Global catastrophe seems to have been narrowly avoided this past weekend, after India and Pakistan agreed to a ceasefire following the most tense military confrontations between the two states in decades. While any ceasefire is inherently fragile and vulnerable to accusations of violations, both sides have stepped back from the brink and seem committed to de-escalation. Any outbreak of violence between two nuclear-armed states is particularly fraught, and the world is right to pay attention. Yet, fortunately, no instance of such confrontation has ever resulted in a nuclear exchange. This is explained by the phenomenon that international relations theorists and security scholars term the stability-instability paradox: Because nuclear weapons fundamentally change the strategic logic of conflict, they may prevent major conflict between two nuclear-armed actors while simultaneously increasing the amount of minor conflict.
Beyond the latest episode, this phenomenon explains the continuation and outcome of much of Pakistan and India’s rivalry since Pakistan’s successful nuclear tests in 1998. Indeed, this is not the first time since they have both demonstrated nuclear capabilities that India and Pakistan have fought each other. Only a year after Pakistan first tested a nuclear weapon, India and Pakistan fought the brief Kargil War over Kashmir. There have been several flare-ups since then, the most recent of which was in 2019. Nor is this the only example of two nuclear-armed states using force against each other: The Soviet Union and China fought a brief but bloody border war in 1969, and more recently Indian and Chinese forces skirmished over their Kashmir border in 2020 and 2021.
Crises between nuclear states are never guaranteed to end so peacefully. They still require delicate diplomacy to manage and de-escalate tensions. Yet, through a better understanding of the strategic logic of nuclear and conventional conflict, we can breathe a sigh of relief at this apparent latest validation of the stability-instability paradox.
Logics of Military Force
There is often a presumption that the risk of nuclear war will prevent two nuclear-armed states from fighting each other directly due to the logic of mutually assured destruction, which many scholars cite as a major reason why there were no large-scale direct confrontations between Soviet and American forces during the Cold War and as a reason why a direct war between China and the United States is unlikely. This is because nuclear-armed powers have sufficient variety in their nuclear weapon delivery systems for their arsenals to survive and deliver a response strike if they were struck first. This, coupled with the fact that a single nuclear weapon has disproportionately devastating attack potential, makes nuclear weapons effectively impossible to defend against and nuclear wars nigh unwinnable. Accordingly, the strategic logic of nuclear weapons is to avoid war altogether.
So, when nuclear-armed states use force against each other, does it mean that deterrence has broken down? Yes and no, with this duality speaking to what underlies the stability-instability paradox: a choice between two different strategic logics. The strategic logic of nuclear weapons works to deter threats against vital interests or state survival. However, because states know that using a nuclear weapon is almost guaranteed to invoke a response in kind, they cannot credibly threaten to use nuclear weapons against attacks that do not reach that threshold. Neither side in a conflict wants to use nuclear weapons, and so will only do so in the most dire circumstances to protect its most fundamental interests.
This means states can safely use conventional force to try to achieve lesser interests, where the strategic logic of nuclear weapons breaks down. There is no single logic of conventional force—it can contribute to compellence and deterrence as well as to brute force outcomes—but overall it involves a conscious decision to signal that a conflict’s stakes do not rise to the level of an existential threat or vital interest. This sets up a situation in which a state can deter another state from threatening its vital interests, while simultaneously fighting that same state over lesser issues using conventional weapons. As long as both states know how the other defines the boundary between the two, they can fight conventionally over anything that does not rise to the level for which they are willing to risk mutually assured destruction. In short, strategic logic operates differently for high- and low-stakes issues, with high-stakes issues favoring nuclear deterrence. Conversely, lower-stakes disputes enable conventional fighting, or even make it likely.
It would be a mistake to automatically infer a cycle of escalation that ends with a nuclear exchange, even when overt conflict breaks out between nuclear-armed states. This would mean that every issue the states would be willing to fight over is an existential one, which is simply not true for the vast majority of war decisions that states have made in the nuclear era. On the contrary, most wars are so-called wars of choice, involving issues for which the combatant states have limited relative intensities of interest. In such wars—which, by definition, fall short of the existential threshold—there is only so much states are willing to spend in terms of blood and treasure to win, with the fighting itself mostly being useful to clarify expectations for both sides on what the outcome of the fighting would be. Once that probable outcome is clear to both sides, there is an incentive to negotiate the war’s end rather than to continue fighting. Stated simply, neither side’s commitment would rise to the imperative to win at all costs.
