For four days in May 2025, two nations armed with thermonuclear warheads fired missiles at each other’s territory, scrambled fighter jets into contested airspace, exchanged drone strikes across the Line of Control, and killed each other’s soldiers and civilians in sustained combat operations, all while possessing arsenals capable of incinerating each other’s cities within thirty minutes of a launch order.

The fact that this sentence can be written as a description of real events rather than a hypothetical scenario represents the single most consequential development in strategic theory since the Cuban Missile Crisis of 1962. India and Pakistan’s May 2025 conflict, triggered by the Pahalgam tourist massacre and culminating in Operation Sindoor’s twenty-three-minute missile barrage, did something that six decades of deterrence scholarship insisted could not happen: it produced sustained conventional warfare between two states possessing thermonuclear weapons. The theoretical literature on deterrence, from Bernard Brodie’s foundational 1946 argument that the atomic bomb had made war between great powers obsolete to Kenneth Waltz’s 1981 claim that proliferation produces stability, rests on a single foundational assumption. That assumption holds that rational state leaders, faced with the possibility of atomic retaliation, will always prefer restraint over escalation. The 2025 conflict did not merely test this assumption. It shattered the assumption’s claim to universality, forced deterrence theorists to confront a case they had dismissed as implausible, and left the international security architecture scrambling to understand why two rational governments chose to fight a war that carried a nonzero probability of annihilation.

The analytical question this article answers is precise: does the 2025 India-Pakistan conflict prove that deterrence works, because escalation stopped short of weapons use, or does it prove that deterrence is failing, because conventional warfare between atomic-armed states was supposed to be prevented entirely? The answer, as this analysis will demonstrate through a systematic comparison with four previous crises and a rung-by-rung mapping of the 2025 conflict against the theoretical escalation ladder, is that both positions contain truth but that the more consequential finding is the second. Deterrence held at the weapons-use threshold but failed catastrophically at the conventional threshold, and the conventional failure exposed structural weaknesses in deterrence theory that no amount of post-hoc rationalization can paper over.

The Framework That Broke: Deterrence Theory Confronts Reality

Before mapping the 2025 conflict against previous crises, the theoretical framework itself requires examination, because the theory’s internal logic determined what analysts expected to happen and explains why reality’s departure from those expectations is so consequential.

Modern deterrence theory rests on four interlocking propositions. The first, articulated by Brodie in 1946, holds that atomic weapons are so destructive that their primary purpose is not to fight wars but to prevent them. The second, developed by Thomas Schelling in the 1960s, holds that strategy in the atomic age is not about capability but about communication: states use their arsenals as bargaining tools, signaling resolve and imposing risks that make adversary aggression unprofitable. The third, most controversially advanced by Waltz in 1981, holds that proliferation enhances stability because every new state possessing the bomb acquires the capacity to deter aggression by threatening retaliation. The fourth, crystallized during the Cold War arms-control negotiations, holds that mutual assured destruction, the condition in which both sides can inflict unacceptable damage regardless of who strikes first, creates a stable equilibrium in which neither side has an incentive to initiate conflict.

These four propositions produced a prediction that the 2025 conflict falsified: atomic-armed states will not fight wars with each other. The prediction was never absolute in academic discourse; scholars like S. Paul Kapur, Paul Staniland, and others had long identified conditions under which deterrence might fail. But the prediction was operationalized in defense planning, diplomatic strategy, and public discourse as though it were absolute. The phrase “the bomb kept the peace” became a truism that shaped everything from NATO doctrine to nonproliferation advocacy. The 2025 India-Pakistan conflict did not merely create an exception to this truism. It exposed the truism as a contingent historical observation about the Cold War rather than a universal law of politics. The Cold War remained peaceful partly because of strategic arsenals and partly because of a dense web of institutional, diplomatic, and economic constraints. South Asia in 2025 lacks many of these non-weapons constraints: there is no economic interdependence comparable to East-West trade, no institutional framework comparable to the OSCE, no crisis-communication architecture comparable to the Washington-Moscow hotline until the DGMO channels were activated under fire. The arsenals in South Asia operate in a thinner institutional environment than their Cold War counterparts did, and the 2025 conflict demonstrates what happens when deterrence is asked to carry the full weight of crisis prevention without institutional support.

The Cases: Five Crises in Comparative Frame

Understanding what the 2025 conflict teaches about deterrence requires placing it in the context of four previous crises that defined the scholarly literature. Each crisis tested a different dimension of deterrence theory, and each produced a different set of lessons that the 2025 conflict either confirmed, complicated, or overturned.

The Cuban Missile Crisis of October 1962 remains the canonical case, and for good reason. When American U-2 reconnaissance aircraft discovered Soviet medium-range ballistic missiles under construction in Cuba, President Kennedy and Premier Khrushchev engaged in thirteen days of brinkmanship that brought the world closer to catastrophe than any subsequent crisis. The resolution, Soviet withdrawal of missiles in exchange for American withdrawal of Jupiter missiles from Turkey and a no-invasion pledge, established the principle that crises between atomic powers are resolved through mutual concession rather than unilateral victory. Cuba established two foundational precedents that deterrence theory subsequently treated as laws. First, both sides demonstrated extreme caution, with Kennedy overruling military advisors who recommended airstrikes and Khrushchev pulling back submarines that had been authorized to use atomic torpedoes. Second, the crisis was resolved without any conventional military engagement between American and Soviet forces. No shots were fired. No territory was contested. The crisis existed entirely in the realm of diplomatic signaling and threat perception. Deterrence theorists extracted the lesson that the bomb produces a “crystal ball effect,” allowing leaders to see the catastrophic consequences of escalation clearly enough to choose retreat.

The Kargil conflict of 1999 introduced a complication that Cuba had not anticipated. When Pakistani soldiers disguised as militants infiltrated positions along the Line of Control in the Kargil sector, India responded with a conventional military operation to retake the captured positions. The fighting lasted from May to July 1999 and produced over a thousand casualties on both sides. Kargil was the first direct military conflict between two declared atomic powers, both India and Pakistan having tested weapons in May 1998, barely a year before the shooting started. The crisis ended when Pakistan withdrew under combined Indian military pressure and American diplomatic pressure. Deterrence scholars initially treated Kargil as vindication: the arsenals constrained the conflict by preventing India from crossing the LoC into Pakistani-administered territory. India fought within Kargil but did not escalate horizontally or vertically. The atomic shadow, scholars argued, held the conflict within manageable bounds. But Kargil also introduced the stability-instability paradox. First articulated by Glenn Snyder and later developed by S. Paul Kapur, the paradox holds that strategic stability, the mutual certainty that neither side will launch a first strike, actually enables conventional instability by creating a shield under which lower-level provocations can occur. Pakistan’s Kargil incursion was itself enabled by its arsenal; without the bomb deterring an Indian full-scale invasion, the operation across the LoC would have been suicidal.

The 2001-2002 India-Pakistan standoff, triggered by the December 2001 attack on India’s Parliament by Jaish-e-Mohammed and Lashkar-e-Taiba operatives, tested a different facet. India mobilized nearly 500,000 troops to the border in Operation Parakram. Pakistan responded with its own mobilization. For ten months, the two armies faced each other, and on at least three occasions, Indian decision-makers considered authorizing military strikes into Pakistani territory. Deterrence scholars pointed to India’s ultimate restraint as evidence that the arsenals had prevented a fourth India-Pakistan war. But the 2001-2002 case introduced another complication: the cost of deterrence through mobilization without action. India’s failure to respond militarily to an attack on its own legislature emboldened Pakistani-based groups to attempt more ambitious operations, contributing directly to the planning environment that produced the 2008 Mumbai attacks. Deterrence prevented war but did not prevent the provocation that invited war, and the absence of a military response invited future provocations of even greater magnitude.

The 2019 Balakot episode compressed several of these dynamics into a shorter timeframe. Following the Pulwama suicide bombing that killed forty Indian paramilitary soldiers, India launched airstrikes against a Jaish-e-Mohammed facility in Balakot, Pakistan, the first time Indian aircraft had struck targets inside Pakistani territory since the 1971 war. Pakistan retaliated the following day with its own airstrikes. The crisis de-escalated within forty-eight hours. Balakot’s contribution was demonstrating that the weapons-use threshold was higher than many analysts had assumed. India struck inside Pakistan with airpower. Pakistan retaliated with airpower. Neither side escalated to missile exchanges, ground operations, or weapons-related signaling. The atomic shadow shaped the conflict by keeping it limited, but it did not prevent the initial strike or the retaliatory response. The escalation ladder had been climbed higher than Kargil, higher than the 2001 standoff, and still the threshold remained above the ceiling of conventional action.

What the Balakot case revealed, and what Kargil and the 2001 standoff had only hinted at, was a pattern of incremental threshold testing. Each successive crisis between India and Pakistan climbed higher on the escalation ladder than the previous one. Kargil produced ground combat in a single sector. The 2001 standoff produced full military mobilization without engagement. Balakot produced airstrikes inside Pakistani territory with retaliatory airstrikes. The pattern suggested that the threshold was not a fixed line but a moving ceiling, pushed upward with each crisis as both sides learned that the previous ceiling had been lower than the actual point of catastrophic escalation. This incremental learning process, in which both sides gained confidence from surviving each successive crisis without crossing the threshold, is itself a source of strategic risk. Confidence gained from previous crises can produce overconfidence in future crises, leading decision-makers to assume that the threshold will always be higher than their current level of action, an assumption that will eventually prove catastrophic if it is wrong.

The 2025 conflict climbed higher still. It climbed so high that the gap between the highest conventional rung reached and the lowest weapons-use rung became, for the first time in the history of the atomic age, genuinely uncertain.

