Net Versity

in Pakistan have repeatedly been involved in major terrorist incidents throughout the world, such as the September 11, 2001, attacks on New York and Washington and the 2005 London subway bombings.¹ They regularly strike government, coalition, and civilian targets in Afghanistan, hampering efforts to stabilize the country. Also, they frequently target India, threatening to incite an Indo-Pakistani conºict that could potentially escalate to the nuclear level. Pakistan-based militancy thus severely undermines regional and international security.

Although this problem has received widespread international attention since 2001 and the advent of the United States’ “global war on terror,” the Pakistan-militant nexus is as old as the Pakistani state. From its founding in 1947 to the present day, Pakistan has used religiously motivated militant forces as strategic tools.² How and why did this situation come about? How has

S. Paul Kapur is Professor in the National Security Affairs Department at the U.S. Naval Postgraduate School and a faculty afªliate at Stanford University’s Center for International Security and Cooperation. The arguments advanced in this article are solely his and do not necessarily reºect the views of any other individual or of the U.S. government. Sumit Ganguly is Professor of Political Science and holds the Rabindranath Tagore Chair in Indian Cultures and Civilizations at Indiana University, Bloomington. He is also Senior Fellow at the Foreign Policy Research Institute.

The authors thank International Security’s anonymous reviewers for helpful comments.

1.                    By “Islamist militants,” we mean nonstate actors who use violence to pursue a sociopolitical agenda based at least in part on their interpretation of Islamic religious principles. We do not label these actors “terrorists” because, although they often attack noncombatant targets, they do not do so exclusively; they also strike military, police, and other government assets. Moreover, they often seek not only to inºuence target audiences, but also to achieve battleªeld victories. Deªnitions vary, but terrorism is usually understood more narrowly as violence by nonstate actors that is (1) directed against noncombatants; (2) intended to coerce or garner support among particular audiences rather than to win on the battleªeld; or (3) both. See, for example, Annual Country Reports on Terrorism, U.S. Code, title 22, sec. 2656(f); Robert A. Pape, Dying to Win: The Strategic Logic of Suicide Terrorism (New York: Random House, 2005), p. 9; and Daniel Byman, Deadly Connections: States That Sponsor Terrorism (Cambridge: Cambridge University Press, 2005), pp. 8 – 9.

2.                    By “Pakistan,” we mean the group of decisionmakers 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 Philip Cohen, The Idea of Pakistan (Washington, D.C.: Brookings Institution Press, 2004), pp. 97–130. The importance of religion to the militants has varied across cases. In some instances, such as the 1947 Kashmir war, the militants were probably as concerned with pillage and plunder as with pursuing religious goals. In other instances, such as the current Kashmir insurgency, the promotion of an Islamist sociopolitical agenda has been one of the militant groups’ primary aims. In all cases, however, Islam has

International Security, Vol. 37, No. 1 (Summer 2012), pp. 111–141

No rights reserved. This work was authored as part of the Contributor’s official duties as an Employee of the United States Government and is therefore the work of the United States Government. In accordance with 17 U.S.C. 105, no copyright protection is available for such works under U.S. law.

111

Pakistan’s strategic use of militants affected its security interests?³ And what general lessons does the Pakistan case suggest regarding states’ use of violent nonstate actors to promote their national security goals? This article addresses these questions.

Scholars have not systematically evaluated Pakistan’s use of militants as a long-term national security strategy. Most describe the broad processes by which Pakistan has become Islamized and militarized in recent decades, thereby creating an environment conducive to the growth of militancy.^[1] Those scholars who do assess Pakistan’s use of militants tend to view the issue narrowly, as a tactic employed since the 1980s to bolster the anti-Indian insurgency in the disputed state of Kashmir, and since the Afghan civil war to support the rise of the Taliban.^[2] Almost all discussions are highly critical of

played a signiªcant role in motivating the militants. See, generally, Praveen Swami, India, Pakistan, and the Secret Jihad: The Covert War in Kashmir, 1947–2004 (London: Routledge, 2006). By “strategic tools,” we mean military instruments intended to advance Pakistani security interests. In using these military instruments, Pakistan has exerted varying levels of supervision over the militants. In some cases, such as the 1965 Indo-Pakistani war, Pakistani ofªcers were embedded with and directly led militant combat units. In other cases, such as the Kashmir insurgency, the Pakistanis have provided the militants with material assistance, advice, and training, but generally have not participated in their operations. In every instance, however, Pakistan has used the militants to carry out violent activity that Pakistani leaders believed would advance their security goals.

3. Leading research on state support for violent nonstate actors primarily examines the impact of state backing on the interests and capabilities of terrorist organizations. See, for example, Byman, Deadly Connections; and Daniel Byman, Peter Chalk, Bruce Hoffman, William Rosenau, and David Brannan, Trends in Outside Support for Insurgent Movements (Santa Monica, Calif.: RAND, 2001). By contrast, we focus on the impact of Pakistan’s support for Islamist militancy on Pakistani interests.

Pakistan’s militant policy, characterizing it as the product of chronic misjudgment and careless decisionmaking—“strategic myopia,” in the words of one scholar.⁶

We show, by contrast, that Pakistan’s use of militancy is not simply the ancillary product of broad social and political changes in the country. Nor is it merely an ill-conceived tactic designed to support the Kashmir insurgency or the Taliban. Rather, it is the centerpiece of a sophisticated asymmetric warfare campaign,^[3] painstakingly developed and prosecuted since Pakistan’s founding. This campaign has constituted nothing less than a central component of Pakistani grand strategy;^[4] supporting jihad has been one of the principal means by which the Pakistani state has sought to produce security for itself.^[5]

Militant Challenge in Pakistan,” Asia Policy, No. 11 (January 2011), pp. 105–137. Seth G. Jones and Fair provide a brief overview of Pakistan’s use of militancy since independence, but they devote the bulk of their attention to assessing Pakistan’s recent counterinsurgency efforts and suggesting means of aligning Pakistani behavior more closely with U.S. security interests. See Jones and Fair, Counterinsurgency in Pakistan (Santa Monica, Calif.: RAND, 2010). Arif Jamal also shows that the use of Islamist militants was among “the ªrst elements of Pakistan’s foreign and defense policy” and “remain[s] so more than sixty years later.” Jamal, however, seeks primarily to provide a highly detailed, descriptive history of Pakistan’s militant strategy, rather than to assess the strategy’s causes and evaluate its consequences. See Jamal, Shadow War: The Untold Story of Jihad in Kashmir (New York: Melville House, 2009), p. 46.

6. Timothy Hoyt, “Pakistani Nuclear Doctrine and the Dangers of Strategic Myopia,” Asian Survey, Vol. 41, No. 6 (November/December 2001), pp. 956–977. See also Ahmad Faruqui, Rethinking the National Security of Pakistan: The Price of Strategic Myopia (Aldershot, U.K.: Ashgate, 2003); Altaf Gauhar, “Four Wars, One Assumption,” Nation, September 5, 1999; Shaikh, Making Sense of Pakistan, pp. 207–208; and Fair et al., Pakistan, pp. 109–112. Important exceptions exist. Byman, for example, identiªes a number of instances in which Pakistan’s use of militancy has paid dividends. He limits his study, however, mainly to Pakistan’s involvement in the Kashmir insurgency since 1989. See Byman, Deadly Connections, pp. 177–178. Mariam Abou Zahab and Olivier Roy characterize Pakistan’s use of Islamist militancy as “a carefully thought-out regional policy,” but they focus more on offering detailed accounts of the policy’s implementation than on evaluating its strategic efªcacy and implications. See Zahab and Roy, Islamist Networks: The Afghan-Pakistan Connection (New York: Columbia University Press, 2004), p. 53.

