Michael Krepon

The risk of a serious crisis between India and Pakistan escalating into a major military conflict is ever present. Diplomatic relations between New Delhi and Islamabad are badly strained. Talks on crucial subjects are on hold, while nuclear capabilities are expanding. Indicators of concern include the high level of civil unrest in the Kashmir Valley, infiltration along the Kashmir divide, and regular fire fights between Indian and Pakistani forces, including the use of artillery, mortars, and small arms.

The refusal of Pakistan’s military and intelligence services to shut down an-ti-India extremist groups, most prominently Jaish-e-Muhammad (JeM) and Lashkar-e-Taiba (LeT), further aggravates tensions and invites future flash-points for conflict. Pakistan’s military and intelligence services deny complicity in JeM’s and LeT’s cross-border operations, but these professions of innocence lack credibility in foreign capitals because infiltrators often need diversionary fire and other kinds of help. Moreover, Pakistani military commanders are expected to be aware of what transpires in their areas of control.

A major terrorist attack on a soft Indian target — iconic government buildings or monuments, religious shrines or temples, international airports, luxury hotels, and other symbols of India’s economic transformation — could spark another conflict. Alternatively, events in Kashmir, military clashes along the Line of Control (LoC) dividing the old princely state of Jammu and Kashmir, or a scenario that has not yet figured prominently in enmity between Pakistan and India could trigger the next crisis on the subcontinent. Whatever the triggering mechanism, another severe crisis would pose serious challenges to U.S. national security interests of preventing uncontrolled escalation and the detonation of nuclear weapons. A serious crisis could also interfere with Washington’s interests in continuing the upward trajectory of its ties with India, while maintaining sound working relations with Pakistan where interests overlap.

U.S. crisis management might be envisioned as a “playbook” offering lessons and management techniques learned from previous crises.1Polly Nayak and Michael Krepon, US Crisis Management in South Asia’s Twin Peaks Crisis , 2nd ed. (Washington, D.C.: Stimson Center, 2014); Polly Nayak and Michael Krepon, The Unfinished Crisis: US Crisis Management after the 2008 Mumbai Attacks (Washington, D.C.: Stimson Center, 2012); Michael Krepon, “Crises in South Asia: Trends and Potential Consequences,” in Crises in South Asia: Trends and Potential Consequences, ed. Michael Krepon and Nathan Cohn (Washington, D.C.: Stimson Center, September 2011), 20-26. This playbook can take physical form, with an outgoing administration providing its successor with files of its best judgments on how to deal with another crisis and the possibility of uncontrolled escalation. Of course, an incoming administration has every right to revise this playbook as it sees fit. It can also question fundamental assumptions about obligatory active U.S. crisis management. Indeed, a playbook cannot be static because the underlying circumstances and context of the next crisis between Pakistan and India will likely be different than the last. U.S.-India relations are in the process of transformation, as are China-Pakistan relations. Adapting a playbook to account for these shifts would reflect sound statecraft. A crisis management playbook is not a public document. The one outlined below is wide-ranging and suggestive.

Section one of this essay outlines contingencies for future India-Pakistan crises and identifies indications and warning signs to focus on. Section two reviews how another crisis could affect U.S. strategic and regional interests. Section three catalogues mechanisms in a notional U.S. crisis management playbook that could be adapted for use in the next crisis.

Contingencies and Warning Signs

One of the two most likely India-Pakistan crisis contingencies that could lead to war involves strikes against iconic targets in India by cadres of groups that have enjoyed safe havens in Pakistan. Severe crises in the past have been triggered by their attacks on the Indian Parliament building in 2001 and on several targets in Mumbai in 2008. There is no shortage of iconic targets within India, including government buildings and monuments, religious gathering sites, shrines and temples, international airports, luxury hotels, and other symbols of India’s economic transformation. Hindu extremists could also spark a crisis by attacking iconic Muslim religious edifices — like the demolition of the Babri Masjid in Ayodhya in 1992, which prompted bombings of the Bombay Stock Exchange and other targets the following year.