The Escalation Ladder: Theory Against Reality

Herman Kahn’s escalation ladder, first published in his 1965 work “On Escalation,” remains the most widely referenced framework for analyzing how conflicts between atomic powers intensify. Kahn proposed forty-four rungs of escalation, from diplomatic disagreements at the bottom to spasm warfare at the top. The ladder was theoretical, designed to help Cold War strategists think systematically about spiraling crises. The 2025 India-Pakistan conflict provided the first opportunity to test the ladder against a real conventional war between atomic-armed states, and the results expose both the framework’s enduring utility and its critical blind spots.

The lowest rungs cover diplomatic protest, economic sanctions, and political maneuvering. The 2025 crisis moved through these rungs in the immediate aftermath of the Pahalgam attack on April 22, 2025. India expelled Pakistan’s remaining diplomatic personnel. India suspended all bilateral trade. India’s External Affairs Ministry issued statements that eliminated every diplomatic off-ramp. Pakistan’s Foreign Ministry responded with counter-statements blaming Indian state terrorism, establishing a rhetorical pattern that persisted through the conflict and beyond. These lower rungs were climbed rapidly, within days rather than the weeks or months that characterized the 2001-2002 standoff, reflecting a decision-making tempo accelerated by the severity of the Pahalgam provocation and the political impossibility of a restrained response.

The middle rungs cover conventional military operations of increasing intensity. India’s Operation Sindoor on May 7, 2025 represents the single most dramatic rung-jump in the history of confrontations between atomic powers. India launched precision missile strikes against nine Jaish-e-Mohammed and Lashkar-e-Taiba facilities within a twenty-three-minute window. The weapons employed, SCALP cruise missiles launched from Rafale fighter aircraft and BrahMos supersonic cruise missiles, represented frontline military systems whose deployment signaled a full military operation rather than a limited punitive strike of the Balakot variety. Kahn’s framework treats conventional strikes against military targets in the adversary’s homeland as a high-rung action. India jumped from diplomatic protest directly to homeland missile strikes, bypassing at least six intermediate rungs that deterrence theory predicted would provide off-ramps for de-escalation.

Pakistan’s response pushed the ladder higher. Within hours of Sindoor, Pakistani artillery opened fire on Indian border positions near Poonch, killing civilians on both sides. Over the following days, Pakistan launched Operation Bunyan-un-Marsoos, claiming retaliatory strikes against Indian military installations. Pakistani and Indian fighter jets engaged in the first dogfight between atomic powers in the jet era. Both sides deployed armed drones in combat roles. India’s S-400 air defense system saw its first combat deployment, engaging Pakistani aircraft and possibly drones in a real-world performance test of Russian military technology against Chinese and Western platforms.

The critical question for deterrence theory is where this conventional escalation sat relative to the weapons-use threshold. Kahn’s ladder places demonstrations, such as a test during a crisis, several rungs above conventional homeland strikes. Tactical use against military targets sits higher still. Strategic exchange occupies the uppermost rungs. In the 2025 conflict, the conventional fighting reached a level within three to five rungs of the demonstration threshold. The exchange of missile fire, aerial dogfights, drone warfare, and artillery bombardment constituted a sustained conventional war, not a crisis, not a standoff. The distinction matters because deterrence theory treats crises and wars differently. Crises are managed through signaling and diplomacy. Wars develop their own momentum, and the operational requirements of conventional combat create escalation pressures that diplomatic signaling cannot fully control.

Caitlin Talmadge’s research on inadvertent escalation during conventional conflict provides the most rigorous framework for understanding why the 2025 conflict’s position on the escalation ladder should alarm strategists. Talmadge argues that conventional military operations can trigger weapons use not through deliberate escalation but through three inadvertent mechanisms. First, conventional strikes may damage command-and-control infrastructure, leading the targeted state to believe a decapitation strike is underway and prompting a launch-on-warning response. Second, conventional operations may threaten delivery systems, such as airfields hosting dual-capable aircraft, creating use-it-or-lose-it pressure. Third, the operational fog of conventional combat may cause decision-makers to misinterpret conventional attacks as atomic ones, particularly when precision-guided munitions strike targets near sensitive facilities. All three of Talmadge’s pathways were plausible during the 2025 conflict. India’s missile strikes targeted military installations in Pakistan, and the question of whether any were co-located with sensitive assets, or perceived by Pakistani decision-makers as adjacent to them, remains unanswered. The aerial engagements involved dual-capable aircraft that can carry both conventional and warhead payloads, and a pilot engaging a target has no way of knowing whether the incoming aircraft is on a conventional or strategic mission.

Talmadge’s framework gains additional relevance when applied to the specific weapons systems deployed during the 2025 conflict. The BrahMos supersonic cruise missile, which India launched during Sindoor, travels at speeds exceeding Mach 2.8, giving the target state less than ten minutes of warning after detection. At these speeds, the decision-making window for determining whether an incoming missile carries a conventional or strategic payload shrinks to minutes, compressing the time available for verification to a duration that may be insufficient for reliable assessment. The SCALP cruise missile, launched from Rafale aircraft, presents a different identification challenge: it is a subsonic, terrain-hugging weapon designed to evade radar detection, meaning that Pakistani surveillance may have had reduced warning time compared to a ballistic trajectory. Both weapons systems create conditions in which Talmadge’s third pathway, misidentification of conventional strikes, becomes not merely theoretical but operationally plausible. The fact that India was simultaneously employing multiple weapons systems with different flight profiles compounded the identification challenge for Pakistan’s defensive radar and command-and-control networks, which had to track, classify, and assess multiple incoming threats at the same time.

The geographic dimension of the escalation ladder deserves separate analysis because the 2025 conflict introduced a spatial complexity that Kahn’s one-dimensional ladder metaphor cannot capture. The strikes covered a geographic arc from Pakistan-administered Kashmir through Punjab to Khyber Pakhtunkhwa, meaning that Pakistani command-and-control had to assess threats across hundreds of kilometers simultaneously. This geographic dispersion created the possibility that strikes in one sector might be misinterpreted in another, particularly if communication between Pakistani military commands was disrupted by the conventional fighting itself. A strike against a Jaish-e-Mohammed camp in Kashmir might be correctly identified as a counter-terrorism operation by the local Pakistani command, while the same pattern of explosions detected by a distant command post with incomplete information might be interpreted as the opening salvo of a broader campaign against Pakistani military infrastructure. The spatial dimension of escalation risk, in which the same action can be interpreted differently depending on the observer’s geographic position and information environment, is a dimension that Kahn’s ladder does not address and that the 2025 conflict exposed as critically important.

Doctrinal Collision: No First Use Against Full Spectrum Deterrence

The 2025 conflict produced a direct collision between India’s declared doctrine and Pakistan’s declared doctrine, a collision that deterrence theory predicted would prevent exactly the kind of conventional war that occurred.

India adopted its doctrine in January 2003, resting on three pillars: credible minimum deterrence, meaning India maintains only the forces necessary to inflict unacceptable damage in retaliation; No First Use, meaning India pledges not to initiate atomic warfare and reserves its arsenal exclusively for retaliation; and massive retaliation, meaning any first strike against India will be met with a response designed to inflict unacceptable damage. The NFU pledge is the most debated element. Critics including Vipin Narang have argued that NFU may not survive a crisis in which India faces an attack on its forward-deployed forces. Defenders argue that the pledge enhances stability by removing Pakistan’s fear of an Indian first strike and thereby reducing pressure for Pakistan to adopt a hair-trigger alert posture.

Pakistan’s posture stands in deliberate opposition to India’s NFU. Pakistan has never adopted a no-first-use pledge and has explicitly rejected the concept. Pakistan’s official posture, articulated through various official and semi-official channels since the establishment of the Strategic Plans Division in 2000, is described as “full spectrum deterrence.” This doctrine holds that Pakistan reserves the right to use its arsenal first if four redlines are crossed: India conquers a large portion of Pakistani territory; India destroys a significant proportion of Pakistan’s armed forces; India imposes an economic blockade; and India engages in political destabilization. The critical element for the 2025 analysis is the deliberate ambiguity of these redlines. “Large portion” is undefined. “Significant proportion” has no numerical threshold. The economic blockade redline, given India’s actual suspension of bilateral trade, may have been technically triggered depending on how expansively Pakistani strategists interpreted it. Feroz Hassan Khan, whose book “Eating Grass” provides the most comprehensive English-language account of Pakistan’s decision-making on these matters, argues that this ambiguity is intentional, designed to create uncertainty in Indian minds and thereby deter any conventional military operation.

The 2025 conflict tested this doctrinal collision directly. India launched conventional strikes against Pakistani territory, precisely the scenario Pakistan’s doctrine is designed to deter. India’s NFU pledge, by guaranteeing restraint, was supposed to provide Pakistan assurance that India’s operations would remain conventional. Pakistan’s full-spectrum posture, by threatening first use, was supposed to deter India from launching conventional operations. Both doctrines failed in their primary deterrent function. India struck despite Pakistan’s threats. Pakistan did not respond with its arsenal despite India striking its territory.

Frank O’Donnell, whose research provides some of the most granular analysis available, offers a framework explaining this outcome. O’Donnell argues that India’s decision to strike was enabled by a judgment that Pakistan’s redlines were not credible at the level of force India intended to employ. India’s strikes were precision-guided, limited in scope, targeted at terrorist infrastructure rather than Pakistani military forces, and framed as counter-terrorism operations. India calculated that Pakistan could not credibly threaten retaliation against an operation whose stated purpose was destroying terrorist camps, because the political cost of escalation over terrorist infrastructure would far exceed the military cost of absorbing the strikes. This calculation proved correct in the immediate term. But the calculation’s correctness does not validate it as a sustainable strategic framework. If the assessment was wrong, the consequences would have been civilizational.

