Magnum Photos

Srinagar, Kashmir, 1948; photograph by Henri Cartier-Bresson

In late October 2008, on the eve of the election that would elevate him to the White House, Barack Obama made some of the most expansive comments about the Kashmir conflict that have ever come from an American presidential candidate. In an interview with Joe Klein of Time magazine, Obama acknowledged that Kashmir’s disputed territory was “obviously a potential tar pit diplomatically,” and yet, he continued:

For us to devote serious diplomatic resources to get a special envoy in there, to figure out a plausible approach, and essentially make the argument to the Indians, you guys are on the brink of being an economic superpower—why do you want to keep on messing with this? To make the argument to the Pakistanis, look at India and what they are doing—why do you want to keep on being bogged down with this particular [issue] at a time when the biggest threat now is coming from the Afghan border? I think there is a moment when potentially we could get their attention. It won’t be easy, but it’s important.

It was refreshing to hear an American politician speak honestly and seriously about Kashmir. Since 1989, when a popular rebellion erupted against Indian misrule, Kashmir’s violence has often been enshrouded by silence. Partly that is because neither the Indian nor the Pakistani government wishes to call attention to its contributions to the conflict. Pakistan’s intelligence service has stoked a low-intensity guerrilla war by funding and arming Islamic radicals and infiltrating them into the Kashmir Valley. India has responded with a brutal counterinsurgency campaign which at its height involved the systematic use of torture and extrajudicial killing. That campaign has lately eased, and the human rights performance of India’s government has improved, but not enough. Protests set off this summer by the shooting deaths of unarmed demonstrators have been the most intense in several years. By mid-August, at least fifty-five people had been killed. Kashmiri rioters have provoked and committed violence this summer, but much of the blame for the high death toll rests with the unprofessional performance of the Indian paramilitaries, whose approach to riot control too often involves indiscriminate firing at crowds.

At least 45,000 people have died violently in Kashmir since 1989. Local human rights groups continue to discover hundreds of unmarked graves containing the bodies of young men either shot in battle or murdered in Indian custody. The victims have left behind 30,000 orphans, according to the International Crisis Group, as well as thousands of widows and at least one thousand “half-widows,” whose husbands are among the missing but have not been proven dead.1

The conflict has again and again spilled outside Kashmir. Just a month after Obama’s interview, on November 26, 2008, ten armed men came ashore in dinghies on Mumbai’s waterfront and embarked on a three-day made-for-television killing spree that left 175 people dead, including nine of the gunmen. Evidence presented in Indian and Pakistani courts has made clear that the attackers had ties to Lashkar-e-Taiba, the largest of the Islamic groups that Pakistan’s army and intelligence service had built up to fight in Kashmir. The Mumbai attack demonstrated that the most radical jihadi groups active in the Kashmir conflict are becoming bolder. Some of them may intend to provoke an Indo-Pakistani war.

Nonetheless, President Obama has never acted on the analysis he mentioned as a candidate. Early on, his administration caved in when India objected to an exploratory plan to include Kashmir among the responsibilities of Richard Holbrooke, the President’s special representative for Afghanistan and Pakistan. Since then the Obama administration has worked quietly to encourage peace talks between India and Pakistan, but without any breakthroughs. As the months have passed, the President’s national security team has become preoccupied by the deteriorating war in Afghanistan and the Taliban-led insurgency in Pakistan. Amid such difficulties, the idea of thinking afresh about a conflict as old and complex as Kashmir’s might seem preposterously ambitious.

But it is essential nonetheless. Mumbai was a warning. American policy has long sought to compartmentalize Kashmir as a problem separate from Afghanistan’s war, the threats posed by al-Qaeda, or Pakistan’s internal violence. That policy is no longer consistent with the facts, and this failure directly threatens American security. In a number of recent cases when radicalized Muslims living in the United States have traveled to Pakistan for training or inspiration, they have connected with groups or networks active in Kashmir. American policy is also outdated with respect to conditions within Kashmir itself. Kashmiris continue to challenge India’s oppressive military presence in the region, and yet overall, the guerrilla war in Kashmir has changed and quieted during the past decade, and new possibilities for a permanent negotiated settlement have emerged.


