Whether states can use force against terrorists based in another country is much discussed. The relevant provisions of the UN Charter do not provide a conclusive answer, but have to be interpreted. The present article suggests that in the course of the last two decades, the Charter regime has been re-adjusted, so as to permit forcible responses to terrorism under more lenient conditions. In order to illustrate developments, it juxtaposes international law as of 1989 to the present state of the law. It argues that the restrictive approach to anti-terrorist force obtaining 20 years ago has come under strain. As far as collective responses are concerned, it is no longer disputed that the Security Council could authorize the use of force against terrorists; however, it has so far refrained from doing so. More controversially, the international community during the last two decades has increasingly recognized a right of states to use unilateral force against terrorists. This new practice is justified under an expanded doctrine of self-defence. It can be explained as part of a strong international policy against terrorism and is part of an overall tendency to view exceptions to the ban on force more favourably than 20 years ago. Conversely, it has led to a normative drift affecting key limitations of the traditional doctrine of self-defence, and increases the risk of abuse.
The legal rules governing the use of force form the core of modern international law. The ban on the use of force is widely held to be peremptory in nature, and has often been described as the ‘cornerstone’ of the modern international system.1 The latter statement in particular suggests a sense of immutability: cornerstones not only cannot be removed (if the edifice resting on them is to stand), but they should better not be moved at all. In that respect, the ‘cornerstone image’, despite its popularity, may convey a wrong impression. The precise scope of the legal rules governing the use of force is by no means beyond debate. It is much discussed whether particular forms of conduct (such as severe coercion not involving military force, low-level military violence, or indirect forms of aggression) are covered by the prohibition. More importantly, there have been protracted debates about exceptions to the prohibition, that is, about justifications that should be available to states using force in order, e.g., to implement a decision by the UN Security Council, to protect human rights of their or of foreign nationals, to fight insurgents operating from foreign territory, or – admittedly one of the more ambitious claims2 – to enforce judgments of international courts.
The UN Charter (UNC) lays down the parameters of these debates. Article 2(4) UNC obliges UN members to ‘refrain in their international relations from the threat or use of force against the territorial integrity or political independence of any State, or in any other manner inconsistent with the Purposes of the United Nations’. In Articles 42, 43 and Article 51, the Charter recognizes two exceptions to this prohibition: forcible enforcement measures within the framework of the organization's collective security system, and the right of self-defence against armed attacks. These provisions lay down an ambitious regime of rules against force. In some cases, states have apparently considered the regime to be too ambitious and have deliberately stepped out of it. These are rare instances, however, and in the clear majority of cases arguments about the legality of forcible conduct are tailored to fit the Charter regime.3 Conversely, the Charter regime on the use of force, notwithstanding its fundamental importance or even its role as a cornerstone, has been anything but static. Faced with challenges such as those referred to in the preceding paragraph, the international community has not formally amended the Charter rules, but has re-appraised them through interpretation.4 In many respects, the interpretation has produced clear and stable results, but it has also led to processes of adaptation and adjustment in the light of new realities or perceptions.5
The following considerations address one particular challenge to the system, and analyse how the international community has responded to it during a particular period of time. The particular period is the years 1989–2009, coinciding with the first 20 years of the European Journal of International Law. The particular challenge is that of terrorism. It will be argued that in the course of two decades the legal rules governing the use of force have been re-adjusted so as to permit forcible responses against terrorism under more lenient conditions. To bring out those developments, the subsequent sections juxtapose the law governing anti-terrorist force as perceived in 1989 (‘The Past’)6 and the present legal regime of 2009 (‘The Present’).7 In order to put developments between 1989 and 2009 in perspective and to avoid the impression that the present state of the law marked an ‘end of history’, a final section briefly speculates how the law may develop in the two decades to come (‘The Future’).8 Before we begin this journey through time, four caveats seem in order:
(i) In line with the overall goal of the symposium, the present article focuses on the rules governing the use of force. It is not a study of terrorism, but assesses the application of the jus ad bellum to the particular problem of terrorism.9 No attempt is made to analyse in any detail what activities can be subsumed under the concept of ‘terrorism’.10 As a working definition, ‘terrorism’ shall be understood to mean an activity ‘intended to cause death or serious bodily harm to civilians or non-combatants with the purpose of intimidating a population or compelling a government or an international organization to do or abstain from doing any act.’11 While leaving open many issues, that working definition seems sufficient for present purposes to describe the type of activity against which states might claim to respond by forcible means. As will be shown, the definitional problem has not paralysed the international community. It has not stopped states from asserting a right to use force against persons or groups they claimed were ‘terrorists’, and it has not stopped others from reacting to those assertions.12
(ii) Focusing on the use of force, the following considerations present only one aspect of the fight against terrorism, and by no means the most relevant. No attempt is made to describe the ever-growing network of international obligations requiring states to, for example, criminalize terrorist activities, prosecute or extradite terrorist offenders, freeze bank accounts of terror suspects,13 or attempt to address causes of terrorism. Non-forcible responses to terror will be taken into account only in the most general way, insofar as their existence reduces the need to resort to force (or their absence increases it).
(iii) In line with the structure of legal rules governing the use of force (which, as will be explored below,14 require states to refrain from the use of force ‘in their international relations’15), the subsequent discussion deals with the extraterritorial, cross-border use of force against terrorists. As a consequence, measures addressed involve the use of force which, even though it may target terrorists, also affects another state's territorial sovereignty.16 In contrast, no attempt is made to scrutinize the legal limits restricting the right of states to employ force against terrorists based in their own territory – a question regulated largely by rules of national law, international human rights law, or possibly also humanitarian law. Furthermore, the less likely scenario of forcible measures directed against terrorists based in neutral areas outside any state's jurisdiction (the High Seas, possibly Antarctica, etc.17) will equally not be covered.
(iv) Finally, in at least one respect, the subsequent analysis will adopt a broad approach. It addresses the different aspects of the international regime governing recourse to force, beginning with the ban on force, but also covering exceptions justifying multilateral as well as unilateral forcible conduct. Of course, some issues (such as the question of self-defence against terrorist acts) will require a broader treatment than others, simply because they have proved so controversial. Yet it is believed that in order to assess the legal regime governing anti-terrorist force in a balanced way, a broader approach is required which helps place that particular controversy in perspective.
The Past: Anti-terrorist Force 1989
The Heyday of a Restrictive Analysis
In order to assess legal developments, it is necessary to revisit the starting-point. For present purposes, that starting-point is the year 1989. It is a starting-point chosen to allow reflection on development during the lifespan of the European Journal. Yet it is also a convenient starting-point because the year 1989 (or, more broadly, the period of the late 1980s) seems to have been the heyday of a particular understanding of the jus ad bellum. Of course, few aspects of the law in this area are ever universally agreed, and it would be wrong to suggest that, 20 years ago, the scope of the international rules on force had been uncontroversial. However, some of the big debates of the time – about wars of national liberation, about the Brezhnev doctrine – have lost (much of) their relevance. If these are left to one side, then, in retrospect, the late 1980s appear to have been the high point of what might be called a ‘restrictive analysis’ – an approach seeking to limit the availability of military force to the largest possible extent. This restrictive analysis was not invented in 1989 nor formulated as a coherent strategy. Yet, 20 years ago it represented the mainstream approach to the jus ad bellum which was arguably more dominant than ever before or thereafter. The mainstream approach was reflected in many of the influential writings of the time, such as the treatment of Articles 2(4), 39–43, and 51 UNC in the first editions of Charter commentaries by Cot/Pellet18 and Simma,19 in the majority of contributions to Antonio Cassese's 1986 collection of essays,20 or in the 1985 proceedings of the German Society of International Law.21 These writings could build on the gradual consolidation of a Charter regime which in 1945 had been revolutionary, but over time had been reinforced through landmark ICJ rulings (such as Corfu Channel and Nicaragua22) and important General Assembly resolutions.23 Of course states, especially the more powerful, were suspicious of the limits placed upon them by the Charter and felt that the drafters had gone too far in their quest to ban force, but ideological confrontation typically precluded agreement on a more flexible re-interpretation.24
For present purposes, the restrictive analysis is crucial not only as a general approach to the jus ad bellum but because it informed the international community's approach to anti-terrorist force. International terrorism of course was a very real threat to many states in the late 1980s. However, the international community typically approached this threat in a ‘contextual’ way, taking account of the causes of terrorism, and was unwilling to condemn it in an unequivocal way.25 Not surprisingly, it failed to adopt a comprehensive anti-terrorism convention.26 While sectoral conventions on specific types of terrorist activities were ratified,27 these tended to approach terrorism as a problem of criminal law to be addressed by means short of (international) military force. Given this lack of consensus, it comes as no surprise that the traditional jus ad bellum viewed assertions of a right to use anti-terrorist force rather sceptically. In line with the dominant, restrictive, analysis of the legal rules governing resort to force, international law as at 1989 effectively ruled out the possibility that states could lawfully resort to forcible measures against terrorists based in another country. This in turn was reflected in the prevailing interpretation of the ban on force (section B), as well as the construction of exceptions to it (sections C and D).
