Chapman Law Review
THE ILLEGALITY OF THE U.S. POLICY OF PREEMPTIVE SELF-DEFENSE UNDER INTERNATIONAL LAW
Copyright © 2005 by Chapman Law Review; Chris Bordelon
In the aftermath of the September 11, 2001 attacks in the United States, domestic and foreign political actors who might otherwise have balked if the world’s lone superpower claimed a broad right to use force abroad instead offered the U.S. a strong showing of goodwill and support. Perhaps interpreting this solidarity as willingness to support American uses of force abroad regardless of their permissibility under international law, the Bush administration articulated a provocative interpretation of the right of self-defense recognized in the United Nations Charter. In a formal policy statement, the administration claimed that this right, which is an exception to the general prohibition of the use of force in international relations, justified the use of preemptive force against “enemies” of the United States.
The Bush administration’s assertion of a right of preemptive self-defense, along with its reliance on this purported right to justify its use of force against Iraq, has drawn considerable criticism on international law grounds. It is debatable whether anticipatory self-defense is permissible under international law. However, the Bush administration’s preemption strategy calls for the use of force beyond the scope of the right of self-defense, even if the right is broadly construed.
The Bush administration released its formal policy statement, “The National Security Strategy of the United States of America” (“NSS”), in September 2002. This document outlines various considerations that are said to guide the administration’s foreign policy decisions. Of particular interest is the document’s treatment of two categories of “enemies:” “terrorists” and “rogue states.” The NSS contains no specific definition of the term “terrorists.” However, the term includes states that “knowingly harbor or … aid” terrorists. The NSS defines “rogue states” as those states with characteristics that match certain enumerated criteria and explicitly includes Iraq and North Korea in this category. The document argues that the emergence of these “new deadly challenges” has effected a “profound transformation” in the “security environment,” and has brought about a “new world” in which “the only path to peace and security is the path of action.” In the context of the NSS, the noun “action,” the verb “act,” and various derivations of these words were used euphemistically to refer to the use of military force by the U.S. against its “enemies.”
The “action” contemplated against terrorists and rogue state is not limited to deterrence of and response to the use or threat of force by these groups, but includes action “against such emerging threats before they are fully formed.” The U.S. will “not hesitate to act alone” to prevent terrorists and rogue states from attacking or threatening to attack, and will “exercise [its] right of self-defense,” as the administration understands it, “by acting preemptively.” The U.S. will “no longer solely rely on a reactive posture as [it did] in the past,” nor will it “let … enemies strike first,” or “remain idle while dangers gather.” Thus, under the NSS, military force is to be used preemptively to “forestall or prevent … hostile acts by our adversaries.” While the administration suggested at least one other justification for using force against Iraq, it relied in part on its asserted right to preemptive self-defense. The commencement of U.S. efforts to put its theory of preemptive self-defense into practice warrants an examination of whether self-defense is a valid justification for using force in these circumstances.
The authors of the NSS believe that the preemptive use of force is compatible with international law. The NSS states that, “[f]or centuries, international law [has] recognized that nations need not suffer an attack” before using force in self-defense “against forces that present an imminent danger of attack.” Preemption, the document asserts, has been regarded as legitimate under international law when undertaken with respect to an “imminent threat–most often a visible mobilization” of military forces. Without further explanation of what “imminent threat” means, the NSS asserts that “[w]e must adapt the concept of imminent threat to the capabilities and objectives of today’s adversaries.” “The greater the threat” posed by these adversaries, the “more compelling the case for taking anticipatory action … even if uncertainty remains as to the time and place of the enemy’s attack.”
The NSS’s brief mention of the legality of the new U.S. strategy provides an incomplete analysis of the relevant norms. The NSS treats the permissible temporal scope of self-defense as a matter of settled law. It thus ignores an ongoing scholarly and international debate concerning when force may first be used. The document also fails to fully consider the application of two customary international law principles, necessity and proportionality, to uses of force called for by the preemption strategy. Moreover, the NSS does not indicate the source of the authority of the U.S. to unilaterally “adapt the concept of imminent threat” to suit its present needs. Analysis of these gaps in the NSS’s consideration of relevant international law suggests that the administration’s confidence in the legality of preemptive self-defense is misplaced.
II. THE UNITED NATIONS CHARTER AND A STATE’S RIGHT OF SELF-DEFENSE
Scholars regard the United Nations Charter (Charter) as the “starting point” for analyzing the relevant norms restricting the ability of states to threaten or use force. The Charter’s text amply demonstrates its drafters’ fundamental concern with limiting instances in which the use of force in international relations would be considered legally permissible. The preamble begins with an expression of the signatories’ determination “to save succeeding generations from the scourge of war,” and describes the “principles and … methods” contained in the substantive portion of the Charter as being aimed at “ensur[ing] … that armed force shall not be used, save in the common interest.” The first of the “Purposes of the United Nations” listed in Article 1 is “[t]o maintain international peace and security.” The second Purpose commits the U.N. to “take … appropriate measures to strengthen universal peace.”
A. Article 2(4)’s Prohibition on the Use of Force
Article 2(4) sets forth a principle that has been described as “the heart” of the Charter, requiring 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.” Other provisions of the Charter require states that become parties to international disputes to “resolve all their disputes by peaceful means.” However, Chapter VII of the Charter contains two exceptions to the general prohibition of the use of force. The first exception allows for the use of force after the Security Council has “determine[d] the existence of any threat, breach of the peace, or act of aggression” and “decide[d]” that “measures shall be taken in accordance with Article … 42.” Article 42 empowers the Security Council to “take such action by air, sea, or land forces as may be necessary to maintain or restore international peace and security.” The second and more frequently invoked exception is the reservation to states of a right of self-defense contained within Article 51. The second exception establishes that “[n]othing in the … Charter … impair[s] the inherent right of individual or collective self-defence if an armed attack occurs against a Member of the United Nations.” When the “Security Council has taken measures necessary to maintain international peace and security,” however, the right to self-defense terminates. Moreover, a procedural limitation applies: any measure undertaken in self-defense must “be immediately reported to the Security Council.”
