4.3.1 ‘If an Armed Attack Occurs …’

At the time of the adoption of the Charter, in light of the experience of World War II, the main threat to international peace and security was envisaged as a threat of armed attack by one state against another state. Accordingly, self-defence is provided for ‘if an armed attack occurs’.Footnote ³⁵ This does not correlate precisely with article 2(4), which prohibits ‘the threat or use of force’, indicating that some such incidents may not justify force in self-defence. Nor is it configured in the same terms as the trigger for SC response under chapter VII of the Charter: ‘breach of the peace, threat to the peace or act of aggression’.Footnote ³⁶ Nevertheless armed force that cannot be justified as self-defence may constitute aggression.Footnote ³⁷ The first question is therefore what constitutes an armed attack. Unlike aggression, there has been no negotiated definition. Authorities suggest that ‘to be deemed an armed attack, an operation must have a minimum “scale and effects”’.Footnote ³⁸ In the Oil Platforms case the ICJ stated the need to distinguish ‘the most grave forms of the use of force (those constituting an armed attack) from other less grave forms’.Footnote ³⁹ While there must be cross-border forcible action to constitute an armed attack, minor localised border skirmishes are unlikely to be accepted as such, even if they involve loss of life,Footnote ⁴⁰ although ‘a cumulative series of minor attacks may constitute an armed attack’.Footnote ⁴¹ A consequence of excluding small-scale violence from the concept of ‘armed attack’ is to rule out the right to self-defence and to minimise the threat of escalation of violence; contrariwise it has been argued that so doing is ‘conceptually confused, inconsistent with customary practice and undesirable as a matter of policy. … The better view is that any deliberate projection of lethal force onto the territory of another state … will normally trigger the application of Article 2(4)’.Footnote ⁴²

Events in the Crimea in 2014 raise the question whether of an armed attack has occurred where there is an undoubted military presence but no use of direct force. Russian troops were legally present in the Crimea in accordance with the terms of a 1997 Treaty (extended in 2010) agreed between Russia and Ukraine.Footnote ⁴³ But the Treaty imposed restrictions on those foreign troops; major troop movements required prior consultation, and numbers could not be unilaterally increased. Russia did not comply with these terms. The 1974 GA definition of aggression provides that ‘[t]he use of armed forces of one State which are within the territory of another State with the agreement of the receiving State, in contravention of the conditions provided for in the agreement’ qualifies as an act of aggression. But, as stated earlier, not every act of aggression constitutes an armed attack, triggering the right to self-defence; is it an armed attack when no shot has been fired? The indisputable show of considerable force made possible the referendum providing for Crimea’s secession from Ukraine,Footnote ⁴⁴ which ensured that the ‘scale and effects’ of the action were grave: annexation of the territory of a sovereign state.Footnote ⁴⁵ As has been pertinently pointed out, the implications for the UN Charter if these events did not constitute an armed attack would be that ‘Ukraine could not lawfully use force against Russian troops to protect territory that undisputedly is part of Ukraine’Footnote ⁴⁶ The affair seemed rife with Cold War assumptions about the use of force by a regional hegemon in its own backyard.Footnote ⁴⁷ On the other hand, the presence of the Russian military did involve violations of human rights on a considerable scale: this included holding a referendum without proper procedures and under military pressure; discrimination against ethnic Ukrainians and Crimean Tartars; infringement of property rights; and control over media and information.Footnote ⁴⁸Indeed, it could be argued that the debate about whether the annexation of Crimea constituted an armed attack obscured the more important practical debate about how to respond to violations of human rights.

In other scenarios the ICJ has been reluctant to recognise all forms of even coercive intervention by a state into another state as an armed attack. In the Nicaragua case the Court found that sending armed bands into the territory of another state could constitute an armed attack if it was of sufficient scale and effects. Assisting rebel forces (e.g., through the provision of weapons or logistical support) is not in the opinion of the Court an armed attack,Footnote ⁴⁹ although it may constitute a threat or use of force, or illegal intervention. In the Court’s view the appropriate legal response to wrongful intervention is not military force in self-defence but the imposition of non-forcible counter-measures. However, these may be practically impossible for a weaker state to impose against a more powerful oneFootnote ⁵⁰ in the same way as power disparities in gender relations may make it impossible for the weaker party (usually the woman) to seek sanctions against perpetrators of violence. This approach allows for instability and insecurity in the target state and the wider region, as experienced in the conflicts in Central America in the 1980s and demonstrated by the Nicaragua case.

4.3.2 Collective Self-Defence, Consent and Invitation

A further complexity is added where the armed attack is against a state other than the one responding, that is, where states A and B seek to react in self-defence to an attack on C, even though they have neither suffered nor anticipate any attack against themselves. The attacked state, C, has the right to collective self-defence, to seek assistance from other states, as Kuwait did following the invasion by Iraq in 1990. In Resolution 678 (29 November 1990) the SC authorised ‘States cooperating with the Government of Kuwait … to use all necessary means’ against Iraq.Footnote ⁵¹ Kuwait’s right of collective self-defence meant that this authorisation was not strictly necessary. Nevertheless, SC approval added legitimacy to the collective military action against Iraq and provided the opportunity for the invocation of Council powers under UN Charter, chapter VII in the changed political environment of 1990, giving rise to the optimism of a New (post–Cold War) World Order.

Defensive alliances have long been part of the European landscape and were instrumental in the advance to war in 1914Footnote ⁵² and 1939. Part of the Cold War landscape was the conclusion of collective self-defence arrangements committing states to respond to an attack on one of their members, such as NATO,Footnote ⁵³ SEATOFootnote ⁵⁴ and ANZUSFootnote ⁵⁵ (all protecting the security of US allies against possible communist attack) and the Warsaw Pact,Footnote ⁵⁶ with membership of the Eastern European states. The Gulf Cooperation Council Peninsula Shield Force was formed towards the end of the Cold War in 1984 in light of the regional instability engendered by the Iran-Iraq war, which broke out in 1980. Some collective self-defence institutions have disappeared (Warsaw Pact), while others have changed their role (NATO acting in the name of humanitarianism in 1999 with the concerns for human security expressed in Chapter 5).