Far from an unmitigated disaster, the strategy has enjoyed important domestic and international successes. Recently, however, Pakistan has begun to suffer from what we call a “jihad paradox”; the very conditions that made Pakistan’s militant policy useful in the past now make it extremely dangerous. Thus, despite its historical beneªts, the strategy has outlived its utility, and Pakistan will have to abandon it if it is to avoid catastrophe.

Below, we trace the evolution of Pakistan’s strategy, from its initial use of tribal forces during the ªrst Kashmir war to its current support for a range of militant organizations in the South Asian region. We show that it emerged in the wake of Partition, speciªcally, out of the new Pakistani state’s acute material and political weakness. Once adopted, the militant strategy became a central component of Pakistani security policy, its sophistication and importance increasing with each subsequent conºict.

In the next section, we evaluate the impact of Pakistan’s militant policy on its security interests. We show that the policy has produced a number of important beneªts. First, it has helped to promote national unity despite Pakistan’s lack of a coherent founding narrative. Second, it has facilitated Pakistani efforts to redress the sharp imbalance in Indian and Pakistani material resources. Third, it has enabled Pakistan to continually challenge Indian control of Kashmir. Finally, it has allowed Pakistan to pursue its strategic goals in Afghanistan without having to intervene there militarily.

We then explain that the weakness that originally made supporting Islamist militants attractive to Pakistan has rendered the strategy extremely dangerous. Speciªcally, Pakistan-based militant organizations have become so powerful that they have begun to exceed the Pakistani state’s ability to control them.

and in principle it need not be violent. Nonetheless, waging war against unbelievers is a prominent form of jihad. See Ignaz Goldziher, Introduction to Islamic Theology and Law, trans. Andras and Ruth Hamori (Princeton, N.J.: Princeton University Press, 1981), pp. 102–103; and John L. Esposito, Islam: The Straight Path (Oxford: Oxford University Press, 2011), pp. 113–115. The Islamist militants whom we discuss in this article, as well as a host of analysts and commentators, commonly use the term “jihad” to refer to the militants’ violent activities. We use the term in that spirit here. Pakistan’s other main grand strategic tools are its nuclear and conventional forces. Nuclear weapons have played a wholly defensive role, deterring Indian conventional and nuclear attacks against the Pakistani homeland. Conventional forces have served a combination of purposes. In some cases, such as the 1947 and 1965 wars, they have joined conºicts that militants had already launched against India. In one instance, the 1971 Bangladesh conºict, they fought a war essentially on their own. In other cases, such as the Kashmir insurgency and the Afghan conºicts, they have avoided direct involvement. And throughout Pakistan’s history, they have provided Pakistan with a robust defense against any Indian conventional attack. Militant forces, by contrast, have served as Pakistan’s primary offensive tool, starting wars in which conventional forces have subsequently participated, as well as waging the Kashmir insurgency and shaping the Afghan security environment, largely on their own. In doing so, they have enabled Pakistan to pursue its most cherished security goals, which have generally required the state to alter the existing strategic environment through offensive military action.

They are increasingly in a position to pursue their own policies, which often damage Pakistani security interests. In addition, supporting militancy has imposed harmful opportunity costs on Pakistan, consuming scarce resources needed for the country’s internal development. Finally, Pakistan’s militant strategy has led India to begin developing signiªcant new offensive capabilities. As a result, Pakistan could soon face a far more potent Indian military and ªnd itself even less secure than it was before.

In the concluding section, we explore possible solutions to these problems and identify broad lessons that may be applicable beyond the Pakistani case. We argue that both India and Pakistan should now adopt policies that depart from their traditional strategic behavior. Pakistan must recognize that its strategy of supporting jihad has become a liability, unequivocally end its support for militancy, and make serious efforts to defeat the militant organizations operating in its territory. India, in turn, could facilitate these efforts by reducing its military pressure on Pakistan. More generally, we argue that the fundamental problems underlying Pakistan’s jihad paradox are likely to apply beyond the Pakistani context. Weak states, which will tend to ªnd the strategic use of nonstate actors particularly attractive, will also be especially prone to losing control of their proxies, suffering damaging opportunity costs, and worsening their security relationships with stronger adversaries. Ironically, state weakness makes the strategic use of nonstate actors at once attractive and dangerous.

The Evolution of Pakistan’s Militant Strategy

The primary motivation for Pakistan’s militant strategy has been its material and political weakness. Pakistan emerged from the ashes of British India’s partition in an extremely vulnerable position.^[6] It had received even less than its ofªcial share of 18 percent of British India’s ªnancial resources and 30 percent of its military assets. East and West Pakistan were separated by 1,000 miles of Indian territory and lacked natural resources, industrial capacity, and strategic depth. And unlike India, which inherited the British Raj’s administrative system, Pakistan had to construct a central government largely from scratch. This impeded its ability to coordinate national, provincial, and local affairs.^[7]Pakistan was not only physically weak; it also lacked a solid ideological foundation. Prior to independence, Muhammad Ali Jinnah and his political party, the All India Muslim League, had based their case for the foundation of Pakistan on what was known as the “two-nation theory.” The theory maintained that India’s Muslim and Hindu communities made up separate nations. Muslims needed their own homeland to practice their religion and culture free from Hindu domination.¹² This necessitated, in the poet Muhammed Iqbal’s words, “the fullest national autonomy” for Indian Muslims.¹³ It did not require a separate state, however. An autonomous Muslim nation could have existed within a larger Indian union under a special power-sharing agreement.¹⁴

The need for Pakistani statehood therefore was not universally obvious to British India’s Muslims. This was especially true in rural areas, beyond the Muslim League’s urban strongholds. In the Northwest Frontier, for example, Pashtuns enjoyed a signiªcant numerical majority and did not fear Hindu domination in an independent India. In addition, partition seriously threatened to damage the interests of the Muslims who would live in the new Pakistan. Those in the Punjab would lose the government ministries that they had come to dominate, as well as highly productive agricultural areas. Those from Bengal would lose the political and economic hub of Calcutta. The ªercely independent tribes of the northwest and Baluchistan would be forced to submit to the central authority of the Pakistani state. In addition, Muslims in other regions, who remained behind as part of independent India, would be left without the support of their Pakistani brethren. 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.”¹⁵

Given these problems, the Muslim League offered only a vague public justiªcation for the creation of Pakistan. In Jinnah’s words, Pakistan would provide a “[s]tate 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-

bridge: Cambridge University Press, 1990), pp. 42, 47; and Ayesha Jalal, Democracy and Authoritarianism in South Asia: A Comparative and Historical Perspective (Cambridge: Cambridge University Press, 1995), p. 18.

12.                 Christophe Jaffrelot, “Islamic Identity and Ethnic Tensions,” in Jaffrelot, ed., A History of Pakistan and Its Origins, trans. Gillian Beaumont (London: Anthem, 2000), pp. 12–13.