The characterization of the Charter’s text as the starting point for analysis of the legality of the use of force is an apt one, in part because the relevant text is fairly brief and the rules it sets forth do not always provide clear answers when applied to particular facts. Throughout the Charter’s history, states that have used force have seized upon potential gaps in the Charter’s prohibition of the use of force in order to justify their actions. For example, states have urged that particular uses of force fell outside the parameters of the prohibition because the force was not used “in the international relations between States.” States have also argued that their particular use of force was not “against the territorial integrity or political independence” of another state, or was not “inconsistent with the Purposes of the United Nations.” States that have used force have sometimes claimed that these phrases give rise to implied exceptions to the prohibition of the use of force, and therefore provide legal justification for intervention in other states to achieve objectives such as the fulfillment of humanitarian need, the attainment of self-determination, the installation of democratic regimes, or the preservation of socialism.
Questions have also arisen regarding what actions constitute “force” for the purposes of Article 2(4). Some argue that that economic or physical actions not involving military action should be treated as involving the use of force, as should indirect support for military action by groups other than the armed forces of the supporting state. Moreover, the Security Council’s authorization of enforcement actions pursuant to Chapter VII is accomplished by passing resolutions that may not clearly define the exact scope of the authorization, providing states with opportunities to justify uses of force that may or may not be authorized depending on how one interprets the resolutions at issue. Similarly, it has also been claimed that Security Council resolutions which authorized force in the past have a continuous and cumulative effect, with the result that the use of force may be resumed at concerned states’ discretion without renewed Security Council authorization.
Notwithstanding the ambiguities and possible loopholes found in Article 2(4), its prohibition on the use of force is widely regarded as barring, as a general rule, the nonconsensual use of force by one state against another. The preemptive measures contemplated by the NSS are not consensual in nature. Moreover, the Bush administration has not sought to justify U.S. action under the NSS by reference to the questionable exceptions to Article 2(4) just described; rather, the administration argues that the unquestionably valid right of self-defense is expansive enough to permit states invoking it to employ preemptive force. Although disagreement exists as to whether non-military action can constitute force, it is widely held that non-consensual armed action by one state against the territory of another– such as that called for by the NSS– constitutes a use of force for purposes of the Charter’s rules. The point at which a state’s support for proxies fighting another state will be deemed a use of force by the supporting state is not entirely clear. There can be no doubt, however, that because the use of force called for by the U.S. strategy consists of action by the American military, the U.S. is responsible for using such force.
B. The Right of Self-Defense Under Article 51
As the preceding discussion suggests, when the U.S. undertakes military action under its NSS strategy, it is likely making use of force within the meaning of Article 2(4). In the future, no prior Security Council authorizations to use force are likely to be available to provide justification when the NSS strategy is applied. However, the argument will be advanced, as it has been in the case of the Iraq war, that the force used by the U.S. is a legitimate exercise of the right of self-defense recognized in Article 51. The text of Article 51, like that of other Charter provisions dealing with the use of force, has given rise to important issues of interpretation. The meaning of the term “armed attack” in Article 51, and the relationship of that term to the “use of force” prohibited by Article 2(4), have been called into question. In particular, controversy exists as to whether or not the two terms are synonymous. The International Court of Justice (ICJ) has suggested that the terms have different meanings because some state actions properly characterized as uses of force are not of sufficient “gravity” to qualify as armed attacks. Thus, according to the ICJ, not every “use of force” is sufficiently serious to be treated as an “armed attack” that would permit a forcible response to be justified as self-defense. With respect to the Iraq war, however, there has been no showing that Iraq used or threatened any force at all in advance of the 2002 U.S. invasion, much less engaged in an armed attack.
It may be argued that the September 11, 2001 attacks constituted an armed attack which ultimately gave the U.S. a right to exercise self-defense against Iraq. Prior to that date, it was not clear whether a terrorist act could be treated as an armed attack; however, the Security Council’s response implied that the September 11 attacks gave rise to an affirmative right to use force in self-defense. While the Security Council did not declare Afghanistan responsible for the attacks, the U.S. justified the war against Afghanistan as an act of self-defense. Most states were receptive to the notion that Afghanistan was sufficiently responsible for supporting the September 11th attackers to permit responsibility for the attacks to be imputed to Afghanistan. The attacks appeared to rise to the level of armed attacks within the meaning of Article 51, so U.S. claims of a right to use force in self-defense against Afghanistan were widely perceived as unobjectionable. While the U.S. made efforts to connect Iraq to the September 11 attacks, the U.S. government could not produce the evidence needed to impute responsibility for those acts to Iraq. Thus, the distinction between the use of force and more substantial armed attacks is inapposite, because Iraq did not use force at all, much less use it in a manner substantial enough to give rise to an armed attack.
Given that Iraq was apparently not responsible for any armed attacks against the U.S., one might think that the U.S. claim that its use of force against Iraq was undertaken in self-defense necessarily fails. The text of Article 51, after all, seems to indicate that the right of self-defense arises only when “an armed attack occurs.” However, in this instance, the text obscures an interpretive dispute that hinges directly on the U.S. claim. States and scholars disagree as to the time at which the right of self-defense becomes available for exercise. Determining the legitimacy of the U.S.’s claim of a right to use preemptive force to defend itself depends upon the extent to which self-defense may be exercised in advance of an armed attack.
