From Kargil to Balakot: Southern Asian Crisis Dynamics and Future Trajectories


The Stimson Center 15 May 2020

Two decade review: South Asia crises will likely escalate further & faster as escalation pressure and crisis management challenges grow Project Part of the Crises & Consequences in Southern Asia Project Program South Asia Download PDF File By Sunaina Danziger  ·  Grace Easterly  ·  Zeba Fazli  ·  Gillian Gayner  ·  Sameer Lalwani  ·  Tyler Sagerstrom  ·  Brigitta Schuchert  ·  Chloe Stein  ·  Elizabeth Threlkeld  ·  Akriti Vasudeva

Two decades since the Kargil Conflict fundamentally reshaped strategic thinking in South Asia, 2019 brought significant developments that suggest further shifts in crisis dynamics are underway across the region. These flashpoints raised pressing questions for analysts seeking to understand how recent trends in nationalism, the rise of China, the growing role of media in forming public opinion, and military modernization shape the escalatory pathways of—and off-ramps from—these contemporary crises in Southern Asia.

From September 17-18, 2019, the Stimson Center’s South Asia Program hosted a workshop on crisis dynamics in Southern Asia with over 25 American, British, Chinese, Indian, and Pakistani analysts. The workshop included a simulation modeling India-Pakistan escalation dynamics and several panel discussions. In hosting this workshop, Stimson sought to pool the collective insights of leading South Asia analysts for debate and discussion centered on a central question: How have crisis dynamics in Southern Asia evolved over the last two decades, and what will the trajectories of crises look like moving forward?

On this page

Executive Summary

The Stimson Center’s South Asia Program convened a workshop in fall 2019 on crisis dynamics in Southern Asia with over two dozen analysts. Several important conclusions emerged from the workshop:

  • Inadvertent escalation risks in South Asia are concerning, and likely to recur. Even in a limited engagement, the risk of India-Pakistan conflict escalation is high because India—perceiving immutable Pakistan support for terrorism—seeks disproportionality in a crisis while Pakistan seeks conventional reciprocity. Both sides routinely miscalculate the other’s resolve and maintain incompatible theories of de-escalation.
  • Studying and learning from past crises does not ensure prevention. The quality of learning varies significantly by country, government, and contingent political interests. Learning the “right” lessons may not necessarily translate into improved decision-making. States can also draw lessons that make them more risk-prone than risk-averse.
  • With Pakistan, India has traded strategic restraint for strategic risk manipulation. India has been “cumulatively emboldened” by the absence of international push-back to behave more assertively towards Pakistan, perhaps risking a commitment trap even as it retains calibrated restraint and bargaining ability when it comes to friction with China.
  • Deterrence and crisis instability will likely intensify alongside further strategic and conventional force modernization. Advances in precision and standoff weapons capabilities enable India and Pakistan to satisfy public demands for retaliation and demonstrations of resolve, even without results. Their incomplete adaptation of counterforce capabilities could stoke false optimism and foster risky decision-making.
  • Future crises can escalate sharply and rapidly, and will be less open to crisis management. Balakot was a watershed moment with several thresholds crossed more rapidly than expected. New technologies compress decision-making timelines, reducing space for de-escalation or third-party intervention. A less active or impartial U.S. role further compounds this challenge.
  • A host of asymmetries limit the scope of bilateral confidence-building and third-party mediation, despite U.S. stakes and a growing Chinese role. Bilateral confidence-building measures (CBMs) are hampered by asymmetries in power, interest, domestic institutions, conceptions of the status quo, and relations with great powers. Third-party involvement is constrained by asymmetric demand from the two sides, perceptions of partiality, and insufficient leverage over nuclear-armed states.


Context for Analysis

Two decades since the Kargil Conflict fundamentally reshaped strategic thinking in South Asia, 2019 brought two significant developments that suggest further shifts in crisis dynamics are underway across the region.  The first came in response to a February 2019 terrorist attack that killed 40 Indian security forces in Indian-administered Kashmir, when Indian warplanes crossed the Line of Control (LoC) and launched airstrikes near Balakot, Pakistan. Pakistan retaliated in kind, shooting down an Indian plane, capturing its pilot, and returning him several days later to terminate the conflict. This episode—the first time in history nuclear powers had exchanged airstrikes on their own territories—may also have triggered preparations for conventional missile strikes and the mobilization of nuclear assets.

The second development came months later when India revoked the Kashmir region’s autonomous status under Constitutional Articles 370 and 35a in what it described as an effort to curb terrorism and promote economic development and political integration. The sudden move drew a fierce Pakistani diplomatic response on the world stage and fears of yet another crisis.

These flashpoints exposed recurring challenges as well as shifting dynamics in Southern Asia. They raise pressing questions for analysts seeking to understand how recent trends in nationalism, the rise of China, the growing role of media in forming public opinion, and military modernization shape the escalatory pathways of—and off-ramps from—these contemporary crises in Southern Asia.

Workshop Details

From September 17-18, 2019, the Stimson Center’s South Asia Program hosted a workshop on crisis dynamics in Southern Asia with over 25 American, British, Chinese, Indian, and Pakistani analysts. The workshop included a simulation modeling India-Pakistan escalation dynamics and several panel discussions. In hosting this workshop, Stimson sought to pool the collective insights of leading South Asia analysts for debate and discussion centered on a central question: How have crisis dynamics in Southern Asia evolved over the last two decades, and what will the trajectories of crises look like moving forward?

Discussion and Findings

The discussions, held under the Chatham House Rule, focused on different dimensions of crisis dynamics in Southern Asia. The first two panels highlighted lessons learned in the first and second nuclear decades. Panelists found that decision-making in New Delhi, Islamabad, and Washington has not necessarily improved over time, with implicit biases, political disincentives, and divergent national priorities influencing states’ abilities to learn from crises and adapt. While India has exuded greater confidence over time, the country’s crisis response has varied across its regional relationships. Whereas India’s behavior towards China cycles between periods of assertiveness and restraint, New Delhi has incrementally employed greater stridency and risk-taking with Pakistan. As a result of this and other factors, participants agreed that crisis escalation across the sub-continent is becoming more likely.

