[Editor’s note from co-editor-in-chief Ryan Goodman: Just Security publishes, on very rare occasions, a long form analysis by a leading scholar or other expert on an important topic. We are very grateful to Professor Claus Kress for this essay on the international legal framework governing the use of force. His analysis is a true tour de force that we believe will add significantly to the work of experts, students of international law, policymakers, journalists, and others who confront these issues.]
“Who killed Article 2(4)?” This is how Professor Thomas Franck in 1970 framed the perceived erosion of the legal prohibition of the use of force especially in the context of the armed conflicts then taking place in the Middle East. Recent events in the same geographic region awaken the memory of this question. On 14 April 2018, the United States, along with France and Britain, launched an air strike against Syrian targets. It would be decidedly too complimentary to claim that the US government made a serious attempt to justify its use of force under international law in its statement in the UN Security Council. Nor do the laconic factual assertions in the letter from the Turkish government to the President of the UN Security Council regarding its military operation in Kurdish-controlled areas of Syria, which commenced in 2018, give the impression that Turkey placed any significant emphasis on the prohibition of the use of force in its deployment decision. It is deeply disturbing that so few years after Russia’s flagrant violation of the prohibition of the use of force in Crimea, two members of NATO apparently felt no real need to justify their use of force under international law. It is also shocking that on 25 March this year, the United States recognized Israel’s annexation of the Golan Heights. This move is unlawful even on the assumption that Israel’s use of force during the Six-Days-War in 1967 was justified as an exercise of the right of self-defense and it breaks with a precious legacy of US foreign policy. With all this in mind, one is tempted to say that the death knell of the prohibition of the use of force resounds even louder today than it did nearly fifty years ago, when Franck asked his alarming question.
Yet, it was and it remains wrong to declare the death of the prohibition of the use of force. While these contemporary challenges are serious, they do not represent the full picture. To this it must first be added the almost unanimous rejection of the US move regarding the Golan Heights in the Security Council and the condemnation of Russia’s use of force in Crimea as unlawful by a large majority in the UN General Assembly. The remarkable international legal diplomatic efforts of the United States after 2001 to develop and promote an international legal doctrine of individual and collective self-defense against transnational terrorist violence and the recent legal diplomacy of United Kingdom and Australia regarding their legal concerns about the justification and scope of a right to pre-emptive self-defense also testify of the continued significance States attach to the prohibition of the use of force. Last but not least, at the end of 2017 the States Parties to the Rome Statute of the International Criminal Court decided to activate the Court’s jurisdiction over the crime of aggression with effect from July 2018. Thus, for the first time in history, the international community has the opportunity to apply the sanction of criminal punishment with a certain degree of consistency in order to avert the danger of erosion of the prohibition of the use of force.
It is noteworthy, however, that the criminal sanction of State leaders for aggression is linked to the central condition that the use of force by persons in such positions manifestly violate the prohibition of the use of force. The highlighted threshold requirement suggests that the current situation of the prohibition of the use of force is not only politically sensitive but also legally complex – at any rate much more complex than it may appear from a quick reading of the relevant provisions of the UN Charter. In fact, there has been considerable controversy over the scope of the prohibition of the use of force for as long as this norm has existed. But looking back on the now almost 75 years during which, in the apt words of the International Court of Justice (ICJ), the “principle of non-use of force” has been a “cornerstone” of the international legal order, certain shifts become discernible. Some formerly central points of contention have, at the least, clearly lost significant weight. But other politically significant uncertainties about the prohibition of the use of force and its exceptions are today subjects of increased discussion.
On the Scope of the Prohibition of the Use of Force: Clarifications
Article 2(4) of the UN Charter prohibits UN Member States from using force directed against the territorial integrity or political independence of another state, or in any other manner inconsistent with the Purposes of the United Nations. For a long time, the above-cited words gave rise to the assertion that the use of force for non-aggressive purposes – such as the protection of endangered nationals abroad or ending horrendous human rights violations – remains outside the scope of the prohibition. In addition, during the Cold War, it was repeatedly argued that the agreed precondition for the prohibition of the use of force, or at least such a broad scope of the prohibition, had fallen away. For this prohibition, it was said, implicitly presupposed that the UN Security Council would fulfill its central task of acting effectively against disturbers of the peace on behalf of the international community. However, such hope was soon to be shattered because in the case of international conflict, the outcome would generally not be effective measures of collective security, but the veto of one of the two world powers in the Council’s vote. These two lines of arguments are largely a thing of the past. In accordance with the will of the creators of the UN Charter, it is rightly no longer seriously contested that the use of force in foreign territory or against otherwise protected positions of a foreign State is subject to the prohibition of the use of force, regardless of the objective pursued by the use of force. The claim that the smooth functioning of the UN Security Council to act constitutes a clausula rebus sic stantibus of the prohibition of the use of force is also rarely heard today. The absence of that idea from prevailing discourse is despite the fact that, after a comparatively short boom in the 1990s, for some time now the collective security system of the United Nations has again proven less effective.
Uncertainties
The prohibition of the use of force has however recently become the object of the debate in a number of other ways.
Cyber Operations
The recent discussion about whether harmful cyber operations may violate the prohibition of the use of force has largely come to the conclusion that a use of force under international law does not depend on conventional weapons being used. Instead of the means, the effect of the State’s action is widely considered to be at the heart of the matter. If a State’s computer operation involves the acute risk to cause damage to life, limb or property in another State with sufficient proximity, then, according to currently prevailing opinion, the threshold under international law for the use of force will be reached. The use of the computer worm Stuxnet against Iranian nuclear facilities, which is traced back to unconfirmed reports about cooperation between the US and Israel and was made public in 2010, offers a relevant example from the recent past. Dreadful scenarios of such operations against nuclear power plants or dams have so far fortunately remained the subject of theoretical analysis.