C. Permissive and Restrictive Interpretations of the Right of Self-Defense
Article 51 must be interpreted in order to determine if the right of self-defense can permit the use of preemptive force. The Vienna Convention on the Law of Treaties, which the U.S. has signed but not ratified, provides a useful guide to the interpretation of the treaty provisions such as Article 51. The Vienna Convention requires that treaties “be interpreted in good faith in accordance with the ordinary meaning to be given to [their] terms … in their context and in the light of [their] object and purpose.” The relevant “context” includes, “in addition to the text, including its preamble and annexes[,]” any agreements or instruments agreed to or accepted by “all the parties in connexion [sic] with the conclusion of the treaty.” In addition to context, the primary guideposts to interpretation are: (1) “subsequent agreement[s]” regarding the treaty’s “interpretation” or “application;” (2) “subsequent practice in the application of the treaty which establishes the agreement of the parties regarding its interpretation;” and (3) “any relevant rules of international law.” The commentary indicates that some dispute exists as to the relative weight to be given to each of the guideposts to interpretation, but concludes that most jurists recognize “the primacy of the text as the basis for the interpretation.”
Scholars endeavoring to interpret Article 51 to determine when the right of self-defense arises have employed methods of interpretation similar to those endorsed by the Vienna Convention, but have drawn different conclusions. The more restrictive position holds that Article 51 forecloses a state’s ability to use force in self-defense before an armed attack occurs. The more permissive position holds that an armed attack need not occur before force may be used in self-defense. Instead, force may be used in anticipation of an attack, in a manner that has been dubbed “anticipatory self-defence.” Force used in an anticipatory act of self-defense must still meet the necessity and proportionality requirements of customary international law, and in the context of anticipatory self-defense, the former requirement demands a showing that the threat is imminent. The Bush administration asserted that this second, permissive viewpoint represents a centuries-old consensus as to the scope of the right. Furthermore, the administration regards the preemptive strategy enunciated in the NSS as an allowable application of the right of self-defense as understood pursuant to the permissive position.
1. Treaty Text
Analysis of Article 51 in light of the Vienna Convention sheds light on the disagreement between the restrictive and permissive schools of thought. The text of Article 51 uses the phrase “if an armed attack occurs” to describe the situation in which the Charter’s operation will not “impair” the right of self-defense. Those advocating the restrictive view may argue compellingly that this language indicates that an actual–rather than a potential–armed attack is needed to trigger the right to use force in self-defense. An armed attack is an event capable of being perceived and identified as such, and like any other event, “occurs” when it “come[s] into existence” or “happen [s],” and not before. Thus, an armed attack must actually happen before it can truly be said that “[n]othing in the … Charter” will “impair the inherent right of … self-defence.” Accordingly, only an armed attack that is happening or has happened can satisfy Article 51’s armed attack requirement.
Moreover, the structure of the language used in Article 51 bolsters the restrictive position. Article 51 impliedly recognizes a general rule of impairment of the right of self-defense when it states that nothing in the Charter will “impair” the right “if an armed attack occurs.” The Article would arguably be devoid of meaning if special provisions did not have to be made in order to preserve a right of self-defense. Thus, the Charter must generally operate to impair the right of self-defense; only in exceptional cases will it be deemed unimpaired. The occurrence of an armed attack gives rise to the only situation in which the Charter states that the right of self-defense is not impaired, however “inherent” it may be. A familiar canon of construction holds that the mention of one thing implies the exclusion of other things; as applied to Article 51, this suggests that explicit exceptions to the general rule of impairment of the right of self-defense should be deemed exclusive. Thus, the language should be read to mean that the Charter has rendered the right of self-defense unavailable prior to the occurrence of an armed attack.
On the other hand, advocates of a permissive interpretation contend that that the text of Article 51 does not purport to grant a right of self-defense to states. The language of Article 51 assumes that such a right already existed when the Charter was signed; indeed, the right is said to be an “inherent” aspect of the sovereignty of states. Article 51’s recognition that states’ self-defense rights are “inherent,” which is defined as “involved in [states’] constitution[s] or essential character” and “belonging by nature” to states, suggests that the argument for a restrictive approach should be turned on its head. The Charter’s drafters described the right of self-defense as “inherent” because they deemed it a fundamental attribute of state sovereignty. Therefore, Article 51 should not be read as restricting the right of self-defense, but as clarifying the right’s continued existence, because the drafters would have been explicit if they wished to place limitations on a right they considered so important.
Scholars adopting a permissive approach endeavor to interpret the phrase “an armed attack” in a manner consistent with their position. Some view the armed attack requirement as a reference not only to actual interstate violence, but also to actions taken in preparation for the attack. Thus, an armed attack may be said to have occurred when some state of preparation is reached, even before one state actually strikes another. The definition of “armed attack” must encompass more than the ultimate act of interstate violence; otherwise, the right would be deprived of the broad scope seemingly appropriate to a right inherent in statehood. An alternative explanation is that Article 51 does not require an armed attack, but merely states one instance in which the customary right of self-defense is preserved. This second understanding is difficult to square with the structure of Article 51. However, the recognized inherent and fundamental need for self-defense against imminent attack, in addition to states’ interpretation of the provision, may be viewed as justifications for the second reading of Article 51.
The Vienna Convention’s “General Rule of Interpretation” does not limit the scope of the “text” considered by an interpreter to the specific portion under scrutiny, but instead calls for examination of the whole “text, including its preamble and annexes,” to aid in interpretation. The restrictive school of thought points to numerous aspects of the Charter that suggest that the avenues left open for the legitimate use of force were meant to be narrow. According to the restrictive view, the scope of the right to self-defense recognized in Article 51 is diminished by the temporal limitation in the armed attack requirement and the Article’s command that the exercise of the right by the states is “immediately reported to the Security Council.” Moreover, the text explains that force used in self-defense must cease when the Council “take[s] measures necessary to maintain international peace and security.” If the “inherent” quality of the right of self-defense is as important to interpreting Article 51 as the permissive position posits — supposedly relaxing or eliminating the text’s requirement of an armed attack — it seems unusual that the right of self-defense would be made contingent upon notification, and subject to limitation by action of a body beyond the state’s control.