The third and fourth panel discussions centered on some of the contributing factors for escalation: military modernization, enhanced nuclear readiness, and counterforce developments in Southern Asia—all of which appear to the state deploying them to lower the expected cost of conflict. Enhanced precision and standoff capabilities seduce leaders into rolling the dice with seemingly clean, low-cost options. Analysts warned that states’ attempts to modernize their militaries may not be successful, putting their leaders in the potentially dangerous position of assuming capabilities exist that are not fully developed. Technological advancements likewise shorten decision-making timelines, creating tighter windows for crisis management. Coupled with the United States’ and China’s evolving roles as de facto mediators, crisis management is growing more complicated, as exemplified during the February 2019 Pulwama-Balakot crisis.

The United States’ uncertain position as crisis and escalation manager was a key focus of discussion in the final panel, which focused on the role of third parties as they navigate strategic relationships across Southern Asia. Third parties’ roles are in flux, participants observed, with the United States playing a less active crisis management role and China remaining hesitant to become directly involved. Several factors, such as new alignment and strategic competition patterns between the United States, China, India, and Pakistan, have complicated mediation. Furthermore, South Asia falls short of at least two conditions cited for successful third-party mediation: that both parties desire a mediator and that the mediator be seen as impartial by both sides.

Unsurprisingly, there were several points of divergence among analysts over the course of the workshop. While participants agreed that prospects for stability in the sub-continent are worsening, they differed over the fundamental factors underlying this trend. For instance, the situation in the disputed Kashmir region emerged as a key point of tension running through the discussions. Analysts disagreed about India’s future objectives, the appropriate international response to rising tensions over Kashmir, and the likelihood that its new status could further stoke regional crises.

What follows is a detailed summary of each exercise and panel from the two-day workshop. This report seeks to capture the range of analysis and opinions—frequently convergent but sometimes divergent—expressed throughout the workshop.

We are grateful to workshop participants and attendees for sharing their analytical insights. We are also indebted to the Department of Energy’s National Nuclear Security Administration, the MacArthur Foundation, and the Carnegie Corporation of New York for their continuing investment in this research, without which this workshop and report would not have been possible.

Conflict Simulation Exercise

Stimson’s South Asia Program convened a simulation exercise on decision pathways to conflict escalation between India and Pakistan from asymmetric, to conventional, to nuclear levels. The participants took on the roles of strategic planners in the Indian or Pakistani governments doing an internal assessment of the probabilities of various escalation routes. In other words, they were asked to consider not what both sides should do, but rather what both sides would do. The goal of the simulation was to generate discussion of escalation and de-escalation pathways considering crisis triggers, strategy and doctrine, capabilities, force posture, and off-ramps.

The simulation generated three major observations:

  • History and precedent will factor into India’s forceful response to cross-border terror attacks. The India team predicted the Indian government would take previous attacks into account in calibrating its response to future attacks, such that casualty tolls of specific incidents might matter less than the overall total of previous attacks, a dynamic termed cumulative retribution. The international community’s tacit acceptance of Indian threshold-crossing in retaliation to attacks has also enabled its approach, under a dynamic of cumulative emboldenment.
  • Each side routinely miscalculates the other’s resolve and intentions. These mutual misperceptions increase the risk of escalation as both sides hold a false confidence in their understanding of the other’s red lines and likely behavior. 
  • The incompatibility of each party’s expectations of victory in limited conflict raises the risk of runaway escalation. Both sides’ implicit theories of de-escalation hold that their adversary will remain relatively restrained or back down, perhaps with the mediation of a third party. The India team exhibited overconfidence in the prospect of off-ramps in situations when the Pakistan team predicted conflict would escalate further, including with the potential use of tactical nuclear weapons against proactive operations.

Indian Cumulative Response Calibration

Early in the exercise, participants agreed that a new threshold had been set for terror attacks that trigger an Indian military response. While players determined that a crisis trigger on the order of the 2008 Mumbai attacks was unlikely to occur within the next year, they noted that the Indian government may perceive attacks on the scale of the 2016 Uri attack or the 2019 Pulwama attack to demand a more forceful response. This idea, termed “cumulative retribution,” 1 predicts that Indian responses to future attacks may take into account pent up desires to inflict punishment for previous attacks in which Indian leaders displayed restraint. Further, the participants agreed that Indian leaders may be willing to up the ante as a result of a parallel concept we describe as “cumulative emboldenment.” Especially in recent years, the international community’s response to Indian behavior has ranged from neutral to congratulatory, even as India crossed previous thresholds and tested boundaries. This may lead the Indian government to believe it has space to raise the bar without incurring significant international push-back in each subsequent crisis.

Throughout the simulation, the India team’s discussion focused primarily on domestic audience pressures, economic ramifications, and the “[public relations] value” of its behavior, while the Pakistan team focused primarily on escalatory potential. For example, the participants were asked to predict the likelihood that India, after a significant terror attack it attributes to elements of the Pakistani government, would opt for Cold Start versus “conventional lite”—the “low end” of its “Proactive Operations” doctrine that involves special operations and/or conventional forces that clearly do not threaten Pakistan’s territorial integrity. In terms of strategic objectives, the India team predicted that Indian government behavior would be shaped primarily by the “commitment trap” of domestic audience pressure rather than by escalation dangers. Indeed, one participant pointed out that limited conventional operations have not worked the way Indians have wanted in the past, so upping the ante in the future is a possibility. The Pakistan team was more focused on the escalatory potential of Cold Start, arguing that the move could trigger Pakistani battlefield nuclear use. There was also a discussion of the tradeoff between operational feasibility and strategic implications. Both teams questioned the ability of the Indian military to mobilize within 72 hours and execute Cold Start effectively; regardless of whether the Indian government might prefer to execute Cold Start, operational constraints could limit that option.

