alone. From 1988 to 2010, India
lost approximately six thousand security-force personnel in Kashmir,
about twice the number of battle deaths as India suffered in either the
first or second Kashmir war, or in the Bangladesh conflict.9 India also lost
fifteen thousand civilians during this period in Jammu and Kashmir, far
exceeding civilian losses in any of its other conflicts.10 In addition, the
insurgency has inflicted major opportunity costs on India. For example,
the Indians currently deploy approximately four hundred thousand per-
sonnel to maintain security in Kashmir.11 This presence requires not only
manpower but also large amounts of money that New Delhi could other-
wise spend elsewhere. Finally, the insurgency has tarnished India’s image.
The tactics that New Delhi has employed to combat the rebels, including
115
kidnapping, torture, and extrajudicial killings, have undermined India’s

standing as a democratic state respectful of human rights and the rule of
law.12 None of these successes of Pakistani strategy have changed the fact
that India remains economically and militarily superior to, and physically
much larger than, Pakistan. As a result, Pakistan continues to suffer from
an acute sense of insecurity vis-à-v is India. Nonetheless, the costs that
the Pakistanis have managed to inflict on the Indians through the use of
militants are not trivial. They have at least partially blunted India’s mate-
rial advantages, making them somewhat less dangerous than they might
otherwise be.
Third, Pakistan’s militant strategy has enabled it to pose an ongoing
challenge to Indian control of Kashmir. Freeing Kashmir from India and
joining it to Pakistan has been one of Pakistan’s central strategic impera-
tives since 1947. Despite its best efforts, Pakistan has not succeeded in
this quest. Jammu and Kashmir remain firmly in Indian control, a situa-
tion that seems unlikely to change in the foreseeable future. Pakistan has,
however, managed to ensure that the territory remains contested, both
militarily and diplomatically. At the military level, India has continually
been forced to fight to retain control of the region. At the diplomatic level,
the Kashmir dispute remains on the agenda of the international commu-
nity, despite India’s claim that it is a solely bilateral issue.13 These accom-
plishments are important. They ensure that although Pakistan has not
prevailed in Kashmir, its cause is not entirely lost. As long as the interna-
tional community remains interested in the Kashmir issue and the rebel-
lion continues to survive, at least a small possibility exists that someday
Pakistan will succeed in its quest. Such an outcome is highly unlikely.
Thanks largely to the militant strategy, however, it is not impossible.
Finally, in Afghanistan, Pakistan’s strategy helped to protect Pakistan
from the dangers of the Soviet invasion, attriting Soviet forces and has-
tening their departure without provoking them to the point of large-scale
retaliation or direct conflict. In addition, the strategy enabled Pakistan
to secure military and financial resources for later use against India in
Kashmir. Subsequently, Pakistan’s militant strategy facilitated a friendly
Taliban government’s ascent to power in Afghanistan. This provided
the Pakistanis with a number of significant benefits, including increased
strategic depth, improved lines of communication to central Asia, and
access to additional militant training grounds. The strategy has also
helped to ensure that, even well after its ouster following the US invasion
of Afghanistan, the Taliban remains a viable entity, continuing to battle
coalition forces for control of the country.14 Meanwhile, Pakistan has
received large amounts of US economic and military assistance—a total
of approximately $24 billion between 2002 and 2012—in return for its
116
participation in US efforts to fight terrorism and stabilize Afghanistan.15

Thus, in Afghanistan, as in Kashmir, Pakistan’s strategy has provided
Pakistan with important near-term benefits and helped to ensure that
a long-term outcome congenial to Pakistani interests remains at least a
possibility.
Pakistan’s militant strategy, then, has yielded some positive results. It
has not enabled Pakistan to achieve its main national goals, particularly
the outright acquisition of Kashmir. Nor has it erased Pakistan’s central
strategic challenges, such as a lack of domestic political unity and material
imbalances with India. It has, however, helped to promote Pakistani inter-
ests in all of these areas, yielding results that, while far from perfect, were
often improvements over what Pakistan might otherwise have achieved.
Most important, it has provided Pakistan with a strategic tool that can
inflict costs on India while avoiding catastrophic defeat, which India has
not yet found a way to counter. It is therefore inaccurate to characterize
the strategy as an abject failure. The reality of the strategy has been con-
siderably more complicated. It is important to keep these facts in mind
when assessing Pakistani strategic behavior in general, and Pakistan’s use
of militants specifically.
Nonetheless, despite its past utility, Pakistan’s strategy has recently
given rise to several extremely dangerous developments. They are so
serious that they far outweigh any further benefits that Pakistan might
hope to gain from the use of militancy. Pakistan has since its founding
suffered from significant material and political shortcomings. Now, how-
ever, problems associated with the militant strategy have helped to make
these shortcomings especially serious and deeply entrenched, rendering
Pakistan a chronically weak state. Weak states exhibit a lack of capacity
to formulate and execute policies designed to provide their citizens with
critical public goods such as internal stability, basic social welfare, and
external security.16 This lowers their people’s quality of life, can trigger
interstate conflict, and can even threaten the state’s survival. Pakistan’s
militant strategy has helped to create a situation in which it suffers from
all of these pathologies.
THE COSTS OF PAKISTAN’S MILITANT STRATEGY
The first major problem with Pakistan’s militant proxy strategy is that the
militant organizations that Pakistan once controlled have increasingly
slipped its grasp. After decades of financial and military support, these
groups are sufficiently strong that they can pursue their own agendas
117
regardless of Pakistani wishes. The results of this change have badly dam-
aged Pakistani security interests.
Militant groups have begun to contest the central government’s author-
ity, in some cases even competing with it for sovereignty over Pakistani
territory. For example, the Tehrik-e-Taliban Pakistan (TTP), or Pakistani
Taliban, has undertaken jihad against the Pakistani state. To this end, it
has launched bombings and suicide attacks against government targets
such as army, police, and intelligence personnel; military bases; and other
security infrastructure. The TTP also has seized control of large sections
of Waziristan, where it has repudiated Islamabad’s writ. There, in accor-
dance with the three goals that it announced at its founding in 2007, the
TTP has imposed an extreme interpretation of sharia law, battled the
Pakistan Army, and participated in attacks on government and coalition
targets in Afghanistan.17 This has led the Pakistani government to station
approximately one hundred and fifty thousand troops in the tribal areas
and to undertake extensive and costly military efforts to rout the insur-
gents. The army, which has borne the brunt of anti-Taliban operations,
has been particularly hard hit, with its soldiers frequently kidnapped and
killed by TTP militants. Pakistani efforts to combat the TTP have also
resulted in widespread civilian deaths and dislocations. This has alien-
ated local populations, thereby compounding the difficulty of Pakistan’s
counterinsurgency campaign.18
The Pakistani government’s efforts to combat the TTP in the tribal
areas have subjected the rest of Pakistan to a bloody cycle of reprisals
between the government and the Tehrik-e-Taliban. For example, the
TTP, in perhaps its most high-profile operation, attacked the Karachi
airport in June 2014, killing twenty-eight people. According to a spokes-
man for the Pakistani Taliban, the attack was launched in retaliation for
the November 2013 killing of TTP chief Hakimullah Meshud in North
Waziristan.19 In response to the Karachi attack, Pakistani Prime Minister
Nawaz Sharif, who had previously favored negotiations with the Tehrik-e-
Taliban, reversed his position and approved extensive military operations
against the group. In response, the TTP promised a campaign of coun-
trywide violence, undertaken by TTP fighters who in recent years have
spread throughout Pakistan, posing as refugees from the tribal regions.
As TTP spokesman Shahidullah Shahid said, “By God, we will soon shake
your palaces in Islamabad and Lahore and burn those to ashes.”20 Thus,
the stakes in the continuing cycle of government–TTP violence increas-
ingly are not limited to military outcomes in the contested tribal areas.
They threaten the broader destabilization of Pakistan proper. As one
prominent Pakistani analyst put it, “establishing control in Waziristan
118
won’t be the biggest issue” in the government’s battle with the TTP. “The
problem will lie in the militants’ pockets of support across the country.”21
Pakistan played an important role in facilitating the rise of the Taliban
in Afghanistan. By contrast, it did not directly create or promote the
establishment of the TTP, which is distinct from the Afghan Taliban.
Nonetheless, Pakistan’s TTP problem has resulted largely from its pursuit
of a militant proxy strategy in Afghanistan. When the Afghan Taliban fell
to coalition forces in 2001, large numbers of its fighters fled the country,
taking refuge in Pakistan’s border regions. Pakistan’s subsequent efforts
to eject these militants from its territory led them to organize into a range
of extremist groups collectively known as the Tehrik-e-Taliban Pakistan,
dedicated not only to battling coalition forces in Afghanistan but also to
fighting a “defensive jihad” against the Pakistani state, which they held
responsible for their plight. This series of events would not have occurred
if the Afghan Taliban had not controlled Afghanistan. And it is unlikely
that the Afghan Taliban would have taken control of Afghanistan had it not
been for extensive Pakistani political, financial, and military support.22 If
Pakistan had not pursed its militant proxy strategy in Afghanistan by pro-
moting the Taliban, then the Pakistani Taliban probably would not exist.
It is also worth noting that while the TTP and the Afghan Taliban
remain distinct organizations, the TTP’s leaders have sworn allegiance
to the Afghan Taliban’s leadership. According to TTP spokesman
Shahidullah Shahid, “we consider [Afghan Taliban leader] Mullah Omar
as the Amir-u l-Momineen.” He “can establish the caliphate” in Pakistan
and Afghanistan.23 The two groups also enjoy close operational coop-
eration. For example, the TTP supplies the Afghan Taliban with trained
fighters recruited from Pakistan’s tribal areas. They may also have coop-
erated on operations such as the 2009 attack on the Central Intelligence
Agency base at Camp Chapman in Khost Province, for which both
groups claimed responsibility.24 Coordination between the Afghan and
Pakistani Taliban has become so extensive that one analyst has called
the organizations “Siamese twins—t wo heads of the same body.”25 Some
observers predict that the groups may more explicitly join forces in the
future, perhaps forming some type of “Afghan-Pakistani Taliban syndi-
cate” to redouble Afghan Taliban efforts to defeat government and coali-
tion forces in Afghanistan, or possibly to intensify the TTP’s jihad against
the Pakistani state.26 Whether or not this particular danger materializes,
it is clear that by facilitating Taliban rule in Afghanistan, Pakistan has
sown the seeds of its difficulties with the TTP.
The Pakistani military has often tried to differentiate between the TTP
and the Afghan Taliban, labeling the former “good” and the latter “bad.”27
As the previous discussion makes clear, however, this is a false dichotomy.
119
The two groups share goals, tactics, and ideologies. They cooperate closely
at an operational level. Moreover, the “bad” Taliban would never have
emerged were it not for the Pakistanis’ role in creating the “good” Taliban.
Far from being wholly distinct entities, then, the Afghan Taliban and the
Pakistani Taliban are two sides of the same coin.
Even Pakistan’s closest militant allies have refused to subordinate their
interests and agendas to Pakistan’s broader strategic imperatives. Lashkar-
e-Taiba (LeT), for example, has adopted far more ambitious goals than
those of the Pakistani government, hoping not just to free Kashmir or to
claim it for Pakistan, but also to conquer India proper. LeT militants have
waged jihad on behalf of Muslim causes in multiple locations, includ-
ing Afghanistan, Chechnya, and the Philippines. Since becoming close
to the Inter-Services Intelligence agency (ISI) in the early to mid-1990s,
however, the group has restricted its focus mainly to Kashmir. This was
only partially the result of Pakistan’s influence; it was also driven by the
group’s own internal considerations. Ideologically, Kashmir is the most
significant of LeT’s concerns. Indian Kashmir is the Muslim territory
occupied by nonbelievers that is nearest to Pakistan. This creates a duty
for LeT’s members to prioritize fighting in Kashmir over waging jihad in
other locations. In addition, the Indian presence in Kashmir is one of the
largest occupations of Muslim territory by non-Muslims in the world.
This obligates members of LeT to pursue jihad in Kashmir before turning
their attention to struggles on behalf of beleaguered Muslims elsewhere.
Finally, LeT views polytheistic India as an especially objectionable foe and
maintains that the Prophet Muhammed offered special rewards to those
Muslims who fought against it.28 This antagonism toward India leads LeT
to prioritize the liberation of Kashmir over other struggles, since freeing
Kashmir necessarily involves struggle against India.
Even as LeT’s ideology leads it to prioritize Kashmir, however, it also
expands the group’s concerns much further afield. India’s presence does
not end at Kashmir’s borders, but rather extends throughout the subcon-
tinent. Therefore, LeT is obligated to fight India across the entire region.
The group thus sees Kashmir as the first in a string of future conquests
that will result in the establishment of Muslim control of all of South Asia.
As LeT official Nasr Javed declared in a 2008 speech, “Jihad will spread
from Kashmir to other parts of India. The Muslims will be ruling India
again.”29
Not only do LeT’s ambitions extend beyond Kashmir throughout
the Indian subcontinent, but also they extend well beyond South Asia.
Indeed, the group views its South Asian struggle as part of a larger effort
to oppose infidel regimes and establish Islamic rule throughout the world,
including in far-off regions such as Europe and North America. In the
120
words of an LeT manifesto, “Until Islam prevails throughout the world

and Allah’s law applies to everyone it is our duty to keep fighting against
the infidels.”30 LeT leaders are unconcerned with the impact of such a far-
reaching agenda on Pakistani interests and undeterred by the prospect
of opposition from Islamabad. According to Javed, “We will continue to
wage jihad and propagate it till eternity… . Nobody can stop it—be it the
U.S. or Musharraf.”31
These problems with Pakistan’s militant proxies are not uniform across
organizations. LeT, for example, appears not to have employed violence
directly against the Pakistani state. Indeed, LeT recruits are apparently
instructed during training never to use their newly acquired skills to
make war on Pakistan, as the group’s doctrine prohibits it from attacking
Muslim governments. In principle, this prohibition holds even when such
governments have fallen into error and strayed from the path of proper
religious behavior. Jaish-e-Mohammed (JeM), by contrast, is subject to
no such doctrinal constraints. JeM has on multiple occasions attacked
Pakistani government targets, including prominent officials, as well as
police, intelligence, and military installations. Perhaps most famously,
JeM was implicated in a nearly successful attempt to assassinate President
Pervez Musharraf in 2003 in retaliation for his cooperation with US anti-
terrorism efforts.32
Even if a group such as LeT is less problematic than some of its peers,
however, it nonetheless poses serious control problems for Pakistan. As
noted earlier, LeT has adopted pan–South Asian goals and a global agenda
that are considerably more expansive than the aims of the Pakistan gov-
ernment. Moreover, the group has publicly repudiated the government’s
attempts to limit its ambitions and activities. LeT could undertake unau-
thorized provocations against India, perhaps on a scale similar to the 2008
assault on Mumbai. India might well blame these actions on the Pakistani
government, even though Pakistani leaders did not countenance them.
The Indians could consequently decide to respond by using military force
against Pakistan, plunging the two countries into conflict. Such an out-
come could be catastrophic for Pakistan, which can ill afford a military
confrontation with conventionally stronger India.
In addition, LeT’s policy of refraining from violence against Pakistan
is less straightforward than it might initially seem. Although LeT has not
itself attacked the Pakistani state, its members have worked closely with
the TTP and various Deobandi groups. LeT members probably have col-
laborated with these organizations in striking Pakistani civilian leader-
ship and military targets. Finally, despite the apparent clarity of LeT’s
prohibition against anti-Pakistani activity, the policy’s future has been
the subject of serious debate within the group. Members have argued that
121
LeT should reconsider its position, given that the Pakistani government
has at times restricted the group’s activities. Even Pakistan’s closest and
most supportive militant allies, then, may pose serious control problems
for it in the years ahead. 33
The second major problem with Pakistan’s militant strategy is the
opportunity cost it entails. Continual support for jihad diverts scarce
national resources from other critical projects, impeding Pakistani inter-
nal development. Pakistan’s education sector offers one of the most urgent
examples of this problem. Only 62 percent of Pakistani primary-school-
aged children and 30 percent of secondary-school-aged children are actu-
ally enrolled in school. Nine percent of primary schools lack a blackboard,
24 percent do not have textbooks, and 46 percent have no desks. A mere
36 percent of public primary schools have access to electricity. And 9 per-
cent of teaching posts at the secondary level remain unfilled. 34 Pakistan
must rectify this situation if it hopes to achieve the long-term economic
growth necessary to afford its citizens a reasonable quality of life. 35
The lack of viable public educational opportunities creates a demand
for private schooling. Part of this demand is met by madrassas, as well as
by less formal mosque schools, which offer poor students free or low-cost
educations based on religious curricula. 36 With their focus on religious
teaching at the expense of such subjects as mathematics, science, and his-
tory, madrassas and other religious schools produce graduates lacking
the skills necessary to run a modern country. They also contribute to the
increasingly Islamist and sectarian tenor of Pakistani society. Survey data
suggest that students and teachers in madrassas are less socially tolerant
and more supportive of religiously motivated violence than their coun-
terparts in public and nonreligious private educational institutions. For
example, they are not as likely to support equal rights for women and
religious minorities, and are more favorably disposed toward militancy
in Kashmir, as well as outright war with India, than their peers in public
and private schools. Finally, madrassas provide manpower for militant
groups. Although madrassas appear to have only weak connections with
militants operating in Kashmir or overseas, suicide attackers in Pakistan
and Afghanistan, as well as sectarian fighters in Pakistan, often come
from madrassa backgrounds.37
A better public education system, offering a broad-based, modern
curriculum, would help to ameliorate these problems. Yet Pakistan has
not chosen to invest in the development of such a system. It has pre-
ferred instead to devote scarce resources to the maintenance of a large
military establishment. The social costs of this military prioritization
are significant. In the late 1980s, Pakistan was spending approximately
7 percent of gross domestic product (GDP) on defense. This spending
122
declined significantly in the following years, falling to roughly half of

late 1980s levels by 2006. Even at these relatively lower levels, however,
defense remained a massive investment, accounting for about 22 percent
of Pakistan’s federal spending. Twelve percent of the country’s federal
budget, or 2.21 percent of GDP, by contrast, was spent on education.
Cutting approximately one percentage point of GDP from the military
budget would have enabled Pakistan to increase education spending by
55 percent. 38
More generally, analysts have shown that lower defense spending in
past decades would likely have yielded significant economic benefits for
Pakistan over the years. This, in turn, could have drastically improved
its present-day economic situation. For example, according to former
Pakistani finance minister Shahid Javed Burki, if Pakistan had limited its
defense expenditures to 2.5 percent of GDP, which would have roughly
matched India’s military spending–to–GDP ratio, Pakistan could have
saved 3 percent of GDP annually. Over fifty years, Burki argues, this would
have resulted in a 50 percent increase in GDP. According to Pakistani
economist Parvez Hasan, halving Pakistani defense spending from 1970
to 2010 could have added two percentage points per year to Pakistan’s
GDP. This would have resulted in an overall economy double the size of
the one that Pakistan actually has. Such economic windfalls could have
been devoted to a range of public goods, including not just education, but
also infrastructure, public health, internal security, and general poverty
alleviation. They also would have made Pakistan a more attractive trad-
ing partner, potentially increasing the incentives for cooperation between
Islamabad and New Delhi. Had this occurred, as Hasan puts it, “the whole
history of the subcontinent could have been a happier one.”39 Despite the
costly trade-offs inherent in its high levels of defense spending, however,
Pakistan continues to invest heavily in the military, maintaining a level of
about 3 percent of GDP in recent years. The situation seems unlikely to
improve in the near future.40
Of course, Pakistan’s asymmetric warfare strategy does not account
for all, or even most, of its military spending. Even if the proportion is
only modest, however, its impact is significant, since relatively small
cuts in defense outlays would substantially increase Pakistan’s ability
to pursue domestic developmental goals.41 More generally, Pakistan’s
asymmetric warfare campaign, by continually provoking India, greatly
increases Indo-Pakistani tensions and helps to create an extremely
hostile and demanding regional security environment. It is this envi-
ronment, in turn, that forces Pakistan to devote such a high level of
resources to ensuring its external security. In addition, the Pakistanis
have had to increase military spending to battle internal insurgencies
123
launched by their erstwhile militant allies such as the TTP.42 In a more

permissive strategic context, with a less threatening international
neighborhood and less domestic instability, the Pakistanis could safely
lower their overall defense budget and invest the savings elsewhere.
Both directly and indirectly, then, Pakistan’s militant strategy diverts
scarce national resources from underfunded sectors that are crucial to
the country’s well-being.
The third problem with Pakistan’s militant strategy is that it is provok-
ing the Indians to develop new military capabilities, which may enable
them to undertake large-scale attacks on Pakistan at short notice. In the
past, India required weeks to move offensive conventional forces from
interior peacetime stations to the Indo-Pakistani border. This delay
allowed Pakistan to prepare its defenses and enabled the international
community to pressure the Indian government to stand down. As a
result, the Indians were sometimes forced to forgo retaliatory military
action even in the face of severe provocations by Pakistan-backed terror-
ists. Perhaps the best example of this problem was the Indian decision
not to strike Pakistan in the wake of a 2001 militant attack on the Indian
parliament.43
To address this problem, the Indians have begun planning to launch
large-scale offensive operations against Pakistan far more quickly in the
future. Under their new approach, which has been labeled the “Cold
Start” Doctrine, the Indians will send multiple integrated elements of
infantry, armor, and artillery across the border within days, rather than
weeks, of a warning order. Previously, under what was known as the
Sundarji Doctrine, the Indians had planned to drive armored forces deep
into Pakistan along a small number of axes. The Indians’ new plans call
for numerous rapid attacks, destroying enemy forces and seizing terri-
tory across a long, shallow line of advance. The Indians hope that this
approach will achieve two main goals. First, its speed will enable India to
strike Pakistan within days of a future terrorist provocation, before the
Pakistanis can prepare their defenses or the international community can
dissuade Indian political leaders from taking offensive action. Second,
Cold Start’s multiple shallow attacks will enable India to inflict significant
punishment on Pakistan without penetrating its territory so deeply as to
exceed Pakistani nuclear thresholds.44
Senior Indian officials, including India’s former army chief, have
denied that the Cold Start Doctrine actually exists.45 Even if Indian plans
have not yet reached the level of doctrinal coherence, however, they are
reorienting India’s approach to problems of time and space in striking
Pakistan. India’s traditional emphasis on slow and deep attacks against
Pakistan is changing to an emphasis on attacks that are fast and shallow.
124
As one senior Indian strategist put it: “Call it what you like, Cold Start,
something else, a doctrine, not a doctrine. This is a new way of thinking
about bringing force to bear against the Pakistanis.”46
Significantly, these conventional military changes not only will affect
Indian strategic planning but also will have a major impact on Pakistan.
Because the Indians will be able to launch a major offensive far more
quickly than before, Pakistan will have to prepare to respond rapidly.
Acquiring the capability to do so will be difficult, however, for India’s
economic and military resources far outstrip Pakistan’s. Despite recent
downturns, India’s economy remains far larger than Pakistan’s, with a
GDP of $1.87 trillion, compared to Pakistan’s $236.6 billion. It is grow-
ing at a rate of 5.5 percent, compared to Pakistan’s 3.7 percent, and was
recently forecast to become the world’s third-largest economy, behind
only China and the United States, by 2030.47 The Indian defense budget,
at over $37 billion, is also far larger than Pakistan’s $7 billion budget and
is growing at a rate of approximately 10 percent per year.48
Pakistan will thus be hard pressed to defend itself against the growing
Indian threat through conventional military means. Instead, Pakistan is
likely to rely more heavily on nuclear weapons.49 Doing so will require
the Pakistanis to threaten to use nuclear weapons first, however, in
response to Indian conventional attacks. Such threats may lack credibil-
ity, for the result of Pakistani nuclear first use would likely be devastating
for Pakistan, which could be irreparably crippled by even modest Indian
nuclear retaliation following a Pakistani first strike. Pakistan therefore
faces a conundrum: To protect itself from the danger of an Indian conven-
tional attack, Pakistan must threaten to trigger a nuclear catastrophe—an
outcome far worse than the conventional war that Pakistan is trying to
deter. Indian leaders may simply not believe that Pakistan would be will-
ing to carry out such a threat.
The Pakistanis are addressing this problem by pursuing a tactical
nuclear weapons capability, for use on the battlefield against Indian mili-
tary forces, rather than against civilian targets in India. To this end, it
has developed and tested the Nasr missile, with a range of 60 kilome-
ters, which Pakistani strategists have dubbed an antidote to Cold Start.
It is also expanding its production of plutonium, which is better suited
than uranium for use in small warheads. 50 A tactical nuclear weapons
capability can help the Pakistanis to increase the credibility of their
nuclear deterrent threat in two ways. First, tactical nuclear weapons are
relatively small and are employed against military targets. The threat to
use them is therefore less momentous, and likely to be more believable,
than a threat to launch large-scale strikes against Indian civilian targets.
125
Second, assuming that launch authority for Pakistani tactical nuclear

weapons is to be predelegated to field commanders at some point in a
crisis, the decision to escalate an Indo-Pakistani conflict from the con-
ventional to the nuclear level may not be fully in the hands of national
leaders. Rather, the decision could be in the hands of more junior mili-
tary officers in the thick of battle, who may be more inclined to use the
weapons than senior officials located in the relative calm of Islamabad or
Rawalpindi. 51 This will increase the likelihood of a conventional Indo-
Pakistani conflict escalating to the nuclear level, regardless of the wishes
of Pakistani national leaders at that time. As a result, Pakistani first-use
threats may become more credible, despite the immense costs and dan-
gers of Pakistani nuclear first use. 52
Together, these Indian and Pakistani strategic developments will result
in compressed crisis escalation timelines in South Asia. Cold Start–like
doctrinal innovations increase the likelihood that a future Indo-Pakistani
militarized dispute will ascend to the level of conventional war more rap-
idly than before, with fewer opportunities for resolution. And the devel-
opment of a Pakistani tactical nuclear weapons capability is likely to make
the potential jump from conventional to nuclear confrontation occur
more quickly as well. 53 By driving aggressive Indian doctrinal innova-
tion, then, Pakistan’s asymmetric warfare campaign threatens to trigger
the very disaster that it was designed to avoid—a direct, large-scale Indo-
Pakistani military confrontation, now with a significant risk of nuclear
escalation. The cost of such an outcome would far outweigh any ben-
efit that Pakistan could hope to achieve through continued support for
militancy.54
Despite its historical utility, then, Pakistan’s militant strategy
has given rise to problems that severely damage Pakistani interests.
Significantly, these problems are largely attributable to a single fac-
tor: the acute weakness of the Pakistani state. It is this weakness that
has enabled the militants increasingly to defy Pakistan, that has forced
Pakistan to choose between supporting jihad and pursuing internal
development, and that makes growing Indian military capabilities so
threatening. Ironically, however, it is also weakness that made the stra-
tegic use of militants attractive to Pakistan in the first place. Pakistan
adopted its militant strategy in response to its own severe material and
political shortcomings. Pakistan is thus caught in a jihad paradox: The
very characteristic of the Pakistani state that makes a policy of support-
ing Islamist militancy useful also makes it extremely dangerous. I dis-
cuss this paradox, and potential solutions to it, at greater length in the
next chapter.
126
CONCLUSION
In this chapter, I offered an assessment of Pakistan’s militant proxy strat-