With respect to Article 51’s interaction with Article 2(4), those urging a restrictive interpretation of the temporal scope of the self-defense right can again invoke the canon that explicit exceptions are deemed exclusive. The general prohibition on the use of force yields only to a limited number of explicit exceptions, including the occurrence of an armed attack. If an armed attack “occurs” when interstate violence actually “happen[s]” and not before, the argument that another, different right of self-defense may be relied upon before interstate violence actually occurs runs afoul of sound textual interpretation.
The right of self-defense recognized in Article 51 is merely an exception to the general prohibition on the use of force contained in Article 2(4). This suggests that “[t]he use of force in self-defense is limited to situations where the state is truly required to defend itself from serious attack. In such situations, the state must carry the burden of presenting evidence to support its actions, normally before these irreversible and irreparable measures are taken.” Thus, the Charter affirmatively requires U.N. members to rely on alternatives to force to resolve their disputes, and suggests a number of peaceful means by which this obligation may be discharged. An annex to the Charter was included to give concrete form to one vehicle for peaceful dispute resolution, the International Court of Justice, the “function [of which] is to decide in accordance with international law such disputes as are submitted to it.” Also relevant in this context is the Charter’s preamble, which explains a central focus of the Charter: the common interest in keeping forcible exchanges between states to a minimum. A restrictive interpretation of Article 51 thus suggests that so long as an “armed attack” remains a possibility rather than an actuality, the Charter’s commands to settle disputes peacefully and to avoid the use of force preclude a state from resorting to arms.
Lastly, the Charter’s other significant exception to the prohibition of the use of force allows for its use pursuant to Security Council decisions. Article 1(1) expresses a preference for the use of “collective measures” as a means of “maintain[ing] international peace and security.” Article 24 reflects the same preference for collective over unilateral action in that it established that United Nations “[m]embers confer on the Security Council primary responsibility for the maintenance of international peace and security.” This preference is further reinforced by the textually subordinate status of self-defense under Article 51, as noted above, in relation to Security Council action. Arguably, these aspects of the text are incompatible with an interpretation of Article 51 that leaves a state free to use force against another state before the Security Council has an opportunity to consider the situation.
Those favoring a permissive approach to interpretation of the right of self-defense are likely to view the same portions of the text differently. First, with respect to the provisions for Security Council involvement found in Article 51, they regard the notification requirement as no more than an administrative or procedural matter, imposing no limits on the substantive scope of the right enjoyed by the state acting in self-defense. As such, the fact that a party exercising the right must give notice to the Security Council should not be read to suggest that the right ought to be narrowly construed. The Council’s apparent ability to terminate a properly-exercised right of self-defense is of little importance to the issue that divides the restrictive and permissive schools of thought — the time when the right may first be asserted. By the terms of the Charter, the right of self-defense can be limited only when the Council acts, and then only when it takes “measures necessary to maintain international peace and security.” Moreover, in practice, this limitation has little effectiveness, as states have exercised what amounts to “concurrent power” with the Council in continuing to defend themselves even after taking “measures,” and possess a wide degree of discretion in deciding when the Council’s action has superseded their right to use force in self-defense. For these reasons, the Council’s ability to terminate the exercise of a state’s right to self-defense should not significantly detract from Article 51’s recognition of self-defense as an inherent right worthy of the liberal interpretation given to it by the permissive school. Additionally, the preference for collective over individual action to “maintain international peace and security” should not be interpreted to imply that the collectivism favored by the Charter is at odds with a broad interpretation of the right of self-defense, because the self-defense right recognized in Article 51 is itself capable of being exercised in a “collective” fashion.
Permissive writers claim that the aforementioned canon of construction should not exclude anticipatory self-defense unless Article 51’s reference to an armed attack intends to foreclose anticipatory action. In their view, the canon is inapplicable because the language requiring an “armed attack” applies only when “the Security Council is acting” and takes “measures necessary to maintain international peace and security.” States’ obligation to settle disputes peacefully is inapposite, because it has already been breached by the time a state exercises its right of self-defense. If the right to use force in self-defense extends to situations preceding the armed attack, a permissive understanding of the right suggests that at some point in a potential attacker’s preparations, the potential defender’s obligation to settle disputes peacefully must yield by reason of the potential attacker’s decision to resort to force.
2. Context and Subsequent Agreements
The text of a treaty, although perceived by many scholars to be of primary importance to the task of interpretation, is not the only relevant consideration. According to the Vienna Convention, the treaty’s “context” should be considered, and certain other items should be “taken into account.” Aside from its own text, the Charter lacks a relevant “context” as defined by the Vienna Convention. The Vienna Convention also calls for consideration of “subsequent agreement[s] between the parties” pertaining to the treaty. Unlike agreements and instruments entered “in connexion [sic]with the conclusion of the treaty to be interpreted,” which form part of the treaty’s “context” and must receive the backing of all other parties to the treaty, subsequent agreements may be relevant to interpretation of a treaty even if they have not been assented to by “all” the parties to the treaty under scrutiny. Therefore, agreements postdating the Charter that are not backed by all parties to it may still be relevant to the Charter’s interpretation. As a result, a number of widely supported General Assembly resolutions may provide insight into Article 51’s interpretation. Although none of these resolutions go directly to the scope of permissible self-defense, what they have to say about the use of force as a general matter may still be relevant to interpreting the scope of permissible self-defense.