Pakistan’s decision not to escalate to weapons use is equally revealing. Pakistan’s conventional military response, artillery fire, airstrikes, drone deployment, and the claimed Operation Bunyan-un-Marsoos, demonstrated that Pakistan possessed conventional options short of catastrophic escalation and was willing to employ them. This contradicts the most alarming pre-conflict assessments, which suggested that Pakistan might resort to its arsenal early in a conflict because its conventional forces could not withstand an Indian offensive. The 2025 evidence suggests that Pakistan’s conventional capability, bolstered by Chinese military hardware and improved air defense systems, provided a buffer between conventional setback and weapons use that the most pessimistic projections had denied existed. The buffer, however, is narrow. The four-day duration raises the question of whether a longer conventional war, one genuinely threatening Pakistan’s military coherence, would have produced different calculations.

The doctrinal collision also revealed the difference between a state’s declaratory doctrine and its operational doctrine. Pakistan’s declaratory posture, full-spectrum deterrence with four undefined redlines, is designed for peacetime signaling, crafted to maximize ambiguity and thereby maximize the deterrent effect on India’s planning. Pakistan’s operational doctrine, revealed through its actual behavior during the crisis, is considerably more restrained than its declaratory posture implies. Pakistan absorbed missile strikes against its territory, aerial combat over its airspace, and the destruction of facilities housing organizations that the Pakistani state has historically protected, all without crossing the weapons-use threshold. This gap between declaratory and operational doctrine has important implications. If India recognizes that Pakistan’s operational doctrine is more restrained than its declaratory one, India may become increasingly willing to push conventional operations further, gambling that the redlines Pakistan advertises will not be operationally enforced. This is a rational but dangerous calculation, because the gap is not constant. Under different circumstances, with different leadership, greater conventional losses, or a more directly threatening operation, Pakistan’s operational behavior might align with its declaratory threats.

The collision’s third revelation concerns the role of framing in managing escalation between armed states. India framed Sindoor as a counter-terrorism operation, not as a military strike against the Pakistani state. This framing was not merely rhetorical; it was a deliberate escalation-management strategy. By framing the strikes as targeting non-state actors rather than sovereign military assets, India provided Pakistan with a face-saving interpretation that did not require a strategic response. Pakistan could absorb the strikes as damage to terrorist infrastructure, which Pakistan officially opposes, rather than as an attack on Pakistani sovereignty, which would require retaliation under its declared doctrine. This framing strategy is both clever and dangerous: clever because it exploits the ambiguity of Pakistan’s relationship with the targeted organizations, and dangerous because it relies on the adversary accepting the attacking state’s frame rather than imposing its own. If Pakistan had rejected the counter-terrorism frame and treated the strikes as an act of war, the outcome might have been catastrophic. The success of India’s framing strategy creates a template that India may replicate in future crises, but each replication increases the probability that Pakistan will reject the frame and respond on the basis of its own interpretation.

The Shadow of the Absent Weapon: How Arsenals Shaped Conventional Decisions

The most analytically challenging dimension of the 2025 conflict is not what the weapons did but what they prevented without being used. The arsenals shaped every conventional decision both sides made, creating pervasive influence impossible to measure precisely but impossible to ignore. Understanding this influence requires examining not only the decisions that were made but the decisions that were not made, the targets that were not struck, the operations that were not launched, and the escalation pathways that were not taken. The negative space of the conflict, the actions that the arsenals prevented, is where the weapons’ influence was most powerful and most consequential.

India’s target selection for Sindoor reveals this influence in the negative space of what was not targeted. India struck terrorist infrastructure, not Pakistani Air Force bases, not army cantonment areas, not naval installations, and not command-and-control facilities. This target selection was driven partly by the stated counter-terrorism objective but also shaped by the atomic shadow. Striking Pakistani military infrastructure would have risked triggering Talmadge’s inadvertent escalation pathways by threatening delivery systems or command-and-control nodes. India’s decision to limit strikes to non-state-actor infrastructure was a deterrence-influenced choice, an acknowledgment that the atomic shadow imposed constraints on target selection that a non-armed adversary would not have imposed. In a conflict with a conventionally-armed-only Pakistan, India’s target list would almost certainly have included Pakistani military assets directly supporting cross-border terrorism, including ISI facilities with documented connections to the targeted organizations, army units providing logistical support to proxy groups, and airfields used for the movement of personnel and materiel across the Line of Control. The arsenal removed these targets from India’s calculus, creating a paradox in which Pakistan’s weapons protected the very military infrastructure that enabled the terrorism that triggered the conflict.

The target-selection constraint operated differently for different categories of military targets, and understanding these differences is essential for projecting how the constraint might evolve in future crises. Targets can be categorized into four tiers based on their proximity to the weapons-use threshold. The first tier, terrorist infrastructure with no co-location near military or strategic assets, carries the lowest escalation risk and is what India actually struck. The second tier, military facilities directly supporting cross-border terrorism but located away from strategic assets, carries moderate escalation risk because striking military targets signals intent against the Pakistani state rather than against non-state actors. The third tier, military facilities with potential co-location near delivery systems or command-and-control nodes, carries high escalation risk because of the Talmadge pathways. The fourth tier, strategic facilities themselves, carries catastrophic risk. India’s 2025 target selection stayed entirely within the first tier. A future crisis, particularly one in which the first-tier strikes proved insufficient to achieve India’s objectives, might push India toward the second tier, testing whether Pakistan’s response changes when the targets are military rather than terrorist. This tier-progression risk is one of the most important dimensions of future India-Pakistan crisis dynamics.

India’s decision not to cross the international border with ground forces represents another deterrence-shaped choice. India’s conventional superiority in ground forces, armor, and mechanized infantry was irrelevant because the atomic shadow made a ground incursion unthinkable. The stability-instability paradox operated with textbook precision: strategic stability enabled conventional instability. India could fire missiles at terrorist camps. India could not march tanks toward Lahore. The distinction between these two actions was defined entirely by the deterrence calculus. The paradox also operated in the maritime domain, where India’s naval superiority over Pakistan is even more pronounced than its ground-force advantage. India did not impose a naval blockade on Pakistani ports despite possessing the capability to do so, because a blockade would have triggered Pakistan’s economic-blockade redline, one of the four declared thresholds for weapons use. India’s naval restraint, like its ground-force restraint, demonstrates the atomic shadow’s power to neutralize conventional military advantages that would be decisive in a non-atomic conflict.

Pakistan’s conventional response was equally shaped by constraints operating in both directions. Pakistan’s artillery bombardment of Indian positions near Poonch represented the upper boundary of Pakistan’s risk tolerance for conventional retaliation. Pakistan could shell border areas, launch retaliatory airstrikes, and deploy drones without triggering Indian retaliation because these actions fell below India’s threshold. But Pakistan could not pursue a larger conventional counter-offensive because any operation threatening Indian territory at scale would have undermined the NFU framework paradoxically protecting Pakistan from Indian first use. Pakistan needed India’s NFU to remain operative because NFU was the only guarantee that India’s arsenal would not be unleashed in response to Pakistan’s conventional retaliation. If Pakistan’s conventional response had been severe enough to compel India to reconsider NFU, Pakistan would have destroyed the very shield keeping Indian warheads in their silos. This reciprocal constraint, in which both sides’ conventional options were bounded by the other side’s deterrence posture, created a conflict zone that was paradoxically both more violent and more controlled than a conflict between non-armed states would have been. More violent because the stability-instability paradox enabled conventional fighting that the arsenals’ deterrent effect was supposed to prevent. More controlled because the arsenals imposed ceilings on escalation that neither side was willing to test.

The influence on the ceasefire that ended the fighting on May 10 is perhaps the clearest example. The ceasefire was negotiated through DGMO hotlines with reported American diplomatic pressure. O’Donnell’s analysis suggests that fear of escalation was a necessary but not sufficient condition. Both sides were aware that continued conventional escalation increased the probability of inadvertent escalation through the Talmadge pathways. The fog of conventional combat, the deployment of dual-capable aircraft, and compressed decision-making timelines all created conditions in which a misidentified strike or communication failure could trigger catastrophe. This awareness contributed to both sides’ willingness to accept a ceasefire even though neither had achieved its maximum conventional objectives. India had struck terrorist infrastructure but had not eliminated the organizational capacity of Jaish-e-Mohammed or Lashkar-e-Taiba. Pakistan had demonstrated the ability to retaliate conventionally but had not reversed the destruction Sindoor inflicted. Both sides accepted incomplete conventional outcomes because the alternative, continued fighting with increasing escalation risk, was worse than an unsatisfying ceasefire.

Command and Control Under Fire: Pakistan’s Architecture in Crisis

The 2025 conflict exposed a dimension of risk that deterrence theory has addressed abstractly but never confronted empirically: the performance of command-and-control systems during sustained conventional combat. This dimension matters because the safety of every person in South Asia, and potentially every person on the planet, depends on the functioning of command-and-control systems that were designed for peacetime conditions and tested only in exercises that cannot replicate the stress, information overload, and communication disruption of actual warfare.

Pakistan’s command-and-control architecture centers on the National Command Authority, established in 2000 and formalized by the National Command Authority Act of 2010. The NCA is chaired by the Prime Minister and includes the Chairman of the Joint Chiefs of Staff Committee, the three service chiefs, and additional military and civilian officials. The Strategic Plans Division, headed by a serving three-star general, serves as the NCA’s secretariat and exercises operational control over Pakistan’s arsenal. In peacetime, Pakistan reportedly maintains its warheads in a de-mated configuration, stored separately from delivery systems, reducing the risk of unauthorized launch but increasing the time required to assemble and deploy weapons in a crisis. This de-mated posture is itself a form of negative control, designed to ensure that no single individual, including the SPD commander, can independently authorize weapons use. The de-mated configuration requires multiple physical steps, retrieving warheads from storage, transporting them to delivery-vehicle locations, mating them to missiles or aircraft, and arming them, each of which is controlled by different elements of the command hierarchy.