From its beginnings, the Kashmir conflict has presented a confounding blend of overt diplomatic disputes and covert war. Jammu and Kashmir was one of several notionally sovereign princely states aligned with the collapsing British Raj that had yet to choose its future at midnight on August 15, 1947, when India and Pakistan were born as independent nations. Maharajah Hari Singh, a Hindu, presided over a Muslim-majority population as well as privileged Hindu classes and a small number of Buddhists. As Singh dithered, Pakistani army officers organized an incursion of Afghan tribesmen into the Kashmir Valley, threatening Srinagar, the summer seat of the maharajah’s throne. Singh turned to India for help; its government insisted it could do nothing until he signed an Instrument of Accession to join India, which the maharajah did, hurriedly, late in October. Pakistan has never accepted the validity of this action; India has always insisted it was proper.

It has long been known that Pakistani army officers had an important part in supporting the incursion that led to Singh’s panicked decision and thus gave birth to six decades of conflict. In Shadow War, a new history of Pakistan’s involvement in Kashmir’s violence, Arif Jamal presents a strikingly continuous narrative in which the tactics of Pakistan’s architects of secret war have changed little to the present day. The principal manager of the Partition-era infiltrations, according to Jamal, was Colonel Akbar Khan, who used the nom de guerre “General Tariq.”

Khan did not always tell his superior officers what he was doing; he diverted ammunition from regular Pakistani army elements; and with other officers, he ran the invading tribesmen’s radio operations. One of his internal memos, referring to the early Kashmiri leader Sheikh Mohammed Abdullah, was entitled “Keep the Pot Boiling in Abdullah’s Kashmir.” None of this would seem unfamiliar to Indian intelligence officers struggling to thwart Pakistani-sponsored infiltrations in recent years. The current chief minister of Indian-controlled Jammu and Kashmir is Omar Abdullah, Mohammed’s grandson; Pakistani officers are presumably still following the precepts of Khan’s memo, without need to alter its title.

The diplomatic issues framing the conflict have changed little in several decades. When the initial Kashmir war was coming to an indecisive end, a United Nations commission adopted a plan to hold a plebiscite in which Kashmiris might choose between India and Pakistan. India initially tolerated this idea but later withdrew its cooperation. Pakistan protested futilely, and in 1965 another war broke out. The territorial divisions that resulted from that conflict have persisted with minor alterations to this day. Pakistan has lately deemphasized the plebiscite, although many Kashmiris hold out hope for it. Of the diverse regions that made up Jammu and Kashmir at Partition, Pakistan today controls a mountainous, Muslim-majority territory known as the Northern Areas, as well as a section of mountains and river valleys that it has dubbed “Azad Kashmir,” or Free Kashmir. India controls the pastoral Kashmir Valley, Hindu-majority Jammu to the southeast, and tiny Ladakh.

In 1971, the two countries fought a third war, which resulted in Pakistan’s defeat and the birth of independent Bangladesh. Afterward India negotiated a victor’s peace called the Simla Agreement. That agreement formally designated the cease-fire line in Kashmir as the “Line of Control” and stipulated that all unsettled disputes would henceforth be resolved only through bilateral negotiations between India and Pakistan; India cited these terms in rejecting the Obama administration’s hints that it would be interested in promoting a settlement.

The cold war’s end saw the beginnings of the renewal of the bloodshed that has stained Kashmir in recent years.2 Indian politicians rigged a local election in 1987, which brought to a boil latent anger over corruption and discrimination against Kashmiri Muslims. The fall of the Berlin Wall and the creation of independent Central Asian states with Muslim majorities and less history of nationhood than Kashmiris felt they possessed stirred hope in the valley that a popular uprising might lead to liberation. A pro-independence guerrilla force, the Jammu and Kashmir Liberation Front (JKLF), dominated the initial revolt.

Pakistan’s principal military intelligence service, the Inter-Services Intelligence Directorate (ISI), sensed an opportunity. Flush with hubris over its successful use of jihadi militias to expel the Soviet Union from Afghanistan, the ISI soon marginalized the JKLF and promoted a pro-Pakistan Islamist force, the Hizbul Mujaheddin. An alphabet soup of other front groups and jihadi liberation armies soon joined in, many of them armed and trained by the ISI.

During the 1990s, when the Hizbul Mujaheddin faltered and some of its leaders showed signs of independent thinking, the ISI built up a more reliable proxy force, Lashkar-e-Taiba, a Saudi-influenced proselytizing organization with ample funds and thousands of recruits drawn from Pakistan’s heartland province of Punjab. The Pakistani army’s reckless self-confidence that it could seize Kashmir by violence climaxed in 1999, when it invaded Kashmir’s northern heights, near Kargil, using soldiers disguised as guerrillas. During the months-long undeclared war that followed, the United States became so concerned that it might escalate into a nuclear exchange that President Bill Clinton intervened to force Pakistan’s withdrawal.