A Robust Interpretation of the Prohibition on Force
By the late 1980s, there had emerged a broad consensus that the prohibition against the use of force was comprehensive in scope, and that it declared every use of force in the international relations of a state to be prima facie illegal.28 As a consequence, the extraterritorial use of force, by one state, against terrorists operating within another state inevitably violated the rule.29
This interpretation of the prohibition marked a compromise between competing schools of thought: on the one hand, it maintained the ‘military’ understanding of force, resisting attempts to open Article 2(4) UNC to other forms of coercion;30 on the other, it interpreted the ban on military force robustly. This robust interpretation flowed naturally from the text of Article 2(4) UNC, but had been challenged in the UN's early days, notably by states and authors insisting that uses of force not directed against another state's territorial integrity or political independence were in line with Article 2(4).31 That argument however (which sometimes was adapted to the use of anti-terrorist force32) was difficult to bring in line with the wording and history of the provision: a cursory reading was sufficient to note that forcible conduct ‘in any other manner inconsistent with the Purposes of the United Nations’ would also be outlawed, and the travaux préparatoires clearly showed that the subsequent inclusion of other elements of the text (such as the reference to ‘political independence’ and ‘territorial integrity’) had not been intended to restrict the scope of the prohibition.33 On that basis, narrow readings of the prohibition had been convincingly dismissed by the ICJ in its Corfu Channel judgment,34 and had lost support ever since. The Court's Nicaragua judgment35 applied this robust interpretation of the Charter-based prohibition to the sphere of customary international law, which according to the Court was very similar in scope.36 It followed that states seeking to use force extraterritorially, as part of an anti-terrorist campaign, required some legal justification.
It deserves to be mentioned though that terrorists were covered by the ban on force merely indirectly. The use of force, by a state, against individuals or groups was as such not sufficient to violate the prohibition.37 That prohibition, as Article 2(4) UNC made clear, only obliged states not to use force ‘in their international relations’. As a consequence, anti-terrorist force could be used as long as it did not concern the scope of states’ international relations. That term, in turn, was read to cover inter-state relations, with cautious extensions to cover the use of force in relations between states and de facto regimes or states and arguably national liberation movements. In the present context, the use of force against terrorists based in another state clearly came within the scope of a state's ‘international relations’, but the indirect way in which Article 2(4) addressed the matter would be relevant to the discussion of exceptions.
The Security Council's Inability to Respond to Terrorist Attacks
Under the Charter system as originally envisaged, collective enforcement action under Chapter VII was to be the main exception to the prohibition on force.38 States seeking to rely on that exception in the fight against terrorism were, however, bound to be disappointed. The main reason for this was obvious. Between 1945 and the late 1980s the Security Council failed to use its authority.39 Even though its decision-making procedures had been applied with considerable flexibility,40 block confrontation paralysed the collective security system of Chapter VII during the first four decades of the UN's existence.41 In no instance had the Security Council qualified a specific act of terrorism (let alone terrorism as such) as a threat to, or breach of, the peace in the sense of Article 39 UNC; as a consequence, it did not take any enforcement action against terrorists.42
The Security Council's paralysis, however, overshadowed another problem: it was by no means clear that, even if it agreed, the Security Council would have been entitled to adopt forcible measures against terrorists. As regards the conditions of Security Council action, it remained at least doubtful whether terrorist attacks could have amounted to a threat to, or breach of, the peace in the sense of Article 39 UNC. Having become used to the Security Council's ever-broadening approaches during two sanctions decades, in retrospect one can hardly fail to be amazed by the caution with which commentators approached Articles 39–42 UNC in the late 1980s. Admittedly, the Council's discretion in interpreting the notion of ‘threat to peace’, of which SC Resolutions 217 and 221 had provided early examples, was widely stressed.43 Yet many writers seemed concerned not to make too much of these precedents. Some even went so far as to read Articles 39–42 UNC in line with the inter-state prohibition on force, requiring at least a threat of force in the sense of Article 2(4) UNC44 – which meant that terrorist acts, unless attributed to a state,45 were outside the Security Council's competence. Furthermore, it seemed clear that states, not non-state actors (such as terrorist organizations), were to be the targets of sanctions.46 Finally, with respect to sanctions involving the use of force, the relationship between Articles 42 and 43 UNC remained uncertain.47 The Charter's drafters had probably intended to couple both provisions, envisaging military enforcement action by UN forces in the sense of Article 43, not by member state forces implementing a Security Council mandate. As is well known, special agreements were not concluded. According to many, this meant that the whole system of military enforcement action under Chapter VII remained ‘inoperative’.48 In any event, commentators writing in the late 1980s were by no means agreed that the two provisions could be easily uncoupled, thus permitting the use of force by states implementing a Security Council mandate.
The Rejection of a Right to Use Anti-terrorist Force Unilaterally
Security Council enforcement action being effectively unavailable, the legal regime governing anti-terrorist force crucially depended on the scope of other exceptions permitting the unilateral use of force. Whether the ‘use of force by a State against terrorists in another country’ could ever be ‘lawful’49 was much discussed. As regards international practice, a number of incidents – among them Israel's anti-terrorist raids since the 1950s, the South African incursions into neighbouring (‘frontline’) states during the 1970s and 1980s, or the United States’ 1986 attacks on Libya – focused international attention. They also helped clarify the legal parameters of the debate in that responding states invoked different legal justifications. The following sections address their claims, distinguishing between the principal argument based on self-defence and additional justifications such as reprisals or hot pursuit.
A Narrow Construction of Self-defence
Self-defence was the principal ground on which states relied in order to justify their use of anti-terrorist force.50 The underlying argument was straightforward: setting out a broad construction of self-defence, states claimed a right to respond to attacks even if these were not carried out by another state. While these claims were made frequently, they were never received favourably by the international community. In fact, during the 1970s and 1980s, the international community rejected them almost systematically.51 To give but some examples, Israel's 1985 raid on the PLO Headquarters outside Tunis was ‘condemn[ed] vigorously’ by the Security Council, which declared it an ‘act of armed aggression … in flagrant violation of the Charter of the United Nations’ and urged other states ‘to take measures to dissuade Israel from resorting to such acts against the sovereignty and territorial integrity of all States’.52 Similarly, the United States’ 1986 raid on targets in Libya, while controversially discussed by commentators, was roundly rejected by the General Assembly as ‘a violation of the Charter of the United Nations and of international law’.53 South Africa's incursions into neighbouring states, if anything, met with stiffer resistance.54
Of course, the international reaction (both the limited support and the overwhelming condemnation) was at least partly motivated by ideological divisions – for political reasons, the international community was likely to reject each and every argument put forward by a then-pariah state such as South Africa. Crucially, however, resolutions such as GA Resolution 41/38 or SC Resolution 573 were also based on legal principle. They applied the restrictive construction of the right of self-defence prevailing at the time to the particular problem of anti-terrorist force. As far as the substantive conditions of self-defence are concerned, the restrictive construction depended on three arguments which, taken together, made self-defence effectively unavailable as a justification for forcible anti-terrorist measures.
First, self-defence against armed attacks by non-state actors was admitted in principle, but only under narrow conditions.55 For an attack to qualify as an ‘armed attack’ in the sense of Article 51 (or its customary equivalent56), the direct attack by a non-state actor had to be attributed to another state under rather stringent rules on attribution. The law on this point was shaped by the ICJ's judgment in the Nicaragua case, which concerned the relationship between a state and rebel forces, but came to define the rules on attribution generally.57 In that decision, the Court (drawing on the General Assembly's Definition of Aggression58) accepted that the jus ad bellum could be violated by ‘the sending by or on behalf of a state of armed bands, groups, irregulars or mercenaries, which carry out acts of armed force against another state’.59 Yet for the conduct of irregular forces to be attributable to a state, that state had to exercise ‘effective control [over] the military or paramilitary operations’ in question,60 whereas logistical or other support was insufficient.61 Self-defence thus depended on complex, and typically fact-dependent, questions of attribution, and required responding states to show a substantial involvement of the territorial state in the very attacks of a terrorist organization against which the response was directed (referred to as ‘effective control’ test).62 As a consequence, only terrorist attacks effectively controlled by another state triggered a right of self-defence. By adopting a restrictive approach to attribution the Court effectively restricted self-defence to the inter-state context. This approach seemed in line with an inter-state reading of the jus ad bellum, took into account the scepticism among UN members against broader readings of self-defence (which would have allowed the abuse of the concept), and for a while was hardly attacked as a matter of principle.63 It should be noted however that the Court's approach really depended on a re-reading of the text of Article 51 UNC: rather than accepting that that provision recognized the right of states to use self-defence ‘if an armed attack occurs’, the Court's Nicaragua judgment (insofar as it shaped the interpretation of that provision) effectively reformulated the provision to allow for responses ‘if an armed attack by another State occurs’.64 The implications of this re-reading would come to haunt the Court some 15 years later.