The latter also applies to such cyber operations that are capable of causing massive malfunctions in so-called critical national infrastructure – in the form of a collapse of the stock exchange system, for example. It is noteworthy here that the prognosis which has been heard on various occasions – and mainly, as it would seem, from western security circles – is that the State affected would classify such an act as a prohibited use of force and, moreover, as an armed attack justifying a use of force in self-defense. This is noteworthy because it would equate a malfunction that has serious economic consequences with physical damage to the opponent. This may claim a certain intuitive plausibility in view of the vast extent of the economic damage in question. But it would mean shaking up a delicate limit to the prohibition of the use of force that has been consistently maintained since 1945 – especially by western States against the opposing demands of developing countries: the exclusion of economic coercion from the prohibition of the use of force. On the horizon of the current discussion, the question has therefore arisen as to whether new vulnerabilities, especially in cyberspace, may lead to change in this long-known basic political pattern in the discussion about the prohibition of the use of force.
De Minimis Threshold?
Significantly greater attention than previously has been given in recent years to the question of whether the violation of the prohibition of the use of force requires a certain intensity of harmful effects. States do not consistently condemn minimally invasive violations of territorial sovereignty as a use of force. Although the Israeli secret service’s kidnapping of Eichmann in Argentina perhaps remains a quite unique case, the evacuation of a State’s own nationals from a situation of non-international armed conflict abroad remains a recurrent event in State practice. The question of whether or under what conditions such rescue operations constitute a use of force is not immaterial. For a use of force is more difficult to justify under international law than non-forcible intrusion in foreign territorial sovereignty. In particular, the possibility of invoking a state of necessity as a legal justification is excluded in the case of forcible measures, but not in the case of other measures. In an evacuation such as that of German citizens from embattled Libya in 2011 (Operation Pegasus), this subtle difference may prove crucial. The German government justified the use of an air force jet by claiming tacit consent of the Libyan government, since the Libyan government had not responded to a prior German request. This interpretation of the silence of the Libyan government, which at that time was clearly greatly occupied elsewhere, is not beyond question. The German government took refuge in this somewhat shaky justification attempt because Germany was not exposed to an armed attack that would have justified a use of force in self-defense, and it was also not clear that such an armed attack was imminent. However, in view of the rapidly escalating conflict in Libya, German nationals there were in present danger of life and limb, which, it could have been argued, created a state of necessity for Germany under international law. If the speedy evacuation of the threatened nationals had remained below the threshold for a use of force due to its minimally invasive character, the German government would have had a plausible necessity argument. This is because the evacuation, as things stood, was necessary to avert the danger to fundamental individual rights, and the interference in the territorial sovereignty of Libya carried much less weight than the danger to those rights. Nevertheless, the German government did not put forward a necessity argument – perhaps because it wished to avoid making the legal claim that the prohibition of the use of force is tied to an intensity threshold. It also bears mentioning in that context that the UK qualified the March 2018 nerve agent attack against Sergei and Yulia Skripal in Salisbury as “an unlawful use of force.” In international legal scholarship, the question of a de minimis threshold for article 2(4) of the UN Charter has recently gained increasing attention.
Use of Force in International Relations
Greater attention is now also being paid to the connection of the prohibition of the use of force with the international relations of the acting State, as expressly set out in article 2(4) of the UN Charter. The undisputed core of this norm lies in the fact that a State which maintains peace in its own territory including by use of necessary force does not use such force in its international relations. However, it is quite uncertain how far this consideration goes. There is much to suggest that it also applies if a State enforces the applicable national law in its territorial sea or in its airspace against a private ship or private aircraft of foreign nationality through forcible means as a last resort. In light of mixed State practice on this point, it is frequently discussed, however, in international legal scholarship whether international relations are engaged when a State, in order to prevent otherwise threatening damage, uses force as a last resort against an intruding foreign warship or military aircraft. In that context, it bears mention that, in June 2019, Iran invoked its right of self-defense after having shot down a US drone allegedly having entered Iran’s airspace for purposes of espionage. Iran thus appeared to accept that shooting down another State’s aircraft in one’s own airspace may amount to a use of force in the international relations to be justified accordingly. The central United States’ claim was that the drone was in international airspace, which is notably consistent with Iran’s implicit legal claim about the relevance of territorial distinctions for the use of force.
While these constellations of legal issues are concerned with the extent to which the focus on international relations limits the scope of the prohibition of the use of force, in the near future the question could also gain importance as to whether international relations may extend beyond the relationship between States. Conventionally, the prohibition of the use of force carries such an inter-State dimension. But this notion fails, for example, when Israel uses force in the Gaza Strip, and it is assumed that the latter is currently not subject to any State sovereignty. Here, the practice of States displays the inclination to go beyond a rigid inter-State perspective and to also apply the protection of the prohibition of the use of force to an area which – like the Gaza Strip – constitutes part of a territorial self-determination unit under international law. It is worth considering going even further to include in the scope of the prohibition areas which are not only State-free at the present moment, but which should remain free from State sovereignty claims permanently. Certainly, it is necessary, for example, to intensify the discussion as to whether the prohibition of the use of force limits State military activities in outer space even beyond the bounds of direct inter-State confrontation. In doing so, closer attention will have to be given to the interplay between the principle of non-use of force and the principle not to use the outer space for other than peaceful purposes.