13.                 Muhammad Iqbal, “1930 Presidential Address to the 25th Session of the All-India Muslim League,” in Latif Ahmad Sherwani, ed., Speeches, Writings, and Statements of Iqbal (Lahore: Iqbal Academy Pakistan, 2005), p. 178.

14.                 Haqqani, Pakistan, p. 5. See also Jalal, Democracy and Authoritarianism in South Asia, p. 14; and Mazhar Aziz, Military Control in Pakistan: The Parallel State (London: Routledge, 2008), pp. 19–20.

15.                 Ayesha Jalal, The Sole Spokesman: Jinnah, the Muslim League, and the Demand for Pakistan (Cambridge: Cambridge University Press, 1985), pp. 2–3. See also Ian Talbot, Pakistan: A Modern History (New Delhi: Oxford University Press, 1999), pp. 66–67, 85, 93; Cohen, The Idea of Pakistan, p. 25 ; and Jaffrelot, “Islamic Identity and Ethnic Tensions,” p. 10.

ciples of Islamic social justice could ªnd free play.”^[8] Such language held widely divergent meanings for different audiences. To some, it suggested that Pakistan would be an Islamic state, governed according to religious principles. To others, it suggested that although Pakistan was to serve as a Muslim homeland, it would do so in a pluralistic and wholly secular manner. This lack of clarity increased Pakistan’s appeal to a diverse range of constituents, but it also created a crisis of identity and deep uncertainty as to Pakistan’s reason for existence. What was Pakistan’s purpose and why was its creation necessary?

No immediate answers to these questions were evident.^[9]

In the wake of Partition, Pakistan thus suffered from severe material and political shortcomings. How could the new Pakistani state redress these problems? Acquiring the Muslim-majority territory of Kashmir, located directly between the new Pakistani and Indian states, offered one means of doing so. 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 conºict, and the region contained important water resources.^[10]

Even more signiªcant, however, the acquisition of Kashmir could strengthen Pakistan’s tenuous political foundations. Pakistani leaders decided that, if their new country were to survive, it would have to be more than just a pluralistic homeland for South Asian Muslims. Rather, it would need to become a state based on a concept meaningful to the majority of ordinary Pakistanis— a state based on Islam.^[11] Establishing Pakistani control over Kashmir could play a central role in this project, afªrming Pakistan’s Islamic identity and supporting the notion that South Asian statehood should be determined on the basis of religion.^[12] Thus, as Farzana Shaikh argues, foreign policy could serve “as a vital compensation for [Pakistan’s] lack of a clearly deªned sense of nationhood.”^[13]

Acquiring Kashmir would not be easy, however. Maharaja Hari Singh governed the territory, and he remained undecided as to whether to join Kashmir to India or to Pakistan.^[14] If Pakistan attempted to seize Kashmir, it could potentially face not only the maharaja’s forces, but also the Indian military, as New Delhi was acutely interested in the territory’s fate.^[15] Given Pakistan’s material weakness, this would be a risky proposition. An important factor offset the danger, however: Pakistan had devised a military strategy that would minimize the direct use of its forces in Kashmir, thus lowering the risk of largescale Pakistani losses.

the ªrst kashmir war

Pakistani strategy followed the broad contours of Pakistan Army Col. Akbar Khan’s secret plan for “armed revolt inside Kashmir.” The Pakistanis ªrst sought to capitalize on brewing unrest in the Poonch region of Kashmir, helping local rebels to transform discontent with the maharaja’s rule into a full-blown revolt. Second, the Pakistanis planned to assist militias of several thousand Afridi tribesmen in launching an external attack on Kashmir. This combination of internal and external pressure, the Pakistanis hoped, would overthrow the maharaja and ensure Kashmir’s accession to Pakistan. Pakistan would avoid direct involvement in the operation, playing only such supporting roles as arming the ªghters; operating their radio network; supplying them with food, clothing, and other matériel; cutting regional road and rail links; and preventing the provision of such essentials as food and gasoline to Kashmir.^[16] The Pakistanis guarded their preparations closely. Khan employed a small group of military and civilian collaborators and kept British ofªcers assigned to the Pakistan Army in the dark. He did, however, eventually secure approval and assistance from key Pakistani ofªcials, including Prime Minister Liaquat Ali Khan. Thus, despite its secrecy, the operation had the imprimatur of Pakistan’s political leadership.^[17]

Internal uprisings succeeded in loosening the maharaja’s grip on power during the ªrst weeks of October 1947.^[18] 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, leading a panicked Hari Singh to appeal for Indian military assistance in repelling the intruders. Although the tribesmen plundered extensively, they were not driven exclusively by the quest for booty; religion also played an important role in motivating them. The militants consistently referred to their attack on Kashmir as “jihad,”^[19] and stated that they were liberating the territory from Hindus for their Muslim brethren.^[20]

Indian forces, airlifted into Kashmir in response to the maharaja’s pleas, soon intercepted the intruders.^[21] Pakistan supported the militants against the Indians, but kept its assistance unofªcial until the spring of 1948, when the army formally took charge of the war effort.^[22] The conºict ultimately ground to a stalemate, ending with a United Nations–sponsored cease-ªre on January 1, 1949. The cease-ªre left one-third of the territory in Pakistani hands and two-thirds under Indian control.^[23]

Despite this inconclusive ending, the ªrst Kashmir war had two important 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 militants therefore constituted Pakistan’s only realistic means of attempting to revise territorial boundaries in Kashmir. Second, the war enhanced Kashmir’s importance to Pakistan; it extended the dispute well past the time of partition and transformed it into a contest of national resolve with India.^[24] Thus, even after the war had ofªcially ended, Pakistani military leaders sought to continue their militant campaign in Kashmir.

the 1965 kashimr war: an enhanced role for militancy

Colonel Khan, who had masterminded Pakistan’s use of tribal forces in 1947 , wrote detailed plans for the Pakistani government to arm and train a Kashmiri “people’s militia” following the ªrst Indo-Pakistani war. These plans resulted in numerous attacks in Jammu and Kashmir during the 1950s.^[25] Thus, by the time of the second Indo-Pakistani conºict over Kashmir in 1965, Pakistani leaders were thoroughly accustomed to employing militants in pursuit of their strategic objectives.^[26] Indeed, the initial phase of Pakistan’s 1965 war plan, code-named Operation Gibraltar, relied almost entirely on irregular forces, and it did so in a manner far more sophisticated than Pakistan’s use of militants in 1947.