The 1965 U.N. Declaration on the Inadmissibility of Intervention in the Domestic Affairs of States and the Protection of Their Independence and Sovereignty (Declaration on the Inadmissibility of Intervention) rejects the use of force in international relations. It declares that “[n]o State may use or encourage the use of … measures to coerce another State … to secure from it advantages of any kind,” and further enjoins States from engaging in “activities directed towards the violent overthrow of the régime of another State.” This language may be understood to condemn acts of self-defense undertaken before, or too far in advance of, the occurrence of an armed attack. However, the Declaration on the Inadmissibility of Intervention goes on to disavow any intention that its language is meant to “affect in any manner the relevant provisions of the Charter.”
The 1970 U.N. Declaration on Principles of International Law Concerning Friendly Relations and Co-Operation Among States In Accordance with the Charter of the United Nations (Declaration on Friendly Relations) demands that states “refrain from the threat or use of force … as a means of solving international disputes,” and affirms that “[a] war of aggression constitutes a crime against the peace” which gives rise to criminal “responsibility.” The Declaration on Friendly Relations provides support for the notion that the right of self-defense is limited rather than open-ended by declaring that forcible “acts of reprisal” are impermissible. Additionally, the Declaration on Friendly Relations indicates that the obligation to settle disputes peacefully continues even after initial attempts to do so have failed. It is hard to reconcile a process in which all parties earnestly endeavor to peacefully settle a dispute with the use of force in advance — or too far in advance — of an armed attack by another disputant, the occurrence of which remains speculative before violence is brought to bear.
An overhasty act of self-defense might well run afoul of a third Assembly resolution, the Definition of Aggression, which states, “[t]he first use of armed force by a State in contravention of the Charter shall constitute prima facie evidence of an act of aggression.” The resolution’s definition of “[a]ggression” may be rebutted only pursuant to a Council determination. No justifications for aggression are identified. Like the Declaration on the Inadmissibility of Intervention, however, both the Declaration of Friendly Relations and the Definition of Aggression contain language indicating that they are not intended to change the Charter’s scope or meaning.
Those favoring a restrictive interpretation of the self-defense right can argue that the aforementioned resolutions are consistent with their view. Specifically, their view is a narrow reading of the right that limits its assertion to situations in which the Charter clearly recognizes its existence–namely, when one state is then undertaking or has already undertaken an “armed attack” against the defender. Adherents of the permissive position may counter that none of the resolutions cited explicitly purports to construe the right of self-defense. Furthermore, those favoring a permissive interpretation may argue that no inferences about the scope of the rules found in the Charter should be drawn. They can ground this argument on traditional reservations about the ability of General Assembly resolutions to produce binding legal effects, and the caveats expressly included in the resolutions that appear to foreclose treatment of the resolutions as authoritative interpretations of the Charter. Those preferring a restrictive approach could respond that the limiting clauses of the resolutions are better understood, not as impediments to the use of the resolutions in construing the Charter, but as indications that the resolutions are intended to restate the Charter’s meaning rather than to modify it. Moreover, resolutions such as the Declaration on Friendly Relations, which received overwhelming support in the General Assembly, may be relevant both to interpreting Article 51 and to independently determining the legality of the use of force by contributing to the development of customary international law.
3. State Practice
The Vienna Convention also calls for consideration of “any subsequent practice in the application of the treaty which establishes the agreement of the parties regarding its interpretation.” On several occasions since the founding of the United Nations, states have used force in a manner seemingly akin to anticipatory self-defense. The U.S. naval blockade of Cuba during the Cuban missile crisis has been cited as an example of a case in which preemptive or anticipatory self-defense was used with international approval. The blockade should properly be regarded as a use of force and was apparently undertaken to forestall the installation and possible future use of Soviet nuclear-armed missiles in Cuba. However, while commentators debated the applicability of the right of self-defense, the official U.S. position did not attempt to justify the blockade on that ground. Instead, it sought to justify it as a measure taken by a regional organization, the Organization of American States, pursuant to Article 52. The American unwillingness to rely on the right of self-defense, especially in view of the strong tendency of states to advance self-defense to legitimize their uses of force, suggests that the U.S. believed that self-defense would have been heavily contested as a justification. Even to the extent that self-defense was raised in discussions in the Council, it cannot be said that the Council recognized the Cuban blockade as legitimate, as opinion was sharply divided, and it is uncertain whether supporters of the U.S. position were motivated by Cold War alliance considerations or by the U.S.’s proffered justifications. Thus, the Cuban blockade provides no real support for a permissive interpretation of the Charter.
A second oft-cited example is the Israeli air-strike against Egypt at the start of the Six-Day War in 1967. Prior to the air strike, the objective of was to destroy Egypt’s Air Force. Egypt had taken a number of provocative actions in short succession that made the onset of hostilities appear imminent, including closing the Straits of Tiran to Israeli traffic. The Israeli strike preceded any interstate violence undertaken by Egypt. Israel sought to justify its action on grounds of self-defense, arguing that the actions taken by Egypt amounted to an armed attack. The Council again split along ideological lines, with the Western bloc supporting the Israeli action and the Soviet bloc opposed to it. However, unlike the example of the Cuban blockade, Israel justified its 1967 air raid on the basis of self-defense. However, Israel’s justification was one of ordinary rather than anticipatory self-defense: it argued that the closing of the Straits was not merely preparation for war, but was tantamount to an armed attack. In any event, the U.N. institutions did not explicitly accept or reject the Israeli claim.