Despite this network of alliances, the right of collective self-defence has been little used in practice, as states tend to avoid military participation in conflicts between other states.Footnote ⁵⁷ This hesitation is perhaps indicative of some unease with the concept of collective self-defence, which moves away from the domestic law analogy and is thus less readily embraced. Only in exceptional circumstances under domestic law can a person justify the use of force in response to an attack on another person, and the recognised collective response is through law enforcement agencies. However, the concept of collective institutional action (or ‘solidarity’) has re-emerged in the context of counter-terrorism. NATO invoked article 5 of the North Atlantic TreatyFootnote ⁵⁸ for the only time in the wake of the terrorist attacks of 11 September 2011;Footnote ⁵⁹ article 222 of the Treaty on the Functioning of the European Union (2010) provides for joint action if a Member State is the object of a terrorist attack;Footnote ⁶⁰ and during the so-called ‘Arab Spring’ the Peninsula Shield force was invoked by Bahrain in response to internal riots and protests. Formal collective defence institutions have been supplemented by more informal coalitions, such as the coalition of the willing against Afghanistan in 2001Footnote ⁶¹ and the coalition of states fighting against IS since 2014. The ad hoc formation of a coalition of willing states has been described as the ‘antithesis of the maintenance of order by the international community’.Footnote ⁶² It dispenses with the need for SC authorisation while claiming the legitimacy associated with multilateralism.

There are also concerns about the potential misuse of the right to collective self-defence. A state might launch an attack against another, purporting to be acting in collective self-defence of that state against an anticipated attack by another state. The case brought by Nicaragua in 1984 against the United States in the ICJ arose out of the widespread violence in Central America in the 1980s, in the Cold War context. The United States claimed that its military actions against Nicaragua were in collective self-defence of El Salvador, which was subject to armed attack by the Sandanista (left-wing) government in Nicaragua. The ICJ took a restrictive approach, which it reaffirmed in the Oil Platforms case, and determined that there are two conditions to the exercise of the right to collective self-defence: the victim state must declare that it has been attacked and must request assistance.Footnote ⁶³ While this provides protection against wrongful claims of self-defence, especially for weaker states, it may delay assistance in an appropriate case until too late.

In 2014 the ‘newly formed’ Government of IraqFootnote ⁶⁴ sought the assistance of the United States in resisting violent armed attacks by the forces of IS.Footnote ⁶⁵ IS forces had come from SyriaFootnote ⁶⁶ and had succeeded in occupying considerable areas across Iraq and Syria, thereby effectively dismantling the border between the two states. Iraq’s request to the United States ‘to lead international efforts to strike IS sites … with our express consent’Footnote ⁶⁷ is a widely accepted example of collective self-defence, which accords with the conditions stipulated by the ICJ (even though the issue of whether self-defence can be claimed against non-state actors as discussed below has hardly been raised). At an International Conference on Peace and Security in Iraq in September 2014, participantsFootnote ⁶⁸ determined that IS constitutes a threat to the international community and underscored the need to remove it from the regions where it had established control ‘by any means necessary, including appropriate military assistance, in line with the needs expressed by the Iraqi authorities, in accordance with international law’.Footnote ⁶⁹ The SC’s welcoming of this ConferenceFootnote ⁷⁰ implicitly approved the military assistance that has been provided by a number of states to Iraqi forces. In June 2015 at a ministerial meeting a ‘small group of the coalition to counter Da’esh [IS]’ reiterated this resolve.Footnote ⁷¹

The United States (and some other states) have also carried out airstrikes in Syria, which had not requested such assistance. Use of force in Syria violates that state’s territorial integrity and is therefore contrary to UN Charter article 2 (4), unless it can be justified. In a letter to the SC, the US ambassador to the UN, Samantha Power, asserted the inherent right of the United States to both individual and collective self-defence, since ‘IS and other terrorist groups in Syria are a threat not only to Iraq, but also to many other countries, including the United States and our partners in the region and beyond’.Footnote ⁷² Neither IS nor ‘other terrorist groups’ (notably Khorasan, said by the Americans to be an ‘al-Qaeda affiliated terrorist group’) has launched an armed attack against the United States (although individual terror attacks may be claimed by IS), so the claim of individual self-defence must rest on questions of imminence, as discussed below. With respect to collective self-defence, the question is whether a request for assistance in self-defence by one state (Iraq) can justify the use of force in another state (Syria). Ambassador Power asserted that this was the case because the governing regime in Syria was ‘unwilling or unable to prevent the use of its territory for such attacks’.Footnote ⁷³

The language of ‘unwilling or unable’ is used elsewhere in international law, for instance in the Rome Statute of the International Criminal Court (ICC) to make a case admissible although it is being investigated or prosecuted by a state with jurisdiction.Footnote ⁷⁴ Although it is not self-evidently applicable to the use of force in self-defence, it appears to have been accepted by UN Secretary-General Ban Ki-moon:

Of course, the language of a ‘threat to international peace and security’ is that of SC competence to authorize force under UN Charter, chapter VII, not that for triggering self-defence under article 51. No SC action has been authorised. Marc Weller argued that it is the attack that triggers the right to self-defence against the perpetrators of the attack, but that if ‘there is no actual or imminent use of force amounting to an armed attack, the doctrine of “unwilling or unable” cannot furnish a legal justification for the use of force in self-defence’. Self-defence is applicable only where an immediate armed response is needed, which ‘implies that the territorial state would not forestall or terminate the imminent threat or attack – if it did, the criterion of necessity would not be met. Hence, there is no additional need to prove that the territorial state is unwilling or unable to act’. He concluded that ‘the infrastructure of IS in Syria is sufficiently closely intertwined with its ongoing operations in Iraq to justify extending the application of the right to self-defense to Syria’.Footnote ⁷⁶ However, this position accepts the collapse of the border between Iraq and Syria, (an illustration of the way that the ‘new wars’ paradigm muddies legal interpretation).Footnote ⁷⁷ Further, the transference of the language of ‘unwilling or unable’ to justifications for the use of force in collective self-defence ‘shields the state using force against terrorists from responsibility for forceful incursions on the territory of another state’,Footnote ⁷⁸ and this argument thus constitutes yet another reinterpretation and thus extension of the right to self-defence in the War on Terror.