Incompatible Theories of De-escalation

Similarly, there was a stark contrast between the two teams’ predictions on the likelihood of escalation and de-escalation. The Pakistan team was more confident that both sides would escalate at any given level, while the India team believed off-ramps and de-escalatory measures would prevail (via escalation dominance). For example, participants were presented with a scenario in which India implemented a variant of Cold Start in response to a terror attack. With India having crossed two of Pakistan’s supposed nuclear “redlines” by seizing significant territory and seriously degrading Pakistan’s military forces (the spatial and military thresholds), analysts on the Pakistan team were confident that India’s actions would trigger Pakistani nuclear use under the “last resort” scenario—battlefield or strategic. The Indian side, in contrast, believing their offensive was bounded and still limited, thought it likely that Pakistan would pursue de-escalation and saw unauthorized use as the more probable (though still remote) route to nuclear use.

Further, when given the scenario of Pakistani deployment (not use) of battlefield nuclear weapons, the India team considered conventional counterforce capabilities against Pakistani nuclear forces and supporting structures to be plausible. Similarly, in the case of Pakistani battlefield nuclear use, the Pakistan team predicted that the Indians would conduct counterforce attacks rather than pursue an off-ramp. Conversely, the India camp saw off-ramps at this stage as extremely likely. Despite India’s doctrine of massive retaliation, the group was confident that this outcome was improbable.

The significant difference of opinion over the probability of an off-ramp may have come from disagreements on when third-party mediation was most likely to take place. While the Pakistan team seemed to agree that third parties have greater chances of preventing escalation early on than later in the conflict, the India team believed that both sides may actually be counting on third-party intervention once the war is hot.

The divergence between the teams’ views on escalation probability likely stemmed from differing perspectives on Pakistan’s strategic objectives and operational plans. The India team agreed Pakistan’s goal in a limited war with India would be to seek conflict termination. Indeed, Pakistan’s conventional inferiority makes any protracted conflict much more costly for the Pakistanis. Thus, the India team was relatively confident in off-ramps and de-escalation. On the other hand, the Pakistan group saw their government’s objective as punishing and embarrassing India, causing them to be more confident that the crisis would escalate.

Probability estimates on operational plans were also noteworthy. In line with its consensus that Pakistan would be most likely to seek de-escalation, the India team believed that Pakistan would want to create some semblance of equivalency—not upping the ante and instead mirroring Indian behavior as a route to de-escalation. One participant on the Pakistan side noted that Pakistan’s response options were extremely restrictive—they were either not feasible (a ground operation), too insignificant (cross-LoC firing), or too escalatory (deployment of tactical nuclear weapons). The Pakistan team assessed that, in terms of operational plans, Pakistan’s cross-LoC options are less effective than India’s. This has practical implications because, despite the India team’s prediction, Pakistan’s ability to respond in a seemingly equivalent manner is in fact limited.

Misperception and Miscalculation

Each side’s misperception and miscalculation of the other’s resolve and capability came through when both teams were asked about relative cost imposition. The India team believed that India would seek to impose high costs on Pakistan, while Pakistan could only impose low costs on India. This fed into an Indian perception of escalation dominance and therefore control. The Pakistan team, however, suggested cost imposition would be symmetrical—i.e. that Pakistan and India would exact relatively even costs during the conflict. This view lends itself to stalemate, which provides some stability but introduces risks of inadvertent escalation. This misperception carries risks. If India is confident in escalation dominance, it is then more likely to partake in escalatory behavior, believing Pakistan will most likely de-escalate. But if Pakistan perceives that it can impose a symmetrical cost on India—and thus claim victory considering its conventional and strategic inferiority and ability to manipulate domestic public opinion—then it may not seek to de-escalate.

The key points of agreement and disagreement that emerged throughout the simulation largely revolved around what one participant termed the “move three dilemma.” After the conflict trigger—in this case, a terror attack—India will retaliate (move one) and Pakistan will respond in kind or scale (move two). Move three is where the crisis will either escalate into general war or de-escalate through an off-ramp. This stage of the crisis was less predictable for either side and was a subject of debate both within and between the teams. Perhaps tellingly, Indian participants rejected the framing of India’s retaliation as the first move, arguing that the original trigger event, a terrorist attack carried out by Pakistan-backed militant groups, constituted “move one.” This disagreement reveals the fundamentally different understandings of crisis escalatory dynamics held by both sides and emphasizes the challenge of preventing escalation when neither side can agree on who went first, and who could, thus, make the last move before a potential off-ramp.

Workshop Panels

Panel I. Organizations, Decision-making, and Risk in the First Nuclear Decade:
Lessons Learned from Kargil, to Twin Peaks, to Mumbai

The first panel discussed the lessons learned in India, Pakistan, and the United States in the first nuclear decade, between 1999 and 2008. Panelists were asked to consider the following questions:

  • What key lessons were learned—and left unlearned—in Delhi, Islamabad, and Washington in the wake of the three crises in Kargil (1999), Twin Peaks (2001-2), and Mumbai (2008)?
  • What is the most underappreciated factor (cause, dynamic, or consequence) from South Asia’s first nuclear decade?
  • Do internal challenges with strategic assessment, organizational competence, and bad decision-making pose the same problems for governments today as they did between 1999 and 2008?
  • Was Pakistan intentionally or unintentionally running nuclear risks during the first nuclear decade?
  • Is a territorial fait accompli-like operation or Mumbai-like operation possible today, and what would be the consequence?

Three key observations emerged from this panel discussion:

  • Learning from crises by New Delhi, Islamabad, and Washington has been inconsistent. In each of the countries in question, state learning has been highly uneven in scope and quality, conditioned by unintentional biases, deliberate political choices, and state priorities.
  • Learning can generate instability rather than stability. Learning has sometimes incentivized belligerence rather than restraint, enhancing instability well into the second nuclear decade.
  • Despite learning, state decision-making has not necessarily improved. While the lenses through which states view crises have shifted from the first nuclear decade to the second, state decision-making during crises did not necessarily improve between 1999 and 2008.

States’ Ability to Learn from Crises

Each country—as well as distinct constituencies within each country—drew different lessons from the crises of this period, participants concluded. Several argued that state learning in general, diffuse as it may be, is not only conditioned by psychological, ideological, and institutional biases, but can also be actively circumscribed by deliberate political choices. Decisions made by leaders and their administrations reflect their personal perspectives, ideologies, and decision-making structures. Their experiences in initial crises color subsequent actions as personalist decision-making structures can outweigh or supplant formal processes. Analysts agreed that as much as states may learn from the crucible of experience over time, successive leaders and decision-makers—often with differing points of view and strategic priorities—must choose to build upon the lessons and learn from the mistakes of their predecessors. All too often this learning does not occur or is circumscribed.