For example, the militants in Operation Gibraltar were much better prepared and organized than the motley array of tribesmen that Pakistan had hastily deployed during the ªrst Kashmir war. Drawn from Razakar and Mujahid militias raised in Pakistani Kashmir, the roughly 30,000 ªghters received six weeks of training in guerrilla warfare techniques. They came under the overall command of the Pakistan Army’s 12th Division. The inªltrators were divided into eight forces, each of which targeted a speciªc area of Indian Kashmir.^[27] In addition to this superior training and organization, Pakistan’s plans were better grounded in Islamic religious tropes in 1965 than they had been during the 1947 war. The operation’s name commemorated Gibraltar’s eighth-century conquest by the Muslim general Tariq bin Ziyad, and many of the invasion units were named after prominent Muslim military heroes such as Salahuddin, Khalid, and Babar. Thus, from its inception, the operation was explicitly religious in tone.^[28]

The Pakistan Army began moving Gibraltar forces into Kashmir in early August 1965, in hopes of stoking anti-Indian sentiment and igniting a rebellion. Taking advantage of the disturbed conditions, the regular Pakistan Army would then launch Operation Grand Slam, a conventional invasion designed to seize the state.³⁷ Vigilant local residents, however, noticed the Pakistani inªltration. They alerted government authorities, who promptly sealed the Line of Control. As a result, Pakistan lost the element of surprise and was unable to move all of its militant forces into Kashmir. Those ªghters who did manage to inªltrate failed to spark the hoped-for uprising.³⁸

Despite these setbacks, Pakistan proceeded with the second phase of its war plan, launching a full-scale, conventional attack against India on September 1 , 1965. A number of inconclusive battles followed, and the war soon bogged down in a stalemate. India and Pakistan subsequently accepted a UN ceaseªre resolution, and by the third week of September, the war was over. Under the postwar settlement, known as the Tashkent agreement, the adversaries agreed to return to the status quo ante.³⁹

Like the ªrst Kashmir war, then, the 1965 conºict did little to change regional borders. It was not, however, a complete failure for Pakistan. Adherence to a militant strategy had again paid dividends, enabling the Pakistanis to inºict signiªcant costs on the Indians in Kashmir, and to bring international attention to the Kashmir dispute, without suffering undue losses. The 1965 conºict thus continued the trend that the 1947 war had begun, reinforcing for Pakistan the notion that irregular forces offered a promising means of eventually securing victory in its ongoing struggle with India for control of Kashmir.

the bangladesh war: an exception reinforces the rule

The next Indo-Pakistani conºict, which occurred a mere six years after the 1965 war, was primarily a conventional military confrontation.⁴⁰ The war’s cata-

www.rediff.com/cms/print.jsp?docpath!/news/2005/sep/19war.htm; and Nawaz, Crossed Swords, p. 206.

37.                 See Sumit Ganguly, “Deterrence Failure Revisited: The Indo-Pakistani Conºict of 1965,” Journal of Strategic Studies, Vol. 13, No. 4 (December 1990), pp. 77–93; and Russell Brines, The IndoPakistani Conºict (New York: Pall Mall, 1968), pp. 301–302.

38.                 See Sumit Ganguly, “Explaining the Kashmir Insurgency: Political Mobilization and Institutional Decay,” International Security, Vol. 21, No. 2 (Fall 1996), pp. 76–107.

39.                 Edgar O’Ballance, “The India-Pakistan Campaign, 1965,” Royal United Services Institution Journal, Vol. 111, No. 644 (November 1966), pp. 330–335.

40.                 Both India and Pakistan did make some use of nonstate actors during the Bangladesh conºict. India supported mukti bahini insurgents ªghting the East Pakistani government while Pakistan supported a number of opposing militant groups, such as the al-Shams and al-Badr brigades. See Richard Sisson and Leo E. Rose, War and Secession: Pakistan, India, and the Creation of Bangladesh (Berkeley: University of California Press, 1990), pp. 181–186, 210–213; Haqqani, Pakistan, pp. 78 –80;

strophic outcome, however, exacerbated the material and political weakness that had originally given rise to Pakistan’s militant policy, and demonstrated the danger of engaging India directly in combat. Thus the Bangladesh war actually increased Pakistan’s future reliance on Islamist militancy as a strategic asset.

Large-scale rioting erupted in East Pakistan in the wake of the ªrst Pakistani national election in October 1970. Millions of refugees began ºowing into India in the spring of 1971 when West Pakistani troops, charged with quelling the uprising, massacred their Bengali countrymen. Unable either to absorb the refugee ºow or to end the crisis by diplomatic means, Indian leaders decided to split East and West Pakistan. Following preemptive Pakistani strikes against Indian air bases on December 3, India attacked East Pakistan with six army divisions. The Indians quickly took Dhaka, and by December 16, Pakistani forces had surrendered.⁴¹

The Bangladesh conºict heightened Pakistan’s long-standing insecurities. At the material level, the war’s outcome was nothing short of disastrous. In the space of approximately two weeks, the Indians had vivisected Pakistan, creating the new state of Bangladesh. In addition, they had seized approximately 5,000 square kilometers of territory and taken roughly 90,000 Pakistani prisoners. This left Pakistan even more vulnerable to India than it had been previously.⁴²

At the ideological level, the Bangladesh conºict badly undermined the twonation theory that had justiªed Pakistan’s founding. A common Muslim identity had failed to unite Pakistanis in the face of ethnolinguistic differences.⁴³ Pakistani leaders believed that, to compensate for this failure, religion would have to play an increased role in Pakistan’s state-building project. Under the leadership of Gen. Muhammad Zia-ul-Haq, Pakistan therefore pursued an Islamization process that included the creation of sharia courts; the appointment of a Muslim consultative assembly; the implementation of charity taxes; the introduction of punishments based on the Koran and sunnah; and the ex-

and Ishtiaq Hossain and Noore Siddiquee, “Islam in Bangladesh Politics: The Role of Ghulam Azam of Jamaat-I-Islami,” Inter-Asia Cultural Studies, Vol. 5, No. 3 (December 2004), pp. 384–399.

41.  See Ganguly, Conºict Unending, pp. 67–69.

42.  Kapur, Dangerous Deterrent; Shahid Amin, Pakistan’s Foreign Policy: A Reappraisal ( Karachi : Oxford University Press, 2002), p. 72; and Hasan Askari-Rizvi, “Pakistan’s Defense Policy,” in Mehrunnisa Ali, ed., Readings in Pakistan Foreign Policy, 1971–1998 (Karachi: Oxford University Press, 2001), p. 211. See also The Hamoodur Rehman Commission of Inquiry into the 1971 War, report as declassiªed by the government of Pakistan (Lahore: Vanguard, 2000).

43.  Ganguly, Conºict Unending, pp. 71–72; and S.M. Burke and Lawrence Ziring, Pakistan’s Foreign Policy: An Historical Analysis (Karachi: Oxford University Press, 1990), pp. 420–421.

pansion of the madrassa system of Islamic education. Schoolbooks were edited to conform to religious teachings, and civil servants were required to perform daily prayers and were graded on religious knowledge and piety.^[29]

The Pakistan Army became a primary object of these Islamizing efforts.^[30]General Zia believed that “[t]he professional soldier in a Muslim army, pursuing the goals of a Muslim state, CANNOT become ‘professional’ if in all his activities he does not take on ‘the colour of Allah.’”^[31] Islamic teaching thus assumed an important role in military education. The army’s Command and Staff College, for example, established a directorate of religious instruction. Ofªcers’ piety became a factor in their career prospects. Islamic teachings were included in promotion exams, and ofªcers deemed insufªciently religious often failed to advance.^[32]

These military and domestic political developments increased Pakistan’s reliance on militancy. The salience of the Kashmir dispute, which was rooted in Pakistan’s identity as a Muslim state and its opposition to “Hindu” India, grew further as a result of Pakistan’s post-Bangladesh Islamization process. The military strategy that could enable Pakistan to seize Kashmir without directly confronting India and risking a catastrophic, Bangladesh-scale defeat, consequently assumed even greater importance. 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.”^[33] Pakistan’s departure from its militant strategy during the Bangladesh war thus reinforced the strategy’s underlying rationale and greatly increased its importance to Pakistani security policy.

