Some Reflections on Self-defence as an Element in Rules of Engagement

Abstract

From 16 to 20 June 2007, the International Security Assistance Force (ISAF) and the Taliban were engaged in a fierce battle over Chora, Afghanistan, resulting in many civilian casualties in and around that capital city. ISAF is a coalition of states established to contribute to the maintenance of security, but which through their frequent engagement in actual warfare have become parties to the armed conflict in Afghanistan. As a result, their actions are governed by international humanitarian law. This includes the prohibition of indiscriminate attacks, i.e. attacks expected to cause civilian casualties at a level excessive in relation to the military advantage anticipated. The hostilities in and around Chora have given rise to the question whether they might have violated this prohibition (a question ultimately answered in the negative). In this debate, self-defence was among the arguments raised in justification. Self-defence usually figures as a standard clause in the rules of engagement. These are texts which, established by commanders, permit or limit the use of force by their armed forces. The chapter briefly discusses the character of these instruments and of the clauses they contain. The focus is in particular on the self-defence clause. Self-defence may be individual or collective, and it may arise on three different levels: as national self-defence, unit self-defence or individual self-defence. National self-defence is the right for states to defend themselves against an attack or imminent attack. Unit self-defence is a notion generally accepted in military practice without having a firm legal basis in most countries. In contrast, individual self-defence is recognised in every domestic legal system. In the closing chapter, the chapter focuses on the relevant Dutch legal system, because the troops involved in the battle over Chora were Dutch forces and collective unit self-defence might have been at issue as an exculpatory argument in that case.

Frits Kalshoven is Professor Emeritus of Public International and Humanitarian Law, Leiden University, The Netherlands. Thyla Fontein is Master of Public International Law, Leiden University.

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

1 Introduction

It is a sad occasion when a collection of chapters is produced in memory of a departed colleague. The loss is all the more poignant if, as in the present case, the person who passed away was in full bloom and everything could still be expected of her. The first author has long had the privilege of knowing and working with Avril McDonald. He is grateful, both, for the opportunity to contribute to this liber in her name, and to carry out this work in cooperation with a promising young lawyer who is at the very beginning of her career.

From 16 to 20 June 2007, the International Security Assistance Force (ISAF) and the Taliban were engaged in a fierce battle over Chora, the capital town of Uruzgan province in Afghanistan. While the Taliban, in its attempt to capture the town, killed a great number of civilians, the counter-attack by an ISAF battalion caused many civilian casualties as well. In the debate that followed, one argument was that the ISAF actions could be justified on the grounds of self-defence, a notion that figures as a standard clause in armed forces’ rules of engagement. In this chapter, we discuss this notion of self-defence and its possible role in legitimising or justifying the use of force in this type of situation.

ISAF, a coalition of states, was set up in December 2001 in reaction to the September 11 attacks on the World Trade Center in New York and the Pentagon in Washington. In an even more immediate reaction, the United States-led Operation Enduring Freedom (OEF) started in October 2001 as a full-fledged fighting force acting in individual and collective self-defence,¹ with two objectives: to disrupt the Al Qaeda bases in Afghanistan and to remove the Taliban de facto government from power.

OEF rapidly achieved the second objective. With the Taliban de facto government removed from Kabul and an Afghan Interim Authority set up in its place, the Security Council on 20 December 2001 authorised the establishment of ISAF, mandating it to assist the Authority in “the maintenance of security in Kabul and its surrounding areas”.² As distinct from OEF, ISAF was not established to fight its own war. As a security force, it was designed at the outset to act in support of the Afghan authorities in and around Kabul in their efforts to ‘maintain security’. To this end, the Security Council authorised ISAF to “take all measures necessary to fulfil its mandate”.³ While the clause ‘all measures necessary’ in a Security Council resolution generally implies authorisation to use lethal force, the quoted phrase limits this to such measures as might become necessary to ‘assist in the maintenance of security’. This task was likely to involve a level of force higher than that allowed in a classical peacekeeping operation, but was intended to remain below the level of all-out warfare. This suited participating governments, who wanted to avoid the suggestion that they might be involved in an armed conflict, let alone one linked to the U.S. ‘War on Terror’.⁴