Following the spate of terrorist attacks by IS, including those ‘in Sousse, on 10 October 2015, in Ankara, on 31 October 2015, over Sinaï, on 12 November 2015, in Beirut and on 13 November 2015 in Paris’,Footnote ⁷⁹ the arguments in favour of military strikes in Syria as self-defence became stronger. For instance, James Stavridis, the former commander of NATO in Europe, argued that NATO should invoke article 5;Footnote ⁸⁰ retaliatory airstrikes by France were described by the foreign minister, Laurent Fabius, as an act of ‘self-defence’;Footnote ⁸¹ David Cameron told the United Kingdom Parliament that the violence meant ‘working with our allies to strike against those who pose a direct threat to the safety of British people around the world’.Footnote ⁸² The then prime minister argued the case for airstrikes against IS in Syria to be ‘founded on the right of self-defence as it is recognised in Article 51 of the UN Charter’.Footnote ⁸³ He asserted this to be supported by UN SC Resolution 2249 of 20 November 2015, which calls upon ‘Member States that have the capacity to do so to take all necessary measures, in compliance with international law, … on the territory under the control of ISIL also known as Da’esh, in Syria and Iraq, to redouble and coordinate their efforts to prevent and suppress terrorist acts committed specifically by ISIL’. Military action was to be carried out in conjunction with political and diplomatic efforts at settlement and humanitarian assistance. While Resolution 2249 neither directly upholds self-defence nor unequivocally authorises military action, the case for self-defence has not apparently been questioned.

Whether a request for intervention has been genuinely made and, if so, whether it has emanated from a legitimate authority within the state in question must always be context-specific. Another factor is whether the alleged attack derives from another state, or from within (but often with suggestions of external support), merging claims of self-defence with those for internal self-determination or democratic choice. During the Cold War both the United States and the Soviet Union made geo-political claims of such requests from governments seeking to shore up their power against ideologically opposed challenge: for instance, with respect to the Dominican Republic (1965) and Granada (1983) in the case of the United States and Hungary (1956), Czechoslovakia (1968) and Afghanistan (1979) in the case of the Soviet Union. In similar terms, in 2014 President Putin claimed that Ukrainian President Viktor Yanukovych had requested military assistance; however, it was doubtful whether he remained in a position to give a valid request.Footnote ⁸⁴ Nor does international law allow a sub-state entity, such as Crimea, to seek external intervention. Putin’s actions with respect to Ukraine have been made subject to sanctions by the United States and EU.Footnote ⁸⁵

The legal basis for Russia’s intervention in Syria is also said to be ‘intervention by invitation’, even though the Russian government has not made a formal statement to this effect. However, two questions arise that militate the validity of this claim. First, can the government of Syria that is being supported by Russia be considered the legitimate authority; even though it is still the government, large parts of the country are no longer under its authority or control. Further, can a government that is committing gross violations of human rights against its own people request assistance from another government, even though the objective is ostensibly to defeat an extremist group, IS, in opposition to that government? Second, related and perhaps more importantly, Russia is violating the principle of distinction, an 'intransgressible principle'Footnote ⁸⁶ of International Humanitarian Law as discussed in Chapter 6, in particular, through air strikes on hospitals and other civilian targets. The commission of war crimes falls under the ius in bello, but it seems that consent to intervention must be vitiated by such acts. This raises similar issues of consent to those discussed with respect to targeted killing later in this chapter.

Despite reports of heavy civilian casualties and a humanitarian catastrophe, the military intervention by Saudi Arabia in Yemen has not been subject to sanctions by the United States or EU.Footnote ⁸⁷ Saudi Arabia’s actions are in response to a request from the deposed President Abd Rabbuh Mansur Hadi for external assistance in self-defence against internal uprising.Footnote ⁸⁸ President Hadi acceded to power following Arab Spring uprisings in 2012 and an election where he was the only candidate; his term of office was due to expire in February 2014 but had been extended for a year without further polling.Footnote ⁸⁹ His authority as the legitimate head of state and thus as able to seek assistance against internal rebellion is seen by some as debateable, although supported by the SC.Footnote ⁹⁰ Saudi Arabia’s military action appears to be in support of protecting its borders and maintaining a regional hegemony against Shi’ite rebels backed by Iran, echoing those Cold War interventions described above. It is backed in its military action by the Gulf Cooperation Council, thus giving it the veneer of collective action.

4.3.3 Imminence: Anticipatory Self-Defence

The risk of escalating violence under the cover of self-defence is theoretically limited by the legal requirements that such action be imminent, necessary and proportionate. However, determining any precise and agreed content to these conditions is problematic. The issue of imminence arises in relation to the notion of anticipatory self-defence. The wording of article 51 is clear: the right to self-defence arises ‘if an armed attack occurs’. A restrictive approach would therefore prohibit any defensive action until this has taken place. But this interpretation is ‘hotly contested by writers and in state practice’.Footnote ⁹¹ It is argued that it would punish the victim and favour the aggressor, supporting a strong argument that a state cannot be expected to wait until an anticipated attack eventuates. This position carries further weight in the context of nuclear weapons and other weapons of mass destruction (WMDs) or fears of terrorist activity. In addition to the interpretive issue is the practical question of what evidence is required to justify such a first strike and who is to make that evaluation. The state that fears that it may become subject to an attack may not be able to make an objective assessment and may act prematurely out of fear, suspicion, on the basis of unreliable intelligence or in bad faith.