The second most likely contingency that could lead to a severe crisis and perhaps war is widespread violence in the Muslim-majority tinderbox of the Kashmir Valley. Public disaffection from Indian rule is extremely high, exemplified by stone-pelting youth and the firing of pellet guns by Indian security forces that have blinded demonstrators. India’s coalition government in Jammu and Kashmir has ruled heavy-handedly and seems unable to make amends with the majority of the population. As civil unrest grows, militant cadres based in Pakistan whose raison d’être is to free “occupied” Kashmir usually increase their infiltration rates, and firefights could intensify between military units along the Kashmir divide.

When bilateral relations deteriorate badly, it is not unusual for firing along the LoC to escalate, for border posts to be increasingly overrun, and for violent extremist groups to engage in increasingly ambitious raids. To stem this progression, Prime Minister Narendra Modi authorized “surgical strikes” across the Kashmir divide in September 2016 after cadres belonging to JeM struck an Indian military base in Uri. In doing so, Modi appears obliged to react in similar fashion after due provocation, while Pakistan’s armed forces are likely to feel obliged to respond with “befitting” responses.

Commando raids across the LoC are not new; publicizing them is. Perhaps caught off guard, Rawalpindi did not respond immediately to Modi’s announcement. Instead, Pakistan questioned its veracity while upping the ante with artillery fire and small unit operations along the Kashmir divide. New Delhi responded in kind. For whatever reason, Rawalpindi was willing to give one “pass” to Indian surgical strikes. If New Delhi again reacts assertively to provocations in the future, Rawalpindi is very likely to as well, because not to do so would constitute a loss of face. The use of drones, helicopters, and perhaps even combat aircraft to accompany cross-LoC raids could enter the picture, complicating escalation control efforts.

Further up the escalation ladder is the Indian Army’s “Cold Start” doctrine of shallow penetrations of Pakistani territory in places of India’s tactical advantage and choosing. The mobilization of Indian integrated battle groups along attack routes into Pakistan would, at least according to Indian military plans, proceed quickly to pre-empt Rawalpindi’s military and Washington’s diplomatic countermoves.

Rawalpindi has demonstrated ways to foil the Indian Army’s plans. It has flight tested capabilities to employ short-range — or (in Western parlance) “tactical” — nuclear weapons to deter India’s Cold Start military plans. To illustrate the seriousness of its deterrent, Pakistan has announced its readiness to use nuclear weapons first, if necessary, even on Pakistani soil. Indian leaders, in turn, have articulated and reaffirmed a doctrine of massive retaliation. They assert that a single detonation on or near a battlefield could trigger a catalytic response.

A scenario that has not yet figured prominently in crises between Pakistan and India could also spark a major confrontation, such as deadly attacks on Indian consulates in Afghanistan or attacks carried out by deeply disaffected Indian Muslims — perhaps in collusion with Pakistan-based groups in response to policies pursued by the Modi government. Pakistan has its own litany of complaint about India’s actions, particularly in Afghanistan and Baluchistan. To date, however, crises on the subcontinent have been triggered along the primary fault lines of India-Pakistan enmity, not in peripheral areas.

Warning indicators for the increased likelihood of the two primary contingencies — an attack on an iconic target in India and frustration by Kashmiri Muslims opposed to Indian governance boiling over — are virtually the same. These indicators are straightforward, readily monitored, and hard to miss. They include the number of attempted LoC crossings from the Pakistani side, the level and intensity of firing across the Kashmir divide, the extent of commando operations to overrun posts, the desecration of bodies, announcements of such operations after the fact (perhaps accompanied by dramatic footage), and “befitting” military responses. Several of these indicators are already evident.

