the gravity requirement

According to Resolution 3314, 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 of such gravity as to amount to the direct state acts of armed force listed in the resolution, or its substantial involvement therein, may amount to an act of aggression. ^{Footnote 16} Thus, for a state to commit vicarious aggression, it is not sufficient to send an armed group to act against another state. It is a requirement that this group engage in acts of armed force comparable in gravity (or effects) to those listed in the resolution (invasion, attack, bombardment, or blockade by the regular armed forces of a state). The ICJ has reaffirmed the scope of this type of armed aggression and stated that this particular provision of Resolution 3314 reflects customary international law, which means that this norm is supported by a general, consolidated practice of states accompanied by opinio juris generalis. ^{Footnote 17} The normative weight of Resolution 3314 is evidenced by the fact that the amendments to the Rome Statute of the International Criminal Court defining the crime of aggression reproduce, word by word, its text. ^{Footnote 18}

What is the difference, if any, between aggression and the use of force, given that Resolution 3314 defines aggression by express reference to the text of Article 2(4) of the UN Charter? Aggression is defined in Article 1 of Resolution 3314 as the “use of armed force by a state against the sovereignty, territorial integrity or political independence of another state, or in any other manner inconsistent with the Charter of the United Nations, as set out in this Definition.” In other words, aggression is an unlawful use of force or the use of force in contravention of the UN Charter or general international law. The novelty introduced by Resolution 3314, however, is the recognition of distinct forms of the use of force since it characterizes aggression as “the most serious and dangerous use of force.” ^{Footnote 19}

The Institut de droit international (IDI) concurs with the approach of differentiating between categories of uses of force in order to regulate the incidence of the right of self-defence. In its 2007 resolution, the IDI distinguished armed attacks of “lesser intensity” from those of “a certain degree of gravity,” stating its understanding that counter-measures in the form of police actions apply to the former and the use of force in self-defence to the latter. ^{Footnote 20} As a matter of principle, this distinction might survive scrutiny, but what acts would qualify as sufficiently grave as opposed to less intense?

Be that as it may, the ICJ endorsed a similar distinction in Military and Paramilitary Activities in and against Nicaragua, maintaining that the proscriptions regarding indirect aggression contained in the Friendly Relations Declaration relate to “less grave forms of the use of force,” while aggression would be the gravest. ^{Footnote 21} From this perspective, the use of force constitutes the genus, and aggression is a species of it, differentiated from other uses of force by virtue of the characteristics enumerated in Resolution 3314. This distinction may seem unwarranted at first sight, but one must understand that the ICJ had a special end in mind — namely to define the conditions under which a state could exercise its inherent right of self-defence in accordance with Article 51 of the UN Charter and customary international law.

Article 51 authorizes a state to use force in self-defence when it is a victim of an armed attack. Therefore, defining what constitutes an armed attack — especially when an armed group or irregular force is involved — is of paramount importance to the determination of situations in which the use of force in self-defence is lawful. The ICJ, citing Resolution 3314, has expressed the view that “the prohibition of armed attacks may apply to the sending by a state of armed bands to the territory of another state, if such an operation, because of its scale and effects, would have been classified as an armed attack rather than as a mere frontier incident had it been carried out by regular armed forces.” ^{Footnote 22} It is the court’s understanding, therefore, that the concept of armed attack, which authorizes resort to the right of self-defence, may include armed activity by armed bands, provided it occurs “on a significant scale” and produces sufficiently grave effects. In line with this finding by the ICJ, an authoritative body of opinion endorses a broad concept of armed attack that includes, as protagonists, non-official agencies such as armed bands and other entities not forming part of the regular forces of the state. ^{Footnote 23}

the accumulation theory

The precise extent and effects required to qualify armed activity as aggression were not defined by the court in Nicaragua or in any other case since, but a good starting point for ascertaining this threshold would be to refer to the specific situations or examples listed in Resolution 3314. Indeed, the court’s view does not seem to sanction a higher or distinctive threshold for armed attacks by non-state entities (as distinct from armed attacks by states directly), since it refers to military actions of the regular armed forces of a state as the relevant point of comparison. ^{Footnote 24}

A point that deserved, but did not receive, the court’s explicit attention in the Nicaragua case is whether a series of successive attacks by non-state entities that are relatively minor in scale and effect could, by accumulation, amount to an armed attack under Article 51 of the UN Charter. Such an accumulation theory would solve the conundrum that an individual attack by a non-state entity is not, as a rule, of sufficient gravity and scale to be classified as an armed attack, such that any military response by the victim state in self-defence would immediately attract charges of being disproportionate and an unlawful act of reprisal. ^{Footnote 25} To be sure, if the accumulation of incidents is disallowed as a basis for invoking the right of self-defence against non-state entities, the unavoidable consequence is an “encouragement for low-grade terrorism” or unremitting indirect aggression that most commonly affects civilians and vulnerable persons. ^{Footnote 26} Bearing in mind the risks associated with outright rejection of the accumulation theory, it is notable that this approach has indeed been ventilated by the court in two cases and that there is a growing body of authoritative opinion and state practice that lend support to it. ^{Footnote 27}

