Article 2(4) of the UN Charter prohibits every member state from using or threatening military force against another country, with only two recognized exceptions: authorization by the UN Security Council under Chapter VII, and self-defense against an armed attack under Article 51.1United Nations. United Nations Charter (Full Text) The rule is short. The disputes over what “force” means, when self-defense begins, and whether any other exceptions exist have filled decades of state practice and case law.
What the Rule Actually Says
The text requires member states to “refrain in their international relations from the threat or use of force against the territorial integrity or political independence of any state, or in any other manner inconsistent with the Purposes of the United Nations.”1United Nations. United Nations Charter (Full Text) That closing clause is not filler. Even a military action that does not aim to seize land or topple a government still violates Article 2(4) if it cuts against the UN’s broader purpose of maintaining peace.
The prohibition binds all 193 member states and applies whether or not a formal declaration of war exists. Cross-border raids, naval blockades, bombardment, troop deployments without consent, and temporary occupations all fall inside its scope. The International Law Commission has identified the ban as a textbook example of jus cogens: a peremptory norm so fundamental that no bilateral treaty or regional agreement can override it.2United Nations. Draft Conclusions on Identification and Legal Consequences of Peremptory Norms of General International Law
What Counts as “Force”
The consensus that emerged from drafting, and has held since, is that “force” here means armed or military force. Some delegations pushed to include economic coercion, and lost. The prohibition instead covers the full range of military conduct, from full-scale invasion to targeted strikes and cross-border abductions. In 1974, the UN General Assembly adopted a Definition of Aggression that listed qualifying acts: invasion, bombardment, blockade, attacks on another state’s military forces, and sending armed bands or mercenaries to carry out attacks of sufficient gravity.3Institute for International Law and Justice. Definition of Aggression, General Assembly Resolution 3314
Economic sanctions and diplomatic pressure sit outside Article 2(4), but they are not lawless. The 1970 Declaration on Friendly Relations states that no country may use economic or political measures to coerce another state into surrendering its sovereign rights.4United Nations. Declaration on Principles of International Law Concerning Friendly Relations and Co-operation Among States A punitive trade embargo may violate the principle of non-intervention without amounting to a use of force. Two different legal regimes, close but not overlapping.
The gravity of the operation does not change the analysis. Some commentators have argued that a single targeted killing, a brief hot-pursuit incursion, or an abduction of one individual falls below a threshold Article 2(4) was meant to address. That view has not gained broad acceptance. The better reading, backed by most state practice, is that Article 2(4) has no minimum-severity requirement. A drone strike that kills one person on foreign soil without consent is still a use of force against that state’s territorial integrity.
Cyber Operations
Whether a cyberattack can qualify as a use of force is one of the most active questions in the field. The prevailing analytical framework, developed in the Tallinn Manual 2.0 (a non-binding expert study commissioned by the NATO Cooperative Cyber Defence Centre of Excellence), applies a “scale and effects” test: a cyber operation is a use of force when its consequences are comparable to a conventional military attack. Relevant factors include severity of harm, immediacy, directness, physical damage, casualties, and military character. A cyberattack that destroys a power grid and causes deaths would almost certainly qualify. Espionage that copies data without disrupting systems almost certainly would not. France, the Netherlands, and Norway have indicated that sufficiently disruptive cyber operations could qualify even without physical destruction.
Threats of Force
Article 2(4) does not wait for shots to be fired. A threat exists when a state signals its readiness to use armed power to force another country to change course. Ultimatums backed by promises of strikes, coercive troop mobilizations along a neighbor’s border, or provocative naval maneuvers in contested waters can all count. The International Court of Justice confirmed the principle in its 1996 Advisory Opinion on the Legality of the Threat or Use of Nuclear Weapons: if using force in a given scenario would violate the Charter, threatening to use it is equally unlawful.5International Court of Justice. Legality of the Threat or Use of Nuclear Weapons
Territorial Integrity and Political Independence
The two interests Article 2(4) names are not decorative. Territorial integrity means a state’s recognized borders cannot be redrawn by military force. No country can lawfully annex another’s land, carve off a region by backing a separatist movement with troops, or impose an occupation to shift boundaries. Conquest has been flatly illegal since 1945.