The situation changed in 2003; in August, NATO assumed the leadership of ISAF,⁵ and in December, the Security Council expanded ISAF’s mandate to cover other Afghan regions.⁶ From that moment onwards, ISAF took over command in an increasing number of Afghan regions, and in July 2006 in the southern region. This included Uruzgan,⁷ a province which at the time was regarded as relatively peaceful, enabling the Task Force Uruzgan (at the time, a Dutch battalion) to engage in numerous activities aimed at improving the living conditions of the local populace. However, the situation did not remain stable, witness the fierce battle waged in June 2007 to prevent the Taliban from capturing the capital. In this ‘battle of Chora’ all kind of weaponry was used, including heavy artillery and long-distance air support. The numerous civilian casualties ISAF’s action had entailed ran counter to what may be the essence of the type of ‘counter-insurgency operation’ it was engaged in, which is to “remain friendly towards the populace while staying vigilant against insurgent actions”.⁸

The question arises whether this state of affairs warranted the conclusion that international humanitarian law (IHL) had become applicable. A first point to note here is that in 2001, the invasion of Afghanistan by the U.S.-led OEF had brought the partners of that coalition into a situation of international armed conflict, with Al Qaeda and Afghanistan (at the time, the Taliban and the Afghan army) as opponents.⁹ In contrast, and although ISAF was established at about the same time as OEF, this body was not set up as a participant to the conflict.

In 2003, with the Taliban ousted from power, the situation in the country changed into one of internal armed conflict with, in jus ad bellum terms, the Taliban in the offence and the Afghan authorities in the defence.¹⁰ ISAF, for its part, assisted the authorities in the maintenance of security. While, as noted before, this could be a fairly peaceful activity, hostilities flared up from time to time, culminating in a case like Chora where ISAF undertook to defend the town against a very determined Taliban attack; a case of open warfare. Given these circumstances, as a matter of jus in bello, ISAF members had become parties to the armed conflict as well.¹¹ Note that qualification of a situation as an armed conflict does not require continuous warlike activity, nor must a party to an armed conflict be set on destroying its opponent: defence is warfare as well.

A next question is whether the conflict the ISAF members had become parties to, amounted to an international or non-international armed conflict. On this, we may rely on the International Committee of the Red Cross (ICRC), which in a recent report notes the emergence of a category of conflicts it indicates as “multinational NIACs” (non-international armed conflicts):

[A]rmed conflicts in which multinational armed forces are fighting alongside the armed forces of a “host” state – in its territory – against one or more organized armed groups. As the armed conflict does not oppose two or more states, i.e. as all the state actors are on the same side, the conflict must be classified as non-international, regardless of the international component, which can at times be significant. A current example is the situation in Afghanistan (even though that armed conflict was initially international in nature).¹²

The ICRC adds that in this case, “[t]he applicable legal framework is Common Article 3 and customary IHL”. Article 3 common to the four Geneva Conventions of 1949 provides basic standards of human behaviour for any situation of armed conflict, but these do not concern actual warfare.¹³ However, the reference to ‘customary IHL’ is meant to cover this lacuna: here, the ICRC has its study on ‘Customary International Humanitarian Law’ in mind.¹⁴ The study identifies 161 rules, with the majority stated to be applicable in non-international armed conflicts as well.

All of this leads to the conclusion that, in theory, Common Article 3 combined with customary law and, in practice, the law of international armed conflict, are indeed applicable to those ISAF activities that were actually governed by IHL. This includes an operation like the defence of Chora, but also in ostensibly more peaceful times, the frequent search for IED, the ill-famed ‘improvised explosive devices’ or the forceful search of houses believed to shelter persons engaged in planting such devices, yet another ‘defensive’ activity that has entailed numerous civilian casualties.

It was precisely after the Chora incident that the self-defence argument cropped up. As mentioned at the outset, it was in this context that the phenomenon of rules of engagement (RoE) entered into play: documents in which self-defence holds a prominent place.

2 Rules of Engagement

The behaviour of military units on mission is governed by a variety of sources, from treaties and national laws down to the operational order of the day. Somewhere in between hover RoE.¹⁵ These are texts designed inter alia to influence conduct and, in particular, the use of force. They are neither formal lawmaking instruments nor, at the other extreme, do they set the operational targets that units are expected to achieve. The U.S. Dictionary of Military and Associated Terms defines them as “[d]irectives issued by competent military authority that delineate the circumstances and limitations under which United States forces will initiate and/or continue combat engagement with other forces encountered”.¹⁶

