Jump to Content Jump to Main Navigation


Forgotten your password?

Problems and Process - International Law and How We Use It by Higgins, Rosalyn (1995-08-24)

14 The Individual Use of Force in International Law

From: Problems and Process: International Law and How We Use It

Rosalyn Higgins

Collective security — Self-defence — Reprisals — Armed forces — Use of force, prohibition — UN Security Council — Self-determination — Humanitarian intervention — Armed attack — Use of force, threat — National liberation movements

(p. 238) 14  The Individual Use of Force in International Law

Since earliest times states have employed military force to pursue their political and economic objectives. As early as the fifteenth century, Grotius insisted that the law of nations limited the use of force to three justifiable causes: ‘defence, recovery of property, and punishment.’1 In that particular sense, it was necessary to restrict the use of force to a just war.2 The Covenant of the League of Nations sought further to control and contain the use of force, without prohibiting it.3 The Kellogg—Briand Pact of 1928 outlawed war as an instrument of national policy. This instrument,4 to which some sixty-three states became party, ‘decoupled’ the justness of the cause from the entitlement to use force. After the cataclysmic events of the Second World War, it was thought necessary to make it even more specific in the UN Charter that force could be used only in self-defence and not to pursue legal rights or genuinely held notions of justice.

The UN Charter thus limited permitted uses of force to self-defence or to collective enforcement action. But it also envisaged that the United Nations would itself provide the mechanisms for asserting legal rights and pursuing political and social justice. There would thus be no need for the individual resort to force. The United Nations was given powers which were intended to allow states to avoid unilateral reliance on the military instrument to guarantee their own security. One cannot understand the post-war debates about the legal limits to the use of force without appreciating that the contemporary norms were predicated upon a Charter system that until now has been impossible to operate. The assertion of legal rights and the pursuit of economic and social justice has been extraordinarily hard to achieve. Indeed, in a decentralized legal order, the notion of justice has often been in the eye of the beholder. And the immediate replacement of wartime co-operation with the Cold War made the collective security system envisaged by the Charter impossible to achieve. It was not possible to set up standing UN forces as envisaged (p. 239) under Article 43 of the Charter; and there was a total absence of political consensus in the Security Council.

Against these realities, states have continued to feel the need to resort to force. There has been a further underlying problem. Not only is the wording of the relevant UN Charter articles full of ambiguity, but the articles were formulated to address the problem of military hostilities between states. In the event, much of post Second World War military history has been about different uses of force—the employment or encouragement of irregulars by one state against another, guerrilla movements, national liberation movements, terrorism. The Charter was also formulated before the development of the atomic bomb. Its provisions were not only predicated upon a collective security system that was never a reality, they did not envisage the new types of violence, and the social conditions that were their origin and their consequence.

Since the collapse of communism in Eastern Europe there have begun to emerge other new realities. The total failure of the collective security system (rooted in confrontation between the West and the Soviet Union) has begun to be replaced by the possibility of co-operation within the UN system. But many political difficulties lie ahead in making that co-operation effective. And the ambiguities of the language of the UN Charter, set against these past and present political difficulties, have generated a variety of legal problems that have had great practical importance. In this chapter we will discuss the question of anticipatory self-defence; what constitutes ‘the state’ for purposes of self-defence; the question of humanitarian intervention; and what constitutes an armed attack. Each problem is set within the Charter relationship established between Article 2 (4) and Article 51.

The Relationship between Article 2 (4) and Article 51

Article 2 (4) provides that ‘All members shall 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.’ At the same time, Article 51 indicates that there are certain uses of force that will not contravene the prohibitions in Article 2 (4). It provides: ‘Nothing in the present Charter shall impair the inherent right of individual or collective self defence if an armed attack occurs against a member of the United Nations, until the Security Council has taken measures necessary to maintain international peace and security.’ It is then provided that these measures be reported to the Security Council. The text makes it clear that, on the one hand, a state may act in self-defence without first securing the permission of the Security (p. 240) Council, whilst, on the other hand, the Security Council retains its responsibility to take such action as it deems fit.