In 2002 the unresolved debate with respect to anticipatory self-defence took on a further twist. In that year the US National Security Strategy addressed the threats of WMD and terrorism. It asserted that international law had long ‘recognized that nations need not suffer an attack before they can lawfully take action to defend themselves against forces that present an imminent danger of attack’, but that the concept of imminence must be adapted to address contemporary threats.Footnote ⁹² It set the stage for the 2003 invasion of Iraq by setting out a claimed right to act unilaterally ‘to exercise our right of self-defence by acting pre-emptively against such terrorists, to prevent them from doing harm’ to the United States – the people and the country.Footnote ⁹³ The core difference between this claim and that to take anticipatory action in self-defence is one of degree: what may be termed pre-emptive self-defenceFootnote ⁹⁴ applies to the military response to some potential future attack, even if the timing and place of such an attack are uncertain, rather than to one that is imminently anticipated. ‘Imminence’ in the terms of the Caroline case means ‘instant, overwhelming, leaving no choice of means and no moment for deliberation’.Footnote ⁹⁵ In her 2002 Wriston Lecture at the Manhattan Institute, Condoleeza Rice commented on the 2002 National Security Strategy’s call for a new concept of imminence: ‘some threats are so potentially catastrophic – and can arrive with so little warning, by means that are untraceable – that they cannot be contained … new technology requires new thinking about when a threat actually becomes ‘imminent’.Footnote ⁹⁶

Thus the argument for pre-emptive self-defence in the case of the 2003 invasion of Iraq is based on a subjective determination that there is a latent threat caused by the nature of the target state as a ‘rogue state’, as seeking to develop a nuclear capacity, or as possessing (and thus potentially using) WMDs.Footnote ⁹⁷ The failure to locate any WMDs in Iraq following the 2003 invasion highlighted the dangers of such claims. A broad interpretation of article 51 allowing for pre-emptive self-defence was not supported by other states, with the United Kingdom, for instance, preferring to base its claim for the legality of the invasion of Iraq on the interpretation of earlier SC resolutions.Footnote ⁹⁸ Nor has it since gained ground with states, commentators or the ICJ. In his report In Larger Freedom the UN Secretary-General considered that ‘Imminent threats are fully covered by article 51’ and that ‘[l]awyers have long recognized that this covers an imminent attack as well as one that has already happened’. Beyond that, however, only the SC has the responsibility to authorize the use of preventive force.Footnote ⁹⁹ The GA in its 2005 Millennium Outcome Document also failed to endorse any notion of pre-emptive self-defence, instead reaffirming ‘that the relevant provisions of the Charter are sufficient to address the full range of threats to international peace and security’.Footnote ¹⁰⁰ However, in rejecting the notion of pre-emptive self-defence, these statements appear to accept the less radical concept of anticipatory self-defence. The ICJ has not clarified the question. In a case brought by the Democratic Republic of the Congo against Uganda, the Court asserted that the use of force in self-defence applies only ‘within the strict confines’ of article 51, which ‘does not allow the use of force by a State to protect perceived security interests beyond these parameters’.Footnote ¹⁰¹ The Court noted that there are other means to protect security interests ‘including, in particular, recourse to the Security Council’,Footnote ¹⁰² a view that is weakened by the political and practical limitations on effective SC response as discussed in Chapter 2.

Despite this first overwhelming rejection of pre-emptive self-defence, in 2006 President George W. Bush reiterated that the doctrine of self-defence needs to be revised and rewritten, including that the requirement that a threat needs to be imminent should be revisited. While this too has not achieved widespread acceptance, a first stage in law-making is articulating what may be perceived as a problem to be addressed. In the decentralised international legal system, lacking a legislative body and compulsory adjudication, the claims of the powerful to change law in accordance with their wishes may carry significant weight, and their practice prove hard to resist. On the other hand, divisions between powerful players may inhibit any movement for legal change.Footnote ¹⁰³ As discussed above, the United States has raised individual self-defence as a basis for their airstrikes in Syria despite there having been no armed attack against it. President Obama stated that IS will ‘ultimately’ pose a threat outside the Middle East, to the United States, Europe and ‘far-flung countries like Australia’.Footnote ¹⁰⁴ Former Prime Minister Cameron made a similar assessment that IS ‘represents a major threat to us, here at home’,Footnote ¹⁰⁵ and that we should not ‘wait until an attack takes place here: we should act in advance, recognising that there are inherent risks in any course’.Footnote ¹⁰⁶ In a further extension of the concept of ‘armed attack’ the threat is also seen as coming from radicalised fighters returning home to the United Kingdom. What is factually uncertain at the time of writing is whether an armed attack is ‘imminent’ such that a right to anticipatory self-defence has arisen, or whether military action against IS in Syria represents acceptance in practice of earlier – pre-emptive – action. Or, as discussed above, there is a growing preference for the rationale that the host state is ‘unwilling or unable’ to take effective action against non-state actors on its territory.

While former President Bush may have considered that taking action in Iraq in 2003 in advance of any anticipated imminent attack would enhance security, it must be asked ‘whose security’? There is an argument for considering that it would increase the security of Israel and that of other states in the region who might have been under threat from Saddam Hussein, but it is harder to see how it enhanced the security of the United States or United Kingdom. In the event, regional security has been undermined by the violence in Iraq and the subsequent emergence of IS.Footnote ¹⁰⁷ It must also be remembered that if pre-emptive self-defence is accepted, other states (including those with nuclear capabilities) would be able to act similarly. This would generate instability and uncertainty: who would have the right to determine the existence of a potential threat? What would be the protection against exaggerated claims for pre-emptive self-defence?Footnote ¹⁰⁸

4.3.4 Necessity and Proportionality

It is equally difficult to give a precise content to the two other conditions for self-defence – the concepts of necessity and proportionality.Footnote ¹⁰⁹ Necessity presupposes that all other possibilities have been exhausted before recourse to force. The threshold is high: the International Law Commission Articles on State Responsibility allow necessity to preclude wrongfulness only if the act (a) ‘is the only way for the State to safeguard an essential interest against a grave and imminent peril and (b) does not seriously impair an essential interest of the State or States towards which the obligation exists, or of the international community as a whole’.Footnote ¹¹⁰ Article 21 of the Articles reiterates the right of self-defence,Footnote ¹¹¹ but does not elaborate any further the meaning of necessity as a condition for self-defence. It could be argued that recourse to force in the name of self-defence of itself impairs an essential interest of the international community as a whole, the right to security, an argument that has greater strength if ‘international community’Footnote ¹¹² is understood as including individuals.

