Hostname: page-component-74656f9884-npj9w Total loading time: 0 Render date: 2026-11-10T06:00:12.023Z Has data issue: false hasContentIssue false
  • English
  • Français

Applying Necessity and Proportionality to Anti-Terrorist Self-Defence

Published online by Cambridge University Press:  05 March 2012

Christian J Tams  and
James G Devaney
Christian J Tams
Affiliation:
Christian J Tams is Professor of International Law at the University of Glasgow (email: christian.tams@glasgow.ac.uk).
James G Devaney
Affiliation:
James G Devaney is PhD candidate at the European University Institute (email: james.devaney@eui.eu).

Abstract

Recent years have seen increasing reliance on self-defence as a concept justifying the use of force against armed attacks by terrorists operating from within a foreign state. While the matter has prompted major debates, a good case can be made that contemporary international law now begins to recognise a right of self-defence against attacks by terrorists. (We will refer to this in the following as ‘anti-terrorist self-defence’, meaning a response directed against an attack carried out by a terrorist organisation.) This article deliberately does not rehearse the major debates about the scope of self-defence, but moves beyond them. It seeks to assess whether and how the two main limitations on the exercise of self-defence – namely necessity and proportionality – could be adapted to self-defence outside the inter-state setting. It shows that both limitations do indeed require some adaptation in order to be applied to anti-terrorist self-defence.

With respect to necessity, we argue that this adaptation is actually a chance to make sense of a requirement that is of very limited practical relevance in the inter-state setting. As the article shows, necessity can assume a crucial role in the anti-terrorist setting: in order for trans-border uses of force to be necessary, states invoking self-defence must demonstrate that the terrorist threat cannot be repelled by the host state (on whose territory the terrorists are based).

As for proportionality, recent debates about anti-terrorist self-defence expose the weaknesses of proportionality as a restraint on self-defence. Construed primarily as a prohibition against excessive force, even in the inter-state setting, proportionality has always been dependent on the nature of the aims pursued in the name of self-defence. Recent anti-terrorist practice suggests that many states are prepared to accept far-reaching uses of force against terrorists. It also suggests that, in applying proportionality, the targeted (or otherwise) character of the response is of crucial importance. The law is clearly in a state of flux, but in debates such as those about Turkey's use of force against the PKK or Israel's military operations against Hamas or Hezbollah, one can see the beginning of a process of developing guidelines on proportionality.

Keywords

use of forceself-defenceterrorismnecessityproportionality

Information

Type
Articles
Information
Israel Law Review , Volume 45 , Issue 1 , March 2012 , pp. 91 - 106
Copyright
Copyright © Cambridge University Press and The Faculty of Law, The Hebrew University of Jerusalem 2012

1. Introduction

Necessity and proportionality have long been recognised as conditions attached to the exercise of inter-state self-defence in international law, despite the fact they are not explicitly mentioned in Article 51 of the UN Charter.Footnote 1 Crucially, Article 51 does not regulate the right to self-defence comprehensively; in order for it to be applicable, ‘external factors’ need to be taken into account. That necessity and proportionality are two such factors is generally agreed. Pursuant to the jurisprudence of the International Court of Justice (the ICJ), ‘[t]he submission of the exercise of the right of self-defence to the conditions of necessity and proportionality is a rule of customary international law’;Footnote 2 in this respect, the Court was surely correct to note that ‘[t]he conditions for the exercise of the right of self-defence are well settled’.Footnote 3 However, whilst they may be ‘well settled’ in principle, the application of necessity and proportionality raises considerable problems. What is more, there is substantial uncertainty as to whether the two requirements can be applied outside the traditional, inter-state self-defence setting, notably to assertions of self-defence against attacks by terrorists. The subsequent discussion focuses on this latter question and proposes how to adapt necessity and proportionality to self-defence outside the inter-state setting.Footnote 4

2. Background

Any assessment of the limitations governing ‘the exercise of the right of self-defence’,Footnote 5 such as necessity and proportionality, presupposes that such a right can be invoked at all. Whilst this presumption presents little difficulty in the case of the traditional inter-state setting, in the present context the corresponding presumption that attacks by terrorists can give rise to a right of self-defence is a matter of considerable debate.Footnote 6 The doctrinal debate will not be rehearsed in the following in any detail, not least because it is so well documented elsewhere.Footnote 7 A quick glance at the specialised literature suggests that the question is not settled; we believe, however, there is now considerable evidence to suggest that the right of self-defence has been ‘opened up’ to admit responses directed against attacks by terrorists, even where these, under traditional rules, could not be attributed to another state. Technically, this ‘opening up’ affects the interpretation of the requirement of an ‘armed attack’ in Article 51 of the UN Charter, which triggers the right of states to self-defence. Traditionally, in line with the dominant inter-state understanding of the jus ad bellum, this ‘armed attack’ requirement had been understood to mean ‘armed attack by another state’, or at least ‘armed attack attributable to another state’Footnote 8 – under reasonably narrow rules of attribution developed notably by the ICJ's Nicaragua judgmentFootnote 9 and subsequently affirmed (and generalised) in Articles 5 and 8 of the ILC's text on state responsibility.Footnote 10 This meant that in order to validly claim a right of self-defence against terrorist attacks, a responding state traditionally had to establish that the terrorists had in fact been under the direction and control of another state, namely the state targeted by the self-defence operation.

Recent practice, while still uncertain, would seem to indicate that the ‘armed attack’ requirement has been broadened to include terrorist attacks even where these could not be attributed to another state under the traditional rules of attribution. The key factor pointing towards such an ‘opening up’ of self-defence is international practice. To be clear, the traditional view had never been uncontested: states such as Israel, Portugal, and South Africa had asserted self-defence against armed attacks by non-state actors.Footnote 11 However, since the 1990s – and arguably in line with the stricter international policy against terrorism – this view has gained support in various forms: a considerable number of states have asserted a right to resort to self-defence against attacks by terrorists that could not be attributed to another state.Footnote 12 Perhaps as importantly, a considerable number of states have accepted their arguments – either endorsing them directly, or remaining silent in the face of assertions of anti-terrorist self-defence.Footnote 13 While each of the many instances remains controversial, three cases – on which the subsequent analysis will draw – may serve to illustrate the gradual shift of view:

  1. (1) Since the 1990s Turkey has repeatedly invoked a right to use force against Kurdish PKK bases in northern Iraq, culminating in Operation Sun, a ground offensive involving several thousand Turkish troops.Footnote 14 The international community largely refrained from condemning this large-scale use of force, with many states stressing the need for a targeted response, but generally showing a ‘mixture of sympathy and concern’Footnote 15 for Turkey's approach.Footnote 16