At any rate, the ICJ also made a brief statement in the Nicaragua case that is open to an interpretation that might protect the vital interests of the victim state even if the accumulation theory is discarded. In response to the United States’ claim that Nicaragua was intervening in neighbouring countries by providing assistance to armed opposition groups and launching transborder attacks, the court observed that “a use of force of a lesser degree of gravity” could entitle the victim state to take “proportionate counter-measures.” ^{Footnote 28} Setting aside the debate as to whether there is a real difference between actions in self-defence and counter-measures in response to foreign intervention, the court is here expressing its view as to what is permissible when acts of intervention involving the use of force, though of lesser intensity, take place. This is important because, as noted above, an attack by an armed group may not, in isolation, be considered of sufficient magnitude and scale to be classified as an armed attack. The use of force as a counter-measure against the non-state perpetrator of the attack and its sponsor state might arguably — according to this interpretation — be lawful, provided it respects the criterion of proportionality.

substantial involvement in an armed group’s attack

In the Nicaragua case, the ICJ drew another important distinction between unlawful armed attacks under Article 51 of the UN Charter (or acts of armed aggression) that involve “the despatch by one state of armed bands into the territory of another state” and “assistance to rebels in the form of the provision of weapons or logistical or other support,” which, according to the court, “may be regarded as a threat or use of force, or amount to intervention in the internal or external affairs of other states.” ^{Footnote 29} This distinction founded the conclusion that assistance to rebels by a state does not authorize the use of armed force in individual or collective self-defence, a view that is shared by some authors. ^{Footnote 30}

One should pause and ponder the broad implications of this proposition. Pursuant to this position, if a state provides training, arms and munitions, financial assistance, food, military equipment, intelligence, and even means of transportation to an armed rebel group that engages in armed activities against another state or its government, it would not be tantamount to armed aggression by the sponsoring state but, merely, a violation of the non-use-of-force principle and/or the non-intervention principle. The armed insurgent group may have been fully empowered to commit armed aggression against the victim state, and it may have actually engaged in military activities on a significant scale, but because it was not sent by the sponsor state, the latter will have committed no armed aggression and should not suffer military retaliation otherwise justifiable on the grounds of self-defence.

Surely this view serves to encourage further acts of intervention and vicarious aggression, for it relieves states from the most undesirable consequence stemming from an act of aggression: a forcible response. Judge Jennings sharply criticized this view in Nicaragua, arguing that “to say that the provision of arms, coupled with ‘logistical or other support’ is not armed attack is going much too far” and that this conclusion was “neither realistic nor just” but, rather, restricted without justification the requirements of lawful self-defence. ^{Footnote 31} Judge Schwebel shared this criticism, adding that “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.” ^{Footnote 32}

An analysis of the travaux préparatoires of Resolution 3314 was undertaken by Judge Schwebel to demonstrate that the ICJ’s decision fundamentally followed the reasoning behind a proposal submitted by a group of small and middle powers that was ultimately defeated in the negotiations leading to the adoption of Resolution 3314. This proposal recognized only direct armed aggression among states and denied recourse to the right of individual or collective self-defence against a state that organized or supported an armed group’s subversive or terrorist activities against other states. ^{Footnote 33}

The ICJ also seems to have neglected an important consideration in advancing its distinction between sending and supporting or assisting armed groups. As noted above, the court considered the whole of Article 3(g) of Resolution 3314 to be reflective of customary international law. Therefore, the substantial involvement of a state in the sending of armed bands that carry out armed aggression against another state ought also to be considered an act of aggression. Having ignored this part of the provision, the issue of what actions might amount to “substantial involvement” passed unnoticed in the court’s decision. However, a tentative, illustrative list of such actions was provided by Judge Schwebel in his dissenting opinion:

Indeed, supporting actions such as these should not be taken as less grievous than the “act of sending” an armed group. After all, an armed group could not be sent — and surely could not launch a successful armed attack on behalf of a state — without such operational and logistical support. As the United States maintained in the Nicaragua case, “while specific attribution is impossible, it is unquestionable that much of this cost — and a large portion of the thousands of deaths which have taken place in the past four years — would not have been incurred but for the substantial support provided by and through Nicaragua to the Salvadoran guerrillas.” ^{Footnote 35} This argument is even more persuasive if one considers the difficulty of proving the “act of sending” as opposed to the “act of providing assistance.” Arms and ammunition, for example, are rather easier to find and trace than orders or directives originating from the sponsoring state, given that vicarious aggression is usually undertaken in such a way as to conceal the involvement of the sponsoring government. ^{Footnote 36}

This debate over the distinction between “sending” and “assisting” resurfaced in the 2005 case, Armed Activities on the Territory of the Congo. There, Uganda argued that the phrase “substantial involvement therein” includes the provision of logistical support and that, as a result, “the giving of logistical support to armed bands with knowledge of their objectives constitutes an armed attack.” ^{Footnote 37} The Democratic Republic of the Congo (DRC) argued, in turn, that substantial involvement requires that the sponsoring government give “specific instructions or directions” to the armed group or that it “controls” the performance of such acts. ^{Footnote 38} This last view, however, seems to be more concerned with the attribution of state responsibility for the actions of the armed group (an argument that surely favoured the DRC’s position in the case, as it is difficult to apply rules of attribution to any particular situation). The ICJ sidestepped the issue raised by Uganda, turning its attention solely to the question of attribution — that is, whether the alleged armed attacks by the armed group against Uganda could be attributed to the DRC. ^{Footnote 39}