Feroz Hassan Khan’s research provides the most detailed publicly available analysis of how this architecture functions under crisis conditions. Khan argues that the system faces a fundamental tension between negative control, preventing unauthorized or accidental launch, and positive control, ensuring that weapons can be delivered when the NCA authorizes their use. In peacetime, the system emphasizes negative control through physical separation, multi-layer authorization, and electronic permissive action links. In a crisis, the system must shift toward positive control, pre-delegating authority, mating warheads to delivery vehicles, and dispersing forces to survivable locations. The shift from negative to positive control is not a binary switch but a graduated process, and the speed at which the process occurs depends on the perceived severity of the threat. A slow, deliberate shift allows for careful verification and reduces the risk of inadvertent escalation. A rapid shift under the pressure of incoming conventional strikes increases the risk of errors, miscommunication, and unauthorized actions by commanders who may interpret the crisis as more severe than the NCA intends.

The 2025 conflict presented this architecture with its first real-world stress test. The open-source record does not reveal what posture changes Pakistan made, and the absence of evidence is itself analytically significant. There are three possibilities. First, Pakistan may have begun mating warheads to delivery vehicles and dispersing forces, a positive-control shift that would have reduced the time between a launch decision and a detonation from hours to minutes. If this occurred, the crisis was more dangerous than the observable conventional fighting suggested, because dispersed and mated warheads are more vulnerable to unauthorized use, command-and-control disruption, and misinterpretation of conventional attacks as strategic threats. Dispersal also increases the number of decision points in the system, because dispersed weapons require dispersed authorization, and dispersed authorization increases the probability that a local commander, operating under the stress and information limitations of combat, will make a decision that the NCA did not intend. Second, Pakistan may have maintained its de-mated peacetime posture throughout, relying on the judgment that India’s strikes did not threaten Pakistan’s survivability. If this occurred, it suggests a higher level of confidence in conventional deterrence than Pakistan’s public posture implies, and it indicates that the full-spectrum deterrence posture is more a political signaling device than an operational military doctrine. Third, Pakistan may have taken intermediate steps, increasing the readiness of certain delivery systems while keeping others in peacetime posture. The third scenario is most likely and most troubling for deterrence theory, because it suggests that force-posture decisions during a crisis are improvisational rather than doctrinal.

The performance of Pakistan’s command-and-control architecture during the 2025 crisis also raises questions about the system’s resilience under conditions of electronic warfare and communication disruption. Modern conventional warfare routinely involves electronic warfare operations designed to degrade the adversary’s communication and surveillance capabilities. If India’s Sindoor operation included electronic warfare components, which precision-guided operations typically do, these operations may have affected Pakistan’s command-and-control communications, at least temporarily. Any disruption of communication between the NCA in Islamabad and field commanders controlling delivery systems would create a gap during which local commanders might be required to make autonomous decisions about weapons readiness, decisions that the system is designed to prevent. The intersection of conventional electronic warfare and strategic command-and-control vulnerability is a dimension of risk that the 2025 conflict highlighted but that no open-source analysis has fully explored, partly because the relevant information is classified and partly because the analytical community has not yet developed frameworks for assessing this intersection.

India’s No First Use Under Pressure: Doctrine Tested

The 2025 conflict also tested India’s NFU from a direction that pre-conflict analysis had not adequately addressed: whether India’s conventional military superiority, combined with NFU, creates a perverse incentive structure in which India can wage unlimited conventional war while sheltering behind the shield that NFU provides. If India pledges never to use its arsenal first, and if Pakistan’s weapons deter India from crossing certain conventional thresholds, then India operates within a conventional window both enabled and constrained by the arsenals on both sides. The 2025 conflict demonstrated the existence of this window with unprecedented clarity.

The pre-2025 scholarly debate about NFU focused almost exclusively on the question of credibility: would India maintain the pledge if Pakistan used tactical warheads against Indian military forces on the battlefield? Narang’s influential 2014 analysis argued that India’s massive-retaliation posture made NFU credible because it promised catastrophic punishment for any first use, whether tactical or strategic. Critics responded that massive retaliation against a tactical strike was not credible because the punishment would be disproportionate to the provocation, creating a deterrence gap at the tactical level. This debate, while analytically important, proved largely irrelevant to the 2025 crisis because the crisis tested NFU in a different dimension entirely. The question in 2025 was not whether India would maintain NFU in response to Pakistani first use but whether India could exploit NFU as a strategic asset in a conventional war that India initiated.

The answer revealed by the 2025 case is that NFU functioned as both a constraint and an enabler. As a constraint, NFU limited India’s ability to threaten catastrophic escalation as a coercive tool during the conventional fighting. India could not say, implicitly or explicitly, “accept our conventional strikes or face atomic retaliation,” because NFU foreclosed this option. As an enabler, NFU provided Pakistan with a baseline assurance that India’s operations would remain conventional, thereby allowing Pakistan to calibrate its response without the fear that any conventional retaliation might trigger an Indian first strike. NFU, paradoxically, made the conventional war possible by reducing Pakistan’s incentive for preemptive escalation. If India had not had an NFU policy, Pakistan’s calculations would have been fundamentally different. Without NFU, Pakistan would have faced a conventional attack from a state that reserved the right to escalate to strategic weapons at any time, creating enormous pressure for preemptive action before India could launch first. With NFU, Pakistan knew that India would not initiate strategic warfare, allowing Pakistan to respond conventionally to India’s conventional strikes without fear of catastrophic retaliation. In this sense, India’s NFU policy may have prevented the 2025 crisis from becoming far more dangerous than it was, by removing one of the most potent drivers of preemptive escalation.

The question of whether NFU held during the 2025 conflict seems straightforward: India did not use its arsenal, so NFU held. But the question is more complex. NFU held at the level of action, but the question of whether it held at the level of signaling is less clear. Reports suggest that Indian strategic forces were placed on heightened alert during the crisis, a standard precautionary measure that does not violate NFU but sends a signal that options are being prepared. This creates a paradox in which NFU is most effective as a deterrent not when it is ironclad but when its credibility is ambiguous. If Pakistan believes India will maintain NFU regardless, Pakistan can escalate conventionally without fear. If Pakistan believes India might abandon NFU under extreme pressure, Pakistan’s conventional retaliation is constrained, reinforcing India’s conventional advantage. India’s interest lies in maintaining NFU as declared policy while allowing sufficient ambiguity about its operational commitment to keep Pakistan uncertain. The scholarly debate about whether India should revise its NFU pledge, engaging analysts from Narang to Rajesh Rajagopalan, may be asking the wrong question. The operational question is how much ambiguity serves India’s strategic interests, and the 2025 case suggests the answer is enough to constrain Pakistan without so much that Pakistan shifts toward preemptive calculations.

The 2025 case also raises a question about NFU’s long-term sustainability that the pre-conflict literature had not adequately addressed. If India can fight a conventional war under the protection of NFU and achieve its immediate military objectives, the incentive to maintain NFU is reinforced: the policy works because it enables India to act conventionally while preventing atomic escalation. But if the 2025 model is repeated, and India conducts progressively more ambitious conventional operations while relying on NFU to prevent Pakistani escalation, the policy may eventually be tested to its breaking point. A future crisis in which India’s conventional operations cause significantly greater damage to Pakistan than Sindoor did, or in which India’s operations directly threaten Pakistani military coherence, could produce a situation in which Pakistan concludes that NFU is no longer protecting Pakistani interests but enabling Indian aggression. At that point, Pakistan’s calculation shifts from absorbing conventional strikes under the protection of India’s NFU to preempting India’s conventional superiority by threatening or conducting first use. The trajectory from the 2025 model to this breaking point is not inevitable, but it is structurally embedded in the interaction between India’s NFU and India’s growing conventional capability, and understanding this trajectory is essential for assessing the long-term stability of South Asia’s deterrence architecture.

Four Days That Rewrote the Rules: The Timeline

The chronological reconstruction of the 2025 conflict’s weapons dimension reveals a tempo of escalation that no existing deterrence model anticipated.

Day One, May 7, 2025, began with Sindoor striking nine targets within twenty-three minutes. The timeline is analytically significant because it compressed the adversary’s response window to a duration shorter than most command-and-control systems are designed to process. Pakistani military command had less than half an hour to determine whether the incoming strikes were conventional or something else, a determination Pakistan’s doctrine required for calibrating its response. The strikes used precision-guided munitions creating blast signatures distinguishable from thermonuclear detonations only through technical analysis requiring time Pakistan did not have. That Pakistan’s response was conventional, artillery within hours, indicates either that Pakistani surveillance confirmed the strikes’ character quickly or that Pakistan’s architecture defaults to restraint in ambiguous situations.

Day Two saw escalation expand across multiple domains. Pakistani artillery intensified along the LoC and the international border in the Punjab sector. India’s Air Force conducted additional sorties to enforce what Indian officials described as air superiority over the engagement zone. Pakistan scrambled fighters, creating conditions for the aerial engagement that represented the first jet-era dogfight between atomic powers, with both sides deploying frontline combat aircraft in contested airspace. The aerial engagement is analytically significant beyond its tactical dimensions because it introduced the problem of dual-capable aircraft identification into a live combat environment. Both India’s Rafale and Pakistan’s F-16 variants are platforms capable of delivering both conventional and strategic payloads, and an adversary tracking an incoming sortie cannot determine the payload type from the aircraft’s radar signature, flight profile, or speed. A Rafale approaching a target at low altitude and high speed looks identical on radar regardless of whether it carries a SCALP cruise missile with a conventional warhead or a weapon with a strategic payload. This ambiguity persists throughout the engagement, from detection through interception to weapons release, creating a window during which the defending side must make decisions about response under fundamental uncertainty about the attacker’s intent.