Intervention by Invitation
Finally, it is accepted in principle that the prohibition of the use of force is not affected if a State has granted prior consent to another State for the use of force in its territory. For a long time the discussion on intervention by invitation circled around the question of whether the government’s power to request military assistance from another State conflicted with the right of self-determination as soon as the violent opposition to the government has escalated into a veritable civil war. However, the recent practice of States cannot be reduced to such simple terms. Both the French intervention in Mali in 2013 at the invitation of the government which was badly harried by the Tuareg and other armed groups, as well as the interventions begun one year later by many foreign States in Iraq to fight the so-called Islamic State at the request of the Iraqi government, took place at a time when internal violence had crossed the threshold of an armed conflict. Nevertheless, these uses of extra-territorial force met with broad international approval.
More persuasive may be the idea that a government could lose its authority to invite foreign States to use force on its territory when it has provoked the armed conflict by crimes under international law against its own population or when it has systematically committed war crimes during the armed conflict. However, the decidedly uncertain international response to the Russian intervention by invitation in Syria launched in 2015 and lasting several years shows that the international legal situation still awaits clarification. It is at least undeniable that the commission of genocide, crimes against humanity and war crimes with the direct involvement of a government does not automatically lead to the government losing its status under international law. Thus, Syria continues to be ruled by Bashar al-Assad at the time of writing these lines.
International law therefore remains strongly committed to the idea of the continuity of a still somewhat effective government. In principle, only the foreseeable permanent and complete loss of effective control over the territory in question creates the conditions for a change of recognized government under international law. In the recent past, however, the viewpoint of democratic legitimacy has sporadically – as in Haiti in 1994 and Sierra Leone in 1997 – led to the continued international recognition of a government that had been overthrown and expelled into foreign exile. A similar picture emerged in 2015 for the protection of an internationally-led transition process after the – actually quite effective – ousting of the Yemeni government from office. At the end of 2017, the international community went a step further in The Gambia and recognized the power of a democratically elected president to invite external intervention to help him achieve effective rule in place of his predecessor, who was reluctant to cede power.
The last four cases have one thing in common: the UN Security Council made clear who it considered to be the government of the State in question at a material moment in time. Thus, the Council played an indirect role in the legalization of those uses of force. By recognizing the legitimate candidate as the government under international law when there was tension between effectiveness and legitimacy, the Council afforded foreign States with the opportunity to resort to the traditional legal basis of intervention by invitation to the benefit of this candidate, without thereby taking too great an international legal risk.
On the Lawful Use of Force: Collective Security
However, according to the text of the UN Charter, the Security Council has much more than just an indirect role to play in the use of force: it has the authority to deploy UN forces in cases of a threat or breach of international peace. According to the drafters of the UN Charter, after 1945 the lawful use of force under international law were to take place first and foremost in the form of genuine world police action. About three quarters of a century after the establishment of the United Nations, one looks back with reverent wonder to the idealism of its founding days. For the idea of deploying a world police force against violators of the peace has remained a dead letter – no member State of the United Nations has placed troops directly at the disposal of the Security Council. Unfortunately, there are no signs that this will change in the near future.
When in 1950 – due to the Soviet empty chair policy in protest against the Chinese representation in the UN – a brief window of opportunity opened for the Security Council to act in response to the North Korean regime’s disruption of peace in the Korean conflict, the Council felt insecure in the wake of the new international legal order and recommended to member States forcible action in a no man’s land somewhere between collective security and collective defense. With the Soviet Union soon returning to its permanent seat on the Security Council, the Council became the captive of the East-West conflict for decades.
It was only after the end of the Cold War, when Saddam Hussein’s Iraq set about annexing the small neighboring Sheikdom of Kuwait, that the UN Security Council was offered the opportunity to bring to life the system of collective security, despite the absence of an international armed force. Finally, the Council responded by authorizing States to take necessary means to restore peace in the region and it was clear to all involved that those “necessary means” also meant the use of force. For a short time, there was a lack of clarity about the international legal classification of the UN Security Council’s approach. One part of international legal scholarship considered that the Council had referred to the right of collective defense of Kuwait against an armed attack. But soon conviction spread that it was not about collective defense, but about collective security. Faute de mieux, despite the lack of a secure textual basis in the UN Charter, the Security Council had decided to put into effect the idea of decentralized collective security. In the absence of an international armed force, States with the appropriate readiness should re-establish world peace with the authorization of the United Nations. In his State of the Union Address, George Bush senior spoke of the dawn of a New World Order. Politically, that was an understandable statement after the long decades of the Cold War. Form a strictly international legal perspective, however, the assessment should have been more sober: The Security Council’s reaction to Iraqi aggression meant no more (and in any case no less) than the late approximation of Council’s practice to the normative aspiration of collective security that had existed since 1945.