A third historical example appears to have received more attention from commentators than the two previously discussed, probably because it presented facts most implicative of the debate over anticipatory self-defense. In 1981, Israel launched another air-raid, this time against an Iraqi nuclear facility under construction near Baghdad. While the facility was not yet operable, and speculative arguments could have been made as to its potential future use as a means of producing nuclear weapons, it was clear that at the time of the strike, the unfinished and inoperative reactor posed no threat to other states. Reliance on a right of preemptive self-defense was unavoidable: Israel aimed to justify its use of force by claiming that force was used to defend against a threat that would become imminent only in the future. The action was met with widespread disapproval, but for different reasons. Some states took a position consistent with the restrictive view and argued that no action of the kind taken could ever be consistent with the Charter. Specifically, Article 51 had limited the scope of the right, making it exercisable only “if an armed attack occurs.” Conversely, other states based their objection to the Israeli action on customary law grounds, arguing that the Israeli strike did not comport with the requirements of necessity and proportionality. The U.S. condemned Israel only for its failure to exhaust peaceful means of resolving the dispute. The U.N. political organs thus condemned the attack without reaching a consensus as to the scope of the right of self-defense.
The historical practice of U.N. members in dealing with claims of anticipatory self-defense suggests that states are not in agreement as to the propriety under Article 51 of actions taken in self-defense prior to the onset of an armed attack. While some states agree that the right of self-defense includes the right of states to deploy force before military force is deployed against them, this is not the majority viewpoint. States rely infrequently on anticipatory self-defense to justify the use of force; instead, they attempt when possible to characterize their actions as supported by the less controversial right of states to use force defensively in response to an armed attack that has occurred or is occurring.
4. Relevant Rules of International Law: Customary Principles of Necessity and Proportionality
Customary international law has an important role to play in an analysis of the legality of the use of force in self-defense. First, according to the interpretive framework set forth in the Vienna Convention, “relevant rules of international law applicable in the relations between the parties” to the Charter must be considered as an aspect of the Charter’s “context” to arrive at a proper understanding of the scope of the right of self-defense recognized therein. General rules of customary international law are as valid and applicable in relations between the parties to the Charter as are the provisions of the Charter itself. The obligations imposed by customary rules exist independently of any treaty provision.
The customary rules factor significantly in the analysis of the availability, timing, and permissible scope of the use of force in self-defense. Many writers cite the early formulation of these rules found in a diplomatic note from U.S. Secretary of State Daniel Webster to the British ambassador in 1842. The note concerned the sinking of an American ship, the Caroline, by the British, which was done in order to prevent the ship from being used to support rebels opposed to the British colonial government. The U.S. demanded that the British justify their action by showing that the “‘necessity of that self-defense [wa]s instant and overwhelming, and leaving no choice of means, and no moment for deliberation.”’ The American note also demanded a showing that the force used involved nothing “‘unreasonable or excessive.”’ Although this note long predated the Charter, and was thus the product of a time when resorting to force was not unlawful in the sense of the prohibition in Article 2(4), it contains a useful elaboration of the customary constraints on self-defense. As the letter indicates, use of force in lawful self-defense must always be both necessary and proportionate.
The concept of necessity limits the use of force to situations in which forcible “measures … are … necessary to respond to” the armed attack or to otherwise repel aggression. Peaceful alternatives must not have been readily available to the state invoking self-defense. The law recognizes that the use of force in self-defense may be necessary for a variety of reasons, such as “to hold the aggressor in check and prevent him from continuing the aggression,” “to recover territory” recently lost to the attacker, or to “repel an attack.” The aim of force used in self-defense cannot be “retaliatory or punitive.” When states rely upon anticipatory self-defense, scholars who accept that doctrine as a valid justification for the use of force generally accept that the requirement of necessity demands a showing that the threat was imminent.
The proportionality requirement, on the other hand, limits the amount and scope of the force that may be used to fulfill the need to defend. The concept that forcible “measures … [must be] proportional to the armed attack” does not demand that the defending state limit the size or sophistication of its forces; nor does it categorically compel a state fighting off an invasion to stop its use of force as soon as the invaders are pushed back beyond the border. However, the force used in self-defense must be “proportional to the offense in its extent, manner, and goal.” The proportionality requirement’s prohibition of “unreasonable or excessive” conduct is, in a sense, similar to the necessity requirement’s prohibition of “forbidden” conduct because in both instances, the conduct that is not permitted will vary with the circumstances. Sufficient force may be used to respond to armed attacks to ensure decisive results, even if to achieve them would require modest increases in the force used or spatial expansions of the conflict. In general, to be proportional to the attack, “the act, justified by the necessity of self-defense, must be limited by that necessity, and kept clearly within it.”
Those favoring a restrictive approach to anticipatory self-defense may argue that these customary doctrines, although understood by some writers to permit anticipatory self-defense, cannot properly be applied when anticipatory force is used. Determinations of necessity and proportionality, after all, “are dependent on the facts of the particular case.” The facts of a case cannot be meaningfully analyzed before they actually exist; analysis before that time is merely speculation. Thus, one who engages in anticipatory self-defense may thwart the application of these rules because the need for force and the necessary amount cannot be ascertained. Before the commencement of an armed attack, the use of force in self-defense can only be based on a guess as to the hostile intentions of the potential attacker. If the guess is incorrect and the attack never would have occurred or could have been peacefully prevented, then there was no need for the use of defensive force in any proportion. Indeed, according to some supporters of the restrictive position, force used in anticipatory self-defense cannot comply with the rules imposed by customary doctrines.
Supporters of the permissive view contend that many writers consider the customary rules to permit the use of anticipatory self-defense. They seek to dispel fears about the potential for anticipatory self-defense to encourage reckless uses of force by reading the requirement of necessity as incorporating a requirement of imminence. The requirement of imminence is theoretically measurable even in the absence of an armed attack; from an objective standpoint, the legality of the anticipatory action can be assessed by considering the relevant facts at the time defensive force was used. The burden of being subject to the use of force initially falls upon the potential attacker, but in light of the rule permitting those subject to attack to take defensive measures, this allocation of risk might be entirely appropriate.