Both sides also began deploying armed drones, adding an unmanned dimension complicating identification for command-and-control systems. A drone strike can be detected and tracked by the same surveillance systems monitoring missile launches, and the signatures of drone-delivered and cruise-missile-delivered munitions are similar enough to create identification ambiguity under combat conditions. The drone dimension introduced another complication that Day One had not presented: the question of attribution. When a missile is launched from a known Indian Air Force aircraft, the attribution is unambiguous. When a drone strikes a target, the attribution depends on tracking the drone to its launch point or identifying its communication links, both of which may be disrupted during active combat operations. A drone strike of uncertain attribution, particularly one that damages infrastructure near sensitive installations, creates an identification problem that is more dangerous than a clearly attributed strike because the defending side must decide how to respond without certainty about who is responsible or what the intent was. Day Two’s multi-domain character, combining manned aircraft, unmanned systems, and artillery across multiple geographic sectors, created an information environment more complex and more dangerous than any single-domain engagement would have produced.

Day Three represented the crisis’s most dangerous phase, though this assessment relies on inference rather than confirmed intelligence. Pakistan launched what it described as Operation Bunyan-un-Marsoos, claiming retaliatory strikes against Indian military positions. The intensity of conventional fighting reached its peak, with both sides committing additional military assets to the conflict zone. Casualty accumulation is a documented driver of escalation pressure because each death creates domestic demand for retaliation that can accelerate beyond decision-makers’ ability to control. The retaliation-demand cycle operates on a ratchet mechanism: each round of casualties creates pressure for a proportionate or escalatory response, and the response generates casualties on the other side that trigger the same cycle. By Day Three, both sides had accumulated enough casualties that the domestic political pressure for continued or intensified military action was competing directly with the strategic pressure for de-escalation, creating a tension that only external intervention could resolve.

American diplomatic intervention reportedly intensified on Day Three, with Washington communicating directly with both capitals about the risks of continued escalation. The American role deserves careful analysis because it highlights the limitations of external mediation between atomic-armed states in active conventional combat. The United States possessed intelligence capabilities, including satellite surveillance and signals intelligence, that provided Washington with some visibility into both sides’ force-posture changes during the crisis. If either side had begun mating warheads to delivery vehicles or dispersing strategic forces, American intelligence would likely have detected the activity and communicated concerns to both New Delhi and Islamabad. American pressure was motivated not only by alliance considerations and regional interests but by the existential risk that weapons use in South Asia would pose to the global strategic order and to the physical safety of American forces deployed across the Indo-Pacific region. Day Three was also the point at which the conflict’s geographic scope reached its maximum extent, with fighting reported along the LoC, the international border in the Punjab sector, and drone operations across multiple sectors. The multi-domain, multi-sector character of Day Three’s fighting created the most complex information environment that either side’s command-and-control architecture had ever faced in real combat conditions. The number of simultaneous engagement areas exceeded the capacity of either side’s surveillance systems to monitor comprehensively, creating blind spots in which actions could be taken or misinterpreted without the knowledge of senior decision-makers.

Day Four produced the ceasefire. The DGMO hotline, a military-to-military communication channel designed for exactly this type of crisis, facilitated the negotiations producing the May 10 cessation of hostilities. The ceasefire’s terms, freezing the military situation in place without resolving any underlying issues, reflected both sides’ recognition that continued fighting carried risks disproportionate to any achievable conventional objective. India had struck terrorist infrastructure but had not eliminated the organizational capacity of Jaish-e-Mohammed or Lashkar-e-Taiba. Pakistan had demonstrated the ability to retaliate conventionally but had not reversed the physical destruction inflicted by Sindoor. Both sides accepted an incomplete conventional outcome because the alternative, continued fighting with accumulating escalation risk, was worse than an unsatisfying truce. The ceasefire’s fragility is itself analytically significant. A ceasefire that resolves the underlying conflict removes the conditions for future escalation. A ceasefire that merely freezes the military situation preserves those conditions intact, guaranteeing that a future provocation of sufficient severity will trigger the same cycle. The 2025 ceasefire belongs to the second category. The organizations responsible for the Pahalgam attack continue to exist. Pakistan’s territory continues to provide sanctuary for groups targeting India. India’s willingness to use military force against Pakistani territory has been demonstrated and will not be un-demonstrated. Every structural condition that produced the 2025 crisis remains operative, and the 2025 precedent has established that conventional military action against an atomic-armed adversary’s homeland is a viable policy option rather than an unthinkable escalation. The four-day duration places the 2025 conflict in a category by itself: longer than any previous India-Pakistan military engagement except the 1999 Kargil conflict, which lasted weeks but involved fighting in a single geographic sector rather than multi-domain operations across the entire front, and far longer than the 2019 Balakot exchange. Four days of sustained conventional combat generated four days of escalation risk, a duration that no Cold War crisis equaled and that no existing crisis-management framework was designed to sustain.

Success or Failure: The Central Debate

Whether the 2025 conflict proves deterrence works or is failing has become the defining debate in contemporary strategic studies.

The “deterrence worked” school argues that the arsenals prevented catastrophe even under extreme conditions of conventional combat. Despite four days of sustained fighting, neither side crossed the threshold. The ceasefire demonstrates that atomic-armed states retain the ability to manage escalation during active combat. This school points to specific mechanisms: both sides maintained communication, both exercised restraint in target selection, and the ceasefire was achieved within four days before escalation risks materialized. For scholars in this tradition, the 2025 conflict is evidence that the “strategic revolution” identified by Robert Jervis remains operative.

The “deterrence failed” school, which this analysis supports, argues that the 2025 case represents a structural failure of deterrence theory’s central prediction. Deterrence is supposed to prevent war between atomic-armed states, not merely prevent weapons use during such a war. If two countries possessing thermonuclear arsenals can fight a four-day conventional war involving missile exchanges and hundreds of casualties, then the arsenals have failed to deter precisely the category of conflict they are supposed to prevent. The argument is not that the weapons are irrelevant. They clearly shaped every dimension of the conflict. The argument is that the version of deterrence theory treating arsenals as a guarantee against war between atomic powers is empirically falsified.

The 2025 evidence supports the “failed” interpretation more strongly for three reasons. First, the speed of escalation, roughly two weeks from Pahalgam to Sindoor, demonstrates that the atomic shadow did not produce the extended deliberation deterrence theory predicts. Kennedy and Khrushchev spent thirteen days in diplomatic maneuvering before resolving Cuba without a shot. India spent fifteen days in political deliberation before launching missile strikes against an atomic-armed adversary’s homeland. The shadow produced less restraint in 2025 than in 1962. Second, the conflict demonstrated that escalation can be driven by domestic political considerations that atomic fear cannot override. India’s leadership faced intense public pressure to respond to Pahalgam with military force, and the political cost of restraint exceeded the security cost of escalation. Deterrence theory assumes leaders prioritize survival over all other considerations, but the 2025 case shows leaders may prioritize political survival over national survival. Third, the ceasefire’s fragility undermines the “worked” interpretation. The ceasefire resolved none of the underlying issues. The organizations responsible for Pahalgam continue to operate. The structural conditions producing the 2025 conflict remain, and each repetition pushes conventional fighting closer to the threshold.

A fourth reason, rarely discussed in the immediate post-conflict analysis but emerging in more reflective scholarly treatments, is that the 2025 conflict’s termination may owe less to the arsenals’ deterrent effect and more to factors that have nothing to do with weapons at all. The ceasefire was produced by a combination of military exhaustion, American diplomatic pressure, and mutual political calculation that the costs of continued fighting exceeded the benefits. If the ceasefire was primarily the product of these non-weapons factors, then the arsenals’ contribution to termination was marginal, and the “worked” interpretation collapses entirely. The counter-argument, that both sides’ awareness of the catastrophic potential influenced every decision including the decision to accept a ceasefire, is plausible but unfalsifiable: it is impossible to prove that leaders would have continued fighting in the absence of the arsenals, because the counterfactual cannot be observed. This unfalsifiability is itself a problem for deterrence theory, because it means that the theory cannot be tested against the very cases it is designed to explain.

The implications of this debate extend beyond the India-Pakistan dyad in ways that should concern every strategic planner on the planet. If the “worked” interpretation prevails, the policy prescription is continuity: maintain existing postures, invest in command-and-control resilience, and trust that the arsenals will continue to prevent catastrophe. If the “failed” interpretation prevails, as this analysis argues it should, the policy prescription is urgent reform: develop crisis-management mechanisms specifically designed for conventional wars between atomic-armed states (which currently do not exist in any formalized architecture between India and Pakistan), address the conventional capabilities that the stability-instability paradox enables, and abandon the comforting illusion that the arsenals make great-power war impossible. The stakes of the debate are measured not in academic prestige but in civilizational survival, because the wrong answer could produce a policy framework that enables the very catastrophe it is designed to prevent.

Global Implications: Beyond the Subcontinent

The 2025 conflict carries implications extending to every dyad on the planet. The lessons are not confined to South Asia because the theoretical frameworks tested, deterrence stability, escalation control, the stability-instability paradox, are universal principles underpinning the entire global strategic architecture.