The UN Charter is intended to provide for a watertight scheme for the contemporary reality on the use of force. Article 2 (4) explains what is prohibited, Article 51 what is permitted. But almost every phrase in Article 2 (4) and Article 51 is open to more than one interpretation. Further, what happens if Articles 2 (4) and 51 are not in fact a watertight system, are not entirely opposite sides of the same coin? Can there be, for example, a use of force that is not against the territorial integrity or political independence of a state (and thus not, on the face of it, violative of Article 2 (4))—but is also not individual or collective self-defence (and thus manifestly permitted under Article 51)? It is unlikely—most uses of force, no matter how brief, limited, or transitory, do violate a state’s territorial integrity. A simple aerial military incursion will do so. So, too, will an attempt to exercise self-help even if in international straits. Self-help is the use of force to obtain legal rights improperly denied. In the Corfu Channel Case5 the United Kingdom engaged in minesweeping in the Corfu Straits (an international strait but also Albanian territorial waters) in order to make effective its legal right to free passage. The Court found such action unlawful—the action violated Albanian territorial sovereignty and legal rights were not to be vindicated through the manifestation of a policy of force.

It has been generally accepted, ever since this clear finding by the Court on the question, that self-help is unlawful under the Charter, notwithstanding the failure of the UN system to ensure that states do get the legal rights to which they are entitled. But, where the physical security of states is concerned, the matter has been more contested. So the inability of the United Nations to provide for the collective security of states has led to a rather more prolonged debate on the legal status of reprisals under the Charter. Reprisals consist of action in response to a prior unlawful military attack, aimed not at defending oneself against an attack as it happens, but rather at delivering a message of deterrence against the initial attack being repeated. Under customary international law, reprisals were lawful if certain criteria were met. These criteria, traditionally attributed to the Naulilaa Arbitration,6 were that there must have been a prior deliberate violation of international law; that an unsuccessful attempt must have been made at redress; and that the action taken in reprisal be proportionate to the injury suffered. Reprisals would necessarily involve a violation of Article 2 (4), however, and, not being self-defence, are not brought within the permissive use of force in Article 51. It is undeniable that post-war (p. 241) state practice has seen a substantial amount of military activity that has been frankly characterized as reprisals—that has been particularly true of the Arab—Israeli conflict in the Middle East. At times it has seemed as if there is an expectation of reprisals in the face of attack, and that the major concern has been as to the proportionality of the reprisals. But the Security Council has repeatedly condemned reprisals (albeit while often failing to condemn equally the prior illegal acts that led to them); and they are condemned in terms in the General Assembly’s Declaration of Principles of International Law Concerning Friendly Relations7 (which does clearly also condemn the organization or encouragement by one state of irregular forces for hostile activity in another). The texts of Articles 2 (4) and 51 clearly do not allow reprisals; and the study of other instruments and practices and judicial decisions does not allow one to conclude that there has been any de facto amendment of the Charter on this point—notwithstanding the fact that, in the absence of effective means of selfprotection, reprisals may be expected to continue.

When a state is not able to engage immediately in action to defend itself, subsequent action can (wrongly) take on the appearance of reprisals, though it is still action in self-defence. Let us imagine that a state is not in a position immediately to resist an invasion; provided that it does repel the invasion as soon as it is able, or as soon as all attempts to secure a peaceful withdrawal have failed, the action will still be self-defence. The position of the United Kingdom in respect of the Falklands/Malvinas, and of Kuwait in respect of Iraq, illustrates the point. In the former, a period of several weeks elapsed between the Argentinian invasion and the arrival in the area of the UK task force. In the latter, nearly five months elapsed between the invasion of Kuwait by Iraq, and the military response by a coalition of UN members.

A particularly acute variation of this problem occurs when the United Nations secures a cease-fire, perhaps with a UN force to oversee the cease-fire—but does not succeed in obtaining the withdrawal of the invading forces. Does the UN cease-fire, and the passage of time (during which the position of the intervening forces becomes entrenched) really preclude the invaded country from liberating its territory? The decision of the Croatian troops on 22 January 1993 to march across UN lines into Serb-held territory within Croatia graphically illustrates the dilemma. It is hard to see that the United Nations’ inability to secure the objectives of its agreed plan8 after a year should extinguish a suspended right of self-defence.

(p. 242) With these preliminary comments behind us by way of background, let us now turn to address each of the problems indicated.