Building from this conversation, participants argued that lesson retention has been a problem in Southern Asia as well as the United States. Analysts pointed out that, in the wake of the first nuclear decade, Pakistan’s government began to believe in nuclear immunity: the notion that Pakistan’s nuclear weapons program saved it from Indian conventional retaliation during crises. Beyond that major lesson, however, very little nuclear learning took place or could take place. Some participants noted that transparent analysis within Pakistan of the state’s actions and blunders (Kargil being the most notorious of the latter) and the ability of civilian authorities to demand accountability for the military’s decisions were and continue to be severely circumscribed. Two decades after the Kargil War, according to some attendees, decision-making processes in Pakistan remain relatively ad hoc, and even its war and nuclear doctrines at times appear fuzzy and incoherent. Coupled with widespread conceptions of nuclear immunity, this now leaves Pakistani decision-makers foundering in unchallenged interpretations of their own history and their options in a future crisis.

In India, meanwhile, participants noted that the challenge of learning from crises of the first nuclear decade is compounded by competing power centers within the bureaucracy and military. Civilian leadership typically relies on advice from experienced bureaucrats when responding to crisis situations, giving civil servants the ability to shape responses and outcomes. While a useful method for incorporating experienced voices into the policy process, it complicates efforts to establish the level of accountability that is central to learning from past strategic errors.

Challenges with processing and learning from previous crises can likewise be problematic for the U.S. government, despite its relatively more transparent decision-making procedures. Lesson retention has been subject to presidential and bureaucratic administrations’ conscious desire to absorb—or buck—the approach of their predecessors. Some analysts suggested that the Trump administration may have resisted—at least initially—applying lessons from its predecessors during the 2019 Pulwama-Balakot crisis.

The Impact of Nuclear Learning on Response Templates

Even when governments learn from crises, their changed behavior may not always enhance stability. Participants teased out this idea by examining Indian Prime Minister Narendra Modi’s decision-making calculus compared to that of his predecessors. Where Indian leaders of the first nuclear decade prized a more restrained approach in the face of crises, the new era has seen a stronger focus on retaliation and score-settling. Today, some analysts argued, the bias leans towards action, even if for political optics more than for strategic effect. As such, Modi’s increasingly aggressive moves in recent crises seem to be building a new Indian response template that may risk greater escalation in the future. From one analyst’s perspective, however, this aggressiveness follows logically from the lessons that India drew in the first and second nuclear decades in which it repeatedly suffered major terrorist spoiler attacks just as peaceful outreach to Pakistan intensified. This pattern generated learning among Indian leaders that any future engagement with Pakistan would only be exploited by spoiler attacks—a conclusion that has generated an increasing appetite for retribution that is shared by the Indian public. In turn, several attendees were concerned that Pakistan, considering its tendency to dismiss Indian retaliation as motivated by domestic pressures rather than a strategic logic vis-à-vis Pakistan, may further reify its perception of its nuclear immunity despite a more aggressive Indian response template.

Participants also focused on shifting crisis perceptions in the United States. Some argued that, in South Asia’s first nuclear decade, Washington viewed India-Pakistan crises primarily through the lens of nuclear risk and escalation management, informed by experience from the Cold War and recent overt nuclearization. Following the 9/11 and Mumbai 26/11 terror attacks, however, this perspective shifted in the second nuclear decade to frame crises as issues of cross-border terrorism. Some mentioned another key shift in the second nuclear decade: while Washington initially aligned closely with Pakistan for support in Afghanistan, the United States subsequently pivoted towards India anticipating its emerging economic and strategic influence. While these changes have shaped U.S. priorities significantly, they have not necessarily improved U.S. decision-making between the first and second nuclear decades. Concurrently, analysts argued, the shift in U.S. perceptions away from the first decade’s discourse of nuclear risk could inadvertently distract U.S. crisis managers from dangerous developments including the future early deployment of readied nuclear weapons in a crisis, as India’s claim of deploying nuclear-powered ballistic missile submarines demonstrated during the Balakot crisis. This served as a sobering reminder that states and decision-makers could draw problematic lessons from the evidence before them and from the experience they accrue.

From the First Nuclear Decade to Future Dangers

Participants agreed that, in some ways, we are now in more dangerous territory than we were between 1999-2008, even as they differed on the impact of issues such as the regional implications of India’s August 2019 revocation of Constitutional Articles 370 and 35A and the likelihood of India pursuing territorial gains in Pakistan-administered Kashmir. However, as one scholar put it, the first decade of nuclear learning in South Asia was about developing conceptual understandings of strategic interests, and indeed, of the dynamics of war, to prevent future crises from escalating. Two decades and several crises since Kargil, most analysts agreed that the states’ desires for the same political and strategic ends have not diminished. Rather, greater sophistication of means, namely their advanced weapons systems, and greater confidence in their ability to calibrate offensive actions to avoid escalation have only intensified regional dangers.

Panel II. Nationalism, Media, and Public Opinion in the Second Nuclear Decade:
Making Sense of Uri, Doklam, and Balakot

The second panel discussed how India and Pakistan’s crisis behavior changed in the second nuclear decade as each continued to develop capabilities, formulate the role nuclear weapons would occupy in defense and security strategies, and learn from crisis episodes with the other. Participants also considered how nationalism, traditional and social media, and leaders’ agency have impacted crisis dynamics on the sub-continent. Panelists were asked to consider the following questions:

  • What are the key similarities and differences between crises in Uri (2016), Doklam (2017), and Balakot (2019)?
  • How does the Modi government’s approach to inter-state relations and crises differ from past administrations’?
  • What is similar or different between India-Pakistan and Sino-Indian border crises?
  • How new and problematic are the roles of nationalism, (social) media, and public opinion? Are they real constraints/drivers or convenient deflections?
  • In response to a future provocation, is there room for Modi to back up or back down, or is he locked into a commitment trap?