  2. (2) Israel relied on self-defence to justify its response to Hezbollah's cross-border operation into northern Israel on 12 July 2006.Footnote 17 To be sure, many states criticised Israel for failing to respect the limits of proportionality.Footnote 18 However, while the scope of the Israeli response may have been excessive (an issue to be addressed below at Section 6), ‘[m]any states expressly recognized and, therefore, acknowledged as a matter of principle that Israel enjoyed a right to respond to the attacks by Hezbollah and Hamas under the law of self-defence’.Footnote 19

  3. (3) Finally, responding to terrorist attacks on US embassies in Nairobi and Dar es Salaam,Footnote 20 the US invoked self-defence in order to prevent their repetition.Footnote 21 Whilst it could not be said that other states uniformly supported the action taken by the US,Footnote 22 international reaction was generally muted,Footnote 23 and most of the US's allies expressed modest support for its position.Footnote 24

These three instances – to which the most obvious case, namely the use of force against those responsible for the attacks of 9/11, must be added – are illustrative of a broader tendency to ‘open up’ Article 51 of the UN Charter to permit self-defence against armed attacks by terrorist actors that could not be attributed to another state under the traditional criteria for attribution.Footnote 25

Controversies and uncertainties, of course, remain.Footnote 26 The ICJ's jurisprudence suggests caution (to say the least);Footnote 27 the majority of states remain unwilling to recognise a broadening of self-defence expressly; and it is by no means certain whether anti-terrorist self-defence should be construed as self-defence against another state (to whom terrorist acts are attributed under more lenient conditions than usual) or as self-defence against non-state actors.Footnote 28 However, notwithstanding these concerns, it seems to us difficult to ignore the trend towards the recognition of a right to use self-defence against terrorist attacks, and it may be appropriate to assess how such a trend would affect the application of the two key limitations governing ‘the exercise of the right of self-defence’.Footnote 29 Against this background, we turn our attention to the question whether necessity and proportionality can be applied to instances of anti-terrorist self-defence. This first requires a closer look at necessity in its traditional inter-state setting.

3. Inter-State Necessity

At its most basic level, as a limit on self-defence in the traditional inter-state setting, necessity is no doubt conceptually important, but of very limited practical relevance. In contrast, it is submitted that necessity can assume a much more important role in evaluating anti-terrorist self-defence.

In the traditional inter-state setting, the most fundamental requirement for necessity is the absence of a reasonable alternative. Given the comprehensive character of the ban on unilateral force in international law, the necessity test is formulated rather strictly: in order to be necessary, measures of self-defence have to be the only reasonable option of addressing the armed attack.Footnote 30 Other responses have to be unavailable; the responding state must have – to use the language of the Caroline formula – ‘no choice of means’.Footnote 31 This aspect of necessity is mainly understood to concern the relationship between forcible and non-forcible unilateral responses.Footnote 32

So why then is it normally so practically irrelevant, especially since nearly all victim states will be in a position to resort to one or more non-forcible options such as breaking-off relations, suspension of treaties or exercise of political pressure? The crucial point is that necessity does not require victim states to exhaust all non-forcible responses before resorting to self-defence, but only those alternatives that are likely to be effective.Footnote 33 A victim state must therefore assess, in good faith and on the basis of the facts available at the time, whether it could repel or avert the attack without resorting to force.Footnote 34 In making this assessment, the real impact of this aspect of the necessity test in the traditional inter-state setting is limited by a number of factors.Footnote 35

First, in requiring a previous ‘armed attack’, international law establishes a high threshold before self-defence becomes an option at all. While the precise relationship between Article 2(4) and Article 51 of the UN Charter is a matter of debate, self-defence under contemporary international law seems available only against qualified uses of force.Footnote 36 For there to be an ‘armed attack’ triggering self-defence, there must be an illegal use of force of some gravity. Whilst this sufficiently serious use of force against the victim state does not in itself mean that any action it takes in self-defence is necessary, it does mean, at the very least, that the argument that such action is unnecessary becomes very difficult to sustain or accept.

Secondly, from the wording of Article 51 (permitting unilateral self-defence only ‘until the Security Council has taken measures necessary to maintain international peace and security’), it seems clear that acts of self-defence could become unnecessary if superseded by institutional enforcement action under Chapter VII of the Charter.Footnote 37 However, while that is accepted in principle, the Security Council rarely disapplies the right of self-defence.Footnote 38 For Security Council actions to qualify as ‘necessary measures’ one would normally expect a binding resolution requiring a stay of violence (for example, the imposition of a cease-fire)Footnote 39 or some other form of sanction directed specifically against the armed attack in question and capable of ending it.Footnote 40 However, such strict Council actions are by no means common; and even where the Council acts, such as in response to the occupation of Kuwait,Footnote 41 it usually affirms the right of self-defence rather than disapplies it.Footnote 42 The controversial case of SC Res 1701 potentially suspending Israel's right of self-defence (although not expressly adopted with reference to Chapter VII) is one of the very few exceptions.Footnote 43

Consequently, as a result of the high threshold of the ‘armed attack’ requirement and the fact that the Security Council very rarely disapplies the responding state's right to self-defence, the necessity requirement is somewhat hamstrung. Of course, states will not usually resort to force lightly, but only after a comprehensive political assessment. That said, as a matter of law, a victim of an armed attack that reaches the threshold required by Article 51 of the Charter would typically be able to establish the necessity of a forcible response. In this respect, it may be said that in the traditional state-setting necessity only rarely excludes the exercise of self-defence.

4. Anti-Terrorist Necessity

Crucially, with respect to anti-terrorist self-defence, necessity is potentially of much greater relevance. The reason for that is rather straightforward: whilst self-defence in the traditional setting involves two states, anti-terrorist self-defence necessarily involves a group (the terrorist entity) in addition to the other two states (or non-states as the case may be). In the traditional setting, it is clear that the response has to come from the victim of the armed attack; in the anti-terrorist setting, however, it is by no means clear who should respond. In other words, ‘there is an additional rung in the ladder’Footnote 44 of potential responses, which allows us to have meaningful discussions about necessity and opens up a new dimension of debates about necessity.

This ‘additional rung in the ladder’ is crucial because repressive measures by the host state against terrorists are clearly an alternative option, a ‘choice of means’Footnote 45 much preferable to the extraterritorial use of armed force by the victim state. Unlike in the traditional inter-state setting, when assessing the necessity of self-defence against a terrorist group, the comparison is not only between forcible and non-forcible responses or between unilateral self-defence and institutional action, but also between unilateral action by the victim state and unilateral action by the host state. As a general rule, for self-defence to be considered necessary, the victim state has to make an attempt to have the host state suppress the terrorist threat. Alternatively, the victim state might attempt to cooperate with the host state against terrorists (so that joint enforcement operations help to avoid resort to force), or seek the host state's consent to extraterritorial anti-terrorist measures.Footnote 46

As a matter of principle, victim states contemplating self-defence against terrorists do not seem to dispute the primacy of host state actionFootnote 47 – in this respect, the crucial role of necessity as a constraint on anti-terrorist self-defence seems to be accepted. However, the question remains under what circumstances responding states can rightfully claim that, notwithstanding the possibility of host state enforcement, their own action in self-defence has become necessary. International practice suggests that this primarily depends on the relationship between the host state and the terrorist organisation in question. While international practice has yet to produce clear-cut rules, it seems to display a rather loose understanding of this particular aspect of ‘necessity’. There are three aspects to this.