egy, discussing its main advantages and disadvantages. In the first sec-
tion, I argued that Pakistan’s strategy has afforded it a number of benefits,
including the promotion of domestic political cohesion, partial mitiga-
tion of India’s material advantages over Pakistan, the ability to ensure
that the Kashmir dispute remained contested and on the international
community’s agenda, and a means of managing the security environment
in Afghanistan. None of these benefits has enabled Pakistan to achieve
its central goal of wresting Kashmir from Indian rule or to eradicate the
country’s deep sense of insecurity vis-à-v is India. Nonetheless, the ben-
efits are not trivial and should be taken into account when evaluating or
explaining Pakistani strategy.
In the chapter’s second section, I argued that Pakistan’s militant strat-
egy has given rise to a number of serious problems that now far outweigh its
advantages. Indeed, they are so serious that they could threaten Pakistan’s
survival in the years ahead. These include Pakistan’s loss of control over its
proxy allies, whose interests increasingly diverge from their patron’s; eco-
nomic opportunity costs, which have stunted Pakistan’s internal develop-
ment; and increased military dangers on the subcontinent, which could lead
to war and potentially even a nuclear exchange between India and Pakistan.
Pakistan thus currently finds itself caught in a jihad paradox. The problems
associated with Pakistan’s militant strategy result primarily from the acute
weakness of the Pakistani state. Yet it is state weakness that has, since inde-
pendence, made the use of militant proxies an attractive strategy for Pakistan.
How can the problems resulting from Pakistan’s jihad paradox be miti-
gated? In chapter 7, I argue that only one possible solution exists: Pakistan
must recognize that its current policy has outlived its utility, abandon its
support for militancy, and work to defeat all militant organizations oper-
ating within its territory. This is unlikely to occur, however, barring a
wholesale change in how Pakistani leaders define the country’s purpose
and identity. The Pakistanis’ use of militancy to undermine the regional
status quo is a symptom of the oppositional nature of the Pakistani state,
which defines itself in contradistinction to India and must therefore chal-
lenge India at every turn. Unless the Pakistanis renounce this fundamen-
tally oppositional approach to state building, there is little chance that
they will abandon militancy. Such a thoroughgoing change is unlikely,
but not impossible. States facing existential challenges have taken simi-
larly radical steps in the past. If the Pakistanis fail to adopt such measures
soon, the prognosis for their country is bleak. Before long, Pakistan may
be consumed by its own strategy.
127
CHAP T ER 7
The Future
Can Pakistan Abandon Jihad?
I n chapter 6, I showed that, despite its historical benefits, Pakistan’s

jihad paradox has become extremely dangerous—so dangerous that, if
Pakistan does not take immediate steps to end its support for militancy, it
could face catastrophe. Surprisingly, Pakistan has largely failed to address
this problem. The Pakistani military has launched extensive operations in
the country’s tribal areas against the Pakistani Taliban, whom it has iden-
tified as “bad” militants. Such efforts have been ineffective in the past.
The Pakistan Army, which is designed primarily for battlefield operations
against Indian conventional forces, is not ideally suited for counterinsur-
gency operations.1 How successful the military now will be in its attempts
to defeat the Tehrik-e-Taliban Pakistan (TTP) remains to be seen.
Regardless of the outcome of its ongoing efforts against “bad” mili-
tants, Pakistan must act decisively against all militants, including those
whom it has traditionally labeled as “good.” Pakistani officials have prom-
ised to do so, stating that they no longer subscribe to the good militant/
bad militant dichotomy. Pakistan will take action, they claim, against the
full spectrum of militant groups, regardless of their allegiance or ideol-
ogy. As Pakistani Defense Minister Khawaja Asif said, “We will eliminate
all sorts of terrorists from our area … without making any differentia-
tion between our Taliban and their Taliban, or good Taliban and bad
Taliban… . They are all bad Taliban. There are no more good Taliban.”2
There is little evidence, however, that Pakistan actually intends to
abandon its selective approach to fighting militancy. For example, despite
assurances that their anti-Taliban offensives would also target the Haqqani
network, the Pakistanis have left the Haqqanis largely unmolested,
128
allowing their personnel to disappear from the tribal areas into Pakistan
proper, as well as into Afghanistan. 3 Elsewhere, the Pakistanis have failed
to take such basic steps as closing training camps for Kashmiri militants,
shuttering Lashkar-e-Taiba’s stronghold at Muridke, significantly restrict-
ing the activities of Jaamat-ud-Daawa’s Mufti Mohammed Saeed, or cut-
ting ties with the Afghan Taliban.
This Pakistani behavior is puzzling. When a national security strat-
egy fails to achieve its objectives, rational leaders should change course,
adopting new strategic approaches to minimize dangers and increase the
likelihood of attaining their goals. Given their militant strategy’s signifi-
cant shortcomings, Pakistani leaders thus should be expected to look for
new ways to produce national security that do not rely on the use of non-
state proxies. Even if the Pakistanis do not abandon their militant proxy
strategy immediately and in full, they should at least be taking concrete
steps away from the policy. This has not yet happened, however. Why is
this the case? Pakistani failure to abandon jihad despite the strategy’s
many dangerous problems is one of the biggest mysteries of Pakistani
security behavior.4 Explaining this past failure is important if we are to
suggest ways in which Pakistan might devise means of altering its behav-
ior in the future.
In this chapter, I argue that several reasons exist for Pakistan’s failure
to abandon its militant proxy strategy. First, the magnitude of the prob-
lems associated with Pakistani strategy has become apparent only rela-
tively recently. For example, truly worrisome Pakistani developmental
indicators, as well as the most dangerous security-related trends in South
Asia, have occurred mainly within the past decade. Until then, Pakistan
appeared to be faring reasonably well, generally avoiding major military
debacles, and actually outperforming India in a number of economic and
developmental areas. Even today, Pakistan does not lag far behind India
on several important indicators of national well-being. Thus, the urgency
of the need to abandon militancy and develop alternative security strate-
gies has not always been apparent.
Second, Pakistani leaders believe that, with nuclear weapons, they
hold a national-security trump card. Whatever else happens in Pakistan’s
security competition with India or other states, nuclear weapons ensure
that the Pakistanis will not suffer catastrophic defeat. This significantly
reduces the Pakistanis’ incentives to abandon their militant strategy and
develop less dangerous means of generating national security.
Third and most important is the character of Pakistan’s state-building
project. Since Pakistan’s founding, it has continually had to compete with
India to meet the demands of its oppositional national narrative. Given
its material imbalance with India, an ongoing jihad employing nonstate
129
C a n Pa k is ta n A b a n d o n J i h a d ? ( 129 )
proxies has offered Pakistan one of the few practical means of doing so.
The costs and risks of its other main option, direct military confrontation
with India, would have been prohibitive. Probably more than any other
factors, these oppositional demands of Pakistani state building prevent
Pakistan from fundamentally changing course and abandoning militancy,
for if it does so, Pakistan will have to cease its ongoing anti-Indian strug-
gle and reach some form of basic accommodation with India. If it makes
such a change, however, the Pakistani state, as defined since its founding,
will cease to exist. Although it will be extremely difficult, this is precisely
what will need to occur if Pakistan is to have any real hope of abandoning
militancy. Only if Pakistan can alter its state-building narrative will it also
be able to change its strategy and avoid looming catastrophe.
WHY PAKISTANI STRATEGY HAS NOT CHANGED
Pakistan’s failure to alter its militant proxy strategy, despite mounting evi-
dence of the strategy’s shortcomings, is a vexing puzzle. What accounts
for Pakistani intransigence on this issue? One possibility is that Pakistani
leaders are simply irrational, unable to devise coherent strategies con-
necting the means that they employ with the ends that they seek. 5 As a
result, the Pakistanis continually make poor policy choices that not only
fail to achieve their goals but also actually undermine them.
The main problem with this explanation is that Pakistani security man-
agers have shown little evidence of irrationality in the past. Indeed, the
military means that they have employed have, in general, been fairly well
tailored to the pursuit of their ends. This is not to argue that the Pakistanis
have never made mistakes, or that their policies have never resulted in
dangerous failures. Mistakes and failures have obviously occurred on a
number of occasions, ranging from the Pakistanis’ long-standing inabil-
ity to dislodge India from Kashmir to Pakistan’s vivisection during the
Bangladesh war. Nonetheless, the main facets of the Pakistanis’ security
behavior, from the use of militant proxies to challenge stronger adver-
saries while avoiding direct military confrontation with them, to the
development of nuclear weapons to insulate Pakistan against large-scale
retaliation by such adversaries, to the recruitment of powerful allies to
provide Pakistan with critical economic and military resources, have both
required a significant degree of finesse to execute and been largely rea-
sonable given the revisionist nature of Pakistan’s goals and its status as a
relatively weak state. It seems unlikely that, given this record, Pakistani
leaders have suddenly become so unable to connect ends and means that
they should be considered irrational.
130
Another possibility is that Pakistan’s failure to change results largely

from pathologies in the structure of its strategic environment. T. V. Paul
calls this Pakistan’s geostrategic curse. According to this argument,
Pakistan’s strategic assets, particularly its position in an extremely impor-
tant part of the world, attract attention and support from powers such
as the United States and China, which seek to exploit Pakistan’s assets
for their own purposes. As a result, Pakistani leaders carefully cultivate
rentier relationships with outside powers but expend little effort devising
serious means of addressing their own country’s host of serious develop-
mental and security-related problems.6
This is an important insight; by creating perverse incentives for
Pakistani leaders, the geostrategic curse no doubt accounts for some of
the country’s failure to devise more successful policies. This does not
mean, however, that the geostrategic curse is the deep cause of Pakistan’s
counterproductive behavior. As Paul makes clear, many other states have
been richly endowed with strategic resources and could therefore suf-
fer from the curse that he identifies. Yet these states do not behave like
Pakistan.7 Turkey, for example, enjoys extensive strategic endowments
and has received even more military assistance from the United States
than Pakistan, but it has managed to succeed in becoming a generally
stable, secular, developmental state.8
What accounts for Turkey’s success? Paul points to a number of causes,
such as the Turkish leadership’s generally pragmatic approach to security
matters, their rejection of an overly competitive or “hyper-realpolitik”
approach to international politics, and their commitment to economic
growth and modernization.9 This suggests, however, that structural stra-
tegic advantages, despite their dangers, do not necessarily doom a state
to Pakistan’s fate. Instead, such a strategically well-endowed state might
come to resemble Turkey, provided that its leaders have similar prefer-
ences to those motivating the Turkish leadership. The deep cause of
Pakistan’s continually counterproductive behavior, it seems, lies not in
structure, but in bad ideas.
A major source of these bad ideas is the Pakistan Army. The army, as
Christine Fair argues, is fundamentally “greedy,” insisting on continuing
Pakistani efforts to alter the status quo for reasons that are not connected
with national security. For example, the army seeks to promote conflict
with India to advance its bureaucratic interests, ensuring that it contin-
ues to receive a disproportionate share of state resources and to dominate
both domestic and foreign policymaking. The army also seeks to pro-
mote continual Indo-Pakistani conflict to advance its ideological com-
mitments, protecting what it sees as Pakistan’s Islamic identity through
ongoing competition with Hindu India. As Fair points out, even if doing
131
so does not result in actual battlefield victories, military leaders view it as

being valuable. Conflict enables the army to reinforce its role as Pakistan’s
ideological protector and undermine India’s dominant position on the
subcontinent.10
Despite the army’s importance, however, these bad ideas, in the first
instance, come from an even more fundamental source—the founding
logic of the Pakistani state. To be sure, the army has internalized and
operationalized this logic. The Pakistani state’s founding narrative pre-
cedes the army’s beliefs and commitments, however. Indeed, its logic pro-
vides the foundation upon which those beliefs and commitments rest.
Since independence, the central narrative of the Pakistani state, given
its weak political foundations, has been oppositional, rooted in the dis-
tinction between Islam and Hinduism. As a result, if it is to remain true to
its state-building logic, Pakistan cannot cease behaving in an oppositional
manner; Pakistan must constantly struggle against India. To abandon this
effort would be to relinquish a central element of the identity and raison
d’être that it has constructed to justify Pakistan’s existence and generate
critical political cohesion. Pakistan thus resembles a fish that cannot stop
swimming. It must continue to move in an oppositional direction if it is to
survive in its current ideological form.
The most clear and dramatic way for Pakistan continually to oppose
India is to seek to undo existing territorial divisions that are seen to favor
Indian interests, particularly in Kashmir. This is the fundamental source
of ongoing Pakistani revisionism. The army is of course deeply committed
to revisionist aims and continually seeks to achieve them. Revisionism,
however, is about more than the army’s military goals, bureaucratic inter-
ests, or ideological commitments. Its source lies in the definitional logic
of the Pakistani state. It is this logic that underlies the army’s continued
revisionism and desire for conflict with India regardless of battlefield vic-
tories or strategic success.
This is one of the reasons territorial revisionism is not limited to the
Pakistani military. Civilian leaders harbor revisionist goals as well. For
example, as noted earlier, Zulfikar Ali Bhutto viewed Pakistan’s claim
to Kashmir as eternal and unchanging. He refused to abandon it even
after suffering catastrophic defeat in the Bangladesh war and despite
the language of the Simla Agreement, in which both India and Pakistan
pledged not to attempt to revise territory on the subcontinent by force.
Benazir Bhutto closely echoed her father’s views, saying that Pakistan
could never relinquish its claim to Kashmir, regardless of the language
of the Simla Agreement.11 In public, she energetically reiterated his claim
that Pakistan was willing to fight one thousand years with India over
Kashmir.12 Nawaz Sharif, despite his deep animosity toward the military,
132
shared the military’s revisionist goals and was well known to support mil-
itant groups operating in Kashmir and elsewhere in the region. Indeed,
Sharif favored negotiating with the Tehrik-e-Taliban even after the army
had decided to undertake large-scale military action against it in 2014. It
is also worth noting that challenging the status quo in Kashmir is popular
throughout Pakistani society. According to recent polling data, 68 percent
of Pakistanis view the current situation in Kashmir as a “very big prob-
lem,” with an additional 18 percent seeing it as a “moderately big prob-
lem”; 79 percent of Pakistanis believe that it is “very important” that the
Kashmir dispute be resolved.13 These views are constantly reinforced in
an array of fora, such as public school curricula, which teach students the
history of jihad, describe Hindus as enemies of Pakistan, and portray the
defense of Pakistan as being synonymous with the defense of Islam, and
in popular political discourse, where civilian leaders describe Kashmir as
Pakistan’s “jugular vein.”14
A strategy of jihad offers one of the few practical means by which
Pakistan can operationalize this foundational revisionism; jihad enables
Pakistan continually to challenge the status quo while avoiding the dan-
gers of direct Indo-Pakistani conflict. Thus, Pakistan’s state-building
narrative and its use of militant proxies are inexorably linked. Pakistan’s
oppositional approach to state building requires it to adopt jihad as an
integral part of its grand strategy—and to continue to employ it regard-
less of whether Pakistan actually achieves its revisionist aims.
Although the logic of Pakistan’s state-building narrative is the major
reason for its ongoing adherence to its militant strategy, a number of other
causes deserve mention. One is the fact that the full extent of the strate-
gy’s dangers became apparent only relatively recently. For example, severe
principal–agent problems between Pakistan and its militant proxies have
arisen in roughly the past decade. Principal–agent issues did, of course,
dog Pakistan’s militant strategy much earlier than this. The Pakistanis
long had difficulty finding allies with the optimal mix of interests and
capabilities that would lead them to do Pakistan’s bidding as effectively
as possible. The task of identifying and recruiting such proxies con-
sumed a great deal of effort, particularly as the Pakistanis managed the
Kashmir insurgency following its outbreak in 1989. Despite these difficul-
ties, however, Pakistani concern focused primarily on the effectiveness
of their allies. The Pakistanis did not fear that the militants would work
directly against them, attacking government infrastructure and person-
nel, attempting to assassinate national leaders, or challenging Islamabad
for control over Pakistani territory. Such problems emerged mainly in the
post-9/11 environment, after Pakistan joined the United States in its war
on terror.
133
Just as the true severity of the militant strategy’s principal–agent prob-

lems has come to light relatively recently, the full extent of the strategy’s
developmental costs and trade-offs has only recently become clear. It
is easy to forget that, for roughly the first four decades of its existence,
Pakistan was considered an economic success story. During the 1950s
and 1960s, the Pakistani economy featured a mix of import substitution
and centralized planning, the maintenance of multiple exchange rates to
encourage exports, and the sale of publicly owned manufacturing con-
cerns to the private sector. The result was an average expansion of 3.1 per-
cent per year during the 1950s and 6.7 percent per year during the 1960s.15
Progress slowed for the first time during Zulfikar Ali Bhutto’s tenure as
prime minister from 1972 to 1977. Bhutto’s socialist proclivities led him
to de-emphasize private-sector growth, adopt a policy of state-directed
industrialization focusing particularly on heavy industry, nationalize
many important sectors of the economy, establish government monopo-
lies of export trades such as cotton and rice, and make large-scale invest-
ments in public- development programs. Combined with exogenous
factors such as the oil shocks of the early 1970s, the Bangladesh war, water
shortages, and poor weather conditions, these policies led to sharp reduc-
tions in private investment and increased deficits. Consequently, growth
slowed to 4.4 percent from 1970 to 1977.16
Fairly soon thereafter, however, under General Zia-ul-Haq, the econ-
omy rebounded. The improvement resulted in part from Zia’s reversal of
a number of Bhutto’s policies, including gradual industrial denational-
ization and some liberalization of trade. In addition, Pakistan benefited
from a massive inflow of US military and development aid in return for
its cooperation in US efforts to oust the Soviet Union from Afghanistan.
Finally, Pakistan received significant remittances from workers over-
seas, which also helped to boost the economy. The result was a return to
roughly 1960s-level growth rates with 6.4 percent average gross domestic
product expansion per year from 1977 to 1988.17
Zia’s reversal of Bhutto-era approaches had been only partial, however.
He maintained a number of statist policies including large-scale invest-
ment in heavy industry, government trade monopolies, and nationaliza-
tion of the banking sector. The resulting inefficiencies helped to create
significant fiscal and current-accounts deficits. The country also suffered
from a balance-of-payments crisis resulting from declining outside assis-
tance as the Afghan war ended. This combination of factors brought
Pakistan’s second growth period to a close in the late 1980s.
Following the Zia years, during the 1990s and into the 2000s, Pakistan
once again adopted new economic policies. This approach emphasized
privatization, investment deregulation, tax reform, and fiscal discipline
134
imposed by outside institutions such as the International Monetary

Fund. The result was intermittent growth. The economy expanded rapidly
during some periods, such as 2004 through 2006, with average growth
of over 7 percent per year. At other times, such as 2009 through 2011,
growth was far slower, averaging less than 2.5 percent per year.18
By briefly recapping Pakistan’s economic history, I seek to put its cur-
rent situation in a larger context. Despite very real present-day problems,
the story of Pakistan’s economy over time has not been one of unmiti-
gated failure. Rather, it has been a story of ups and downs, with failures
resulting primarily from an array of nonmilitary challenges including
problematic domestic development strategies and exogenous shocks
ranging from resource shortages to poor weather. In fact, Pakistan’s big-
gest impediment to achieving sustained economic growth has probably
been its failure to impose a broad-based taxation regime on the country.19
Overall, as S. Akbar Zaidi writes, “despite the rise-and-fall pattern of GDP
growth over the past 50 years, Pakistan’s economy has grown on average
by more than 5 percent per year. This is no mean achievement.”20 Indeed,
as Parvez Hasan points out, only a small number of countries in East Asia
managed to achieve similar rates of economic expansion between 1960
and 1990. Pakistan’s overall growth rate since its founding has exceeded
that of most developing countries.21
Even if one ignores the larger historical picture and focuses just on
the recent past and the present, Pakistani performance, particularly in
comparison to its main competitor, India, is mixed. India has a larger
economy, with a gross domestic product of approximately $1.6 trillion,
compared to Pakistan’s roughly $237 billion. In addition, Indian eco-
nomic performance over the past decade has far outmatched Pakistan’s.
During this period, India has enjoyed unprecedented economic expan-
sion, with growth of roughly 7 to 10 percent per year between 2003 and
2010. Pakistani performance, meanwhile, hit a slump in 2007 and fell
as low as 1.6 percent per year in 2010. Since then, however, the situation
has changed. Pakistani performance has improved, with growth rates
of 4 percent in 2012 and 6.1 percent in 2013, while India’s has fallen off
substantially, with growth of 4.7 percent in 2012 and 5 percent in 2013. 22
Indian and Pakistani figures for gross domestic product per capita, mean-
while, are closely matched—$1,299 for Pakistan compared with $1,499
for India in 2013.23 Other indicators also put India and Pakistan in a
roughly equal light. For example, Pakistan’s adult literacy rate is 54.9 per-
cent versus 62.8 percent for India. And life expectancy at birth in Pakistan
is 66.4 years, compared with 66.2 years in India.24
My purpose here is not to deny that the Pakistani economy is seri-
ously underperforming, or that, as a result, the country’s domestic
135
development is lagging dangerously. Nor is it to deny that Pakistan’s

militant proxy strategy, and the competitive international environ-
ment to which it gives rise, is a major drag on Pakistani economic
growth and the source of significant developmental trade-offs. All
of these problems are real, and they are growing more serious over
time. Nonetheless, we must keep them in context when trying to
understand Pakistan’s continued adherence to its militant strategy.
Pakistani leaders have not been ignoring catastrophically bad growth
and development data for decades while blindly adhering to security
policies that were worsening the situation. In truth, Pakistan’s past
economic record has been mixed, featuring both periods of consider-
able growth and periods of significant contraction. Overall, its perfor-
mance has been equal to or better than many peer countries. Severe
underperformance has been relatively recent, with Pakistani growth
declining precipitously only during the past decade. Even with its cur-
rent problems, however, on a number of important indicators Pakistan
is roughly evenly matched with its main adversary, India—despite
widespread celebration of India in recent years as an economic suc-
cess story. This helps to explain some of Pakistan’s lack of urgency in
renouncing its militant proxy strategy, despite the strategy’s deleteri-
ous economic and developmental effects.
The third major reason for Pakistan’s failure to abandon its militant
strategy is that the full impact of the strategy’s deleterious effects on
Pakistan’s external security has only recently become clear. As we have
seen, the use of militants traditionally enabled Pakistan to challenge the
regional status quo while minimizing the likelihood of full-scale conflict
with India. This was one of the strategy’s most powerful attractions for
Pakistani security managers. Full-scale Indo-Pakistani conflict became
even less probable once Pakistan acquired what it viewed as its trump
card in its competition with India—a nuclear weapons capability. Prior to
this there was always a danger that the Indians could launch a large-scale
conventional military attack against the Pakistanis in retaliation for their
ongoing provocations. Pakistan’s acquisition of nuclear weapons, how-
ever, made such an Indian response extremely dangerous, and thus even
less likely than it had been previously. Pakistan’s reaction was to engage
in even more adventurism than it had done before, thereby making the
subcontinent considerably less stable. Peace in South Asia was about
five times more likely prior to nuclearization than it was after India and
Pakistan acquired nuclear weapons. Militarized disputes between India
and Pakistan were approximately four times more common after nuclear-
ization than they were when the subcontinent was nonnuclear. Pakistan
initiated the vast majority of these disputes.25
136
India was slow to develop a response to this changed state of affairs.