Anticipatory Self-Defence

Article 51 allows self-defence only when an armed attack has occurred. Does this mean that a state has, quite literally, even when it sees that it is about to be attacked, to wait until the blow has been struck before defending itself? In this context, what is meant by the reference in Article 51 to an ‘inherent’ right of self-defence? Is it simply embellishment? Or does it rather serve to carry forward into Article 51 the customary international law on the matter? Under customary international law, self-defence fell to be tested against the criteria enunciated by US Secretary of State Webster in his diplomatic note to the British in the context of the Caroline Case of 1842.9 He stated, in a widely accepted dictum, that anticipatory self-defence must be restricted to those cases where the necessity ‘is instant, overwhelming, and leaving no choice of means, and no moment for deliberation’. That formulation has significance in several ways for the UN Charter system. It must be remembered that the Charter does indeed have its own procedures for dealing with international threats to peace. If the threat is one that could reasonably be contained or turned aside through calling an emergency meeting of the Security Council, the criteria of the Caroline probably will not be met. At the same time, in a nuclear age, common sense cannot require one to interpret an ambiguous provision in a text in a way that requires a state passively to accept its fate before it can defend itself. And, even in the face of conventional warfare, this would also seem the only realistic interpretation of the contemporary right of self-defence.10 It is the potentially devastating consequences of prohibiting self-defence unless an armed attack has already occurred that leads one to prefer this interpretation—though it has to be said that, as a matter of simple construction of the words alone, another conclusion might be reached.

Of course, abusive claims may always be made by states claiming to act in anticipatory self-defence. But in a decentralized legal order that is always possible; there is no avoiding the judgment that third parties will have to make on the claims in the light of all the available facts. But the (p. 243) Webster formula, although suggested so long ago, seems still very useful in providing the required balance between allowing a state to be obliterated and encouraging abusive claims of self-defence. It still has great operational relevance and is an appropriate guide to conduct.

What Constitutes ‘the State’ for Purposes of Self-Determination

Article 2 (4) prohibits the threat or use of force against ‘any state’—a precondition to the entitlement of any right of self-defence. What does this term ‘any state’ mean? Certainly the use of force is prohibited against a state’s dependent territories overseas, as much as against the metropolitan state itself. The military action in the Falklands was a use of force against a state—albeit actually against a dependency of the United Kingdom—and in turn allowed a right of self-defence to the United Kingdom, even though the ‘self’ was some eight thousand miles away. Again, as a matter of policy, that must be the right interpretation, otherwise all territories other than independent metropolitan territories could be exposed to hostile military action with impunity.

But there is a rather more difficult issue: can a use of force against one’s nationals abroad be termed an attack against a state? The importance of the question is apparent. The only use of force that is lawful is self-defence. States think it very important to be able to protect their nationals abroad—but they can do so only if the consequential use of force is justified as self-defence. And it can be justified as self-defence only if there has been an attack upon ‘the state’.

When we spoke about jurisdictional matters, we discussed the taking of jurisdiction over persons on the basis of passive personality—that they had harmed one‘s nationals abroad. We noted it was a controversial basis of jurisdiction. Here the focus is not on the taking of jurisdiction (or bringing someone to trial) but of the actual use of force, this time not necessarily against the individual wrongdoer, but against the state that is believed to have been responsible for the harmful acts. In an interesting interrelationship, questions of jurisdiction, use offeree, and state responsibility all bear on the same facts.

We have over the last few years seen a tendency to claim to be able to act in self-defence in response to attacks upon one’s nationals. The existence of international terrorism has undoubtedly been an impetus to this claim. States have seen indiscriminate bombings and killings directed at their diplomats overseas, at their military personnel, and, quite frequently, at civilians too. It is the hallmark of terrorism, that violence is not conducted in the state-to-state encounters of armies and airforces envisaged by the Charter, but by these covert, violent, and frequently indiscriminate means. (p. 244) A claim to be acting in self-defence has been made by the United States in the 1986 bombing raid on Libya, which it believed to have been responsible for a series of attacks on US nationals in Rome, Vienna, and Berlin.11 Some of these attacks were on diplomats; some on servicemen at a night club; and others upon civilians at an airport.

Even if the occasion for invoking self-defence in these circumstances has in this era of terrorism become more frequent, the idea of treating an attack on one’s nationals as an attack on the state is not entirely new. Professor Bowett has suggested that ‘it is perfectly possible to treat an attack on a state’s nationals as an attack on the state, since population is an essential ingredient of the state’.12 Professor Bowett also finds it significant that the General Assembly’s 1970 Declaration of Principle on Friendly Relations contains no prohibition on the protection of nationals, though reprisals are expressly denounced. But the Declaration is dealing with what constitutes an unlawful action under Article 2 (4), not what constitutes a lawful defence under Article 51. It is not surprising that reprisals are denounced as an unlawful use of force. The question of whether one’s nationals may be defended by reliance on Article 51 did not arise in that context.