the afghan war: setting the stage for insurgency in kashmir

As the Pakistanis’ militant campaign to acquire Kashmir became more important in the wake of the Bangladesh war, their ability to prosecute it increased. The reason was yet another turn of events seemingly unrelated to Kashmir— the Soviet Union’s 1979 invasion of Afghanistan. The Pakistanis had longstanding interests in Afghanistan, such as formalizing the two countries’ international border, acquiring strategic depth, gaining access to Central Asia, and minimizing India’s regional presence.^[34] 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-Afghanistan border, and coercive trade measures aimed at Kabul. The Pakistanis had also employed a militant campaign against Afghanistan, training approximately 5,000 Afghan Islamists at camps within Pakistan from 1973 to 1977 and then sending them home to destabilize the government.^[35]

Signiªcantly, the Pakistanis viewed the Afghan war as an opportunity to pursue their interests not just in that country, but also in Kashmir. General Zia believed that, if Pakistan could serve as the primary conduit for international aid to the Afghan resistance, it could inºate the cost of its support for the war effort and then divert the proªts from U.S. 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 conºict 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.”^[36]

To this end, the Pakistanis ensured that all assistance to the Afghan resistance ºowed through their Inter Services Intelligence (ISI) agency. By 1987 , approximately 20,000 ªghters per year were receiving training at seven Pakistani camps, with more being trained in Afghanistan. Pakistan also funneled tens of thousands of tons of arms and munitions per year to the Afghan mujahideen. In return, the Pakistanis extracted as much money and matériel as possible from the United States and other sponsors such as Saudi Arabia. General Zia eventually secured from the United States a package of $3.2 billion, a deal to purchase F-16 ªghter aircraft, and promises to reduce pressure on Pakistan over its nuclear program and human rights record.^[37]

Even as they supported U.S. efforts in Afghanistan, then, the Pakistanis were setting the stage for a new, expanded Kashmir campaign. Pakistan’s newfound resources enabled them to employ militants in the territory even more extensively than they had previously. In the mid-1980s, the Pakistanis struck an agreement with the Jammu and Kashmir Liberation Front (JKLF) to provide military and ªnancial support for the Front to ªght 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 insurgents 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.”^[38] 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 Front with military and ªnancial backing. Meanwhile, the JKLF established roughly 300 sleeper cells in Indian Kashmir and waited for instructions from Front leaders in Pakistan.^[39]

By the time Soviet troops began withdrawing from Afghanistan in 1988, the Pakistanis had thus signiªcantly bolstered their ability to ªght a Kashmiri proxy war. Moreover, just as Zia had predicted, a preoccupied United States turned a blind eye toward their activities. As long as the Pakistanis were helping to oust the Soviets from Afghanistan, U.S. leaders were not concerned with such peripheral matters as Kashmir.^[40] As we explain below, the Pakistanis lost little time in exploiting this windfall.

the kashmir insurgency

In 1988, a violent anti-Indian rebellion erupted in Jammu and Kashmir. The Pakistanis did not cause the discontent underlying the uprising,^[41] but they played a crucial role in transforming spontaneous, decentralized opposition to Indian rule into a full-ºedged insurgency. During 1988, the ISI had repeatedly pushed the JKLF to launch attacks in Kashmir, at one point threatening to betray Front leaders to the Indian security services if it did not act soon. The Front had been delaying the start of its campaign so that it could carry out more extensive preparations. Under Pakistani pressure, however, its leaders ªnally gave in and orchestrated two bombings in Srinagar in July 1988.^[42] This was the insurgency’s opening salvo.

The Pakistanis spent the subsequent decades carefully managing the insurgency, using their military, ªnancial, and political resources to determine its character and trajectory. Their ªrst major step after facilitating the outbreak of the rebellion was to undermine the position of the Jammu and Kashmir Liberation Front. The Front was at least notionally secular and was also a genuinely nationalist organization, dedicated to achieving Kashmiri independence rather than joining Kashmir to Pakistan. Therefore a signiªcant tension existed between the JKLF’s goals and those of Pakistan. In addition, the JKLF’s cadre were not hardened ªghters, and thus did not prove to be as militarily effective as the Pakistanis had hoped.^[43]

The Pakistanis therefore decided to replace the JKLF with more useful partners. To this end, during the early 1990s Pakistan shifted its support from the Front to the Hizb-ul-Mujahideen (HUM). HUM based its agenda on that of its political patron, the Jamaat-e-Islami, which promoted Islamist domestic policies and advocated the accession of Kashmir to Pakistan. HUM used its newfound resources to wage war against the JKLF. This devastated the Front and enabled HUM to emerge as the dominant militant group in Kashmir, a status that it maintained through the rest of the 1990s.^[44] During this period, violence in Kashmir skyrocketed. For example, attacks on Indian forces increased from about 50 in 1989 to more than 3,400 in 1993, while security-force and civilian deaths jumped from roughly 90 to approximately 1,230.^[45]

The Pakistanis’ decision to support the Hizb-ul-Mujahideen thus paid handsome dividends. Nonetheless HUM, similar to the JKLF before it, ultimately proved to be less than an ideal partner for Pakistan. The group was composed primarily of Kashmiris, who had brethren in Indian Jammu and Kashmir. HUM members thus worried about the consequences of their violent activities, and in important instances, they were willing to compromise with India. In addition, they publicly encouraged Kashmiri separatist groups to resolve their differences with India in an amicable fashion. This behavior seriously undermined Pakistani leaders’ faith in the HUM. The Pakistanis therefore soon began to recruit a new class of militant groups as their partners in the Kashmir jihad. Like Hizb-ul-Mujahideen, these organizations promoted Islamist sociopolitical agendas and sought Kashmir’s accession to Pakistan. Unlike HUM, however, they consisted mainly of non-Kashmiris, who were unconcerned with the insurgency’s impact on the people of Kashmir. Thus these mehmaan mujahideen, or guest ªghters, were not amenable to any form of compromise with India and did not hesitate to employ extreme violence in pursuit of their ends. One of their hallmarks was fedayeen attacks, in which heavily armed assailants launched high-risk operations against prominent military and civilian targets, often ªghting to the death. They also did not hesitate to target Muslims who ran afoul of their philosophy.^[46]

The Pakistanis quickly recognized the utility of these new organizations, which included Lashkar-e-Taiba (LeT), Jaish-e-Mohammed, al-Jehad, al-Omar, Harkat-ul-Ansar, and the Ikhwanul Muslimeen.^[47] They afforded the groups extensive ªnancial, logistical, and military support, and the results were impressive. The advent of their fedayeen tactics, for example, coincided with a rise in fatalities among Indian forces to their highest-ever level in the Kashmir insurgency. Even more important, the groups took jihad beyond Kashmir into the Indian homeland, with attacks on such locations as Delhi’s Red Fort. These organizations thus quickly assumed the role of Pakistan’s primary proxy forces. By the mid-1990s, Lashkar-e-Taiba and Jaish-e-Mohammed had emerged as the two leading groups.^[48]