Three key observations emerged from this panel discussion:

  • “Strategic restraint” is a thing of the past. Indian strategic restraint has gradually been diluted in the second nuclear decade, whether in terms of responding to cross-border terrorism emanating from Pakistan or reacting to Chinese adventurism on the border.
  • India’s crisis behavior differs towards Pakistan and China. Despite India’s broadly increasing assertiveness, there is a fundamental difference in its crisis response and behavior with Pakistan as compared to China. With Pakistan, India is wont to employ more risk-taking strategies, and its actions are aimed at coercing Islamabad to change its proxy war tactics. With China, in contrast, India’s actions are aimed at terminating the crisis quickly and in a way that leaves space for dialogue and accommodation. 
  • Deliberate crisis escalation may become more frequent. New Delhi’s new strategy of manipulating risks is likely to make future crises between India and Pakistan more dangerous.

India’s Waning Strategic Restraint

Analysts agreed that India appears to be shedding its strategic restraint in response to cross-border terrorism from Pakistani territory. Several pointed to India’s surgical strikes along the Line of Control with Pakistan after the Uri attack in September 2016 and the Balakot airstrikes after the Pulwama attack in February 2019 as indicators of this evolution. More broadly, some participants argued that New Delhi’s sudden withdrawal of autonomy provisions in Indian-administered Kashmir in August 2019 offered further evidence of restraint’s demise.

Analysts were divided, however, on the underlying factors dictating this newfound Indian assertiveness. One school attributed it to the leadership of Indian Prime Minister Narendra Modi and the identity politics and nationalist ideology of his ruling Bharatiya Janata Party government. Another school posited that India’s increasingly aggressive response to recent provocations from Pakistan is a result of cumulative emboldenment—the sense that New Delhi feels empowered by the lack of significant international push-back against actions that cross salient thresholds, such as the Balakot airstrikes.

Participants also wrestled with whether and how domestic audience pressures have played into India’s more muscular response to provocations from Pakistan. Some argued that Modi has locked himself into a commitment trap by making national security an election issue, rendering any potential return to strategic restraint a risky electoral strategy. Others argued that the Modi government maintains agency in choosing appropriate responses based on its ability to shape public opinion and social media reactions to its advantage, but to date has shown no inclination to adopt a more restrained posture. One analyst contended that a commitment trap is only a trap if you don’t seek to follow through on the commitment. 

India’s Approaches Towards Pakistan and China

While participants agreed that India’s assertiveness was also evident in its bold course of action towards China during the Doklam standoff of summer 2017, several pointed out that the country calibrates its approach towards crises with Beijing differently than with Islamabad. In contrast to its publicized Uri surgical strikes and Balakot airstrikes against Pakistan, Modi’s government handled the Doklam standoff in a much more subdued manner, as one participant pointed out. Another added that after Doklam, India made the political decision to avoid further crises with China and sought a modus vivendi through the Wuhan summit. India’s responses to Pakistan, however, grew increasingly bellicose with crises from Uri to Balakot.  

New Delhi’s differing approaches towards Islamabad and Beijing, one participant argued, should be seen in the context of three global events: 1) U.S. retrenchment, 2) the rise of populist governments worldwide, and 3) the rise of China. As the war-weary United States begins to retreat from the global stage and nationalist governments flourish, India has seized opportunities to take unilateral actions to secure its interests and respond more muscularly to both of its neighbors without opprobrium from the international community. However, the analyst added, because India appreciates its military imbalance with a rising China and its dependence on Chinese trade and investment for economic development, New Delhi has carefully calibrated its response toward Beijing accordingly. India’s conventional superiority over Pakistan, in contrast, has had the opposite effect.      

Relatedly, several participants argued that Beijing had also recalibrated its policy towards India as a result of New Delhi’s bold action during the Doklam standoff. One analyst explained that, though China has traditionally seen India as a less developed country and thus a low-level threat, Chinese officials determined they had underestimated India in the aftermath of its assertive action in Doklam. Going forward, Beijing is likely to both continue supporting a weaker Pakistan to maintain balance of power in the region while also seeking to cooperate with India more broadly against the Global North. Chinese fears about New Delhi aligning with the U.S. Indo-Pacific strategy will likewise lead to further Chinese concessions towards India. One such concession was evident when China removed its technical hold on the UN’s terrorist designation of Masood Azhar, the mastermind of the Mumbai terror attacks.

Future Challenges 

Several analysts agreed that New Delhi, by revoking Article 370, aimed to make Pakistan irrelevant to the Kashmir dispute. One warned that India’s objective was to make the Line of Control the international border and force Pakistan to accept these “facts on the ground” over time. If this is the case, analysts expected many more crises to befall India-Pakistan relations as India’s new strategy of manipulating risks is likely to contribute to further escalation. Multiple participants agreed that the lesson India took from the Balakot airstrikes was that it could get away with significant cross-border conventional operations without risking nuclear escalation. This strategy assumes, however, that India would achieve escalation dominance, which it does not yet possess. Some analysts argued that India’s risk manipulation strategy would likely incentivize Pakistan to escalate crises more quickly rather than backing down, setting the stage for more intense clashes. The result could be re-hyphenation of India and Pakistan—the opposite outcome of what India had set out to achieve—and more dangerous regional crises.

Panel III. Military Modernization, Readiness, and Doctrine:
Impacts on Deterrence and Crisis Stability

The third panel sought to unpack the drivers of military modernization in South Asia and the relationship between modernization, crisis behavior, and strategic objectives. The panel paid particular attention to the challenges, risks, and consequences of these developments in terms of strategic and crisis stability. Participants were asked to consider the following questions:

  • What are the important conventional or strategic changes underway in the Indian, Pakistani, and Chinese military capabilities in the region, if any? Is this military modernization or arms racing?
  • Will these changes bring greater deterrence stability or instability over the next decade?
  • Will these changes bring greater crisis stability or instability over the next decade?
  • Will greater standoff, speed, and precision capabilities have an impact on decision-making or state interactions?
  • Are military buildups more important than territorial disputes in fueling rivalry and tensions?