A large part of Shadow War chronicles the ISI’s complex relationship with Hizbul Mujaheddin, a relatively obscure subject for readers outside of South Asia. Jamal’s research, which includes interviews with militants, reveals valuable and original detail but it suffers considerably from a lack of analytical perspective. The book also provides little new insight into Lash- kar or other groups that have more recently shaped events in Kashmir and beyond. Despite these limitations, Shadow War is a credible and courageous effort by a Pakistani journalist to document the unaccountable, opaque policies of his country’s intelligence service.

Jamal’s account effectively ends in 2004. Since then, the ISI has increasingly lost control of militants based in Punjab; sections of these groups, including Lashkar, Jaish-e-Mohammed, and the anti-Shia group Sipha-e-Sabha, have recently aligned with the Pakistani Taliban in a revolutionary uprising against the Pakistani state, and even against the ISI. These networks are responsible for some of the devastating suicide bombings that have taken hundreds of lives in Punjabi cities such as Lahore during the past two years. Suicide bombers have also launched unprecedented attacks against Pakistani military and sectarian targets in Azad Kashmir.

Mike King

At the same time, without much fanfare, India has largely prevailed in its counterinsurgency campaign in the Kashmir Valley, at least provisionally. More than a thousand civilians died in Kashmir in 2001; last year, the figure was seventy-two. The number of dead will climb in 2010 because of this summer’s rioting and the indiscriminate shootings by Indian paramilitaries, but even so, the death toll will remain a small fraction of what it was during the 1990s. Islamist guerrillas stage occasional attacks and have prevented tourism and economic activity from returning to normal, but the guerrillas have been isolated militarily. State elections late in 2008 drew a turnout of more than 60 percent overall, although less in the disputed heart of the valley. Even there, political participation has increased and popular support for Islamist guerrillas from Pakistan is fading.

India has drawn down some of its troop deployments and turned greater responsibility for security over to the locally recruited police. Yet not only has India’s government failed to professionalize its methods for riot control in Kashmir and elsewhere, it has left in place draconian emergency laws and has granted de facto immunity to Indian commanders for appalling human rights violations in the past. This summer’s rioting has demonstrated the persistent resentment toward India felt by many Kashmiris. Overall, conditions in the valley have improved and India has established a position of control, but these achievements are not likely to prove sustainable in the absence of a broader settlement.

Nor should the United States or its European allies become complacent about the conflict. Kashmir retains an important place in Pakistan’s sense of grievance; as an enduring cause, it is a source of radicalization and recruitment, one that offers to jihadis international legitimacy and even Pakistani state sanction. The authors of a recent International Crisis Group report describe the present danger:

Another Mumbai-like attack would have a devastating impact on bilateral relations [between India and Pakistan] and could conceivably bring the nuclear-armed neighbours to the brink of war. Moreover, the militants pose an equal threat to the Pakistani state and its citizens.

The Kashmir conundrum is comparable to the Israeli-Palestinian conflict in the sense that the outlines of a permanent two-state settlement have already been negotiated; what’s missing is the political leadership, the public confidence, and the security conditions required to conclude that settlement and to defend it against the violent reaction it would inevitably provoke from Islamist and Hindu nationalist extremists.

For all of Kashmir’s unrest, the Line of Control has provided a remarkably stable basis for a settlement. As Howard Schaffer points out in The Limits of Influence, a thorough and intelligent history of American diplomatic intervention in the Kashmir dispute, the valley, with a population of just under five million, is today “the only part of the pre-1947 state in which the majority of the population is so seriously discontented with the status quo that it wishes to break its link with the country that administers it.” Whether even the valley today is the exception Schaffer describes is debatable; Kashmiri separatist leaders continue to talk sporadically with New Delhi and Islamabad about forms of autonomy that might be established within the Indian and Pakistani constitutions.