The distinction between “sending” and “assisting” does not appear in the definition of aggression contained in the 2005 African Union Non-Aggression and Common Defence Pact. This treaty defines an act of aggression as “the sending by, or on behalf of a Member State or the provision of any support to armed groups, mercenaries, and other organized trans-national criminal groups which may carry out hostile acts against a Member State, of such gravity as to amount to the acts listed above, or its substantial involvement therein”. ^{Footnote 40} Equating assistance to sending highlights the connection between the resulting armed action perpetrated by the armed group and the political objectives of the sponsoring state. It is indeed difficult to conceive that a state would provide substantial support to an armed group without knowledge of this entity’s aims and agenda, including the eventual targets of its armed actions. Otherwise, how could the supporting state, for example, be reassured that the armed group would not turn its weapons against that state’s own population and officials? Why would a state spend money, intelligence, and other resources, and “donate” weapons and equipment, unless it sees the actions of the armed group as advancing its own foreign policy agenda?

For these reasons, substantial support should be taken as presumptive evidence of the act of sending the armed group, or of the fact that the armed group is acting on behalf of the supporting state, assimilating this armed group to a de facto organ of that state. In conclusion, notwithstanding the court’s position and reservations, the foregoing considerations suggest that it is unjustified to exempt a state from responsibility for the commission of armed aggression when it has played a significant role in encouraging and enabling armed attacks by a non-state armed entity.

the origin of the non-state entity

The ICJ’s advisory opinion in Legal Consequences of the Construction of a Wall in the Occupied Palestinian Territory (Israeli Wall) set out another condition for the exercise of self-defence by the victim state: that the attack by the armed group originate from abroad. ^{Footnote 41} In a similar vein, the North Atlantic Treaty Organization’s (NATO) first invocation of Article 5 of the North Atlantic Treaty initially sought to determine whether the 11 September 2001 armed attack suffered by the United States had come from abroad. ^{Footnote 42} Unlike the court in Israeli Wall, however, NATO did not view the origin of the attack as a condition precedent for the exercise of self-defence but, instead, as a question relating to whether the situation came within the solidarity clause under the North Atlantic Treaty. ^{Footnote 43} The reasoning underlying such an inquiry was that armed attacks emanating from domestic non-state entities should be dealt with by the security apparatus of the state under its domestic legal system rather than by outside intervention. However, the question still remains as to whether foreign state assistance to a domestic armed opposition group that enables it to carry out domestic armed attacks will attract the international responsibility of the foreign state concerned (for violation of the non-intervention and/or non-use of force principles).

general considerations

There is sufficient evidence in state practice and support from law-determining agencies to authorize the view that when a victim state exercises its right to self-defence against armed aggression by a foreign-based, or foreign-supported, armed group, it may, under certain circumstances, be entitled to extend this forcible response to the territory of the state that is hosting, supporting, or harbouring the group. The reach of the right to self-defence against non-state entities located in the territory of another state is grounded in a solid legal tradition. It can be traced back to the old legal maxim sic utere tuo ut alienum non laedas, which applies to an owner of property (in this case, the territorial state), requiring it to use its property in a way that does not cause injury or harm to others. ^{Footnote 85} This axiom has become a generally accepted principle of international law applicable to the broad field of international responsibility, ^{Footnote 86} including in connection with the use of international watercourses, ^{Footnote 87} territorial waters, ^{Footnote 88} other marine areas, ^{Footnote 89} and outer space, ^{Footnote 90} as well as generally in relation to the protection of the environment. ^{Footnote 91} No less important, though, is its association with the doctrine of the abuse of rights. As Hersch Lauterpacht has pointed out, a “large part of the law of intervention is built upon the principle that obvious abuse of rights of internal sovereignty, in disregard of the obligations to foreign states … constitutes a good legal ground for dictatorial interference.” ^{Footnote 92}

This principle has been affirmed in authoritative international jurisprudence. The Island of Palmas arbitration award of 1928 asserted the duty of every state, arising from its territorial sovereignty, to protect the rights of other states within its territory. ^{Footnote 93} In the 1949 Corfu Channel case, the court affirmed what it considered to be a general and well-recognized principle, namely “every state’s obligation not to allow knowingly its territory to be used for acts contrary to the rights of other states.” ^{Footnote 94} Finally, the same principle underlies Article 3(f) of Resolution 3314, which characterizes as an act of aggression “the action of a state in allowing its territory, which it has placed at the disposal of another state, to be used by that other state for perpetrating an act of aggression against a third state.” ^{Footnote 95} The UN secretary-general’s Survey of International Law in Relation to the Work of Codification of the International Law Commission summarized well the state of international law on this matter back in 1949: “There has been general recognition of the rule that a state must not permit the use of its territory for purposes injurious to the interests of other states in a manner contrary to international law. This rule has been applied, in particular, with regard to the duty of states to prevent hostile expeditions against the territory of their neighbours.” ^{Footnote 96}