The number and scope of attacks on religious processions, gatherings, temples, or mosques are important indicators of a downward spiral. The bloodier the attack, the more likely it could prompt a strenuous reaction. The most egregious, but far from the only, case of targeting a religious site was the 1992 demolition of the Babri Masjid in Ayodhya, India, by a mob incited by leaders of the Bharatiya Janata Party and its affiliates. Retaliatory attacks in Bombay the next year resulted in over 250 fatalities, including the bombing of the Bombay Stock Exchange. Further examples of soft targets include numerous religious temples in Varanasi, a holy city within the parliamentary district represented by Modi. Within India, there is a trend of increased targeting of places of worship, mostly directed at mosques.

The 2001 attack on the Indian Parliament building occurred after a progression of smaller-scale attacks culminating in the truck bombing of the State Assembly building in Srinagar. This attack did not elicit a reaction by New Delhi. The incidence and patterns of militant attacks against India can provide important indicators of bigger explosions to come.

If a crisis becomes serious enough to trigger preparations for a limited conventional war, key indicators would include troop movements from garrisons and heavy equipment from storage areas. Military doctrines are spring-loaded to take offensive or blocking action. Time is of the essence to gain military advantage and to foil opposing military logistical gains. An intense crisis and perhaps warfare are in the offing if civil traffic on railways is subordinated to military needs and if ammunition accompanies troops to fighting corridors. Movements of nuclear-capable delivery vehicles to satellite bases or toward fighting corridors would be particularly ominous, as would the stand-down of aircraft operations to achieve peak readiness as a crisis builds.

Attempts to improve bilateral relations could also prompt explosive reactions. The most notable example so far was then Prime Minister A.B. Vajpayee’s trip to Lahore to improve ties after the 1998 nuclear tests. Pakistan’s military leadership’s rejoinder was to send Northern Light Infantry troops across the Kashmir divide, triggering the 1999 Kargil War. On three occasions, Prime Minister Modi has tried to employ gestures to improve ties with Pakistan, only to be rebuffed by forays carried out by anti-India groups against Indian diplomatic and military outposts.2Prime Minister Modi’s invitation to Prime Minister Sharif to attend Modi’s inauguration (2014), India and Pakistan’s joint statement condemning terrorism (2015), and Modi’s surprise Christmas Day visit to Lahore (2015) were all followed by terrorist attacks against India within days. Michael Krepon, “Kashmir and Rising Nuclear Dangers on the Subcontinent,” Arms Control Wonk, January 17, 2017. As long as Pakistan’s security apparatus views a normal neighborly relationship with India as inimical to its interests, it can easily utilize proxies to disrupt diplomatic progress.

Implications for U.S. Interests

Regardless of how a future crisis is sparked, the failure of escalation control could have dire consequences for the region, for reducing nuclear dangers globally as well as regionally, and for U.S. national security’s regional and strategic interests.

Great damage to U.S. strategic interests would result if escalation was uncontrolled and nuclear weapons were used on a battlefield for the first time since 1945. If the “nuclear taboo” were broken, the global nuclear order would be badly shaken. The Nuclear Nonproliferation Treaty is likely to suffer significant further weakening. A resumption of nuclear testing by several states could also follow. The environmental and humanitarian consequences of the battlefield use of nuclear weapons could be severe depending on how many and where the detonations occur, placing a premium on crisis managers to intervene quickly under extraordinarily intense pressures.

A second U.S. strategic interest is to prevent the theft and unauthorized use of nuclear materials and weapons. During intense crises, nuclear assets are likely to be moved to avoid being targeted, to signal Pakistan’s military resolve, and (from Pakistan’s perspective) to focus Washington’s attention on crisis management. Nuclear safety and security are subject to compromise when nuclear assets are moved. The theft of a nuclear warhead or fissile material by an extremist group could result in nuclear-armed terrorists targeting the United States and other states.