First, states that actively support terrorists are unlikely to take effective enforcement action against them.Footnote 48 In this case, reactions by the victim state will be considered necessary as host state enforcement is not a realistic option. Of the various instances of anti-terrorist self-defence, this is presumably the least controversial. A clear example of this can be seen in the international response against Al Qaeda operating from Afghanistan. In that case, at least from the late 1990s, the international community took the view that the Taliban regime did support the terrorists; as a consequence, the necessity of some action against Al Qaeda targets in Afghanistan was not disputed: the Taliban, it was felt, would simply not repress their activities.Footnote 49

Secondly, where a host state does not actively support terrorists (or there is no evidence to support such an allegation), the matter is more complex.Footnote 50 Practice nevertheless seems to indicate that acts of self-defence may be considered necessary if the host state tolerates terrorist activities on its territory or harbours terrorists. This is evidenced, for example, by the international response to Turkey's invasion of northern Iraq (accepted largely because of the passive attitude adopted by the northern Iraqi authorities).Footnote 51 Additionally, states such as Russia or Iran have claimed a right to use force where host states had tolerated (as opposed to supported) terrorists operating from within their soil.Footnote 52 Legally, such an approach might be explained by reference to the duty of states to suppress terrorist activities.Footnote 53 Such a duty is enshrined in modern Security Council resolutions and anti-terrorism conventions and it may also derive from the general duty of states not to knowingly allow their territory to be used for violations of the rights of other states.Footnote 54 It seems also implicit in modern international conventions such as the 2005 African Union Non-Aggression and Common Defence Pact, which expressly qualifies as an act of aggression the harbouring of terrorists, as well as any provision of support for them.Footnote 55

Whether the duty to suppress terrorist activities has actually been violated (and self-defence thus becomes available) may, of course, be difficult to establish in practice. The Turkish example referred to above suggests that where terrorists have operated over a considerable period of time, and where they have repeatedly mounted attacks, the host state's unwillingness to suppress their activities would be considered to be established. Repeated international appeals to the host state to take action against terrorists operating on its territory would seem to be another relevant factor, especially if these are translated into specific commitments (whether deriving from Security Council sanctionsFootnote 56 or under international agreementsFootnote 57).

Finally, recent practice indicates that acts of self-defence can even be considered necessary where the host state is merely unable to suppress terrorist activities on its territory. For example, the Israeli invasion of Lebanon initially was considered to amount to self-defence by many states, even though Lebanon arguably was not in a position to suppress Hezbollah's activities.Footnote 58 Other states invoking self-defence have equally asserted a right to act if the host state seemed incapable (rather than unwilling) to suppress terrorist threats.Footnote 59 Two considerations may be offered to support that position: (i) the line between situations of ‘unwillingness’ and ‘inability’ may be extremely thin, and not suited to distinguish illegal acts of force from instances of potentially necessary self-defence; (ii) in addition, and perhaps more importantly, states that are unable to suppress terrorist activities on their territory would still be under a duty to do so since an inability to comply does not relieve states of their obligations.Footnote 60

As is clear from this brief summary, the primacy of host state action (as an aspect of the necessity test) is undisputed in principle; however, responding states have pushed the boundaries of ‘necessity’ far. It will remain to be seen whether, as practice consolidates, claims to use self-defence against states that are unable or unwilling to suppress terrorist activities on their territory will continue to be accepted. In our view, the recent broadening of Article 51 of the Charter will only be sustainable if the application (and understanding) of necessity reflects the exceptional nature of self-defence. Much, of course, will depend on the willingness of victim states to disclose evidence that had led them to take action in the first place. In this respect, it is surprising that so far the international community has shown a considerable reluctance to assess the legality of self-defence operations through the lens of necessity. While actual uses of self-defence have been discussed, notably with regard to their proportionality (a matter to be addressed belowFootnote 61), claims that action was ‘necessary’ have not been seriously probed – in fact, states relying on self-defence have often not been asked to ‘make their case’ before international forums such as the Security Council.Footnote 62 While the full potential of the necessity requirement thus has yet to be realised, the preceding considerations suggest that necessity indeed should be the key factor in debates about anti-terrorist self-defence.

5. Inter-State Proportionality

Proportionality is considered to be the more complex limitation restricting the exercise of self-defence.Footnote 63 Unlike necessity, which concerns the availability of self-defence, proportionality limits the scope and intensity of the response.Footnote 64 In considering proportionality as a limitation on the right to self-defence it is submitted that, compared to other areas of international law where proportionality features, proportionality as a restraint on the use of force justified as self-defence is comparatively weak. This conceptual weakness is clearly brought out by the recent practice of anti-terrorist self-defence; however, in the international debate about specific anti-terrorist operations a number of guidelines on the application of proportionality would seem to be emerging.

The requirements of proportionality apply in many areas of international law, from human rights law to international humanitarian law, but also in the rules governing countermeasures or self-defence.Footnote 65 While always concerning a balancing of interests, it is crucial to note that the application of proportionality in these different fields serves different functions, which in turn influence the formulations of the respective proportionality equations.Footnote 66 In the context of self-defence – notwithstanding the frequent affirmation of the principle, and the need to distinguish proportionality under the jus ad bellum from proportionality restrictions imposed by the jus in bello Footnote 67 – there is a surprising degree of uncertainty about the functions of the proportionality test. A number of misleading statements by the ICJ seem to suggest that what is required is a quantitative evaluation of the harm inflicted on the victim state, which would have to be compared with the harm inflicted by the self-defence operation.Footnote 68 In contrast, most commentators rightly note that self-defence, even when directed against an ongoing attack, aims to avoid future consequences (such as the occupation of the victim state, the loss of territory, and so on). It is therefore a response against a threat as much as against an attack already inflicted. What is more, it is a response that under Article 51 is acknowledged as an ‘inherent right’ – in other words, a perfectly legal form of conduct.

Taking into account both factors, it is submitted that the proportionality of self-defence operations needs to be assessed in a prospective way, comparing the scope and intensity of the response to the threat posed by the armed attack, and bearing in mind international law's recognition of a ‘superior right’Footnote 69 to repel armed attacks by means of self-defence. Notably, this means that the response may very well be of greater intensity than the initial armed attack.Footnote 70

As perceptively noted in the Chatham House principles on self-defence, ‘[t]he force used, taken as a whole, must not be excessive in relation to the need to avert or bring the attack to an end. The physical and economic consequences of the force used must not be excessive in relation to the harm expected from the attack’.Footnote 71 In applying this prohibition against excessive force, commentators have typically taken into account factors such as the gravity of the initial armed attack and the likelihood of its success; the scale and intensity of the response, its geographical and temporal scope and the amount of collateral damage sustained.Footnote 72 However, none of these factors is particularly specific, and none of them can be a substitute for an overall balancing of interests in which everything depends ‘on the contingencies of an operation’.Footnote 73

6. Anti-Terrorist Proportionality

The general weakness of proportionality as a restraint on self-defence is reflected in debates about anti-terrorist self-defence. In fact, if anything, recent anti-terrorist practice seems to have exacerbated proportionality's conceptually weak condition. It has done so because (on the limited evidence available so far) self-defence operations conducted against terrorists seem to be characterised by two peculiar features.