What there is some evidence of in customary international law is the right to humanitarian intervention on behalf of threatened citizens abroad. Sometimes that was spoken of as a free-standing right, but sometimes it was expressed in terms of an exercise in self-defence. In any event, it is now clear that, particularly in the face of terrorism, the claim to self-defence goes beyond a claim to rescue nationals, but in fact covers acts that really have a different character. The claim was explained by former US Ambassador to the United Nations, Jean Kirkpatrick, in this way: ‘The prohibitions against the use of force in the Charter are contextual, not absolute … The Charter does not require that people submit supinely to terror, nor that their neighbours be indifferent to their terrorisation.’13 One can be sympathetic with the sentiment but note also that the language of self-defence is being invoked to cover military responses that really bear the characteristics of reprisals or retaliation. This reality is again emphasized by the example of the US military air-strike against Libya in 1986 in response to perceived terrorism against nationals. Having characterized it as self-defence, the President then spoke of the need to (p. 245) respond to abuse of one’s nationals; and the US Permanent Representative spoke of the action being designed to ‘disrupt Libya’s ability to carry out terrorist acts and to deter future terrorist acts by Libya’. The former is the language of retaliation, the latter of reprisals. Neither is really the language of self-defence.14

Humanitarian Intervention

Under contemporary international law, may a state militarily intervene in another territory to rescue citizens under threat? Under customary international law, such activity was widely tolerated. But is it still allowed under the Charter? Let us examine the legal and policy issues.

Even minor military incursions are unlawful uses of force. It is quite clear, from the practice under the Charter and otherwise, that the Charter law does not simply prohibit major clashes between entire armies, while allowing smaller scale military interventions. Attacks by single planes, for example, are as much a violation of Article 2 (4) as would be an attack by a squadron. And it is not really feasible to engage in a rescue operation of threatened nationals without engaging in some use of force, which is prohibited by the terms of Article 2 (4).

But does that dispose of the matter? There are several reasons for thinking that it does not. First, what Article 2 (4) prohibits is the use of force against the territorial integrity or political independence of a state, or in any other manner inconsistent with the purposes of the United Nations. It can easily be seen that even a single plane attacking a country is a use of force against its territorial integrity. But is the answer so clear when the military intervention is not an attack on the state as such, but an operation simply designed to be able to rescue and remove one‘s threatened citizens? Is that really a use of force against the territorial integrity of a state, or is it not rather a violation of sovereignty—in the same way as a civilian aircraft which enters airspace without permission will surely be violating sovereignty—but still not attacking the state or its territorial integrity? It would seem that hostile intent, coupled with military activity against the state (and beyond the minimum needed for the rescue), is what would distinguish a violation of sovereignty from an attack upon a state‘s territorial integrity.

If we can satisfy ourselves that humanitarian intervention does not violate the prohibition against the use of force against a state’s territorial integrity, then we can feel fairly confident that no other prohibition in (p. 246) Article 2 (4) is being violated. A military action to end a hijacking, for example, would not be force against a state‘s political independence (unless it was intended to overthrow the government), and nor would it seem to be contrary to the purposes of the Charter, being directed towards the preservation of human life.

There is a different way of looking at the whole question—instead of looking to see whether a humanitarian intervention violates Article 2 (4), looking instead at the permitted use of force under Article 51. That approach focuses rather on self-defence, and brings us back to the question of harm to one’s nationals and self-defence. It is very similar to, but not quite the same as, the question we asked ourselves before. Instead of saying ‘Is an attack on a foreign citizen an attack on the state, which therefore entitles self-defence?’ the question is the simpler one of whether a state can claim that military action to rescue one’s citizens is an exercise of self-defence. Again, cautious support has sometimes been offered for this view. Professor Sir Humphrey Waldock (later Judge Waldock), giving this General Course in 195215 said that a state could use force to rescue nationals ‘as an aspect of self defence’, if the threat of injury was imminent, if there was a failure or inability on the part of the territorial sovereign to protect them, and if the measures of protection were strictly confined to the object of protecting them against injury.’ These criteria would all seem to have been met in the Entebbe situation. There an Israeli civilian airliner was hijacked to Entebbe; the then President, Idi Amin, far from endeavouring to negotiate the safe release of the passengers, provided further arms for the hijackers and ominously separated the Jewish from the non-Jewish passengers. The dangers seemed extremely imminent and the rescue operation was directed only at procuring the safety of the passengers.

The following may be noted: a claim of humanitarian intervention based on self-defence could only be advanced in respect of nationals, because it is predicated on the argument that the state is being harmed through injury to its nationals, and can therefore respond in self-defence. But a claim of humanitarian intervention based on the argument that no violation of Article 2 (4) is entailed, would not logically be limited to the protection of one’s own nationals. Either Article 2 (4) is or is not violated by such activity—but nothing turns upon whether those being rescued are nationals or not.