A severe crisis between India and Pakistan has the potential to worsen U.S. ties with both countries, as well as China. Alternatively, a crisis could reaffirm stronger ties with India, clarify further the costs of Pakistan’s ties to extremist groups, and deepen a pattern of cooperation with Beijing on regional crisis management. Maintaining strong ties to India, sound working relations with Pakistan on matters of mutual interest, and a crisis management partnership with China are important U.S. regional security interests.

Having strengthened its political and defense ties with Washington, New Delhi now expects the United States to be an ally rather than an “honest broker” in a future crisis, in addition to de-escalating tensions with Pakistan. Washington would, in turn, require Beijing’s help to defuse the crisis, rather than backing Pakistan in ways that make crisis management and de-escalation more difficult.

Strong and effective U.S. crisis management efforts could yield dividends elsewhere. Other allies in other regions would be monitoring U.S. crisis management efforts on the subcontinent to draw conclusions and set expectations in the event that a crisis erupts in their own regions.

The Limits of U.S. Diplomacy

U.S. diplomatic initiatives have not laid the conditions for crisis prevention on the subcontinent, let alone conflict resolution. Nor have Indian and Pakistani diplomatic initiatives. In theory, one way to prevent the next crisis on the subcontinent would be to assist in a diplomatic settlement of the Kashmir dispute. In reality, Washington has neither had the power nor the interest to do so. If Washington were to wade into the Kashmir morass when conditions were not amenable for success, the situation on the ground would likely worsen — even if such an initiative were acceptable to New Delhi, which is extremely unlikely.

If Pakistani and Indian leaders do not have the interest or political will to resolve the Kashmir dispute, Washington will not be able to manufacture either commodity. The best preventive options for another serious Kashmir crisis are improved Indian governance and far more restrictive rules of engagement for Indian security forces. Although Washington could quietly encourage New Delhi to take positive steps in this direction, only New Delhi can pursue these initiatives.

Washington’s leverage on Pakistan to take fundamental shifts away from violent anti-India extremist groups was never greater than immediately after the 9/11 attacks and the 2001 Parliament attack, but promises by Pakistan’s military leader at the time, General Pervez Musharraf, to rein in these groups were short-lived. Pakistan’s military and intelligence services have so far apparently concluded that the costs of cutting ties with perceived strategic assets remain greater than the reputational costs of maintaining existing links. (A very different calculus has been applied to the Tehrik e-Taliban Pakistan, which has trained its fire against the state.) The absence of a major attack on an iconic Indian target since 2008 suggests the possibility that Pakistan’s military and intelligence services have worked out understandings with groups such as the JeM and LeT to conduct operations on a much lower scale. Scales of violence can, however, become sliding.

The success of previous U.S. crisis management efforts has been possible because Indian and Pakistani leaders have wished to avoid war and escalation. In the crucible of the 2001-2 Twin Peaks crisis and the deeply embarrassing (to Pakistan as well as India) 2008 Mumbai crisis, Indian prime ministers of two very different coalition governments concluded that the prospect of battlefield gains could not justify the risks of escalation and to national economic growth. In these two crises, New Delhi’s decisions to stand down were facilitated by pledges elicited by U.S. crisis managers from Pakistan’s military leaders to shut down anti-India extremist groups. But there is no evidence that Indian leaders took these pledges seriously.

What, then, might be done to avoid the next serious crisis? Indian and Pakistani prime ministers are likely to want to encourage greater trade and cross-border investments but have been foiled by Pakistan-based spoilers. The advent of a newly elected Pakistani government in 2018 could provide another opportunity for Modi to seek improved relations. Private U.S. messaging to encourage Modi to try again would do no harm and could be helpful. If, however, Modi does try again, public as well as private messages directed at Pakistani audiences would be advisable, pointedly noting why Modi’s previous initiatives have failed and placing the burden on Pakistan’s military and intelligence services to prevent spoilers from once again taking blocking steps.