In state practice, reference can be made to the Caroline affair (1837–41), a historic, landmark international dispute from which key requisites for the lawful exercise of self-defence were derived. ^{Footnote 97} This case concerned a cross-border armed operation by British forces against the steamboat Caroline, which was being used to provide assistance to a rebel band located at the border between then Upper Canada and the US state of New York. The ensuing exchange of notes between the British and US governments indicated that both nations agreed with the general principle that limited cross-border armed operations against armed opposition groups located in another state’s territory were permissible only as an extraordinary measure of self-defence, provided the requirements of necessity and proportionality were fully satisfied. ^{Footnote 98}

One should not under-estimate the continuing relevance of the Caroline case for the appraisal of state practice. In examining the guiding criteria of proportionality and necessity for the lawful exercise of self-defence, for example, scholarship continues to reference the Caroline affair. ^{Footnote 99} Similarly, and notwithstanding the significant normative developments that have taken place since the adoption of the UN Charter regarding the use of force, the Caroline affair continues to be referenced in the literature, particularly with respect to the right of anticipatory self-defence in the face of an imminent armed attack as well as to the right of self-defence against non-state actors. ^{Footnote 100}

Drawing from an extensive analysis of states’ domestic and international practice, Lauterpacht reached the conclusion that international law “imposes upon the state the duty of restraining persons resident within its territory from engaging in such revolutionary activities against friendly states as amount to organized acts of force in the form of hostile expeditions against the territory of those states.” He thought this obligation would flow not only from the law of neutrality but also, primarily, from the “well-established customary rule that the territory of a state must not be allowed to serve as a base for military or naval operations against another state.” ^{Footnote 101}

When this legal principle is violated with the consequence that an armed attack is suffered by another state, some of the most fundamental rights of the victim state are violated. Surely inspired by the vital role played by this principle in the maintenance of international peace and security, the UNSC occasionally calls on states to “prevent armed individuals and groups from using their territory to prepare and launch attacks on neighbouring countries” ^{Footnote 102} or stresses that “every Member State has the duty to refrain from organizing, instigating, assisting or participating in terrorist acts in another state or acquiescing in organized activities within its territory directed towards the commission of such acts.” ^{Footnote 103}

In view of these legal considerations, the IDI has conceded the lawfulness of the use of force in self-defence against a state that directs, instructs, or controls a non-state entity that launches a cross-border armed attack. ^{Footnote 104} This view is warranted by the fact that in these cases the non-state entity acts as a de jure or de facto agent of the directing, instructing, or controlling state, such that international responsibility can be attributed to that state. ^{Footnote 105} It is submitted, however, that the scope of self-defence may be extended to reach the territory of a host state irrespective of whether it has played a role in directing, instructing, and controlling the non-state entity, particularly in situations where the host state is unwilling or unable to restrain and disband the armed group.

host state unwillingness or inability to control the non-state entity

It is generally recognized that failed or failing states are the main safe haven for armed groups due to the existence of ungoverned or ill-governed spaces in their territory. ^{Footnote 106} What, then, is the appropriate response when the host or harbouring state is both a failed state and unwilling or unable to control or neutralize armed bands operating from its territory? In determining a state’s responsibility for armed attacks launched against foreign targets by non-state entities located in its territory, the approach adopted by the ICJ seems to take into account the territorial government’s capacity to exercise its authority and jurisdiction effectively. In the Armed Activities case, Uganda contended that the DRC had breached its duty of vigilance by tolerating or acquiescing in the cross-border armed activities of anti-Ugandan rebel groups operating from DRC territory. ^{Footnote 107} Counsel for Uganda argued for full responsibility of the host state and the consequential applicability against it of the right to self-defence, asserting that when a state “tolerates the activities of armed bands and the armed attacks which they launch against a neighbouring state, the failure to control renders the state harbouring such armed bands susceptible to action in accordance with Article 51 by the victim state.” ^{Footnote 108} Sound as this viewpoint may seem, it is at variance with the position maintained by the ICJ. The argument was dismissed by the court on the grounds that the DRC’s omission was justified by the difficult terrain where the armed groups operated and the lack of central government control over that territory. ^{Footnote 109}

Judge Kooijmans found this view too simplistic, inasmuch as it exempted the state from showing that it had made an effort to fulfil its duty of vigilance by adopting measures calculated to prevent rebel bands from launching cross-border attacks. ^{Footnote 110} Indeed, it should be insufficient for a host state simply to allege, as a circumstance precluding the wrongfulness of its conduct, powerlessness to exercise control over part of its territory or armed groups located there, given that the state’s “inability to discharge the duty does not relieve it of the duty.” ^{Footnote 111} The state should have to show efforts in good faith to prevent its territory from being used for acts contrary to the rights of other states. Concrete actions of prevention (through domestic law enforcement measures) and cooperation with potential or actual victim states, designed to address promptly the threat emanating from its territory, ought to be the standard of conduct by which wrongfulness is judged. Along these lines, the IDI recognized, in its 2007 resolution, an international legal obligation of cooperation in this situation: “The state from which the armed attack by non-state actors is launched has the obligation to cooperate with the target state.” ^{Footnote 112}

