In the summer of 1999, India and Pakistan went to war, again. Pakistan had secreted a sizeable force in remote outposts in the high mountains near Kargil, in the northern part of the Indian state of Jammu and Kashmir. When India discovered the Pakistani forces build-up, it mounted a hurried and initially haphazard response to dislodge the invaders, deploying heavy reinforcements near Kargil and mobilizing its air force for daring air strikes. India’s forces fought tenaciously, for weeks, with soldiers often scaling sheer cliff-faces and fighting hand-to-hand against the enemy, to painstakingly recapture the mountainous territory, peak after peak.
India also did something surprising. Unlike in previous wars against Pakistan, in 1965 and 1971, Indian forces never crossed over into Pakistani territory during the 1999 Kargil War. The cabinet had set a limit: None of India’s ground or air forces were to cross the Line of Control (LoC), a line that separates Indian- and Pakistani-controlled parts of disputed Kashmir and serves as the de facto boundary between the two countries. Even when Indian operations were failing in initial weeks and the Army prepared for a large counter-offensive elsewhere into Pakistan, the order to expand the war never came. India, it turns out, fought with remarkable restraint.
To many observers, the obvious reason for this restraint was India and Pakistan’s new status as declared nuclear-armed countries. Both nations had first tested nuclear weapons in May 1998, only one year before the Kargil War started. India and Pakistan had possessed deployable nuclear weapons for years beforehand, including during previous crises (Hagerty 1995). But now they were two openly-declared nuclear powers at war. For nuclear deterrence optimists, the Kargil War is a clear-cut case of nuclear-armed countries at war intentionally limiting their military operations to avoid escalation (Ganguly 2008). From this perspective, deterrence almost self-evidently worked.
The fledgling deterrence between India and Pakistan, however, does not accurately explain Indian behavior. Other strategic conditions weighed at least as heavily in keeping India’s response to Pakistan’s incursion restrained and the war limited to Indian territory. Among these conditions, three such factors proved most important: India had purely defensive war aims; it had political motivations to appear measured and responsible; and it had a readymade threshold, in the LoC, to set clear boundaries to its military operations. These factors were, by chance, all in effect in 1999—but they are now all obsolete. In the next crisis or war with Pakistan, India may still be deterred from an escalatory dynamic which could quickly lead to nuclear war, but many of the previous safeguards that kept the Kargil War limited no longer exist.
Limited war in the mountains
In the Kargil War’s initial stages, India underestimated the threat. Its early encounters with the Pakistani fighters, apparently scattered in pockets many miles apart, gave little clue about the size of the infiltration. The Pakistanis dressed and spoke as if they were mujahideen from the northern tribal areas, but in fact they were Pakistani paramilitaries and Army officers, orchestrating an elaborate deception plan to occupy land on India’s side of the LoC. For weeks, the Indian Army was certain it was intercepting militants routinely on their way to the Kashmir valley, where an insurgency had festered for the better part of a decade. It was convinced this was a local problem that would be summarily resolved—until it wasn’t.
As Indian assaults met with well-armed and organized Pakistani resistance, the military leadership came to realize it was facing a formidable opposing force that would require a concerted military campaign. At cabinet meetings in late May 1999, the government approved the use of air power to attack the Pakistani positions, which it had earlier refused to do. But the government insisted that ground and air operations remain on the Indian side of the LoC. This was a firm limitation on military operations, one that the military accepted without question and would never reconsider throughout the course of the war.
Given the remoteness and tactical difficulties of the battlefield, the Indian Army took weeks to deploy and acclimatize heavy reinforcements. With local assaults making no headway against the well dug-in Pakistanis, pressure in New Delhi was mounting to open a new front where India might have greater advantages and gain some leverage. But the Army never needed to seek cabinet permission for such escalation.