Although the Pakistani government has repeatedly promised to move against these organizations, it has never seriously done so. For example, following Lashkar-e-Taiba and Jaish-e-Mohammed’s implication in a 2001 attack on the Indian parliament, the Pakistani government ofªcially banned the groups. President Pervez Musharraf, meanwhile, publicly pledged to prevent Pakistani soil from being used as a launching pad for future terrorism. Nonetheless, LeT remained active and a few months later was implicated in the massacre of soldiers’ families at an Indian military base at Kaluchak in Jammu and Kashmir.^[49]

In the following years, the Indian government pinned a number of attacks within India on Lashkar-e-Taiba, charges that Pakistan either ignored or denied.^[50] Meanwhile, the group continued to operate openly within Pakistan, regularly holding large rallies at its headquarters on the outskirts of Lahore.^[51]Even after overwhelming evidence linked Lashkar-e-Taiba to the 2008 terrorist assault on Mumbai, Pakistani authorities declined to move decisively against the group.^[52] They placed Haªz Muhammad Saeed, leader of LeT’s charitable front organization, Jamaat-ud-Dawa (JuD),⁶⁸ under house arrest, but he was released for want of evidence. Soon thereafter, a Lahore court dismissed terrorism charges against Saeed.^[53] The Pakistani government has thus consistently allowed Lashkar-e-Taiba and other such groups to operate with relative impunity, ensuring that more than six decades after partition, the territorial division of India remains violently contested.

shaping post-soviet afghanistan

Despite Kashmir’s importance, it is not Pakistan’s sole regional concern; as noted above, Pakistan also has long-standing strategic interests in Afghanistan. After the Soviet Union’s defeat, the Pakistanis began working to shape Afghanistan’s postwar strategic environment. They did so in essentially the same manner as they have in Kashmir, using Islamist militants to promote their interests without subjecting Pakistani forces to the costs and risks of direct conºict.

Following the Soviet Union’s withdrawal in 1989, Afghanistan collapsed into a civil war between various warlords and ethnic factions. The Pakistanis were deeply involved in this struggle from the beginning. They initially supported a number of Afghan political parties, including the warlord Gulbuddin Hekmatyar’s Hizb-ul-Islami. After tensions among these organizations erupted into violence, the Pakistanis settled on Hekmatyar. The Pakistanis were impressed with Hekmatyar’s discipline and ruthlessness. They also believed that, as an ethnic Pashtun and the product of Pakistani military training during the early 1970s, he would serve as a pliant ally. Hekmatyar ultimately proved to be militarily ineffective, however, and by 1994 it had become clear that the Pakistanis would have to look elsewhere for an Afghan proxy.^[54] They decided on the Taliban.

The Pakistanis found the Taliban appealing for a number of reasons. First, Benazir Bhutto’s government already had a relationship with the group. In 1993, the political party Jamaat-e-Ulema-Islam (JUI) had aligned with Bhutto’s Pakistan People’s Party (PPP). The JUI then developed a close association with Gen. Naseerullah Babar, Bhutto’s minister of the interior. Also, one of the Jamaat’s principal leaders, Maulana Fazlur Rehman, 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 network of madrassas, or religious schools, in Pashtun areas of the North West Frontier Province, Federally Administered Tribal Areas, and Baluchistan. The Taliban, which literally means “students,” emerged largely from these Pakistani madrassas in 1994. Thus, from their inception, the Taliban enjoyed close ties to the Pakistani government. In addition to these connections, the Pakistanis were attracted to the Taliban’s success on the battleªeld. 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. It was becoming clear that the Taliban, who similar to Hekmatyar’s forces were predominantly Pashtun, would be both a potent military force and a potential long-term ally. General Babar now referred to them as “our boys.”^[55]

As with Kashmiri militant groups, the Pakistanis gave the Taliban extensive assistance. For example, General Babar created an 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 electricity in Kandahar; Frontier Corps paramilitary forces helped to construct 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 ªre in conjunction with Taliban operations. Pakistani efforts intensiªed after the Taliban captured Herat in late 1995, giving them control of western Afghanistan and demonstrating that they could operate effectively beyond the Pashtun South. This assistance was essential to the Taliban’s success; their ascendance to power, which was complete by 2000 , could not have occurred without Pakistani backing.^[56]

Despite their debt to Pakistan, the Taliban exhibited considerable independence upon taking power, diverging from Islamabad’s preferences in important areas. For example, they refused to formalize the Afghanistan-Pakistan border in accordance with Pakistani wishes. Nonetheless, the Taliban cooperated with the Pakistanis on a number of important matters, including the establishment of militant training camps in Afghanistan, and they were indebted to Pakistan for facilitating their rise to power. The Pakistanis therefore viewed a Talibanled Afghanistan as a friendly state on their western ºank, providing them, in Larry Goodson’s words, with nothing less than a “proxy army in

Afghanistan.”^[57] Whatever the Taliban’s problems, then, the Pakistanis considered them a stabilizing force and deserving of ongoing assistance. As a senior Pakistani diplomat put it, “We will support whoever can bring stability to Afghanistan. If they are angels, nothing like it. And if they are devils, we don’t mind.”^[58]

Under strong pressure from the United States, Pakistan joined the U.S.-led coalition to oust the Taliban from Afghanistan after the September 11 terrorist attacks. The Pakistanis did not wholly abandon their erstwhile allies, however. For despite assuming a central role in the global war on terror, Pakistan had not altered its Afghan agenda; as before, the Pakistanis desired a friendly regime in Kabul to facilitate their acquisition of strategic depth, access to Central Asia, training of militants, exclusion of India from the area, and favorable resolution of border issues. The Pakistanis continued to believe that the Taliban ªt this description better than any of Afghanistan’s other contenders for power.

The Pakistanis therefore worked to prevent the United States from eradicating the Taliban in the wake of the September 11 attacks, airlifting trapped Taliban forces, Pakistani military and security personnel, and other ªghters out of Kunduz in January 2002.^[59] In the following years, they provided the Taliban and associated Afghan militant groups such as the Haqqani network with extensive ªnancial, logistical, and intelligence support, while accepting billions of dollars in U.S. aid to de-Talibanize the country.^[60] Pakistani assistance to these organizations was so extensive that Joint Chiefs of Staff Chairman Adm. Michael Mullen publicly characterized them as “proxies of the government of Pakistan” that were “hampering efforts to improve security in Afghanistan” and “spoiling possibilities for broader reconciliation.”^[61] Pakistan’s Afghan strategy has thus closely mirrored its policy in Kashmir, using militant proxies to promote its security interests while avoiding the dangers of direct military intervention.

Evaluating Pakistan’s Militant Strategy

From the 1947 Kashmir conºict to the current war in Afghanistan, Islamist militants have constituted one of the primary means by which Pakistan has sought to promote its security interests. How successful has Pakistan’s use of militancy actually been? Has the strategy made Pakistan more secure? As noted above, most analysts believe that Pakistan’s militant strategy has been disastrous, badly damaging its security interests. Although there is a good deal of truth in this view, the overall results of Pakistani policy have been more complicated. Pakistan’s use of Islamist militants has in fact yielded a number of signiªcant beneªts.