Necessity also has a temporal element. States may be unwilling to undertake time-consuming steps and to wait to see if a forceful response is in fact necessary. For example, in August 1990 the SC imposed economic and other sanctions on Iraq in response to its military action against Kuwait.Footnote ¹¹³ In November 1990, before there had been time to see if they would be effective, ‘Member States cooperating with the Government of Kuwait’ were authorised by the Council ‘to use all necessary means’ (i.e., military force) in January 1991 if Iraq had not withdrawn from Kuwait before that date.Footnote ¹¹⁴ Similarly, the United States determined that Al Qaeda, which was being sheltered by the unrecognised Taliban regime in Afghanistan, was responsible for the attacks on 11 September 2001. The United States demanded that Osama Bin Laden be handed over. When this did not happen it led the coalition of the willing to take military action against Afghanistan, claiming the action to be a lawful exercise of self-defence. This action commenced on 7 October 2001 before the United States had had time to determine whether the handover of Bin Laden could be negotiated.Footnote ¹¹⁵ The military response could be understood as a reprisal (prohibited under the GA Declaration on the Principles of Friendly Relations)Footnote ¹¹⁶ or as pre-emptive action against further attacks. However, the international community widely accepted the US action against Afghanistan as legal self-defence.

Proportionality in the jus ad bellum is another slippery concept. Vaughan Lowe has asked whether ‘the degree of force [is] to be calibrated against the scale of violence that would – might – occur if the threat against which the state is defending itself were to be realised? Or is proportionality to be measured against the force necessary to prevent – or to respond – to the attack?’Footnote ¹¹⁷ In the Nicaragua case the ICJ favoured the second approach, thereby limiting forceful action in self-defence to what is needed to reply to an attack.Footnote ¹¹⁸ This is consistent with the approach of Professor Ago in the International Law Commission, who determined that the concept of proportionality ‘was that which was proportionate to repelling the attack, and not a requirement of symmetry between the mode of the initial attack and the mode of response’.Footnote ¹¹⁹

It is evident that a determination of legality would differ according to which approach to proportionality is preferred. This has particular application in the context of terrorist attacks. For example, at the outset of Operation Cast Lead – Israel’s military action against the Gaza Strip in December 2008 through January 2009 – the state’s permanent representative at the UN, Gabriela Shalev, sent identical letters to the UN Secretary-General and the President of the SC stating:

The military operation involved a week-long air attack, a ground invasion (with continuing air cover) and shelling of the coast by the Israeli navy. Statistical data of casualties differ as does determination of the proportion of civilian casualties, not least because of the problems of definition and identification of individuals as civilians.Footnote ¹²¹ In September 2009, the UN Fact-Finding Mission to Gaza (Goldstone Report) reported that NGOs estimated the overall number of Palestinian deaths as between 1,387 and 1,417; the Gaza authorities put the number at 1,444 and the Israeli Government at 1,166. There was also extensive damage to civilian and social infrastructure. The Government of Israel also stated that three Israeli civilians and one soldier were killed in southern Israel by rocket and mortar attacks launched by Palestinian armed groups, and nine Israeli soldiers were killed during the fighting inside the Gaza strip, four as a result of friendly fire.Footnote ¹²²

As is apparent from the letters submitted by the Israeli permanent representative to the UN, Israel justified this military action as legal self-defence against ongoing terrorist attacks from Gaza. The Goldstone Report found that more than 8,000 rockets and mortars had been launched from Gaza into southern Israel since April 2001, that they had caused relatively few fatalities and physical injuries among the residents of southern Israel and that property damage was also not extensive.Footnote ¹²³ As well as rockets failing to strike persons and property, casualties are reduced by the warning systems and protective bomb shelters within Israel. What, of course, must not be underestimated is the deep insecurity and fear that these attacks cause to the affected populations, as well as psychological trauma. Israel unarguably has the right to defend itself against armed attacks, but was Operation Cast Lead a proportionate exercise of this right? The Goldstone Report does not address this question as its mandate referred only to the conduct of the military operation (jus in bello and human rights), not its initial lawfulness (jus ad bellum). The answer may depend on the basis for determining proportionality; if the criterion is quantitative assessment of harm, it appears disproportionate, if, however, it is the intensity of action necessary to prevent further attacks, then it may appear proportionate.

That the latter is not necessarily borne out by the facts is suggested by the reality that the ceasefire negotiated in June 2008 was more successful in lowering the level of violence than the military action. In any event the assessment cannot be made in these simplistic and militaristic terms but should also factor in the context, including the occupation, the economic blockade of Gaza, Hamas control in Gaza and the stalled peace process.

It could also take into account the broader impact of the conflict on both populations, but ‘assessment of overall civilian harm and deaths … do[es] not form part of the proportionality jus ad bellum equation … because to a degree, these are the expected consequences of the use of force’.Footnote ¹²⁴ Certain harms are excluded from any determination of proportionality, such as ‘long term civilian casualties resulting from starvation or disease’, displacement, refugee flowsFootnote ¹²⁵ and disruption to children’s education contributing to continued poverty and unemployment. The particular impact of such harms on women ‘represents an unacknowledged gendered consequence of force’, even force in self-defence.Footnote ¹²⁶ Heathcote uses feminist analysis to argue for the reconfiguration of the requirements of proportionality and necessity to take account of women’s experiences of conflict in a way that would contribute to de-gendering article 51.Footnote ¹²⁷ She points to Judith Gardam’s suggestion that the principles of necessity and proportionality could offer real constraints to the use of force by states if they were ‘refined and developed to produce the outcomes ostensibly claimed by their presence’.Footnote ¹²⁸ These outcomes should include the impact of military force on the populations of both the targeted state and of the state purportedly acting in self-defence. Such an understanding would move the principles beyond an assessment that takes into account solely the contingencies of the conflict and the state’s military capabilities.