Another U.S. publication, entitled ‘Legal Support to Military Operations’, adds that RoE must be “reviewed by legal advisors for compliance with applicable law and policy”.¹⁷ A slightly older manual provides in somewhat broader terms that RoE “may take the form of execute orders, deployment orders, memoranda of agreement, or plans”, and it states that “[p]roperly developed rules of engagement fit the situation and are clear, reviewed for legal sufficiency, and included in training […] Rules of engagement vary between operations and may change during an operation. Adherence to them ensures Soldiers act consistently with international law, national policy, and military regulations”.¹⁸

Rules of engagement are widely used, by states individually and, with the growing frequency of combined operations such as OEF and ISAF, by coalition forces as well. NATO document MC 362/1, NATO Rules of Engagement, broadly defines RoE as “directives to military forces (including individuals) that define the circumstances, conditions, degree, and manner in which force, or actions which might be construed as provocative, may be applied”.¹⁹ The Sanremo Handbook on Rules of Engagement,²⁰ a publication of the Sanremo International Institute of Humanitarian Law, stays closer to the military language of the U.S. Dictionary when it defines RoE as texts “issued by competent authorities and [that] assist in the delineation of the circumstances and limitations within which military forces may be employed to achieve their objectives”. It adds that “[w]hatever their form, [RoE] provide authorisation for and/or limits on, among other things, the use of force, the positioning and posture of forces, and the employment of certain specific capabilities”.²¹ RoE “are not used to assign missions or tasks [or] to give tactical instructions”, such matters “are assigned through Operations orders and other similar instruments of command and control”.²²

The Handbook notes that “[t]he conduct of military operations is governed by international law, including the law of armed conflict (LOAC)”. It adds that “[b]oth nations and individuals are obliged to comply with LOAC”.²³ That may be so, but current history provides many examples of parties to armed conflicts, and especially armed opposition groups, that order their forces to use violence against the civilian population in general or specified (categories of) civilians, or against objects under special protection, such as hospitals or vehicles provided with the Red Cross or Red Crescent emblem, all of this in blatant violation of the law.²⁴ It does not seem unlikely that such parties state their intentions in texts comparable to the RoE of the Sanremo Handbook. At all events, it should be noted that respect of the law is not an inherent element of RoE.

The Sanremo Handbook refers to the reverse possibility, of RoE prescribing conduct that remains below the upper limit of lawful violence: “in some circumstances, [states] may limit permissible levels of incidental injury or collateral damage to levels below that acceptable under LOAC […]”. It accordingly “allows for the creation of RoE that provide for the conduct of operations in compliance with national policy”.²⁵ Depending on its formulation, such a provision may amount to a standing order under the applicable domestic military law.

The Handbook distinguishes between many different situations, each with their own peculiarities regarding permissible or advisable recourse to violence, and lethal force in particular. Rather than entering into these many possibilities, we refer the reader to the long list of detailed suggestions set forth in Annex B, the ‘Compendium of ROE’.²⁶ We may recall the two situations mentioned in the Introduction: OEF, a full-fledged armed conflict, and ISAF, a security operation.²⁷

ISAF’s operations in support of a friendly government have led to frequent review of the RoE, in particular, as the political need to avoid casualties among the civilian population grew. It should be noted here that adaptability to changes in the physical or political field is a general characteristic that sets RoE apart from treaties and other sources of international law regulating the conduct of war.

3 Self-defence as a Key Element in Rules of Engagement

As briefly mentioned in Part I, a predominant feature of many contemporary RoE is the insistence on self-defence. The very first point the addressees may note is that they have the right to self-defence; and, as if this were not enough, the text may close with a repetition of the reminder of their right to self-defence. A soldiers card issued in December 1992 to members of a U.S. unit engaged in a relief operation in Somalia notes that “[n]othing in these Rules of Engagement limits your right to take appropriate action to defend yourself and your unit”; goes on to confirm that “[y]ou have the right to use force to defend yourself against attacks or threats of attack”, and, removing even the last conceivable bit of doubt, recalls that the addressee must always be “prepared to act in self-defense”.²⁸

An annex to the Sanremo Handbook provides model RoE cards which summarise “the key ROE principles regulating the use of force by individuals for a particular mission”.²⁹ Cards are provided for three distinct situations involving military action indicated as ‘self-defence’, ‘peace operations’ and ‘armed conflict’, with the level of permitted force increasing with every step on the ladder. Self-defence refers to the classical peacekeeping mission that does not in and of itself include active recourse to force; peace operations constitute the middle category that includes today’s notions of security assistance and counter-insurgency, and armed conflict refers to the situation of plain warfare.