In the absence of active U.S. diplomacy seeking to improve India-Pakistan relations, the most useful tool for preventing another severe crisis on the subcontinent might well be U.S. intelligence collection and sharing. Washington might be able to prevent a serious crisis by obtaining timely warning of an impending attack and sharing this information with both Indian and Pakistani officials. Broader and deeper intelligence cooperation between the United States and India would be unsettling for Pakistan, but it could also have some useful deterrent effect in preventing crisis-triggering explosions.

What other moves by Washington might decrease the likelihood of another crisis on the subcontinent? The Trump administration has mostly dispensed with diplomatic nuance toward Pakistan, and the U.S. Congress has begun to impose heavy penalties for Rawalpindi’s choices by reducing Coalition Support Funding and the denial of credits to help finance combat aircraft. Some also propose declaring, or threatening to declare, Pakistan a state sponsor of terrorism. A renewed threat to do so might prompt temporary and cosmetic changes in Rawalpindi’s actions, just as threats by the George W. Bush administration after 9/11 had this effect. If, however, the United States labels Pakistan a state sponsor of terrorism, Washington would be without any levers to influence Pakistan’s choices regarding nuclear weapons, proliferation, and counterterrorism — choices Washington cares about greatly. The United States as well as Pakistan would both lose by acting on this threat, and its hollow repetition diminishes U.S. credibility.

Withdrawing Pakistan’s status as a “major non-NATO ally” could clarify its growing isolation from the West even if this decision no longer has practical effect. Congress already limits the types of military equipment available to Pakistan relative to India and denies favorable financing for arms sales.

The United States has tried to sensitize Pakistan and India to the dangers of an intensified nuclear arms competition. During a crisis in 1990 triggered by large-scale Pakistani military exercises, U.S. emissaries encouraged the adoption of confidence-building measures (CBMs) and nuclear risk-reduction measures (NRRMs). Regrettably, neither side has viewed these measures as having intrinsic value; instead, they are viewed as devices to adopt to fend off external pressures after crises or to trade for something deemed more important.

The last measure negotiated was in 2007 on procedures to reduce risks from accidents relating to nuclear weapons, which has been subsequently renewed. While Washington would ruffle feathers by calling out India and Pakistan on their lack of progress in reducing nuclear dangers by diplomatic means, no harm can come from private and public messaging that responsible nuclear stewardship can be demonstrated by adopting additional CBMs and NRRMs.3Frank O’Donnell, “Launching an Expanded Missile Flight-Test Notification Regime,” Stimson Center, Off Ramps Initiative, March 23, 2017, As for cooperation on preventing the theft of a nuclear device or fissile material, it would be very difficult for either country, after repeatedly downplaying this risk, to acknowledge security shortfalls or its occurrence. In particular, Rawalpindi would be concerned about triggering U.S. military operations within Pakistan — akin to the May 2011 raid on Osama bin Laden’s compound in Abbottabad — if they acknowledged the theft of a nuclear device or its key components.

The Trump Administration and the Increased Complexity of Escalation Control

In the 1999 Kargil, 2001-2 Twin Peaks, and 2008 Mumbai crises, the Bill Clinton and George W. Bush administrations had no indecision about taking the lead in crisis management. It is not obvious that the Trump administration would react in the same way. There is no obvious lead crisis manager within the administration, and it remains unclear how the application of an “America first” approach to the region would fare against the very substantial U.S. interests — including preventing uncontrolled escalation and nuclear weapons’ use — that could be placed at risk in the event of another severe crisis. It remains possible that the Trump administration could adopt a lower profile during a crisis to place more of the burdens of crisis management on Indian and Pakistani decision makers. If, however, the Trump administration, like its predecessors, decides to play an active crisis management role, it would have a well-worn playbook to draw upon, as discussed below.