First, recent international practice would seem to indicate that self-defence operations against terrorists can pursue far-reaching aims. For example, Turkey's Operation Sun campaign against the PKK in northern Iraq in 2008 was justified as a broad response that would finally weaken PKK for good.Footnote 74 Similarly, Israel indicated that, in addition to repelling or averting specific attacks, its intervention in Lebanon was designed to weaken Hezbollah decisively,Footnote 75 while the US strikes on Sudan and Afghanistan have been considered to be ‘punitive rather than defensive’.Footnote 76 These examples suggest that the ‘defensive’ model of self-defence as a response against specific armed attacks, limited to repelling them (or averting imminent future attacks), has come under strain. To the extent that it has, this will of course have implications for the application of the proportionality test: if self-defence can aim to decisively weaken a terrorist organisation, then the use of force will only be disproportionate if it is excessive in relation to this wider aim. Accordingly, proportionality becomes even more difficult to apply meaningfully.

Secondly, recent anti-terrorist practice also seems to stretch the temporal limits of self-defence. As is well-known, recent years have witnessed high-profile debates surrounding anticipatory self-defence.Footnote 77 What is more, the increasing popularity of the ‘accumulation of events’ doctrine (permitting the construction of an armed attack out of a series of small, one-off attacks) undermines narrow temporal notions of self-defenceFootnote 78 – if an armed attack is construed out of a series of pinpoint attacks stretching over time, the requirement that the response would have to be temporarily linked to the armed attack is weakened. In short, the ‘normative drift’Footnote 79 that has affected the law of self-defence over the last two decades is clearly reflected in the application of proportionality in underlining the vagueness of the requirement and adding to its uncertainty.

Nevertheless, it would be wrong simply to dismiss proportionality as a mere ‘rhetorical tool’.Footnote 80 It is submitted that, inherent vagueness and normative drift notwithstanding, proportionality does have a role to play. This is so because, unlike necessity (the potential of which, as noted above, remains to be realisedFootnote 81), proportionality is a key factor in international debates about anti-terrorist self-defence. As proportionality is regularly discussed, there is a considerable deal of international practice to draw on, comprising justifications of self-defence operations as well as international reactions. Notwithstanding the many conceptual uncertainties surrounding proportionality, these debates constitute a rudimentary evaluation process by which the international community applies general normative commands to specific patterns of fact,Footnote 82 and which benefits from broad participation.Footnote 83 Whilst any normative guidance gleaned from this rudimentary evaluation process remains general, recent practice would seem to prompt three observations:

  1. (1) Debates about Turkey's Operation Sun and Israel's intervention in Lebanon suggest that the international community will be willing to tolerate as proportionate rather drastic measures against terrorists. In long-term struggles, such as in the case of Turkey–PKK and Israel–Hezbollah, many states seem prepared to accept that self-defence may involve weakening the terrorists as opposed to repelling their attack. What is more, the fact that harm inflicted on terrorists massively exceeds the harm sustained by the victim state's citizens in itself does not seem to present major problems. Turkey's invasion of northern Iraq may serve to illustrate the point: it was prepared by massive air raids and artillery fire and involved the deployment of large numbers of fully armed ground troops, which occupied a sector of foreign territory. Furthermore, on Turkey's own admission, casualties among the PKK were 20 times higher than the number of Turkish nationals killed in attacks preceding the invasion.Footnote 84 Nevertheless, the scale of the invasion was not generally considered to be excessive; in fact, many states lauded the precision of Turkey's use of force, without discussing the number of losses among the PKK.Footnote 85 This in turn suggests that where responses target terrorists, counterforce may be massively more intensive than the initial armed attacks.Footnote 86 What is more, recent practice suggests that geographical factors that may be considered relevant to the proportionality of inter-state self-defence are of limited relevance: hence states hit by terrorist attacks on their home soil have asserted a right to respond against terrorists at their base – and even where their conduct was not generally accepted, the fact that the self-defence operation had carried the fight against terrorism into far-away, remote countries seemed to be a factor of limited relevance.

  2. (2) In contrast, recent practice suggests that self-defence will be considered to be excessive (and thus disproportionate) if it is directed against the host state rather than the terrorists operating from within it. In fact, with the exception of Operation Enduring Freedom targeting Taliban forces, states asserting a right to use anti-terrorist self-defence have accepted that limitation: as noted by Kimberley Trapp, ‘all invocations of the right to use force in self-defence against terrorist attacks since the 2001 US-led campaign in Afghanistan purport to be based on a right to use defensive force against non-state terrorist actors, not against the state acquiescing in their activities’.Footnote 87 Conversely, where responding states were perceived as targeting the host state directly, their response was quickly criticised as disproportionate – as evidenced most clearly by the international reaction to Israel's attack on the Beirut airfield, which seems to have been a key turning point in the international community's assessment of the war of July 2006.Footnote 88

  3. (3) Finally, in assessing the proportionality of anti-terrorist self-defence operations, the prevention of civilian suffering seems to have become the paramount concern. Put differently, while harm sustained by terrorists seems to be tolerated, even where it reaches considerable proportions, the international community has protested against self-defence operations that were perceived as ‘lacking focus’ and leading to widespread harm among the civilian population of the host state. Again, this seems to be brought out clearly by a comparison between Operation Sun and the Israeli intervention in Lebanon: the former was perceived as a targeted operation in which Turkey avoided unnecessary civilian losses while hitting the PKK hard. In contrast, Israel's intervention in Lebanon (and, even more so, its subsequent Operation Cast Lead in the Gaza strip) were perceived as inflicting unacceptable levels of harm on the civilian population – and thus criticised as being excessive and disproportionate.

Admittedly none of this is dogmatically satisfying. What is more, as the process of norm application is largely left to the discretion of states assessing a self-defence operation, it becomes increasingly open to biases (whether cognitive, political or other). But as the international community begins to come to terms with, and gains experience in, classifying anti-terrorist measures as self-defence, it produces an ever-growing body of evidence on how to apply proportionality, which can be utilised in future settings.Footnote 89

7. Concluding Remarks

It is clear that anti-terrorist self-defence presents specific and grave challenges to the law of self-defence. However, with respect to necessity, the challenge should in fact be seen as an opportunity to make sense of a requirement that has traditionally been of limited relevance. In order for necessity to assume its rightful role, states will have to engage in a discourse that has so far been missing. States can do so simply by asking for evidence to be produced and positions explained on why exactly resort to force was necessary. By doing so, they would bring out clearly that the state invoking self-defence against terrorist entities has to establish that such action was necessary.