Political independence protects a state’s right to govern itself without foreign military coercion. Using armed forces to overthrow another government, install a preferred leader, or dictate domestic policy at gunpoint strikes at the core of what the rule exists to prevent. The two protections work in tandem: one shields physical territory, the other shields decision-making authority.
Use of Force Is Not Always an Armed Attack
One of the most consequential distinctions in this area is the line between a “use of force” (prohibited by Article 2(4)) and an “armed attack” (which triggers self-defense under Article 51). Not every illegal use of force qualifies as an armed attack. The ICJ drew this line in the 1986 Nicaragua v. United States case, distinguishing “the most grave forms of the use of force (those constituting an armed attack) from other less grave forms.”6Institute for International Law and Justice. Case Concerning Military and Paramilitary Activities In and Against Nicaragua
The practical consequence is significant. The Court found that arming and training rebel groups, as the United States did with the Nicaraguan contras, was a use of force but did not amount to an armed attack. Nicaragua could not invoke collective self-defense in response to that specific conduct.6Institute for International Law and Justice. Case Concerning Military and Paramilitary Activities In and Against Nicaragua Merely funding insurgents, the Court added, did not even reach the level of a use of force, though it was an unlawful intervention. A state suffering a lesser use of force can pursue diplomatic remedies and Security Council action. It cannot shoot back and call that self-defense.
Exception One: Security Council Authorization
The first path to lawful force runs through the UN Security Council. Under Chapter VII, the Council can determine that a situation is a threat to the peace, a breach of the peace, or an act of aggression.7United Nations. UN Charter, Chapter VII: Action with Respect to Threats to the Peace, Breaches of the Peace, and Acts of Aggression Once it does, military force is not the first step. Article 41 lets the Council impose non-military measures first: economic sanctions, arms embargoes, travel bans, severance of diplomatic relations.8United Nations. Charter of the United Nations – Article 41
Only when those measures are inadequate or have failed can the Council authorize military action under Article 42. That article allows air, sea, and land operations “as may be necessary to maintain or restore international peace and security.”7United Nations. UN Charter, Chapter VII: Action with Respect to Threats to the Peace, Breaches of the Peace, and Acts of Aggression The 1950 Korean War resolution and the 1990 authorization to expel Iraq from Kuwait are the textbook examples. In each case, a formal Security Council resolution supplied the legal basis for coalition operations.
The system has a built-in weakness: the veto. Any of the five permanent members—the United States, United Kingdom, France, Russia, and China—can block a Chapter VII resolution no matter how the other members vote. When a permanent member is party to a conflict or has strategic reasons to shield an ally, the Council is paralyzed. That happens often, and it explains much of the distance between the Charter on paper and the Charter in practice.
Exception Two: Self-Defense Under Article 51
The second path is the inherent right of self-defense. Article 51 preserves this right when “an armed attack occurs against a Member of the United Nations,” and only “until the Security Council has taken measures necessary to maintain international peace and security.” Self-defense is meant to be a bridge, not a substitute, for collective security. States exercising the right must report their actions to the Security Council immediately.9United Nations. Charter of the United Nations – Article 51
The right covers both individual and collective self-defense. A state under attack can defend itself, and other states can come to its aid at its request. But the trigger is high: an actual armed attack. Lesser forms of hostile conduct, including arming insurgents, do not meet the threshold and do not activate Article 51.
Necessity and Proportionality
The Charter’s text does not spell out limits beyond the armed-attack trigger and the reporting requirement. Customary international law fills the gap with two constraints. The response must be necessary: no realistic peaceful alternative was available, and the threat demanded an immediate military reply. The response must also be proportionate: the defending state cannot exceed what is required to repel the attack. A country that suffers a limited border incursion and responds by invading and occupying the attacking state’s capital has almost certainly gone too far. The ICJ has consistently treated both requirements as binding customary law.
Anticipatory Self-Defense
Whether a state can act before an armed attack lands is one of the most contentious questions in this field. The Charter’s text says the right arises “if an armed attack occurs,” which on a strict reading requires waiting for the blow. Many states and scholars argue that waiting for the first strike in the age of ballistic missiles and weapons of mass destruction is suicidal, and that self-defense must include responding to imminent attacks. This position draws on the 1837 Caroline incident between Britain and the United States, which produced a widely cited formula: anticipatory force is permissible only when the threat is instant and overwhelming, there is no moment for deliberation, and the response is proportionate.