With very limited diplomatic leverage to prevent crises, U.S. crisis management has focused on mitigating the risks and dangers of escalation control. In the 2001-2 and 2008 crises, this meant trying to persuade the Indian prime minister not to respond to a grave provocation with the use of force. U.S. suasion barely succeeded after the 2001 attack on the Indian Parliament; it was a far easier “sell” after the 2008 Mumbai attacks. U.S. crisis management succeeded in both cases because two very different Indian prime ministers came to the same conclusion — that no battlefield gains could justify risks to national economic growth and the uncertainties of another war with Pakistan.

Crisis management is now far more difficult. First, the promises elicited from Pakistani leaders in previous crises have not been honored, and new promises will have little credibility unless backed by highly unusual actions. Second, Prime Minister Modi has signaled that he will not be deterred from retaliating if warranted by the level of provocation — this was, after all, the reason why he authorized surgical strikes after the September 2016 attack at Uri. Consequently, Indian retaliation will be widely presumed if warranted by the provocation. Moreover, New Delhi would expect strong backing from Washington whenever and however it chooses to retaliate. Once India retaliates, a “befitting” response is likely from Pakistan’s military, which would otherwise lose face. And what then? The primary task of crisis managers would then be to control escalation not after the initial provocation, as in previous crises, but after second- and third-order strikes.

Possible Elements of a U.S. Crisis Management Diplomatic Playbook

Every administration that has engaged in high-stakes crisis management on the subcontinent has learned lessons that have been passed down to its successors. This “playbook” includes the choreography of visits and phone calls by foreign leaders. In previous crises, U.S. officials have presumed that Indian leaders would not initiate conflict if a major world leader or top U.S. official were en route to the region. Regular high-level visits by senior U.S. officials were also deemed advisable, as they could prove useful after a crisis erupted. While it is true that decisions are made on the basis of perceived national interests and not on personal ties, personal relationships still matter, as they could provide insights to help defuse a severe crisis.

The absence of senior U.S. officials with knowledge of the region can become an obvious deficit during a deep crisis. If the Trump administration wishes to engage in an active crisis management role but has no seasoned crisis manager, it might consider “deputizing” someone from the outside who has the requisite skills and experience.

In the past, Beijing has played a subordinate and supportive role in U.S. crisis management efforts. As China’s stake in Pakistan grows, its role in crisis management could change. It might be wise to add an agenda item for high-level meetings with Chinese officials on contingency planning in the event of another severe crisis on the subcontinent.4Yun Sun, “Create a Channel for a U.S.-China Dialogue on South Asia,” Stimson Center, Off Ramps Initiative, August 10, 2017, Beijing might still be content to play a subordinate role, leaving the heavy lifting to Washington, but close collaboration would still be necessary.

Routinized intelligence cooperation prior to a crisis could be essential in the event that a crisis erupts. U.S. crisis managers usefully conveyed intelligence assessments of disastrous potential outcomes during the 1990 crisis triggered by large-scale Pakistani military exercises. This sobering information still has utility. In the 2008 Mumbai crisis, Washington assisted New Delhi by offering close cooperation with the Federal Bureau of Investigation on matters of domestic terrorism and forensics. Making public new U.S.-Indian counterterrorism interactions could have modest deterrent benefit. This could be pursued on many fronts by different intelligence agencies and their Indian counterparts. In addition, it seems essential to continue to help Pakistan, to the extent possible, on its counterterrorism challenges.

There could well be value in sensitizing publics about the humanitarian, social, environmental, and food security consequences of nuclear exchanges. There are no downside risks to engaging scientific and environmental experts in both countries to join U.S. and international experts in assessing the modeling of the consequences of nuclear detonations on the subcontinent. Washington would do well to encourage both governments to allow their nongovernmental experts in climatology, environmental science, agriculture, and other disciplines to carry out joint studies on the humanitarian and environmental consequences of nuclear exchanges.5Arka Biswas, “A Joint India-Pakistan Initiative on the Humanitarian Impact of Nuclear Weapons,” Stimson Center, Off Ramps Initiative, May 4, 2017,; and Saira Bano, “Convene a Joint Commission on the Consequences of a Nuclear War in South Asia,” Stimson Center, Off Ramps Initiative, October 17, 2017,