With respect to proportionality, the challenge is considerable and it is especially difficult in the case of anti-terrorist entities. Proportionality is never going to be a very strict limit on self-defence because it is mainly understood as a prohibition against excesses. However, recent practice makes it clear that states are in fact making use of the limited restraining potential that proportionality offers. More specifically, the proportionality debates we have witnessed since the late 1990s illustrate that the process of developing guidelines on proportionality is under way. Recent practice suggests that the targeted (or otherwise) character of the response appears to be the key consideration informing proportionality judgments.

Whilst the current situation is far from doctrinally satisfying, it is perhaps at least pragmatically acceptable. After all, it must be borne in mind that the very idea of anti-terrorist self-defence in the sense of Article 51 remains controversial, and that it significantly stretches the traditional understanding. We are witnessing the adaptation of necessity and proportionality to a new reality, with all the uncertainty that such a readjustment process involves. During this process dogmatic uncertainty will not prevent states from pronouncing on the legality of self-defence operations and it is these pronouncements that will form the basis of tomorrow's more specific guidelines.

References

1 UN Charter, art 51, does not mention ‘proportionality’/‘proportionate’ at all, and ‘necessity’/‘necessary’ only in relation to the Security Council: cf art 51(1): ‘… until the Security Council has taken measures necessary to maintain international peace and security’.

2 Legality of the Threat or Use of Nuclear Weapons, Advisory Opinion, 8 July 1996 [1996] ICJ Rep 226 (‘Nuclear Weapons’), [41] (citing Military and Paramilitary Activities in and Against Nicaragua, Merits, Judgment, 27 June 1986 [1986] ICJ Rep 14 (‘Nicaragua’), [176]). The fullest account of the ICJ's jurisprudence on self-defence can be found in Green, James A, The International Court of Justice and Self-Defence (Hart 2009)Google Scholar.

3 Oil Platforms (Islamic Republic of Iran v US), Judgment, 6 November [2003] ICJ Rep 161 (‘Oil Platforms’), [76]. In its work on state responsibility, the International Law Commission (ILC) even suggested that ‘the requirements of proportionality and of necessity [were] inherent in the notion of self-defence’: Responsibility of States for Internationally Wrongful Acts (2001), annexed to UNGA Res 56/83, 12 December 2001, and reproduced in Crawford, James (ed), The ILC's Articles on State Responsibility (CUP 2002)Google Scholar], commentary to art 21, para 6. See further Gray, Christine D, International Law and the Use of Force (3rd edn, OUP 2008) 147Google Scholar; Gardam, Judith Gail, Proportionality, Necessity and the Use of Force in International Law (CUP 2004) 138ffCrossRefGoogle Scholar; Roucounas, ‘Emmanuel, ‘Preliminary Report on Self-Defence’ (2007) Annuaire de l'Institut de droit international 72, 99Google Scholar.

4 Both necessity and proportionality come ‘in a number of different guises and forms’: Lubell, Noam, Extraterritorial Use of Force against Non-State Actors (OUP 2010) 64CrossRefGoogle Scholar. In addition to the jus ad bellum, they operate as restraints under rules governing the conduct of hostilities, or jus in bello. Proceedings such as Nuclear Weapons (n 2) suggest that the two different sets of limitations are not always clearly separated, and frequently an unnecessary use of force may equally violate the jus in bello. That, however, is not necessarily the case, and the two tests follow a different rationale – the jus ad bellum regulates the entitlement and scope to use force, whereas necessity and proportionality under the jus in bello govern the employment of force in concrete military encounters, such as issues of targeting. These latter considerations will largely be left to a side in the following, and taken up only to the extent that they influence the necessity and proportionality tests under the jus ad bellum. For a useful discussion see Cannizaro, Enzo, ‘Contextualizing Proportionality: Jus ad Bellum and Jus in Bello in the Lebanese War’ (2006) 88 International Review of the Red Cross 779CrossRefGoogle Scholar.

5 Nuclear Weapons (n 2) [41].

6 Especially given the position taken by the ICJ in Legal Consequences of the Construction of a Wall in Occupied Palestinian Territory, Advisory Opinion, 9 July 2004, [2004] ICJ Rep 136 (‘Wall’), [139]; cf the Separate Opinion of Judge Higgins, [33]–[35] and the Declaration of Judge Buergenthal, [6], as well as the following voices in the literature: Cassese, Antonio, ‘Terrorism is also Disrupting Some Crucial Legal Categories of International Law’ (2001) 12 European Journal of International Law 993CrossRefGoogle Scholar; Stahn, Carsten, ‘Nicaragua is Dead, Long Live Nicaragua!’ in Walter, Christian and others (eds), Terrorism as a Challenge for National and International Law: Security versus Freedom? (Springer 2004) 827Google Scholar; Kammerhofer, Jörg, ‘The Armed Activities Case and Non-State Actors in Self-Defence Law’ (2007) 20 Leiden Journal of International Law 89CrossRefGoogle Scholar; Tomuschat, Christian, ‘Der 11 September und seine rechtlichen Konsequenzen’ (2001) 28 Europäische Grundrechte-Zeitschrift 535Google Scholar; Murphy, Sean D, ‘Self-Defense and the Israeli Wall Opinion – An Ipse Dixit from the Court?’ (2005) 99 American Journal of International Law 62CrossRefGoogle Scholar and ‘Terrorism and the Concept of “Armed Attack” in Article 51 of the UN Charter’ (2002) 43 Harvard International Law Journal 41Google Scholar; Krajewski, Markus, ‘Selbstverteidigung gegen bewaffnete Angriffe nicht-staatlicher Organisationen’ (2002) 40 Archiv des Völkerrechts 183Google Scholar; Ruys, Tom and Verhoeven, Sten, ‘Attacks by Private Actors and the Right of Self-Defence’ (2005) 10 Journal of Conflict and Security Law 289CrossRefGoogle Scholar; Antonopoulos, Constantine, ‘Force by Armed Groups as Armed Attack and the Broadening of Self-Defence’ (2008) 55 Netherlands International Law Review 159CrossRefGoogle Scholar; Klein, Pierre, ‘Le droit international à l'épreuve du terrorisme’, 321 Recueil des Cours (2006) 396ffGoogle Scholar; Gray (n 3) 193ff); van Steenberghe, Raphaël, ‘Self-Defence in Response to Attacks by Non-State Actors in the Light of Recent State Practice: A Step Forward?’ (2010) 23 Leiden Journal of International Law 183CrossRefGoogle Scholar; Moir, Lindsay, Reappraising the Resort to Force (Hart 2010), especially 22–31Google Scholar; Tsagourias, Nicholas, ‘Non-State Actors in International Peace and Security: Non-State Actors and the Use of Force’ in d'Aspremont, Jean (ed), Participants in the International Legal System: Multiple Perspectives on Non-State Actors in International Law (Routledge 2011) 326Google Scholar.