In mid-June, Indian forces finally achieved their first breakthrough, capturing a massive mountain feature known as Tololing. In subsequent weeks, with ground and air forces adapting to the conditions, they achieved more tactical successes. The Pakistan government, maintaining the claim that the incursion was the action of non-state militants, dared not reinforce its troops. Consequently, the Pakistani positions became more difficult to resupply, their morale collapsed, and their defensive line crumbled. In desperation, Pakistan’s Prime Minister Nawaz Sharif hurriedly invited himself to Washington in July, to plead for a US-brokered ceasefire. President Clinton, instead, insisted on Pakistani withdrawal as a precondition of any further diplomatic process. Days later, suffering a rapidly deteriorating tactical situation, Pakistan agreed to India’s terms for a ceasefire. It began a phased withdrawal, which concluded on July 26, 1999.
Although the war had remained limited—with India fighting back within the geographic confines of Kargil, and Pakistan not making any credible threats of nuclear escalation—Washington was anxious about the risk of escalation. Coinciding with Sharif’s visit to Washington, the White House received intelligence of possible heightened Pakistani nuclear readiness, although Indian leaders and subsequent Pakistani observers alike dismissed that reporting as a misinterpretation.
The risk of a nuclear reaction would surely have been higher had India launched a full-blown counteroffensive. In earlier crises—such as the “Brasstacks Crisis” of 1987-88 and the “Compound Crisis” of 1990—the simple threat of a major Indian military offensive, without crossing the border, had elicited nuclear threats from Pakistan in return. In Kargil, India had enticing military reasons to expand the war. Air strikes across the LoC at Kargil would have more quickly isolated and weakened the infiltrators. Attacks across the LoC elsewhere, or even the international border further south, would have given India bargaining leverage if operations in Kargil failed. Despite the potential operational benefits of escalating, the Indian government was adamant that India would remain restrained. That forbearance proved critical in limiting nuclear risk in the first ever—and only—war between two nuclear-armed countries.
The actual reasons for India’s restraint
In Kargil, the Indian military did have standing contingency plans to strike across the LoC and it mobilized to prepare for a counteroffensive. The military did instinctively prefer a wider war, but the government from the onset of the conflict had ordered Indian forces not to cross the LoC. This led some analysts and historians to argue that India’s restraint was motivated by deterrence—that is, by a fear of nuclear escalation.
India’s leaders were generally aware of nuclear dangers. Only months after the nuclear tests of May 1998, India and Pakistan had initiated a new high-level diplomatic process aimed at reducing nuclear risk. The diplomacy culminated with the Lahore Declaration of February 1999, which explicitly noted that the two sides’ new nuclear weapons status imposed an added responsibility to avoid conflict. Both countries also committed to a series of nuclear confidence-building measures, such as the notification of nuclear-related accidents and upcoming ballistic missile tests. In parallel, the Indian military began considering new war-fighting concepts and plans, with a view to keeping conventional conflicts well below the nuclear threshold. India’s Chief of Army Staff, Gen. V.P. Malik, began deliberating on the dynamics and requirements of limited war to fight without provoking a nuclear retaliation (Malik 2006). Indian political and military leaders alike acknowledged the responsibilities and risks associated with their new nuclear status and accepted the necessity for restraint to avoid escalation.
However, that awareness does not actually explain India’s strategy during the Kargil War. According to firsthand accounts in memoirs and interviews I conducted of former political and military officials, India’s restrain during the Kargil War did not directly spring from concerns of nuclear escalation. Instead, three other specific strategic factors enabled and encouraged restraint.
First, India’s war aims focused on defense, rather than coercion. Its priority was to evict the Pakistani intruders, rather than to impose any costs or otherwise coerce Pakistan. India demonstrated that a strategically defensive campaign could be successfully fought without tactical offensives into enemy territory. This probably came at an operational cost, however. For instance, the Indian air force could not fly over the LoC on bombing runs or strike enemy support targets across the LoC, actions which could have dislodged the Pakistani intruders more quickly. Despite facing a clear foreign incursion into its territory, India limited itself neatly to defensive-only war aims.