The general question has yet to be judicially determined, though it did arise in an incidental way in the Tehran Hostages Case before the International Court of Justice in 1980.16 The Court was seized of an application by the United States to deal with the merits of that issue—(p. 247) namely, whether the State of Iran was in violation of the Vienna Convention on Diplomatic Relations 1961 or international law more generally, by any attributability to it of the acts of those who had taken US diplomats in Tehran hostage. The matter had already been, for several months, the subject of attempts at resolution elsewhere—there had been Security Council resolutions, a UN fact-finding commission, and an Order of the Court calling for the release of the hostages. No progress had been made. While the merits of the case were before the Court, the United States engaged upon an ill-fated military attempt at rescuing the hostages. If one takes the Waldock tests, one question immediately presented itself: whether the hostages were in immediate danger of injury or harm (over and above the harm already occasioned by their very detention). The Court carefully did not pronounce upon the lawfulness or not of the US action, but in some carefully chosen phrases indicated that it thought it inappropriate for the action to have been mounted while the matter was before the Court.

Many writers do argue against the lawfulness of humanitarian intervention today.17 They make much of the fact that in the past the right has been abused. It undoubtedly has. But then so have there been countless abusive claims of the right to self-defence. That does not lead us to say that there should be no right of self-defence today. We must face the reality that we live in a decentralized international legal order, where claims may be made either in good faith or abusively. We delude ourselves if we think that the role of norms is to remove the possibility of abusive claims ever being made. The role of norms is the achievement of values for the common good. Whether a claim invoking any given norm is made in good faith or abusively will always require contextual analysis by appropriate decision-makers—by the Security Council, by the International Court of Justice, by various international bodies. We can think of recent invocations of the right of humanitarian intervention—ranging from the Belgian and French interventions in Stanleyville in 1963, to the Israeli intervention in Entebbe in 1976, to the US intervention in Grenada in 1987. We are all capable of deciding, on the facts at our disposal, in which of these foreigners were really at imminent risk, which interventions were bona fide for reasons of humanitarian necessity, and which were not. Nor am I persuaded by another, related argument sometimes advanced—that humanitarian intervention should be regarded as impermissible, because, in the international legal system, there is no compulsory reference to impartial decision-makers, and states finish up judges in their own cause. (p. 248) There are a variety of important decision-makers, other than courts, who can pronounce on the validity of claims advanced; and claims which may in very restricted exceptional circumstances be regarded as lawful should not a priori be disallowed because on occasion they may be unjustly invoked.

What Constitutes an Armed Attack?

Article 2 (4) requires states to refrain from the threat or use of force against the territorial integrity or political independence of other states. In so far as Article 51 is meant to be the obverse side of this prohibition, it might have been expected that Article 51 would provide a right of self-defence if such prohibitions were violated. But it does not—at least not in matching terms. We have already seen that Article 51 does not provide for any self-defence against a threat of force, although the threat is a violation of Article 2 (4). For self-defence to be a legitimate response to a threat of force, the threat would have to meet the Webster tests in the Caroline.

The other puzzle has always been that self-defence is not even permitted for the other prohibited act—a ‘use of force’ against the territorial integrity or political independence of a state. It permits it only for ‘an armed attack’. Two questions are apparent. The first is whether any non-military coercion can be deemed to trigger the right to self-defence. The second is whether all uses of force are in fact armed attacks.

The answer to the first question is straightforward. In spite of occasional colloquial reference to ‘economic aggression’, nothing in Article 2 (4) deals with economic or diplomatic duress. The Charter implicitly accepts that it cannot regulate political influence and economic pressure. It may be undesirable, but it is not unlawful under the Charter. And there is certainly no suggestion that it gives rise to any right of military response.

The second question of whether all prohibited force is in fact an armed attack has been examined in great detail by the International Court of Justice in the Nicaragua v. United States Case. But as long ago as 1970 the problem had been identified with striking prescience by Professor T. Franck. Pointing out that wars of national liberation were leading to new kinds of assistance being given, he noted that ‘those new kinds of assistance do not fit comfortably into conventional international concepts and categorizations’. He continued: ‘Insofar as one state merely encourages guerrilla movements within another, an “armed attack”, at least in the conventional sense, cannot be said to have taken place. The more subtle and indirect the encouragement, the more tenuous becomes the analogy to an “armed attack”.’18 He then observed: ‘Since the Charter speaks only of (p. 249) a right to defend against an armed attack, the international community is left to ponder what principles govern the right to retort in instances of lesser trespass.’19 Franck observed that a ‘line of continuity’ runs from invasions by tanks and divisions through training, arming, sheltering and infiltrating neighbouring insurgents, all the way down to hostile radio propaganda calling for revolution in a foreign country’.20 But he thought one still could not ignore the differences.