7 See references in n 6. For Tams' view on the matter see Tams, Christian J, ‘The Use of Force against Terrorists’ (2009) 20 European Journal of International Law 359Google Scholar; Tams, Christian J, ‘Light Treatment of a Complex Problem’ (2005) 16 European Journal of International Law 963CrossRefGoogle Scholar; Tams, Christian J, ‘Swimming with the Tide, or Seeking to Stem It? Recent ICJ Rulings on the Law of Self-Defence’ (2005) 18 Revue Québécoise de Droit International 275Google Scholar; and Bruha, Thomas and Tams, Christian J, ‘Self-Defence against Terrorist Attacks. Considerations in the Light of the ICJ's Israeli Wall Opinion’ in Dicke, K and others (eds), Weltinnenrecht. Liber Amicorum Jost Delbrück (Duncker and Humblot 2005) 85Google Scholar.

8 According to Judge Kooijmans, ‘this has been the generally accepted interpretation for more than 50 years’, which anti-terrorist practice had called into question: see Wall (n 6), Separate Opinion of Judge Kooijmans, [35].

9 Nicaragua (n 2) [109] and [115].

10 See Responsibility of States for Internationally Wrongful Acts (n 3), commentary to art 8(3)–(5).

11 For references cf Gray (n 3) 99–103.

12 In addition to the states involved in the instances mentioned in the text (Israel, US, Turkey), these include, for example, the following: Australia, Russia, Iran, Colombia, and states participating in Operation Enduring Freedom. For references see Tams, ‘The Use of Force against Terrorists’ (n 7) 378–81.

13 See references ibid.

14 See the detailed account by Ruys, Tom, ‘Quo Vadit Jus ad Bellum? A Legal Analysis of Turkey's Military Operations against the PKK in Northern Iraq’, (2008) 12 Melbourne Journal of International Law 334Google Scholar; and further Wettberg, Gregor, The International Legality of Self-Defence Against Non-State Actors (European University Studies 2007) 144–51Google Scholar for details.

15 Ruys (n 14) 344.

16 For example, the Belgian Foreign Minister noted that the Turkish action ‘was precisely targeted and aimed only at PKK targets, without harming the population of northern Iraq or local factions’: Questions et réponses écrites, Chambre des représentants de Belgique, 25 February 2008, QRVA 52 010, 1357; available at http://www.dekamer.be/QRVA/pdf/52/52K0010.pdf. His Dutch colleague observed that ‘the Turkish actions appear to be restricted to specific actions against PKK targets in the border area of northern Iraq’: Maxime Verhagen, ‘Beantwoording vragen van het lid Van Bommel over een Turkse invasie in Noord-Irak’, Ministerial statement, 3 March 2008, as cited in Ruys (n 14) 363.

17 See UN Doc A/60/937-S/2006/515, Letter sent to the Security Council dated 12 July 2006. For a comprehensive account see Ronen, Yaël, ‘The 2006 Conflict in Lebanon’ (2008) 9 Yearbook of International Humanitarian Law 362–93CrossRefGoogle Scholar; see also Secretary-General's Report on UNIFIL, UN Doc S/2006/560, 22 July 2006; and Gray (n 3) 237–44. For a detailed legal analysis cf Zimmermann, Andreas, ‘The Second Lebanon War: Jus ad Bellum, Jus in Bello and the Issue of Proportionality’ (2007) 11 Max Planck Yearbook of United Nations Law 99Google Scholar; and Cannizaro (n 4) 779.

18 States including Russia, Argentina, Japan and France: see UN Doc S/PV.5488, 13 July 2006; UN Doc S/PV.5489, 14 July 2006; UN Doc S/PV.5493, 21 July 2006; UN Doc S/PV.5493 (Resumption 1), 21 July 2006.

19 Van Steenberghe (n 6) 193. During the Security Council meeting of 14 July 2006, only China and Qatar expressly rejected Israel's reliance on self-defence: see UN Doc S/PV 5489. In contrast, Israel's right was recognised by Argentina (ibid 9), Japan (12), the UK (12), Peru (14), Denmark (15), Slovakia (16), Greece (17), France (17) and the USA (17). The UN Secretary General also acknowledged Israel's right to defend itself against Hezbollah attacks under UN Charter, art 51: see UN Doc S/PV.5492, 20 July 2006, 3; UN Doc S/PV.5498, 30 July 2006, 3.

20 For a clear account see Lobel, Jules, ‘The Use of Force to Respond to Terrorist Attacks: The Bombing of Sudan and Afghanistan’ (1999) 24 Yale Journal of International Law 537Google Scholar.

21 UN Doc S/1998/780, 20 August 1998.

22 China, Russia, a number of Arab states and the Non-Aligned Movement condemned it, as did (at least implicitly) UN Secretary General Kofi Annan: cf Schneider, Howard and Boustany, Nora, ‘A Barrage of Criticism in Mideast; US Accused of “State Terrorism”’, Washington Post (21 August 1998) A20Google Scholar.

23 See Lobel (n 20) 556–57, for much more on this point; and Wettberg (n 14) 157–59, for a summary of responses.

24 See the references in Drozdiak, William, ‘European Allies Back US Strikes; Japan Says It “Understands”’, Washington Post (21 August 1998)Google Scholar A20.

25 For references to, and controversial assessments of, many other instances, see the works cited in nn 6 and 7.

26 These are very clearly set out in the works by Gray (n 3) and Klein (n 6) as well as Corten, Olivier, ‘Self-Defence Against Terrorists: What can be Learned from Recent Practice (2005–2010)?’ (2010) 109 Kokusaiho Gaiko Zassi (Journal of International Law and Diplomacy) 129Google Scholar.

27 See the position taken by the ICJ in Wall (n 6) [139].

28 On this compare, for example, Trapp, Kimberley N, ‘The Use of Force against Terrorists: A Reply to Christian J. Tams’ (2009) 20 European Journal of International Law 1049CrossRefGoogle Scholar and Tams, Christian J, ‘The Use of Force against Terrorists: A Rejoinder to Federico Sperotto and Kimberley N. Trapp’, (2009) 20 European Journal of International Law 1057Google Scholar.

29 Nuclear Weapons (n 2) [41] (emphasis added).

30 Gardam (n 3) 148–50; Gray (n 3) 150; Gazzini, Tarcisio, The Changing Rules on the Use of Force in International Law (Manchester University Press 2006)Google Scholar 146.

31 cf the British–American correspondence relating to the Caroline incident (Note from Mr Webster to Mr Fox, 24 April 1841) in 29 BFSP 1129, 1138 (1840–41) (‘the Caroline correspondence’).