The ICJ has never directly endorsed anticipatory self-defense, and its case law suggests reluctance to expand Article 51 beyond its text. Academic consensus is growing that a constrained right to act against genuinely imminent attacks exists under customary law, but the meaning of “imminent” remains contested. The further a state moves from responding to an attack in progress toward striking preemptively against a speculative future threat, the weaker its legal footing.
Non-State Actors and the “Unable or Unwilling” Doctrine
Article 51 was drafted with state-on-state attacks in mind. Whether self-defense reaches non-state armed groups operating from another country’s territory is a question the Charter does not answer directly. After September 11, 2001, a broad consensus formed that self-defense can be exercised against non-state groups responsible for armed attacks of sufficient scale, even when the host state did not direct or control the group’s actions.
The harder question is what happens when the host state is not complicit but simply cannot suppress the threat. The “unable or unwilling” doctrine holds that a defending state may use force on another state’s territory when that state cannot or will not address the armed group operating within its borders. Over a dozen states, including the United States, United Kingdom, Turkey, Australia, and several European nations, have invoked this doctrine in Article 51 notifications to the Security Council. Critics counter that it effectively allows one state to judge another’s sovereignty away based on a subjective assessment of “inability.” The doctrine remains deeply contested.
Contested Justifications Beyond the Two Exceptions
Several other justifications for force have been advanced over the decades. None commands the same legal consensus as Chapter VII authorization or Article 51 self-defense, and each sits in tension with Article 2(4).
Humanitarian Intervention and R2P
Whether states can use force to stop mass atrocities without Security Council authorization is one of the most polarizing questions in international law. Proponents argue that Article 2(4) cannot have been meant to shield governments committing atrocities against their own populations, and that a narrow right of humanitarian intervention exists when the Council is deadlocked. Some read the “territorial integrity or political independence” language as permitting brief interventions that do not aim to annex territory or change a political system.
The 2005 World Summit Outcome document tried to chart a middle path with the Responsibility to Protect framework. All UN member states agreed that each state bears primary responsibility for protecting its own population from genocide, war crimes, ethnic cleansing, and crimes against humanity, and that the international community should act “through the Security Council, in accordance with the Charter, including Chapter VII” when a state manifestly fails.10United Nations. 2005 World Summit Outcome R2P as formally adopted routes enforcement through the Council, not around it. Unilateral humanitarian intervention without Council authorization remains illegal under the prevailing view, and the vast majority of governments have formally rejected any unilateral right to intervene on humanitarian grounds.
Rescue of Nationals Abroad
States have occasionally used force to evacuate their own citizens from dangerous situations abroad: hostage rescues, emergency evacuations during civil wars, and similar operations. The legal theory frames these as limited exercises of self-defense that do not target the host state’s territorial integrity or political independence, provided the operation is confined to removing the nationals and stays within necessity and proportionality. Whether they genuinely fit within Article 51 or represent a tolerated violation of Article 2(4) depends on who is doing the analysis. Rescuing states invoke self-defense; states whose territory is entered call it a sovereignty violation.
Intervention by Consent
When a state’s recognized government invites foreign military forces onto its territory, the intervention generally does not violate Article 2(4), because it does not operate “against” the host state. This is one of the most common bases for foreign military presence worldwide. The consent must be genuine, freely given, and issued by an authority recognized under international law as competent to speak for the state. Where consent becomes murky—a government contested by rival factions, a leader whose legitimacy is in dispute—the legal foundation crumbles quickly. A government also cannot use consent to authorize acts that would themselves violate international law.
The Distance Between Law and Practice
The prohibition binds all 193 UN member states and carries the highest possible status in international law.11United Nations. About Us Violations can produce international condemnation, ICJ proceedings, General Assembly resolutions, and Security Council sanctions. Enforcement has always been the system’s weakest link. The same veto that can block authorization of collective force can block condemnation of a permanent member’s own aggression.
The prohibition is not empty for that reason. It shapes how states justify their actions. Even governments that violate Article 2(4) almost always claim an exception rather than argue the rule does not apply to them. That aggressors feel compelled to invoke self-defense or humanitarian necessity, however implausibly, confirms that the norm still functions as the baseline against which international conduct is measured. The ongoing debates over cyber operations, non-state actors, and humanitarian intervention are not signs the rule is failing. They are signs the international community is still working out how a text written in 1945 applies to threats its drafters never imagined.