Subsequently, the practice of a decentralized collective security authorization model has not only been sustained, but has also been applied by the Security Council in cases of serious internal violence. In 1991, when the Council identified a threat to international peace in connection with Saddam Hussein’s terrorization of the Shiite and Kurdish parts of the Iraqi civilian population, it still attached great importance to identifying a connection to inter-State relations by referring to cross-border refugee flows. However, this effort was already quite strained in 1991, and subsequently the Council has repeatedly classified a human tragedy resulting from the domestic commission of crimes under international law and/or non-international armed conflicts as threats to international peace and security. This practice was formally endorsed in 2005 by the UN General Assembly when it embraced, in its World Summit Document, the idea of the subsidiary international responsibility to protect civilian populations who are victims of grave crimes under international law. While this has contributed to a debate about possible constraints on a permanent member in “vetoing” the authorization of a protective use of force by the Security Council, the idea of a duty to take forcible collective security action has not consolidated into a norm under international law. However, the Security Council’s power to authorize a use of force is no longer in question when it comes to protecting a civilian population that has been the victim of genocide, crimes against humanity or the systematic commission of war crimes in cases where the national government in question does not fulfill its responsibilities or is itself behind the attacks. Accordingly, the legality of the Security Council’s decision in 2011 to authorize the use of force to protect the civilian population in Libya met with widespread approval.
Strong objections arose only when the intervening States were later held to have abused their mandate by lending support to the violent overthrow of Gaddafi. This shows, on the one hand, that since 1990 States have managed to clarify the core of the legal framework of the lawful use of force within the context of the system of collective security established by the UN Charter. On the other hand, the Libya case confirms the experience already gained in 2003 through the use of force by a Coalition of the Willing in Iraq, that individual cases may lead to great controversy over the interpretation of the relevant Security Council resolutions. The anger of more than a few States – notably Brazil, Russia, India, China and South Africa – over what they believed to be excessive NATO action in Libya may have contributed to the Security Council’s failure in the rapidly emerging Syrian drama. Ultimately, however, the national interests that had profoundly diverged in this conflict proved to be decisive in the context of a recent declining willingness on the part of States to put the idea of collective security ahead of such interests. As a result, the practice, regrettably, has now quite clearly moved away from the normative ideal of using force, as a rule, within the framework of collective security. In the more recent State practice, the debate again primarily revolves around the justification and the limits of the lawful unilateral use of force, even if the corresponding military operations are usually carried out collectively, that is by ad hoc coalitions of States. It is all the more important to note that, in accordance with the text of the UN Charter, only the right to individual and collective self-defense is recognized as an exception to the prohibition of the use of force outside the framework of collective security. Nothing close to a noteworthy tendency in State practice to invoke a state of necessity or a right to reprisals as the legal justification for a use of force, as had happened quite often before 1945, can be identified today.
Self-Defense
For all that, the right of a State to defend itself (individual self-defense) or to defend another State at its request (collective self-defense) is – besides intervention by invitation – the focus of recent State practice. Similar to the prohibition of the use of force, a shift in the focus of the discussion is also to be observed in the area of the international law on self-defense. For decades, the central issue was whether the right of self-defense was limited to the case of an armed attack or whether certain violations of the law below that threshold may also be averted through a use of force. Whether States such as the United Kingdom, Israel and the United States – which on earlier occasions had advocated that article 51 of the UN Charter recognizes the broader version of the right of self-defense in conformity with pre-Charter customary international law – would today commit to a narrower reading remains uncertain. It is unmistakable, however, that today States generally attach greater importance to establishing a connection to an armed attack when they invoke the right of self-defense than in the first decades after 1945. This trend may certainly be seen as a consolidation of international peace and security law. But it should not be overlooked that some old controversies are now being pursued in the form of a dispute over the correct interpretation of the concept of armed attack. More specifically, there is uncertainty about the intensity threshold of a use of force, the nature of the attacker, and the object of attack. Closely related to this are controversies about when the right to self-defense begins and ends.
Intensity Threshold
The International Court of Justice did not comment in any detail on the right to self-defense until 1986, in the Nicaragua case. In that case, the Court gave the impression of seeking to impose strict limits on this right to use force through a very narrow interpretation of an armed attack. An armed attack was, it was said verbatim, the most grave form of the use of force. This sounded as if only massive military operations may be defended against by using force. This raised the question whether the Court really intended that a State must endure hostile use of force below such a high intensity threshold even where a non-forcible response is unavailable. In 2003, in the Oil Platforms case, the Court upheld the wording of the most grave form of use of force, but seriously considered the possibility that a use of force against a single warship could trigger the right to self-defense. With this clarification, the ICJ may have largely defused a self-created problem. This is worth mentioning because the relevance of this question very much persists in practice as not only in Iran’s invocation of the right of self-defense to justify the shooting down of a US drone in June 2019 demonstrates.
Transnational Non-State Armed Attack?
Far more controversial remains the question of the attacker. In the Nicaragua case, the ICJ stated – without further explanation and without the judgment resting on this point – that the right of self-defense required an armed attack by a State. According to this reading of the UN Charter, which the Court has not abandoned ever since, a State which is attacked by non-State actors from outside the State is prevented from using force against the non-State actors in their foreign host territory, even if the attack cannot otherwise be effectively defended against. Even before 1986, the legal interpretation advocated by the Court was anything but unchallenged among States. However, this fact has remained obscured because in the first decades after the entry into force of the UN Charter, the counter-position was most prominently claimed by Israel, Portugal and South Africa. However, from the outset these States had no prospect of their legal claim being internationally accepted. Israel had maintained (and continues) military occupation of areas dedicated to Palestinian self-determination, Portugal fought a frowned-upon battle for the preservation of its colonies and South Africa was an international pariah due to its racist regime. The fact that, for example, Iran and Turkey also reacted militarily to transnational non-State use of force and claimed that they were legally entitled to do so remained largely unnoticed.