32 See Ago, Roberto, Eighth Report on State Responsibility (Addendum), Yearbook of the International Law Commission 1980, vol II/1, 60Google Scholar, para 120: ‘The reason for stressing that action taken in self-defence must be necessary is that the state attacked (or threatened with imminent attack, if one admits preventive self-defence) must not, in the particular circumstances, have had any means of halting the attack other than recourse to armed force. In other words, had it been able to achieve the same result by measures not involving the use of armed force, it would have no justification for adopting conduct which contravened the general prohibition against the use of armed force. The point is self-evident and is generally recognized; hence it requires no further discussion’ (emphasis added).

33 This latter qualification is frequently overlooked. See, however, the very clear formulation in Principle 3 of the Chatham House Principles: Wilmshurst, Elizabeth, ‘Principles of International Law on the Use of Force by States in Self-Defence’, Chatham House, ILP WP 05/01, October 2005; reproduced in (2006) 55 International and Comparative Law Quarterly 963 (‘Chatham House Principles’)CrossRefGoogle Scholar.

34 See commentary to Principle 3 of the Chatham House Principles, ibid.

35 Dinstein, Yoram, War, Aggression, Self-Defence (4th edn, CUP 2005) 237CrossRefGoogle Scholar; Schachter, Oscar, International Law in Theory and Practice (Martinus Nijhoff 1991) 152Google Scholar.

36 This approach has remained controversial, but would seem to follow from the ICJ's jurisprudence in Nicaragua (n 2) [191], Oil Platforms (n 3) [51] and [62], and Armed Activities on the Territory of the Congo (Democratic Republic of Congo (DRC) v Uganda) Judgment, 19 December 2005 [2005] ICJ Rep 201 (‘DRC v Uganda’) [147]). See also the Eritrea/Ethiopia Boundary Commission's Award on Ethiopia's Jus ad Bellum Claims 1–8, 19 December 2005 (2006) 45 ILM 430. For details see Green (n 2) 31–41; and for a useful summary van Steenberghe (n 6) 202–04.

37 This implies a general statement on the relationship between self-defence and collective security measures, expressing the Charter's preference for ‘institutional’ action over ‘private’ reactions: cf Stanimir Alexandrov, A, Self-Defense against the Use of Force in International Law (Kluwer 1996) 104–05Google Scholar and 130ff for details and references.

38 The point is much neglected. For comment see Dinstein (n 35) 215–16; Gray (n 3) 126–28.

39 Dinstein (n 35) 216.

40 Hence the Security Council's determination, in SC Res 502 (1982), that Argentina's invasion of the Falklands/Malvinas amounted to a breach of the peace was as such not sufficient to disapply the UK's right of self-defence. See further Gray (n 3) 125.

41 UNSC Res 661(1990), UN Doc S/RES/661, 6 August 1990.

42 This in fact was already envisaged by Kelsen, Hans, The Law of the United Nations (2nd edn, Stevens & Sons 1951) 800–02Google Scholar.

43 See SC Res 1701(2006), 11 August 2006, which required Hezbollah to cease all attacks and Israel to refrain from ‘all offensive military operations’ (emphasis added). For debates about the scope of this disapplication of self-defence see UN Doc S/2007/147, 14 March 2007, at 3–8; for a discussion of the status of the resolution (which did not mention ‘Chapter VII’ directly but used the language of art 39) see Lapidoth, Ruth, ‘The Legal Effect of Security Council Resolution 1701 (2006) on Cessation of Hostilities in Lebanon’ in Fischer-Lescano, Andreas and others (eds), Frieden in Freiheit = Peace in liberty = Paix en liberté: Festschrift für Michael Bothe zum 70. Geburtstag (Nomos/Dike 2008)Google Scholar 187).

44 Noam Lubell, ‘Use of Force against Individuals and Groups Abroad’, PhD thesis, University of Essex (2008) at 47. The formulation does not seem to be used in Lubell (n 4).

45 cf Caroline correspondence (n 31).

46 See Principle 6 of the Chatham House Principles (n 33): ‘If the right of self-defence … is to be exercised in the territory of another state, it must be evident that that state is unable or unwilling to deal with the non-state actors itself, and that it is necessary to use force from outside to deal with the threat in circumstances where the consent of the territorial state cannot be obtained’.

47 This is brought out implicitly through the fact that in order to justify their conduct, states asserting a right to use self-defence against terrorist attacks stressed that host states had taken no action. Hence in order to justify their conduct in the instances mentioned above in Section 2 ‘Background’, Turkey and the US asserted that the respective host states affected by their operations had been unwilling or unable to prevent the terrorist threat or attack. Israel went further in that it alleged (controversially) that Lebanon had supported Hezbollah's activities and therefore bore responsibility.

48 Principle 6 of the Chatham House Principles (n 33); and further Ruys and Verhoeven (n 6) 316.

49 Paust, Jordan J, ‘Use of Force against Terrorists in Afghanistan, Iraq and Beyond’ (2002) 35 Cornell International Law Journal 533, 540–43Google Scholar.

50 Ruys and Verhoeven (n 6) 317.

51 cf Sarah Rainsford, ‘Iraq Troop Withdrawal Baffles Turks’, BBC News (UK) 29 February 2008, available at http://news.bbc.co.uk/2/hi/europe/7272108.stm.

52 See, for example, Letter from the Permanent Representative of the Russian Federation to the United Nations addressed to the Secretary-General, UN Doc S/2002/1012, 11 September 2002 (seeking to justify Russia's use of force on Georgian territory and claiming that ‘[t]he continued existence in separate parts of the world of territorial enclaves outside the control of national governments, which, owing to the most diverse circumstances, are unable or unwilling to counteract the terrorist threat is one of the reasons that complicate efforts to combat terrorism effectively’). Iran has on several occasions invoked UN Charter art 51 to justify the use of force against bases of the Mujahedin-e Khalq Organization (MKO) on Iraqi territory – see Wandscher, Christiane, Internationaler Terrorismus und Selbstverteidigungsrecht (Duncker & Humblot 2006) 148–49Google Scholar; Wettberg (n 14) 151–52. While Iraq denounced the use of force as an act of aggression, the international community did not condemn it. Similarly, Iran's incursions into Iraqi territory in pursuit of Kurdish armed bands (labelled ‘organized terrorist mercenaries’) avoided international criticism: see Franck, Thomas, Recourse to Force (CUP 2002)CrossRefGoogle Scholar 64. There was little evidence to suggest that the conduct of the MKO (let alone that of Kurdish insurgents) could have been attributed to Iraq.

53 See explanation to Principle 6 of the Chatham House Principles (n 33).

54 See UNGA Res 2625(XXV), 24 October 1970, Principle I (10); ICJ, Corfu Channel Case, Judgment, 9 April 1949, [1949] ICJ Rep 4.

55 cf art 1(c)(xi): ‘The following shall constitute acts of aggression: (…) the encouragement, support, harbouring or provision of any assistance for the commission of terrorist acts and other violent transnational organized crimes against a Member state’; available at http://www.africa-union.org/root/au/documents/treaties/text/Non%20Aggression%20Common%20Defence%20Pact.pdf.