Cooperation is a particularly relevant tool for any failed state that wants to compensate for its lack of adequate resources to deal with armed groups in its territory. Inaction or omission, by contrast, may give rise to charges of aggression. According to the 1933 Convention for the Definition of Aggression (London Convention), for instance, the failure to discharge the obligation of cooperation could be considered an act of aggression. ^{Footnote 113} This convention classifies as an act of aggression the “provision of support to armed bands formed in its territory which have invaded the territory of another state, or refusal, notwithstanding the request of the invaded state, to take, in its own territory, all the measures in its power to deprive those bands of all assistance or protection.” ^{Footnote 114} Despite its conventional nature and res inter alios acta effect, this provision is at least recognition of the importance attached by the parties to appropriate efforts to counter the transnational threat posed by armed groups. Later, at the Nuremberg trials, the general normative significance of the London Convention was affirmed when the chief prosecutor for the United States called it “one of the most authoritative sources of international law” on the subject of aggression and made express reference to its provisions in order to define what constitutes an aggressor state. ^{Footnote 115}

The view that states are under a legal duty of prevention and cooperation in regard to armed actions of non-state entities located in their territory seems to be in accord with state practice. In the Caroline case, for instance, the United States was accused of providing a safe haven to rebel bands operating against British possessions in what later became Canada. The US government argued that its conduct was in conformity with international law (in particular, the non-intervention principle), drawing Britain’s attention to preventive measures taken in that regard, such as its laws on neutrality and an act of Congress adopted in 1838 ^{Footnote 116} intended to restrain “military enterprises from the United States into the British Provinces.” ^{Footnote 117}

A clearer example, involving a failed state and its efforts to protect the rights of other states from acts of domestic armed groups, is found in the 2011 cooperation arrangement between the government of Kenya and the transitional federal government (TFG) of Somalia. ^{Footnote 118} This agreement was designed to deal with the threat posed by the armed group Al-Shabaab, which, operating from Somali territory, had launched successive armed attacks against both Kenya and Somalia, injuring and killing scores of government officials and civilians. By this arrangement, the two governments agreed to cooperate in “undertaking security and military operations, and to undertake coordinated pre-emptive action and the pursuit of any armed elements that continue to threaten and attack both countries.” ^{Footnote 119} On the basis of this agreement, Kenya was able to pursue, into Somali territory, members of Al-Shabaab that were participating in cross-border attacks against it. In conveying the text of the agreement to the president of the UNSC, Kenya made it clear that it would be using force as “remedial and pre-emptive action,” thus indicating the legal basis of its actions: anticipatory or pre-emptive self-defence against a non-state entity, the extraterritorial scope of which would be sheltered by the said agreement. ^{Footnote 120}

As the ICJ has noted, good faith in international relations is a vital principle of international law, whose function is to guide the fulfilment of legal obligations. ^{Footnote 121} In cooperating with Kenya to address this common threat, the Somali government seems to be discharging its duty of vigilance in good faith while, at the same time, protecting its sovereign rights (since bilateral cooperation is a manifestation of the territorial state’s consent rather than a violation of its sovereignty or political independence).

By contrast, Afghanistan and Sudan have been accused of not showing good faith efforts to fulfil their duties as territorial sovereigns. When the United States launched military operations directed at Al-Qaeda targets in response to the 1998 armed attacks against its embassies and nationals, it sought to justify the cross-border nature of its acts by referring to the fact that Al-Qaeda had maintained a network of facilities in Afghanistan and Sudan from which terrorist attacks had been launched against American targets and that the governments of Sudan and Afghanistan had not acted upon repeated efforts “to shut these terrorist activities down and to cease their cooperation with the Bin Laden organization.” ^{Footnote 122}

Similar grounds were alleged by Turkey in 2015 when it identified ISIL as the entity responsible for cross-border armed and terrorist attacks against Turkish nationals and members of its armed forces. Invoking Article 51 of the UN Charter, it launched military operations against ISIL targets located in Syria. Turkey justified its military actions in self-defence by arguing that Syria had become a “safe haven” for ISIL, an area used “for training, planning, financing and carrying out attacks across borders platform from which it launched armed attacks against Turkey.” ^{Footnote 123} Most significantly for present purposes, Turkey contended that the Syrian government was “neither capable of nor willing to prevent these threats emanating from its territory which clearly imperil the security of Turkey and safety of its nationals.” ^{Footnote 124}

Back in 1958, France justified military measures against Tunisian forces and rebel bands located in Tunisian territory by reference to numerous border violations and incursions into French territory by Algerian rebel bands, allegedly aided and abetted by the Tunisian authorities, and to the refusal of Tunisian authorities to take the necessary measures to prevent the recurrence of such incidents, despite several “warnings and suggestions” issued by France. ^{Footnote 125}