Second, India’s wider geopolitical ambitions at the time of the Kargil War encouraged restraint. The government’s strategic priority was to recover its international standing and build a productive relationship with the United States. Throughout the year preceding the war, India had been roundly marginalized as a pariah state—a distinction it shared with Pakistan—because of its nuclear tests. Indian Prime Minister Atal Bihari Vajpayee and his government very pragmatically recognized that they had to build their legitimacy as a new but responsible nuclear power, particularly if they wanted to lift the sanctions that much of the international community, led by Washington, had imposed after the tests.
According to several firsthand accounts and confirmed in my interviews, restoring India’s international status was the primary motivation for restraint during the war (Singh 2006, Sinha 2020, Das 2021). India portrayed itself as a purely defensive, responsible country that resisted taking advantage of the situation for self-gain, and sought only the restoration of the status quo. This strategy proved to be sagacious. Washington had maintained a close security partnership with Pakistan throughout the 1980s and considered both India and Pakistan to be equally reckless with their nuclear testing in 1998. Now, however, the United States was identifying Pakistan as the aggressor in the Kargil War, and supporting India’s position wholly. As a result, President Bill Clinton was careful to align closely with India’s policy as he negotiated with Pakistan Prime Minister Sharif in July 1999 (Riedel 2002).
The third factor contributing to Indian restraint was the availability of a clear threshold for military operations. The position of the LoC was well understood by both sides—albeit with some minor ambiguity between outposts—and it became a clear boundary for Indian operations. This sharply delineated geographic threshold allowed India to signal its credibility, to both Pakistan and the international community, about its officially stated position that India’s actions were not intended to be escalatory.
The clarity of escalatory thresholds is especially important in the India-Pakistan rivalry because the two nuclear powers are adjacent, with potential battlefields being relatively small. The center of the Pakistani metropolis of Lahore is scarcely 20 kilometers from the Indian border; the small handful of major north-south lines of communication are similarly vulnerable. In such an environment, relatively minor gradations of force or geographic spread could transform a contained battle into a perceived strategic or even existential threat to Pakistan. Indian decisionmakers recognized this during the Kargil War. General Malik surmises, for example, that a decision to cross the LoC would have been accompanied by more serious concerns over how to avoid nuclear risk (Malik 2006). The LoC represented not only a geographic boundary, but also a strategic boundary within which Indian forces could fight with little risk of escalating the conflict.
Together, these three strategic factors proved essential to restrain Indian military operations during the Kargil War. But had Indian attacks not achieved the breakthroughs they did in mid-June, the Army would very likely have sought and gained cabinet permission to expand the war. If tactical imperatives had demanded an escalation, an unrestrained India would have had to depend on nuclear deterrence—untested in wartime—to keep the war limited. It so happened that India’s strategy of restraint prevailed and ensured that both countries did not have to rely on deterrence. Despite what deterrence optimists trumpet, the Kargil War between the two nuclear rivals was, in fact, a very weak test of deterrence.
The current erosion of India’s restraint strategy
The strategic factors encouraging restraint that shaped the course of the Kargil War were all contingent on the conditions of the time. In today’s India-Pakistan relations, these factors hold different values. Not only do they offer little prospect of keeping a war limited, but they could have the opposite effect—encouraging riskier war plans.
First, the most likely conflict scenarios between India and Pakistan involve India pursuing coercive, not defensive, war aims (Kapur 2008). The 1999 Kargil War was the last time Pakistan launched a military bid to capture part of Indian-controlled territory. It has since retained its basic strategic goal of contesting Indian control over Jammu and Kashmir and prioritized instead a campaign of supporting cross-border terrorist attacks. These attacks have triggered multiple crises already and would be the likely casus belli for an India-Pakistan war. In 2001 and 2002, India ordered a general mobilization, known as Operation Parakram, threatening a punitive attack on Pakistan, until international pressure convinced it to stand down. In 2016, the Indian military launched a small infantry raid across the LoC, and in 2019 it launched an air strike on a purported terrorist facility in Balakot, inside undisputed Pakistani territory. In each one of these instances, the Indian military actions were in response to terrorist attacks and aimed at forcing Pakistan to halt its provocations. Unlike during the Kargil War, such measures require some form of offensive action in Pakistan, which are unavoidably riskier steps on the escalation ladder.