Thus, for a number of years, Pakistan could feel relatively safe challenging
the regional status quo with nonstate actors from behind a nuclear shield.
The Indians’ lack of an answer to the Pakistanis’ strategy was partially a
function of inadequate resources and partially the result of a dearth of
fresh ideas. As noted earlier, India has made progress on both of these
fronts. Its economic boom has enabled India to invest heavily in new
military capabilities. In addition, the Indians are developing new ways
of thinking about the problems of time and space in attacking Pakistan,
like Cold Start. This is likely to improve India’s ability to punish Pakistan
quickly in the event of a future crisis.
These developments have happened in roughly the past decade. Rapid
Indian economic growth did not begin until the early 2000s, and Indian
leaders commenced Cold Start–like strategizing after India’s failure to
undertake military action against Pakistan during the 2001–02 parlia-
ment attack crisis. In fact, the most dangerous security-related develop-
ment on the subcontinent—the entry of the nuclear–jihad nexus into a
second phase with Pakistan’s pursuit of a tactical nuclear capability—is
even more recent than this. Pakistan began seeking to develop tactical
nuclear weapons in response to Indian Cold Start–like planning in the
years following the parliament crisis. Its first test of the Nasr missile did
not occur until 2011. Thus, as with principal–agent problems and eco-
nomic costs, the Pakistanis have had a relatively small amount of time
to appreciate fully the external dangers that the use of militants has
caused them. It is therefore not wholly surprising that these dangers have
not yet led them to abandon a strategy that they have employed for over
sixty years.
Even if the Pakistanis’ long-standing attachment to their militant
strategy is more understandable than some would suggest, however, the
fact remains that they now must abandon it. As I explained in detail in
chapter 6, the strategy’s costs and dangers have become so great that
Pakistan faces a serious risk of catastrophe if it fails to do so. How likely
are the Pakistanis actually to alter their behavior? Under what circum-
stances might such a change possibly occur?
Some scholars believe that Pakistan’s policy may change if external
actors increase its incentives for good behavior. For example, Stephen
Cohen argues that the United States should pressure the Pakistan Army
to abandon its proxy strategy by conditioning further US military aid on
Pakistani progress in fighting militancy. The United States could make
this arrangement more palatable by agreeing, in return for Pakistani coop-
eration, to assist Pakistan in furthering the peace process with India.26
Other scholars highlight the need for the Pakistani leadership to change
137
its policy priorities. T. V. Paul, for example, maintains that Pakistani lead-
ers must de-emphasize “hyper-realpolitik” projects such as competition
with India and instead pursue developmental goals, such as improved
education, healthcare, and infrastructure, and fuller integration into the
global economy.27 The work of still other scholars suggests that Pakistan is
unlikely to give up its militant strategy unless the army undergoes a thor-
ough ideological transformation. Christine Fair, for example, argues that,
for Pakistan to cease its efforts to challenge India and overturn regional
territorial arrangements, army leaders would have to renounce their deep
commitment to revisionism in South Asia.28
All of these scholars make excellent points. One of the United States’
few levers of influence with Pakistani leaders is its ability to control the pro-
vision of large amounts of military and developmental aid to their coun-
try. Although placing conditions on this aid cannot force the Pakistanis
to behave according to US wishes, doing so may at least encourage them
to reconsider the wisdom of particularly dangerous policies. In addition,
citizens of a more prosperous, healthier, better-educated Pakistan are less
likely to view continued conflict with India as attractive and are more
likely to see the opportunity costs of such policy as prohibitive. Finally, a
Pakistani army that is ideologically dedicated to revising the status quo in
South Asia is unlikely to abandon the use of militant proxies, which are its
most effective tool for achieving this goal.
Even if all of these conditions were met, however, it is doubtful that
Pakistan would fully disassociate itself from militancy. The reason is that,
at root, Pakistani strategy is not necessitated by lax donor control over for-
eign aid, misdirected Pakistani spending priorities, or even a revisionist
culture within the Pakistan military. Pakistan’s militant strategy is neces-
sitated, rather, by Pakistan’s state-building logic. Pakistan is by definition
an oppositional state, created in contradistinction to India. As long as it
remains so, it will need to challenge existing territorial arrangements on
the subcontinent. And, given its weak status relative to India, this will
require Pakistan to continue to employ some version of its militant strat-
egy. If it does not do so, Pakistani policy will be at odds with the country’s
core national purpose. Thus, in principle, even ideological transformation
within the army would be insufficient to fully divert Pakistan from its cur-
rent course. The army’s dedication to revisionism comes from a source
deeper than its own bureaucratic culture—it comes from the country’s
oppositional state-building narrative, which informs the full range of
Pakistani policies and institutions, the military included. Reconsidering
jihad would thus require an intellectual reformulation of the Pakistani
state. As Jean-Luc Racine puts it, continued oppositional relations with
India “weigh so heavily on the experience of ‘being Pakistan’ that no one
138
could change them without rethinking the ideological foundations of the

nation.”29
Consequently, solving Pakistan’s jihad problem will be possible only
if it changes the nature of its national narrative, ensuring that opposition
to India is no longer a central purpose of the Pakistani state. Such an out-
come, of course, would be extremely difficult to achieve, as it would in
effect require Pakistan to reverse the efforts of six decades of state build-
ing. Success would not, however, be wholly impossible. States facing exis-
tential threats have made this type of thoroughgoing change in the past.
The Soviet Union offers a dramatic example. In the late 1980s, Russian
leaders renounced Communism as their organizing political principle
and dissolved the Soviet Union, relinquishing control over the territories
of the former USSR, as well as the states of Eastern Europe.
The causes of this transformation are the subject of considerable debate.
Some scholars maintain that Mikhail Gorbachev and his allies reversed
course primarily because of the severe economic and security challenges
that the Soviet Union faced. With the Soviet economy in the doldrums
and the military falling behind its increasingly technologically sophisti-
cated Western adversaries, Soviet leaders recognized that they could not
maintain their empire without inviting financial ruin or triggering a cata-
strophic war. They therefore dissolved the Soviet Union and released its
satellite states in an effort to ensure Russia’s survival. 30
Other scholars argue that the Soviet Union’s transformation resulted
from a normative change in Soviet leaders’ ideas about the nature and
purpose of their country. Over the course of their careers, Gorbachev
and his colleagues had been exposed to Western, liberal economic and
political principles, which they found far more desirable than the Soviet
Union’s command economy and totalitarian politics. They therefore
decided to transform their country from a Communist dictatorship that
forcibly controlled Eastern Europe and was locked in a security competi-
tion with the West to a democracy at peace with its neighbors and the
West, that participated in transnational trade and economic networks,
and that addressed its security concerns through deliberation rather than
coercion. 31
Regardless of which account of Soviet motivations is correct, the fact
is that Mikhail Gorbachev and his allies in the Kremlin decided that the
USSR’s state-building project was deeply flawed and unsustainable. In
Gorbachev’s words, “By the mid-1980s our society resembled a steam
boiler. There was only one alternative—either the Party … would lead
a process of change … or it would preserve and protect the former sys-
tem. In that case an explosion of colossal force would be inevitable.”
Soviet leaders therefore undertook a project of fundamental national
139
transformation. It was, as Gorbachev put it, “not some kind of cosmetic

maintenance job, but rather a complete overhaul of the extremely cen-
tralized, bureaucratized, ideology-ridden system.” They were able to
achieve this goal even after roughly seven decades of Communist rule.
Significantly, the United States, under the George H. W. Bush administra-
tion, exercised considerable restraint during this period and did not seek
to exploit the Soviets’ military and political vulnerabilities as they were
undergoing this process. 32
The Soviet case differs from Pakistan in innumerable ways and can-
not offer a roadmap to guide any potential Pakistani transformation. It
does, however, suggest three points that may be applicable to Pakistan.
First, the Soviet experience suggests that, in principle, transformational
political change in Pakistan would not be impossible. Reformulating
Pakistan’s oppositional national narrative would of course be extremely
difficult. But it is not obvious that doing so would be harder than transi-
tioning Russia away from Communist rule. 33 Second, the Soviet case sug-
gests that, whatever else might be needed, successful transformation in
Pakistan would hinge upon extraordinary personal leadership. The Soviet
Union’s material and political woes long preceded Gorbachev’s ascen-
dance. Yet, prior to Gorbachev, radical change in the vein of glasnost and
perestroika was unthinkable. Fundamental transformation became pos-
sible only with the arrival of new leadership. The same is likely to hold true
in Pakistan. A new approach to state building will require a leader who,
like Gorbachev, decides to deal with a set of long-standing challenges in a
radically different manner than his or her predecessors, and has the politi-
cal acumen to guide the process to fruition. Success would also require
Indian leaders to display some of the astuteness and sensitivity of the Bush
administration, allowing Pakistan to transform itself without seeking to
exploit its attendant vulnerabilities. In addition to avoiding triumphalist
political posturing and rhetoric, this might include at least temporarily
forgoing military expansion particularly threatening to Pakistan and
Cold Start–like doctrinal changes. Although many Indian leaders believe
that it is not their responsibility to make Pakistan feel more secure, reas-
surance in this case would be very much in their interest.
Finally, as critics of ideas-based theories of international relations point
out, there is no way of predicting when national leaders able to evange-
lize and implement new ideas will emerge. 34 The experience of the Soviet
Union, however, suggests that an existential national crisis can serve as a
catalyst for the rise of such leadership. If that is the case, then a glimmer
of hope may exist for Pakistan, which is in just such a situation now. The
crisis may create the urgency necessary to convince new leaders to gamble
on radically different approaches to conceptualizing the Pakistani state.
140
If this does not occur, it is doubtful that other policy fixes, such as condi-
tioning foreign aid or even implementing military reform, will be effective
in diverting Pakistan from its current course. The purpose and meaning
of Pakistan as an oppositional state dedicated to competing with India
will remain unchanged. Continued adherence to a strategy of jihad will be
the only logical result, just as it has been for the past six decades.
MILITANT STRATEGIES BEYOND PAKISTAN: SOME

GENERAL OBSERVATIONS
Although this study has focused on Pakistan’s strategic use of militancy,

its relevance is not limited to the Pakistani case. Its findings include a
number of general observations regarding the use of militant proxies that
can apply to other states as well. I discuss these observations briefly next.
I offer them not as iron-clad predictions, but rather as general points that
can be helpful in anticipating at least some of the potential thinking and
behavior of states that may employ militant proxy forces in the future.
First, close relationships between sponsor states and nonstate proxies
will not necessarily be the result of mistake, accident, or abstract sociohis-
torical forces. The use of nonstate proxies offers cost, operational, and bar-
gaining advantages to sponsors that can allow them to pursue goals that
would otherwise be beyond their reach. Sponsors are therefore likely to
use them deliberately, as part of sophisticated security strategies. Indeed,
they may even build larger national strategies at least partially around the
use of proxy forces. Scholars and policymakers should underestimate nei-
ther the attraction nor the potential efficacy of such an approach for pos-
sible sponsor states.
Second, the success or failure of a militant proxy strategy depends to
a significant degree on the political environment in which it is employed.
Conventional military operations can succeed through brute force, sim-
ply by overpowering an opponent. A proxy strategy, however, relies less
on force and is likely to be most successful where it has indigenous allies
or sympathizers. Thus, it will often be particularly sensitive to structural
variables such as political and economic conditions within the target
country, which will affect the population’s attitudes toward outside inter-
vention. The best operational plan may fail if those conditions are not
favorable. The more aggrieved the population is, the more likely they will
be to support or tolerate the militants, and the more likely the sponsor
state’s strategy will be to succeed.
Third, a nonstate proxy strategy can be especially effective when a spon-
sor combines it with a nuclear weapons capability. In such a combined
141
strategy, nonstate proxies can serve as offensive forces that defeat the tar-
get state’s denial efforts, while nuclear weapons provide the sponsor with
protection against target-state punishment. Sponsors that possess both of
these capabilities will be particularly formidable adversaries. Target and
other states will likely have considerable difficulty coercing them into
abandoning their militant strategies.
Fourth, control challenges inherent in principal–agent relationships
can cut in opposite directions for sponsor states. They can be problem-
atic, preventing sponsors from extracting optimal performance from their
proxies, potentially rendering sponsors unable credibly to commit to end-
ing their proxy campaign, and even allowing proxies violently to oppose
sponsor states. Control challenges also can be useful to sponsors, how-
ever, enabling them to demand a particularly high price for the difficult
task of reining in their militant allies. Sponsors therefore must manage
the principal–agent balance carefully. They need to maintain enough con-
trol over their allies to ensure that they receive acceptable service and can
credibly commit to ending their proxy campaign. Simultaneously, spon-
sors must avoid appearing to have so much control that they can call off
the militants with ease. This is likely to be a difficult balance to strike.
Fifth, a nonstate proxy strategy is likely to be especially attractive to
weak states, which lack the wherewithal to challenge stronger adversar-
ies in conventional military confrontations. This would seem to suggest
that sponsors might be convinced to abandon their proxy campaigns if
their adversaries or other states assuaged their security concerns, making
them feel less threatened and obviating their need for aggressive behav-
ior. 35 Although this approach may work in some cases, in others it may
not be effective. The reason is that although the appeal of a militant strat-
egy could be rooted in state weakness, the deep cause of the goals that
the state seeks to accomplish through the strategy might not be security
related. Rather, the state’s goals might result from ideological commit-
ments and normative concerns, rather than from security. If this is the
case, then no amount of reassurance will lead the sponsor to abandon its
proxy campaign. It will do so only if the adversary capitulates or if the
sponsor state undergoes significant ideational change.
Finally, the relationship between state weakness and adherence to a
militant strategy is paradoxical. The strategy is especially attractive to
weak states that are dissatisfied with the status quo but unable to face
stronger adversaries in a conventional conflict. That weakness, however,
can make sponsor states vulnerable to a range of serious problems associ-
ated with the use of militant proxies. These problems can be so serious as
to threaten the survival of the sponsor states. The leaders of weak states
tempted to adopt a militant proxy strategy thus need to consider their
142
policy options carefully. The weaker they are, the more likely a nonstate
proxy strategy is to appear attractive—and the more likely such a strat-
egy is to harm them severely. States that decide to follow Pakistan’s stra-
tegic logic, then, should remember that doing so may well land them in
Pakistan’s predicament.
143
NOT E S
CHAPTER 1
1. During the Cold War, US security policy sought to achieve the twin goals of deter-
rence and containment of the Soviet Union. The United States did not wish to force
the Soviets out of their existing positions in Eastern Europe and elsewhere, but
rather sought to prevent them from expanding their holdings to new territories. The
September 11, 2001, terror attacks convinced US leaders that this type of approach,
with its focus on maintaining the status quo, was no longer a feasible basis for US secu-
rity policy. The 9/11 attacks showed that nonstate actors such as terrorists could inflict
major damage on the United States. If, in the future, those actors were armed with
nuclear weapons, their impact on the United States could be truly catastrophic. Since it
was not clear that terrorists willing to sacrifice their lives could be deterred, US leaders
decided that they had to adopt a different tack, actively rolling back terrorists and the
regimes that supported them. In some cases, this required launching large-scale wars
such as those in Afghanistan and Iraq. See National Security Strategy of the United States
(September 2002), http://w ww.state.gov/documents/organization/63562.pdf.
2. See, for example, James A. Piazza, “Rooted in Poverty? Terrorism, Poor Economic
Development, and Social Cleavages,” Terrorism and Political Violence, Vol. 18
(2006), pp. 159–77; Alan B. Krueger and Jitka Maleckova, “Education, Poverty and
Terrorism: Is There a Causal Connection?” Journal of Economic Perspectives, Vol. 17,
No. 4, (Fall 2003), pp. 119–4 4; “The Mind of the Terrorist: A Review and Critique
of Psychological Approaches,” Journal of Conflict Resolution, Vol. 49, No. 1 (February
2005), pp. 3–42; Rex A. Hudson, “The Sociology and Psychology of Terrorism: Who
Becomes a Terrorist and Why?” Library of Congress Report (September 1999); Robert
Pape, Dying to Win: The Strategic Logic of Suicide Terrorism (New York: Random
House, 2006); and Max Abrahms, “What Terrorists Really Want: Terrorist Motives
and Counterterrorism Strategy,” International Security, Vol. 32, No. 4 (Spring 2008),
pp. 78–105.
3. By “Islamist militants” I mean nonstate actors violently pursuing a sociopolitical agenda
based at least partially on their interpretation of Islamic religious principles. I do not
label these actors “terrorists” because, although they often attack noncombatants, they
do not do so exclusively; they also strike military, police, and other government tar-
gets. Moreover, they often seek not only to influence particular audiences but also to
achieve battlefield victory. Although definitions vary, terrorism is usually understood
more narrowly, as violence by nonstate actors that is (1) directed against noncom-
batants, (2) intended to coerce or garner support among particular audiences rather
than to win on the battlefield, or (3) both. See, for example, Annual Country Reports on
Terrorism, U.S. Code, Title 22, sec. 2656(f); Pape, Dying to Win, p. 9; and Daniel Byman,
144
( 144 ) Notes
Deadly Connections: States That Sponsor Terrorism (Cambridge: Cambridge University

Press, 2005), pp. 8–9.
4. Of eleven terrorist attacks between 2001 and 2010 that killed more than one hundred
people and were not associated with ongoing interstate wars, four were perpetrated
by groups operating from Pakistan. Russian organizations, which had the next high-
est representation, accounted for only two incidents. Of seventeen attacks that killed
between fifty and one hundred people, nine were perpetrated by groups operating
from Pakistan. Russian organizations, again with the next highest representation,
accounted for just three incidents. Finally, al-Qaeda operatives in Pakistan coordi-
nated significant components of the September 11, 2001, terror attacks, which killed
over three thousand people and were by far the deadliest attacks during this period.
See RAND Database of Worldwide Terrorist Incidents, http://w ww.rand.org/nsrd/
projects/terrorism-i ncidents.html.
5. US Government, Report of the National Commission on Terrorist Attacks Upon the
United States (Washington, DC: US Government Printing Office, 2004), p. 167; and
Stephen Kurczy, “Top 5 Al Qaeda-l inked Militants Pakistan Has Captured,” Christian
Science Monitor, May 3, 2011.
6. See Taha Siddiqui, “Will Pakistan Release Its Osama bin Laden Abbottabad Report?”
Christian Science Monitor, October 31, 2012.
7. Souad Mekhennet and Michael Moss, “Europeans Get Terror Training Inside
Pakistan,” New York Times, September 10, 2007; and Alexandra Topping, “7/7 Gang
Given Bomb-Making Advice from Pakistan, Inquest Hears,” Guardian, February
2, 2011.
8. Angel Rabasa, Robert D. Blackwill, Peter Chalk, Kim Cragin, C. Christine Fair, Seth
Jones, Nathaniel Shestak, Brian A. Jackson, Brian Michael Jenkins, and Ashley J. Tellis,
“The Lessons of Mumbai,” RAND Occasional Paper (2009); and “26/11 Mumbai
Attacks: Pakistani Officials Reveal Attackers’ Training,” Times of India, November
11, 2012.
9. Christopher Hinton, “Iraq War Ends with a $4 Trillion IOU,” Wall Street Journal,
December 15, 2011.
10. In the Iraq war, 4,424 US personnel died, and in the Afghan conflict, 2,168 died.
At least 114,000 civilians are estimated to have died in Iraq. Approximately 14,728
Afghan civilians died since the United Nations began keeping statistics on civilian
casualties in Afghanistan in 2007. See US Department of Defense Casualty Records,
http://w ww.defense.gov/news/casualty.pdf; Dan Murphy, “The Iraq War Death Toll?
At Least 162,000 and Counting,” Christian Science Monitor, January 2, 2012; and Sayed
Salahuddin, “Afghan War’s Civilian Deaths Fell in 2012, U.N. Says,” Washington Post,
February 19, 2003.
11. Robert T. Batcher, “The Consequences of an Indo- Pakistani Nuclear War,”
International Studies Review, Vol. 6, No. 4 (December 2004), pp. 151–52.
12. “Pakistan: The World’s Most Dangerous Place,” Economist, January 3, 2008.
13. Of the respondents, 31 percent identified Pakistan as posing the greatest security chal-
lenge to the United States, while 15 percent chose China. See “The FP Survey: The
Future of War,” Foreign Policy (March 2012).
14. By asymmetric warfare I mean war waged by a militarily weaker state against a sig-
nificantly stronger adversary. For similar definitions see T. V. Paul, Asymmetric
Conflicts: War Initiation by Weaker Powers (Cambridge: Cambridge University Press,
1994), p. 3; and Ivan Arreguin-Toft, “How the Weak Win Wars: A Theory of Asymmetric
Conflict,” International Security, Vol. 26, No. 1 (Summer 2001), p. 94. India has con-
sistently exceeded Pakistan’s conventional military capabilities by a ratio of well over
2:1 and is widely recognized as being the stronger power in the Indo-Pakistani conflict
relationship. See S. Paul Kapur, Dangerous Deterrent: Nuclear Weapons Proliferation
and Conflict in South Asia (Stanford: Stanford University Press), pp. 22–23.
145
Notes ( 145 )
15. By Pakistan I mean the group of decision makers who determine the Pakistani state’s
national security policy. Despite periods of nominally civilian rule, in practice this group
has almost always been the leadership of the Pakistan Army. See Stephen P. Cohen, The
Idea of Pakistan (New Delhi: Oxford University Press, 2004), pp. 97–130; and Mazhar
Aziz, Military Control in Pakistan: The Parallel State (London: Routledge, 2008).
16. The importance of religion as a motivator has varied across cases. In some instances,
such as the 1947 Kashmir war, the desire for pillage and plunder probably drove the
militants as much as the pursuit of religious goals. In other instances, such as the cur-
rent Kashmir insurgency, one of the militant groups’ primary aims has been the pro-
motion of an Islamist sociopolitical agenda. In all cases, however, Islam has played a
significant role in motivating the militants.
17. See, for example, Ralph Peters, “America Plays the Fool in Pakistan’s Double
Game,” New York Post, July 27, 2010; James P. Farwell, “U.S. Must Turn Up the
Heat on Pakistan. Here’s How to Make That Happen,” Christian Science Monitor,
October 27, 2011; and “Pakistan’s Next Batch of Intrigues,” Economist, November
17, 2011. Reports of Pakistani duplicity are so widespread that the topic has
become the subject of popular satire. See “Pakistani Intelligence Announces Its
Full Cooperation with U.S. Forces During Upcoming Top-S ecret Drone Strike on
Al-Q aeda at 5:23 a.m. Near Small Town of Razmani in North Waziristan,” Onion,
June 9, 2011.
18. See, for example, Nicholas Schmidle, “Getting Bin Laden: What Happened That
Night in Abbottabad,” New Yorker, August 8, 2011; Lawrence Wright, “The Double
Game: The Unintended Consequences of American Funding in Pakistan,” New Yorker,
May 16, 2011; and Jeffrey Goldberg and Marc Ambinder, “The Ally from Hell,” Atlantic
(December 2011). Even this more careful category of journalistic work at times borders
on demonization. For example, a recent discussion characterized Pakistani security
policy as “terrible, horrible, no-good [and] very bad,” a “race to annihilation” fueled
by “fear, paranoia, and a deep sense of insecurity.” See Tom Hundley, “Pakistan’s
Terrible, Horrible, No-Good, Very Bad Idea to Develop Battlefield Nukes,” Foreign
Policy, September 5, 2012.
19. See, for example, Arif Jamal, Shadow War: The Untold Story of Jihad in Kashmir
(New York: Melville House, 2009); and Bruce Reidel, Deadly Embrace: Pakistan,
America, and the Future of the Global Jihad (Washington, DC: Brookings
Institution, 2011).
20. See, for example, Stephen Tankel, Storming the World Stage: The Story of the Lashkar-
e- Taiba (New York: Columbia University Press, 2011); and Carlotta Gall, The Wrong
Enemy: America in Afghanistan 2001-2014 (Boston: Houghton Mifflin Harcourt, 2014).
21. See Hussain Haqqani, Pakistan: Between Mosque and Military (Washington,
DC: Carnegie Endowment, 2005); Zahid Hussain, Frontline Pakistan: The Struggle
with Militant Islam (New York: Columbia University Press, 2008); and Hassan
Abbas, Pakistan’s Drift Into Extremism: Allah, the Army and America’s War on Terror
(New York: M. E. Sharpe, 2004).
22. See Byman, Deadly Connections; C. Christine Fair, Fighting to the End: The Pakistan
Army’s Way of War (Oxford: Oxford University Press, 2014); and T. V. Paul, The
Warrior State: Pakistan in the Contemporary World (Oxford: Oxford University
Press, 2014).
23. Aqil Shah, The Army and Democracy: Military and Politics in Pakistan (Cambridge,
MA: Harvard University Press, 2014); Husain Haqqani, Magnificent Delusions: Pakistan,
the United States, and an Epic History of Misunderstanding (New York: Public Affairs,
2013); Cohen, The Idea of Pakistan; and Owen Bennett Jones, Pakistan: Eye of the Storm
(New Haven, CT: Yale University Press, 2002). See also Ahmed Rashid, Descent into
Chaos: The United States and the Failure of Nation Building in Pakistan, Afghanistan and
Central Asia (New York: Public Affairs, 2008); and Farzana Shaikh, Making Sense of
Pakistan (New York: Columbia University Press, 2009). I do not suggest that litera-
ture in this camp wholly ignores Pakistan’s use of Islamist militants as strategic tools.
My point is simply that its main purpose is to not to explain and assess Pakistan’s
146
( 146 ) Notes
militant strategy. Partial exceptions include Praveen Swami, India, Pakistan and the
Secret Jihad: The Covert War in Kashmir, 1947-2004 (London: Routledge, 2006), which
focuses on Pakistan’s use of militancy in the Kashmir conflict; and C. Christine Fair,
Keith Crane, Christopher S. Chivvis, Samir Puri, and Michael Spirtas, Pakistan: Can
the United States Secure an Insecure State? (Santa Monica, CA: RAND, 2010), which
assesses Pakistani security strategies, including support for militancy, in light of US
strategic interests.
24. See “LeT Are Freedom Fighters, Musharraf Says,” United Press International, October
11, 2010; “Pakistani Prime Minister Nawaz Sharif ’s Address to the Nation, July 12,
1999,” http://w ww.satp.org/satporgtp/countries/pakistan/document/papers/paki-
stani/_ pm_nawaz.htm; and Barbara Crossette, “India’s Growing Peril: Kashmir and
Punjab Separatism,” New York Times, April 17, 1990.
25. The title of Hassan Abbas’s book, Pakistan’s Drift into Extremism: Allah, the Army,
and America’s War on Terror, captures this tendency to downplay the agency of the
Pakistani government in forging and exploiting links with Islamist militancy. It
characterizes the connection between Allah and the army as having resulted from
drift, rather than from the deliberate actions of Pakistani political and military
leaders.
26. On the persecution of the Ahmadis, see Antonio R. Gualtieri, Conscience and
Coercion: Ahmadi Muslims and Orthodoxy in Pakistan (Toronto: Guernica Press, 1989).
27. This shortcoming is not limited to broad overviews of Pakistani political history.
As noted earlier, other studies focusing specifically on Pakistan’s relationship with
Islamist militancy also trace the emergence of the Pakistan–m ilitant nexus to the
Zia era.
28. By “jihad” I mean violence intended at least partially to advance the actor’s view of
Islamic sociopolitical or strategic principles. Jihad, which literally means the struggle
to follow God’s will, in principle need not be violent. Waging war against nonbe-
lievers is, however, a prominent form of jihad. See Ignaz Goldziher, Introduction to
Islamic Theology and Law, trans. Andras and Ruth Hamori (Princeton, NJ: Princeton
University Press, 1981), pp. 102–3; and John L. Esposito, Islam: The Straight Path
(Oxford: Oxford University Press, 2011), pp. 113–15. The militants whom I discuss
in this book commonly use the term “jihad” to refer to their own violent activities.
I use the term in that spirit here. It is important to note that Pakistani leaders have
also used Islamist militants domestically. In doing so they have sought to curb Shiite
influence in the country, extend the government’s reach more fully across Pakistani
territory, and strike electoral bargains to gain favor with and manipulate particular
constituencies. See Zahab and Roy, Islamist Networks, p. 27; Fair et al., Pakistan, pp.
90–91; and Paul Staniland, “Beyond the Monopoly of Violence: Militancy and the
State in Pakistan,” Working Paper (September 2012). Although these domestic uses
of militancy are important, they are not my focus in this book; I am interested primar-
ily in the Pakistani state’s use of militancy against external adversaries. This external
policy has, of course, often had internal ramifications. For example, as I argue later,
the policy has helped to promote domestic political cohesion in the absence of a coher-
ent narrative justifying the existence of the Pakistani state. Such internal benefits,
however, have resulted from an externally directed strategy, rather than from a policy
that was primarily directed inward. It also is important to recognize that not all of
Pakistan’s militant allies have been Islamist in nature. For example, the Pakistanis pro-
vided extensive support to Sikh militants seeking to establish an independent state of
Khalistan in the Indian Punjab, and to rebels in India’s northeastern state of Nagaland.
See Haqqani, Between Mosque and Military, pp. 270–72; D. B. Shekatkar, “India’s
Counterinsurgency Campaign in Nagaland,” in Sumit Ganguly and David P. Fidler,
eds., India and Counterinsurgency: Lessons Learned (Abingdon, United Kingdom:
Routledge, 2009), pp. 9–27; and Subir Bhaumik, Troubled Periphery: Crisis of India’s
North East (New Delhi: Sage, 2009), pp. 153–81. Such campaigns inflicted consider-
able costs on India, and many of the arguments that I make in this book would apply
147
Notes ( 147 )
to them as well. Nonetheless, they are not my primary concern in this project, which
focuses on Pakistan’s use of Islamist militancy and the central role that it has played in
Pakistani grand strategy.
29. See Barry Posen, “The Case for Restraint,” American Interest (November/December
2007); and Robert Art, A Grand Strategy for America (Ithaca, NY: Cornell University
Press, 2004), pp. 1–2 .
30. Nuclear weapons have helped Pakistan to aggressively challenge the status quo, but
only when employed in combination with other types of capabilities. Specifically,
nuclear weapons have insulated Pakistan against Indian retaliation in response to
provocations by Pakistani conventional forces and by Pakistan-supported militants.
This has enabled the Pakistanis to use these forces more aggressively than they other-
wise would have. Pakistani nuclear weapons themselves, however, played a defensive
role, deterring any potential Indian retaliatory attack against Pakistan. See generally
Kapur, Dangerous Deterrent, pp. 32–63.
31. The Pakistan Army’s doctrine, known as offensive defense, focused on repelling an
Indian offensive by launching attacks into Indian territory. Although blunting an
Indian conventional military attack of course remains a major goal, the Pakistan
Army appears recently to have shifted its focus to defeating insurgents operating
on Pakistani territory. See R. S. N. Singh, “Pakistan’s Offensive Defense Strategy,”
Indian Defence Review, February 18, 2011; Kapur, Dangerous Deterrent, p. 51; Stephen
P. Cohen, The Pakistan Army (New Delhi: Oxford University Press, 1998), p. 145;
“New Doctrine: Army Identifies ‘Homegrown Militancy’ as Biggest Threat,” Express
Tribune, January 3, 2013; and Mohammad Jamil, “Military Doctrine Reviewed,”
Nation, January 9, 2013.
32. On these cases see Byman, Deadly Connections, pp. 79–115; Central Intelligence
Agency, “Soviet Support for International Terrorism and Revolutionary Violence,”
Special National Intelligence Estimate, May 27, 1981; and Steve Coll, Ghost Wars: The
Secret History of the CIA, Afghanistan and Bin Laden from the Soviet Invasion to September
10, 2001 (New York: Penguin, 2004).
33. General scholarship on states’ strategic use of nonstate actors has sought primar-
ily to determine the policy’s impact on the actors, rather than on the states that
deploy them. Daniel Byman, for example, asks what effect state support has had on
the interests, capabilities, and behavior of leading terrorist groups. Although he also
seeks to determine why states support terrorists, he does not address specifically
the effects of state sponsorship on the strategic interests of the sponsors themselves.
Byman, Deadly Connections, pp. 1–2 0, esp. p. 16; and Daniel Byman, Peter Chalk,
Bruce Hoffman, William Rosenau, and David Brannan, Trends in Outside Support for
Insurgent Movements (Santa Monica, CA: RAND, 2001). Pakistan-specific work also
largely ignores the question of how the strategic use of Islamist militancy has affected
Pakistan’s own domestic and international interests. As noted earlier, scholarship
focusing directly on the Pakistan–m ilitant connection has provided detailed empiri-
cal accounts of Pakistan’s use of militancy, tried to explain the broad social and politi-
cal trends underlying it, and discussed the tactical use of militancy in Kashmir. But
this literature has not carefully evaluated Islamist militancy’s utility as a decades-old
component of Pakistani grand strategy.
34. Timothy Hoyt, “Pakistani Nuclear Doctrine and the Dangers of Strategic
Myopia,” Asian Survey, Vol. 41, No. 6 (November–D ecember 2001), pp. 956–7 7.
See also Ahmad Faruqui, Rethinking the National Security of Pakistan: The Price
of Strategic Myopia (Aldershot: Ashgate, 2003); Altaf Gauhar, “Four Wars, One
Assumption,” Nation, September 5, 1999; and Shaikh, Making Sense of Pakistan,
pp. 207– 8 .
35. Fair, Crane, Chivvis, Puri, and Spirtas, Pakistan, p. 112. For partial exceptions, see
Byman, Deadly Connections, pp. 177–78; Mariam Abou Zahab and Olivier Roy,
Islamist Networks: The Afghan-Pakistan Connection (New York: Columbia University
Press, 2004), p. 53; and Fair, Fighting to the End, p. 6.
148
( 148 ) Notes
CHAPTER 2
1. In principle, proxy forces need not be limited to nonstate actors; one state could also
use another state’s conventional military forces to promote its own strategic interests.
During the Cold War, for example, the United States and Soviet Union often used
other states, such as North and South Korea and Vietnam, to fight wars promoting US
and Soviet interests while avoiding the dangers of direct superpower confrontation.
Although a strategy of using nonstate proxies shares some similarities with the use of
state-level proxies, the two approaches differ in important ways. For example, nonstate
forces are likely to be smaller and less technologically sophisticated than conventional
militaries. Using such forces as proxies will probably be considerably less expensive
and more easily deniable than backing another state’s conventional military. Nonstate
proxies can also offer their sponsors more operational flexibility than state-level forces.
For example, militants can be deployed to attack targets in distant, powerful coun-
tries that state-level proxies would be unable or unwilling to strike. Differences and
similarities aside, Pakistan’s policy has been to use nonstate proxies, rather than state-
level proxy forces, to promote its strategic interests. Since I wish to explain and assess
Pakistani behavior, my specific focus in this book is states’ use of nonstate actors such
as militants or terrorists as proxy forces.
2. Answering this question also can improve our understanding of cases beyond Pakistan
and South Asia. Pakistan, after all, is not the only state that might wish to use militant
proxies to further its strategic interests. Indeed, the frequency of interstate war has
declined significantly since World War II; it is today a rare event. Increasingly, states
have substituted alternative strategies for outright conflict with their adversaries. The
use of nonstate proxies is one such alternative. An understanding of the general logic
of a proxy strategy can help us to anticipate the circumstances under which such a
strategy might be attractive and how it might be employed. See Idean Salehyan, “The
Delegation of War to Rebel Organizations,” Journal of Conflict Resolution, Vol. 54, No.
3 (2010), p. 494; and Byman, Deadly Connections, pp. 1–4.
3. See, for example, Walter Laqueur, The New Terrorism: Fanaticism and the Arms of
Mass Destruction (New York: Oxford University Press, 1999); Caleb Carr, The Lessons
of Terror: A History of Warfare Against Civilians (New York: Random House, 2002);
Abrahms, “What Terrorists Really Want,” pp. 78–105; Andrew H. Kydd and Barbara
F. Walter, “The Strategies of Terrorism,” International Security, Vol. 31, No. 1 (Summer
2006), pp. 49–80; and Pape, Dying to Win.
4. See, for example, David B. Carter, “A Blessing or a Curse: State Support for Terrorist
Groups,” International Organization, Vol. 66, No. 1 (January 2012), pp. 129–51;
Marianne Heiberg, Brendan O’Leary, and John Tirman, eds., Terror, Insurgency, and
the State: Ending Protracted Conflicts (Philadelphia: University of Pennsylvania Press,
2007); Daniel Byman, Deadly Connections; and Byman, Chalk, Hoffman, Rosenau,
and Brannan, Trends in Outside Support for Insurgent Movements. Scholarship that
does directly address the issue of sponsor-state interests focuses primarily on the
difficulties that sponsors may face in getting militant proxies to do their bidding. It
usually does not systematically explore the military advantages that would lead a
state to utilize nonstate proxies in the first place. See, for example, Boaz Atzili and
Wendy Pearlman, “Triadic Deterrence: Coercing Strength, Beaten by Weakness,”
Security Studies, Vol. 21, No. 2 (2012), pp. 301–35; and Thomas H. Johnson,
“Financing Afghan Terrorism: Thugs, Drugs, and Creative Movements of Money,”
in Jeanne K. Giraldo and Harold A. Trinkunas, eds., Terrorism Financing and State
Responses: A Comparative Perspective (Stanford: Stanford University Press, 2007),
pp. 111–13.
149
Notes ( 149 )
5. Salehyan, “The Delegation of War to Rebel Organizations,” pp. 495–9 7.