5. Object and Purpose of the Treaty
Under the interpretive approach of the Vienna Convention, a treaty should be considered “in the light of its object of purpose.” The purpose to be examined is not merely the purpose of the individual rule under consideration, but that of the treaty as a whole. Here, both sides of the debate over the permissibility of anticipatory self-defense can marshal compelling arguments. Those favoring a restrictive approach are aided by the Charter’s manifest focus on keeping interstate violence to a minimum. An interpretation of the right of self-defense that unnecessarily increases the likelihood that force will be used, such as the interpretation articulated by the permissive school, is inconsistent with this overriding goal. Their interpretation allows the self-defense exception to override the prohibition on using force. The permissive interpretation gives rise to circumstances in which creative argument might justify any use of force by connecting its use to speculation about some future threat that might reasonably–whether rightly or wrongly–be perceived as imminent. In an era where deadly weapons are capable of widespread and serious harm, the dangers of resorting to force are compounded.
Similarly, the Charter is designed to ensure the operation of mechanisms guided toward the peaceful settlement of international disputes. Achievement of this purpose may be thwarted if the right of self-defense is construed too broadly. If states can anticipate potential attacks and respond with force before they occur, the underlying disputes that engendered hostility between the states may give rise to interstate violence before mechanisms that promote peaceful settlements of international disputes have a chance to work. The time available for peaceful dispute resolution to function is reduced, possibly by a significant amount, depending upon the accuracy of the estimation as to the imminence of the attack by the state exercising its right of self-defense.
Additionally, the Charter’s language was meant to be functional in nature and capable of practical application. The restrictionist school of thought believes that Article 51 imposes such a rule in the context of self-defense: the use of force is justified if an armed attack occurs, and is not if an armed attack has yet to occur. Those advancing a permissive interpretation of the right of self-defense favor a considerably less certain analysis based only upon the customary standard rather than the objectively determinable criteria of the occurrence of an armed attack. The rule as permissively interpreted is more difficult to apply, and the risk of error is increased. The benefits of a clear rule are lost. Requiring an armed attack before using force in self-defense clearly informs states of the activity they must avoid in order to prevent force being used against them. The permissive approach also leaves states less certain as to what preventive measures they can take to avoid the permissible use of force against them.
Those favoring the permissive approach can agree that the Charter was meant to produce a soundly-functioning international system, but need not accept the restrictive approach’s viewpoint of what such a system entails. Under this view, the Charter established a framework within which international peace and security can be maintained with the states themselves as primary beneficiaries. States, and their sovereignty and integrity, were to be preserved by the U.N. rather than subsumed into a world government. Diminution of the scope of the “inherent” right of self-defense would amount to an unwarranted intrusion into a fundamental aspect of sovereignty. Self-defense is too fundamental to this sovereignty to have artificial time limits placed upon it by the Charter or customary law.
In addition, the Council has been only modestly successful in performing its primary function of achieving sustained peace. The Charter was intended to be a holistic scheme for ensuring international peace, safeguarding states from aggression and preserving the right of self-defense as a supplement to the collective mechanism embodied in the Council. The Council’s limited success thus far suggests that the purpose of maintaining international peace and security may not be fulfilled unless force in self-defense may be exercised in more than the narrowest of circumstances.
Furthermore, as a practical matter, the permissive school can argue that the Charter should not be interpreted as a suicide pact that illegalizes defensive actions necessary for a state to preserve itself. Since the Charter was signed in June 1945, the weapons available to states have become even more destructive. Such weapons may enable an attacker to strike a decisive blow against its victim in one stroke. By the time an attack is underway, it may simply be too late to take measures in self-defense. If the law guarantees that the attacker may strike first, it has effectively rewarded that state’s violence with military advantage. The purpose of the Charter was certainly not to encourage violence or to deprive states of the ability to take reasonable steps to defend themselves. The restrictive view impedes resort to the right of self-defense, potentially making resort to violence acceptable and advantageous to states in a position to take advantage of others that scrupulously follow the law.
III. EVALUATION OF THE ASSERTED RIGHT TO USE PREEMPTIVE FORCE IN SELF-DEFENSEEE
The divide between the interpretive approaches to the right of self-defense runs deep, and the state of the law with respect to the availability of anticipatory self-defense is likely to remain unclear. However, fairly widespread agreement exists as to the substance of the customary rules limiting the exercise of self-defense. Even those who assert the validity of anticipatory self-defense generally agree that customary rules limit its exercise. Regardless of one’s position on the interpretation of Article 51, using force in self-defense in contravention of this minimal customary standard will assuredly violate international law.
The preemptive use of force in self-defense, which was enunciated as the U.S. strategy in the NSS and was suggested as a possible justification for the Iraq war, must be evaluated in light of these standards. Initially, the NSS’s suggestion that agreement on the permissibility of anticipatory self-defense has existed “[f]or centuries” must be a reference to the period before recourse to force was prohibited as a general matter by the Charter, subject to a carefully-worded recognition of the right to self-defense. Since the Charter’s adoption, two schools of thought on anticipatory self-defense have existed. The restrictive position, far from viewing such action as capable of being taken “lawfully” or with “legitimacy,” categorically bars using force in self-defense before an armed attack occurs. The NSS’s preemptive strategy does not comport with this position because the U.S. plans to use defensive force long before potential attackers are prepared to act.