The US-China relationship faces a version of the stability-instability paradox mirroring the India-Pakistan dynamic. Strategic stability between Washington and Beijing creates space for conventional competition in the South China Sea, the Taiwan Strait, and the broader Indo-Pacific. The 2025 case demonstrates that this space can produce actual warfare, not merely competition, if the political stakes are high enough. A crisis over Taiwan could produce an escalation dynamic with the same characteristics: rapid movement from diplomatic crisis to military operations, constrained but real conventional combat, and a ceasefire driven by fear rather than conflict resolution. The Taiwan scenario carries an additional complication: the involvement of multiple armed states in a single crisis, creating a multi-player escalation dynamic even harder to manage than the bilateral India-Pakistan case.

The NATO-Russia relationship faces its own escalation risks illuminated by the 2025 case. Russia’s doctrine, reserving the right of first use in response to a conventional attack threatening the state’s existence, is structurally similar to Pakistan’s full-spectrum posture. The 2025 case demonstrates that a state with a first-use doctrine will nonetheless absorb significant conventional punishment before invoking its ultimate options, but that the threshold is undefined and may be lower than the attacking state assumes. The lesson for NATO is that Russia’s threshold is both higher than Moscow’s rhetoric suggests and lower than NATO might hope, creating a zone of uncertainty within which conventional operations carry risks that cannot be precisely quantified. Russia’s recent doctrinal evolution, including statements suggesting that warheads could be used in response to conventional threats, parallels Pakistan’s rhetorical trajectory before 2025. Pakistan’s rhetoric proved calibrated for deterrent effect rather than operational guidance. Russia’s rhetoric may follow the same pattern, but relying on an adversary’s rhetoric being bluster is dangerous strategic gambling.

The Israel-Iran dynamic represents a potential future case the 2025 conflict previews. If Iran acquires the bomb, the Israel-Iran relationship enters a stability-instability dynamic in which strategic deterrence enables sub-threshold conventional and proxy competition. The 2025 case suggests this competition can escalate to direct military engagement without triggering the ultimate escalation, but with risks that are difficult to manage. Israel’s strategic doctrine, historically relying on conventional superiority, would face the same constraint India’s superiority faced: weapons on the other side limit conventional superiority’s utility by creating escalation ceilings.

The North Korea dimension challenges conventional wisdom about Pyongyang’s program. The standard analysis treats North Korean warheads as a deterrent against American regime change. But the 2025 case demonstrates that arsenals serve a dual function: deterring existential threats while enabling lower-level provocations. The stability-instability paradox operates on the Korean Peninsula as it does in South Asia. A North Korean provocation of sufficient severity could trigger a conventional military response testing the same escalation dynamics the 2025 conflict exposed. The specific parallels between the South Asian and Korean cases are instructive. Both involve an asymmetric conventional balance, with the conventionally superior power (India in South Asia, the US-South Korea alliance on the Korean Peninsula) constrained by the inferior power’s arsenal from exploiting its conventional advantage. Both involve a history of provocations that test the superior power’s willingness to absorb attacks without responding militarily. And both involve domestic political dynamics in the superior power that create pressure for military responses to provocations, pressure that the inferior power’s arsenal may not be sufficient to deter. The 2025 India-Pakistan case demonstrated that this pressure can produce sustained conventional warfare despite the presence of arsenals on both sides. A Korean crisis with similar characteristics, a North Korean provocation severe enough to generate irresistible political pressure for a South Korean or American military response, could produce a conflict with escalation dynamics that the 2025 South Asian case previews.

The broader implication for the global strategic order is that the post-1945 assumption of atomic peace, the idea that these weapons prevent great-power war, is a historically contingent observation rather than a theoretical certainty. The Cold War remained peaceful partly because of strategic arsenals and partly because of a dense web of institutional, economic, and diplomatic constraints that supplemented deterrence. As the global security environment evolves, with new armed states, weakening institutional constraints, intensifying political pressures, and the proliferation of precision-guided conventional weapons that blur the line between conventional and strategic operations, the conditions under which deterrence fails at the conventional level multiply. The 2025 India-Pakistan conflict is the first demonstration of this failure. It may not be the last. The intellectual challenge for strategists, diplomats, and scholars is not to explain away the 2025 case as an anomaly but to incorporate it into a revised understanding of how deterrence operates in the real world rather than in the theoretical models that the 2025 conflict has exposed as inadequate. The practical challenge is even more urgent: to build the crisis-management institutions, communication architectures, and arms-control frameworks that the 2025 case demonstrated are missing, before the next crisis between atomic-armed states produces a conventional war that is less fortunate in its outcome than the one the world survived in May 2025.

The Inference Gap: What We Cannot Know

Strategic decision-making is the most classified dimension of any military crisis, and intellectual honesty requires acknowledging that the most important questions about the 2025 conflict may never be answered from open-source materials alone. The gap between what happened and what we know happened is not merely an inconvenience; it is a structural feature shaping how lessons are drawn and how policies are formulated.

Several critical questions remain unanswered. Did Pakistan’s National Command Authority convene, and what options were presented? Did the Strategic Plans Division initiate any warhead-to-delivery-vehicle mating procedures? Were submarines deployed to survivable bastions? Did Indian strategic forces receive readiness orders? Were there moments when either side’s decision-makers believed an attack was imminent? Were there communication failures or near-misses that have not been reported? Did either side misidentify a conventional strike, and how long did any ambiguity persist? Did electronic warfare operations disrupt command-and-control communications at any point during the four days, and if so, were field commanders required to make autonomous decisions about force readiness?

The answers would fundamentally alter the analysis. If both sides maintained peacetime postures throughout, the crisis was dangerous but manageable, and the “worked” interpretation gains strength. If either side shifted to higher readiness, the crisis was more dangerous than observable fighting suggests, and the “failed” interpretation gains urgency. If there were moments of genuine ambiguity, misidentified strikes or communication gaps during which field commanders operated without clear guidance from senior leadership, the crisis may have been far more dangerous than any public analysis captures, and the margin between the ceasefire and catastrophe may have been measured in minutes rather than days.

The inference gap operates differently for different categories of information. Some information, such as force-posture changes, is classifiable and will eventually be declassified as the political sensitivity fades over decades. The Cuban Missile Crisis was widely regarded as the most dangerous moment in the atomic age for decades, but subsequent revelations, including the existence of tactical warheads in Cuba that neither Kennedy nor most Soviet commanders knew about, demonstrated the crisis was far more dangerous than contemporaneous analysis recognized. Other information, such as the subjective perceptions of decision-makers during moments of crisis, may never be recoverable because it was never documented and because memory is unreliable under conditions of extreme stress. A Pakistani field commander who momentarily believed that incoming Indian missiles might be carrying strategic warheads would not necessarily have documented that belief, and the absence of documentation does not mean the perception never occurred. The inference gap for subjective perceptions is permanent, meaning that any analysis of how close the 2025 crisis came to catastrophe is necessarily incomplete in a dimension that may be the most important dimension of all.

The historical precedent for post-hoc revelations changing the assessment of crisis severity should give every analyst pause. Beyond Cuba, the 1983 Able Archer exercise, a NATO military exercise that Soviet intelligence interpreted as possible cover for a Western first strike, was assessed at the time as a minor incident but was later revealed to have triggered genuine preparations for retaliatory launch on the Soviet side. The 1995 Norwegian rocket incident, in which a scientific research rocket was briefly interpreted by Russian radar operators as an incoming American missile, reached the level of President Yeltsin being presented with the launch briefcase before the rocket’s trajectory was correctly identified. In both cases, the initial assessment of crisis severity was dramatically lower than the reality later revealed. The 2025 India-Pakistan conflict almost certainly contains similar hidden dimensions, and the full assessment of how close the crisis came to catastrophic escalation may not be available for years or decades, if it becomes available at all.

The analytical approach this article adopts is to identify what can be inferred from observable actions, public statements, and the structural logic of both sides’ postures while being transparent about the limits of inference. Observable actions, the target selection, the weapons employed, the duration of the fighting, and the terms of the ceasefire, provide a foundation. Public statements during and after the crisis provide additional data, though statements made during a military crisis are shaped by propaganda requirements as much as by operational reality. The structural logic of both sides’ postures provides a framework for inferring what force-posture changes were likely, even when they cannot be confirmed. The resulting analysis is necessarily probabilistic rather than definitive, and readers should treat specific claims about force-posture changes during the crisis as informed inferences rather than established facts.

Where the Comparison Breaks Down

The comparison between the 2025 conflict and previous crises breaks down at several critical points requiring acknowledgment to avoid misleading conclusions.

The Cuba comparison breaks down on three dimensions. First, Cuba involved superpowers with global responsibilities. India and Pakistan in 2025 were regional powers whose conflict, while consequential in its strategic dimension, did not threaten the international system’s fundamental structure. The restraint Kennedy and Khrushchev exercised was shaped by awareness of responsibility for global stability, a responsibility neither Modi nor Pakistan’s leadership carried. Second, Cuba breaks down on information symmetry. Both Cold War powers possessed sophisticated intelligence providing relatively accurate pictures of each other’s deployments and intentions. The United States had U-2 reconnaissance photography showing Soviet missile sites with high resolution. The Soviet Union had intelligence on American military mobilization. This information symmetry, while imperfect, contributed to rational decision-making by reducing the uncertainty that can trigger inadvertent escalation. The India-Pakistan dyad in 2025 possessed less information symmetry, particularly regarding each other’s force postures and command-authority decisions during the crisis. This asymmetry made rational decision-making harder and inadvertent escalation more probable. Third, Cuba breaks down on the question of conflict character. Cuba was a crisis, not a war. No shots were fired between American and Soviet forces. The 2025 conflict was a war, with sustained combat producing casualties on both sides. The escalation dynamics of a crisis, in which both sides maneuver for position through signaling and diplomatic communication, differ fundamentally from the escalation dynamics of a war, in which both sides are simultaneously fighting and trying to prevent the fighting from crossing the catastrophic threshold. The dual requirement of fighting and restraining is far more cognitively and institutionally demanding than the requirement of signaling alone, and the 2025 case demonstrates that this dual requirement can be sustained for at least four days but that the strain on command-and-control systems increases with each passing day.