6. This may have resulted from the fact that much of this literature was produced dur-
ing the Cold War, when such state-to-state proxy relationships occupied center stage.
See, for example, Bertil Duner, “Proxy Intervention in Civil Wars,” Journal of Peace
Research, Vol. 18, No. 4 (1981), pp. 353–61; and Yaacov Bar-Simon-Tov, “The Strategy
of War by Proxy,” Cooperation and Conflict, Vol. 19 (1984), pp. 263–73.
7. I assume a weak sponsor and strong adversary because, as I explain later, a militant
proxy strategy is likely to be especially attractive to a weak state seeking to challenge a
stronger opponent. Not surprisingly, this is Pakistan’s position relative to India.
8. Byman, Deadly Connections, p. 5. See also Sehehyan, “The Delegation of War to Rebel
Organizations,” p. 503; and Chris Loveman, “Assessing the Phenomenon of Proxy
Intervention,” Conflict, Security and Development, Vol. 2, No. 3 (2002), p. 33.
9. There is disagreement on this point in the literature. Some scholars hold that proxy
strategies will be appealing to relatively strong states, which will be better able to exe-
cute them than weaker countries. Others maintain that weak states will tend to prefer
proxy strategies, to compensate for their material disadvantages. See Salehan, “The
Delegation of War to Rebel Organizations,” p. 508; and Cecily G. Brewer, “Peril by
Proxy: Negotiating Conflicts in East Africa,” International Negotiation, Vol. 16 (2011),
p. 142. I argue that, although they may face execution problems, weak states are most
likely to prefer the use of militant proxies, as the need to minimize costs will be more
urgent for them than for strong states. As I will show, chronic state weakness is what
motivated Pakistan to develop and deploy a proxy strategy.
10. A weak state would also realize these benefits by using a militant proxy strategy against
opponents that were equally or less powerful. The advantages of such a strategy are
more important, however, where a weak state faces a more powerful adversary.
11. For discussions on the concepts of denial and punishment, see Robert Jervis, The Illogic
of American Nuclear Strategy (Ithaca, NY: Cornell University Press, 1989), pp. 8–13;
Robert A. Pape, Bombing to Win: Air Power and Coercion in War (Ithaca, NY: Cornell
University Press, 1996), pp. 13, 19; and Thomas C. Schelling, Arms and Influence (New
Haven, CT: Yale University Press, 1966), pp. 1–3 4.
12. The level of danger for the weak state is not fixed. It varies with the power differential
between the two states. The larger the difference is in power between the adversaries,
the greater the danger for the weaker state.
13. Although the strong state would inflict some of these costs on the weak state’s home-
land, rather than on its military forces, the costs should still be considered as result-
ing from the strong state’s denial campaign. This is the case because these costs would
arise directly from escalation of the initial battlefield confrontation. They would not
result from separate strong-state attempts to inflict pain on the weak state unrelated to
the battlefield confrontation.
14. Deterrence is the use of conditional threats to convince an opponent not to take a par-
ticular proscribed action. Deterrence occurs when, in the opponent’s judgment, the
expected costs of action (the cost of an action multiplied by the probability of actually
incurring that cost) outweigh its expected benefit (the benefit of an action multiplied
by the probability of actually realizing that benefit). In a case of deterrence by denial,
the opponent arrives at this judgment because of the high likelihood that it will face an
extremely effective defense. The opponent could reach this judgment even if it would
achieve significant benefits by launching a successful attack. As long as the likelihood
of actually achieving success and realizing those benefits is low, the expected benefit of
attacking is likely to be low as well. See Jervis, The Illogic of American Nuclear Strategy,
pp. 8–13; Pape, Bombing to Win, pp. 13, 19; and Schelling, Arms and Influence, pp. 1–3 4.
15. It is also possible, however, that a strong-state punishment campaign would have the
opposite effect on weak-state leaders’ political standing. There is significant evidence
that attacks on their homeland can unite populations around the war effort and harden
their attitudes toward the enemy. Strategic bombing campaigns, for example, have
150
( 150 ) Notes
generally not led states to surrender and instead have increased their resolve. See Pape,
Bombing to Win.
16. On the challenges of preventing terrorist penetration of national borders, see, for
example, K. Jack Riley, “Border Security and the Terrorist Threat,” Testimony before
the House Homeland Security Committee, RAND Corporation Testimony Series,
August 8, 2006.
17. See C. Christine Fair, Urban Battlefields of South Asia: Lessons Learned from Sri Lanka,
India, and Pakistan (Santa Monica, CA: RAND, 2004), p. 6.
18. See Robert A. Pape, “The Strategic Logic of Suicide Terrorism,” American Political
Science Review, Vol. 97, No. 3 (August 2003), p. 346.
19. India has long suffered from this problem, which has been exacerbated by its use of
heavy-handed counterinsurgency tactics. Despite recent improvements stemming in
part from New Delhi’s attempts to foster economic development and political liber-
alization in the region, anti-I ndian sentiment in Kashmir remains high. See Parvaiz
Bukhari, “India-Controlled Kashmir Seethes as Curfew Extends to Seventh Day,”
Christian Science Monitor, February 15, 2013; and Zahid Rafiq, “Armed Resistance
Reemerging in Kashmir,” Christian Science Monitor, June 25, 2013.
20. See Praveen Swami, “Terrorism in Jammu and Kashmir in Theory and Practice,”
in Kanti P. Bajpai and Harsh V. Pant, eds., India’s National Security: A Reader (New
Delhi: Oxford University Press, 2013), p. 125.
21. For example, Pakistan has consistently denied that it uses Islamist militants as
proxy forces against India. See Jayshree Bajoria and Eben Kaplan, “The ISI and
Terrorism: Behind the Accusations,” Council on Foreign Relations Backgrounder,
May 4, 2011.
22. Terrorists or militant organizations, by contrast, may issue manifestos explaining
their philosophy and justifying their behavior. Al-Qaeda, for example, has offered
numerous public justifications for its global terrorist campaign. See Osama Bin Laden,
“Declaration of Jihad, August 23 1996,” in Bruce Lawrence, ed., Bruce Howarth, trans.,
Messages to the World: The Statements of Osama Bin Laden (London: Verso, 2005).
23. Other scholars have also noted that deniability can insulate sponsor states from target-
state retaliation. See, for example, Byman, Deadly Connections, p. 22. I unpack its oper-
ational advantages in greater detail, however.
24. Identifying high-value targets can present a serious obstacle to the prosecution of
successful punishment campaigns. During the Cold War, US strategists expended
considerable resources attempting to determine what targets to hold at risk to deter
Soviet aggression. Did Soviet leaders most value their populations? Their military
assets? Their own lives? The answers to these questions were hotly contested and
remained the subject of energetic debate throughout the Cold War period. See Charles
L. Glaser, Analyzing Strategic Nuclear Policy (Princeton, NJ: Princeton University
Press, 1990), pp. 49–57; and Lawrence Freedman, The Evolution of Nuclear Strategy
(New York: Palgrave Macmillan, 2003), pp. 132–36.
25. Indeed, the entire conflict process may be seen as one of bargaining, where fighting
erupts because the two sides lack information about each other’s capacity and resolve
that would enable them to strike a mutually acceptable deal prior to the outbreak of
hostilities. Once capability and resolve become clear through combat, the antagonists
are better able to negotiate an acceptable settlement. See James D. Fearon, “Rationalist
Explanations for War,” International Organization, Vol. 49, No. 3 (Summer 1995),
pp. 379–414.
26. These specific phenomena are known as adverse selection and moral hazard problems.
See Joseph A. Stiglitz, “Principal and Agent,” in John Eatwell, Murray Milgate, and
Peter Newman, eds., The New Palgrave: A Dictionary of Economics (London: Macmillan
Press Limited, 1987), pp. 966–71; and Stephen A. Ross, “The Economic Theory of
Agency: The Principal’s Problem,” American Economic Review, Vol. 63, No. 2 (May
1973), pp. 134–39.
27. Navin A. Bapat, “Understanding State Sponsorship of Militant Groups,” British Journal
of Political Science, Vol. 42, No. 1 (January 2012), p. 5.
151
Notes ( 151 )
28. This discussion suggests that a degree of weakness is useful to the sponsor state as it
attempts to extract maximum payment from its adversary in exchange for ending a
proxy campaign. A relatively strong sponsor that exercised tight control over its prox-
ies would have less difficulty in calling them off and thus would probably have to settle
for a smaller payoff than a weaker sponsor for which calling off proxies was a more dif-
ficult task. It is important to note, however, that while a degree of weakness will be use-
ful to the sponsor state, excessive weakness will undermine its bargaining efforts. If the
sponsor is too weak, it will be unable to credibly commit to reining in its proxies; the
adversary simply will not believe the sponsor’s promise to end the conflict in exchange
for payment. Ideally, then, the sponsor would be sufficiently weak that controlling
its proxies is difficult and deserving of a payment premium, but not so weak that its
promise to end the proxy campaign lacks credibility. See Bapat, “Understanding State
Sponsorship of Militant Groups,” p. 16.
29. See Ibid.
30. Militant proxies could also diverge from sponsor-state preferences in the opposite
direction, behaving less aggressively than their sponsors would like. This would not be
particularly dangerous to the sponsor state, however. It would merely result in a lack
of progress toward the sponsor’s goals. To remedy the situation, the sponsor would
simply have to look for other, more aggressive proxies to employ.
31. Note that even if the proxy acted without the sponsor’s approval, and the sponsor
denied responsibility and disavowed the proxy’s action, the adversary could blame
the sponsor and attack it anyway. This could occur because the adversary’s leadership
does not believe sponsor-state denials. Alternatively, it could occur because domestic
political pressure to retaliate becomes overwhelming in the adversary state, forcing its
leaders to respond militarily, regardless of their beliefs regarding sponsor-state respon-
sibility for the proxy’s actions.
32. Stephen Tankel, “Lashkar-e-Taiba: Past Operations and Future Prospects,” National
Security Studies Policy Paper, New America Foundation (April 2011), p. 3.
33. Geoffrey Kambere, Puay Hock Goh, Pranav Kumar, and Fulgence Msafir, “The
Financing of Lashkar-e-Taiba,” Counterterrorism Exchange, Vol. 1, No. 1 (August
2011), p. 7; and Press Trust of India, “Lashkar’s Annual Military Budget Is $5.2
Million,” Hindu, December 6, 2010.
34. Tankel, “Lashkar-e-Taiba,” pp. 6–12; and Press Trust of India, “Lashkar’s Annual
Military Budget.”
35. Kambere, Goh, Kumar, and Msafir, “The Financing of Lashkar-e-Taiba, pp. 12–13.
36. Note that Pakistan was estimated to have spent approximately $50 million per year
supporting various Kashmiri militant groups during the mid-to late 1990s. This sug-
gests that the range of estimates offered earlier is well within the realm of plausibility.
See ibid., p. 7.
37. The 1999 Kargil war, in which forces from the Pakistan Army’s Northern Light
Infantry (NLI) clashed with the Indian Army, is a partial exception. Even in this case,
however, the Pakistanis avoided the use of wholly conventional military forces. Prior
to Kargil, the NLI was a paramilitary unit within the Pakistan Army. It did not become
a full-fledged army unit until after the conflict, probably as a reward for its faithful
service during Kargil. See John H. Gill, “Military Operations in the Kargil Conflict,”
in Peter R. Lavoy, ed., Asymmetric Warfare in South Asia: The Causes and Consequences
of the Kargil Conflict (Cambridge: Cambridge University Press, 2009), p. 120.
38. See S. Paul Kapur, “The Kashmir Dispute: Past, Present, and Future,” in Sumit
Ganguly, Andrew Scobell, and Joseph Chinyong Liow, eds., Routledge Handbook
of Asian Security Studies (Abingdon, UK: 2010), pp. 107–8; Press Trust of India,
“All-Weather LoC Fence in Kashmir: Experts to Visit Border Soon,” Times of India,
October 21, 2012; Times of India News Service, “Rashtriya Rifles to Have 30
New Battalions,” Times of India, August 2, 2001; and Times News Network, “900
Black Cat Commandos out of VIP Duty to Fight Terror,” Economic Times, October
16, 2012.
39. See Indian Express, Inside 26/11 (New Delhi: Express Group, 2009).
152
( 152 ) Notes
40. See Parvaiz Bukhari, “Human Rights Report Names Names in Kashmir, Invokes
International Law,” Christian Science Monitor, December 6, 2012.
41. Indeed, commentators have gone so far as to suggest that these cultural ties might help
to drive a more substantive rapprochement between the two countries. See C. Raja
Mohan, “Punjabiyat and India-Pakistan Ties,” Hindu, February 16, 2004.
42. High-level Indian and Pakistani decision makers have made this point to the author
on numerous occasions. It is important to emphasize that my claim here is not that
shared history would necessarily prevent India and Pakistan from launching large-
scale attacks against each other’s populations—only that their common past would be
an impediment to doing so and make such an outcome less likely.
43. See “India’s Muslims: Growing, and Neglected,” Economist, March 2, 2013.
44. India has a long history of such communal unrest, beginning with large-scale Hindu–
Muslim violence that may have killed up to one million people during partition.
More recently, hundreds died in Hindu–Muslim riots in 2002 in Gujarat following
a dispute over a mosque built on the site of a Hindu temple. See Christophe Jaffrelot,
“Communal Riots in Gujarat: The State at Risk?” Heidelberg Papers in South Asian
and Comparative Politics, Working Paper 17 (July 2003).
45. See Ashley J. Tellis, Stability in South Asia (Santa Monica, CA: RAND, 1997).
46. In the late 1980s, the Pakistanis developed a de facto nuclear capability; although
they had not yet tested nuclear weapons, they probably could have assembled one in
short order if necessary. Pakistan possessed an overt nuclear capability after the 1998
nuclear tests. See Kapur, Dangerous Deterrent, pp. 10–11.
47. Shortly after Kargil, Indian leaders, including Army Chief General V. P. Malik and
Defence Minister George Fernandes, articulated the principles of a new, “limited war”
concept, under which India would fight and prevail over Pakistan while controlling the
escalation ladder and ensuring that the dispute did not result in a nuclear confronta-
tion. See “Closing Address by Gen. V.P. Malik, COAS and Chairman Chiefs of Staff
Committee, at a National Seminar on the Challenge of Limited War: Parameters and
Options,” Institute for Defence Studies and Analyses, January 2000, http://w ww.idsa-
india.org/chief6-2 000.html; and C. Raja Mohan, “Fernandes Unveils ‘Limited War’
Doctrine,” Hindu, January 24, 2000.
48. During the Kargil conflict, for example, the Indian government planned to allow its
forces to cross the Line of Control into Pakistani Kashmir if necessary, but refused
to consider violating the international border and entering Pakistan proper. During
a subsequent Indo-Pakistani crisis following a terrorist attack on the Indian parlia-
ment, the Indians planned a series of shallow retaliatory attacks across the interna-
tional border, while rejecting the option of striking deeper into Pakistani territory. See
Kapur, Dangerous Deterrent, pp. 120–31. As noted earlier, nuclear weapons’ limitation
of Indian military options has emboldened the Pakistanis, who believe that they can
continue to pursue their militant strategy in relative safety, insulated from the danger
of Indian retaliation by their nuclear shield. See ibid., pp. 14–31.
49. Attacks on militant training camps in Pakistani Kashmir are often suggested as a
potential means of Indian retaliation against Pakistan. See, for example, Sadanand
Dhume, “India Stands Up to Pakistan: A Delhi Less Patient with Islamabad Is Good
for Regional Stability,” Wall Street Journal, January 16, 2013.
50. For an encyclopedic listing see K. Santhanam, Sreedhar, Sudhir Saxena, and Manish,
Jihadis in Jammu and Kashmir: A Portrait Gallery (New Delhi: Sage Publications, 2003).
51. This is true even of extremely detailed, carefully researched analyses. See, for exam-
ple, Swami, who says only that Islamist militants receive “official support” from the
“Pakistan military and intelligence establishment.” Swami, “Terrorism in Jammu and
Kashmir,” in Bajpai and Pant, eds., India’s National Security, pp. 125–2 6.
52. For details on this crisis see V. K. Sood and Pravin Sawhney, Operation Parakaram: The
War Unfinished (New Delhi: Sage Publications, 2003).
53. Kapur, Dangerous Deterrent, pp. 131–39.
54. This is not to suggest that potential conventional military costs and nuclear dangers
had no effect on Indian forbearance during this crisis. Such concerns, particularly
153
Notes ( 153 )
regarding the possibility of nuclear escalation, mattered. For example, nuclear dan-
ger led the Indians to rule out launching a large-scale conventional attack that could
threaten Pakistan with catastrophic defeat. Nonetheless, the belief that coercive
diplomacy had worked against Pakistan was crucial to India’s decision to stand down;
without it, the Indians would likely have undertaken military operations against the
Pakistanis. This Indian belief, in turn, resulted from assurances by American officials,
who were convinced that the Pakistan government was sufficiently separate from the
militants that it should not be punished for launching the attacks and would work to
prevent similar militant violence in the future. See ibid.
55. Agence France Presse, “No Rush Against Anti-I ndia Militants Says Musharraf,” Dawn,
November 10, 2010.
56. Lionel Beehner, “Musharraf ’s Taliban Problem,” Council on Foreign Relations
Backgrounder, September 11, 2006.
57. Kanchan Lakshman, “The Expanding Jihad,” South Asia Intelligence Review, Vol. 6,
No. 2 (February 18, 2008), http://w ww.satp.org/satporgtp/sair/A rchives/6 _ 32.htm;
Lashkar-e-Taiba, “Hum Jihad Kyon Kar Rahein Hein” [Why Do We Wage Jihad?],
S. Paul Kapur, trans., p. 2; and Lakshman, “The Expanding Jihad.”
58. See C. Christine Fair and Seth G. Jones, “Pakistan’s War Within,” Survival, Vol. 51, No.
6 (December 2009/January 2010), pp. 162–69, 173–78, 182.
59. See Pervez Musharraf, In the Line of Fire: A Memoir (New York: Free Press, 2006), pp. 201–
204. Stephen Philip Cohen, “The Nation and the State of Pakistan,” Washington Quarterly,
Vol. 25, No. 3 (Summer 2002), pp. 115–16; Gul, The World’s Most Dangerous Place, pp. 16–
17, 159–60; Hussain, Frontline Pakistan, pp. 33–75, 110–13; Samina Yasmeen, “Pakistan’s
Kashmir Policy: Voices of Moderation?” Contemporary South Asia, Vol. 12, No. 2 (June
2003), p. 12; and Ziring, “Weak State, Failed State, Garrison State,” p. 190.
60. Douglas Lynd, The Education System in Pakistan: Assessment of the National Education
Consensus (Islamabad: UNESCO, 2007), pp. 7, 8, 16.
61. S. Paul Kapur’s discussions with senior Indian strategists, New Delhi, July and
September 2010; and Walter C. Ladwig III, “A Cold Start for Hot Wars? The Indian
Army’s New Limited War Doctrine,” International Security, Vol. 32, No. 3 (Winter
2007/08), pp. 159–6 0, 164–6 6.
62. For an extensive debate over the likelihood of such escalation see Sumit Ganguly and
S. Paul Kapur, India, Pakistan, and the Bomb: Debating Nuclear Stability in South Asia
(New York: Columbia University Press, 2010).
63. This is not to suggest that militants played no role in the Bangladesh war. As I explain
in chapter 4, both sides employed such forces extensively. Nonetheless, the conflict
was, on balance, primarily conventional in nature.
CHAPTER 3
1. The partition, which split independent India from Pakistan in 1947, triggered large-
scale Hindu–Muslim violence and population shifts. The division claimed between
one hundred thousand and one million lives, and approximately fifteen million people
left their homes to resettle in the new Indian or Pakistani states. See Yasmin Khan, The
Great Partition: The Making of India and Pakistan (New Haven, CT: Yale University
Press, 2008).
2. The territorial division between India and Pakistan had been decided by two commis-
sions headed by English jurist Sir Cyril Radcliffe over a roughly one-month period dur-
ing the summer of 1947 in New Delhi. One of Radcliffe’s main qualifications for the
job was that he had never been to India and was unfamiliar with the issues and actors
involved in the partition. He thus was thought likely to act impartially. Nonetheless,
Radcliffe’s decisions were the source of significant controversy. See Alistair Lamb,
Kashmir: A Disputed Legacy (Karachi: Oxford University Press, 1993), pp. 103–17.
3. See Khan, The Great Partition, pp. 143–6 6.
154
( 154 ) Notes
4. Ian Talbot, Pakistan: A Modern History (New Delhi: Oxford University Press, 1998),