4.4.1 Self-Defence in Response to Attacks against Nationals

The use of force in defence of nationals is a version of both self-defence (attacks on the nationals of a state constitutes an armed attack on the state) and of humanitarian intervention, as discussed in Chapter 5. There has been a long history of powerful states intervening in weaker ones on this basis, although only on a few occasions since World War II.Footnote ¹²⁹ Incidents during the Cold War, for instance, by Belgium in the Congo (1960), the United States in Grenada (1983) and Panama (1989) and Israel at Entebbe (1976) have all been extensively discussed.Footnote ¹³⁰

The claim has become another example of the stretching of self-defence in the twenty-first century as President Putin has claimed that Russia’s actions in Georgia in 2008 and in the Crimea in 2014 were for the protection of Russian nationals against armed attacks. In both cases, Russian troops were already deployed in the area, as peacekeepers in Georgia and as military deployment in Crimea. Putin’s legal justification with respect to the Crimea was expressed in a letter to the Federal Council, the upper house of Parliament that has the constitutional right to approve deployment of Russian armed forces outside the territory of the Federation. He noted ‘the extraordinary situation in Ukraine, the threat to the lives of citizens of the Russian Federation. Our compatriots, the personnel of the military contingent of the Armed Forces of the Russian Federation deployed on the territory of Ukraine’.Footnote ¹³¹ Putin has argued that the Federal Council’s authorisation also covers the deployment of Russian troops in Eastern Ukraine, although he has said ‘I hope very much I do not have to take advantage of this right.’Footnote ¹³² Nevertheless, despite denials, the evidence that Russian soldiers and special forces are operating in the region is rather compelling.Footnote ¹³³ In explaining to the Duma Russia’s actions with respect to the Crimea, President Putin emphasised that attempts had been made in Ukraine ‘to deprive Russians of their historical memory, even of their language and to subject them to forced assimilation’ and that the rights of ethnic minorities – Russians – had been violated. He continued that when ‘residents of Crimea and Sevastopol turned to Russia for help in defending their rights and lives’, Russia could not ignore their plea: ‘we could not abandon Crimea and its residents in distress. This would have been betrayal on our part’.Footnote ¹³⁴ On 27 March 2014 the GA called upon states not to recognise any change in the status of the Autonomous Republic of Crimea and to refrain from actions or dealings that might be interpreted as such.Footnote ¹³⁵

A further twist is the process of ‘passportisation’Footnote ¹³⁶in both Georgia and Ukraine. The issuance of Russian passports designated people as Russian who were therefore to be defended by Russia.Footnote ¹³⁷ Leaving aside whether Russian nationality for these people is ‘real and effective’Footnote ¹³⁸ and even if it is accepted that an attack on nationals validates intervention for their defence,Footnote ¹³⁹ there must be evidence of such attacks and the requirement of proportionality must be fulfilled. However, there has been no credible evidence of such violence against ethnic Russians in Ukraine (including Crimea) or of any such attacks being imminent.Footnote ¹⁴⁰

4.4.2 Armed Attack by Non-State Actors

In contemporary political violence, neither the attackers nor the defenders are necessarily states. The classic case is the terrorist attacks on New York and Washington, D.C., on 11 September 2001; similarly, the ‘public brutality and indoctrination’Footnote ¹⁴¹ committed by IS in Syria and Iraq demonstrate the ongoing capabilities of non-state actors for extreme violence.

While flying civilian planes into the World Trade Center and the Pentagon indisputably constituted attacks of a high threshold of violence, they were carried out by non-state actors. Article 51 does not explicitly say that the armed attack must emanate from a state, although the UN Charter regulates relations between sovereign states and is predicated upon that basis. This starting point is also explicit in the GA’s 1974 definition of aggression, which states that ‘aggression is the use of armed force by a State’.Footnote ¹⁴² The definition of the crime of aggression by the Review Conference of the ICC similarly states that ‘an “act of aggression” means the use of armed force by a State against another State’.Footnote ¹⁴³ However, the African Union has adopted a definition that encompasses the use of armed force ‘by a State, a group of States, an organization of States or non-State actor(s) … against the sovereignty, political independence, territorial integrity and human security of the population of a State Party to this Pact’.Footnote ¹⁴⁴ The inclusion of aggression as a crime subject to the jurisdiction of the ICC constitutes the criminalisation of war.Footnote ¹⁴⁵ This approach assumes that aggression can be committed by an individual responsible for its ‘planning, preparation, initiation or execution’.Footnote ¹⁴⁶ However, the individual must be linked to the state, for it applies to a person ‘in a position effectively to exercise control over or to direct the political or military action of a State’,Footnote ¹⁴⁷ not apparently to the same acts of a non-state military or terrorist group.

The ICJ has not directly considered whether there is a right of self-defence against the acts of non-state actors, but by indicating that the inherent right of self-defence arises in the case of armed attack by one state against another state has apparently adopted a restrictive, state-based approach.Footnote ¹⁴⁸ The majority in the case of Armed Activities on the Territory of the Congo (involving claims by the DRC against Uganda) did not determine ‘whether and under what conditions contemporary international law provides for a right of self-defence against attacks by irregular forces’. Judges Simma and Kooijmans, however, directly addressed the issue; Judge Simma considered that SC Resolutions 1368 and 1373 (adopted in the aftermath of 11 September 2001) ‘cannot but be read as affirmations … that large scale attacks can qualify as “armed attacks”’ within the terms of article 51, and both he and Judge Kooijmans considered that armed attacks by non-state actors should be subject to the same tests as those committed by states.Footnote ¹⁴⁹ Judge Buergenthal in the Wall case thought likewise, noting that

The ICJ’s majority position has not been uncritically accepted, not least through concern about its consequences. For instance, the unwillingness of the majority in the ICJ to affirm a right to self-defence against an attack by non-state actors might encourage states to use non-uniformed armed rebels who are indistinguishable from the local population to carry out their illegal aggressive plans.Footnote ¹⁵¹ Jensen argues that in this way states could ‘orchestrate large scale armed violence without creating a right to self-defense for their victims and simultaneously increasing the survivability of their attackers by clothing them in the protection of civilians’.Footnote ¹⁵² However, it can be countered that states would remain responsible for any illegal act attributable to them in accordance with the International Law Commission’s Articles on State Responsibility, giving rise at least to non-forcible counter-measures in response. In the Wall case Judge Buergenthal considered that the Court had failed to address not only the question of a right to self-defence against the acts of non-state actors, but more broadly that of Israel’s right to self-defence from continuous terrorist attacks, or the extent to which the wall ‘is a necessary and proportionate response to these attacks’.Footnote ¹⁵³ Such an analysis would seek to weigh the impact of the wall on the Palestinians’ right to self-determination and human rights against Israel’s right to security, and of Israelis to human security. On the one hand, the construction of a wall is less harmful to human security than a military assault such as took place in Operation Cast Lead in 2008–2009, Operation Pillar of Defense in November 2012 or Operation Protective Edge in the summer of 2014; on the other hand, the wall has divided communities, cut Palestinians off from their sources of livelihood and limited their freedom of movement in harmful ways.