Secondly, the right of self-defence was narrowly construed in another respect as well. The Court's Nicaragua judgment confirmed that self-defence should be available only in response to grave infractions of the prohibition against the use of force. Again, just as with issues of attribution, such a threshold requirement was not easily applied and specific incidents remained controversial.65 Yet as a matter of principle, the Court's clear message that it would ‘be necessary to distinguish the most grave forms of the use of force (those constituting an armed attack) from other less grave forms’66 was clear, and came to shape the dominant understanding of self-defence as a defence against qualified uses of force.67 This narrow interpretation could draw on the differences in wording between Article 2(4) UNC on the one hand (‘any … use of force’) and Article 51 UNC on the other (‘armed attack’).68 Yet it seemed to imply that states would have to turn the other cheek, or at least forgo forcible responses, when faced with lesser breaches of Article 2(4) UNC – an apparent implication which remained controversial.69 With respect to extraterritorial anti-terrorist violence, the threshold requirement was crucial as attacks by terrorists were likely (although not necessarily) to be of lesser intensity than attacks by organized state military forces, and thus might fail to meet the required threshold.70 To avoid that conclusion, notably Israel took the view that ‘continuous pin-prick assaults’, if part of a general strategy, could be ‘apprais[ed] … in their totality as an armed attack’.71 But, by and large, the accumulation doctrine was received unfavourably and Israel's reliance on it was not accepted in discussions in the Security Council.72
Finally, under the traditional approach, the right of self-defence was limited by a ‘functional argument’:73 Article 51 and customary international law entitled states to ‘resort to force only defensively, in the presence of an armed attack and to the extent necessary to repel it’.74 While the special problem of anticipatory responses was much discussed,75 it was accepted in principle that ‘self-defence [was] not an open-ended instrument but only has the aim of repelling armed attacks and provisionally guaranteeing the security of states’.76 This functional reading implied requirements of necessity and proportionality limiting the scope of the right, but also meant that there had to be a temporal link between the measures of self-defence and the attack against which they were directed, sometimes referred to as the requirement of ‘immediacy’.77 Unless the ‘accumulation doctrine’ was accepted (which, by and large, it was not78), this meant that responses against terrorist attacks of an instant character could not easily qualify as self-defence, as ‘coming after the event and when the harm has already been inflicted’, they could not ‘be characterized as a means of protection’.79
Taken together, these factors meant that under the traditional understanding dominant in the late 1980s, states seeking to justify the extraterritorial use of force against terrorists faced almost insurmountable hurdles.
Exclusion of Other Justifications
Finally, the traditional jus ad bellum narrowed to options for states to use force against terrorist attacks because it tended to view self-defence and Security Council enforcement action as the only available exceptions to the ban on the use of force. Insofar as states invoked other grounds, they were likely to be accused of undermining the Charter regime specifically recognizing two exceptions only. The underlying approach – one might speak of the ‘exclusivity thesis’ – flowed naturally from the text and spirit of the UN Charter.80 Admittedly, it seemed more difficult to accept in practice: when looking at forcible conduct relevant in the fight against terrorism, humanitarian interventions to rescue nationals held captive abroad were the most obvious exception,81 yet the legality of such actions remained controversial, and the practice was sometimes explained as an aspect of self-defence, so as to accommodate the exclusivity thesis.82 On balance, however, the exclusivity thesis was a strong factor leading to the progressive exclusion of traditionally accepted exceptions to the ban on the use of force, and also precluded the acceptance of new exceptions.
As regards potential new exceptions invoked to justify anti-terrorist measures, South Africa's attempt to introduce an expanded doctrine of hot pursuit by land may serve to illustrate the point.83 The doctrine, in South Africa's argument, justified limited breaches of Article 2(4) UNC, namely incursions into foreign territory as part of an on-going pursuit of offenders. Hot pursuit was introduced as a novel justification permitting responses where reliance on self-defence failed, e.g., for lack of attribution. Yet, as Christine Gray noted rather benevolently, ‘this doctrine was not well-received’84 – in fact it was resoundingly rejected in SC Resolution 568 (1985), in which the Council ‘denounce[d] and reject[ed] racist South Africa's practice of “hot pursuit”’ as an attempt to ‘destabilize and terrorize Botswana and other countries’. This denunciation of course was part of a general condemnation of apartheid South Africa, yet it was based on the legal principle that the Charter's exceptions to Article 2(4) UNC were exhaustive.85
With respect to formerly accepted exceptions, the doctrine of armed reprisals (sometimes described as ‘forcible self-help’) probably has been the most prominent ‘victim’ of the exclusivity thesis.86 Under the pre-Charter rules, the right to take armed reprisals had been part of the regular rules on countermeasures – accepted in principle, subject to the requirements of necessity and proportionality, and to be distinguished from self-defence by its character as an enforcement measure aimed at inducing compliance with international law.87 Not requiring an ‘armed attack’ and not limited to defensive reactions, the concept of armed reprisal of course would have been a convenient tool in the fight against terrorism; it might have justified forcible responses to the use of force below the threshold of an armed attack, where the threat was no longer immediate, or where the real purpose of the response was to deter future attacks, or even to punish offenders.88 Convenient though it was, the doctrine plainly could not be squared with the Charter's comprehensive ban on force, and was denounced ‘on every possible occasion’.89 Actual instances of state practice (such as the British bombing of Fort Harib in Yemen in 1964) were condemned from the 1960s;90 from 1970 onwards, important documents such as the Friendly Relations Declaration91 or the Helsinki Final Act92 almost inevitably contained generalized exclusionary clauses, e.g., obliging states ‘to refrain from acts of reprisal involving the use of force’.93 It does not come as a surprise that states, even when they took anti-terrorist measures the real aim of which was to retaliate hardly ever portrayed their conduct as an ‘armed reprisal’, but instead invoked self-defence, even where that on the facts seemed less plausible.94
The Past: Taking Stock
The terrorist threat for many states was very real 20 years ago. The problem existed and was addressed by an emerging regime of international rules against terrorism. Yet within that regime the use of force against terrorism did not play a central role. Insofar as the international community engaged in the fight against terror, it largely followed a ‘criminal law strategy’,95 seeking to criminalize certain terrorist activities and to improve cooperation between states. The ‘military mode of operation’,96 approaching terrorism by cross-border military force, remained exceptional. To give but one illustration, Gilbert Guillaume's 1989 Hague Lecture on ‘Terrorisme et droit international’ addressed sectoral conventions, issues of humanitarian law, and criminal cooperation at some length, and then treated forcible responses against terror in a mere five pages.97
It could do so because, as far as forcible responses were concerned, states had agreed on a general ban, but were unwilling to apply the existing exceptions to the fight against terrorism. There was little to expect from a collective security system paralysed by block confrontation. The Security Council was incapable of addressing terrorist threats, let alone doing so by forcible means. This left the unilateral option, which some states – notably Israel, South Africa, and the United States – did assert. However, their claims, whether based on self-defence or some other legal construction, were never accepted. To the clear majority of states, 20 years ago, to admit the unilateral use of force against terrorists meant to invite abuse. The restrictive approach was easy to bring in line with the wording of the UN Charter. Influenced by experience of two world wars, the drafters had excluded ‘the unilateral use of force … as far as possible’.98 Article 2(4) UNC placed a heavy onus on states using force. Self-defence as the only relevant exception allowing for unilateral force was accepted as a temporary right aimed at repelling armed attacks. Subsequent practice and jurisprudence, true to the drafters’ intention to minimize the availability of lawful force, construed this exception narrowly, and rejected attempts to maintain or introduce unwritten exceptions.
Admittedly, in some respects, the seemingly rigid rules were handled rather flexibly. During the 1960s and 1970s, freedom from colonial domination seemed a ‘legitimate cause’, and the fight against it arguably could be pursued by means of force.99 But for obvious reasons that argument was not applied to the fight against terrorism. Divided by political ideology, the international community did not consider ‘freedom from terrorism’ a legitimate cause which would have required a flexible application of the rules against force or should be fought by military means. In some instances, it may have lacked the power actually to prevent the use of force against terrorists, but it was not prepared to give blessing to such conduct.
The Present: Anti-terrorist Force 2009
The Restrictive Analysis under Pressure
The last two decades have presented states with ample opportunity to revisit the rules governing anti-terrorist force. On the face of it, the Charter regime is the same: Articles 2(4), 42/43, and 51 UNC still apply. As often, states have shown a remarkable reluctance formally to change existing rules. In fact, even reform initiatives set up to study initiatives such as the High Level Panel on Threats, Challenges and Change deny the need for reform.100 However, the restrictive regime dominant 20 years ago has come under strain. Within the collective security system, the Security Council has reinvented itself and has used its powers creatively.101 While that development has mostly been welcomed, states and commentators increasingly question whether the drafters of the Charter were wise to exclude ‘the unilateral use of force … as far as possible’.102 Outside the anti-terrorism context, this is brought out by renewed debates about humanitarian intervention or force against pirates just as new discussions about intervention in failed states or forcible interdiction at sea: sceptics upholding the restrictive view may still draw support from the text of the Charter rules (as traditionally interpreted), but by and large seem to be on the defensive.