pp. 98–9 9; Ayesha Jalal, The State of Martial Rule: The Origins of Pakistan’s Political
Economy of Defense (Cambridge: Cambridge University Press, 1990), pp. 42, 47.
5. Muhammad Ali Jinnah, “Achievements of the First Year: A Firm Basis for Optimism,”
Message to the Nation, August 14, 1948; and “Students’ Role in Nation Building,”
Speech at Dhaka University, March 24, 1948, in Quaid-I-A zam Mohammad Ali
Jinnah: Speeches as Governor- General of Pakistan, 1947– 1948, pp. 181 and 107.
See also Safdar Mahmood, Pakistan: Political Roots and Development 1947-1999
(Oxford: Oxford University Press, 2000), pp. 185–86.
6. Jalal, The State of Martial Rule, pp. 42, 47; Ayesha Jalal, Democracy and Authoritarianism
in South Asia: A Comparative and Historical Perspective (Cambridge: Cambridge
University Press, 1995), p. 18; Talbot, Pakistan, p. 100; and Cohen, The Idea of Pakistan,
p. 41.
7. Christophe Jaffrelot, “Islamic Identity and Ethnic Tensions,” in Christophe Jaffrelot,
ed., Gillian Beaumont, trans., A History of Pakistan and Its Origins (London: Anthem
Press, 2000), pp. 12–13.
8. Ibid., pp. 10–11.
9. Stanley Wolpert, Jinnah of Pakistan (New Delhi: Oxford University Press, 1984),
pp. 27, 35, 45. The reasons for Jinnah’s change from a staunch advocate for Hindu–
Muslim unity to a champion of communal separatism are the source of considerable
speculation. Animosity between Jinnah and Congress leader Jawaharlal Nehru, whom
Jinnah often found to be condescending and dismissive, may have played a role in this
shift. See Wolport, Jinnah, pp. 147–4 8.
10. Muhammad Iqbal, “1930 Presidential Address to the 25th Session of the All-I ndia
Muslim League,” in Latif Ahmad Sherwani, ed., Speeches, Writings and Statements of
Iqbal (Lahore: Iqbal Academy Pakistan, 2005), pp. 9, 11.
11. Jalal, Democracy and Authoritarianism, p. 15.
12. Quoted in Wolpert, Jinnah, pp. 128, 147.
13. Jalal, Democracy and Authoritarianism, pp. 15–16, 25.
14. Ayesha Jalal, The Sole Spokesman: Jinnah, the Muslim League and the Demand for
Pakistan (Cambridge: Cambridge University Press, 1985), pp. 171–72.
15. Haqqani, Pakistan: Between Mosque and Military, p. 5. See also Jalal, Democracy and
Authoritarianism, p. 14; and Mazhar Aziz, Military Control in Pakistan: The Parallel
State (London: Routledge, 2008), pp. 19–2 0.
16. Iqbal, “1930 Presidential Address,” p. 15.
17. Jalal, Democracy and Authoritarianism, p. 14. See also Mazhar Aziz, Military Control in
Pakistan, pp. 19–2 0; and Haqqani, Pakistan, p. 5.
18. Ayesha Jalal, The Sole Spokesman, pp. 2–3. See also Talbot, Pakistan, pp. 66–67, 85, 93,
97; Cohen, The Idea of Pakistan, p. 25; Jaffrelot, “Islamic Identity and Ethnic Tensions,”
p. 10; and Jalal, Democracy and Authoritarianism, p. 23.
19. Mohammad Ali Jinnah, “A Call to Duty,” October 11, 1947, Karachi, Pakistan, in
Quaid-I-A zam Mohammad Ali Jinnah, p. 31.
20. See, for example, Mohammad Ali Jinnah, “Presidential Address to the Muslim League,
Lucknow, 1937,” http://w ww.columbia.edu/itc/mealac/pritchett/0 0islamlinks/t xt_
jinnah_ lucknow_1937.html; and Jinnah, “Presidential Address to the Muslim League,
Lahore, 1940,” http://w ww.columbia.edu/itc/mealac/pritchett/0 0islamlinks/t xt_
jinnah_lahore_1940.html
21. Jinnah, “Presidential Address to the Constituent Assembly of Pakistan on 11th August
1947,” in Quaid-I-A zam Mohammad Ali Jinnah, p. 17.
22. Talbot, Pakistan, p. 5.
23. Ibid., p. 94; Haqqani, Pakistan, pp. 5–6, 26–27; and Shaikh, Making Sense of Pakistan,
pp. 1–4, 14–15.
24. Paul, Warrior State, p. 95.
155
Notes ( 155 )
25. “Sixty Years of U.S. Aid to Pakistan: Get the Data,” Guardian, July 11, 2011; Cohen,
The Idea of Pakistan, pp. 55–56; Talbot, Pakistan, p. 96; and Jinnah, “Students’ Role in
Nation Building,” in Quaid-i-A zam Mohammad Ali Jinnah, pp. 106–12.
26. See Kapur, Dangerous Deterrent, p. 48.
27. George Cunningham, “Frontier Policy,” Memo to Liaquat Ali Khan, September 20,
1947, India Office Records, British Library, MS EUR D 670/13. See also Lawrence
Ziring, Pakistan: At the Crosscurrent of History (Oxford: Oneworld Publications, 2003),
pp. 69–70; and Ilhan Niaz, The Culture of Power and Governance of Pakistan, 1947-2008
(Oxford: Oxford University Press, 2010), pp. 77–79.
28. See Haqqani, Pakistan, pp. 2–18; and Mohammed Ayoob, “Two Faces of Political
Islam: Iran and Pakistan Compared,” Asian Survey, Vol. 19, No. 6 (June 1979),
pp. 536–37.
29. Iqbal Singh Sevea, The Political Philosophy of Muhammad Iqbal: Islam and Nationalism
in Late Colonial India (Cambridge: Cambridge University Press, 2012), pp. 127, 128,
138, 147.
30. Allah Bukhsh K. Brohi, “Preface,” in S. K. Malik, The Quranic Concept of War (New
Delhi: Adam Publishers & Distributors, 1992).
31. Not all Muslim religious thinkers shared these views regarding Islam’s relationship
with the nation-state. The deobandi scholar Hussain Ahmad Madani, for example,
believed that the Koran and other sources demonstrated Islam’s compatibility
with territorially based political communities. See Seava, The Political Philosophy
of Muhammad Iqbal, pp. 1, 157. Even if views such as Iqbal’s were not unanimous
among philosophers, however, they did have a significant influence on Pakistani
state building and came to be reflected in official thinking on this issue. It is worth
noting in this vein that The Quranic Concept of War is a book authored by a serv-
ing Pakistani general officer, includes a forward by General Zia, contains a pref-
ace by a senior Pakistani diplomat who became advocate general of Pakistan, and
became assigned reading in the Pakistan Army. See Patrick Poole and Mark Hanna,
“Publisher’s Preface” and Joseph Myers, “Introduction,” in The Quranic Concept of
War. See also Joseph C. Myers, “The Quranic Concept of War,” Parameters (Winter
2006– 0 7), pp. 108–21.
32. Olivier Roy, “Islam and Foreign Policy: Central Asia and the Arab-Persian World,” in
Jaffrelot, ed., A History of Pakistan and Its Origins, p. 134.
33. Shaikh, Making Sense of Pakistan, p. 180.
34. See Kapur, “The Kashmir Dispute,” in Ganguly and Scobell, eds., Handbook of Asian
Security, p. 103.
35. Hari Singh also may have delayed his decision in the hope that he could somehow
strike a bargain with India and Pakistan by which Kashmir would retain its inde-
pendence. See Sumit Ganguly, The Crisis in Kashmir: Portents of War, Hopes of Peace
(Cambridge: Cambridge University Press, 1999), p. 9.
36. See Robert G. Wirsing, India, Pakistan, and the Kashmir Dispute (New York: St. Martin’s
Press, 1998), pp. 46–53; Alex Van Tunzelmann, Indian Summer: The Secret History of
the End of an Empire (London: Simon and Schuster, 2007), p. 285; Judith M. Brown,
Nehru: A Political Life (New Haven, CT: Yale University Press, 2003), pp. 177–79; and
Sharif al Mujahid, ed., In Quest of Jinnah: Diary, Notes, and Correspondence of Hector
Bolitho (Oxford: Oxford University Press, 2007), p. 84.
37. Akbar Khan, Raiders in Kashmir (Karachi: Pak Publishers Limited, 1970), pp. 10–13.
38. Ibid. See also Shuja Nawaz, Crossed Swords: Pakistan, Its Army, and the Wars Within
(Oxford: Oxford University Press, 2008), pp. 43–4 6; and Wirsing, India, Pakistan, and
the Kashmir Dispute, pp. 42–4 6.
39. Victoria Scholfield, Kashmir in Conflict: India, Pakistan, and the Unending War
(London: I. B. Taurus, 2000), pp. 41–42.
40. See Khan, Raiders in Kashmir, pp. 11–23; Nawaz, Crossed Swords, pp. 42–49; Jamal,
Shadow War, pp. 45–50, 56; and Andrew Whitehead, A Mission in Kashmir (New
Delhi: Viking Penguin, 2007). See also Shuja Nawaz, “The First Kashmir War
Revisited,” India Review, Vol. 7, No. 2 (April–June 2008), pp. 115–5 4.
156
( 156 ) Notes
41. Khan, Raiders, pp. 12–14, 16, 20; C. Dasgupta, War and Diplomacy in Kashmir, 1947-
1948 (New Delhi: Sage Publications, 2002), pp. 38–39; and Whitehead, A Mission in
Kashmir, pp. 136–37.
42. Nawaz, Crossed Swords, p. 48.
43. See Khan, Raiders, p. 22.
44. See Haqqani, Pakistan, p. 29; Margaret Bourke-W hite, Halfway to Freedom: A Report on
the New India in the Words and Photographs of Margaret Bourke-W hite (New York: Simon
and Schuster, 1949), pp. 207–8; and Whitehead, A Mission in Kashmir, chapters 3 and 7.
45. Author’s personal communication with Andrew Whitehead, January 2011.
46. See Kapur, “The Kashmir Dispute,” pp. 103–4.
47. Nawaz, Crossed Swords, pp. 48–68.
48. Sumantra Bose, Kashmir: Roots of Conflict, Paths to Peace (Cambridge, MA: Harvard
University Press, 2003), p. 41. See also Sumit Ganguly, Conflict Unending: India-
Pakistan Tensions Since 1947 (New Delhi: Oxford University Press, 2002), pp. 15–30.
49. See Jawaharlal Nehru, “Facts Relating to Kashmir,” Press Conference Statements in
New Delhi, January 2, 1948, in Jawaharlal Nehru, Independence and After: A Collection
of Speeches, 1946-1949 (New York: John Day Company, 1950), pp. 66, 69.
50. Von Tunzelmann, Indian Summer, pp. 306–7; Dasgupta, War and Diplomacy, p. 111.
51. Von Tunzelmann, Indian Summer, pp. 295, 302–3. See also Philip Ziegler, Mountbatten
(New York: Alfred A. Knopf, 1985), pp. 447–50, 504; Narendra Singh Sarila, Once a
Prince of Sarila: Of Palaces and Tiger Hunts, of Nehrus and Mountbattens (London: I. B.
Taurus, 2008), pp. 286–87. Note that Mountbatten later believed that Britain had gone
too far in support of the Pakistanis and urged his government to adopt an approach
more sensitive to Indian interests. See Ziegler, Mountbatten, p. 450.
52. The resolution required the Indians to withdraw from Kashmir as well, but only
after the Pakistanis had done so. See United Nations Security Council Resolution 47
(1948), https://w ww.mtholyoke.edu/acad/i ntrel/k ashun47.htm
53. Wirsing, India, Pakistan, and the Kashmir Dispute, p. 64.
54. Praveen Swami, India, Pakistan, and the Secret Jihad, p. 26; and Ziring, Pakistan, pp.
40–41.
CHAPTER 4
1. Khan, Raiders, pp. 125–2 6; and Jamal, Shadow War, pp. 71–73.
2. Jamal, Shadow War, pp. 72–73; and Swami, India, Pakistan and the Secret Jihad,
pp. 34–37.
3. Gul Hassan Khan, Memoirs of Lt. Gen. Gul Hassan Khan (Karachi, Pakistan: Oxford
University Press, 1993) pp. 115–17; and Nawaz, Crossed Swords, pp. 205–6 .
4. See Ganguly, The Crisis in Kashmir, pp. 49–53; Bose, Kashmir, pp. 78–83; and
M. Rafique Afzal, Pakistan: History and Politics, 1947-1971 (Karachi, Pakistan: Oxford
University Press, 2001), pp. 302–3.
5. Afzal, Pakistan: History and Politics, pp. 276– 78, 303– 5; Safdar Mahmood,
Pakistan: Political Roots & Development, 1947- 1999 (Karachi, Pakistan: Oxford
University Press, 2000), pp. 204–9; Ganguly, Conflict Unending, pp. 40–43; Brown,
Nehru; Ziring, Pakistan, pp. 103–4; Ganguly, The Crisis in Kashmir, pp. 53–55; and
Cohen, The Idea of Pakistan, p. 104. Ironically, the purpose of the Pakistanis’ study of
guerilla warfare at US military institutions was to learn how to fight against insurgents,
rather than to develop expertise in supporting them. See Cohen, The Idea of Pakistan,
p. 104. For a detailed overview of developments in Pakistani thinking during this
period regarding the use of militants see Fair, Fighting to the End, pp. 227–38.
6. Paul, Asymmetric Conflicts, pp. 120–22; Khan, Memoirs, pp. 116–17, 180–81; Afzal,
Pakistan: History and Politics, pp. 306–7; and Nawaz, Crossed Swords, pp. 206–7.
7. Khan, Memoirs, pp. 168–69.
157
Notes ( 157 )
8. Ministry of Defence, Government of India, S. N. Prasad and U. P. Thapliyal, eds., The
India-Pakistan War of 1965: A History (Dehra Dun: Natraj Publishers, 2011), pp. 51–52.
9. Afsir Karim, “The 1965 War: Lessons Yet to Be Learnt,” http://w ww.rediff.com/cms/
print.jsp?docpath=/news/2 005/sep/19war.htm; and Nawaz, Crossed Swords, p. 206.
11. See Sumit Ganguly, “Deterrence Failure Revisited: The Indo-Pakistani Conflict of
1965,” Journal of Strategic Studies, Vol. 13, No. 4 (December 1990), pp. 77–93; Russell
Brines, The Indo-Pakistani Conflict (New York: Pall Mall, 1968), pp. 301–2; and
Shaukat Qadir, “1965: Operation Grand Slam,” Daily Times, October 4, 2003. Note
that, although the Pakistani strategy consisted of two phases, Gibraltar and Grand
Slam, Gibraltar was the centerpiece of Pakistan’s effort. It was the Gibraltar force that
would actually infiltrate Kashmir, stoke an uprising, and capture the territory for the
Pakistanis. Operation Grand Slam was designed largely to prevent Indian interference
in this project, cutting Kashmir off from India and thereby ensuring that India could
not undo the gains that the Gibraltar militants had already made.
12. Prasad and Thapliyal, The India Pakistan War of 1965, p. 56.
13. See Sumit Ganguly, “Explaining the Kashmir Insurgency: Political Mobilization and
Institutional Decay,” International Security, Vol. 21, No. 2 (Fall 1996), pp. 76–107;
Bose, Kashmir, p. 84; Prasad and Thapliyal, The India-Pakistan War of 1965, p. 56;
Afzal, Pakistan: History and Politics, p. 307; and Nawaz, Crossed Swords, pp. 207–8.
14. Edgar O’Ballance, “The India- Pakistan Campaign, 1965,” Royal United Service
Institution Journal, Vol. 111, No. 644 (November 1966), pp. 330–35; Khan, Memoirs,
pp. 182–85; and Ganguly, Conflict Unending, pp. 44–4 6.
15. The Indians did not attempt to launch punishment operations against Pakistan dur-
ing the 1965 war. Since there is no evidence that they actively contemplated such
operations, I do not argue that a lack of punishment resulted from the advantages of
Pakistan’s proxy strategy. It seems, rather, that punishment was simply a nonissue dur-
ing this conflict.
16. Of course, the Indians did discover the Pakistani infiltration and move to seal the Line
of Control, thereby preventing the further entry of militants into Indian territory. By
this point, however, a sizeable militant force had already infiltrated Indian Kashmir. It
is highly unlikely that Pakistan would have been able to enter Kashmir with an equally
large conventional military force.
17. Prasad and Thapliyal, The India-Pakistan War of 1965, p. 60.
18. O’Ballance, “The India-Pakistan Campaign,” p. 332; Nawaz, Crossed Swords, p. 208;
and Prasad and Thapliyal, The India-Pakistan War of 1965, p. 81.
19. Paul, Asymmetric Conflicts, p. 112; Talbot, Pakistan, pp. 176–77; and Nawaz, Crossed
Swords, p. 207.
20. This supports my point in c hapter 2 regarding the differences between militants and
special operations forces. The officers in charge of the Gibraltar force were special-
operations personnel, drawn from the Pakistan Army’s Special Services Group. They
were impeded, however, by their lack of local skills and knowledge—skills and knowl-
edge that the militants possessed. As I argued, militants will often have an advantage
over special operations forces, and sponsor states may find them preferable for this
reason.
21. Nawaz, Crossed Swords, p. 208; and Afzal, Pakistan: History and Politics, p. 307.
23. Both India and Pakistan did make use of nonstate actors during the Bangladesh con-
flict. India supported Mukti Bahini insurgents fighting the East Pakistani government,
while Pakistan supported a number of opposing militant groups, such as the al-Shams
and al-Badr brigades. As I explain later, however, nonstate actors served primarily as
force multipliers, supporting conventional military operations. They were neither the
primary combatants nor the decisive factor in the conflict. See Richard Sisson and Leo
E. Rose, War and Secession: Pakistan, India, and the Creation of Bangladesh (Berkeley
and Los Angeles, CA: University of California Press, 1990), pp. 181–86, 210–13;
158
( 158 ) Notes
Haqqani, Pakistan, pp. 78–80; and Ishtiaq Hossain and Noore Siddiquee, “Islam in
Bangladesh Politics: The Role of Ghulam Azam of Jamaat-I-Islami,” Inter-A sia Cultural
Studies, Vol. 5, No. 3 (December 2004), pp. 384–9 9.
24. Talbot, Pakistan, pp. 24–25; and Sisson and Rose, War and Secession, pp. 8–16.
26. In East Pakistan, the league won 75 percent of the vote. Sisson and Rose, War and
Secession, p. 32.
27. Stanley Wolpert, Zulfi Bhutto of Pakistan: His Life and Times (New York: Oxford
University Press, 1993), pp. 145–49; and Ganguly, Conflict Unending, pp. 53–6 0.
28. The purpose of these groups was not to engage the Indian military, but rather to act as
counterinsurgency forces and target Pakistan’s domestic enemies within the country.
See Haqqani, Pakistan, pp. 79–80.
29. Archer Blood, Cable to Secretary of State William Rodgers, “Selective Genocide,”
March 28, 1971; and Kenneth Keating, Cable to Secretary of State William
Rodgers, “Selective Genocide,” March 29, 1971, http://nsarchive.gwu.edu/NSAEBB/
NSAEBB79/
30. See Gary J. Bass, The Blood Telegram: Nixon, Kissinger, and a Forgotten Genocide
(New York: Alfred A. Knopf, 2013), pp. 49–62, 119–33, 191; Dennis Kux, The United
States and Pakistan 1947- 2000: Disenchanted Allies (Baltimore: Johns Hopkins
University Press, 2001), pp. 186–88; Ziring, Pakistan, pp. 124–25; and Ganguly, Conflict
Unending, p. 61. Significantly, the Pakistan government’s Hamoodur Commission,
which subsequently investigated the army’s conduct during Operation Searchlight,
determined “it is clear that during and after the military action excesses were indeed
committed on the people of East Pakistan” by the Pakistan Army. It recommended
the trial and punishment of those personnel involved for “acts of wanton cruelty and
immorality against our own people.” See Hamoodur Commission Supplementary
Report (1974), as declassified by the Government of Pakistan, pp. 36–37, http://i mg.
dunyanews.tv/i mages/docss/hamoodur_rahman_commission_report.pdf
31. Ganguly, Conflict Unending, pp. 61–62; and Sisson and Rose, War and Secession,
pp. 207–8.
32. Sisson and Rose, War and Secession, pp. 181–86; Ganguly, Conflict Unending, p. 62;
Jones, Pakistan, p. 169; and Nawaz, Crossed Swords, pp. 283, 299–300.
33. Department of State Telegram, Sydney Sober to William P. Rodgers, “Conversation
with Pak Army Chief of Staff: Indo-Pak Military Confrontation,” August 11, 1971,
in Roedad Khan, ed., The American Papers: Secret and Confidential India, Pakistan,
Bangladesh Documents: 1965-1973 (Karachi, Pakistan: Oxford University Press,
1999), p. 641; Bass, The Blood Telegram, pp. 267–68; Nawaz, Crossed Swords, pp. 298–
99; and Sisson and Rose, War and Secession, p. 214.
34. See John J. Mearsheimer, Conventional Deterrence (Ithaca, NY: Cornell University
Press, 1983), pp. 35–43 and 205–6 .
35. Katherine Frank, Indira: The Life of Indira Nehru Gandhi (New York: Houghton
Mifflin, 2002), pp. 335–37; and Bass, The Blood Telegram, pp. 250–57.
36. Nawaz, Crossed Swords; and Ganguly, Conflict Unending, p. 68.
37. Nawaz, Crossed Swords, pp. 301–2 .
38. See Ganguly, Conflict Unending, pp. 67–69; and Sisson and Rose, War and Secession,
pp. 230–3 4.
39. Hasan-A skari Rizvi, “Pakistan’s Defense Policy,” in Mehrunnisa Ali, ed., Readings in
Pakistan Foreign Policy, 1971-1978 (Karachi, Pakistan: Oxford University Press, 2001),
pp. 210–11; and Sisson and Rose, War and Secession, p. 233.
40. Kapur, Dangerous Deterrent, pp. 67– 68; Shahid Amin, Pakistan’s Foreign
Policy: A Reappraisal (Karachi, Pakistan: Oxford University Press, 2002), p. 72; and
Rizvi, “Pakistan’s Defense Policy,” in Ali, ed., Readings in Pakistan Foreign Policy,
p. 211. See also Hamoodur Rehman Commission Supplementary Report.
41. Zulfikar Ali Bhutto, Speech at Lahore Airport, June 28, 1972.
42. See Text of Simla Agreement, http://w ww.mea.gov.in/i n-focus-a rticle.htm?19005/
Simla+Agreement+July+2+1972
159
Notes ( 159 )
43. Zulfikar Ali Bhutto, Speech at Pakistani Institute of International Affairs, Karachi,
July 31, 1972.
44. Quoted in Wolpert, Zulfi Bhutto, p. 195.
45. Ibid., pp. 191–92.
46. Shaikh, Making Sense of Pakistan, p. 54; Ganguly, Conflict Unending, pp. 71–72; and
S. M. Burke and Lawrence Ziring, Pakistan’s Foreign Policy: An Historical Analysis
(Karachi, Pakistan: Oxford University Press, 1990), pp. 420–21.
47. Ziring, Pakistan, pp. 164–65, 171, 183, 184–85; Talbot, Pakistan, pp. 270–83; Shahid
Javed Burki, Pakistan: Fifty Years of Nationhood, 3rd ed. (Boulder, CO: Westview, 1999),
pp. 51–53; Mir Zohair Hussain, “Islam in Pakistan Under Bhutto and Zia-u l-Haq,” in
Hussin Mutalib and Taj ul-Islam Hashmi, eds., Islam, Muslims, and the Modern State
(New York: St. Martin’s, 2004), pp. 60–68; Abbas, Pakistan’s Drift into Extremism,
pp. 103–7; Pervaiz Iqbal Cheema, The Armed Forces of Pakistan (New York: New York
University Press, 2001), p. 150; Hussain, Frontline Pakistan, p. 19; Craig Baxter,
“Restructuring the Pakistan Political System,” in Shahid Javed Burki and Baxter, eds.,
Pakistan Under the Military: Eleven Years of Zia ul-Haq (Boulder, CO: Westview, 1991),
pp. 29–30; and Haqqani, Pakistan, pp. 131–57. Note that, in addition to promoting
national unity, Zia hoped that Islamicization would increase his personal legitimacy
after his execution of Bhutto. See William L. Richter, “The Political Dynamics of
Islamic Resurgence in Pakistan,” Asian Survey, Vol. 19, No. 6 (June 1979), pp. 551–
52; Aqil Shah, “Pakistan’s ‘Armored’ Democracy,” Journal of Democracy, Vol. 14, No. 4
(October 2003), p. 33; and John L. Esposito and John Obert Voll, Islam and Democracy
(New York: Oxford University Press, 1996), p. 109.
48. Jahan Dad Khan, Pakistan Leadership Challenges (Karachi, Pakistan: Oxford University
Press, 1999), p. 167; and Cheema, The Armed Forces, p. 147.
49. Muhammed Zia ul-Haq, “Foreword,” in Malik, The Quranic Concept of War.
50. Ibid.
51. Rizvi, Military, State and Society, pp. 245– 2 6; and Hussain, Frontline Pakistan,
pp. 19–2 0.
52. Ziring, Pakistan, pp. 131–32. See also Vali Nasr, “National Identities and the India-
Pakistan Conflict,” in T. V. Paul, ed., The India-Pakistan Conflict: An Enduring Rivalry
(Cambridge: Cambridge University Press, 2005), pp. 178–2 01; and Shaikh, Making
Sense of Pakistan, p. 187.
53. Feroz Hasan Khan, Eating Grass: The Making of the Pakistani Bomb (Stanford: Stanford
University Press, 2012), p. 87.
54. See Kapur, Dangerous Deterrent, pp. 72–73.
55. Ibid., pp. 32–63.
56. Thomas Barfield, Afghanistan: A Cultural and Political History (Princeton, NJ: Princeton
University Press, 2010), pp. 225–38; and Larry P. Goodson, Afghanistan’s Endless
War: State Failure, Regional Politics, and the Rise of the Taliban (Seattle: University of
Washington Press, 2001), pp. 54–58.
57. The border between Pakistan and Afghanistan, known as the Durand Line, was drawn
in 1895 to separate Afghanistan from British India. Since 1947, Pakistan has con-
sidered the line to mark its border with Afghanistan. Afghan leaders, however, have
refused to formalize it. See Robert D. Crews and Amin Tarzi, “Introduction,” in Crews
and Tarzi, eds., The Taliban and the Crisis of Afghanistan (Cambridge, MA: Harvard
University Press, 2008), pp. 16–17; and Barfield, Afghanistan, pp. 153–55. Although
a direct Soviet attack on Pakistan was possible, Pakistani leaders thought that the
Soviets were far more likely to have designs on access to the Iranian port of Chah Bahar
on the Arabian Sea than on Pakistani territory. Thus, they believed that Iran was prob-
ably in more direct military danger as the result of the Soviet invasion of Afghanistan
than was Pakistan. Still, the Soviets’ presence on the Pakistani border was worrisome,
particularly in light of possible Indo-Soviet cooperation jointly to threaten Pakistan.
See Wirsing, Pakistan’s Security Under Zia, pp. 36–39; and Talbot, Pakistan, p. 268. See
also Deepa M. Ollapally, “The Evolution of India’s Relations with Russia: Tried, Tested,
and Searching for Balance,” in Sumit Ganguly, ed., India’s Foreign Policy: Retrospect
160
( 160 ) Notes
and Prospect (New Delhi: Oxford University Press, 2010), pp. 227–29; and Mahmood,
Pakistan: Political Roots and Development, pp. 210–22.
58. Kux, The United States and Pakistan, pp. 234–42; Talbot, Pakistan, p. 249; Nawaz,
Crossed Swords, p. 370; Robert G. Wirsing, Pakistan’s Security Under Zia, 1977–
1988: The Policy Imperatives of a Peripheral Asian State (New York: St. Martin’s Press,
1991), pp. 34–37; and Tanner, Afghanistan, p. 250.
59. Marvin G. Weinbaum, “The Politics of Afghan Resettlement and Rehabilitation,” Asian
Survey, Vol. 29, No. 3 (March 1989), pp. 287–307; Pervaiz Iqbal Cheema, “Impact of
the Afghan War on Pakistan,” in Ali, ed., Readings in Pakistan Foreign Policy, pp. 333–
39; and Wirsing, Pakistan’s Security Under Zia, pp. 43–4 4.
60. Ahmed Rashid, “The Afghan Conundrum,” in Maleeha Lodhi, ed., Pakistan: Beyond
the “Crisis State” (Karachi, Pakistan: Oxford University Press, 2011), pp. 312–14;
Ziring, “Weak State, Failed State, Garrison State,” p. 184; Goodson, Afghanistan’s
Endless War, p. 165; Shaikh, Making Sense of Pakistan, pp. 200–8; Cohen, The Idea of
Pakistan, pp. 190, 195; and Fair et al., Pakistan, pp. 110–12.
61. Rizwan Hussain, Pakistan and the Emergence of Islamic Militancy in Afghanistan
(Aldershot: Ashgate, 2005), pp. 61–75, 78–80; and Imtiaz Gul, The World’s Most
Dangerous Place: Pakistan’s Lawless Frontier (New York: Viking, 2009), pp. 2–3.
62. Wirsing, Pakistan’s Security Under Zia, pp. 27–2 8, 54.
63. Barnett R. Rubin, Afghanistan from the Cold War Through the War on Terror
(Oxford: Oxford University Press, 2013), pp. 81, 83; Max Boot, Invisible Armies: An
Epic History of Guerrilla Warfare from Ancient Times to the Present (New York: Liveright,
2013), pp. 495–98; Kux, The United States and Pakistan, p. 258; Talbot, Pakistan,
pp. 268–69; and Wirsing, Pakistan’s Security Under Zia, p. 54.
64. Kux, The United States and Pakistan, pp. 252–53, 262; Swami, India, Pakistan, and the
Secret Jihad, pp. 144–45; Wirsing, Pakistan’s Security Under Zia, p. 38; and Tanner,
Afghanistan, p. 250.
65. Tanner, Afghanistan, p. 247; Goodson, Afghanistan’s Endless War, pp. 50, 64; and
Barfield, Afghanistan, pp. 237–38.
66. Goodson, Afghanistan’s Endless War, pp. 64–67; Tanner, Afghanistan, p. 255; and
Lester W. Grau, The Bear Went Over the Mountain: Soviet Combat Tactics in Afghanistan
(New York: Routledge, 2005), pp. 77, 201.
67. Ali Ahmad Jalali and Lester W. Grau, The Other Side of the Mountain: Mujahideen
Tactics in the Soviet-Afghan War (Quantico, VA: United States Marine Corps Studies
and Analysis Division, 1995), pp. XX, 3, 37; Grau, The Bear Went Over the Mountain,
pp. 135, 197–98, 202; Goodson, Afghanistan’s Endless War, pp. 66–67; Barfield,
Afghanistan, p. 235; and Bruce Reidel, “Comparing the U.S. and Soviet Experiences in
Afghanistan,” CTC Sentinel, Vol. 2, No. 5 (May 2009), p. 2.
68. Tanner, Afghanistan, pp. 264–65; Kux, The United States and Pakistan, p. 281; Milt
Bearden and James Risen, The Main Enemy: The Inside Story of the CIA’s Final Showdown
with the KGB (New York: Random House, 2003), pp. 241–4 6; and Rafael Reuveny
and Aseem Prakash, “The Afghanistan War and the Breakdown of the Soviet Union,”
Review of International Studies, Vol. 25 (1999), p. 697.
69. Tanner, Afghanistan, pp. 266–67; Bearden and Risen, The Main Enemy, pp. 246–5 4;
and Kux, The United States and Pakistan, p. 281.
70. Reuveny and Prakash, “The Afghanistan War,” pp. 697–98, 705–6 .
71. Ibid, pp. 697–98; Artemy Kalinovsky, “Decision-Making and the Soviet War in
Afghanistan,” Journal of Cold War Studies, Vol. 11, No. 4 (Fall 2009), pp. 60–61; Tanner,
Afghanistan, p. 268; “Text of Gorbachev Statement Setting Forth Soviet Position on
Afghan War,” New York Times, February 9, 1988.
72. See Bill Keller, “Last Soviet Soldiers Leave Afghanistan,” New York Times, February
16, 1989.
73. Ziring, The War in Afghanistan, pp. 64, 54.
74. This account draws on Arif Jamal’s interviews of senior Pakistani military officers and
Kashmiri leaders close to Zia, including Maulana Abdul Bari, founding amir of Jamat-
i-Islami of Azad Jammu and Kashmir. They are discussed in Jamal, Shadow War, and in
161
Notes ( 161 )
S. Paul Kapur’s conversations with Jamal, New Brunswick, New Jersey, February 2011.
See also Shaun Gregory, “The ISI and the War on Terrorism,” Brief No. 28 (Pakistan
Security Research Unit, January 24, 2008), pp. 4–7; Christophe Jaffrelot, “A Fruitless
Search for Democracy,” in Jaffrelot, ed., A History of Pakistan and Its Origins, pp. 79–
80; and Bruce Reidel, Deadly Embrace: Pakistan, America and the Future of the Global
Jihad (Washington, DC: Brookings Institution, 2011), p. 26.
75. Wirsing, Pakistan’s Security Under Zia, p. 56.
76. Steve Coll, Ghost Wars: The Secret History of the CIA, Afghanistan, and Bin Laden,
from the Soviet Invasion to September 10, 2001 (New York: Penguin, 2004), pp. 60–
74; Mohammad Yousaf and Mark Adkin, Afghanistan–The Bear Trap (Havertown,
PA: Casemate, 1991), pp. 81, 96, 98, 117; Dennis Kux, The United States and Pakistan,
pp. 252–53, 274; Swami, India, Pakistan, and the Secret Jihad, pp. 144–45; and Abbas,
Pakistan’s Drift into Extremism, p. 110.
77. Jamal, Shadow War, p. 112.
78. Ibid., pp. 109–12, 126–2 8; Kapur’s discussion with Jamal; and S. Paul Kapur’s dis-
cussion with a senior Indian counterinsurgency force field commander and a retired
senior Indian intelligence officer, New Delhi, September 2010.
79. Swami, India, Pakistan, and the Secret Jihad, p. 142; and Ahmed Rashid, Taliban: Militant
Islam, Oil and Fundamentalism in Central Asia (New Haven, CT: Yale University
Press, 2000).
80. Goodson, Afghanistan’s Endless War, p. 68; and Nawaz, Crossed Swords, pp. 377–79.
CHAPTER 5
1. See Ganguly, “Explaining the Kashmir Insurgency”; Ashutosh Varshney, “India,
Pakistan, and Kashmir: Antinomies of Nationalism,” Asian Survey, Vol. 31, No. 11
(1991), pp. 1014–17; Anand Mohan, “The Historical Roots of the Kashmir Conflict,”
Studies in Conflict and Terrorism, Vol. 15, No. 4 (1992), pp. 302–3; and Simon Jones,
“India, Pakistan, and Counterinsurgency Operations in Jammu and Kashmir,” Small
Wars & Insurgencies, Vol. 19, No. 1 (2008), pp. 5–6 .
2. Ganguly, The Crisis in Kashmir, pp. 96–9 7; and Sten Widmalm, “The Rise and Fall of
Democracy in Jammu and Kashmir,” Asian Survey, Vol. 37, No. 11 (1997), pp. 1009–23.
3. Ganguly, The Crisis in Kashmir, pp. 98–9 9.
4. Swami, India, Pakistan and the Secret Jihad, pp. 160–63.
5. Ibid., pp. 162–63.
6. Jones, “Counterinsurgency Operations in Jammu and Kashmir,” pp. 8–9.
7. Jamal, Shadow War, pp. 128–30; Kapur’s discussion with Jamal; and Kapur’s dis-
cussion with a senior counterinsurgency force field commander and a retired senior
Indian intelligence officer.
8. See Ganguly, The Crisis in Kashmir, pp. 65– 73; Wirsing, India, Pakistan, and
the Kashmir Dispute, pp. 113–18; Bose, Kashmir, pp. 107–35; and Iffatt Malik,
Kashmir: Ethnic Conflict International Dispute (Oxford: Oxford University Press,
2002), pp. 158–6 0, 283.
9. Jagmohan, My Frozen Turbulence in Kashmir (New Delhi: Allied Publishers, 2002),
pp. 518–19.
10. Figures from South Asian Terrorism Portal, http://w ww.satp.org/satporgtp/coun-
tries/i ndia/states/jandk/data_sheets/casualtiesmilitency.htm
11. See text of Terrorism and Disruptive Activities (Prevention) Act (1987), http://w ww.
satp.org/satporgtp/countries/i ndia/document/actandordinances/TADA.HTM
12. Amnesty International, “India: The Terrorist and Disruptive Activities (Prevention)
Act: The Lack of Scrupulous Care” (November 1984), pp. 1–4.
13. Victoria Schofield, Kashmir in Conflict: India, Pakistan and the Unending War
(London: I. B. Taurus, 2003), pp. 156–57, 168–70; and Asia Watch and Physicians
162
( 162 ) Notes
for Human Rights, The Human Rights Crisis in Kashmir: A Pattern of Impunity
(New York: Human Rights Watch, 1993).
14. Schofield, Kashmir, p. 169.
15. Swami, India, Pakistan and the Secret Jihad, pp. 166–67.
16. Jammu and Kashmir Liberation Front, “Our Ideology” and “Our Aims and Objectives,”
http://jklfajkgbzone.org/i ndex.php/2 012-11-2 3-13-2 4-14/2 012-12-0 2-19-0 8-3 4/
our-ideology and http://jklfajkgbzone.org/i ndex.php/2 012-11-23-13-2 4-14/2 012-12-
02-19-08-3 4/our-a ims-a nd-objectives
17. Bose, “The JKLF and the JKHM,” in Heiberg et al., eds., Terror, Insurgency, and the
State, pp. 232–3 4; and Kapur’s discussion with a senior Indian counterinsurgency
force field commander and retired senior Indian intelligence officer.
18. Jamal, Shadow War, p. 146.
19. Ibid., p. 143. See also Swami, India, Pakistan and the Secret Jihad, pp. 178–80; and
Navnita Chadha Behera, Demystifying Kashmir (Washington, DC: Brookings
Institution Press, 2006), pp. 151–67.
20. See Bose, “The JKLF and JKHM,” pp. 237–38; John R. Schmidt, The Unraveling: Pakistan
in the Age of Jihad (New York: Farrar, Straus and Giroux, 2011), pp. 81–82; Schofield,
Kashmir in Conflict, pp. 157, 174–75; and Hussain, Frontline Pakistan, p. 25.
21. Data from South Asia Terrorism Portal, http://w ww.satp.org/satporgtp/countries/
india/states/jandk/data_sheets/casualtiesmilitency.htm
22. Bose, Kashmir, p. 133–35; Schofield, Kashmir in Conflict, pp. 198–9 9; Shale Horowitz
and Deepti Sharma, “Democracies Fighting Ethnic Insurgencies: Evidence from
India,” Studies in Conflict & Terrorism, Vol. 31, No. 8 (2008), pp. 767–68; and Jones,
“Counterinsurgency Operations in Jammu and Kashmir,” pp. 14–15.
23. S. Paul Kapur’s discussion with Indian counterinsurgency force field commander, New
Delhi, September 2010.
24. Security force deaths, however, remained high, at nearly 260. Figures from South
Asia Terrorism Portal, http://w ww.satp.org/satporgtp/countries/i ndia/states/jandk/
data_sheets/casualtiesmilitency.htm
25. Schmidt, The Unraveling, pp. 84–85; Swami, India, Pakistan and the Secret Jihad,
pp. 180, 189–9 0; Hussain, Frontline Pakistan, pp. 56–57; and Kapur’s discussion with a
senior Indian counterinsurgency force field commander.
26. S. Paul Kapur’s discussion with Indian counterinsurgency force field commander
and senior retired intelligence official, New Delhi, September 2010; Tankel, Storming
the World Stage, pp. 51–5 4; and Haqqani, Pakistan: Between Mosque and Military,
pp. 287–89.
27. For profiles of these groups, see Muhammad Amir Rana, A to Z of Jehadi Organizations
in Pakistan, trans. Saba Ansari (Lahore: Marshall Books, 2004); and Santhanam,
Sreedhar, Saxena, and Manish, Jihadis in Kashmir.
28. Tankel, Storming the World’s Stage, pp. 40–4 4.
29. Ibid., pp. 60–61.
30. Lashkar- e -
Taiba was not the only militant organization to employ fedayeen
attacks. Other groups, such as Jaish-e -Mohammed, used these tactics as well.
LeT, however, is recognized as having pioneered fedayeen operations. See Bose,
Kashmir, pp. 140–47; Tankel, Storming the World’s Stage, p. 63; Praveen Swami,
“Lashkar Honed Fidayeen Skills in Srinagar Attacks,” Hindu, December 1, 2008;
C. Christine Fair, “Antecedents and Implications of the November 2008 Lashkar-
e-Taiba (LeT) Attack upon Several Targets in the Indian Mega-City of Mumbai,”
RAND, Testimony before the House Homeland Security Committee, March 11,
2009, http://w ww.rand.org/c ontent/d am/r and/pubs/testimonies/2 009/RA ND_
CT320.pdf, p. 8.
31. Tankel, Storming the World’s Stage, pp. 63–6 4.
32. Figures from South Asian Terrorism Portal, http://w ww.satp.org/satporgtp/coun-
tries/i ndia/states/jandk/data_sheets/casualtiesmilitency.htm
33. See Arif Ashfaq v. State of NCT of Delhi, Supreme Court of India, August 10, 2011,
pp. 1–4.
163
Notes ( 163 )
34. Although the Indian government accused both Lashkar- e-