Only the devastating terrorist attacks of 11 September 2001 placed the core issue squarely on the international legal agenda. Since then, it has been discussed with great intensity, as to whether the attacker must be a State in all circumstances. Immediately after the terrorist attacks, the United States started its military operation Enduring Freedom in Afghanistan. In the context of its international legal justification, the United States first of all took the position that a State itself would become the attacker if it purposefully harbored, as did Afghanistan (at that time largely ruled by the Taliban), a non-State organization (e.g., al-Qaeda) launching a cross-border attack from its territory. An essential part of the US international legal political strategy was thus to defuse the international law challenge of massive non-State transnational armed force through a generous interpretation of the concept of State armed attack. But the international legal viability of this line of justification was and is doubtful. Because in principle under international law, an apparently private act is only to be classified as a State act if the State controls the private actor. Whether a lex specialis on attribution applies to the classification of an armed attack and whether the deviation from the general rules on attribution could go as far as the United States asserted war against terror remains far from settled.
Regardless, even the US itself has not gone so far as to maintain that even a State which is unable to prevent a non-State actor from using force from its territory becomes an attacker. A few years later, such a constellation of events was to move Syria to the center stage of international law. In the territory of this State already tormented by a terrible civil war, to all misfortune the so-called Islamic State also gained a foothold, to bring (also) from there the neighboring Iraq to dire straits militarily. At no time was this so-called Islamic State a State under international law. It has rather remained a non-State transnational terrorist organization throughout its existence. However, from 2014, the US, along with a number of other States have claimed the right to defend Iraq, at the latter’s request, against cross-border attacks by the Islamic State on Syrian territory. In doing so, the Allies made clear to the UN Security Council the legal conviction that, under certain circumstances, a non-State armed attack could justify the exercise of the right of individual and collective self-defense. This legal position has not gone unchallenged among States, however. To the contrary, it would seem that those States that categorically reject the idea of a right of self-defense in case of a non-State armed attack have become increasingly more vociferous.
For UN member States, the relevant legal question is primarily a matter of treaty interpretation rather than of the identification of the applicable rule of customary international law. In fact, the wording of article 51 of the UN Charter does not require that the armed attack should be carried out by a State. However, reference to the open wording of this provision does not suffice to justify why a State (and its population) must accept forcible defensive measures by another State against non-State violent actors in its territory when it cannot effectively counteract them. In other words, the question is whether it can be said that the State suffering the non-State armed attack finds itself in a situation of self-defense vis-à-vis the host State of the non-State attackers. Fortunately, the scholarly debate increasingly focuses on this key aspect. According to one view, the liability of the host State to endure defensive action by the attacked State requires, at the very least, an internationally wrongful conduct of the host State which is absent where this State is unable to take effective action against the private attackers. The contrary view maintains that the responsibility for internationally wrongful conduct and the liability to endure defensive action follow different guiding principles: This view does not dispute the fact that a determination of responsibility for internationally wrongful conduct requires the violation of an international legal obligation, so that the host State, for those purposes, must, to the very least, have failed to exercise due diligence with respect to the non-State armed attack emanating from its territory. Yet, this view disputes that a negative finding on the host State’s responsibility for internationally wrongful conduct implies denying the same State’s liability to endure defensive action. Instead, it is argued that if, under the circumstances, only cross-border defensive action promises to be an effective remedy against the non-State armed attack, the host State, as a corollary to its claim to sovereignty over the territory concerned must bear the cost of this defensive action. Such types of defensive action may not invariably be classified as self-defense in national legal orders, but it may be characterized as a special form of self-defense under international law where it is impossible to fall back on the residual category of a state of necessity.
On this legal debate, there is no consensus in sight. After all, the intensive discussion of recent years appears to have contributed to a growing awareness that there is a difference in normative terms as to whether a State itself carries out an attack or if it simply does not prevent an attack from being carried out from its territory – and that this difference should have implications for the proportionality of defensive measures. As a rule, it would seem to be widely accepted that the use of force in self-defense may, if necessary, have a higher intensity than the armed attack it is responding to. But this principle is then in question when a State does not itself carry out the attack, but is simply unable to stop a powerful violent non-State actor operating from its territory. Such a State – and this means, above all, its population – should not in the course of forcible defensive measures suffer much more serious damage than the threatened consequences for the defending State itself – and its people – of the non-State armed attack. In a nutshell: The special character of the right of self-defense under consideration should translate into a special limitation flowing from the principle of proportionality.
Internal Armed Attack against Foreign Nationals?