Legal debates about anti-terrorist force bear distinctive features, but by and large fit in with the general development. An increasing number of states considers terrorist activities to be a threat which has to be addressed through multilateral or unilateral action, including by forcible means. There is still no comprehensive anti-terrorism convention, but special sectoral treaties have mushroomed, and have been complemented by far-reaching anti-terrorism rules enacted as part of secondary United Nations law.103 These new rules are informed by a new ‘a-contextual’ approach which is willing to ignore the root causes of terrorism and denounce it as criminal irrespective of its motives.104 ‘Freedom from terrorism’ thus is increasingly regarded as a universal community value.105 Conversely, the fight against terrorism is increasingly regarded as a legitimate cause which might warrant a ‘military approach’106 and allow re-adjustments to the jus ad bellum. To analyse this development in a more nuanced way it is necessary to revisit the three different aspects of the traditional regime – the ban on force, the collective security option, and the exception(s) allowing for unilateral force.
The Robust Interpretation Affirmed
Very little needs to be said on the ban on force. Notwithstanding the introductory comments just made, in this respect there have been few developments. To begin with, the ban on force still, as 20 years ago, does not prohibit the use of force against terrorists as such, but only in international relations between states. As such it covers the extraterritorial use against terrorists based in another country, but does so only indirectly. As regards the scope of the prohibition, even though the traditional regime has come under strain, the comprehensive ban on force has not been seriously questioned. Admittedly, in heated exchanges (both within and outside the anti-terrorism context), some commentators have rehearsed old arguments about the allegedly limited scope of the prohibition,107 while others (notably in the United States) have suggested the Charter system had become obsolete.108 But these claims have been few and far between, and the argument underlying them has met with the very limited amount of attention that it deserved.109 As under the old law, the contemporary jus ad bellum of course is capable of distinguishing between grave and minor infractions of the prohibition against force. International lawyers however seem as unwilling as before to read into Article 2(4) UNC any de minimis exception. Instead, the robust interpretation continues to enjoy the support of the large majority of states and commentators.110
The Security Council's New Activism (and Its Limits)
In contrast, the collective security system has confronted the problem of terrorism; as a consequence, it is today a real possibility that states using force against terrorists should be in a position to do so with the blessing of the Security Council. This is a result of (i) the Security Council's renaissance since 1989 and (ii) its active role in the fight against terrorism. As far as the use of force against terrorists is concerned, the new potential has however not been fully used to date (iii).
(i) As is well-known, in the last two decades the Security Council has asserted its role in the international system and has adopted a very liberal interpretation of its powers under Chapter VII UNC.111 In fact, so completely has it re-invented itself that worries about the Council's paralysis have quickly given way to concerns about the limits of its powers.112 While indeed worrying in many respects, the Security Council's practice has quickly disposed of two legal problems which might in the past have limited the potential for collective action against terrorism. First, the Council has ‘uncoupled’113 Articles 42 and 43 UNC.114 It is beyond doubt today that on the basis of Article 42 UNC, the Council can authorize military measures by coalitions of the willing, and that this authorization should justify the – otherwise illegal – use of force. Secondly, no one seriously questions the Council's right to qualify as a ‘threat to peace’ situations which have nothing to do with the use of inter-State force,115 Chapter VII has thus been uncoupled from Articles 2(4) and 51 UNC. Both developments, it is submitted, are fully in line with the text and spirit of the Charter. Especially the second development has made it possible for the Security Council to take a leading role in the fight against terrorism.
(ii) A leading role it has indeed assumed. There are two aspects to this. First, practice since 1989 makes it abundantly clear that acts of terrorism can amount to threats to peace in the sense of Article 39 UNC.116 The Council has stated so time and again, both with respect to concrete instances of terrorist violence117 and in a more principled manner.118 With respect to the former, it may be noted that it has addressed acts of terrorism without an evident international element.119 As regards statements of principle, SC Resolution 1566 (2004) is particularly clear; in it the Council, acting under Chapter VII, ‘[c]ondemns in the strongest terms all acts of terrorism irrespective of their motivation, whenever and by whomsoever committed, as one of the most serious threats to peace and security’.120 Practice thus has clarified the substantive conditions for Security Council enforcement action, and leaves no doubt that the Security Council can adopt sanctions against terrorists and terrorism.
Secondly, as far as actual sanctions are concerned, the picture is more nuanced. Of course, once the Security Council has qualified an act of terrorism as a threat to the peace, the road towards sanctions under Articles 41 and 42 UNC is in principle open. Yet the Council has used these provisions rather differently. Its new activism is based on enforcement measures of a non-military character. In fact, with respect to Article 41 UNC, there is very little the Council has not done, and it may have exceeded its competences more than once in the process. To give just some examples, during the last two decades, as part of a fight against terrorism, the Council has set up a special anti-terrorism committee;121 it has ordered member states to freeze bank accounts of terror suspects,122 to prosecute specific terrorist acts,123 and to extradite terror suspects;124 and it has even assumed the role of a legislator fast-tracking the usual, and cumbersome, treaty-making process.125 With regard to non-military sanctions, the fight against terrorism indeed has become a catalyst for an ever-broader understanding of the Security Council's competences.
(iii) In contrast, the Security Council has not so far authorized the use of anti-terrorist force as a military sanction.126 While stretching the interpretation of Article 41 UNC, it has refrained from applying Article 42 UNC in the fight against terrorism. This does not mean that it has not contemplated military sanctions. In SC Resolutions 1368 and 1373, the Council expressly noted that the attacks of 9/11 had triggered a right of self-defence – an issue to be addressed below127 – but this amounted to a multilateral endorsement of a claim to use force unilaterally, rather than multilateral enforcement action in the sense of Article 42.128 The reasons for this absence of practice under Article 42 UNC remain a matter for speculation. With respect to the 9/11 attacks, there is little doubt that the Council would have authorized enforcement action had the United States wanted to adopt a ‘multilateral’ approach.129 In most other cases, military measures were not considered useful, and not contemplated by the victim state. Insofar as the Security Council has adopted general ‘law-making’ resolutions, its conduct has not concerned specific violent acts, and thus military sanctions have simply not been called for. Finally, although developments during the last two decades have removed the legal obstacles which might have previously prevented collective action against terrorists, the political hurdles for Security Council sanctions remain the same: in particular, sanctions require the support (or at least acquiescence) of all five permanent members (the P5). And while the consensus among the P5 during the last decade has been astonishing, the issue of military enforcement measures remains sensitive, and agreement cannot be taken for granted.
To sum up, unlike 20 years ago, it is beyond doubt today that the Security Council can authorize military measures against terrorists, and thereby justify the extraterritorial use of force by a state implementing that mandate. To date, however, such military enforcement action has remained a theoretical possibility. Recent practice has clarified that international law permits it, but also shows that it remains at best an exceptional option.130
A Broader Right to Use Force Unilaterally?
The key developments during the last two decades affect the rules governing the unilateral use of force against terrorists. Unlike with respect to the multilateral option, there has been a considerable body of practice – states exercising force against terrorists have, expressly or by implication, moved beyond the traditional regime. This body of recent practice needs to be briefly surveyed (1) before it can be evaluated, especially in the light of recent jurisprudence (2).
International Practice: Beyond the Traditional Approach
As noted above, ever since 1945, states have used force against terrorist threats; yet their practice for a long while was sparse, and typically critically received by the international community. The last two decades have seen a considerable shift. The number of states which claim a right to take forcible anti-terrorist measures has markedly increased, while the willingness of other states to condemn such measures has decreased. The situations in which force has been used (or a corresponding right has been asserted) vary considerably, but have almost exclusively been explained as exercises in self-defence.
(i) To begin with the most obvious piece of evidence, there was general agreement that the United States could resort to measures of self-defence in response to the 9/11 attacks.131 This it did from October 2001 by launching Operation Enduring Freedom.132 That operation has now been on-going for 7½ years and has served as a justification for forcible measures against Al-Qaeda and Taleban targets, but also includes a ‘maritime component’. While initial debates about the conditions of self-defence have ebbed away, over the years there has been growing concern that Operation Enduring Freedom overstretched the limits of self-defence.133 Still, a series of international relations continues to underline the importance of the operation as part of the international efforts to stabilize the situation in Afghanistan.134
(ii) The response against the 9/11 attacks is not an isolated incident.135 Quite to the contrary, in a variety of instances, states have reacted against terrorist attacks by using massive military force, including the invasion of states from which terrorists were operating.136
Just as in the past, Israel has remained one of the most ardent supporters of a broad right to self-defence. In 2003, in response to a suicide bombing in Haifa it bombed Palestine camps north of Damascus.137 In the summer of 2006, following rocket attacks against it by the Lebanon-based Hezbollah, Israel responded first with bombardments and then an invasion of Lebanon.138 The international community's reaction to the raids of October 2003 as well as to the July War of 2006 was mixed. There was broad agreement (with respect to both conflicts) that Israel's use of force had been disproportionate. However, a considerable number of states, especially with respect to the July 2006 war, in principle accepted Israel's right to use force against terrorist organizations such as Hamas or Hezbollah. Israel itself was at pains to attribute Hezbollah's conduct to Lebanon and Syria, but did not claim that these states had controlled and directed Hezbollah's conduct.