Taiba and Jaish- e-
Mohammed of launching the attack, only members of Jaish-e-Mohammed were later
convicted for participating in the operation. See Vanessa Gezari, “Indian Suspect
Claims Pakistan Aided Parliament Attack,” Chicago Tribune, December 21, 2001;
Arundhati Roy, “ ‘And His Life Should Become Extinct’: The Very Strange Story of the
Attack on the Indian Parliament,” Outlook, October 30, 2006; and “Timeline: From
the Attack on Parliament to Afzal Guru’s Hanging,” CNN IBN, February 3, 2013,
http://ibnlive.in.com/news/t imeline-f rom-t he-attack-on-parliament-to-a fzal-g urus-
hanging/371739-3.html
35. See C. Christine Fair, “Lashkar-e-Tayiba and the Pakistani State,” Survival, Vol.
53, No. 4 (2011), pp. 14–15; and Ryan Clarke, “Lashkar-I-Taiba: The Fallacy of
Subservient Proxies and the Future of Islamist Terrorism in India,” U.S. Army War
College, Strategic Studies Institute, The L’Etort Papers, March 2010, http://w ww.stra-
tegicstudiesinstitute.army.mil/pubs/d isplay.cfm?pubID=973, pp. 44–45.
36. See Clarke, “Lashkar-I-Taiba,” pp. 2, 17–19; and Neil Padukone, “The Next al-
Qaeda? Lashkar-e-taiba and the Future of Terrorism in South Asia,” World Affairs
Journal, November/ December 2011, http://w ww.worldaffairsjournal.org/a rticle/
next-a l-qaeda-lashkar-e-taiba-a nd-f uture-terrorism-south-asia
37. Tankel, Storming the World’s Stage, pp. 60, 128–29; Swami, India, Pakistan and the
Secret Jihad, pp. 193–95; and Kapur’s discussion with a senior Indian counterinsur-
gency force field commander and a retired senior Indian intelligence official.
38. Tankel, Storming the World’s Stage, pp. 207–10.
39. Scott-Clark and Levy, The Siege.
40. For detailed discussions of the 2001–02 crisis, see V. K. Sood and Pravin Sawhney,
Operation Parakram: The War Unfinished (New Delhi: Sage, 2003); Polly Nayak and
Michael Krepon, “U.S. Crisis Management in South Asia’s Twin Peaks Crisis,” Report
No. 57 (Washington, DC: Henry L. Stimson Center, September 2006); and Sumit
Ganguly and Michael R. Kraig, “The 2001-2 002 Indo-Pakistani Crisis: Exposing
the Limits of Coercive Diplomacy,” Security Studies, Vol. 14, No. 2 (Summer 2005),
pp. 290–324.
41. See Sood and Sawhney, Operation Parakram, pp. 59–144; Alex Stolar, “To the
Brink: Indian Decision-Making and the 2001-2 002 Standoff,” The Henry L. Stimson
Center, Report No. 68, February 2008, http://w ww.stimson.org/i mages/uploads/
research-pdfs/To_t he_ Brink.pdf
42. See President Pervez Musharraf ’s Address to the Nation, Islamabad, January 12,
2002, http://w ww.millat.com/president/1020200475758AMword%20file.pdf; Alan
Sipress and Rajiv Chandrasekaran, “Powell ‘Encouraged’ by India Visit,” Washington
Post, January 19, 2002; Robert Marquand, “Powell Tiptoes Indo- Pak Divide,”
Christian Science Monitor, January 18, 2002; and Susan Milligan, “India-Pakistan
Standoff Easing, Powell Says,” Boston Globe, January 18, 2002.
43. Peter Chalk, “Lashkar-e-Taiba’s International Focus and Its Growing Links with al-
Qaeda,” Terrorism Monitor, Jamestown Foundation, July 29, 2010, http://w ww.james-
town.org/programs/g ta/single/?tx_tt news[tt_ news]=36683&tx_tt news[backPid]=
26&cHash=fc945260f6; and Tankel, Storming the World’s Stage, pp. 125–27.
44. Kapur’s discussion with former Indian National Security Advisor Brajesh Mishra,
May 2005.
45. Government of India, Ministry of Home Affairs, Annual Report 2006-2 007, p. 143.
46. Shaun Waterman, “Pakistani General Denies Terror Links,” Washington Times, July
28, 2010.
47. Tankel, Storming the World’s Stage, pp. 69–72.
48. Alex Rodriguez, “Pakistani Militant Groups Out in the Open,” Los Angeles Times,
May 8, 2010. See also “JuD Starts ‘Hate America’ Campaign on Davis Issue,” Deccan
Herald, February 22, 2011; Tankel, Storming the World’s Stage, p. 128; Saeed Shah,
“Pakistan Floods: Jamaat-ud-Dawa, Islamists Linked to India’s Mumbai Attack, Offer
Aid,” Christian Science Monitor, August 4, 2010; National Counterterrorism Center,
“Lashkar-e-Tayyiba,” http://w ww.nctc.gov/site/g roups/let.html; and United Nations
164
( 164 ) Notes
Department of Public Information, “Security Council Al-Qaida and Taliban Sanctions