An international body that has taken a different approach from that of the ICJ is the International Fact-Finding Mission into the conflict in Georgia that was constituted by the EU after the Russian intervention in South Ossetia in August 2008. In an extremely detailed report it directly addressed the issue of non-state actors and asserted that UN Charter articles 2(4) and 51 both apply to the acts of non-state entities (in this instance South Ossetia and Abkhazia).Footnote ¹⁵⁴ The consequence of this conclusion is that the actions of such entities may both constitute a wrongful use of force or an armed attack and that they may act legally in self-defence. Consequentially other actors (in this instance Russia) may claim to be acting in collective self-defence of those entities, an expansive interpretation with the potential for escalation of violence.Footnote ¹⁵⁵

Other states as well as the United States have used force in response to armed attacks by non-state actors, including Israel in Lebanon in 2006 and Turkey’s incursions into northern Iraq in 2008 against Kurdistan Worker’s Party (PKK) bases situated there. Kress notes that the lack of international criticism of Turkey ‘cannot be ignored’.Footnote ¹⁵⁶ After careful consideration of the relevant state practice and commentaries, he concludes that despite the ICJ’s contrary position, since 11 September 2011 there exists a right of self-defence against armed attacks committed by non-state actors.Footnote ¹⁵⁷ However, the parameters of this have not been determined; any such use of force must be subject to the requirements of necessity and proportionality, which, as discussed above, are especially difficult to apply in the context of terrorism.

If self-defence against an armed attack by a non-state actor is accepted as justifiable, the next question is: who is the legitimate target of such coercive response? In Afghanistan, the military action of the coalition of the willing was not directed solely at Al Qaeda but also at the Taliban, the then (unrecognised) government in Afghanistan, justified on the grounds that it had harboured the terrorists. In international law terms, providing a safe place for terrorist activity initiated from its territory amounts to a direct internationally wrongful act, attributable to the state. But it is not clear whether there is a right to self-defence against a state simply for the harbouring of terrorists. In the Nicaragua case the ICJ required a state to exercise ‘effective control’ over non-state groups in order for their acts to be attributable to the state.Footnote ¹⁵⁸ President Bashar al-Assad in Syria cannot be said to be ‘harbouring’ IS, but nor, of course, is he in ‘effective control’ of the territories captured by their forces. He has been engaged in civil war in Syria since 2012. Nevertheless, as discussed above, the United States claims that its actions in bombing IS bases in Syria are in individual and collective self-defence that is justified by Syria’s inability or unwillingness to control IS. This claim conflates internal conflict, consent, individual and collective self-defence, and collapses state boundaries to justify widespread military action in the war on terror. Crawford has suggested that a nuanced approach would be to accept that the right to self-defence applies only with respect to the acts of states, but to seek relaxation of this ‘effective control’ test, thereby making the acts of non-state actors more readily attributable to the state.Footnote ¹⁵⁹ However, he also suggests that ‘it may be better to leave the use of force against non-state actors to the Security Council’.Footnote ¹⁶⁰ The claim of self-defence may be stronger against a ‘failed’ state that is unable or unwilling to control the acts of non-state bodies on its territory, as indicated by Judges Kooijmans and Simma in the Armed Activities case, discussed above. It is unclear whether the Taliban in 2001 were able to exercise ‘effective control’ over the use of their territory for these ends, or to respond to the United States’ demands to surrender Osama Bin Laden. Following the defeat of the Taliban, continued strikes against Al Qaeda were justified by the consent of the newly elected government and the presence of a multilateral force (ISAF) mandated by SC resolution.Footnote ¹⁶¹

4.4.3 Targeted Killings

The other aspect of the use of force in counter-terror operations is the use of ‘targeted killings’ that have come into prominence since the beginning of the twenty-first century. They can take many forms, although particular publicity has been given to killing by unmanned aerial vehicles, or drones. Philip Alston, the former special rapporteur on extrajudicial, summary or arbitrary executions, a special procedure of the UN Human Rights Council, has identified the following distinguishing features of targeted killings:Footnote ¹⁶² they involve premeditated lethal force, intentionally directed at an identified individual by a state agent or an organised armed group. There is a possibility (often a high possibility) that other non-targeted people, including undoubtedly civilians, will be also killed. Targeted killings may take place outside the targeting state’s own territory, as, for example, with the killing by Israeli agents of a Hamas leader in a hotel in Dubai, United States drone attacks in Pakistan, Yemen and Afghanistan and, of course, the killing of Osama Bin Laden in Pakistan. Unmanned aerial attacks are favoured by those states with the requisite technology: they are relatively cheap and safe from the perspective of the attacking state, certainly in comparison with interventions by ground forces. As well as the high casualty rate, living under the threat of drone attacks is debilitating, can be destructive of traditional ways of life and undermine community coherence;Footnote ¹⁶³ it can also promote radicalisation and the desire for revenge attacks.

The drone campaign greatly intensified during President Obama’s administration. Such attacks are shrouded in secrecy, giving rise to a lack of detailed knowledge about the incidence, location, targets of individual attacks and civilian deaths and casualties.Footnote ¹⁶⁴ The legality of drone attacks is contested and has been the subject of a number of UN reports from different perspectives; different conclusions are reached according to analysis under the jus in bello, human rights law or refugee law, contributing to the lack of legal certainty and human insecurity.