Furthermore, since the 1990s, Turkey has repeatedly invoked a right to use force against Kurdish PKK bases in northern Iraq.139 The 1990s saw frequent incursions. A decade later, cross-border attacks of October 2007 led to an unprecedented escalation, culminating in ‘Operation Sun’, a ground offensive during which several thousand Turkish troops invaded northern Iraq in late February 2008. The international community's reaction was characterized by a ‘mixture of sympathy and concern’140 for Turkey's conduct. Just as with respect to the July War, states stressed the need for reactions to be proportionate, and on that basis criticized the Turkish use of force. Most reactions however ‘carefully refrained from formally condemning Turkey's behaviour’.141
(iii) When looking at uses of force below the threshold of invasions proper, the number of instances in which states have used force against terrorist attacks increases considerably. Not all of them are well documented, but to support the argument made here, it may be sufficient briefly to refer to the following incidents:
In 1998, in response to attacks on US embassies in Kenya and Tanzania, the United States bombarded a pharmaceutical plant in the Sudan (allegedly used by terrorists) and a terrorist base in Afghanistan.142 To justify its conduct, the United States referred to Article 51 UNC but did not allege any substantial involvement of Afghanistan and/or Sudan in the activities. The international community's reaction was mixed, ranging from condemnation (especially of the attacks on Sudan) to open or tacit approval. Similarly (though involving an instance of alleged ‘state terrorism’), the United States had fired missiles on the headquarters of the Iraqi Intelligence Service in Baghdad in 1993, in response to an alleged assassination attempt on President Bush.143
From the mid-1990s, Iran on several occasions invoked Article 51 UNC to justify the use of force against bases of the Mujahedin-e Khalq Organization (MKO) on Iraqi territory.144 While Iraq denounced the use of force as an act of aggression, the international community did not condemn it. Equally ‘uncommented’145 remained Iran's incursions into Iraqi territory in pursuit of Kurdish armed bands (labelled ‘organized terrorist mercenaries’). There was little evidence suggesting that the conduct of the MKO (let alone that of Kurdish insurgents) could have been attributed to Iraq under the traditional ‘direction and control’ test.
In 2000 and again in 2004, Russia asserted a right to respond extraterritorially to Islamic terrorists.146 In 2007, following attacks by Chechen rebels, it conducted air strikes against Chechen bases in the Pankisi Gorge in Georgia, claiming that Georgia ‘had been unable to establish a security zone in the area of the [Russian–Georgian] border, continues to ignore Security Council Resolution 1373 and does not put an end to the bandit sorties and attacks on adjoining areas of Russia’.147 Responses were mixed, but again there was no principled condemnation that would have denied Russia's right to use force extraterritorially.
In March 2008, Colombian forces moved into Ecuadorian territory in pursuit of rebels belonging to FARC (which it considers a terrorist organization).148 The OAS qualified the operation as a ‘violation of [Ecuador’s] sovereignty’;149 other international organizations were largely silent; the United States expressed support.
(iv) The examples mentioned so far involve the actual use of force by states. The new trend they reflect is confirmed by statements. Russia's assertion of a broad right to use force extraterritorially has been referred to already.150 Along similar lines, Australia claimed a right to use force extraterritorially against terrorists threatening to attack Australia or its citizens following the Bali bombings of October 2002.151 As for more principled statements, the 2005 African Union Non-Aggression and Common Defence Pact expressly qualifies the harbouring of terrorists, as well as any provision of support for them, as an act of aggression.152 Finally, the United States’ 2002 National Security Strategy went well beyond these claims; it famously asserted a right of pre-emptive self-defence against non-imminent threats, in particular those by terrorist organizations.153
(v) Finally, these instances seem part of a broader trend among states to exercise force against attacks by non-state actors – attacks which would have been difficult to justify under the traditional approach. While not specifically relying on a right to use force against terrorist attacks, states like Rwanda, Tajikistan, or Burma/Myanmar have all responded to cross-border attacks by insurgents or rebels. For that purpose, they have all moved troops into neighbouring states even though these, under the traditional rules of attribution, could hardly be said to have directed or controlled the insurgents.154
The brief summaries provided in the previous paragraphs of course cannot replace a detailed assessment, but clearly point in one direction: the international community today is much less likely to deny, as a matter of principle, that states can invoke self-defence against terrorist attacks not imputable to another state. Instead debate has shifted towards issues of necessity and proportionality (i.e. the scope of self-defence measures). This is particularly clear in the international community's responses to Israel's repeated claims to use self-defence, in particular the July War of 2006.155 The vigorous and principled condemnation of the 1970s and 1980s has been replaced by concerns that Israel's actions should remain proportionate (which often they have not been). The traditional approach seeking to minimize the availability of lawful force in that respect has come under pressure – as Tom Franck noted already in 2002, assertions of a right to exercise self-defence against terrorist and other non-state attacks ‘are no longer exceptional claims’.156
Assessing Recent Practice
While Franck's observation is shared by many commentators, it is much more difficult to assess how recent practice can be fitted into the traditional legal regime. This is mainly due to two factors. First, the more recent practice affects many central features of the traditional, restrictive approach. Secondly, the new trends emerging in practice are clearly discernable, but may require further consolidation before bringing about a readjustment of the law. This is in particular because recent rulings by the international courts have addressed many aspects of the jus ad bellum and need to be taken into account. The current law thus is in many respects in a state of flux. That said, in at least one respect states and courts have been clear: they have treated the new practice under the rubric of self-defence, and have not ‘invented’ new exceptions to the use of force. This aspect needs to be addressed before the challenges to the traditional understanding of self-defence can be evaluated.
The ‘Ritual Incantation’157 of Self-Defence
The new practice summarized above is marked by its diversity. States have used anti-terrorist force in very different situations, ranging from ‘on the spot’ reactions to cross-border violence to long-term campaigns with broadly defined objectives (notably Operation Enduring Freedom). In addition to repelling attacks, their use of force has typically served non-defensive purposes, notably as a means of retaliation (e.g., in the United States’ raid on Baghdad) or as a means of enforcing international rules against terrorism (e.g., Russia's attacks on Georgian territory). Given this diversity, it is interesting to note that, almost inevitably, states seeking to justify their conduct have invoked the right of self-defence. In contrast, they have not re-opened debates about the permissibility of armed reprisals, even where their actions seemed to follow the logic of retaliation.158 What is more, although there has been much talk about the ‘enforcement paradigm’,159 states have not asserted a right to enforce international law against terrorists as a self-standing exception to the ban on force, but considered it as a sub-set of self-defence.160 The diversity of new, anti-terrorist practice thus stands in stark contrast to the almost monotonous assertions of self-defence. Anti-terrorist practice thus supports Christine Gray's more general observation that references to self-defence today may almost amount to the ‘ritual incantation of a magic formula’.161
On the face of it, the ‘ritual incantation’ of self-defence may be comforting. For present purposes, it narrows down the field of inquiry considerably. If states do not invoke other grounds (even though they may better reflect their actual conduct), they do not seem to consider that these other grounds would afford justification. To take but one example, if states resort to what may look like reprisals, but then do all they can to explain their conduct as an act of self-defence, they – quite correctly162 – do not seem to think that a plausible case for the legality of armed reprisals can be made out.163 In that respect, it may be said that, notwithstanding the new practice, at some level the traditional jus ad bellum continues to function. States do not publicly question the main pillars on which it rests. Discussion still takes place within the parameters of the traditional system; that system has not lost its capacity to channel debates.
On the other hand, while the pillars of the system still stand, they may be eroding rather fast. Invocations of Article 51 UNC (or even ‘ritual incantations’) may indicate compliance with the system's logic, but could also be signs of its degeneration. The point was very clearly made by Daniel Bethlehem, who noted ‘[t]he reliance by States on self-defence in virtually every conceivable circumstance’, which in turn had led ‘to normative drift, as attempts have been made to stretch the concept’.164 As will be shown, this ‘normative drift’ indeed affects key features of the regime of self-defence. The subsequent sections address three aspects of the traditional regime which have come under particular strain, and in so doing seek to give a balanced account of the present state of the law governing anti-terrorist force.
What Remains of the Strict Rules on Attribution?
Much of the discussion so far has centred on the inter-state reading of self-defence. Given the number of responses to terrorist attacks, many commentators have queried whether the ‘effective control’ test of attribution was still valid. The issue has prompted rather heated exchanges among commentators,165 especially after the International Court of Justice had seized the opportunity to address the matter. In retrospect, some claims made during these debates appear exaggerated. Especially in the immediate aftermath of the 9/11 attacks, many commentators proclaimed a radical re-reading of Article 51 UNC, which completely dispensed with the need for attribution as set out in the Nicaragua judgment.166 Following this new approach, states could resort to self-defence against all types of armed attacks, irrespective of any state involvement. Predictably, this radical re-reading has prompted much discussion, not least by the International Court of Justice.