First, Pakistan’s militant strategy has helped to ensure that, despite its lack of a coherent founding narrative, the country has had a plausible reason to exist. This reason has been primarily negative, based on opposition to so-called Hindu India, rather than on a positive vision of what Pakistan might create or become, but it has been important nonetheless. Pakistanis’ sense of participation in a larger common project has been real and widespread, giving rise to what Christophe Jaffrelot describes as a “strong nationalism.” This has helped Pakistan to endure the host of severe ideological and material challenges that it has faced since independence; in the face of war and crisis, Pakistanis have demonstrated a high degree of loyalty to the state.^[62] Without it, Pakistan may not have been able to survive.^[63]

Second, Pakistan’s militant strategy has helped to compensate for its material imbalance vis-à-vis India. To be sure, Pakistan remains the weaker party in the Indo-Pakistani relationship, but Pakistan’s strategy has taken a signiªcant toll on India over the decades. 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 6,000 security force personnel in Kashmir,^[64] about twice the number of battle deaths as India suffered in either the ªrst or second Kashmir war, or in the Bangladesh conºict.^[65] India also lost 15,000 civilians during this period in Jammu and Kashmir, far exceeding civilian losses in any of its other conºicts.^[66] In addition, the insurgency has inºicted major opportunity costs on India. For example, the Indians currently deploy approximately 400,000 personnel to maintain security in Kashmir.^[67]This presence requires not only manpower, but also large amounts of money that New Delhi could otherwise spend on domestic development. Finally, the insurgency has tarnished India’s image. The tactics that New Delhi has employed to combat the rebels, including kidnapping, torture, and extrajudicial killings, have undermined India’s standing as a democratic state respectful of human rights and the rule of law.^[68]

Third, Pakistan’s militant strategy has enabled it to pose an ongoing challenge to Indian control of Kashmir. Although Pakistan has not succeeded in wresting Kashmir from India, it has managed to ensure that the territory remains contested, both militarily and diplomatically. At the military level, India has continually been forced to ªght to retain control of the region. At the diplomatic level, the Kashmir dispute remains on the agenda of the international community,⁸⁵ despite India’s claim that it is a solely bilateral issue. Both of these accomplishments leave open the possibility, however small, that someday Pakistan will succeed in its quest to alter territorial boundaries in Kashmir.

Finally, in Afghanistan, Pakistan’s strategy facilitated a friendly Taliban government’s ascent to power. In addition it has helped to ensure that, even well after their ouster, the Taliban remain a viable entity, continuing to battle coalition forces for control of the country.⁸⁶ Meanwhile, Pakistan has received large amounts of U.S. ªnancial and military aid in return for its participation in the anti-Taliban campaign. Thus in Afghanistan, as in Kashmir, Pakistan’s strategy has provided Pakistan with important near-term beneªts and has ensured that a long-term outcome congenial to Pakistani interests remains a possibility.

Pakistan’s militant strategy has thus not been an abject failure. To the contrary, it has successfully promoted Pakistani interests on a number of important fronts. Nonetheless, despite its past utility, Pakistan’s strategy has recently given rise to several extremely dangerous developments. First, the militant organizations that Pakistan once controlled have increasingly slipped its grasp. After decades of ªnancial and military support, these groups are sufªciently strong that they no longer necessarily bend to Pakistan’s will. Indeed, the militants can now pursue their own agendas regardless of Pakistani wishes. The results of this change have damaged Pakistani security on two distinct levels.

First, militant groups have begun to challenge the central government for sovereignty over Pakistani territory. For example, groups such as the Tehrik-eTaliban (TTP) have seized control of large sections of South Waziristan. There they have repudiated Islamabad’s writ, imposed an extreme interpretation of sharia law, and participated in attacks on government and coalition targets in

Conºict, Vol. 8, No. 4 (May 1996). See also Wirsing, India, Pakistan, and the Kashmir Dispute, pp. 154 – 162; and Schoªeld, Kashmir in Conºict, pp. 168–172.

85.                 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:// www.un.org/en/peacekeeping/missions/unmogip/; and Chidanand Rajghatta, “Obama Mulls Clinton as Special Envoy on Kashmir,” Times of India, November 7, 2008.

86.                 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. Gen. David Petraeus characterized coalition progress against the Taliban as “fragile and reversible.” Statement of Gen. David

H. Petraeus before the United States House Armed Services Committee on Afghanistan, March 16 , 2011.

Afghanistan.^[69] These actions have forced the Pakistani government to undertake extensive military operations to rout the insurgents. The ensuing bloodshed has imposed signiªcant human and ªnancial costs on the Pakistan government, and has alienated local populations, thereby compounding the difªculty of Pakistani counterinsurgency efforts.^[70]

Second, militant organizations have refused to subordinate their interests to Pakistan’s broader strategic imperatives. For example, in the aftermath of events such as September 11 and a December 2001 terrorist attack on the Indian parliament, Pakistan was forced to limit its support for militancy, in some cases even outlawing jihadi groups. The militants viewed this shift not as a necessary response to changed international circumstances, but rather as an act of treason. Instead of tempering their behavior, they launched a violent retaliatory campaign against the Pakistani government, which included attempts to kill Musharraf and other leaders, as well as attacks on army, intelligence, and police installations.^[71]

The militants have also 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. As LeT ofªcial Nasr Javed declared in a 2008 speech, “Jihad will spread from Kashmir to other parts of India. The Muslims will be ruling India again.”^[72] Signiªcantly, the group’s ambitions do not end on the subcontinent; Lashkar-e-Taiba views its South Asian struggle as only part of a larger, global jihad. In the words of an LeT manifesto, “Until Islam prevails throughout the world and Allah’s law applies to everyone it is our duty to keep ªghting against the inªdels.”^[73] The group is unconcerned with the impact of such a maximalist agenda on Pakistani interests, and is 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.”^[74]

The second major problem with Pakistan’s militant strategy is the opportunity cost that it entails. Continual support for jihad diverts scarce national resources from other critical projects, impeding Pakistani internal development. Pakistan’s education sector offers one of the most urgent examples of this problem. In 2007, only 62 percent of Pakistani primary school–aged children, and 30 percent of secondary school–aged children, are actually enrolled in school.^[75] Yet Pakistan allocated 22 percent of federal spending, amounting to 3.8 percent of gross domestic product (GDP), to the military in 2006. Twelve percent of the federal budget, or 2.21 percent of GDP, by contrast, went to education. Cutting approximately one percentage point of GDP from the military budget would enable Pakistan to increase education spending by 55 percent.^[76]

Of course, Pakistan’s asymmetric warfare campaign does not account for all, or even most, of its military spending. But even if the proportion is only modest,^[77] its impact is signiªcant; as noted above, small cuts in defense outlays would substantially increase Pakistan’s ability to pursue domestic developmental goals. More generally, Pakistan’s asymmetric warfare campaign, by continually provoking India, helps to create an extremely hostile and demanding strategic environment on the subcontinent. It is this environment, in turn, that forces Pakistan to devote such a high level of resources to ensuring its external security. In a more permissive strategic context, the Pakistanis could safely lower their overall defense budget. 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 danger of Pakistan’s strategy is that it is provoking the Indians to develop new military capabilities, which will enable them to undertake largescale 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 terrorists.^[78]

To remedy this problem, the Indians are increasing the offensive capabilities of their forces along the international border, which previously played a primarily defensive role. And unlike their earlier plans to drive deep into Pakistan along a small number of axes, they are preparing to rapidly launch numerous shallow attacks along a broad swath of Pakistani territory. The Indians expect that this new approach, often referred to as Cold Start, will enable them to strike Pakistan within days of a future provocation, before the Pakistanis can prepare their defenses or the international community can dissuade India from taking offensive action.^[79] Pakistan’s asymmetric warfare strategy, by driving aggressive Indian military innovation, thus threatens to trigger the very disaster that it was designed to avoid: a direct Indo-Pakistani military confrontation. Signiªcantly, 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.^[80] The danger of such escalation is likely to increase with Pakistan’s development of a tactical nuclear weapons capacity, which is designed to lower Pakistan’s nuclear threshold and enhance the credibility of its nuclear threats in the face of Indian conventional advances.^[81]

Despite its historical utility, then, Pakistan’s militant strategy has given rise to problems that severely damage Pakistani interests. Signiªcantly, these problems are largely attributable to a single factor: 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 strategic use of militants attractive to Pakistan in the ªrst place; Pakistan adopted its militant strategy in response to its own material and political shortcomings. Pakistan is thus caught in a jihad paradox: the very characteristic of the Pakistani state that makes a policy of supporting Islamist militancy useful also makes it extremely dangerous.