The Kargil comparison breaks down because both countries’ arsenals in 1999 were small, delivery systems immature, and neither had fully developed doctrine or operational plans. By 2025, both possessed diversified triads with estimated arsenals of 150 to 200 warheads each. The arsenals of 2025 were qualitatively different from 1999, and escalation dynamics between states with mature forces differ from dynamics between states with embryonic capabilities. A state with five warheads and a limited number of aircraft capable of delivering them faces fundamentally different strategic choices than a state with 170 warheads deployable through land-based missiles, aircraft, and submarine-launched systems. The former must consider whether using its limited arsenal achieves any strategic objective; the latter possesses the redundancy and diversity to contemplate tactical use while preserving a strategic reserve. Kargil’s deterrence lessons, derived from an era when both arsenals were small and fragile, may not transfer to an era when both are large and diversified. Kargil also breaks down on geographic scope. Kargil was fought in a single mountainous sector of roughly one hundred kilometers along the LoC. The 2025 conflict involved operations across the entire LoC, along the international border in the Punjab and Rajasthan sectors, and potentially in the maritime domain. The geographic expansion of the 2025 conflict relative to Kargil increased the number of potential escalation flashpoints and reduced both sides’ ability to maintain situational awareness across the entire conflict zone. A localized conflict in a single sector is manageable because senior decision-makers can focus their attention and communication resources on a limited area. A dispersed conflict across a thousand-kilometer front is exponentially harder to control, because the number of independent decision points multiplies and the probability that at least one of those points produces an escalatory action increases.

The Balakot comparison breaks down in duration and intensity. Balakot was a twenty-four-hour strike-and-counterstrike cycle that de-escalated before it could develop the operational momentum of a sustained military campaign. The 2025 conflict lasted four days with sustained multi-domain operations. The difference between a twenty-four-hour exchange and a four-day war is not merely quantitative but qualitative. Extended combat creates escalation pressures that brief exchanges do not: casualty accumulation creating domestic pressure for retaliation, potential munitions depletion pushing toward unconventional alternatives, and operational fog increasing misidentification probability. Balakot proved India and Pakistan could exchange a single airstrikes cycle without escalation. The 2025 conflict proved they could fight a multi-day war without crossing the ultimate threshold. These are different proofs with different implications, and the second should not be treated as merely a scaled-up version of the first. The escalation risks of a four-day war are not four times the risks of a one-day exchange; they are qualitatively different because the operational dynamics of sustained combat, including logistics, reinforcement, intelligence degradation, and command fatigue, introduce variables that a single-day exchange does not produce.

The most significant limitation of the comparative framework is that every previous crisis ended without conventional military engagement between the atomic-armed states, with partial exceptions of Kargil (limited to one geographic sector) and Balakot (limited to one cycle of airstrikes). The 2025 conflict is the first case of sustained, multi-domain conventional warfare between two states possessing thermonuclear weapons. There is no direct historical precedent for what happened in May 2025, which means that the comparative analysis, while useful for identifying continuities and departures from established patterns, cannot fully capture the novelty of the case. The 2025 conflict is not the latest entry in a series of similar events. It is the first entry in a new category of events, one that deterrence theory assumed would never exist.

What the Comparison Teaches

The 2025 India-Pakistan conflict teaches five lessons about deterrence that cannot be derived from any previous case and that should reshape both theory and policy.

The first lesson is that the stability-instability paradox is an operational reality producing sustained conventional warfare. The paradox was identified decades ago but treated as a tendency rather than a law. The 2025 case demonstrates that the paradox can produce high-level conventional conflict, including missile exchanges, not merely the low-level provocations previous cases generated. The paradox operated not as a background condition but as the organizing logic of the conflict: every conventional decision both sides made was enabled by the strategic stability that the arsenals provided and constrained by the catastrophic potential those same arsenals represented. India could launch precision strikes against terrorist camps because the stability-instability paradox guaranteed that Pakistan would not respond with its arsenal to limited counter-terrorism operations. Pakistan could retaliate with artillery and airstrikes because the paradox guaranteed that India’s NFU pledge would hold as long as Pakistan’s response remained conventional. The paradox created a combat zone bounded above by the weapons-use threshold and below by the political minimum of acceptable military action, and both sides fought vigorously within that zone. The policy implication is that arms-control frameworks must address conventional capabilities that deterrence shields enable, because it is at the conventional level, not the strategic level, where the most dangerous fighting will occur.

The second lesson is that deterrence operates as a ceiling rather than a floor. The arsenals prevent weapons use. They do not prevent war. The distinction seems obvious after the fact, but the theory’s strongest versions claimed they prevented all war between armed states, a claim the 2025 case empirically falsifies. Crisis-management mechanisms must be designed for conventional wars, not merely standoffs. The DGMO hotline that contributed to the 2025 ceasefire was a crisis-management tool designed for limited border incidents, not a sustained multi-domain military campaign. If the conflict had lasted longer or escalated further, the hotline’s capacity to manage escalation would have been tested beyond its design parameters. The lesson for the international community is that every dyad of states possessing thermonuclear arsenals needs a crisis-management architecture designed for the worst case scenario that the 2025 conflict demonstrated is possible: sustained conventional war with accumulating casualties, multiple domains of engagement, and compressed decision-making timelines. Currently, no such architecture exists between India and Pakistan. Its absence during the 2025 crisis was a source of risk that both sides and the international community were fortunate to survive.

The third lesson is that domestic political pressure can override deterrence-induced restraint. India’s decision to launch Sindoor was driven by the political impossibility of absorbing Pahalgam without a military response. Deterrence theory assumes leaders will always choose restraint when the alternative carries catastrophic risk, but this assumption fails when the political cost of restraint exceeds the perceived probability of escalation. The Pahalgam attack killed twenty-six people, including foreign tourists, in a spectacularly violent assault on a symbolically important tourist destination. The domestic political environment in India after Pahalgam left the government with no politically viable option other than military action. The choice was not between peace and war; it was between military action with some risk of catastrophic escalation and political collapse with certain consequences for the ruling government. When framed this way, the decision to accept some risk of escalation in exchange for political survival is rational in the narrow sense, even if it is irrational in the broader sense of risking civilizational destruction for political advantage. The lesson is that the domestic political dynamics within states possessing such arsenals are as important to stability as the military balance between them. Any assessment of escalation risk that ignores domestic politics is incomplete, and any crisis-management framework that assumes rational, security-maximizing behavior by leaders under domestic political pressure is built on a foundation that the 2025 case has cracked.

The fourth lesson is that the escalation ladder is not a ladder at all. Kahn’s metaphor implies sequential movement, each rung providing an opportunity to stop. The 2025 conflict demonstrated rung-skipping, jumping from diplomatic protest to homeland missile strikes and bypassing intermediate steps. Rung-skipping eliminates signaling opportunities, compressing the time available for de-escalation. India moved from economic sanctions and diplomatic expulsion directly to cruise-missile strikes against Pakistani territory, bypassing the intermediate rungs of naval blockade, cross-border special-operations raids, limited artillery exchanges, and air patrols that the ladder metaphor suggests would provide opportunities for both sides to signal their intentions and adjust their behavior. The absence of these intermediate steps meant that Pakistan had to respond to a fait accompli rather than to a sequence of signals, reducing Pakistan’s ability to calibrate its response and increasing the probability of overreaction. The lesson for crisis management is that escalation between atomic-armed states may proceed by jumps rather than steps, and that every crisis-management protocol currently in operation, designed for sequential escalation with opportunities for mid-crisis communication, may be obsolete in an era of precision-guided, rapid-tempo military operations.

The fifth lesson is the most consequential. The 2025 conflict demonstrated that a conventional war between atomic-armed states can end without weapons use, but it did not demonstrate that such a war will always end this way. The sample size is one. One case of conventional war that terminated short of catastrophic escalation does not establish a pattern, let alone a law. If the 2025 outcome is treated as evidence that such wars can be managed safely, it will lower the threshold for future conventional action, increasing frequency and, by probability laws, increasing the likelihood that one of them crosses the ultimate threshold. The greatest danger may not be the 2025 conflict itself but the lessons decision-makers draw from it. If the lesson is that atomic-armed states can fight conventional wars without catastrophic consequences, the world has become more dangerous, not less, because the perceived cost of conventional war has decreased while the actual risk of catastrophic escalation remains unchanged. The correct lesson is the opposite: the 2025 conflict succeeded in avoiding catastrophe partly through skill, partly through restraint, and partly through luck, and the share attributable to luck cannot be determined. A policy framework that relies on luck as a component of its risk calculus is not a policy framework at all; it is a gamble that compounds with each repetition.

The 2025 crisis timeline and the detailed signaling analysis provide granular accounts. The lessons of the 2025 conflict and the future of India’s counter-terror doctrine extend this analysis into the policy domain. The foundational question, whether India and Pakistan have entered an era in which conventional wars between atomic powers become periodic features of their relationship rather than unthinkable events, remains the most important unanswered question in contemporary strategic studies.

Frequently Asked Questions

Q: Did India and Pakistan almost have an atomic war in 2025?