The ICJ's New Uncertainty
Faced with these new challenges, the International Court of Justice initially seemed willing to defend its traditional approach. In the Israeli Wall case, in a ‘telegraphic’167 statement ‘startling in its brevity’,168 it observed that ‘Article 51 of the Charter … recognizes the existence of an inherent right of self-defence in the case of armed attack by one State against another State’, and hence did not justify Israeli measures aimed at preventing attacks by terrorists operating from within the occupied territories.169 One year later, the Court's majority was far more equivocal. It rejected Uganda's reliance on self-defence as a response to armed attacks by a rebel movement operating from within the Democratic Republic of the Congo (DRC), since these could not be attributed to the DRC.170 In a curious turn of argument, the Court then however expressly left open the question ‘whether and under what conditions contemporary international law provides for a right of self-defence against large-scale attacks by irregular forces’.171 Both pronouncements were criticized by individual judges. Predictably, a number of judges were not convinced by the majority's (lack of) reasoning in the Israeli Wall opinion and drew attention to the more recent practice. More importantly, Judges Buergenthal, Kooijmans, and Simma expressly accepted that self-defence was available against armed attacks ‘even if [these attacks] cannot be attributed to the territorial State’.172
Towards a More Lenient Standard of Attribution
Given the Court's new uncertainty and the amount of new state practice, the better view indeed is that the traditional rules have been modified. This in fact seems to have become the prevailing understanding.173 There is, however, much discussion about the way the new developments can be accommodated. In this respect, the academic debate, suggesting a clear distinction between state attacks and non-state attacks, may have been based on false premises. As noted above, the Nicaragua case did not rule out that attacks by non-state actors could be promoted to the level of state attacks, if only the state was sufficiently involved in them; as a consequence, it may have been more flexible than defenders of the traditional approach acknowledged.174 Conversely, the radical re-reading, by focusing on non-state attacks as such, seemed to take self-defence out of its regulatory context and fails to recognize that it serves as an exception to the comprehensive ban on the inter-state use of force.175 The radical re-reading of self-defence may have provided a justification for the attack on terrorists, but could not explain why states were entitled to violate the territorial sovereignty of the state in which they were based.176
In between strict adherence to tradition and radical departure, the more convincing way to accommodate the new practice is to opt for an approach which retains the traditional understanding of self-defence as a justification for the use of force between states, but recognizes the existence of special rules on attribution of terrorist activities.177 This more moderate (but still important) re-reading indeed seems to be borne out by state practice: states invoking self-defence do make an effort to identify links between the territorial state and the terrorist organization in question.178 What they no longer seem to do is to identify links that are strong enough to amount to ‘effective control’ as required by the Nicaragua test. Instead, contemporary practice suggests that a territorial state has to accept anti-terrorist measures of self-defence directed against its territory where it is responsible for complicity in the activities of terrorists based on its territory – either because of its support below the level of direction and control or because it has provided a safe haven for terrorists. In short, pursuant to this more moderate re-reading, modern practice points towards a special standard of imputability in relations between terrorist groups and host states, arguably most closely resembling international rules against ‘aiding and abetting’ illegal conduct.179 The contours of this new test have yet to be firmly established, and it lies in the nature of things that the broadening of attribution standards increases the risk of abuse. However, it is submitted that the concept of ‘aiding and abetting’ in terrorist activities captures the essence of new practice while still maintaining some degree of predictability. Notably, it broadens the forms of support which trigger a territorial state's responsibility, but does not lose sight of its intention. At the same time, the test seems flexible enough to accommodate issues such as the international condemnation of a state's conduct (e.g. through Security Council resolutions calling on a state to repress terrorists operating on its territory), but also clarifies that where a state is unaware of terrorist conduct it will not be exposed to forcible responses.180
The gradual recognition of this new standard of attribution indeed marks a departure from the restrictive test enunciated in Nicaragua. Curiously, this departure comes at a time when the Nicaragua test has matured into a general (residual) rule of attribution affirmed in the ILC's work on state responsibility181 and confirmed in the Court's jurisprudence.182 However, the novelty of the approach should not be overstated. Rules on attribution are not set in stone; Article 51 UNC does not lay down a particular approach. The traditional approach requiring ‘effective state control’ may have become accepted over time, but it was a standard developed by the Court, not God-given.183 As a consequence, a move towards a more lenient standard of attribution should not be seen as revolutionary (let alone as blasphemy), but as a process of reform. It brings the new law in line with views expressed in Judge Jennings’ dissent, notably his plea for more flexible standards of attribution,184 and in that respect may have been another illustration of minority views slowly gaining ground. What is more, the more lenient standard of attribution can be said to reflect the growing determination of the international community's fight against terrorism.185 Finally, by employing the notion of complicity, it builds on a form of involvement which, pursuant to Article 16 of the ILC's Articles and the ICJ's jurisprudence, qualifies as wrongful,186 and which the African Union's Non-Aggression and Common Defence Pact describes as an aggression.187
None of this suggests that the move towards a more lenient rule of attribution is a beneficial development. However, it is a move which the international community seems willing to accept and which should be seen as a process of reform rather than a revolution.
What Remains of the Threshold Requirement?
The second element of the traditional understanding – the threshold requirement – has received less attention. Still, the brief survey of practice suggests that states have claimed a right to respond to breaches which, in themselves, may not have qualified as a ‘most grave for[m] of the use of force’. By way of example suffice it to mention the responses, by Israel and Turkey, which were prompted by cross-border attacks below the threshold of an armed attack.188 These developments indicate that the threshold requirement distinguishing armed attacks from ‘lesser sins’ may need to be re-visited. In this respect, two observations are in order.
Affirmation in Principle
While state practice suggests a more lenient approach, the distinction as such has been defended rather vigorously. In fact, the recent jurisprudence of international courts and tribunals if anything affirms it. While the ICJ seems to have second thoughts about the ‘state attack requirement’, it has reaffirmed the threshold requirement on various occasions. In the Oil Platforms case, it expressly affirmed the distinction between ‘most grave’ and ‘less grave forms’ of the use of force in the context of inter-state conflicts.189 In , insofar as it left open whether states could respond to ‘attacks by irregular forces’,190 it contemplated self-defence only if directed against ‘large scale attacks’.191 When looking beyond the ICJ, the threshold requirement was equally decisive for the award of the Eritrea/Ethiopia Boundary Commission, whose decision hinged on the distinction between ‘geographically limited clashes’ and armed attacks triggering a right of self-defence.192 Recent jurisprudence thus suggests that the distinction between armed attacks and more limited uses of force is still very relevant.193
A More Flexible Application (and Its Risks)
On closer scrutiny, there may have nevertheless been some development. When applying the threshold requirement, states seem to have shown a new willingness to accept the ‘accumulation of events’ doctrine which previously had received little support.194 This doctrine of course affects many aspects of the law on self-defence.195 With respect to the threshold requirement, it must be assessed whether a series of minor incidents, taken together, can be said to reach the threshold of an armed attack. In reacting to Israel's and Turkey's practice, few states have expressly endorsed this doctrine. By implication, the large number of states accepting Turkey's and Israel's claim to self-defence however seemed to accept it, at least in situations involving constant terrorist attacks which are part of a deliberate policy of violence.196 As regards recent jurisprudence, the doctrine was much discussed by the litigants in the Cameroon–Nigeria, DRC–Uganda, and Oil Platforms cases. The Court did not expressly pronounce on the matter, but equally seemed inclined to accept it – hence its statement, in Oil Platforms, that ‘even taken cumulatively’ a series of incidents did not qualify as an armed attack on the United States.197
These statements suggest a trend towards the recognition of the ‘accumulation doctrine’, but may require further consolidation. Clearly, the doctrine appeals to those who have long criticized the gap between Articles 2(4) and 51 UNC, implying that states had to accept low-level uses of force.198 By recognizing the possibility of accumulation, the international community might close this gap. However, recognition of the accumulation doctrine is not the only way to achieve that goal. In his separate opinion in the Oil Platforms case, Judge Simma favoured an alternative approach, admitting ‘proportionate defensive [forcible] measures’ against uses of force not qualifying as an armed attack; this in turn may have been one factor leading him to reject the ‘accumulation doctrine’.199
Both arguments suggest that while the threshold requirement is maintained as such, it is increasingly being re-interpreted – either by admitting the possibility of accumulation or by recognizing a right of low-level counterforce. Both approaches indicate increased opposition to the narrow construction of self-defence set out in Nicaragua and to the gap resulting from it. On policy grounds, that gap indeed seems difficult to defend, and attempts to close it should be viewed favourably. In terms of the law, the argument in favour of ‘proportionate defensive measures’ is difficult to square with the comprehensive ban on force and requires the abandonment of the ‘exclusivity thesis’. In contrast, to re-interpret the threshold requirement in the light of the accumulation doctrine may present the more feasible approach. However, as will be shown in the next section, the ‘accumulation doctrine’, while closing the gap between Articles 2(4) and 51 UNC (and doing so in an elegant way), produces serious side-effects: it undermines the temporal dimension of self-defence and risks turning a temporal right into an open-ended licence to use force. For that reason, it can only be hoped that the international community will not embrace it and, if anything, accept a more limited right to use low-level counterforce against cross-border violence.