Absence of effort in good faith to control and disband armed groups and to cooperate with a target state might thus be seen as just cause for that state’s use of force in self-defence in the territory of the host state. Conversely, to require the prior consent of the host state or a reasonable basis for attributing responsibility to that state might be considered an impediment to the exercise of the right of self-defence. ^{Footnote 126} It is no wonder, therefore, that states continue to advance claims on the former basis. In 2002, Russia alleged that armed and terrorist attacks were being launched against Russia from a territorial enclave along the border between Georgia and Russia. It claimed that such enclave was outside the control of Georgia, which seemed “unable or unwilling to counteract the terrorist threat” and that Georgia had not accepted several proposals for joint cooperative action to halt the cross-border incursions or, alternatively, requests for extradition of the perpetrators. Russia concluded with a warning that if Georgia is “unable to establish a security zone … and does not put an end to the bandit sorties and attacks … we reserve the right to act in accordance with Article 51 of the Charter of the United Nations.” ^{Footnote 127}

The fulfilment of the obligation to cooperate with the victim state is often encouraged or endorsed by the UN — in particular, the UNSC — which considers cooperative arrangements a necessary element of a desirable political answer to cross-border attacks by armed groups involving two or more states. At the same time, a study of the practice of the UNSC shows a reluctance to adjudicate conflicting legal claims and attribute responsibility, as this might prejudice a political settlement that would be acceptable to the governments concerned. ^{Footnote 128} In 1998, amid the humanitarian crisis in Darfur, the UNSC received letters from Chad and Sudan in which each accused the other of aggression through planning, training, arming, financing, and sending of armed groups involved in armed attacks against it. ^{Footnote 129} Both states claimed to have acted in self-defence in response to such aggression. The UNSC adopted a resolution under Chapter VII of the UN Charter that made no determination as to which state had committed aggression or which could claim a lawful exercise of its right of self-defence. Instead, it offered the ingredients of a political solution to the crisis, expressing its concern “at the activities of armed groups and other attacks,” demanding the immediate cessation of violence by those armed groups and encouraging the governments of Sudan, Chad, and the Central African Republic to “ensure that their territories are not used to undermine the sovereignty of others” and to cooperate “with a view to putting an end to the activities of armed groups in the region and their attempts to seize power by force.” ^{Footnote 130}

negative consequences of extending the use of force in self-defence to the territory of the host state

As a corollary to the failure of a host state to fulfil its duty to prevent and cooperate, a state that suffers an armed attack from an armed group based in the host state ought to be able to implement forceful measures in self-defence against that group despite the fact that it is headquartered abroad. ^{Footnote 131} The application of this principle, however, is not free from considerable difficulties. First, there is no doubt that any such forceful measures in self-defence would prima facie be deemed to violate the host state’s territorial integrity and political independence and, thus, be regarded by that state as an infringement of the non-use-of-force principle and its sovereignty. For instance, when military forces from Uganda and Rwanda entered the DRC under the justification of defending themselves against armed groups operating from DRC territory, the DRC considered itself a “victim of armed aggression, as defined in resolution 3314.” ^{Footnote 132}

In addition, third states might consider the host state to be the victim of foreign aggression and, hence, offer military aid (possibly under the justification of collective self-defence), further aggravating the conflict. For instance, in 1999, following an express invitation by the Congolese government, Uganda’s and Rwanda’s actions in the DRC triggered military assistance to the DRC by three other African nations (Namibia, Angola, and Zimbabwe) on the basis of their commitments under the Treaty of the Southern African Development Community. ^{Footnote 133} However, that said, no state should be allowed, as a matter of principle, to hide behind the shield of its sovereignty when it is pursuing shady foreign policy goals in connection with unlawful and vicarious uses of force.

Recognizing the windows of opportunity that would be open to great powers by recognition of this principle, authors have raised the objection that it would allow the strongest to act as iudex in causa sua, blurring the line between defensive and offensive (or aggressive) action and, ultimately, undermining the rule of law in the international community. ^{Footnote 134} There is no question that the exercise of any right might be subject to abuse, undermining the rules-based international order. As Judge Lauterpacht has noted, every state is expected to avoid crossing “the imperceptible line between impropriety and illegality, between discretion and arbitrariness, between the exercise of the legal right ... and the abuse of that right.” ^{Footnote 135} The risk of an abuse of right is bound to be particularly high if the intervening state is a great power, is acting on the basis of its own unilateral assessment of the facts, and if reprisal and punishment, rather than self-defence, are the true impetus for its actions. ^{Footnote 136} One is also aware that in the misapplication of the principle, a powerful state might resort to false claims and staged incidents in order to justify armed intervention against weaker states.

The problem with this objection is that it applies equally to all principles and norms associated with the use of force in international relations. One cannot dissociate power politics from international politics. From a legal viewpoint, the misapplication or misinterpretation of any legal principle is always a possibility, given the limitations of an international legal system that reflects the decentralized nature of international society (further discussed later in this article). Legal justifications will ultimately be judged by members of the international community, and any resort to armed force is subject to the mechanisms of control and supervision of the UN collective security system (however (in)effective they might be).

necessity in self-defence against non-state entities

Necessity dictates that no military intervention abroad in self-defence would be justified if the host or harbouring state takes immediate and effective measures to neutralize, disarm, and disband the armed group, adopts measures to hold accountable those responsible for crimes under international law (or is willing to extradite them), and shows a disposition to compensate for damages incurred by the victim state. This would be the ideal type of response from the host state, but one must concede that this attitude is likely to be rather unusual, such that it would be more realistic to expect a more measured reaction by the host state, one that takes into account its resources, its degree of control over the territory, and its foreign policy goals.