In 1976, Israel used force at the airport in the Ugandan city of Entebbe to free its own nationals, who had been taken hostage and threatened with death following a plane hijacking. In this case, too, the question arose as to the nature of the attacker, for the correctness of the Israeli claim that Uganda’s government had collaborated with the hijackers remained uncertain. But above all, the Entebbe case today stands for the legal question of what the object of an armed attack must be in order to trigger the right of self-defense. Article 51 of the UN Charter speaks of the right of self-defense in the event of an armed attack against a member of the United Nations. What is highly controversial for the attacker – as just outlined – is thus not in question for the object of attack: the prerequisite is an armed attack against a State. Such a case is certainly given when the attack is directed against State territory. It is not seriously disputed that this basic case is to be equated with use of force against extra-territorial sovereign manifestations of a State, such as warships or even embassies. On the other hand, for a long time it has hardly been seriously considered whether circumstances can be imagined under which the use of force against foreign nationals assumes the quality of an armed attack against their home State. Against this is the self-evident principle that the State must be able to enforce its monopoly on the use of force in its territory also against foreign nationals. Increasingly, however, consideration is being given to whether this principle prevents speaking of an armed attack against another State, when, as in the case of a hostage-taking, an internal use of force is manifestly targeted against members of another State without any arguable link to the enforcement of national law of the State using force. The example of the Entebbe case illustrates the shift in the focus of the discussion. In its earlier argument in the UN Security Council, Israel did not emphasize the claim of being the target of an armed attack. Instead, it was said that the rescue operation had responded to the customary understanding of the right of self-defense, and that this had been unaffected by article 51 of the UN Charter. In a substantially similar case today, it may be confidently assumed, that Israel would take the position of having become the victim of an armed attack. This does not mean that the international community has come to a consensus on this point. After Russia abandoned the former Soviet position and has come to pronounce – clearly since 2008 in the Caucasus conflict with Georgia – its support for the defense of its own nationals abroad, the camp in favor of such a right has become stronger. But the Russian mass naturalization of Georgian nationals may have heightened the concern of smaller States over misuse of this legal argument. Moreover, the notion of protecting a State’s own nationals abroad for not a few States remains bound up with the degrading historical experience of western gunboat diplomacy in the colonial era – even though, fortunately, the invocation of the concept of the forcible protection of nationals abroad in order to protect mere property interests is now a thing of the past.
Lawful Preventive Self-defense?
Just as old as the dispute over the forcible protection of nationals abroad is that over preventive self-defense. In 1837, when Britain and the United States had a dispute over a British raid on the steamboat Caroline in American territory, the two counterparts agreed that it was permissible to forcibly prevent an armed attack when it was imminent. The Nuremberg military tribunal was to confirm this famous precedent of the Caroline in 1946. Almost at the same time, however, a dispute ensued about whether the UN Charter had abolished this old right of limited preventive self-defense. The opponents of this view also took the position here that article 51 of the UN Charter did not affect the pre-Charter customary right of self-defense. Meanwhile, it has become preferred instead to justify the same result by arguing that the reference to armed attack in article 51 leaves room for the inclusion of armed attacks that are imminent. However, the wording of this provision does not provide anything close to clear support for this interpretation. For this reason alone, it is not surprising that the dispute over preventive self-defense continues.
In 2002, the United States fueled this dispute by making it part of its national security strategy, in the context of both transnational terrorist threats and of the stubbornly resistant Saddam Hussein, to be able to forcibly counter the threat of use of weapons of mass destruction by non-State armed forces or by rogue States even in the case of uncertainty as to the time and place of the attack. This American legal claim had what it takes to break the international legal cornerstone of the prohibition of violence from its anchoring. For this, however, there was – and still is – no willingness on the part of the international community. Accordingly, the push by George W. Bush’s government was overwhelmingly met internationally with anxious rejection. Even the UN Secretary-General raised his voice in a warning in 2005. At the same time, however, he argued in favor of the legal view that current international law allows preventive self-defense as soon as an armed attack is imminent. The presumably resonant hope of charting the line of a sustainable compromise was not fulfilled. Instead, Kofi Annan’s remarkably extroverted statement promptly called forth a long line of opponents to any form of preventive self-defense.
There is thus no doubt that the legality of preventive self-defense remains doubtful under international law in the event of an imminent armed attack. In three similar – possibly even coordinated – advances in recent years, however, no such doubt was felt. In quick succession, the United States (in April 2016), the UK (in January 2017) and then Australia (in April 2017), each in the form of a speech by a high-ranking government lawyer, set out the firm conviction of having the right to preventive self-defense in such a case. The requirement of imminence would be – this was the consistent international law diplomatic key message – today, unlike in the classical correspondence in the Caroline case, no longer exclusively understood temporally. The message appears to be chiefly based on the consideration that a certain advance in the timing of preventive action could be inevitable in order to ensure effective defense against hard-to-pinpoint new (especially: terrorist, non-state) enemies and (perhaps: cyber) weapons. To what extent this view of preventive self-defense will be adopted, cannot yet be reliably estimated. In fact, Brazil, has pushed back against softening the temporal element of imminence, and France has most recently joined this side of the debate in favor of a restricted definition of imminence. To be sure, the recent legal claims of the United States, United Kingdom and Australia lags behind the above-mentioned passage in the national security strategy of the United States in 2002. The UK Attorney General, for example, stated, “It is absolutely not the position of the UK Government that armed force may be used to prevent a threat from materializing in the first place.” Equally undeniable is the legitimacy of the concerns underlying the legal diplomacy of the US, UK and Australia. On the other hand, it must not be ignored that any advancement in the timing of preventive action increases the risk of error or rather, erroneous forecasting. This risk can be illustrated by a historical precedent, the grave consequences of which continue to this day: Israel justified the beginning of its military operation in the Six Day War of June 1967 under international law by the decision made in the previous month by Egypt to close the Strait of Tiran to Israeli maritime traffic. However, Israel was also clearly pointing to a future armed attack by Arab States under Egyptian leadership, whose imminence was indicated by not a few signs. However, until now it is unclear whether Gamal Abdel Nasser had actually made this decision and was about to attack Israel. In the recent literature, such a decision is sometimes strongly put in doubt. In view of the present state of the discussion it can therefore not be ruled out that Israel waged the Six Day War on the erroneous assumption of having to prevent an armed attack.
When does the Right of Self-Defense End?