How can the problems resulting from Pakistan’s jihad paradox be mitigated? In the article’s concluding section, we 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 operating within its territory. Signiªcantly, the burden of such a shift need not fall wholly on Pakistan. India could cooperate as well, delaying some of its planned military initiatives so that Pakistan can concentrate less on defending itself against external threats and more on pursuing the critical goal of internal stability.

Conclusion

Since its founding in 1947, Pakistan has used Islamist militants as a central tool of its grand strategy. Although often derided as a shortsighted failure, this policy has achieved important domestic and international successes. Paradoxically, however, the weakness of the Pakistani state has now made support for jihad as dangerous to Pakistan as it once was useful.

Addressing the problems associated with Pakistan’s jihad paradox will be extremely difªcult, though not impossible. Doing so will require decisive action on the part of both Pakistan and India that transcends the boundaries of each state’s traditional strategic behavior. Pakistani leaders must recognize that although asymmetric warfare paid dividends in the past, that era is over; today the costs of supporting jihad far exceed its advantages. Therefore, Pakistan must fully cease its support for all militant organizations and take forceful action to crush those operating in its territory.

This applies not just to “bad” militants such as the Pakistani Taliban battling government forces in South Waziristan; it applies to what Pakistan has considered to be “good” militants, such as the Afghan Taliban, as well. The activities of “good” militants bleed into Pakistan proper, with the Afghan Taliban facilitating the operations of their Pakistani brethren.^[82] Even if the Pakistanis succeed in routing groups such as the Pakistani Taliban from their own country, the militants will soon return if they are allowed to ºourish in Afghanistan. Pakistan cannot have it both ways. It must entirely cease its support for militant groups, regardless of their location or political agenda.

Abandoning jihad will not be easy. The Pakistani security establishment has been following its militant strategy for more than six decades. Overcoming the resulting bureaucratic ideology, interests, and inertia to fundamentally change Pakistani policy will pose extreme challenges. In addition, the Pakistanis would be hard pressed to defeat the militants even if they genuinely wanted to do so; the jihadis have proven adept at battling Pakistani security forces, which are ill suited for counterinsurgency operations.^[83] Regardless of these difªculties, the Pakistanis have little choice but to try to change course. The costs of failing to do so are likely to be disastrous.

India, for its part, must stop viewing Pakistan’s militant strategy primarily as a compellence problem. Even the most sophisticated plans for coercing Pakistan into abandoning its strategy will be of limited utility given Islamabad’s lack of full control over the militant organizations. Furthermore, many of the coercive tools that India is acquiring, such as enhanced conventional military capabilities, are inappropriate to the task. They cannot prevent further terrorist attacks in Kashmir or in India proper, and by threatening Pakistan, they will create incentives for it to rely more heavily on both militant proxies and nuclear weapons to defend itself. At the very least, India’s coercive efforts will divert the Pakistanis’ attention outward, and away from the urgent internal tasks of defeating the militants and promoting internal stability.^[84]

The Indians could instead consider temporarily reducing Pakistan’s external security concerns, affording Islamabad the opportunity to turn its attention inward. This would entail delaying initiatives such as Cold Start, which will impede Pakistani efforts to focus on domestic problems. Such restraint would be difªcult to muster, given the severe costs that Pakistan’s militant campaign has inºicted on India. In addition, it could prove to be ineffective; the Pakistanis might misread Indian forbearance as evidence that their militant policy in fact was working. And, as noted above, the Pakistanis would be hard pressed to defeat the jihadis even if they wanted to do so. Indian restraint could, however, help to create a critical opportunity for Pakistan to abandon militancy, an outcome that would signiªcantly improve Indian and regional security. In the end, if a temporary policy of restraint appeared to be ineffective, the Indians could revert to their earlier military plans.

Regardless of its external security posture, India must do better at protecting itself internally against militant attacks. This will require thoroughly overhauling its approach to domestic security, which has been severely wanting. For example, even though Indian authorities had credible warnings of an impending seaborne operation against Mumbai, the November 2008 attacks took them completely off guard. And once the assault began, Indian forces required three days to rout a small group of terrorists. Substantive improvements in Indian border and maritime security, critical infrastructure, interagency intelligence sharing, and federal and state law police capabilities appear to be under way, and must continue.^[85]

Finally, the Indian government should take steps to address the legitimate concerns of its Muslim population, both in Kashmir and in India proper. In Kashmir, the Indians must go well beyond vague promises of greater state autonomy to address long-standing grievances regarding large-scale troop deployments, routine harassment of citizens, extra-judicial killings, and the use of torture against suspected insurgents. In the rest of India, the government must ensure that Muslims enjoy rights that other citizens take for granted, such as nondiscrimination in housing and employment. Although many Muslims have achieved economic success and attained prominent positions in Indian society, large portions of the population still face discrimination. This could be rectiªed through steps such as more aggressive recruiting of Muslims into the elite Indian Administrative Service, ensuring greater representation in the police and military, and facilitating better access to higher education.^[86]Such measures would reduce the incentives for outsiders to launch further anti-Indian terrorism and lower the likelihood that foreign militants will ªnd collaborators inside India.¹⁰⁵

More generally, the Pakistani case shows that nonstate actors can potentially serve as effective strategic tools, enabling weak states to pursue goals that might otherwise be well out of reach, such as fostering domestic political cohesion; inºicting serious military, economic, and reputational costs on stronger adversaries; challenging existing territorial arrangements; and shaping regional strategic environments. In doing so, nonstate actors can assume a major role in a state’s security policy. Such an outcome may result not from strategic “myopia,” but rather from a weak state’s rational assessment of its goals and capabilities.

The Pakistani case also suggests, however, that the type of state for which a nonstate actor strategy is likely to appear useful may also ªnd such a strategy to be particularly problematic. Nonstate actors are helpful military tools for states that lack the resources to pursue their goals through more conventional means.^[87] This weakness, however, will also make such states especially prone to losing control of their proxies, suffering damaging opportunity costs, and reducing their security by provoking stronger adversaries, just as Pakistan has been doing. Ironically, then, state weakness makes the strategic use of nonstate actors simultaneously attractive and dangerous. Other states contemplating similar policies should take the example of Pakistan’s jihad paradox as a sobering lesson.