The only difference with nuclear-armed states in wars of choice is that they self-select into the conventional weapons logic; in choosing the war, they choose the strategic logic of conventional weapons by default. There are some vital interests that nuclear-armed states would choose to secure through nuclear deterrence and other interests that they would choose to secure through conventional deterrence—and if deterrence fails, fight a war over using conventional means. It stands to reason, then, that nuclear-armed states do not truly contemplate nuclear escalation to secure interests that they have remanded to the conventional realm.
Residual Risks
The risk of escalation is not zero, however, and the self-selection into conventional weapons logic is not all-comforting. A primary concern is the relatively sparse empirical support for the strategic logic of nuclear weapons. While there are good reasons to think nuclear weapons have successfully deterred great power conflict, this nuclear peace may also be a function of the small number of nuclear states. The longer-established nuclear states have historically exercised great restraint when it comes to their nuclear arsenals. Even when retaliating or defending against attacks by nonnuclear adversaries, nuclear powers have held back from using all available force at their disposable, as evidenced by the United States after the 9/11 attacks, Russia during its current war with Ukraine, and Israel’s muted response to recent Iranian ballistic missile attacks.
Nonetheless, the successful deterrence of nuclear war during the Cold War was not necessarily easy or a foregone conclusion. As Thomas Schelling emphasizes, conflict in a nuclear world is fundamentally a contest of wills and risk-taking. The only way to win in a world characterized by mutually assured destruction is to get the other side to back down by convincing its leaders that you have a stronger interest than they do (despite the self-selection dynamic described above). This is often accomplished by demonstrating your willingness to actually use your nuclear weapons, which can of course heighten the risk of nuclear war. Managing this risk requires not only a shared interest in avoiding nuclear war, but also a shared understanding of how to signal red lines and resolve.
As nuclear weapons have proliferated to states with less stable governments or more hostile relations with neighbors, however, the risks of misunderstandings, accidental use, and imprudent escalation understandably amplify concerns that a nuclear war of some scale is simply a matter of time. New and aspiring nuclear-armed states may lack sufficient experience developing shared understandings with their adversaries, or may not have the institutional safeguards to insure against accidental or unauthorized deployments. In this context, India and Pakistan may prove to be the most responsible and rational of new and aspiring members to the nuclear club.
Lastly, the development of tactical nuclear weapons blurs the line between where each strategic logic operates. While proponents believe they offer another mechanism to signal rising stakes without escalating immediately to mutually assured destruction, critics have raised important questions about whether such a message can be sent clearly amid the fog of war. In this case, it would be difficult for either side to know whether the use of a tactical nuclear weapon signals an immediate shift toward the high stakes of nuclear logic or not.
Coming to an Understanding
States have always used a variety of means to indirectly communicate their intentions and capabilities. Often, decisions about when, where, and how military force is used—or not used—are intended to convey a message about what will come next, if the situation does not change. For example, while it is easy to assume that Iran’s ineffectual strikes against Israel merely indicate military weakness, some argue that they were carefully calculated to demonstrate precision-targeting capabilities without further raising the stakes by creating more casualties.
Nuclear crises are no different. Analysts have paid careful attention not only to Russian President Vladimir Putin’s language around nuclear weapons, but also when and how he has mobilized nuclear forces. These actions can serve as signals of the intensity of his interest in the conflict to other actors—especially the United States—with the goal of deterring their deeper involvement. Likewise, India’s mobilization of its Territorial Army, while appearing escalatory, may simultaneously have sent strong signals of resolve that ultimately helped bring the conflict to a resolution. It is notable (and fortunate) that neither India nor Pakistan felt the need to signal their willingness to use nuclear weapons during this most recent crisis. However, as the example of Putin’s nuclear signaling shows, there would have been plenty of other opportunities to signal that the stakes were rising to nuclear levels.
This all presumes that strategic logics are only shaped endogenously by technology and environment, however. A concern that should animate us all is the possibility that strategic logics are shaped exogenously by the international order as well. The current American-built liberal international order is undergoing a period of profound challenge from other great powers. Historically, transitions of international orders have been violent, and only since World War II have they encompassed the possibility of multiple strategic logics that are now available to nuclear-armed states. If order shapes logic, then at some point the continued challenge to and evolution of the extant international order could usher in a new strategic nuclear logic—an invalidation of the stability-instability paradox—that would not be known until, say, India and Pakistan have their next crisis. The sigh of relief after a successful future de-escalation may not come as quickly and fully as it did this time.
Dr. Max Margulies is the chief research officer of the Modern War Institute at West Point.
Colonel Patrick Sullivan, PhD, is the director of the Modern War Institute at West Point.
The views expressed are those of the authors and do not reflect the official position of the United States Military Academy, Department of the Army, or Department of Defense.