Only in the recent past has the situation been put into the broader perspective that, in addition to the delicate issues just outlined with regard to when the right to self-defense begins, difficult problems also arise regarding when such a right ends. The prevailing view today is that self-defense against an armed attack is allowed only until it has been halted and – as far as required – repelled. The difficulty of reconciling this standard with the practice of States is illustrated by operation Enduring Freedom. The terrorist attack of 11 September 2001 had come to its terrible conclusion that same day, and afterwards there were no attackers in America who would have to be pushed back across the border. However, in October, after the Taliban failed to respond to US requests to extradite Osama bin Laden, the country’s extensive military operation against al-Qaeda and the Taliban in Afghanistan began. The United States did not refer to an imminent further terrorist attack, but put the armed attack of 9/11 at the center of its argumentation. At the same time, however, the United States claimed to be active in preventing and deterring future attacks. The beginning of Operation Enduring Freedom met with almost unanimous approval worldwide.
There are two ways of interpreting this State practice, if one does not wish to classify Enduring Freedom in spite of the almost unanimous judgment of the international community as being illegal from the beginning: If one adheres to the formula of halting and repelling the armed attack as the sole aim of lawful self-defense, then one would have to start with the notion of armed attack itself and be prepared to pull together a chain of violence deployed from the same source and connected by sufficiently short time intervals into a continuous armed attack. Without such an extended definition of the concept of armed attack, there would be only the possibility of recognizing as the goal of self-defense the prevention of a renewed attack by the same aggressor in the near future which would otherwise be reasonably feared. The debate over the controversial demarcations in question has so far led to no clear consensus.
It is to be distinguished from the separate US legal claim that al-Qaeda’s armed attack continued beyond the destruction of this organization’s strongholds in Afghanistan in 2001 and 2002 because al-Qaeda had been joined by “associated forces.” The implied assertion that it is possible to amalgamate the use of force emanating from different terrorist cells spread over several State territories as one continuing armed attack if these cells are somehow – perhaps only by common ideology – “associated” is unconvincing and risks resulting in an endless individualization of the international legal concept of self-defense. The use of force against the so-called Islamic State in Syria is likely to give rise to a similar temptation and the international response to any such individualization remains to be seen.
Humanitarian Intervention – or Rather: Lawful Force in Defense of a Civilian Population under Attack?
The preceding overview has made it clear that the right of self-defense recognized by article 51 of the UN Charter is marked by significant grey areas. However, as already stated, it is certain that only States are holders of this right. In fact, the right of an attacked State to defend itself of course subsumes the defense of its population. But if a population becomes the target of an – in extreme cases, genocidal – attack by its own government, the attack is not directed against the respective State, which is why its right to individual self-defense does not arise. For the same reason, a foreign State cannot invoke the collective right of self-defense in order to come to the military assistance of a population in dire need under attack by its own government. Such use of force by a foreign State is subject to the prohibition on the use of force laid down in article 2(4) of the UN Charter. That’s because the almost unanimous reading of this article, as discussed above, captures uses of force for non-aggressive purposes. The interim conclusion follows that current international law grants the right of individual and collective self-defense in the case of an armed attack on a warship, but not in the case of an armed attack by a government against its own population.
Now it is also beyond dispute today – and this has already been mentioned – that the UN Security Council can authorize States within the collective security system of the UN Charter to militarily come to the aid of a civilian population against its own government in extreme cases. However, shortly after the formalization of that understanding, it has become clear in Syria that this principle cannot be relied on for the Council to ensure that the international community’s subsidiary responsibility for the protection of imperiled civilian populations is consistently exercised. Of course, it does not contradict the idea of a broadly consistent exercise of the subsidiary international responsibility to protect if the Council opposes military action because, judging from the situation, it either does not offer a serious chance of rescue or involves the risk of unjustifiable collateral damage to innocent people. However, if at least one permanent member ties the hands of the Council out of a national power calculus, a highly precarious state of tension emerges, which the International Commission on Intervention and State Sovereignty established by Canada and staffed by eminent personalities expressed in 2001 as follows:
“It is a real question in these circumstances where lies the most harm: in the damage to international order if the Security Council is bypassed or in the damage to that order if human beings are slaughtered while the Security Council stands by.”
If one seeks in the practice of States for their answer to this question, the result is a highly complex picture. However, this does not apply to the level of abstract statements. For since 1999, the Group of 77 (which has grown to 134 States) have repeatedly expressed their belief that there is no basis in international law for so-called (unilateral) humanitarian intervention. Accordingly, there is no indication in the 2005 World Summit Outcome on subsidiary international responsibility to protect that an intervention to protect a civilian population as a last resort may also be permitted outside the UN collective security framework. Nevertheless, the picture is no longer completely clear. For as always in international law, the focus must also be on how States have actually behaved in concrete incidents.
In the 1970s, the background of horrendous human rights violations in Bangladesh (East Pakistan), Cambodia (Pol Pot), and Uganda (Idi Amin), lead to the use of force by India, Vietnam, and Tanzania respectively. None of these conflicts provides a strong precedent for the emergence of an exception to the prohibition of the use of force. After all, none of the three States in their international legal justification appeared to decisively rely on the protection of the respective civilian population. Nevertheless, it is noteworthy that in the case of Tanzania’s armed intervention, in which neither of the two superpowers had skin in the game, the international response came quite a long way toward tacit approval. In 1990, the States of the West African Economic Community used force with the explicitly stated goal to protect the civilian population in Liberia, even at a time when the absence of an invitation to intervene from the government of Liberia was beyond question. This use of force was not condemned internationally. On the contrary, the President of the Security Council announced that the members of the Council welcomed the intervention. The African States have not abandoned the idea of militarily assisting a civilian population in acute need, even in the aftermath of this incident – and despite the statements of the Group of 77. This is evidenced by article 4(h), which was included in the Constitutive Act of the African Union after the genocide in Rwanda. The article provides for the possibility of an African use of force, without the consent of the territorial State at the time, to stop international crimes committed in Africa, a possibility which is “only” linked to the condition of a prior decision of the Assembly of the Union.