56 Notably in the case of the Taliban and Al Qaeda: see, for example, UNSC Res 1214(1998), UN Doc S/RES/1214, 8 December 1998; and UNSC Res 1267(1999), UN Doc S/RES/1267, 15 October 1999.

57 For example, the specific obligations relating to the suppression of PKK activities under the 1999 Washington Accord. For details see Ruys (n 14).

58 nn 17–19.

59 See, for example UN Doc S/2002/1012 (n 52) in respect of Russia's claim referred to above; and cf Tams, ‘The Use of Force against Terrorists’ (n 7) 379–81 for a summary of similar claims.

60 A point rightly stressed in the comment to Principle 6 of the Chatham House Principles (nn 33 and 46).

61 See below, Section 6 ‘Anti-Terrorist Proportionality’.

62 A point rightly stressed by Lobel (n 20) 552–53.

63 Roucounas (n 3) 99.

64 Trapp, Kimberley N, ‘Back to Basics: Necessity, Proportionality, and the Right of Self-Defence against Non-State Terrorist Actors’ (2007) 56 International and Comparative Law Quarterly 141, 146CrossRefGoogle Scholar; Green (n 2) 86; Lubell (n 4) 63–64.

65 See Delbrück, Jost, ‘Proportionality’ in Bernhardt, R (ed), Encyclopedia of Public International Law, vol III (1997), 1140Google Scholar; and the detailed treatment by Franck, Thomas, ‘On Proportionality of Countermeasures in International Law’ (2008) 102 American Journal of International Law 715CrossRefGoogle Scholar.

66 A point rightly stressed by Lubell (n 4) 63–64.

67 See Nuclear Weapons (n 2) [42]; Lubell (n 4), 63–64; and cf above n 4.

68 For the ICJ's approach see the following two statements made in the Nicaragua and DRC v Uganda cases: (i) ‘Whatever uncertainty may exist as to the exact scale of the aid received by the Salvadorian opposition from Nicaragua, it is clear that these … United States activities … could not have been proportionate to that aid.’ (Nicaragua (n 2) [237]); (ii) ‘[T]he taking of airports and towns many hundreds of kilometres from Uganda's border would not seem proportionate to the series of transborder attacks it claimed had given rise to the right of self-defence’ (DRC v Uganda (n 36) [147]). For comment, see notably Christodoulidou, Theodora and Chainoglou, Kalliopi, ‘The Principle of Proportionality in Self-Defence and Humanitarian Intervention’ (2007) 20 Humanitäres Völkerrecht 79, 83Google Scholar: ‘The above analysis of [ICJ jurisprudence] seems to argue that the ICJ's approach to proportionality, albeit epigrammatically at times, is to strike a balance between the attack and the wrong provoking it.’ See also Frederic L Kirgis, ‘Some Proportionality Issues Raised by Israel's Use of Armed Force in Lebanon’, ASIL Insight No 10/2006: ‘These ICJ pronouncements focus primarily on weighing the intensity of the defensive force against the intensity of the force defended against.’ Available at http://www.asil.org/insights060817.cfm#_ednref3.

69 Cannizaro (n 4) 785.

70 The point may be illustrated by reference to the Falklands/Malvinas conflict of 1982, during which the British self-defence operation led to the loss of 907 lives, compared to 0 lives lost during Argentina's invasion; and still it was not seriously considered to be disproportionate.

71 Principle 5 of the Chatham House Principles (n 33).

72 For details and useful comment see notably Gardam (n 3) 159–79.

73 cf O'Connell, Daniel Patrick, The Influence of Law on Sea Power (Manchester University Press 1975)Google Scholar 64.

74 cf Wettberg (n 14) 144–51, for details. cf Rainsford (n 51).

75 cf Cabinet Communiqué of 16 July 2006: ‘Prime Minister Olmert said that we would continue in this campaign as necessary and added that it is clear to all that we will not stop until we can tell the Israeli people that the threat hanging over it has been removed.’ Available at http://www.mfa.gov.il/MFA/Government/Communiques/2006/Cabinet+Communique+16-Jul-2006.htm.

76 Gray (n 3) 197. See Lobel (n 20) 537, for details.

77 See Gill, TD, ‘The Temporal Dimension of Self-Defence: Anticipation, Pre-Emption, Prevention and Immediacy’ (2006) 11 Journal of Conflict and Security Law 361CrossRefGoogle Scholar; Hamid, Abdul Ghafur, ‘The Legality of Anticipatory Self-Defence in the 21st Century World Order’ (2007) 54 Netherlands International Law Review 441CrossRefGoogle Scholar; and Gray (n 3) 160ff.

78 Whether such an ‘accumulation of events’ triggers a right of self-defence is much discussed: for a clear summary of debates see, for example, Gray (n 3) 155; and cf the ICJ's statement, in Oil Platforms (n 3) [64] that ‘even taken cumulatively’ a series of incidents did not qualify as an armed attack on the United States. Since the question primarily concerns the interpretation of the ‘armed attack’ requirement, it is left to a side here. It should, however, be noted that recognition of the doctrine would help to avoid the sometimes artificial separation of multiple attacks into separate incidents; on the other hand, it would risk turning self-defence into an open-ended licence to use counterforce.

79 Daniel Bethlehem, ‘International Law and the Use of Force: The Law As It Is and As It Should Be’, written evidence to the Select Committee on Foreign Affairs, Minutes of Evidence, 8 June 2004, para 21; available at http://www.publications.parliament.uk/pa/cm200304/cmselect/cmfaff/441/4060808.htm.

80 cf Gardam (n 3) 187.

81 See Section 4, ‘Anti-terrorist necessity’.

82 For much more on this process see Franck (n 65).

83 cf Gray (n 3) 154, who notes that proportionality has come to be seen as a convenient yardstick against which states' responses can be assessed – precisely so as to avoid complex debates about the availability of self-defence: ‘In Security Council debates states have thus been able to avoid going into doctrinal disputes as to whether self-defence is wide or narrow; they can simply say that the use of force was not necessary or proportionate and therefore illegal.’

84 The Turkish military claimed that approximately 540 PKK members were killed during the bombardments and the ground offensive, which also led to the destruction of 800 shelters, weapon stores and other PKK positions.

85 See n 16.

86 In fact, it is telling that after a detailed study, one commentator observed that ‘the Turkish intervention poses few problems in terms of proportionality’: Ruys (n 14) 344.

87 Trapp (n 64) 155.

88 For statements to this effect see, for example, UN Doc S/PV.5492, 20 July 2006, 3 (Secretary-General) or UN Doc S/PV.5489, 30 July 2006, 9 (Argentina).

89 Franck (n 65) 764: ‘Dispute by dispute and adjudication by adjudication … international law is developing rather like the common law, broadening precedent by precedent.’