Security Dilemma

When Safety Breeds Danger

In 1936, the British political scientist John Herz identified a mechanism that would prove central to understanding why states that want only to be safe often end up at war. When one state builds weapons or forms alliances to protect itself, other states cannot be certain whether these measures are genuinely defensive or preparation for aggression. Prudence demands they assume the worst and respond in kind—increasing their own military capabilities to hedge against a threat that may not exist. The first state, observing its neighbor’s buildup, concludes that its original fears were justified and increases its efforts further. The cycle continues, with each side’s defensive measures appearing threatening to the other, until the spiral produces the very conflict both sides sought to avoid. This mechanism—the security dilemma—is one of realism’s most important contributions to understanding why international politics is so prone to conflict, even among states with no aggressive intentions.

The security dilemma matters because it explains something that simpler theories of conflict cannot: why war can arise from fear rather than ambition, from misperception rather than malice, from the structure of the international system rather than the character of individual states. If wars occurred only because aggressive leaders sought conquest, preventing them would be straightforward—identify the aggressors and stop them. But the security dilemma shows that war can occur between status quo powers—states seeking only to preserve what they have—because the international system’s anarchic structure makes their defensive measures indistinguishable from offensive preparations.

The Mechanism

Anarchy and Uncertainty

The security dilemma arises from two conditions that define international politics:

Anarchy: There is no world government, no international police force, no authority that can guarantee a state’s safety. States exist in what realists call a “self-help” system—they must provide for their own security because no one else will. This structural condition creates a permanent incentive to arm and ally, regardless of anyone’s intentions.

Uncertainty about intentions: States cannot know with certainty whether other states’ military capabilities are intended for defense or offense. A new missile system might be designed to deter attack—or to enable one. A military alliance might be purely defensive—or it might be the prelude to aggression. Even intelligence agencies cannot reliably determine another government’s true intentions, which may change with new leaders, new circumstances, or new opportunities.

The combination of these two conditions produces the dilemma: a state that arms for defense may inadvertently signal hostile intent, provoking the very threat it sought to deter. The result is an arms race spiral in which both sides become less secure the more they spend on security.

The Spiral Model

Political scientist Robert Jervis formalized the security dilemma’s logic in his influential 1978 article “Cooperation Under the Security Dilemma.” Jervis identified two variables that determine the severity of the dilemma:

The offense-defense balance: When offensive military capabilities are dominant—when it is easier to attack than to defend—the security dilemma is most severe. States must arm heavily because vulnerability to attack is high, and each side’s buildup appears threatening because offensive capabilities dominate. When defense is dominant—when fortifications, terrain, or technology favor the defender—the dilemma is mitigated. States can protect themselves without threatening others.

Offense-defense distinguishability: When offensive and defensive weapons are indistinguishable—when the same tanks, aircraft, and missiles serve both purposes—the security dilemma is severe because states cannot signal defensive intent through their force structure. When offensive and defensive weapons are distinguishable—as when coastal fortifications clearly differ from expeditionary forces—the dilemma is reduced because states can demonstrate defensive intent through their military posture.

The worst case—offense dominant and indistinguishable—creates the most dangerous security environment. Both sides must arm heavily, and neither can credibly signal defensive intentions. The best case—defense dominant and distinguishable—allows states to protect themselves without threatening others.

Historical Examples

The Anglo-German Naval Race (1898-1914)

The naval arms race between Britain and Germany before World War I is a textbook case. Germany’s decision to build a battle fleet—justified by Kaiser Wilhelm II and Admiral Tirpitz as necessary for Germany’s status as a world power and the defense of its colonial interests—was perceived by Britain as an existential threat. British security depended on naval supremacy; a German fleet capable of challenging the Royal Navy threatened Britain’s food supply, its empire, and its survival.

Britain responded with the revolutionary HMS Dreadnought (1906), rendering all previous battleships obsolete and resetting the competition. Germany accelerated its own building program. Each escalation confirmed the other side’s fears. By 1914, the naval race had poisoned Anglo-German relations so thoroughly that what might have been a manageable diplomatic competition became an existential confrontation. Neither side wanted war—but the security dilemma made war more likely.

The Cold War Arms Race

The Cold War produced the most dangerous security dilemma in history. The United States and Soviet Union, emerging from World War II as the world’s only superpowers, faced a classic spiral:

The Soviet Union, devastated by a war that killed 27 million of its citizens, sought security through territorial buffers in Eastern Europe and military buildup. The United States, confronting Soviet expansion, perceived aggressive intent and responded with Containment, the NATO alliance, and a massive nuclear arsenal.

The nuclear arms race exemplified the dilemma’s logic. The United States built hydrogen bombs because it feared Soviet nuclear capability; the Soviet Union built hydrogen bombs because it feared American nuclear capability. By the mid-1980s, the two sides possessed over 70,000 nuclear warheads—enough to destroy civilization many times over. Neither side wanted Armageddon; both achieved a situation where Armageddon was one miscalculation away.