Some purchase might be gained at the front end of a severe crisis if leaders on both sides have invested political capital in trying to improve relations. Symbolic gestures by Indian leaders, however, are unlikely to be decisive in convincing Pakistan’s military and intelligence services to turn the page. If Modi tries once again to improve ties, Washington could provide reinforcement by privately as well as publicly placing Pakistan on notice that it would be held responsible for blocking actions taken by anti-India militant groups that enjoy safe havens.

One way to defuse escalation during a crisis might be to call for a ceasefire along the Kashmir divide. A generally accepted rationale for ceasefires is for the sake of innocent civilians who bear the brunt of violence along the LoC. However, it is hard to implement ceasefires when Kashmir is on the boil, and they are easily broken.

Washington could also let it be known that it is considering the removal of nonessential U.S. government workers and the issuance of travel advisories to the region as a severe crisis unfolds. The U.S. ambassador to India issued these directives in the latter half of the Twin Peaks crisis — not to be manipulative but out of genuine concern to remove large numbers of U.S. nationals from harm’s way. The ambassador’s decision had the effect of halting U.S. business delegations to India and prompting New Delhi to take steps to wind down this extended crisis.

An activist U.S. administration’s playbook might also include steps to consider after a crisis has de-escalated. One step might be to publicly encourage the negotiation and adoption of new CBMs and NRRMs. Private messages and the talking points of U.S. government spokespersons along these lines have had little effect. Public statements by high-level officials might therefore be warranted. Leaders in both countries are sensitive to the argument that responsible nuclear stewardship requires steps to reduce nuclear dangers, especially when nuclear capabilities are advancing on several fronts. These messages would, however, be viewed as hypocritical unless Washington is also pursuing nuclear risk-reduction measures with Moscow and Beijing.


In the past, U.S. crisis managers have succeeded because Indian leaders have wanted them to succeed. New Delhi has not retaliated against deeply embarrassing and horrific acts of violence initiated or abetted by Pakistan’s military and security forces, prioritizing India’s economic growth against the uncertainties of warfare and escalation control.

The dynamics of crisis management and escalation control are more complicated now. Indian Prime Minister Modi has in effect forecast surgical strikes in the event that provocations cross unspecified thresholds. Pakistan’s military is likely to retaliate as it deems necessary, leaving the next decision in New Delhi’s hands. Previous U.S. administrations adopted an activist crisis management approach; the Trump administration’s approach remains unclear.

There are persuasive reasons for New Delhi and Rawalpindi to seek to control escalation. Pakistan’s future depends increasingly on a stable climate for Chinese investments and India’s growing power is directly linked to open-ended economic growth. Uncontrolled escalation leading to a limited conventional war and the possibility of crossing the nuclear abyss would be ruinous to these plans.

Nonetheless, the conditions for another serious crisis exist between India and Pakistan, and will continue as long as powerful interests in Pakistan remain irreconcilable to a normal relationship with India — a condition reinforced by deep tensions between Muslim-majority Kashmir and New Delhi.

Michael Krepon received the Carnegie Endowment's Thérèse Delpech Memorial Award in 2015 for lifetime achievement in nongovernmental work to reduce nuclear dangers. Krepon co-founded the Stimson Center in 1989, served as Stimson’s President and CEO until 2000, and continues to direct Stimson’s programming on nuclear and space issues. He was appointed the University of Virginia’s Diplomat Scholar, where he taught from 2001-10. He is the author and editor of 21 books, most recently The Lure and Pitfalls of MIRVs: From the First to the Second Nuclear Age (Stimson Center, 2016). He worked previously at the Carnegie Endowment for International Peace, the U.S. State Department’s Arms Control and Disarmament Agency during the Carter Administration, and on Capitol Hill.