The Scope of the Right
Debates about the re-interpretation of the ‘armed attack’ requirement have occupied most of the academic discussion. However, the real ‘normative drift’200 discernable in recent practice on self-defence concerns not the conditions under which states can invoke self-defence, but the scope of the right. In particular, recent practice seems to have largely abandoned the functional understanding of self-defence as a protective means of ‘repelling armed attacks’.201 This in turn raises doubts not only about the temporal limits of self-defence, but also about the inherently defensive character of the right.
The Temporal Limitation
The temporal limitation of self-defence has come under pressure from different directions. States’ assertions of a right of pre-emptive self-defence (e.g., in the United States’ 2002 National Security Strategy) present the most obvious challenge. That challenge has led to a more flexible handling of the immediacy criterion, but in its radical form seems to have been resisted. By and large, few states were willing to accept the United States’ assertion of a right of pre-emptive self-defence.202 The United Kingdom's Attorney General, not otherwise known principally to oppose the United States’ position, was adamant that ‘international law permits the use of force in self-defence against an imminent attack but does not authorise the use of force to mount a pre-emptive attack against a threat that is more remote’.203 In his report In Larger Freedom, Kofi Annan equally made it clear that Article 51 UNC ‘covers an imminent attack’, but ‘[w]here threats are not imminent but latent, the Charter gives full authority to the Security Council to use military force, including preventively, to preserve international peace and security’.204 Finally, the ICJ's judgment in DRC–Uganda points in the same direction: in response to Uganda's assertion of a right to defend security interests (i.e., interests not under an imminent threat), the Court observed: ‘Article 51 … does not allow the use of force by a State to protect perceived security interests …. Other means are available to a concerned State, including, in particular, recourse to the Security Council.’205 These statements suggest that the doctrine of pre-emptive self-defence has been rejected, and with it the boldest attempt so far to turn Article 51 UNC into a carte blanche for forcible intervention.
This rejection may however have come at a price. While rejecting pre-emptive self-defence, the international community – almost as part of a ‘bargain’ – seems to move towards accepting the doctrine of anticipatory self-defence. The statements just cited206 are evidence of a more flexible approach. Admittedly, the matter is far from settled,207 but when looked at from a distance the price to pay seems small, and a re-adjustment of the rules would not necessarily conflict with the wording of Article 51 UNC.208 Perhaps more importantly, recent practice suggests that the problem of anticipatory self-defence is increasingly overshadowed by debates about the ‘accumulation of events’ doctrine.209 As noted above, that doctrine may serve to broaden the notion of armed attack. Yet it also, and much more fundamentally, affects the temporal dimension of self-defence. If attacks can be accumulated, then a response will satisfy the immediacy requirement even if it comes too early or too late to repel the single incident which prompted it. The risks of such an approach are readily apparent – in the extreme scenario, a state facing continuous ‘pin prick attacks’ by a terrorist movement can rely on self-defence to justify the use of force sine die. While understandable as a means of re-defining the gravity of armed attacks, the ‘accumulation doctrine’ thus effectively undermines the immediacy requirement characterizing the temporal right of self-defence.
As noted above, the ‘accumulation doctrine’ has received increased support in recent practice, but has not officially been endorsed.210 The preceding considerations suggest that its growing acceptance presents grave risks. The risks are borne out particularly clearly by Operation Enduring Freedom, an operation initially based on a broad, yet defensible reading of Article 51 UNC, which has turned into a self-perpetuating military campaign serving a range of objectives. It is submitted that that campaign has clearly overstretched the boundaries of even the broadest understanding of self-defence.211 One can only hope that the international community's willingness to accept a ‘quasi-permanent’ state of self-defence will remain an isolated deviation from the general rule, perhaps to be explained by the singular impact of the 9/11 attacks which triggered it.
The Defensive Character of Measures
Equally worrying is the tendency of the international community to accept claims based on self-defence which in reality do not serve a defensive purpose. As noted above, under the traditional, restrictive, doctrine, self-defence was not available against ‘instant’ terrorist attacks which were completed before the victim state could react.212 The ‘accumulation doctrine’, if accepted, of course might help avoid that result. Still, even under that doctrine, measures of self-defence in principle would have to serve a ‘defensive’ purpose. The survey of practice suggests that this requirement today is applied very flexibly. States have labelled as self-defence a whole range of measures which did not serve a defensive purpose. For example, when justifying its 1993 attack on Baghdad or the 1998 bombardment of Sudan/Afghanistan, the United States invoked Article 51 UNC, but argued in terms of retaliation, not protection.213 Similarly, Russia, explaining its attacks on the Pankrisi Gorge, invoked a right to enforce international law, not to defend itself.214 By the same token, Iran's pursuit of Kurdish fighters into Iraqi territory did not serve to ‘repel an attack’ (whether grave enough/imminent or not) but to arrest criminals.215 Of course it is often difficult to re-establish the motives prompting a state to use force, and the line between defence and retaliation may at times be difficult to draw. Yet one cannot fail to note that in recent years states have invoked self-defence to justify conduct which primarily served non-defensive purposes.
Again, it may be too early to tell whether recent practice will modify the existing standards. In the light of claims to ‘retaliatory self-defence’, the dangers of the recent trend need to be recalled. By accepting that states merely ‘pay lip-service to the need to act in self-defence’216 while in reality pursuing other objectives, the international community seems to give up an inherent feature of the right of self-defence, namely its defensive character. The result may be rather paradoxical: while unequivocally condemning the doctrine of armed reprisals, the international community seems indeed – as Pierre Klein aptly noted – gradually to accept armed reprisals disguised as self-defence.217 In so doing, it may re-introduce an altogether flexible exception to the ban on force which had been considered illegal for decades, and abandon an inherent feature of the right of self-defence.
The Present: Taking Stock
The preceding considerations suggest that in many respects the last two decades have transformed the rules governing forcible responses against terrorism. The international community's growing determination to fight terrorism has not left the jus ad bellum unaffected. The extraterritorial use of force remains prima facie illegal, but justification seems much more readily available than 20 years ago. Insofar as the UN's collective security system is concerned, this has been uncontroversial. It is beyond doubt today that the Security Council can authorize the use of force against terrorists. However, the political limits remain. Even in times of worldwide consensus, individually affected states may prefer to act unilaterally. More importantly, even a ‘legislating’ Security Council seems to lack the political will to decide on military enforcement against terrorists.
As far as forcible measures are concerned, the contemporary ‘strong international policy against terrorism’218 therefore has been implemented outside the Security Council framework. This has brought about new uncertainty about the scope of exceptions to the ban on force. Whether the new practice is a temporary aberration or ushers in a new era in which the jus ad bellum is applied with greater flexibility may in some respects be too early to tell. Yet three points can be made: (i) the new practice is more than a response to the 9/11 attacks. On frequent occasions, during the last two decades, many different states have asserted a right to use force against terrorists, and their conduct has been viewed rather favourably by the international community. (ii) The new practice is exclusively justified under an expanded doctrine of self-defence, even though it may bear little resemblance to classical forms of self-defence. (iii) As far as possible re-adjustments of the law of self-defence are concerned, the debate about the state or non-state origin of the attack seems overstated. While there seems to emerge a new, lower standard of attribution, the new practice can be explained as a reform of the previous, restrictive approach set out in Nicaragua. The real ‘normative drift’219 concerns two other aspects of the law on self-defence: the immediacy test and the requirement that measures be defensive in character. By increasingly abandoning these limitations, the international community runs the risk of transforming a temporary defensive right into an open-ended instrument for forcible intervention.