The race also extended to conventional forces, missile defense, space-based weapons, and submarine technology. Each advance by one side compelled a response by the other. The security dilemma did not cause the Cold War—genuine ideological hostility and geopolitical competition drove it—but it amplified the competition far beyond what rational calculation of threats required.

US-China Competition

The contemporary Great Power Competition between the United States and China exhibits classic security dilemma dynamics. China’s military modernization—the world’s largest navy by vessel count, the DF-21D “carrier killer” missile, the construction of artificial islands in the South China Sea—is described by Beijing as defensive: protecting Chinese Sovereignty, securing trade routes, and deterring foreign intervention. The United States and its allies perceive these same capabilities as offensive: designed to project power, intimidate neighbors, and establish Chinese hegemony in the Western Pacific.

Washington’s response—the Quad, AUKUS, enhanced deployments to Japan and the Philippines, freedom of navigation operations in the South China Sea—is described by the United States as defensive: maintaining the rules-based order, deterring aggression, and reassuring allies. Beijing perceives these measures as containment: an American strategy to encircle China, deny it its legitimate interests, and preserve American dominance.

Each side’s defensive measures fuel the other’s threat perception. The result is an accelerating competition that neither side may want but neither can safely stop—the security dilemma operating at the grandest scale.

Can the Security Dilemma Be Mitigated?

Arms Control

The Cold War demonstrated that arms control, while imperfect, can moderate the security dilemma. The ABM Treaty (1972) limited missile defenses, preserving mutual vulnerability and the logic of deterrence. SALT and START reduced nuclear arsenals. The INF Treaty (1987) eliminated an entire class of weapons. These agreements did not resolve the underlying competition, but they slowed the spiral and reduced the risk of accidental war.

The deterioration of arms control since 2019—American withdrawal from the INF Treaty, the expiration of the Open Skies Treaty, uncertainty about New START’s successor—has removed constraints that took decades to build. The return of unconstrained arms competition between three nuclear powers (the US, Russia, and China) represents a failure of the mechanisms that previously moderated the dilemma.

Defensive Military Postures

States can mitigate the security dilemma by adopting military postures that are clearly defensive. Switzerland’s militia-based army, designed for territorial defense and incapable of power projection, signals defensive intent without ambiguity. Japan’s postwar constitution (Article 9) renounced war and limited its Self-Defense Forces to defensive capabilities—though these constraints have been progressively reinterpreted.

In practice, most major powers maintain forces capable of both offense and defense, making it difficult to signal purely defensive intent. China’s carrier-building program and long-range missiles serve both defensive and offensive purposes. NATO’s eastward expansion can be read as defensive (protecting new members) or offensive (encircling Russia). The ambiguity is structural, not a matter of messaging.

Transparency and Communication

Direct communication between adversaries can reduce misperception. The “hotline” established between Washington and Moscow after the Cuban Missile Crisis enabled leaders to communicate during crises, reducing the risk that misunderstanding would escalate to war. Military-to-military contacts, shared early warning data, and confidence-building measures (advance notification of military exercises, observer exchanges) can moderate the security dilemma by reducing the uncertainty that fuels it.

Limitations of the Concept

Critics argue that the security dilemma is sometimes invoked too broadly—used to explain conflicts that are better understood as the product of genuine aggression rather than mutual fear. Russia’s invasion of Ukraine in 2022 has been interpreted by some through the security dilemma lens (NATO expansion threatening Russia’s security), but others argue that this framing excuses Russian aggression and ignores the agency of the states that chose to join NATO precisely because they feared Russia.

The distinction matters. Not every arms buildup is defensive; not every conflict is a tragic misunderstanding between status quo powers. Some states genuinely seek expansion, domination, or revision of the international order. The security dilemma is a powerful explanatory tool, but it should not obscure the reality that some conflicts arise from aggression rather than mutual fear.

Sources & Further Reading

  • Perception and Misperception in International Politics by Robert Jervis — The foundational work on how cognitive biases and the security dilemma produce conflict between states that may prefer peace, with extensive historical case studies.

  • Cooperation Under the Security Dilemma by Robert Jervis (World Politics, 1978) — The seminal article that formalized the offense-defense balance and distinguishability variables, establishing the security dilemma as a central concept in international relations theory.

  • The Tragedy of Great Power Politics by John J. Mearsheimer — Offensive realism’s argument that the security dilemma is irresolvable because states can never be certain of others’ intentions, making power maximization the only rational strategy.

  • The Security Dilemma: Fear, Cooperation, and Trust in World Politics by Ken Booth and Nicholas J. Wheeler — A comprehensive examination of the concept’s theoretical foundations, historical applications, and implications for contemporary security policy.