A state can clearly engage in the use of force either directly or through the acts of irregulars for whom it has assumed responsibility. Law-making resolutions of the United Nations have consistently opposed such indirect military hostile uses of force. The General Assembly Declaration of 1965 on the Inadmissibility of Intervention stipulated that ‘no state shall organise, assist, foment, finance, incite or tolerate subversive, terrorist or armed activities directed towards the violent overthrow of the regime of another state, or interfere in civil strife in another state’.21 There are similar prohibitions in the so-called Friendly Relations Declaration, which forbade ‘the organisation of irregular forces or armed bands within the territory … of another state‘.22 Neither of these clauses makes an exception either for wars of national liberation or for the promotion of self-determination.

At the same time, and in parallel, the ideas concerning self-determination—first in its colonial context, and then more generally—took hold. How was self-determination to be accomplished against a recalcitrant colonial power save through wars of national liberation? At the same time as the law-making resolutions prohibited outside support for such activities, General Assembly resolutions began calling for support, moral and material, for such activities. In the 1970s this relatively low level transnationally supported violence was largely directed against certain remaining colonial powers. But in the 1980s it started also to be used by those who had previously opposed its legality. Indirect military assistance was provided by the West to the Mujahadeen fighting the Soviet Union in Afghanistan. This time the description was not ‘wars of national liberation’ but ‘assistance in self-defence against an invader’. By the end of the decade it was being suggested in some quarters that armed force could be used to ensure that peoples had the right to freedom of choice in their own country, when faced with oppressive regimes. Internal self-determination became the continuum of wars of national liberation, in the sense that each strongly depended on outside training, finance, and arming. Different countries engaged in various of these activities, apparently regardless of the general prohibitions, and while also denouncing the legality of the action of the other in lending such assistance.

(p. 250) The matter has come to a head in the case of Nicaragua v. United States23 though I emphasize that the type of indirect military action and support in which the United States there engaged was just one of a long series of comparable activities by many states, each of which insisted that their action was lawful (because it was for a war of national liberation, or to assist others in ridding themselves of invaders, or to allow the peoples to overthrow a dictator).

In the Nicaragua v. US Case the Court found, as findings of fact, that the United States for a period provided funds for military and paramilitary activities by the Contras—Nicaraguans in opposition to their own government, but whose military activity had been characterized by the United States as defence in aid of Honduras, Costa Rica, and El Salvador against Nicaraguan incursions into their territories. Rejecting the self-defence arguments, the Court found this US assistance unlawful. The Court also found that ‘an intermittent flow of arms’ had gone from Nicaragua to the armed opposition in El Salvador, but that the flow was not on a significant scale and the evidence was insufficient to hold the Sandanista government responsible. The Court then further found that certain transborder military incursions into Honduras and Costa Rica were imputable to the Nicaraguan government.

The Court next addressed the question as to what military acts would constitute ‘armed attack’. Citing the General Assembly Resolution on the Definition of Aggression, the Court said that an armed attack could include not merely action by regular armed forces across an international border, but also ‘the sending by or on behalf of a state of armed bands, groups, irregulars or mercenaries, which carry out acts of armed force against another state’ such as to amount to an actual armed attack rendered by regular forces. The Court found this to represent customary international law (without elaborating how it reached that view) and continued:

The Court does not believe that the concept of ‘armed attack’ includes not only acts by armed bands where such acts occur on a significant scale but also assistance to rebels in the form of the provision of weapons or logistical or other support. Such assistance may be regarded as a threat or use of force …’24

To summarize, an armed attack could take place directly, through the use of one’s own forces, or indirectly, through armed bands or irregulars. The key is the scale of the activity. If it is not very substantial, it may still be an unlawful use of force, but it will not be an armed attack—and hence no self-defence may be used against it. That finding has occasioned a torrent of criticism, the critics contending that it is an encouragement for lowgrade terrorism because the state at whom it is directed cannot use force in (p. 251) self-defence against it.25 As Judge Schwebel put it in his dissenting opinion: ‘The Court appears to offer—quite gratuitously—a prescription for overthrow of weaker governments by predatory governments while denying potential victims what in some cases may be their only hope of survival’—he is there referring to seeking assistance through collective self-defence.26