Three key observations emerged from this panel discussion:

  • There are significant pressures for counterforce development in Southern Asia. These include: 1) trilateral cascade pressures between China, India, and Pakistan; 2) first mover incentives; and 3) Indian nuclear deterrence credibility.
  • Military modernization efforts do not ensure modernization success. Counterforce capabilities are particularly difficult to master. Partial or unsuccessful counterforce capabilities put a state in a worse position than having no counterforce at all by incentivizing preemption and inflating false optimism, leading to risky decision-making.
  • Prospects for stability are declining. While analysts diverged on the relationship between modernization and crisis behavior, they agreed on the bottom line: prospects for stability are diminishing, and new capabilities are likely to intensify this trend.

Drivers of Modernization

Analysts found that there is a cascading dynamic of external pressures driving military modernization in Southern Asia. One participant argued that the trilateral dynamic among China, India, and Pakistan makes the modernization arms race even more dynamic than in other contexts, as it pressures conventionally weaker powers like Pakistan to offset the capability gap with nuclear weapons and asymmetric warfare. Another analyst pointed out the risk of first mover incentives: since New Delhi and Islamabad are only 429 miles apart, preemption, decapitation of command-and-control centers, and pre-delegation of nuclear launch authority will always appear attractive due to time constraints.

Participants agreed that counterforce pressures are particularly prevalent in India. Because of its longstanding (though now-weakened) no first use (NFU) nuclear doctrine, India faces a stark dilemma when it comes to deterring Pakistan from using battlefield nuclear weapons on its own territory against a potential Indian conventional assault. Pakistan’s reliance on tactical nuclear weapons thus creates pressures for India to develop more advanced capabilities to preempt and limit damage. Pakistani leaders have publicly accused Indian leaders of bluffing when they threaten nuclear use against a Pakistani city in response to Pakistani tactical use on its own territory as called for under the massive retaliation doctrine. Deterrence suffers without threat credibility.

Strategic Ends

The group contemplated the relationship between modernization, crisis behavior, and strategic objectives in the context of the Pulwama-Balakot episode. One participant argued that the crisis portrayed “resolve without results” in that each side’s new capabilities—standoff, precision-guided munitions, net-centric warfare, etc.—did not function as anticipated. Another analyst held that Indian conventional capability development and employment is geared towards its strategic objective of creating space to retaliate against Pakistan without crossing the nuclear threshold. Others, however, argued that, unless the objective is to satiate a domestic audience, Indian behavior is inherently non-strategic. More advanced arms that create space to retaliate without crossing a nuclear threshold are unlikely to punish Pakistan enough to coerce a change in its behavior. This approach also works against Indian interests by re-hyphenating the India-Pakistan relationship, leading to potentially escalatory crises and putting the leadership in a commitment trap. The participants agreed that, while previous Indian administrations saw strategic restraint as a guiding objective, the Modi government appears to have shifted in both its approach and its strategic objectives.

Challenges, Risks, and Consequences

Another theme throughout the conversation was the difficulty of operationalizing counterforce capabilities. In other words, just because modernization is underway does not mean it will be successful. Participants pointed out that modernization requires not only acquisition, but advanced software, referencing intelligence, surveillance, reconnaissance functions, and intelligence fusion as examples of necessary but difficult capabilities to master. This means that states may still be willing to test adversaries who possess visible counterforce hardware because they discount their opponents’ software—i.e. the extent to which a counterforce attack could be successfully carried out based on training, systems integration, information sharing, and other factors.

Furthermore, analysts agreed that partial or unsuccessful counterforce capabilities put a state in a worse position than not having counterforce capability at all. These capabilities could force a state like India into a use-or-lose dilemma and create pressures for preemption in a crisis. Because of the ever-present security dilemma, such capabilities could also be read as indicators of malicious intentions, therefore intensifying arms races. Nascent Indian ballistic missile defense (BMD), for example, provides little to no strategic benefits but has triggered development within China and Pakistan of new missiles and countermeasures designed to defeat BMD.

The group further discussed whether modernization might create false optimism in terms of capability and escalation controllability among state leadership, leading to risky decision-making. One analyst argued that Pakistan’s tactical nuclear weapons invite preemption, complicate command-and-control, and make movement difficult. Pakistan may also be unprepared to operate on a nuclear battlefield. Indian modernization, combined with its conventional superiority vis-à-vis Pakistan (and perception of escalation dominance), may instill overconfidence in escalation controllability in the event of a crisis.

Finally, the group agreed that modernization also produces more opportunities for reliability failures and nuclear accidents. Managing a nuclear arsenal—command-and-control, nuclear security, and nuclear surety—requires humans to operate these systems, and humans are fallible. To the extent that modernization is pursued, and especially the development of counterforce capabilities, control is likely to be distributed to more individuals overseeing more advanced technology with fewer built-in checks and balances. Several analysts noted that the United States has endured several close calls related to systems and human error, and such risks in South Asia should not be overlooked.

The conversation ended on a cautionary note. One analyst warned that the norm of non-use, the centerpiece of what is left of the global nuclear order, is eroding. At the same time, however, South Asian leaders have shown extraordinary restraint when it comes to non-use. A few participants stressed that restraint must not be taken for granted, as no nuclear state has ever been put in a position in which it believes its survival is at stake, with the possible exception of Israel during the 1973 Arab-Israeli War. Modernization—combined with vulnerable geography, close proximity, and reduced decision times—threatens to create scenarios in which nuclear-armed states feel less and less secure. Relying on the restraint of individual leaders may not be enough to prevent these crises from going nuclear.

Panel IV. Conflict Escalation, De-Escalation, and Termination: Prospects and Challenges

The fourth panel centered on the significance of the Balakot crisis, conflict management, and prospects and mechanisms for de-escalation in Southern Asia. Panelists were asked to consider the following questions:

  • How might you imagine a war breaks out and evolves between India-Pakistan or India-China? (Or India-China/Pakistan?) Could you imagine any non-traditional pathways?
  • What are the key variables that shape the escalatory potential of an emerging conflict: geography, domain, casualties, overt/covert, domestic/international political context, etc.?
  • Will potential conflicts be short, protracted, intense, or mild?
  • What are the routes for de-escalation or termination after a conventional war breaks out, and when are they likely to emerge?
  • Have de-escalatory pathways changed since previous conflicts. If so, why?