The 2025 conflict carried a nonzero probability of catastrophic escalation, but the probability’s magnitude is genuinely uncertain because the most relevant data, force-posture changes during the crisis, remains classified. What can be said with confidence is that the conflict reached a level of conventional intensity never before occurring between atomic-armed states. The theoretical literature identifies multiple pathways through which this fighting could trigger inadvertent escalation, including misidentification of conventional strikes, damage to command-and-control systems, and use-it-or-lose-it pressure on delivery systems. Whether any pathways were activated during the crisis is unknown, but the conditions for their activation were present.

Q: Does the 2025 conflict prove deterrence works?

The “worked” interpretation rests on the observation that the conflict ended without weapons use, proving that the arsenals prevented catastrophe even under extreme conditions. This interpretation is correct as far as it goes but incomplete. Deterrence theory’s strongest versions claim that the arsenals prevent not only weapons use but all war between atomic-armed states. The 2025 conflict falsifies this stronger claim. Whether this represents deterrence “working” or “failing” depends on what one expects the arsenals to deter.

Q: How close did the 2025 crisis come to weapons use?

Observable indicators suggest the crisis remained below the threshold by a margin whose width is uncertain. India’s target selection avoided Pakistani military installations. Pakistan’s conventional response remained within bounds that India’s NFU framework permitted without reconsideration. The ceasefire was achieved within four days, before ammunition depletion or casualty accumulation created irresistible pressure. These indicators suggest a meaningful margin, but one whose size cannot be specified.

Q: What is Pakistan’s deterrence doctrine?

Pakistan’s posture is full-spectrum deterrence, reserving the right of first use if four redlines are crossed: territorial conquest, armed-forces destruction, economic blockade, or political destabilization. The doctrine is deliberately ambiguous about the thresholds defining each redline. Feroz Hassan Khan’s “Eating Grass” provides the most comprehensive account of the doctrine’s origins.

Q: What is India’s No First Use policy?

India’s NFU, codified in 2003, pledges that India will not initiate atomic warfare. The arsenal is reserved exclusively for retaliation. The doctrine promises massive retaliation in response to any first use against India. NFU’s most debated dimension is whether it would survive a scenario in which Pakistan uses tactical weapons against Indian military forces, because massive retaliation against a limited tactical strike would be disproportionate.

Q: Did either side signal readiness during the crisis?

The open-source record provides limited evidence of weapons-related signaling. Neither side publicly invoked its arsenal. Pakistan’s ISPR referenced “full-spectrum deterrent capability” without specifying, a formulation allowing signaling without explicit threats. India’s communications focused on the counter-terrorism framing. The absence of explicit signaling is analytically significant, suggesting both sides were aware that overt threats would raise escalation risk.

Q: Has the 2025 conflict changed deterrence theory?

The 2025 conflict has introduced a new empirical category: sustained conventional warfare between atomic-armed states. Previous cases involved limited engagements that deterrence theorists incorporated without fundamental revision. The 2025 case, with its multi-domain character and unprecedented escalation, demands more than incremental adjustment. The stability-instability paradox has been validated as operational reality. The escalation ladder has been revealed as more permeable than Kahn’s metaphor suggests. The assumption that the arsenals prevent all war between armed states has been empirically falsified.

Q: Could a future India-Pakistan crisis go further?

The structural conditions producing the 2025 conflict persist: Pakistan-based groups continue to target India, India has demonstrated willingness to respond with military force, and both sides possess arsenals capable of destroying each other’s major cities. Each repetition of the crisis cycle has produced a higher military response. The 2001 Parliament attack produced mobilization without engagement. The 2008 Mumbai attacks produced diplomatic fury without military action. The 2016 Uri attack produced cross-LoC surgical strikes. The 2019 Pulwama bombing produced airstrikes inside Pakistani territory. The 2025 Pahalgam massacre produced missile exchanges, aerial combat, and a four-day conventional war. If this escalatory pattern continues, a future crisis could push conventional fighting closer to the weapons-use threshold. The probability of crossing that threshold in any single crisis is low but nonzero, and accumulates across repeated crises.

Q: Were dual-capable aircraft deployed?

Both sides deployed aircraft capable of delivering warheads. India’s Rafales are dual-capable platforms. Pakistan’s F-16s are integrated into its delivery plans. Deploying dual-capable aircraft during conventional combat creates one of Talmadge’s inadvertent escalation pathways: the adversary cannot distinguish between a conventional sortie and a strategic delivery mission based on aircraft type alone.

Q: Did Pakistan’s full-spectrum posture influence India’s decisions?

Pakistan’s posture influenced India by constraining target selection and operational scope rather than preventing military action. India targeted terrorist infrastructure rather than military installations, reflecting implicit acknowledgment that striking military targets could trigger redlines. Pakistan’s doctrine failed at the macro level (it did not prevent Sindoor) but succeeded at the micro level (it constrained the operation’s parameters).

Q: How did the arsenals shape the conflict without being used?

The arsenals shaped the conflict through five mechanisms. First, constraining India’s target selection. Second, preventing India from using ground forces. Third, constraining Pakistan’s retaliation to levels permitting India’s NFU to stand. Fourth, motivating the ceasefire by raising escalation risks. Fifth, attracting international diplomatic pressure from actors intervening because of the catastrophic consequences conventional fighting could produce.

Q: Were submarines deployed?

Neither side has confirmed deploying armed submarines. India’s Arihant-class SSBN is believed operational. Pakistan’s sea-based capability is less developed. If either deployed submarines, the deployments would represent a shift toward survivable second-strike postures, making the crisis more dangerous in one sense (more warheads at sea in uncertain command environments) and less in another (assured second-strike capability reduces first-strike incentives).

Q: Does the 2025 crisis validate or undermine deterrence theory?

The crisis validates the narrow version (arsenals prevent weapons use) while undermining the broad version (arsenals prevent all war). The distinction matters because the broad version has been the basis for decades of strategic policy. If the arsenals only prevent their own use, the implications for military planning, crisis management, and arms control are fundamentally different from those implied by a theory treating them as guarantors of peace.

Q: What role did the United States play?

American diplomatic pressure reportedly contributed to the ceasefire. The United States possesses intelligence capabilities providing visibility into both sides’ posture changes. American involvement was motivated not only by alliance considerations but by the existential risk that weapons use in South Asia would pose to the global strategic order.

Q: What is the escalation ladder and how does it apply?

The escalation ladder, developed by Herman Kahn in 1965, analyzes how conflicts between atomic powers intensify through steps from diplomatic disagreement to all-out warfare. The 2025 conflict climbed higher than any post-Cold War crisis, reaching rungs including homeland strikes and multi-domain conventional warfare. The ladder’s limitation, exposed by 2025, is that real escalation does not proceed one rung at a time.

Q: How did the 2025 conflict differ from Balakot?

The 2025 conflict differed in duration (four days versus twenty-four hours), intensity (sustained multi-domain operations versus a single airstrikes cycle), and weapons employed (precision cruise missiles versus unguided bombs). These differences mean the 2025 conflict occupies a fundamentally different position on the escalation ladder.

Q: Why did deterrence fail to prevent the 2025 conflict?

Deterrence failed because the political costs of restraint exceeded the deterrence-induced costs of action. The Pahalgam massacre created irresistible domestic pressure for a military response, and India calculated that the risk of catastrophic escalation from a limited counter-terrorism strike was acceptable relative to inaction’s political consequences.

Q: What did the conflict reveal about Pakistan’s command and control?

The conflict revealed, through inference, that Pakistan’s command architecture is more resilient and cautious than pessimistic pre-conflict assessments predicted. Pakistan absorbed four days of conventional military pressure without crossing the threshold, suggesting the National Command Authority maintained control and that Pakistan’s redlines are operationally higher than its declaratory posture implies.

Q: What is the significance for the global strategic order?

The 2025 conflict is the most significant development in the global strategic order since the Cold War’s end. It demonstrates that atomic-armed states will fight conventional wars, reveals rapid escalation bypassing intermediate rungs, exposes the gap between declaratory doctrine and operational behavior, and creates a precedent that may lower the perceived cost of conventional military action between armed states.

Q: How has India’s strategic posture changed after 2025?

India has demonstrated willingness to use precision-guided force against Pakistan-based terrorist infrastructure, establishing a new baseline. NFU remains officially in place but with productive ambiguity. Defense procurement has accelerated toward precision-strike capabilities, longer-range munitions, and enhanced command-and-control integration. The shift from strategic restraint to calibrated offensive capability is the most significant doctrinal evolution since 1998.

Q: What is the stability-instability paradox and how did it apply?

The paradox holds that strategic stability paradoxically enables conventional instability by creating a protective shield. It operated with extraordinary clarity in 2025. Both sides’ arsenals created a ceiling preventing weapons use but also a floor of conventional safety beneath which both felt free to operate. India launched missile strikes confident Pakistan would not escalate. Pakistan retaliated confident that India’s NFU protected it. The paradox explains how a crisis between atomic-armed states produced more conventional violence, not less.

Q: Could better crisis communication have prevented the conflict?

Better infrastructure might have slowed escalation but is unlikely to have prevented the conflict entirely. India and Pakistan lack several institutional mechanisms that helped manage Cold War crises: a direct head-of-state channel, pre-negotiated crisis protocols, and agreed rules of engagement. The absence of these mechanisms contributed to escalation speed and ceasefire difficulty.

Q: What lessons should other armed states draw from 2025?

Every state possessing such weapons should draw three lessons. First, the arsenals prevent weapons use but not conventional war, requiring defense planning for sustained conventional combat with similarly armed adversaries. Second, escalation involves dramatic jumps eliminating mid-crisis diplomacy opportunities, requiring crisis protocols designed for rapid escalation. Third, domestic political pressure can override deterrence-induced restraint, making internal political dynamics as important to stability as the military balance.