Committee Adds Names of Four Individuals to Consolidated List, Amends Entries of
Three Entities,” December 10, 2008.
49. Tankel, Storming the World’s Stage, p. 70.
50. Angel Rabasa, Robert D. Blackwill, Peter Chalk, Kim Cragin, C. Christine Fair,
Seth Jones, Nathaniel Shestak, Brian A. Jackson, Brian Michael Jenkins, and Ashley
J. Tellis, “The Lessons of Mumbai,” RAND Occasional Paper (2009), pp. 1–2 , 8, 13,
16, 19. Sebastian Rotella, “David Headley, Witness in Terror Trial, Ties Pakistani
Spy Agency to Militant Group,” Washington Post, May 23, 2011; Sebastian Rotella,
“Mumbai Case Offers Rare Picture of Ties Between Pakistan’s Intelligence Service,
Militants,” Pro Publica, May 2, 2011; and Jason Burke, “Pakistani Intelligence Services
‘Aided Mumbai Terror Attacks,’ ” Guardian, October 18, 2010.
51. See Salman Masood, “Terror Suspect Cleared Again in Pakistan,” New York Times,
October 12, 2009.
52. Declan Walsh, “Pakistani Militant, Price on Head, Lives in Open,” New York Times,
February 6, 2013.
53. Anatol Lieven, Pakistan: A Hard Country (New York: Public Affairs, 2011), p. 171; and
“Pakistan’s Jamaat ‘Ban’ Lie Nailed,” Times of India, January 12, 2009.
54. At present, Lashkar-e-Taiba probably ranks as the most important South Asian ter-
rorist group. See Indrani Bagchi, “Is Lashkar New al-Qaida?” Times of India, July 4,
2009; and Husain Haqqani, “The Ideologies of South Asian Jihadi Groups,” in Hillel
Fradkin, Husain Haqqani, and Eric Brown, eds., Current Trends in Islamist Ideology,
Vol. 1 (Washington, DC: Hudson Institute, 2005), pp. 24–25; and Tankel, Storming
the World’s Stage, p. 1.
55. Riaz Mohammad Khan, Afghanistan and Pakistan: Conflict, Extremism, and Resistance
to Modernity (Washington, DC, and Baltimore: Woodrow Wilson Center Press and
Johns Hopkins University Press, 2011), pp. 39–52.
56. Khan, Afghanistan and Pakistan, pp. 52–53; Griffin, Reaping the Whirlwind, pp. 21–22;
and Coll, Ghost Wars, pp. 117–19.
57. James Ferguson, Taliban: The Unknown Enemy (Cambridge, MA: Da Capo Press,
2010), pp. 52–5 4; Khan, Afghanistan and Pakistan, pp. 51–52.
58. Rashid, Taliban, p. 26; Ferguson, Taliban, p. 51; Michael Griffin, Reaping the
Whirlwind: The Taliban Movement in Afghanistan (London: Pluto, 2001), pp. 21, 70;
Larry P. Goodson, Afghanistan’s Endless War: State Failure, Regional Politics, and the
Rise of the Taliban (Seattle: University of Washington Press, 2001), pp. 110–11; and
Coll, Ghost Wars, pp. 119–2 0.
59. See Rashid, Taliban, pp. 26, 186–9 7.
60. Ibid., pp. 89–9 0.
61. Ibid., pp. 28–29, 34; Goodson, Afghanistan’s Endless War, pp. 107–8, 111; C. Christine
Fair, The Madrassah Challenge: Militancy and Religious Education in Pakistan
(Washington, DC: US Institute of Peace, 2008), p. 57; Griffin, Reaping the Whirlwind,
pp. 72–73; Coll, Ghost Wars, pp. 289–9 7; and Hussain, Frontline Pakistan, pp. 28–29.
62. Goodson, Afghanistan’s Endless War, pp. 111–13; Rashid, Taliban, pp. 40, 184–85;
Griffin, Reaping the Whirlwind, p. 94; Jeffrey Dressler, “The Haqqani Network: From
Pakistan to Afghanistan,” Institute for the Study of War (October 2010), pp. 7–8; Gul,
The World’s Most Dangerous Place, pp. 148–56; and John Burns, “In Newly Won Afghan
Region, Taliban Consolidate Their Hold,” New York Times, May 26, 1997.
63. Rashid, Taliban, p. 186.
64. Goodson, Afghanistan’s Endless War, pp. 110, 114; Hussain, Frontline Pakistan, p. 30;
and Rashid, Taliban, p. 29.
65. S. Paul Kapur’s discussion with senior Pakistani diplomat, Islamabad, Pakistan,
December 2011.
66. Musharraf, In the Line of Fire, p. 202.
67. Estimates of the number of evacuees range from the hundreds to approximately five
thousand. Although the bulk of the evacuations occurred in November 2001, they
may have continued for some time, into January 2002. See Rashid, Descent into Chaos,
165
Notes ( 165 )
pp. 90–93; Seymour M. Hersh, “The Getaway,” New Yorker, January 28, 2002; and
Press Trust of India, “India Protests Airlift of Pak Fighters from Kunduz, Fears They
Will Enter Kashmir,” Indian Express, January 2002.
68. See Jayshree Bajoria and Eben Kaplan, “The ISI and Terrorism: Behind the
Accusations,” Council on Foreign Relations, Backgrounder, May 4, 2011, http://
www.cfr.org/pakistan/i si-terrorism-behind-accusations/p11644; Matt Waldman,
“The Sun in the Sky: The Relationship Between Pakistan’s ISI and Afghan Insurgents,”
London School of Economics, Crisis States Research Centre, Discussion Paper 18,
June 2010, pp. 12–17; Bruce O. Riedel, “A New Pakistan Policy: Containment,”
New York Times, October 14, 2011; Peter Tomsen, “Pakistan: With Friends Like
These …,” World Policy Journal, Vol. 28, No. 82 (2011): pp. 86–88; Thomas F. Lynch
III, “The 80 Percent Solution: The Strategic Defeat of bin Laden’s al-Qaeda and
Implications for South Asian Security,” in Peter Bergen and Katherine Tiedemann,
eds., Talibanistan (New York: Oxford University Press, 2013), pp. 406–8; and Adrian
Hänni and Lukas Hegi, “The Pakistani Godfather: The Inter-Services Intelligence
and the Afghan Taliban 1994-2 010,” Small Wars Journal, Vol. 9, No. 4 (2013),
pp. 6–11.
69. Vahid Brown and Don Rassler, Fountainhead of Jihad: The Haqqani Nexus, 1973-
2012 (New York: Oxford University Press, 2013), pp. 152–55; Anand Gopal, Mansur
Khan Mashud, and Brian Fishman, “The Taliban in North Waziristan,” in Bergen
and Tiedemann, eds., Talibanistan, pp. 132–38; Jeffrey A. Dressler, “The Haqqani
Network: From Pakistan to Afghanistan,” Institute for the Study of War, Afghanistan
Report, October 6, 2010, http://w ww.understandingwar.org/sites/default/fi les/
Haqqani_ Network_Compressed_0.pdf, pp. 9, 11– 12; Jeffrey A. Dressler, “The
Haqqani Network: A Strategic Threat,” Institute for the Study of War, Afghanistan
Report 9, March 2012, http://w ww.understandingwar.org/report/haqqani-network-
strategic-t hreat, pp. 11–15; and Eric Schmitt, “White House Backs Blacklisting
Militant Organization,” New York Times, September 6, 2012.
70. Brown and Rassler, Fountainhead of Jihad, pp. 162–81; and Dressler, “The Haqqani
Network: A Strategic Threat,” p. 8.
71. Pervez Musharraf, In the Line of Fire: A Memoir (New York: Free Press, 2006), pp. 201–
4; Waldman, “The Sun in the Sky,” pp. 1–2 6; Seth G. Jones, “Pakistan’s Dangerous
Game,” Survival, Vol. 49, No. 1 (Spring 2007), pp. 17–19; and Dressler, “The Haqqani
Network,” p. 37.
72. Senate Armed Services Committee, “Statement of Admiral Michael Mullen,”
September 22, 2011, p. 3.
73. Department of Defense Press Briefing by General Joseph F. Dunford on Operations in
Afghanistan in the Pentagon Briefing Room on Afghanistan, March 13, 2014, http://
www.defense.gov/Transcripts/Transcript.aspx?TranscriptID=5392
74. Government of India, Ministry of Home Affairs, Annual Report 2006-2 007, p. 143.
75. Ibid.
76. This is not to argue that India never succeeded in preventing militant infiltration, cap-
turing militants in its territory, or thwarting ongoing militant operations. It did so on
many occasions. My point is simply that, despite some successes, India failed at these
tasks much of the time. This resulted in thousands of militant infiltrations and attacks,
as well as well over one thousand civilian and security-force deaths per year.
77. Department of Defense, “Report on Progress Toward Security and Stability in
Afghanistan,” December 2012, pp. 3, 18. See also p. 21.
78. Ibid., p. 152.
79. See Seth G. Jones, “Counterinsurgency in Afghanistan,” RAND, Counterinsurgency
Study, Vol. 4, 2008, pp. 54–61; Seth G. Jones, “What Went Wrong in Afghanistan?
Allowing a Sanctuary in Pakistan,” Foreign Policy, March 5, 2013; Thomas H. Johnson
and M. Chris Mason, “No Sign Until the Burst of Fire: Understanding the Pakistan-
Afghanistan Frontier,” International Security, Vol. 32, No. 4 (2008), pp. 43–58;
Steven Metz, “Destroy the Taliban’s Sanctuary,” Joint Forces Quarterly, Vol. 54, No. 3
(2009), p. 62; Gilles Dorronsoro, “Afghanistan: The Impossible Transition,” Carnegie
166
( 166 ) Notes
Endowment for International Peace, Carnegie Papers, June 2011, pp. 15–17; Antonio
Giustozzi, Koran, Kalashnikov and Laptop: The Neo-Taliban Insurgency in Afghanistan
(New York: Columbia University Press, 2008), pp. 190, 202; Jason Motlagh, “The
Taliban’s Changing, and Deadly, Tactics,” Pulitzer Center on Crisis Reporting, July
1, 2010, http://pulitzercenter.org/a rticles/talibans-changing-a nd-deadly-tactics-
afghanistan; Thomas H. Johnson, “Taliban Adaptations and Innovations,” Small Wars
& Insurgencies, Vol. 24, No. 1 (2013), p. 6; Kenneth Katzman, “Afghanistan: Post-
Taliban Governance, Security, and U.S. Policy,” Congressional Research Service,
RL30588, March 4, 2014, p. 17; Tom V. Brook, “Afghan Insurgent Tactics Shift to
Dodge Airstrikes,” USA Today, January 20, 2009; Fred Kaplan, “The Insurgent’s
Playbook,” Slate, February 27, 2012, http://w ww.slate.com/a rticles/news_ a nd_
politics/ w ar_ s tories/ 2 012/ 0 2/ t he_ k illing _ o f_ t wo_ a merican_ o fficers_ i n_
afghanistan_ i s_destroying_t he_t rust_ needed_to_ rebuild_t he_c ountry _. html;
Dressler, “The Haqqani Network: A Strategic Threat,” pp. 31–35; Johnson, “Taliban
Adaptations,” pp. 7–8, 10–2 0; Department of Defense, “Report on Progress Toward
Security and Stability in Afghanistan,” p. 152; Defense Casualty Analysis System,
“U.S. Military Casualties—Operation Enduring Freedom (OEF) Casualty Summary
by Casualty Category,” https://w ww.dmdc.osd.mil/dcas/pages/report_oef_t ype.
xhtml; and Susan Chesser, “Afghanistan Casualties: Military Forces and Civilians,”
Congressional Research Service, December 6, 2010, pp. 2–4.
80. In Musharraf ’s Words: “A Day of Reckoning,” New York Times, January 12, 2002.
CHAPTER 6
1. Olivier Roy, “Islam and Foreign Policy: Central Asia and the Arab-Persian World,” in
2. See Mumtaz Ahmad, “The Crescent and the Sword: Islam, the Military, and Political
Legitimacy in Pakistan, 1977-1985,” Middle East Journal, Vol. 50, No. 3 (Summer
1996), pp. 381–86.
3. Jean-Luc Racine, “Living with India: Relations Between Pakistan and India,” in
4. Tariq Ali, Can Pakistan Survive? The Death of a State (New York: Random House,
1984); Rashid, Descent into Chaos, pp. 33–37; National Intelligence Council, Global
Trends 2025: A Transformed World (Washington, DC: US Government Printing
Office, 2008), pp. 45, 72; and Ahmad Faruqui, “Can Pakistan Survive?” Outlook India,
January 23, 2009.
6. Cohen, The Idea of Pakistan, p. 270; and Christophe Jaffrelot, “Nationalism Without
a Nation: Pakistan Searching for Its Identity,” in Jaffrelot, ed., Pakistan: Nationalism
Without a Nation (New York: Palgrave, 2002), p. 7.
7. S. Paul Kapur’s discussion with a senior serving Pakistani military officer (2013).
8. See Shah, The Army and Democracy, pp. 14–15; Haroon K. Ullah, Vying for Allah’s
Vote: Understanding Islamic Parties, Political Violence, and Extremism in Pakistan
(Washington, DC: Georgetown University Press, 2014), p. 25; Niaz, The Culture of
Power and Governance in Pakistan, p. 190; Ziad Haider, “Ideologically Adrift,” in Lodhi,
ed., Pakistan: Beyond the “Crisis State,” p. 123; and Aminah Mohammad-A rif, “The
Diversity of Islam,” in Jaffrelot, ed., A History of Pakistan and Its Origins, pp. 233–3 4.
9. See South Asia Terrorism Portal, “Jammu and Kashmir Data Sheets: Fatalities in
Terrorist Violence Since 1988,” http://w ww.satp.org/satporgtp/countries/i ndia/
states/jandk/data_sheets/a nnual_casualties.htm; and Correlates of War Project, Inter-
State War Data, ver. #4.0, http://w ww.correlatesofwar.org/
10. South Asia Terrorism Portal, “Jammu and Kashmir Data Sheets.” Indian civilian losses
in its other conflicts with Pakistan were negligible. As Ganguly points out, these wars
167
Notes ( 167 )
were fought by “gentlemen’s rules,” which eschewed the use of violence against non-
combatants. See Sumit Ganguly, “Wars Without End: The Indo-Pakistani Conflict,”
Annals of the American Academy of Political and Social Science, Vol. 541 (1995),
pp. 167–78.
11. Although India would have to maintain some troop presence in Jammu and Kashmir
even under normal conditions, its current commitment is far larger than would be nec-
essary in the absence of the rebellion.
12. Human Rights Watch, India’s Secret Army in Kashmir: New Patterns of Abuse Emerge
in the Conflict, Vol. 8, No. 4 (May 1996). See also Wirsing, India, Pakistan, and the
Kashmir Dispute, pp. 154–62; and Schofield, Kashmir in Conflict, pp. 168–72.
13. For example, the United Nations has, since 1947, maintained a military observer group
in Kashmir, and the United States has recently broached the idea of helping the two
sides to resolve the Kashmir dispute. See United Nations Military Observer Group in
India and Pakistan, http://w ww.un.org/en/peacekeeping/m issions/u nmogip/; and
Chidanand Rajghatta, “Obama Mulls Clinton as Special Envoy on Kashmir,” Times of
India, November 7, 2008.
14. See, for example, Ken Dilanian and David S. Cloud, “U.S. Intelligence Reports Cast
Doubt on War Progress in Afghanistan,” Los Angeles Times, December 15, 2010.
General David Petraeus characterized coalition progress against the Taliban as “frag-
ile and reversible.” Statement of Gen. David H. Petraeus before the House Armed
Services Committee on Afghanistan, March 16, 2011.
15. Susan B. Epstein and K. Alan Kronstadt, “Pakistan: U.S. Foreign Assistance,”
Congressional Research Service Report (July 2013), pp. 10–11.
16. See T. V. Paul, South Asia’s Weak States: Understanding the Regional Security Predicament
(Stanford: Stanford University Press, 2010), p. 5.
17. The TTP’s three main goals are to enforce Sharia, battle coalition forces in Afghanistan,
and engage in defensive jihad against the Pakistan Army. See Qandeel Siddique,
“Tehrik-e-Taliban Pakistan: An Attempt to Deconstruct the Umbrella Organization
and the Reasons for Its Growth in Pakistan’s North-West,” Danish Institute for
International Studies Report, Volume 12 (2010), pp. 4–8; and Fair and Jones,
“Pakistan’s War Within,” pp. 162–69, 173–78, 182.
18. Rob Crilly, “Peace Talks Collapse as Taliban Execute 23 Pakistani Soldiers,”
Telegraph, February 17, 2014; Karen DeYoung, “Pakistan Plans Military Operation
in North Waziristan, Targeting Extremist Groups,” Washington Post, February 25,
2014; Fair et al., Pakistan, pp. 94–95; and Siddique, “Tehrik-e -Taliban Pakistan,”
pp. 71–7 2.
19. Although it was carried out by an American drone strike, the militants viewed
Meshud’s killing as part of the Pakistani government’s broader campaign against the
Taliban. See Jon Boone, “Pakistani Taliban Claim Karachi Attack and Leave Peace
Talks in Crisis,” Guardian, June 9, 2014.
20. Declan Walsh, “In Drive Against Militants, Pakistani Airstrikes Hit Strongholds,”
New York Times, June 17, 2014. Note that a willingness on the part of some TTP ele-
ments to negotiate with the Sharif government may have led the rest of the group to
demonstrate its opposition to such compromise in dramatic fashion. See Jason Burke,
“Tehrik-e-Taliban Pakistan Asserts Itself After Split by Attacking Karachi Airport,”
Guardian, June 9, 2014.
21. Walsh, “In Drive Against Militants.”
22. Particularly provocative was the Pakistan Army’s 2007 takeover of the Lal Masjid in
Islamabad. Since approximately 70 percent of the Masjid’s students were from the
Federally Administered Tribal Areas and Khyber Pakhtunkhwa, the militants viewed
the army’s action as a direct affront. This appears to have been a catalytic event in the
founding of the TTP. See Akbar Ahmed, Thistle and Drone: How America’s War on
Terror Became a War on Tribal Islam (Washington, DC: Brookings Institution Press,
2013); and Saiddique, “Tehrik-e-Taliban Pakistan,” pp. 9–10.
23. Nazar ul Islam, “The Caliphate Cometh: The Pakistani Taliban Open Up,” Newsweek
Pakistan, February 11, 2014.
168
( 168 ) Notes
24. Michael Kugelman, “When the Afghan and Pakistani Taliban Unite,” Foreign Policy,
March 25, 2014, http://southasia.foreignpolicy.com/posts/2 014/03/25/when_t he_
afghan_a nd_ pakistani_taliban_u nite
25. Saiddique, “Tehrik-e-Taliban Pakistan,” p. 9.
26. Kugelman, “When the Afghan and Pakistani Taliban Unite.”
27. See Khaled Ahmed, “The Fiction of ‘Good’ and ‘Bad’ Taliban,” Friday Times,
September 15, 2012; Fair and Jones, “Pakistan’s War Within,” pp. 162, 181; and Jones,
“Pakistan’s Dangerous Game,” pp. 19–2 0.
28. Tankel, Storming the World Stage, pp. 6, 40–41; and Haqqani, “The Ideologies of South
Asian Jihadi Groups,” pp. 24–25.
29. Kanchan Lakshman, “The Expanding Jihad,” South Asia Intelligence Review, Vol. 6, No.
2 (February 18, 2008), http://w ww.satp.org/satporgtp/sair/A rchives/6 _ 32.htm
30. Lashkar-e-Taiba, “Hum Jihad Kyon Kar Rahein Hein” [Why Do We Wage Jihad?],
S. Paul Kapur trans., p. 2.
31. Lakshman, “The Expanding Jihad.”
32. See Musharraf, In the Line of Fire, pp. 201–4; Cohen, “The Nation and the State of
Pakistan,” pp. 115–16; Gul, The World’s Most Dangerous Place, pp. 16–17, 159–6 0;
Hussain, Frontline Pakistan, pp. 33–75, 110–13; Samina Yasmeen, “Pakistan’s Kashmir
Policy: Voices of Moderation?” Contemporary South Asia, Vol. 12, No. 2 (June 2003),
p. 12; Ziring, “Weak State, Failed State, Garrison State,” p. 190; John R. Schmidt, The
Unraveling: Pakistan in the Age of Jihad (New York: Farrar, Straus and Giroux, 2011),
p. 128; “Jaish Behind Attempt to Kill Musharraf,” Daily Times, January 1, 2004; and
Tankel, Storming the World’s Stage, pp. 30, 64.
33. See Clarke, “Lashkar-I-Taiba,” pp. vi, 7–8, 10, 12, 44, 78–79; Fair et al., Pakistan,
pp. 54– 55, 101– 2; Zahid Hussain, Frontline Pakistan, pp. 53, 60– 61; Nicholas
Howenstein, “The Jihadi Terrain in Pakistan: An Introduction to the Sunni Jihadi
Groups in Pakistan and Kashmir,” Research Report No. 1 (Pakistan Security Research
Unit, February 5, 2008), p. 21; and Tankel, Storming the World’s Stage, p. 44.
34. Douglas Lynd, The Education System in Pakistan: Assessment of the National Education
Consensus (Islamabad: UNESCO, 2007), pp. 7, 8, 16.
35. Pakistan’s current rate of economic growth—averaging 2.9 percent between 2009 and
2014—is, in the words of the Economist, “dismally low,” creating large-scale unemploy-
ment and “bleak prospects” for the future, particularly for Pakistan’s large youth popu-
lation. “Pakistan’s Economy: The Urdu Rate of Growth,” Economist, February 15, 2014.
36. The proportion of full-time Pakistani schoolchildren attending madrassas is probably under
10 percent. The proportion of children supplementing their educations through mosque
schools is likely to be considerably higher. See Fair, The Madrassah Challenge, p. 10.
37. Ibid., pp. 10–11.
38. Fair et al., Pakistan, p. 124; Pakistan Ministry of Education, “FAQs,” http://w ww.moe.
gov.pk/faqs.htm. See also Jessica Stern, “Pakistan’s Jihad Culture,” Foreign Affairs, Vol.
79, No. 6 (November/December 2000), pp. 115–2 6.
39. Shuja Nawaz and Mohan Guruswamy, “India and Pakistan: The Opportunity Cost of
Conflict” (Washington, DC: Atlantic Council, 2014), p. 5.
40. Stockholm International Peace Research Institute, Military Expenditure Database
2014; and Kamran Yousef, “Defence Budget Hiked by 15 Percent,” Express Tribune,
June 13, 2013.
41. Not surprisingly, Pakistan does not publicize the portion of its defense budget devoted
to supporting militancy. For a detailed discussion of the costs of fielding a militant orga-
nization such as Lashkar-e-Taiba, see Wilson John, The Caliphate’s Soldiers: Lashkar-
e-Tayyeba’s Long War (New Delhi: Amaryllis, 2011). See also Peter Chalk, “Pakistan’s
Role in the Kashmir Insurgency,” Jane’s Intelligence Review, Vol. 13, No. 9 (September
2001), pp. 26–27.
42. See Zeeshan Haider, “Militancy-H it Pakistan Ups Defense Spending by 17 Percent,”
Reuters, June 5, 2010.
169
Notes ( 169 )
43. See Sood and Sawhney, Operation Parakaram; and Harinder Singh, “Rethinking
India’s Limited War Strategy,” Working Paper (New Delhi: Institute for Defence
Studies and Analysis, December 2010).
44. S. Paul Kapur’s discussions with senior Indian strategists, New Delhi, July and
September 2010; and Ladwig, “A Cold Start for Hot Wars?” pp. 159–6 0, 164–6 6.
45. See Manu Pubby, “No ‘Cold Start’ Doctrine, India Tells U.S.,” Indian Express,
September 9, 2010.
46. S. Paul Kapur’s discussion with senior Indian strategist closely involved with doctrinal
planning, September 2010.
47. World Bank, World Development Indicators: India, http://data.worldbank.org/coun-
try/i ndia; World Bank, World Development Indicators: Pakistan, http://data.world-
bank.org/country/pakistan; World Bank, Global Economic Prospects: South Asia,
http://w ww.worldbank.org/en/publication/g lobal- e conomic-prospects/d ata?vari
able=NYGDPMKTPKDZ&region=SAS; and “India Likely to Become 3rd Largest
Economy by 2030: Report,” Economic Times, November 6, 2013.
48. See Usman Ansari, “Pakistan Again Boosts Defense Budget, but at a Smaller Rate;”
Defense News, June 5, 2014; and Laxman K. Behera, “India’s Interim Defence Budget
2014-15: An Appraisal,” Institute for Defence Studies and Analyses Comment,
February 23, 2014.
49. S. Paul Kapur, “Ten Years of Instability in a Nuclear South Asia,” International Security,
Vol. 33, No. 2 (Fall 2008), p. 90.
50. Shashank Joshi, “Pakistan’s Tactical Nuclear Nightmare: Déjà vu?” Washington
Quarterly, Vol. 36, No. 3 (Summer 2013), p. 161.
51. Pakistan has traditionally maintained tight, centralized control over its nuclear
weapons. During a crisis, however, lines of communication between the weap-
ons and higher headquarters may get overrun, rendering the weapons unus-
able. To avoid this situation, some form of predelegation of launch authority to
field commanders during a crisis seems likely. See Barry Posen, Inadvertent
Escalation: Conventional War and Nuclear Risks (Ithaca, NY: Cornell University
Press, 1991), pp. 1–4 ; Khan, Eating Grass, pp. 331-3 32; Rajesh Basrur, “South
Asia: Tactical Nuclear Weapons and Strategic Risk,” RSIS Commentaries 65/
2011, April 27, 2011, pp. 1–2 ; Naeem Salik, “Tactical Nuclear Weapons and
Deterrence Stability,” Working Paper, Naval Postgraduate School (2012), p. 3;
Zafar Jaspal, “Tactical Nuclear Weapon: Deterrence Stability Between India and
Pakistan,” Naval Postgraduate School, Working Paper (2012), p. 9; and Joshi,
“Pakistan’s Tactical Nuclear Nightmare,” p. 165. This discussion is also based on
the author’s private discussions with Pakistani military officers and strategic ana-
lysts in Islamabad, Pakistan, in December 2011.
52. The development of a tactical nuclear weapons capability can thus be viewed as a
Pakistani precommitment strategy, creating a system in advance that will make it dif-
ficult for Pakistan not to use nuclear weapons in the event of an Indian attack. On pre-
commitment strategies see Schelling, Arms and Influence. Tactical nuclear weapons can
also serve in a denial capacity, defeating attacking forces on the battlefield and thereby
blunting enemy offensives. Recent studies suggest that Pakistani weapons are unlikely
to be effective in this role, however. See Joshi, “Pakistan’s Tactical Nuclear Nightmare,”
p. 163. In reality, Pakistani tactical nuclear weapons are likely to be most useful as
means of credibly threatening escalation of a conflict from the conventional to the
nuclear level.
53. See Ganguly and Kapur, India, Pakistan, and the Bomb, pp. 77–78.
54. S. Paul Kapur’s discussions with senior Pakistani military officers, Islamabad, Pakistan,
December 2011. See also Usman Ansari, “Pakistan Tests ‘Nuke-Capable’ Short-R ange
Missile,” Defense News, April 20, 2011; and Rajesh Basrur, “South Asia: Tactical
Nuclear Weapons and Strategic Risk,” No. 65, S. Rajaratnam School of International
Studies Commentaries (April 27, 2011), pp. 1–3.
170
( 170 ) Notes
CHAPTER 7
1. See Clarke, “Lashkar-I-Taiba ,” pp. 41–45; Fair et al., Pakistan, pp. 94–100; Gul, The
World’s Most Dangerous Place, pp. 179–86; and Hussain, Frontline Pakistan, pp. 141–53.
2. Saeed Shah, “Pakistan Vows to Target All Militants in Tribal Area,” Wall Street Journal,
July 6, 2014.
3. Saeed Shah, Safdar Dawar, and Adam Entous, “Militants Slip Away Before Pakistan
Offensive,” Wall Street Journal, July 17, 2014.
4. For a discussion of this point, see Fair, Fighting to the End, pp. 3–4, 18.
5. Although numerous definitions of rationality exist, I define it simply as an actor choos-
ing means that are designed to achieve the ends that the actor seeks. Note that this
definition does not pass judgment on the actor’s preferred ends. The fact that an actor
is pursuing goals that some might consider foolish or undesirable does not make the
actor irrational. Note also that under this definition, an actor’s mistakes or failures do
not preclude rationality. Assuming that the actor behaved in a manner that it believed
was most likely to achieve its desired ends, the actor would be rational, even if it was
ultimately proved to be wrong. For a discussion of rationality in this spirit, see Richard
A. Posner, “Rational Choice, Behavioral Economics, and the Law,” Stanford Law
Review, Vol. 50 (1997), p. 1551, 1553, 1554.
6. Paul, The Warrior State, pp. 17–23, 116–2 6. Theoretically, Paul’s argument is closely
related to an extensive literature on the relationship between natural resource endow-
ments and economic development. As this literature shows, states that are richly
endowed with natural resources, and that depend heavily on their export to gener-
ate gross domestic product, often underperform economically. This “resource curse”
results from a range of causes, including a lack of incentives for leaders to develop
optimal policies and institutions, particularly with respect to rule of law and property
rights; the production and export of resources crowding out other economic sectors
such as manufacturing; problems associated with commodity price volatility on world
markets; and an abundance of natural resources encouraging armed conflict. Paul’s
argument regarding Pakistan’s “geostrategic curse” rests on similar if not identical
logic. See Jeffrey A. Frankel, “The Natural Resource Curse: A Survey,” Working Paper
15836, National Bureau of Economic Research (March 2010), pp. 2–3.
7. This is also the case with the more general resource curse in the economic literature.
The negative correlation between states’ export of natural resources as a percentage
of gross domestic product and economic performance is only modest. In many cases,
states that depend heavily on resource exports perform well economically. Thus, the
most accurate view of the relationship between resources and development is not that
rich resource endowments necessarily lead to poor economic performance. Rather, it
is that rich resource endowments do not necessarily result in economic success. See
Frankel, “The Natural Resource Curse,” p. 3.
8. Between 1946 and 2010, Turkey received over $40 billion in total military assistance
from the United States, ranking it fourth among aid recipients, behind Israel, Vietnam,
and Egypt. Pakistan, by contrast, received over $11 billion during this period. It
ranked twelfth, behind such countries as France, Greece, and Italy. See US Overseas
Loans and Grants, United States Agency for International Development.
9. Paul, Warrior State, pp. 153–63.
10. Fair, Fighting to the End, pp. 6–7.
11. S. Paul Kapur’s discussion with Benazir Bhutto, 2004.
12. Shekhar Gupta, “Playing with Fire,” India Today, May 31, 1990.
13. Pew Global Attitudes Project, “2012 Spring Survey Topline Results,” June 27, 2012,
pp. 47, 53. Popular attitudes toward the militant groups are also fairly lenient. Only
32 percent of Pakistanis favor use of military force against militants in tribal areas.
171
Notes ( 171 )
And only 23 percent of Pakistanis have a “highly unfavorable” view of Lashkar-e-
Taiba, with 41 percent either claiming not to have an opinion or refusing to answer the
question. Ibid., pp. 57–58. It is possible, of course, that some of these public statements
and survey responses have resulted from military pressure. In most cases, however,
they seem unconnected to or even divergent from the military’s views. For example,
Zulfikar Ali Bhutto clung to his revisionist position even when Pakistan was in no posi-
tion to pursue change in Kashmir or confront India militarily. Nawaz Sharif has, in
important cases, shown even more tolerance for militants than has the military. And
ordinary Pakistanis responding to polling questions are not under military pressure.
See Victor Mallet and Farhan Bokhari, “Army Support Bolster’s Pakistan’s Nawaz
Sharif,” Financial Times, August 18, 2015.
14. See Azhar Hussain and Ahmad Salim with Arif Naveed, Connecting the Dots: Education
and Religious Discrimination in Pakistan (Washington, DC: US Commission on
Religious Freedom, 2011), pp. 15, 42–4 4, 47–4 8; Tim Craig and Haq Nawaz Khan,
“In Pakistan’s Ideological War over Textbooks, Helen Keller Doesn’t Make the Cut,”
Washington Post, February 14, 2015; and “Kashmir Is Pakistan’s Jugular Vein: Nawaz
Sharif,” Economic Times, February 5, 2015. Despite their concerns regarding Kashmir,
Pakistanis view it as being less important than a host of other problems, such as crime,
lack of employment, and corruption. See Pew Global Attitudes Project, “Pakistani
Public Opinion Ever More Critical of U.S.,” June 27, 2012, p. 23.
15. Naved Hamid, “Rethinking Pakistan’s Development Strategy,” Lahore Journal
of Economics, Special Edition (September 2008), pp. 47–4 8; and Parvez Hasan,
“Learning from the Past: A Fifty-Year Perspective on Pakistan’s Development,”
Pakistan Development Review, Vol. 36, No. 4 (Winter 1997), p. 356.
16. Hamid, “Rethinking Pakistan’s Development Strategy,” p. 48; and Hasan, “Learning
from the Past,” pp. 358–59.
17. Hamid, “Rethinking Pakistan’s Development Strategy, pp. 47–49.
18. Ibid., p. 49; and Asian Development Bank, Pakistan: Economy, http://w ww.adb.org/
countries/pakistan/economy; World Bank, GDP Growth (annual %), http://data.
worldbank.org/i ndicator/N Y.GDP.MKTP.KD.ZG
19. See generally S. Akbar Zaidi, “Pakistan’s Roller-Coaster Economy: Tax Evasion Stifles
Growth,” Carnegie Endowment for International Peace, Policy Brief 88 (September
2010), pp. 1–11.
20. Zaidi, “Pakistan’s Roller-Coaster Economy,” p. 5.
21. Hasan, “Learning from the Past,” p. 356.
22. World Bank, Annual GDP Growth.
23. World Bank, GDP per Capita, World Bank Open Data, http://data.worldbank.org/
indicator/N Y.GDP.PCAP.CD
24. UNICEF, Statistics, India, http://w ww.unicef.org/i nfobycountry/i ndia_statistics.
html; and UNICEF, Statistics, Pakistan, http://w ww.unicef.org/i nfobycountry/paki-
stan_ pakistan_statistics.html
25. Kapur, Dangerous Deterrent, pp. 24–25, 26.
26. Cohen, The Idea of Pakistan, pp. 310–11.
27. Paul, Warrior State, pp. 196–9 7.
28. Fair, Fighting to the End, p. 19.
29. Jean-Luc Racine, “Living with India,” in Jaffrelot, ed., A History of Pakistan and Its
Origins, p. 112.
30. Stephen Brooks and William Wohlforth, “Power, Ideas, and the End of the Cold
War: Reevaluating a Landmark Case for Ideas,” International Security, Vol. 25, No. 3
(Winter 2000/01), pp. 5–53.
31. Robert G. Herman, “Identity, Norms and National Security: The Soviet Foreign Policy
Revolution and the End of the Cold War,” in Peter J. Katzenstein, ed., The Culture of
National Security: Norms and Identity in World Politics (New York: Columbia University
Press, 1996), pp. 271–316.
32. Mikhail Gorbachev, Memoirs (New York: Doubleday, 1995), pp. 348–49; Archie
Brown, The Gorbachev Factor (New York: Oxford University Press, 1997), pp. 240–42;
172
( 172 ) Notes
Andrei Grachev, Gorbachev’s Gamble: Soviet Foreign Policy and the End of the Cold War
(Cambridge: Polity, 2008), pp. 180–83; and Don Oberdorfer, From the Cold War to a
New Era: The United States and the Soviet Union, 1983-1991 (Baltimore: Johns Hopkins
University Press, 1998).
33. This is by no means to suggest that Gorbachev’s efforts were wholly successful, or that
today’s Russia comports with the vision of the future that Gorbachev held while he was
in power. It is simply to point out that Gorbachev and his allies were able to respond to
an existential crisis in the Soviet Union by fundamentally changing the meaning and
purpose of the Soviet state.
34. Stephen M. Walt, “International Relations: One World, Many Theories,” Foreign Policy,
Vol. 110 (Spring 1998), pp. 29–4 6.
35. Where violent state behavior results from insecurity, the international relations lit-
erature generally prescribes reassurance, to reduce the state’s motive for violence. See
Robert Jervis, “Cooperation Under the Security Dilemma,” World Politics, Vol. 30, No.
2 (January 1978), pp. 176–214.
173
INDE X
Abdullah, Farooq, 82 Bari, Maulana Abdul, 160–61n74