What I find puzzling about the Court’s reasoning is this. It refers to the Assembly Resolution on the Definition of Aggression27 which states that an armed attack occurs if the use of force by bands or irregulars is equivalent to an armed attack by the regular forces of a state. But how much force does one need by the regular force of a state before it is ‘an armed attack’ and allows of self-defence? If a division of troops rolls over the border, is the decision as to whether force can be used to repel them the level of force they are using? By adopting the unsatisfactory definition of the General Assembly Aggression Resolution, and proclaiming it customary international law, the Court appears to have selected criteria that are operationally unworkable. When a state has to decide whether it can repel incessant low-level irregular military activity, does it really have to decide whether that activity is the equivalent of an armed attack by a foreign army—and, anyway, is not any use of force by a foreign army entitled to be met by sufficient force to require it to withdraw? Or is that now in doubt also? Is the question of level of violence by regular forces not really an issue of proportionality, rather than a question of determining what is ‘an armed attack’?

Two final points: the Court was purporting to deal with customary international law rather than the Charter; the Court in terms avoided pronouncing upon the implications of all this for the question of whether there exists a right of anticipatory self-defence.

The Doctrinal Debates

All of the questions we have discussed are difficult. And it is not surprising that such difficult questions have occasioned different views. What has been interesting is that the debate on these issues has been one that has not merely addressed the specifics of the particular problem to hand. It has been conducted at the level of legal philosophy. In a series of vigorous exchanges (in which Professors Schachter, Henkin, Franck, Gordon, and Reisman have been leading protagonists), some general questions have (p. 252) been asked that are applicable to every one of the problems we have discussed. Are the shortcomings of the international system (the failures of the United Nations, the violations of the Charter, the massive violations of human rights, the frequent absence of democracy) such that the limits on the use of force contained in Article 2 (4) and Article 51 should be set aside?28 Or should the limitations in those provisions, and other relevant norms of international law, be respected, even though that self-restraint would appear, in our imperfect world, to consolidate the position of the wrongdoer at any particular moment?29 It is important to understand that this absorbing recent debate is not simply about whether texts should be interpreted as far as possible to support democracy and the values protected by international law. All the protagonists in the debate would agree on this. None of them takes a strict constructionist or conservative approach to textual interpretation. The argument addresses a more radical argument—namely, whether the failure of the international system, coupled with fundamentally changed circumstances since the time when the relevant texts were agreed, makes preferable unilateral action for the common good even if it is at variance with the norms articulated in the Charter and elsewhere.

This question is answered in the affirmative by Professor Reisman, who emphasizes that ‘norms are instruments devised by human beings to precipitate desired social consequences. One should not seek a point-for-point conformity to a rule without constant regard for the policy or principle that animated its prescription, with appropriate regard for the factual constellation in the minds of the drafters.’30

My own position is that, if it is felt that the erstwhile articulation of norms no longer serves community interests, then those norms can properly be subjected to processes for change. The normal processes for change will include non-compliance. New, or refined, norms often emerge from a process of widespread non-compliance with old norms. But there is a distinction between non-compliance, on the one hand, and interpretation infra legem to achieve certain outcomes, on the other. And we should not pretend that they are the same. We should, moreover, be very sure that the (p. 253) norms as presently articulated are so irredeemably inappropriate to the factual realities that we do indeed wish to undermine them. I believe that the application of Article 2 (4) and Article 51 has been very unsatisfactory. But I am not yet convinced that they have no useful purpose to perform or that unilateral outcome-directed action without reference to common norms is not dangerous.

How does this philosophical debate apply to the issues we have been discussing? How, for example, are we to interpret ‘state’, when deciding the reach of the protection allowed by international law against terrorism? The underlying question is a serious and real one, especially for lawyers who believe that international law is not just the dissecting of words as if they have no social context or political reality. And is humanitarian intervention to be permitted? And should indirect force not be allowed if it is to overthrow tyranny and establish democratic choice? In each case there are two essential choices, well illustrated by the permissibility of indirect force. The first choice is to say that, if one is to retain any control over this downward spiral into violence, the basic prohibition against the use of force, direct or indirect, by regulars or irregulars, must be maintained, with the sole exception being for self-defence. The other view is to say that, in a decentralized legal order, each action will have to be looked at on its merits and in context, and appraisals made as to whether the purpose of the military assistance is to support or to crush the values of human respect and liberty that international law should be promoting.

I generally believe that, in our decentralized legal order, facts must be looked at, and legal views applied, in context. But I also believe such policy choices are appropriate when the legal norms leave open alternative possibilities. I believe that to be the case, for example, on the question of humanitarian intervention. But I do not believe the question of the use of indirect force to be comparable. It seems clear to me that such force is prohibited by the relevant legal instruments, and that the common good is best served by terming the indirect use of force unlawful, regardless of the objectives in a particular case.