As recently as January 2004, the former Pakistani President Pervez Musharraf and then Indian Prime Minister Atal Vajpayee agreed to launch talks that would lead to “a permanent settlement of all bilateral issues, including Jammu and Kashmir, to the satisfaction of both sides.” To assure India that he would not revert to Pakistan’s historic practice of talking peace in public and fomenting guerrilla violence in secret, Musharraf pledged that he would not “permit any territory under Pakistan’s control to support terrorism in any manner.” Since then, Indian and Pakistani leaders have made fitful, incomplete, but nonetheless remarkable progress in defining the terms of a settlement of Kashmir and other territorial disputes. Some of this negotiation has occurred at periodic high-level summits staged in public, but Indian and Pakistani envoys have carried out much of the serious work in secret back-channel talks held in hotels in Dubai, London, Bangkok, and elsewhere.3

Before he was forced from office in 2008, Musharraf seems to have concluded that Pakistan might achieve more of its goals in Kashmir through peaceful political negotiations than by continuing to infiltrate Islamic radicals. India’s surging economic growth and Pakistan’s desire to benefit in an atmosphere of normalized relations and growing trade were also factors in his willingness to seek compromise, according to Pakistani officials involved. In any event, under Musharraf, cross-border infiltration of guerrillas into Kashmir from Pakistan slowed to a virtual trickle by late 2006.

By early 2007, according to Indian and Pakistani officials involved, the two sides had completed the outline of an agreement on Kashmir. It remains unclear which provisions had been agreed on and which issues had been designated for future negotiation, but at the heart of the breakthrough lay an emphasis on autonomy and special status for residents of the valley and Azad Kashmir, including the right to move freely across the Line of Control. Before the deal could be announced, however, Musharraf’s grip on power slipped; as Islamist violence within Pakistan intensified in mid-2007, it became implausible to announce such a risk-taking peace deal, for which neither the Indian nor the Pakistani public was prepared. Musharraf has since been succeeded by a weak elected government in Pakistan as well as new army leadership that appears more focused on quelling internal revolt and managing Pakistan’s interests in Afghanistan.

The specific settlement proposals developed three years ago remain sound. Schaffer’s history traces in rich detail the roots of these ideas. As far back as 1956, President Dwight Eisenhower proposed an American role in negotiations that would divide Jammu and Kashmir, in a settlement linked to expansive aid and investments in Indian and Pakistani economic development by the United States and other wealthy countries. More recently, official envoys and nongovernmental organizations such as the Washington-based Kashmir Study Group have debated and refined ways to grant substantial autonomy to Kashmiris while preserving Indian and Pakistani sovereignty. The technical provisions of a final deal are less important than how its balance is ultimately perceived by the Indian and Pakistani publics. To be workable, an agreement must be favorable enough to Kashmiris for Pakistan’s army to save face, but not so favorable as to provoke an unmanageable backlash from Hindu activists in India.

Schaffer describes the elements of an agreement that he believes are necessary: the Line of Control should become an international border; the border should be “sufficiently porous to allow for the easy movement of people and goods across it”; all populations in the original Jammu and Kashmir state should enjoy some measure of autonomy; and “joint institutions” should be established “on an all-Kashmir basis [to] play a role in managing noncontroversial matters affecting Kashmiris on both sides of the line.”

Schaffer also adopts the traditional view that the United States should advance these ideas, if at all, only through “quiet diplomacy.” He is an accomplished retired diplomat; his profession overvalues discretion. The rationale for caution is that a more active American approach would produce an unfavorable reaction within India from Hindu nationalist politicians, the media, and even the secular Congress Party–led government, which would likely feel compelled to assert its nationalist credentials to defy perceived American meddling. Therefore, this thinking goes, the costs of American visibility on Kashmir will outweigh any benefits.

These assumptions are outdated for three reasons. First, the emerging strategic alliance between the United States and India is becoming strong enough to withstand media noise and political disagreements that earlier would have been more consequential. American presidents and negotiators helped bring peace to Northern Ireland by openly advocating positions that British leaders did not always accept, without consequence to the Anglo-American alliance. Fundamentally, it was in the shared interest of the United States, Ireland, and Britain to end Irish Republican Army terrorism, just as it is in the shared interest of India, Pakistan, and the United States today to end terrorism by Kashmir-driven groups based in Pakistan.

Second, because the outlines of a Kashmir settlement have already been identified by India and Pakistan, but would require political cover to be announced and implemented, it is essential that the United States, Europe, Japan, Middle Eastern governments, and other nations prepare to strongly support any final agreement, and to consider how to invest in its success, economically and otherwise. Only by open discussion and active advocacy for a political solution that protects the interests of Kashmiris, Pakistan, and India can such a preparatory atmosphere be created.