Abdullah, Sheikh, 53 Bengal, 38
Afghanistan Bhutto, Benazir, 5, 98, 131
mujahideen in, 9, 71–78, 79, 84, 104 Bhutto, Zulfikar Ali
Pakistan and: Hizb-i-Islami and, 72–73, Awami League and, 60–61
109; Islamist militants and, 8, 22; Bangladesh war and, 66–67
mujahideen and, 71–78, 79, 84, 104; economy and, 133
Soviet invasion and, 70–71; Taliban execution of, 71
and, 79–80, 81, 103, 104, 105–7, Islamist militants and, 7
115–16; warlords and, 96–98 on Kashmir, 131, 171n13
punishment campaigns and, 106–8 nuclear weapons and, 68–70
Soviet invasion of, 52, 70–71, 115 second Kashmir war and, 54, 59
Taliban in: characteristics of,98–9 9; bin Laden, Osama, 1, 101
Pakistan and, 79–80, 81, 98–102, 103, Black September, 9
104, 105–7, 109, 115, 118–19 Blood, Archer, 61
United States and, 9, 72–73, 106, 115–16 Brohi, Allah Bukhsh K., 41
Ahmad Khan, Syed, 35 Burki, Shahid Javed, 122
al Omar, 90 Bush, George H. W., 139
Ali Khan, Liaquat, 40, 44 Bush, George W., 101
Ali Khan, Rao Farman, 64 Byman, Daniel, 147n33
al-Jehad, 90
al-Qaeda, 101, 150n22 Central Intelligence Agency (CIA), 72, 118
Armitage, Richard, 94 Central Treaty Organization
Asif, Khawaja, 127 (CENTO), 39–4 0
Atlee, Clement, 47 China, 40, 54, 65–6 6, 130
Aurora, Jagjit Singh, 64 Cohen, Stephen, 136
Awami League, 60–61 “Cold Start” Doctrine, 123–2 4, 125, 136
Ayub Khan, 53, 54–55 Cold War, 148n1, 150n24
Correlates of War Project, 13
Babar, Naseerullah, 98, 99 Cunningham, George, 40
Balochistan, 35, 38, 44, 98
Bangladesh war (1971) Daoud, Mohammad, 70
economy and, 133 Dar, Abdul Majid, 89
militant strategy and, 8, 51–52, denial campaigns
64–70, 79 Afghanistan and, 106
origins and events of, 59–6 4 India and, 22–23, 46–47, 57, 58, 104
Pakistani conventional forces and, 22, militant proxy strategy and,
30, 51–52 15–18, 30–31
bargaining, 18–19, 26–27, 31 Soviet Union and, 77–78
174
( 174 ) Index
deterrence, 15 punishment campaigns by, 23–2 6,

Dostum, Abdul Rashid, 96–9 7 106–7, 157n5
Dunford, Joe, 102 Sundarji Doctrine and, 123
Durand Line, 98, 100, 159n57 See also Kashmir
Indian Air Force, 62, 63–6 4
East Pakistan, 8, 33, 60–6 4 Indian Army, 44, 62, 64, 93–94
The Economist (newspaper), 2 Indian Councils Act (1909), 35
economy, 133–35 Indian National Congress (INC), 35–37,
education, 27, 68, 121 38, 82–8 4
Indian Navy, 62, 63–6 4
Fair, C. Christine, 130–31, 137, 146n23 Indian parliament attack (2001), 26, 92,
Falah-e-I nsaniyat, 95 107, 123, 136
fedayeen attacks, 91 Indo-Pakistani War (1947)
Fernandes, George, 152n47 militant strategy and, 8–9, 22, 29,
Foreign Policy (journal), 2 43–50, 51
Pakistani conventional forces and, 8, 22
game theory, 13 results of, 52
Gandhi, Indira, 63, 66 Indo-Pakistani War (1965)
Gandhi, Mohandas Karamchand, militant strategy and, 8–9, 22, 29, 51,
33–3 4, 36 52–59, 78–79, 103
Gandhi, Rajiv, 82 Pakistani conventional forces and, 8, 22
Ganguly, Sumit, 83 International Monetary Fund (IMF),
Goodson, Larry, 100 39–4 0, 133–3 4
Gorbachev, Mikhail, 75, 138–39 international relations (IR) scholarship, 13
Great Britain, 47–4 8 International Security Assistance Force
(ISAF), 106
Hamoodur Commission, 158n30 Inter-Services Intelligence (ISI)
Haqqani, Jallaludin, 101–2 Afghan war and, 72
Haqqani network, 101–2 , 127–2 8 Haqqani and, 101
Harakat-u l-Mujahideen, 92 Kashmir insurgency and, 85, 90
Hari Singh, Maharaja, 42, 43–4 4, 45 Lashkar-e-Taiba and, 91, 93, 95, 119
Harkat-u l-A nsar, 90 Pashtuns and, 98
Hasan, Parvez, 122 second Kashmir war and, 54
Hazbatral Affair, 53 Iqbal, Muhammed, 35–36, 40–41
Hekmatyar, Gulbuddin, 72–73, 96–98, Iran, 9
99, 109 Islamist militancy in Pakistan
Hinduism, 131 Afghanistan and: mujahideen and,
Hizb-i-Islami, 72–73, 109 71–78, 79, 84; Taliban and, 79–80,
Hizbollah, 9 81, 98–102, 103, 104, 105–7, 109, 115,
Hizb-u l-Mujahideen (HM), 88–9 0, 108 118–19; warlords and, 96–98
Bangladesh war and, 51–52, 64–70, 79
Iftikharuddin, Mian, 43–4 4 benefits of, 21–27, 30–31, 46–49, 77–79,
Ikhwanul Muslimeen, 90 111, 112–16
India costs of, 27–2 8, 31, 111–12, 116–125
“Cold Start” Doctrine and, 123–2 4, failure to abandon, 127–136
125, 136 first Kashmir war and, 46–50, 51, 52
denial campaigns by, 22–23, 46–47, 57, as grand strategy, 8–10
58, 104 Haqqani network and, 101–2
economy of, 134, 136 international stability and, 1–2
Muslims in, 23, 35–39 Kashmir insurgency and, 81,
nonstate actors and, 61–62, 157n23 84–96, 103–5
175
Index ( 175 )
Lashkar-e-Taiba and, 90–93 Kashmir insurgency

nuclear weapons and, 69–70 costs to India of, 30, 114–15
origins of, 31, 32–43 origins of, 82–8 4
possible changes in, 136–140 Pakistan’s role in, 8, 9, 22, 30, 79, 81,
potential objections to arguments 84–96, 103–5
on, 28–30 Keating, Kenneth, 61
price of, 21–22, 31, 46, 121–23 Khan, Abdul Hamid, 62–63
problems with, 81–82, 108–9 Khan, Akbar, 43–45, 52–53, 54
role of religion in, 67–68, 78–79, 112–14 Khan, Ismael, 96–9 7
scholarship on, 3–7 Khan, Sher, 44
second Kashmir war and, 51, 52–59,
78–79, 103 Lashkar-e-J hangvi, 92
shortcomings of, 114 Lashkar-e-Taiba (LeT)
Soviet Union and, 52 ambitions of, 27
funding of, 21–22
Jaamat-ud-Daawa, 128 Indian parliament attack and, 26, 92
Jaffrelot, Christophe, 113 Mumbai attacks and, 1–2 , 22–23, 93,
Jaish-e-Mohammed (JeM) 95, 105
fedayeen attacks and, 162n30 Pakistan and, 90–96, 108–9,
Indian parliament attack and, 26, 93 119–121, 128
Kashmir insurgency and, 90 Red Fort attack and, 92
Pakistan and, 92, 120 League of Nations, 47
Jalal, Ayesha, 36, 38 Lebanon, 9
Jamaat Ulema-e-Islam (JUI), 98 liberalism, 13
Jamaat-e-Islami (JI), 88 London bomb attacks (2005), 1
Jamaat-ud-Dawa (JuD), 95
Jamal, Arif, 160–61n74 Madani, Hussain Ahmad, 155n31
Jammu & Kashmir National Conference Malhotra, Jagmohan, 85–86
(JKNC), 82–8 4 Malik, A. M., 64
Jammu and Kashmir Liberation Front Malik, Akhtar, 55
(JKLF), 76, 84–85, 86–88 Malik, S. K., 41
Janjua, M. K., 44 Malik, V. P., 152n47
Javed, Nasr, 119, 120 Massoud, Ahmed Shah, 97
jihad, definition of, 146n28 Masud, Tommy, 44
jihad paradox, 10, 28, 31, 125, 126, 127 mehmaan mujahideen, 90–92, 108. See also
Jinnah, Mohammed Ali, 34, 35–36, 37, Lashkar-e-Taiba (LeT)
38–39, 42, 44 Meshud, Hakimullah, 117
Middle East, 47
Kargil war (1999), 8, 24, 30, militant proxy strategy
151n37, 152n48 benefits of, 14–19, 21, 30–31, 32
Kashmir costs of, 20–21, 31, 32
Jammu and Kashmir Liberation Front general observations on, 140–42
and, 76 India and, 61–62, 157n23
Lashkar-e-Taiba and, 119 scholarship on, 13
role and importance of, 40–43, 49, 100, vs. state-level proxies, 148n1
115, 131–32 See also Islamist militancy in Pakistan
role of militant proxy strategy Mohammed, Khalid Shaikh, 1
in, 32–33 Mountbatten, Louis, 45, 47
Simla Agreement and, 66–67, 131 Muhammedan Anglo-Oriental College
See also Indo-Pakistani War (1947); (Aligarh), 35
Indo-Pakistani War (1965) Mukti Bahini, 61–62, 64–65
176
( 176 ) Index
Mullen, Michael, 102 Pashtuns and, 98

Mumbai attacks (2008), 1–2 , 22–23, 93, role of, 130–32
95, 105 Pakistan Navy, 62
Musharraf, Pervez Pakistan People’s Party (PPP), 60–61, 98
Islamist militants and, 5, 27 Palestine, 47
Jaish-e-Mohammed and, 92, 94, 120 Partition (1947), 33–4 0
Lashkar-e-Taiba and, 94 Pashtuns, 37–38, 98
Taliban and, 101 Paul, T. V., 39, 130, 137
on terrorism, 107 principal–agent problems, 19, 20, 31,
war on terror and, 100 48–49, 132–33, 141
Muslim League, 35–37, 38–39, 42 punishment campaigns
Muslim United Front (MUF), 83–8 4 Afghanistan and, 106–8
India and, 23–2 6, 106–7, 157n5
Najibullah, Mohammad, 75 militant proxy strategy and, 15,
Nasr missiles, 124, 136 18, 30–31
Nehru, Jawaharlal Soviet Union and, 78
first Kashmir war and, 45, 47 Punjab, 35, 38
on Indian National Congress, 36 Punjabiyat, 23
Jinnah and, 154n9
Kashmir and, 42 The Quranic Concept of War
second Kashmir war and, 52, 53–5 4 (Malik), 41
Niazi, Amir Abdullah Khan, 64
Nixon, Richard, 63 Racine, Jean-Luc, 113, 137–38
North Korea, 148n1 Radcliffe, Cyril, 153n2
Northern Light Infantry (NLI), 30, 151n37 Rahim, Khwaja Abdul, 44
North-West Frontier Province (NWFP), Rashid, Ahmed, 100, 101
35, 37–38, 44, 98 realism, 13
nuclear weapons Red Brigades, 9
as security challenge, 2 Red Fort attack (2000), 92
as strategic tool, 8, 24, 68–70, 124–25, Rehman, Maulana Fazlur, 98
128, 135–36 Rehman, Sheikh Mujibur, 60–61
United States and, 71 religion, 67–68, 78–79, 112–14, 131–32
revisionism, 131–32
Operation Enduring Freedom (OEF), 106 Rodgers, William, 61
Operation Gibraltar, 55–56, 57, 59, 103 Roy, Olivier, 41, 112
Operation Grand Slam, 56–57, 58
Operation Jackpot, 64 Saeed, Hafiz Mohammad, 95–96
Operation Parakram, 93–94 Saeed, Mohammed, 128
Operation Searchlight, 61 Saudi Arabia, 72
Saur Revolution (1978), 70
Pakistan Air Force, 62 secrecy, 28–29, 30
Pakistan Army secularism, 112
compared to Indian Army, 62–63 September 11 attacks (2001), 1, 2, 100
counterinsurgency and, 127 Shahid, Shahidullah, 117, 118
Islamization of, 68 Shaikh, Farzana, 42
Lashkar-e-Taiba and, 91, 93 Sharif, Nawaz, 5, 117, 131–32, 171n13
Operation Gibraltar and, 55–56, 57, Simla Agreement, 66–67, 131
59, 103 Sind, 35
Operation Grand Slam and, 56–57, 58 South Korea, 148n1
Operation Searchlight and, 61 Southeast Asia Treaty Organization
Pakistani Taliban and, 117 (SEATO), 39–4 0
177
Index ( 177 )
Soviet Union Cold War and, 148n1, 150n24

Cold War and, 150n24 Indian parliament attack and, 26
invasion of Afghanistan by, 52, 70–71, 115 Pakistan and: Bangladesh war and,
nonstate actors and, 9, 148n1 65–6 6; bilateral security
Pakistan and, 159n57 relationships and, 40; geostrategic
transformation of, 138–39 curse and, 130; India and, 94, 107–8 ,
Stinger antiaircraft missiles, 74–75 136; influence on, 136–37; Kashmir
strong nationalism, 113 and, 76; nuclear weapons and, 71;
Sundarji Doctrine, 123 Taliban and, 102; war on terror and,
Supreme Court of India, 86 100–101, 132
Swami, Praveen, 84, 87, 146n23, 152n51 Soviet Union and, 139
Turkey and, 130
Talbot, Ian, 39, 113 war on terror and, 26–27, 81,
Taliban 100–101, 132
characteristics of, 98–9 9
Pakistan and, 79–80, 81, 98–102, 103, Vietnam, 148n1
104, 105–7, 115, 118–19
See also Tehrik-e-Taliban war on terror, 26–27, 81, 100–101, 132
Pakistan (TTP) West Pakistan, 33, 60–61
Tankel, Stephen, 95 Whitehead, Andrew, 45
Tariq bin Ziyad, 45, 55 Wirsing, Robert G., 72
taxation, 134 World Bank, 39–4 0
Tehrik-e-Taliban, 27 World Food Programme, 71
Tehrik-e-Taliban Pakistan (TTP), 117–19,
120, 122–23, 127, 132 Yahya Khan, 60–61, 64
Terrorist and Disruptive Activities
(Prevention) Act (TADA, 1987), 86 Zaidi, S. Akbar, 134
Turkey, 130 Zia-u l-Haq, Muhammad
Afghan war and, 73, 75–76, 84, 100
United Nations (UN), 45, 47, 57, 66, 95 economy and, 133
United Nations Commissioner for Islamist militants and, 5, 7
Refugees, 71 on Islamization, 68
United Nations Security Council, 48, 52 Kashmir insurgency and, 85
United States The Quranic Concept of War
Afghanistan and, 9, 72–73, 106, 115–16 and, 155n31
Bangladesh war and, 61, 65–6 6 Ziring, Lawrence, 68, 75
178

Jihad As Grand Strategy - Islamist Militancy, National Security, and The Pakistani State

Uploaded by

Copyright:

Available Formats

You might also like

Jihad As Grand Strategy - Islamist Militancy, National Security, and The Pakistani State

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Jihad As Grand Strategy - Islamist Militancy, National Security, and The Pakistani State

Uploaded by

Copyright:

Available Formats

i

Jihad as Grand Strategy

Published in the United States of America by Oxford University Press

© Oxford University Press 2017

All rights reserved. No part of this publication may be reproduced, stored in

You must not circulate this work in any other form

CIP data is on file at the Library of Congress

T errorism’s ascendance as one of the world’s leading strategic dan-

( 2 ) Jihad as Grand Strateg y

militant group Lashkar-​e-​Taiba. They trained for their mission under

The Pakistan–​militant nexus has come under intense public scrutiny in

( 4 ) Jihad as Grand Strateg y

independence, the impact of the Iranian revolution and the anti–​Soviet-​

its motives, strengths, and weaknesses, and help to devise techniques of

( 6 ) Jihad as Grand Strateg y

Prior scholarship thus has made important contributions to our under-

( 8 ) Jihad as Grand Strateg y

subsequently participated, such as the 1947 and 1965 wars. In addition,

( 10 ) Jihad as Grand Strateg y

Nonetheless, the reality of Pakistan’s militant policy is considerably

The Plan of the Book

The book proceeds as follows: Chapter 2 discusses the general logic of

( 12 ) Jihad as Grand Strateg y

Chapter 4 examines the evolution of Pakistan’s militant strategy from

The Logic of a Militant

P akistan’s use of militancy is widely dismissed as being a foolish and

( 14 ) Jihad as Grand Strateg y

THE BENEFITS OF A MILITANT PROX Y STRATEGY

The first benefit of militant proxy strategies is straightforward: They are

There are two main avenues of military response open to a strong

( 16 ) Jihad as Grand Strateg y

It is important to note that militant proxies do not undermine strong-​

( 18 ) Jihad as Grand Strateg y

the attacks, it may still be unsure as to who is facilitating their activity by

( 20 ) Jihad as Grand Strateg y

THE COSTS OF A MILITANT PROX Y STRATEGY

The benefits of a militant proxy strategy are likely to be particularly

The logic of a militant proxy strategy thus involves a complex mix of

THE BENEFITS AND COSTS OF PAKISTAN’S

Pakistan’s use of Islamist militants has consistently exploited the cost,

( 22 ) Jihad as Grand Strateg y

since Pakistan supports multiple militant organizations, let us assume

( 24 ) Jihad as Grand Strateg y

provocations. From an operational standpoint, the Indians have not

India could have attempted to avoid these problems by designing

( 26 ) Jihad as Grand Strateg y

( 28 ) Jihad as Grand Strateg y

avoid: a direct Indo-​Pakistani military confrontation. Significantly, given

Here, I consider a number of potential counterarguments to the claims

( 30 ) Jihad as Grand Strateg y

During the Kargil conflict, offensive forces were not members of

undermine the adversary’s ability to respond to weak-​state provocations,

PARTITION AND THE SOURCES OF PAKISTAN’S

The primary motivation for Pakistan’s initial resort to a militant proxy

( 34 ) Jihad as Grand Strateg y

of cash balances until January 1948, after intense pressure, including a

Indian Muslims needed their own autonomous homeland to practice

militant group Lashkar-e-Taiba. They trained for their mission under

The Pakistan–militant nexus has come under intense public scrutiny in

independence, the impact of the Iranian revolution and the anti–Soviet-

It is important to note that militant proxies do not undermine strong-

avoid: a direct Indo-Pakistani military confrontation. Significantly, given

undermine the adversary’s ability to respond to weak-state provocations,

institutions. Indeed, he had believed that the nation-state was nothing

all-weather friend. When such assistance was not immediately forthcom-

most hardline Islamist groups, such as Gulbuddin Hekmatyar’s Hizb-i-

abilities and significantly increased the pace and intensity of anti-Soviet