Three key observations emerged from this panel discussion:

  • Balakot was a watershed moment. Many argued that both India “breaking the Kashmir taboo” by striking targets in the Pakistan mainland and the United States stepping back as crisis manager may have changed expectations for escalation—and perceptions of risk manipulation possibilities—in future regional crises. 
  • The era of extended crises is over. India and Pakistan are more likely to experience a short rather than protracted crisis. New technologies including unmanned aerial vehicles, cyber capabilities, and long-range missiles have made decision-making timelines shorter, meaning there is a more limited timeframe in which crisis de-escalation can occur.
  • Future crises are less prone to “management.” Participants questioned the changing roles and impact of third-party crisis managers in the region, including whether the United States acted as a conflict enabler or manger during the Pulwama-Balakot crisis and the potential future role of China. Many saw the United States as stepping back from crisis management, though it remains the key third-party powerbroker in the region.

Balakot’s Lasting Impact

Participants agreed that the Balakot airstrikes within Pakistan proper (beyond Pakistan-administered Kashmir) and relative lack of U.S. involvement made Balakot fundamentally different from past crises. Some questioned the degree to which “luck” was involved in de-escalation and the precedents this crisis set for India and Pakistan in terms of the rules of engagement, dynamics of deterrence, and risk manipulation.

Most analysts agreed that the Balakot crisis could not have followed the path it did to de-escalation without a certain amount of “luck,” and several analysts commented that the captured pilot provided a fortuitous off-ramp. One analyst countered that such a reading discounts the decisions made by both sides, arguing that luck was perhaps necessary but not sufficient for de-escalation. In this telling, de-escalation followed a conscious calibration by both sides after an assessment that each had gotten what it wanted: India carried out military operations to punish Pakistan and re-establish deterrence while Pakistan acquitted itself militarily and gained from a magnanimous gesture. The remaining risk is inadvertent escalation—a risk that may be magnified in future crises if the lessons both sides take from Balakot are successfully manipulating risk, controlling escalation, and claiming victory for domestic consumption.

Participants debated the extent to which each side was able to control the escalation of the crisis. As one analyst argued, the captured pilot was only one instance of “luck” during the crisis. Had the Balakot airstrikes led to the death of civilians or had the pilot died after being captured by civilians in Kashmir, the crisis may have had a different outcome. Moreover, emerging developments in doctrine and posture, including India’s seeming ambiguity over NFU and Pakistan’s growing reliance on tactical nuclear weapons, threaten deterrence stability and risk dangerous escalation in future crises. Given the extent to which contemporary and future crises are likely to be incited by proxies and disaffected populations, several analysts argued that de-escalation is likely to be increasingly challenging under the charged context of India’s recent abrogation of Kashmir’s autonomous status and the resulting outcry in Pakistan.

A Shortened Crisis Management Timeline

Analysts agreed that technological advancements and the speed of (dis)information have made opportunities for de-escalation or third-party involvement much more limited than in previous crises. One analyst argued the window for crisis management in a Cold Start-style military engagement would be as short as 48 to 72 hours before the conflict threatened to encroach on nuclear thresholds and trigger uncontrolled escalation.

The shifting geographies of conflict contribute to this shortened timeline. A conflict that follows the Balakot precedent and reaches targets outside of the disputed Kashmir region could be more unpredictable due to the absence of equivalent target sets or shared understandings of reciprocal exchanges and tacit bargaining. While a crisis’s duration is likely to be shorter given limits on munitions, crises may be more intense in the future because they involve probing for new thresholds that are not clearly established or salient. As dual-use and nuclear platforms are readied for use in the battlespace, disinformation and public messaging could induce panic or anticipation of rapid use, which could intensify escalation. Degeneration of forces will also be a challenge for de-escalation as more intense future crises will likely involve significant movement of materiel and/or personnel that will have to be withdrawn in the aftermath of a crisis.

Fluctuating Crisis Management Roles

Analysts agreed that the international context of crises in South Asia is changing. While many participants pointed out that the United States had taken a step back during the Balakot crisis, some argued that Washington had enabled the crisis by seeming to embolden India. One analyst commented that the United States had effectively given Modi the “green light” to escalate, both in Balakot and in potential future crises, through public and reported private statements. Participants debated whether U.S. involvement was inherently positive, with one analyst noting the moral hazard problem of Pakistan and India overestimating U.S. willingness to intervene in a crisis before it had escalated out of control. Participants perceived China, meanwhile, as increasingly drawn into a crisis management role while lacking the full capacity and willingness to step in when required.

Analysts reiterated the importance of past de-escalation measures such as enacting travel advisories, removing non-essential personnel, issuing travel warnings to limit business travel, and spreading information on the impacts of nuclear use. Regarding India-Pakistan bilateral mechanisms, establishing an additional backchannel for Indian and Pakistani military and civilian representatives to use during crises could be another means of de-escalation. While noting the importance of such bilateral approaches, analysts were pessimistic that such mechanisms would be adopted in the near term.

Panel V. Third Parties, Confidence Building, and the Future of Territorial Dispute

The fifth panel discussed the role of third parties in Southern Asian crisis management. Panelists were asked to consider the following questions:

  • What is the role of third parties in conflict management and resolution? Has this shifted?
  • Are confidence building measures (CBMs) possible? Which parties must be involved?
  • Can the region tolerate the continuation of indefinite territorial disputes? Is ugly stability possible into the future?
  • Does instability really threaten any state’s core equities? Does instability serve anyone’s interests?