The use of targeted killings also raises questions under the jus ad bellum.Footnote ¹⁶⁵ In justification of targeted killing in the territory of another state, the US State Department has asserted that since the United States ‘is in an armed conflict with Al-Qaeda, as well as with the Taliban and associated forces, in response to the horrific 9/11 attacks, [… it] may use force consistent with its inherent right to self-defense’.Footnote ¹⁶⁶ As discussed above, claims of self-defence in response to terrorist acts highlight legal uncertainties with respect to when the right arises, against whom and the application of the principles of proportionality and necessity. For instance, bearing in mind that there is no international legal definition of terrorism, what acts constitute an armed attack giving rise to self-defence, and what is the threshold of violence to constitute an armed attack for the purpose of self-defence? The asserted right of self-defence in the ongoing war against Al Qaeda (and associated suspected terrorist armed groups) extends it spatially and territorially. Former Prime Minister Cameron’s use of the self-defence justification for the killing of two UK citizens, Reyaad Khan and Ruhul Amin, by drones in Syria in August 2015Footnote ¹⁶⁷ puts international law as interpreted within the framework of the War on Terror above the claims of UK criminal law – bringing the ‘outside’ ‘inside’.Footnote ¹⁶⁸ This is a one-sided application of the War on Terror since those who commit terrorist acts within the United Kingdom are rightly subjected to criminal procedures.

In a 2013 statement President Obama averred that US counter-terrorism was no longer to be defined by a ‘boundless “global war on terror,” but rather as a series of persistent, targeted efforts to dismantle specific networks of violent extremists that threaten America’.Footnote ¹⁶⁹ Although he reaffirmed the US right to self-defence in what he termed ‘a just war, – war waged proportionately, in last resort and in self-defence’, Obama also asserted that drone attacks would not be undertaken where there is the ability to capture and detain individuals: ‘our preference is always to detain, interrogate and prosecute’.Footnote ¹⁷⁰ But strikes will be undertaken against ‘terrorists who pose a continuing and imminent threat to the American people, and where there are no other governments capable of effectively addressing the threats’.Footnote ¹⁷¹ While the objective of constraining drone attacks is welcomed, the speech lacks precision and specificity as to what constitutes ‘a continuing and imminent threat’ and how the capability of other governments to address effectively threats is to be judged. Nor did President Obama indicate any deference to the requirement of reporting to the SC.

Justifications based on self-defence are not necessary where the state in which the attack takes place has consented, thus precluding the wrongfulness of the intervention.Footnote ¹⁷² This is said to be the case with respect to states such as Pakistan, Yemen and Somalia. But as with ‘invitation’ what does consent mean? There has been little debate about the validity of ‘consent’Footnote ¹⁷³ in the context of what are described as ‘weak’ or ‘failing’ states and many uncertainties. Is tacit or implied consent sufficient? What does ‘consent’ mean where states do not have effective control over all their territory? Consultation is not the same as consent; Pakistan is said to have been ‘consulted’ when a US commando operation killed Osama Bin Laden in a suburban house near Abbotsville, but, apparently, the government was not actually informed of the details for fear that the information would leak and the attack would be foiled. Even if consent has been given for the killing of an identified person on the territory of another state, that state remains bound by its own human rights obligations, unless it has formally derogated from them to the extent that it is legally able to do so.Footnote ¹⁷⁴ A state cannot abdicate its own responsibilities to those within its territory by allowing another state to attack them or otherwise violate their human rights, for instance, through wrongful arrest or rendition.Footnote ¹⁷⁵

4.4.4 Recasting Self-Defence: The ‘Bethlehem Principles’

In light of the uncertainties surrounding the scope of self-defence, in particular its applicability in response to terrorist attacks by non-state actors, the desirability of a draft ‘Declaratory Resolution on Self-defence’ has been raised.Footnote ¹⁷⁶ While no such instrument has been forthcoming, the former advisor to the UK Foreign and Commonwealth Office, Daniel Bethlehem, has put forward a set of what he has called ‘Principles Governing Self-defence against Non-State Actors’.Footnote ¹⁷⁷ They have been formulated after discussions with foreign ministries, defence ministries and legal military advisors. Drawing upon these discussions, Bethlehem maintains that his objective is to bridge the gap between academic discourse and the operational needs of states and to bring some clarity and precision to when states may use force in the territory of another state against an armed attack by a non-state entity. Among the sixteen principles articulated by Bethlehem are the following:

The Principles warrant – and have received – careful analysis. They have been subject to some considerable and wide-ranging criticism. They appear to conflate the jus ad bellum and jus in bello (when force may be used and against whom).Footnote ¹⁷⁹ In fact, they express the inter-relation between all the different ways in which the concept of self-defence has been reinterpreted and stretched since the end of the Cold War, and more particularly since the onset of the War on Terror. Throughout, Bethlehem claims that legitimate action depends upon there being a ‘reasonable and objective basis’, but there is, of course, no independent authoritative body able to make such a determination. In reality it depends upon the subjective interpretation by the state concerned, which is likely to be biased and may be misleading. For instance, before the military action against Iraq in 2003 Secretary of State Powell purported to present ‘reasonable and objective’ evidence of Iraq’s possession of WMDs, which turned out to be false. What is the basis for deciding that a state is ‘unwilling’ to curtail terrorist activities on its territory? Many of the criteria lack precision. What constitutes ‘vulnerability to future attack’? Who has the final word as to whether some other course of action is ‘reasonable and effective’? Some statements are excessively ‘terse’: ‘The term ‘armed attack’ includes both discrete attacks and a series of attacks that indicate a pattern of continuing armed activity’;Footnote ¹⁸⁰ ‘self-defense must be proportionate to the threat that is faced’. The assertion that they are without prejudice to the UN Charter presents them as within the framework of existing international law, and as such legitimate. However, the suggestion of a right to self-defence ‘in other circumstances’, including that a state’s ‘imperative interests’ may be the target of imminent or actual attack go well beyond the wording of UN Charter, article 51 –‘if an armed attack occurs’. Defining state consent as encompassing ‘strategic or operational, generic, ad hoc, express or implied’ consent expands the concept of ‘valid’ consent as understood in other areas of international law (for instance, treaty law) and as such could contribute to the fragmentation of international law.Footnote ¹⁸¹ In sum, the Principles seek to legitimate an extensive interpretation of self-defence in order to address ‘the realities of contemporary threats’. They reinforce the political stance of the United States and other Western powers; indeed, they seem to ‘uniquely benefit’ the US practice of targeted killings.Footnote ¹⁸² The Principles do not, of course, constitute international law, but once articulated such ideas tend to develop a life of their own. In terms of security, they seem likely to escalate violence by allowing for subjective determination of the imminence of an armed attack by the state using force. They assume military action and enable, rather than constrain, violence.