In 1991, for the first time after the entry into force of the UN Charter, “Western” States intervened militarily to end grave human rights violations. France, the United Kingdom, and the United States set up a no-fly zone outside of the UN collective security framework in Iraq to protect Kurds in the north and Shiites in the south from Saddam Hussein’s murderous rampage. It was the British who placed the aspect of averting an exceptionally severe humanitarian emergency at the center of their international legal justification. In this case, too, the international reaction came fairly close to tacit approval. Then, in 1999, came Kosovo. The UN Security Council had been alarmed by the threat of a humanitarian catastrophe. However, even after the failure of all attempts to reach Slobodan Milosevic by negotiation, the Council gave no green light to a military collective security measure to protect the tormented Kosovo Albanian population. In this situation, the NATO States decided to intervene without a UN mandate. While States like Germany maneuvered their justification under international law, the United Kingdom unreservedly claimed the legal right to use force as a last resort to avert an overwhelming humanitarian catastrophe. The opposition of the Group of 77 to this position at the level of abstract opinions has already been mentioned. In the UN Security Council, however, Russia’s attempt to condemn the military operation in Kosovo as a violation of the prohibition of the use of force failed by a clear 12 votes to 3. In 2013, following the repeated use of chemical weapons in Syria, the UK made it clear that it maintained its interpretation of international law. In line with this, the UK justified its involvement in the 14 April 2018 air strikes in Syria as protecting the Syrian population from further chemical attacks. The fact that the stated purpose of the allied air strikes was confined to the deterrence of the future use of one type of weapon out of a far wider weapon arsenal in use against the civilian population casts serious doubt on the precedential value of this incident in the context of defending a civilian population under attack. It is also clear that the use of force of 14 April 2018 has not widely been accepted as lawful and that it has been condemned as unlawful use of force by some States. At the same time, it is difficult to ignore the remarkable extent of support voiced internationally for the allied action, though largely couched in political terms.
Not restricted to the Assad regime’s use of chemical weapons and therefore particularly noteworthy is the position that the Arab League set out in its Doha Summit in 2013, without this having attracted much attention so far. There, the Arab States claimed the right to militarily assist the rebels of the Free Syrian Army against Assad. This, it is expressly stated in the relevant declaration, is about the defense of the Syrian people. Thus, another group of “non-Western” States has taken a position under international law in a concrete conflict that is not consistent with the rigorous rejection of a right to humanitarian intervention in the declarations of the Group of 77. At any rate, the Doha Declaration appears notable in that it substitutes the defense of a people for the concept of humanitarian intervention. For the latter immediately calls to mind the memory of the heavily burdened historical idea of a European mission civilisatrice. In contrast, the concept of defense of a population adequately expresses the fact that the population under attack must at the center of the exception to the prohibition of the use of force in question.
The outlined practice of States is multi-layered and its interpretation raises difficult questions pertaining to the way the subsequent practice to a treaty and rules of customary international law are to be identified. It is not possible to say that this practice has developed to a point where a right to collectively defend a civilian population as a last resort in the event of an attack by its own government has consolidated. But neither can such a right be denied today with the same certainty as when the UN Charter entered into force. For the present, it can be said with some certainty that the matter will remain on the international agenda as a burning issue, as governments misuse as an instrument of terror against their own population the right entrusted to them to use force, and national power calculus prevents the UN Security Council from assuming the subsidiary international responsibility to protect a civilian population in the most dire need.
Cornerstone with Grey Areas
In his famous dissent to the verdict in the Tokyo trial against the Japanese aggressors, the Indian judge Radhabinod Pal also spoke out for a political reason against the criminalization of a war of aggression under international law: such criminalization deprives peoples under colonial rule of the opportunity to shake off their yoke with external military assistance (Boister/Cryer, Documents on the Tokyo International Military Tribunal, p 910 para 239). In December 1961, India followed Pal’s words by deeds when it ended by force Portuguese colonial rule over a number of exclaves on the Indian sub-continent, including Goa. To criticisms based on the prohibition of the use of force, India proudly responded that forcibly eliminating the colonial remnants in question was a question of higher justice. Today, the question of a use of force in defense of a people under colonial rule hardly arises anymore. Instead, Pals’s old objection has come in the new guise of the emergency call for a use of force in defense of civilian populations unbearably assaulted by their governments. This question loomed large when States sought to define the crime of aggression – the legal term preferred over war of aggression today. Not only the United States has expressed the fear that with the criminalization of aggression the International Criminal Court would be subjected to the utmost tension in having to punish State leaders, because they decided as a last resort to use force to stop the commission of crimes under international law. This fear explains, to an important extent, why diplomats agreed in 2010 to furnish the definition of the crime of aggression with the additional requirement that the use of force by the State must manifestly violate the UN Charter. The other legal complexities of the current situation of the principle of non-use of force outlined in this essay explain further why the State conduct element of the crime is limited to manifest violations: The “cornerstone” of international law is as stable today as it was in 1970. But it remains surrounded by a grey area. It has a different shape today than it did in 1970, but it certainly covers no less significant terrain than it did then. And no matter how much one complains: in the short term, no substantial increase in determinacy is in sight.