Conclusion and Coda
The preceding discussion suggests that the law governing anti-terrorist force is in a process of change. This process may be looked at from different angles. The growing international consensus against terrorism has entailed far-reaching legal consequences. International rules requiring state cooperation or active state conduct against terrorism have multiplied. More important, at least for present purposes, is the militarization of the fight against terrorism. International law now accepts that the fight against terror may require the use of extraterritorial force – certainly within the multilateral context, but possibly also outside it. To return to a point made earlier, a Hague Lecture on ‘Terrorisme et droit international’ today would have to devote more than five pages to the problem of forcible measures.220
As regards the jus ad bellum, developments during the last two decades indicate that the law is capable of adaptation. The growing consensus against terrorism has put pressure on the traditionally restrictive regime. This pressure has affected the interpretation of exceptions to the ban on force. These are increasingly construed broadly, so as to take account of the denunciation of terror. The broad construction can be easily accommodated within the United Nations system of collective security. With respect to unilateral force, it requires a far more difficult adjustment of the traditional rules, and is proving much more controversial. Yet the process of adjustment seems well under way. The point can by illustrated by reference to a statement made some 25 years ago, in Security Council debates on one of the more high-profile instances of anti-terrorist force. To justify his country's raid on Tunis, Israel's representative then argued that ‘Tunisia knew very well what was going on in this extraterritorial base, the planning that took place there, the missions that were launched from it, and the purposes of those missions: repeated armed attacks against my country and against innocent civilians around the world. Tunisia, then, actually provided a base for murderous activity against another State and, in fact, the nationals of many States who are the objects and victims of this terrorist organization.’ And, further, ‘[a] country cannot claim the protection of sovereignty when it knowingly offers a piece of its territory for terrorist activity against other nations’.221
Twenty-five years ago that statement was roundly rejected, partly because Security Council members evaluated the facts differently, but partly because they did not accept Israel's legal argument. As a consequence, 14 members of the Security Council agreed to condemn Israel's conduct in the strongest terms, with only the United States abstaining.222 How the Security Council would respond were it to hear the same argument again today remains a matter for speculation. Much depends of course on the facts and the credibility of a state's claim. Yet the underlying legal claim argument – that states aiding and abetting terrorists abuse their sovereignty and must accept some form of counter-action – has become a standard formula of modern debates and would probably meet with approval of some and tacit agreement of many states.
The consequences, and implications, of the new, flexible approach are readily apparent. By recognizing a broad construction of the exception, the international community, in a specific field of international relations, increasingly seems to free states from constraints imposed by the ban on force.223 In that respect, developments in the law governing anti-terrorist force fit in well with a new willingness to accept (or at least consider justifiable) the use of force where it serves a legitimate cause, notably borne out by renewed debates about the legality of humanitarian interventions. This new flexibility of course comes at a price: the broadly construed exceptions to the ban on force can be abused – a risk which is much increased by the worrying tendency to extend self-defence in time or even to accept ‘retaliatory self-defence’. Yet the preceding discussion also suggests that during the last two decades the international community has been prepared to take that risk. Readjustments of the jus ad bellum are not deduced from some legal principle, but borne out by the actual practice of states, which at least during the last two decades has recognized the right of states to use anti-terrorist force if this served to avert threats and no other means seemed available. Whether this flexible approach persists will depend not least on the prudence of states in making use of their ‘reappropriated’224 right.
Coda – The Future: Anti-terrorist Force 2029
If the jus ad bellum has evolved during the last 20 years, then it is likely to evolve further over time. If it is accepted that some aspects of the rules on anti-terrorist force are in a process of readjustment, then they may even require some further development, either confirming recent trends or reversing them. So how might the law develop? How might commentators, 20 years from now (possibly in a symposium celebrating the 40th anniversary of the European Journal), reflect on ‘the use of force against terrorists’? Prediction, as has been observed,225 ‘is very difficult, especially if it's about the future’, so the subsequent paragraphs are speculative. However, three scenarios can be envisaged.
A Return to the Criminal Law Strategy
The first scenario is that of a de-militarization of the fight against terrorism and of a return to the criminal law strategy. This would presuppose further progress in the move towards a criminalization of terrorism and the establishment of a much more effective system of international cooperation. As part of an ever ‘strong[er] international policy against terrorism’,226 states would need to accept – whether in international conventions or through Security Council legislation – a broader range of obligations relating to the treatment of terrorists and terrorist organizations. If existing treaties are to serve as a model, the future multilateral regime is likely to include enhanced duties to criminalize and prosecute terrorist activities, arrangements for cooperation in criminal matters, as well as in the fight against financiers of terrorism; in addition, one might hope that it also includes safeguards protecting individual rights of terror suspects. As a residual option, the international community might also establish international judicial bodies competent to prosecute terrorist activities – maybe eventually even a ‘terrorism chamber’ of the International Criminal Court. Just as it has done with respect to other international tribunals, the Security Council would be well-placed to assume the role of an enforcement agent, e.g., by requiring states to hand over terror suspects, failing which they might face sanctions. In fact, the (no doubt reformed) Security Council of 2029 might even call upon a standing anti-terrorist force to arrest terror suspects and hand them over to the ICC's terrorism chamber. While the latter development may require some imagination, the criminal law scenario could build on the groundwork laid in existing anti-terrorism conventions and on precedents of previous Security Council action.227 In contrast, an effective criminal law strategy, backed up by Security Council sanctions, would reduce the need (perceived or real) for unilateral force. The broad construction of self-defence emerging during the 1990s and 2000s might in retrospect then appear as no more than a temporary aberration.
A More ‘Protean’ Jus ad bellum
The second scenario is less optimistic, but may be more likely. It is based on the assumption that the trend towards unilateral anti-terrorist force consolidates. If states continue to exercise anti-terrorist force, how would this consolidated body of practice affect the state of the law? On one level the answer is clear: many aspects of the jus ad bellum are flexible enough to accommodate change (and indeed may have accommodated it already). It is submitted that the Charter does not preclude the recognition of a right to use self-defence against states harbouring terrorists, if that is required to avert an imminent armed attack. By the same token, the Charter does not rule out the ‘accumulation doctrine’, although that doctrine presents grave risks. In fact, the Charter may even tolerate the trend towards retaliatory defence, even though one might hope that states abusing the self-defence concept for non-defensive purpose would be more open about their real intentions. In short, if the international community agrees on the broader reading of Article 51 UNC emerging during the last two decades, then that re-interpretation will become accepted over time. Commentators writing in 2029 might in retrospect point to the 1990s and 2000s as the crucial period in which the broader reading had gained ground. Perhaps they might even view the flexible approach to anti-terrorist force as one of the key elements of a general move towards what Sean Murphy has labelled a ‘protean jus ad bellum’.228 The gradual recognition of a right to use anti-terrorist force from the 1990s may in retrospect have paved the way towards a new understanding of the rules on force – a ‘protean’ approach striking a new balance between the absence of force and the protection of common values, permitting states to disregard constraints of the Charter in defence of community goals. If this is to happen (and one may dread the thought), then the commentator writing in 2029 might describe the fight against terror as a catalyst of indeed revolutionary change.
Extraterritorial Enforcement Jurisdiction over Terrorists
Finally, there may be a third scenario transcending the dichotomy between criminal and military anti-terrorism strategies. A commentator writing in 2029 might be able to report on the emergence of a regime of international enforcement jurisdiction over terrorists. That regime may be based on international conventions or Security Council authorisations, or both. It would allow states to use force extraterritorially as a designated measure of law enforcement. It may have drawn inspiration from long-established rules governing enforcement measures against pirates on the high seas,229 or controversial attempts to re-write the principles governing the interdiction of ships,230 but would need to move beyond these so as to allow for enforcement on foreign soil. Presumably, the right of states to enforce anti-terrorist rules would be subject to rather strict conditions: for example, it could require the responding state to provide evidence of the territorial state's failure to act (thus taking up considerations of necessity and complementarity), and to respond only against individuals or organizations included on an internationally-agreed ‘black list’.
The new regime of extraterritorial enforcement jurisdiction over terrorists would be based on the conviction that the fight against terrorism requires a military component, but it would channel military measures into a regime of enforcement jurisdiction established in the interest of the international community. As such, it would recognize that forcible measures against terrorists in recent years have typically been based on the logic of enforcement, not defence. Unlike other possible scenarios, a regime of extraterritorial jurisdiction would address terrorists directly (just as Article 110 of the Law of the Sea Convention (LOSC) addresses pirate ships directly) rather than indirectly applying the inter-state rules on force to them.
Evidently, the prospects for such an enforcement regime are rather slim. It presupposes a willingness of states to accept a far-reaching exception to present-day rules governing enforcement jurisdiction. Still, the possibility should not be completely ruled out. Recent Security Council resolutions have revived and broadened the long-dormant rules governing enforcement jurisdiction over pirates, so as to permit enforcement measures within sovereign spaces of another state.231 The fate of the ‘proliferation security initiative’ abandoning the principle of exclusive flag state jurisdiction in favour of some form of ‘public interest enforcement’ is by no means sealed. It is entirely possible that these developments should signal a broader approach to enforcement jurisdiction generally which, by 2029, could be applied to terrorists as well.
The different options are not mutually exclusive. Twenty years from now, the fight against terror may still be fought, as today, with criminal and military means, possibly complemented by an internationalized enforcement regime of limited application (e.g., in failed states). In fact (and that might be scenario 4), the fight against terror may have effectively been abandoned if different factions within the international community again revert to the rhetoric of ‘our’ versus ‘your’ terrorists, the former fighting for legitimate causes, the latter engaging in criminal activities. If there is one lesson to draw from developments during the last two decades, then it is that no scenario can be excluded. Who, in the 1980s, would have expected the international community to move towards an unconditional condemnation of terrorist activities? Who, in 1989, would have expected the Security Council to adopt a comprehensive, non-military, anti-terrorism programme through a series of resolutions? If the international community is capable of maintaining a strong stance against terrorism, then there is no reason to expect that the jus ad bellum should be immune from (further) change.