Moreover, even if the NSS is considered in light of the permissive approach to the right of self-defense, the preemptive strategy does not pass muster. First, even a permissive interpretation of Article 51 cannot completely do away with the requirement of an armed attack. To the extent that preemptive self-defense is asserted to justify the use of force against speculative threats that bear no relation to any armed attack, the Charter will bar the claim. Second, customary rules are likely to foreclose the permissibility of preemptive self-defense if circumstances arise in which Article 51 does not. Unjustifiable preemptive self-defense differs from potentially permissible anticipatory self-defense in that it allows states to unleash force against one another with much greater ease. Anticipatory self-defense, as advocated by most adherents to the permissive interpretation of self-defense, is impermissible if a state cannot demonstrate that a threat actually exists and is imminent. In contrast, preemptive self-defense, such as that called for by the NSS, permits the use of force by states that merely feel threatened, without requiring a showing of objectively verifiable indications that the asserted threat warrants the use of force in reply. Proportionality becomes virtually impossible to measure as to preemptive force, because no attack has occurred, is occurring, or is planned as to which the use of force must be proportionate. A legal regime that would permit the preemptive use of force in self-defense would be far too careless in its handling of the use of interstate force, a fearsome villain who the parties to the Charter knew from personal experience had to be kept in shackles of law.
Second, those advocating a permissive approach agree with their restrictive-minded counterparts that the use of force in self-defense must be restrained by the customary law concepts of necessity and proportionality. However, one cannot assess the facts and circumstances that give rise to the necessity of the use of force, and confine its proportions, before the attack actually occurs. For adherents of the restrictive view, this is one reason why an armed attack must actually have occurred before force is used in self-defense. For adherents of the permissive view, the requirement that the armed attack be imminent assuages concerns about the availability of facts sufficient to draw conclusions about whether a given use of force is justified.
The NSS, however, does not require a showing of an imminent threat before authorizing preemptive uses of force. The administration explains that, “as a matter of common sense and self-defense, America will act against such emerging threats before they are fully formed.” Indeed, the U.S. strategy expressly seeks to “adapt the concept of imminent threat,” apparently allowing for force to be used “preemptively,” “even if uncertainty remains as to the time and place of the enemy’s attack,” and even before “dangers gather.” U.S. action against Iraq demonstrates that the NSS’s notion of imminence does not require an immediate or tremendous threat by a potential attacker. Indeed, if the Iraq situation is any guide, a threat may be sufficient to trigger NSS authorization to use force in purported self-defense if it is little more than plausible or even a hypothetical threat.
If the rules regulating the use of force are relaxed to reflect U.S. policy, conflicts may arise that could be prevented if the rules governing self-defense continue to require a truly imminent threat. The U.S. position may also produce interstate violence by creating an increased incidence of improper uses of force by states that feel less reluctant to use force purportedly in self-defense than in the past. Moreover, a standard that relies on a considerably less imminent threat is more likely to give rise to mistakes by the defenders as to whether they faced a threat at all. As the evidentiary problems with the case for self-defense in Iraq demonstrate, the U.S. government will experience mistakes with its current understanding of the self-defense rule. The abandonment of the requirement of an imminent threat leaves no principled means by which the law can distinguish force used in self-defense from aggression. The Bush administration’s understanding of the constraints of customary law on self-defense is at variance with that law in its present form.
Thus, the U.S. position in the NSS may be best described as an argument for a change in the law. Some writers have argued, on the assumption that the use of force in self-defense can be permitted before an armed attack occurs, that the war in Iraq is state practice that provides evidence of a change in the customary law loosening the constraints of necessity and proportionality. Some scholars have gone further, arguing that the prohibition on the use of force has come to lack vitality, and that the use of force for purposes such as preemptive self-defense is no longer forbidden. These arguments are misplaced. The creation of a new customary rule traditionally requires “an established, widespread, and consistent practice” combined with “evidence of a belief that this practice is rendered obligatory by the existence of a rule of law requiring it.” The practice of justifying the use of force on grounds of preemptive or anticipatory self-defense is not consistent or widespread. Furthermore, the psychological element of opinio juris is not present: the U.S. has conceded that “[it] must adapt the concept of imminent threat” as currently understood, and many other states apparently view the U.S. justification of self-defense for the preemptive use of force in Iraq as inappropriate. This refusal to treat divergent state practice as an emerging customary rule suggests the continued vitality of existing rules. Finally, from a policy standpoint, it cannot be denied that the elimination or substantial weakening of the rules restricting the use of force would have a deleterious effect on international order, potentially unleashing anarchy as states other than the U.S. come to view the use of force as a permissible means of resolving their disputes.
Perhaps to avoid the negative impact of other states’ assertions of a right to use preemptive force, the NSS appears to have taken an extreme approach. The NSS can be read to argue that different rules should apply to constrain the use of force by the U.S., and that these rules should be relatively permissive. This approach, however, is difficult to square with the U.N.’s fundamental “principle of the sovereign equality of all its Members.” While states are not equal in other respects, each should be treated as a formal equal of the others for purposes of applying the rules of international law. Exceptional approaches subvert this important principle. They place international law in the unsatisfying position of serving as an ongoing and evolving apologia for the actions of powerful states. They are also inconsistent with basic notions of fairness in an international system that has long been governed by rules that are the product of states’ mutual consent.
The strategy of preemptive self-defense articulated by the U.S. in the NSS, and apparently applied against Iraq, is at odds with current norms governing the use of force in self-defense. In order for the preemptive use of force to be lawful notwithstanding the Charter’s prohibition of the use of force, U.S. strategy would have to comport with the Charter and customary law relevant to the right of self-defense. Application of either of the prevailing interpretations of the right of self-defense leads to the conclusion that the U.S. strategy does not comply with international law. The U.S. strategy calls for resorting to force at an earlier stage than the Charter and customary law permit. Consequently, unless and until changes to the law are made that render the U.S. position regarding preemptive force compatible with the law governing self-defense, the U.S.’s “threat or use of force” in the circumstances identified in its strategy will remain inconsistent with international law.