Second, India’s contemporary geopolitical goals center more on the expression of independent power than the quest for international legitimacy. In no small part, India’s restraint during the Kargil War and its subsequent rapprochement with Washington allowed India to establish itself as key member and defender of the international order. The United States has become a source of real-time operational support and long-term technological gain, but the balance has shifted since 1999. The United States now also eagerly courts India as an indispensable partner. In fact, the Modi government has pugnaciously declared India as an independent power center in an emerging multipolar system (Jaishankar 2020). When New Delhi’s policy diverged from Western policy over the Russian invasion of Ukraine, Indian leaders unapologetically explained that they were serving Indian interests and owed nothing to what they considered a hypocritical European agenda (Laskar 2022).
Third, there is currently no clear escalatory threshold that India can credibly claim to respect. New Delhi certainly no longer sees the LoC as an inviolable instrument of the status quo, as it did in Kargil. Rather, India’s crisis behavior in recent years has crossed previously inviolable thresholds, probably to intentionally create uncertainty and deterrence in the enemy’s mind. In its “surgical strike” of 2016, India crossed the LoC. In the 2019 Balakot air strike, it crossed both a geographic threshold, in attacking undisputed Pakistani territory beyond Kashmir, and a force-level threshold, in using air power. The escalating nature of this chain of crises has set precedents and suggests that a much wider range of military actions are now on the table—actions that were unthinkable during the Kargil War.
Restraint can no longer be considered a predictable hallmark of Indian strategic behavior. India has now traded force for risk, supplementing smaller attacks with higher levels of unpredictability. This risky behavior, however, despite appearing helpful for deterrence, is dangerous for creeping every crisis closer to nuclear escalation.
In future crises, Indian military operations in Pakistani territory may not be contained within clear thresholds. Indian planning very likely incorporates some limits to the types of force or targets it considers, but these are not apparent to outside observers—including its enemy. Likewise, Pakistan has signaled that it has set multiple red lines for nuclear use, but those red lines are only vaguely stated—presumably deliberately—to retain some freedom of action and impose caution on Indian military planners. Given the reciprocal opacity of limits and red lines, Indian actions against Pakistan, unlike during the Kargil War, will not have clear thresholds with which to signal its restraint and reduce the risk of escalation.
The Kargil-era pillars of restraint have crumbled, and observers of the India-Pakistan conflict should take note. In 1999, the interplay of these strategic conditions determined India’s response, rather than concern about Pakistani nuclear retaliation. In a future crisis, India would likely go on the offensive rather than stay on the defensive, seek to display its self-confident power rather than seek to win international acceptance; and lack any clear thresholds to limit military operations. Now more than ever, battlefield unpredictability could determine whether a conflict remains limited or escalates. If Pakistan was to put up stiffer than expected resistance, India would be pressured to escalate further to achieve some measure of success. In future crises or open conflicts, an India no longer restrained by old strategic conditions may be tempted to take greater risks, testing nuclear deterrence in unprecedented and potentially cataclysmic ways.
This article is based on research funded by the Stanton Foundation and administered through the Center for International Security and Cooperation, Stanford University.
The author wishes to thank Paul Kapur for helpful comments on an earlier draft.
The Bulletin elevates expert voices above the noise. But as an independent, nonprofit media organization, our operations depend on the support of readers like you. Help us continue to deliver quality journalism that holds leaders accountable. Your support of our work at any level is important. In return, we promise our coverage will be understandable, influential, vigilant, solution-oriented, and fair-minded. Together we can make a difference.