Finally, the interests that the United States has in the Kashmir conflict are greater now than at any time in the postwar period. American efforts to prevent a second Taliban revolution in Afghanistan and to quell Islamist rebellion within Pakistan are unlikely to succeed if ISI continues its three-decade practice of using jihadi groups to wage their own brand of war against India. The only way to gradually reduce ISI’s influence within the Pakistani establishment and to strengthen more progressive civilian leaders is to pursue a broader normalization of economic and political ties between Pakistan and India. That in turn will require a durable settlement in Kashmir. Silence and indirectness about the conflict is no longer workable.

A reconsidered American approach to Kashmir should return first of all to the tone of Obama’s Time interview: honest talk about an admittedly difficult problem. More such straight talk is now required. The United States and India share an interest in the emergence of a stable, economically successful Pakistan with an army that believes it is in Pakistan’s national interest to stop fomenting jihadi violence in Afghanistan and India. It is difficult to imagine that such a Pakistan will evolve if groups such as Lashkar are not disarmed, delegitimized, and defunded. And it is difficult to imagine that such an achievement would be possible in the absence of a political settlement that satisfies the great majority of Kashmiris and delivers economic benefits to Pakistan, such as preferential access for textiles to American markets, as well as water and energy security. President Obama and his foreign policy team should articulate this alternative to the status quo before Indian and Pakistani publics, without embarrassment.

They should also hold the ISI accountable when credible evidence appears that it is continuing to arm Islamist militias fighting in Afghanistan or Kashmir. A recent report by a Harvard University researcher who interviewed Taliban commanders in Afghanistan argues that the ISI’s training and support for Afghan Taliban guerrillas remain far more extensive than either Pakistan or the Obama administration has acknowledged.4 Pakistan has been designated a major non-NATO ally of the United States and enjoys substantial aid and access to military equipment as a result. Showering the army and the ISI with support without holding them accountable for funneling terrorists into neighboring countries will only make the problems and threats facing the United States in South Asia worse. Pakistan has often evaded accountability by keeping its support for Islamist groups hidden and by pleading, credibly, that it lacks the capacity to fully control terrorist groups operating on its soil.


Farooq Khan/EPA/Corbis

Kashmiri Muslim protesters throwing stones at Indian policemen, Srinagar, August 30, 2010

Kashmir offers a rare case where the ISI’s conduct can be measured. Pakistani forces maintain a security zone on their underpopulated side of the Line of Control that makes it all but impossible for guerrillas to cross into India-controlled Kashmir without army cooperation. According to the latest annual report of the Indian Ministry of Defense, between April 2009 and February 2010, Indian forces defeated attempted infiltrations thirty-three times, suggesting a rate of just less than one attempt per week, not including successful crossings. Even this relatively modest number should be unacceptable to the United States. If the Indian reporting of infiltration rates is inflated, then the United States, Britain, or NATO should undertake independent monitoring and announce the results. If Pakistan’s performance is judged to be unacceptable, then the Pakistani army should be held accountable and transparent reports made available to the Indian and Pakistani publics.

A similar public assessment should also be made of ISI support for the Afghan Taliban. Public exposure is not only the best disinfectant against the poison the ISI spreads in South Asia; such reporting would also advance American interests. India has justifiably been dissatisfied with the extent of Pakistan’s crackdown on Lashkar and similar groups since the attack in Mumbai. If another such attack occurs, American credibility as a source of reliable information, and as a vehicle for effective pressure on Pakistan to cease illegal support for terrorist groups, will be important to defuse a potential war crisis. That credibility should be established now, before such a crisis occurs.

The United States should speak out with equal clarity about continuing Indian human rights violations—and the immunity of the security forces—inside Kashmir, and more broadly about the benefits that would flow from a final political settlement and the normalizing of Indo-Pakistani relations. The US should acknowledge the legitimacy of Pakistan’s historic concerns about the rights of Kashmiri Muslims while insisting that only a peaceful political process to secure those rights is legitimate.

The United States does not need to intervene directly in Kashmiri negotiations to support the Indo-Pakistani peace process. It does, however, need to rediscover the sense of urgency and international leadership that characterized its engagement with Kashmir in the 1950s and early 1960s. Schaffer reports that in 1956, President Eisenhower wrote identical letters to the leaders of India and Pakistan, urging them forward. In the absence of a Kashmir settlement and the removal of other obstacles to normalized Indo-Pakistani relations, Eisenhower wrote, “the peaceful, progressive economic development which each nation desires…cannot succeed.” That has not changed.

—September 2, 2010

This Issue

September 30, 2010