As for the victim state, it seems unreasonable to require that it await action by the harbouring or host state before taking forcible measures in self-defence in order to halt or repel an ongoing armed attack; after all, such a situation would leave the victim state “no choice of means, and no moment for deliberation.” ^{Footnote 138} Equally, there is no reason why the victim state should await measures against the non-state entity by the host state when the latter is clearly unable or unwilling to take such measures. For instance, in 1996, Israel took military measures against Hezbollah, alleging an exercise of its right of self-defence under Article 51 of the UN Charter. During the ensuing debates in the UNSC, the representative of Egypt conceded that “firing Katyusha rockets across borders is indeed a proscribed act which must cease forthwith.” However, instead of upholding Israel’s right to self-defence, it proceeded to argue that Israel and Lebanon should have utilized the mechanisms provided by the armistice agreement in force and that these two countries could also have reached an agreement on security guarantees to be implemented by the UN. ^{Footnote 139} This prescription could perhaps have played some role in settling the conflict in the long run and functioned as an element of a more comprehensive, regional settlement. At the time, however, Lebanon was unable to negotiate or implement any agreement with Israel for the simple fact that it clearly lacked control over its southern territory and the Hezbollah organization. The representative of Egypt must have known this since the UNSC had tried without success, since 1974, to assist Lebanon in re-establishing its authority and control over the area. ^{Footnote 140} In cases like these, to expect the victim state to first seek the host state’s cooperation in controlling and suppressing the armed group amounts to a denial, in practice, of the right of self-defence.

proportionality in self-defence against non-state entities

Proportionality would in principle constrain the military operation in self-defence to the targeting of the military assets of the armed group, in a way that complies with international humanitarian and human rights law. In contrast, a comprehensive military operation intended concurrently to target the non-state entity and promote regime change in the host state would be difficult to justify legally, as this would constitute a patent infringement of the non-intervention principle and would be hard to reconcile with the legitimate ends of the right to self-defence.

As it stands, the law seems to be unsettled as to whether the defending state is entitled to engage in “hot pursuit” of elements of the armed group into the host state’s territory without the consent of the latter. ^{Footnote 141} It might be argued that the forces of the victim state should stop at the border, giving the host state the opportunity to take immediate measures to prevent future attacks against the victim state. A factor that should be taken into account in the legal assessment of such a situation is whether the attack is one of a series of previous attacks that the host state has failed to prevent. If that is the case, there ought to be a legal presumption against the host state that it is unlawfully using the armed group as a proxy to promote its national and foreign policy objectives. As Derek Bowett has observed, “it is conceivable that the only measures of protection against subsequent attacks or incursions by armed bands may lie in pursuing them to destruction across the frontiers of a neighbouring state.” ^{Footnote 142}

If such a right of hot pursuit, regardless of the territorial state’s consent, is to become generally recognized in state practice, there should be some degree of formal or informal coordination with the host state in order to avert the extension of the military action to the host state’s military or civilian assets. Targeting the host state’s military forces should be lawful only in self-defence — that is, in the face of an actual or imminent armed attack by the host state in the course of the victim state’s defensive operations. Equally, armed incursions in hot pursuit should not, as a rule, lead to the occupation of the host state. These legal standards, if followed, should limit the scope (and possibly the duration) of the armed conflict.

In practice, however, it would be extremely difficult to judge competing legal claims in a situation of hot pursuit. There are examples of state practice in which hot pursuit has been alleged, although such claims have usually been met with a conflicting claim of an unlawful armed attack by the other party. One such situation took place in 1997 when Angola alleged self-defence to justify the actions of its armed forces, which had entered Congolese territory in pursuit of armed groups immediately following the latter’s armed attacks. ^{Footnote 143} Similarly, a finely crafted legal justification for a cross-border military operation was advanced by Iran in 1996 in order to demonstrate full respect for the standards set by international law. First, Iran set out the factual basis for the right claimed (alleging that armed terrorist groups had launched transborder armed attacks against Iranian border towns from Iraq) and the scope of the right exercised (specifying that its forces had “pursued the retreating armed groups” and “targeted their training camps in Iraq”). It then proceeded to argue that the legal requisites for a lawful exercise of the right to self-defence had been satisfied, noting that the measures taken were both necessary and proportional, the operation had already been concluded, and that it respected the territorial integrity of Iraq (supposedly due to immediate withdrawal). ^{Footnote 144}

There is some authority — though it is disputed — in support of the view that a state can use force to prevent the continuation of attacks in the foreseeable future. The view that self-defence should be solely a lawful instrument to stop an armed attack in progress by a non-state entity is deemed to be more in accord with the spirit of the UN Charter, which was designed to restrict the unilateral use of force in international relations so as to privilege the operation of the collective security system. ^{Footnote 145} This is why, in the UN Charter framework (pursuant to Article 51), recourse to self-defence must be followed by a notification to the UNSC, which is then expected to take the necessary measures in order to restore international peace and security. Thus, the system presumes that the UNSC will intervene and act to halt the aggression and prevent its recurrence, leaving the victim state with no need or justification to proceed with its own forceful measures.