Book II, De jure belli ac pacis, ch. 1, ss. 1.4 and 2; ch. 2, s. 13.

‘For Grotius, his formulated doctrine of the “just war” was a method of controlling the unbridled power of states to act as they chose’ (R. Higgins, ‘Grotius and the Development of International Law in the United Nations Period’, in H. Bull, B. Kingsbury and A. Roberts (eds.), Hugo Grotius and International Relations (1990), 267.

See Arts. 15 and 16.

1928 General Treaty for the Renunciation of War, 94 LNTS 57.

ICJ Reports (1949) 3.

Naulilaa Case (Germany v. Portugal) 2 RIAA 1011.

GA Res. 2625 (XXV) 1970, A/8028 (1970).

Return of certain lands to Croatia; demilitarization of Krajina and its placing under UN supervised autonomy; and return of refugees to their homes. See SC Res. 743 (1992) establishing UNPROFOR, and approval of the associated UN plan in SC Res. 740 (1992).

British Parliamentary Papers, lxi; British and Foreign State Papers, xxix. 1129. Schachter states that this formulation cannot be said to have reflected state practice (‘The Right of States to Use Armed Force’ (1984) 82 Michigan Law Review 1620 at 1635).

10  For the contrary view—namely, that an armed attack must have occurred—see L. Henkin, ‘Force, Intervention, and Neutrality in Contemporary International Law’ (1963) Proc. ASIL 147, 166; P. Jessup, Modern Law of Nations (1948), 164–7.

11  1989) 87 Dept. State Bull. 87; and M. Leich, ‘Contemporary Practice of the US Relating to International Law’ (1986) 80 AJIL 612 at 632 for US explanations. See also GA Res. 43/38, 1986.

12  D. Bowett, ‘The Use of Force for the Protection of Nationals Abroad’, in A. Cassese (ed.), The Current Legal Regulation of the Use of Force (1986), 39.

13  Quoted in V. Nanda, ‘The United States Armed Intervention in Grenada—Impact on World Order’ (1984) 14 California Western International Law Journal 395 at 418.

14  For comment on ‘expanded’ claims to self-defence, see O. Schachter, ‘Self Defence and the Rule of Law’ (1989) 83 AJIL 259.

15  Recueil des cours (1952, II), 451 at 467.

16  ICJ Reports (1980) 3.

17  e.g. I. Brownlie, ‘Humanitarian Intervention’ in J. N. Moore (ed.), Law and Civil War in the Modern World (1974), 217 at 217–18; R. Falk, Legal Order in a Violent World (1968), 339; E. Jiminez de Arechaga, General Course, Recueil des cours (1978, I), 116. L. Henkin, General Course, Recueil des cours (1989, IV), 154.

18  T. Franck, ‘Who Killed Article 2 (4)?’ (1970) 64 AJIL 809 at 812.

19  Ibid.

20  Ibid. 813.

21  GA Res. 2131 (1965).

22  GA Res. 2625 (XXV) 1970..

23  ICJ Reports (1986) (Merits) 14.

24  Ibid. 104.

25  See e.g. A. Sofaer, ‘Terrorism and the Law’ (1986) 64 Foreign Affairs 901 at 919; M. Reisman, ‘Old Wine in New Bottles: The Reagan and Brezhnev Doctrines in Contemporary International Law and Practice‘ (1988) 13 Yale Journal of International Law 171 at 195–6.

26  ICJ Reports (1986) 14 at 350.

27  GA Res. 3314 (XXIX) 1974.

28  This view has been forcefully argued by M. Reisman: ‘Criteria for the Lawful Use of Force in International Law’ (1985) 10 Yale Journal of International Law 279; ‘Coercion and Self-Determination: Construing Charter Article 2 (4)’ (1984) 78 AJIL 642; and ‘Article 2 (4): The Use of Force in Contemporary International Law’ (1984) Proc. ASIL 74–87. See also E. Gordon, ‘Article 2 (4) in Historical Context’ (1985) 10 Yale Journal of International Law 271.

29  This is the position taken by O. Schachter: ‘In Defence of International Rules on the Use of Force’ (1986) 53 University of Chicago Law Review 113; ‘The Legality of ProDemocratic Invasion’ (1984) 78 AJIL 645; and ‘Self-Judging Self-Defence’ (1987) 19 Case Western Reserve Journal of International Law 121. Also T. Franck, ‘Who Killed Article 2 (4)?’ (1970) 64 AJIL 809.

30  Reisman, ‘Criteria for the Lawful Use of Force’, 283.