Three key observations emerged from this panel discussion: 

  • China will be an increasingly relevant third party in South Asia. While the United States should aim to establish channels for dialogue between India and Pakistan, China can help reinforce a policy of restraint in Islamabad.
  • Asymmetries complicate third-party mediation. Asymmetries between India and Pakistan, in terms of power, perceptions of the status quo, institutional architecture, and relations with the United States, have made meaningful third-party mediation nearly impossible, at least in the short term.
  • South Asia does not meet the preconditions for successful third-party mediation. The three preconditions for third-party mediation are: 1) both parties must seek it, 2) third parties must have sufficient leverage, and 3) the third-party mediator must be relatively impartial to both sides. The second precondition may be met by the United States, but  the first and third are clearly lacking.

The Changing Nature of Mediation

Participants discussed the nature of third-party mediation and intervention in the event of a crisis in South Asia, agreeing that the ever-present shadow threat represented by India and Pakistan’s nuclear capabilities dramatically increases the stakes of any regional confrontation. Third parties—most notably the United States and China—have two principal crisis management roles. Wielding international influence, third parties can encourage preemptive measures to prevent a crisis scenario from unfolding. Third parties can also intervene through hard and soft power channels to de-escalate a developing crisis scenario. Participants concluded that persisting asymmetries between India and Pakistan, the regional influence of international terror cells and non-state actors, the U.S. tilt towards India, and enduring Chinese influence in Pakistan create an inhospitable climate for mediation.

In the short and medium terms, the international community is well-served having the United States and China managing crises in South Asia. In the long term, however, external interventions could prove counterproductive to the extent that they disincentivize potential bilateral CBMs. Even in the absence of CBMs or meaningful dialogues, there may be some scope for self-interested, unilateral but coordinated arms control measures. There may also be opportunities of leverage to push for accommodations by India and Pakistan, including India’s desire to host a successful G20 summit in 2022.

The United States and China’s Role

Each participant described the United States and China as the two likely and viable third-party actors and noted that both could play a role in crisis management rather than conflict resolution. Analysts largely agreed that the United States lacks a long-term policy vision for the region. The United States’ most significant regional intervention has been in Afghanistan, and it continues to monitor terrorist activity across the Pakistan-India and Pakistan-Afghanistan borders. Participants disagreed somewhat on the United States’ role in recent confrontations between India and Pakistan. One analyst noted that, despite public narratives describing the United States’ diminishing role and China’s increasingly expansive one, Washington may have played a more active role in defusing the Balakot crisis through private channels than was initially acknowledged. Another analyst argued that even while the United States helped de-escalate the February 2019 crisis, other allied governments have assessed that early U.S. actions and statements, including the perceived “green light” for retributive action, had a permissive effect on crisis escalation.

While the participants agreed that the United States could attempt to rekindle dialogue between India and Pakistan, they remained pessimistic about the United States finding the appropriate channels and opportunities to do so. China, more entangled in regional great power dynamics, could help enforce a policy of restraint in Islamabad, such that limited confrontations do not risk the use of nuclear weapons. Such an approach would require overcoming the increasingly competitive and adversarial nature of the U.S.-China relationship to initiate a cooperative dialogue with Beijing that works to both prevent as well as defuse South Asian crises. Given the United Kingdom’s stakes in the region, it is possible the former colonial power could also play a supporting role in that dialogue.

With respect to long-term, third-party mediation, all analysts agreed that the United States and China ought to take steps to encourage existing channels for dialogue and focus on the region’s future stability. At least in theory, there may be an appetite for dialogue among some constituencies in both India and Pakistan. Third parties might observe regional dynamics to see how the revocation of Article 370 affects Pakistan’s negotiating parameters moving forward. To break the stalemate, it might be necessary to re-center the dialogue on counterterrorism challenges from groups representing a threat to both sides.

Relationship Asymmetries

Asymmetries pose significant roadblocks to successful dialogue and mediation. Analysts suggested that India has defined its identity as one that is escaping hyphenation with Pakistan. In fact, while it is generally assumed all states are interested in strategic stability, several argued that India’s incentives and ambitions may not be congruent with such balance. India’s civilian-led government also struggles to find appropriate interlocutors/channels for trusted, meaningful dialogue and bargaining mechanisms with a de facto military-led government in Pakistan. Furthermore, Pakistan still faces a principal-agent problem wherein its non-state actors cannot be fully controlled. To the extent that militant groups view the Pakistani state as insufficiently able or willing to act, particularly in light of India’s recent actions in Kashmir, they might be tempted to take action on their own accord. One analyst noted that the extent to which Lashkar-e-Taiba finds itself in the “decider” seat in this context is particularly troubling. Third, with India’s August 5 move in Kashmir, there is a new asymmetry in the status quo. India is likely to define this “new normal” as the starting point for any potential negotiations, while Pakistan’s narrative of 70 years depends on resisting it. Finally, U.S. strategic and economic interests have undoubtedly tilted towards India as Washington depends less on Pakistan to fight al-Qaeda, draws down its presence in Afghanistan, and elevates major power strategic competition above counterterrorism. This may result in greater deference to Indian prerogatives.

Preconditions for Third-Party Mediation

According to one analyst, three preconditions are necessary for a successful third-party mediation: 1) both conflicting parties must want it, 2) the third party/parties must possess leverage, and 3) the third-party mediator(s) must be relatively impartial to both states. Participants agreed that the first and third preconditions are absent in South Asia, and that this presents both an immediate and long-term challenge to mediation. India generally eschews third-party involvement, and one analyst noted it now seeks to more aggressively veto it.

Even if the conditions were met, another analyst noted that the academic literature on third parties does not inspire confidence. Third parties may be able to stop violence in the short term but generally do not resolve long-term disputes. Third-party mediation also generally achieves less success at the interstate level than at the intrastate level during civil conflicts. Moreover, both India and Pakistan seem to have mobilized constituencies tolerant of high-cost crises, rendering the stakes “indivisible”—all or nothing—and thus less susceptible to a mutually accommodative settlement. Finally, in the event of renewed tensions, the fog of war makes even the best-laid plans for mediation challenging. Nevertheless, the medium-term costs of instability to the Pakistani economy and the long-term costs of instability to Indian society and its international image could be quite dire.

Long-term CBMs would depend on toning down the nationalistic rhetoric that has made both India and Pakistan relatively impervious to productive third-party mediation, decreasing the influence of violent non-state actors, and building working channels for intraregional dialogues moving forward.