The question arises, however, whether the victim state is under a legal obligation to cease any current or future acts of self-defence against the non-state armed group as soon as the UNSC adopts collective measures on the basis of Chapter VII of the UN Charter. There are those who argue that, once the UNSC is seized of the situation and exercises its responsibilities under the UN Charter, the right to self-defence loses its object and ceases to be applicable, without prejudice to its resurrection should a new armed attack occur. ^{Footnote 146} This debate actually arose during consideration by the UNSC of the Falkland Islands/Malvinas conflict in 1982. Argentina argued for the cessation of all hostilities (thus affecting the use of force by the United Kingdom in self-defence) after the UNSC adopted a resolution addressing the situation, while the United Kingdom contended that the UNSC’s measures had to be effective first in order that international peace and security be restored. ^{Footnote 147} The practice of the UNSC seems to support the latter stance since it shows UNSC resolutions that recall the inherent right of self-defence while imposing binding coercive and non-coercive measures under Chapter VII. ^{Footnote 148} The lesson to be learned is that the right to self-defence remains applicable until collective security measures have proven to be effective in restoring international peace and security, thus eliminating any imminent threat to the state concerned.

This view is vindicated by the recognized fact that the collective security system is imperfect at best. The fact that the UN Charter safeguards the inherent right of self-defence represents a recognition (or premonition) by the drafters that collective measures will not always work due to anticipated political restrictions surrounding the smooth operation of the UNSC. The latter’s functioning is limited by the prerogative of the veto enjoyed by the permanent members of the UNSC, any of which can block action by the organ; not to mention the political agendas of all other states or coalitions of states represented in the organ from time to time, which can impair a sound and objective legal judgment or political settlement of any dispute. ^{Footnote 149} The range of allies and sympathizers the victim or host states can mobilize will determine in large measure the degree of support for their cause and the legal justifications they adduce. Moreover, collective measures short of the authorization of force, such as provisional measures and sanctions, may not work as planned. For instance, in the Falkland Islands/Malvinas conflict, UNSC Resolution 502 demanded an “immediate withdrawal of all Argentine forces” from the islands, but this demand was not fulfilled despite the binding nature of the decision and the good offices and mediation efforts of the UN secretary-general, the Peruvian president, and the US secretary of state. ^{Footnote 150} Asserting a continuing right of self-defence, the British military forces finally landed in the islands and retook them by force, exactly forty-nine days after the adoption of Resolution 502.

If account is also taken of legal developments that would extend the scope of the general right to self-defence to include imminent attacks, it must be admitted that states are allowed to resort to force in self-defence even after the “first wave of attacks” is complete, provided there are reasonable grounds to believe that further attacks are imminent. ^{Footnote 151} Endorsement of this legal view is indeed found in state practice. For instance, in 1998, the United States justified its defensive action in response to armed attacks against its embassies and nationals abroad by reference to warnings by Al-Qaeda that strikes would “continue from everywhere” as well as to evidence that “further such attacks were in preparation.” ^{Footnote 152} In view of this alleged imminent threat, the United States argued that it “had no choice but to use armed force to prevent these attacks from continuing.” ^{Footnote 153} A slightly broader legal rationalization (for it did not refer to specific threats made) was advanced by the United States to justify its response to the terrorist attacks of 2001, as it alleged that defensive measures were designed to “prevent and deter further attacks on the United States.” ^{Footnote 154} With regard to the same attacks, the British government set out the prevention of the “continuing threat of attacks from the same source” as the aim of its defensive action against Al-Qaeda. ^{Footnote 155} The British attorney general’s reply to a question posed in Parliament on the matter clarified that “it must be right that states are able to act in self-defence in circumstances where there is evidence of further imminent attacks by terrorist groups, even if there is no specific evidence of where such an attack will take place or of the precise nature of the attack.” ^{Footnote 156}

The problem with this approach, however, is that it may confound self-defence and armed reprisal, at least in the eyes of the host state and its allies, giving rise to claims of an unlawful use of force. It must be mentioned in this regard that Chatham House adopts a slightly more restrictive view regarding imminent attacks by non-state entities, asserting that the territorial state should be given leeway to take domestic measures against the entity and that self-defence ought to be allowed only in cases of “compelling emergency.” ^{Footnote 157}

Authors make a distinction between deterrence, punishment, and prevention for the purpose of determining the legitimate ends of self-defence and, therefore, whether its invocation is proportional or not and, hence, lawful or unlawful. ^{Footnote 158} This intellectual exercise is certainly useful for academic purposes. In practice, however, one can hardly be expected to identify and separate the three purposes in any given situation since they are all inter-connected. In other words, there will always be an element of retribution, prevention, and deterrence attached to any purported defensive action. It is indeed the act of punishment — expected by the population of the victim state — that adds credibility to the threat of future, automatic retaliation in case of further attacks, which in turn is relevant to the prevention and deterrence of repeat conduct by the non-state entity (and its sponsoring or host state). ^{Footnote 159} Having said that, the language of the legal justification presented by the state exercising its right to self-defence should be carefully drafted in order to avoid any direct reference to retribution, focusing instead on self-defence, prevention, and deterrence. In this way, the state may avert the charge of having launched an unlawful armed reprisal. ^{Footnote 160}