The model RoE cards each open with the identical message that “nothing in your ROE limits your right to take action in self-defence”, adding that only such force may be used as is “necessary to neutralise the threat”, this includes deadly force.³⁰ The Handbook distinguishes three types of self-defence: national self-defence, unit self-defence and individual self-defence. National self-defence is “the defence of a nation, a nation’s armed forces, and a nation’s persons and their property”; unit self-defence is “the right of unit commanders to defend their unit, other units of their nation, and other specified units” and individual self-defence is “the right of an individual to defend himself or herself (and in some cases other individuals)”.³¹

What do these clauses stand for; what practical significance can they have and what law, if any, do they refer to? Here, some general points may be noted at the outset. First, the fact that a RoE is in force presupposes the existence of one of the three types of situation, mentioned above, where the military may have an active role to play. And, second, recourse to any right of self-defence in a legal procedure can only arise if the claimant stands accused of an act in contravention of a rule of (international or domestic) law or other norm-setting provision, for instance, the applicable RoE.

Taking the right of national self-defence first, this obviously refers to the right of the state to defend itself against foreign armed attacks, as provided in Article 51 of the UN Charter. Individual claims might conceivably arise if the military authorities of a country at war stand accused of having made the armed forces conduct hostilities in flagrant violation of IHL; the argument might be that the war would otherwise have been lost. At the International Criminal Court, the claim would be expressly excluded; Article 31 (1b) of the Statute provides that “[t]he fact that [a] person was involved in a defensive operation conducted by forces shall not in itself constitute a ground for excluding criminal responsibility”. Apart from that, the treaties in force make the point clear that the law must be respected “in all circumstances”, including the case where the war has taken a wrong turn.³²

Other scenarios might be envisaged where lower personnel of a country at war claim that an act in violation of the law or of the applicable RoE was committed in defence of the nation. However, the chance that the claim is honoured in a court of law seems slim indeed, given the lack of a basis in existing criminal legislation.

In effect, the real significance of the national self-defence clause may lie outside the legal realm, conveying the message that “all is well, you are fighting for a just cause”. This may actually reflect the internationally recognised legitimacy of an operation. Where authorisation was not sought (because the state has decided to commit an act of aggression) or has been withheld (as with the invasion of Iraq in 2003 by the U.S.-led ‘coalition of the willing’), the message that “you have the right of self-defence” may even more clearly act as a moral boost, proclaiming that “the world may have refused our claim but we are doing the right thing”.

The next item on the list, unit self-defence, leads to similar questions. There is no doubt that the notion of ‘unit’ has a strong resonance in many armed forces. Company commanders may have become friends at the military school and soldiers probably were in the same boot camp. Units of these lower echelons may be strongly inclined to assist each other; and if this results in a violation of their RoE, the excuse that it was a matter of collective unit defence may lie readily at hand. Again, however, unit self-defence appears not to have found a place in many existing criminal legislations³³ and, again, the true significance of the right of unit self-defence may lie in its cohesive power.³⁴

There remains the right of individual self-defence as specified on a soldiers card, i.e. self-defence by individual members of armed forces.³⁵ Here, the situation is the opposite of the two preceding cases: the claim of individual self-defence in justification of an unlawful act is as universally recognised as individual claims of national or unit self-defence are unknown. The facts underlying claims arising in the context of military operations will obviously differ from what normally happens on the domestic scene. Even so, it is generally accepted that in the military sphere, for a claim of self-defence to be accepted, it must have been both necessary as a last resort and proportionate to the perceived threat. Indeed, the RoE itself may include these requirements in the phrase formulating the individual right of self-defence.

Another matter is that domestic legislation and practice on self-defence are not identical in all countries. We will not discuss these differences, confining us further down to the situation in the Dutch legal system. It remains to consider here in what circumstances recourse to the right of self-defence will be necessary.

Take the case of an ISAF soldier who, on a mission expected to involve the use of force, advances in terrain under Taliban control: if he receives and returns fire, this is just warfare, and there is no place for a claim of individual self-defence. However, when the soldier kills an unidentified person who turns out to have been an unarmed civilian, the soldier is guilty of an unlawful killing and, if prosecuted, will escape punishment if his appeal to the right of self-defence is honoured.

The general conclusion is that a claim of individual self-defence in a warlike situation may enter into play when a soldier carries out an act which, whether or not part of the task he was ordered to perform, appears to have resulted in a violation of the applicable law of armed conflict. Whether the claim is honoured ought to depend exclusively on the facts of the case. This should in principle be a matter for a court to decide. In practice, it is equally possible that as a matter of policy, claims of self-defence are accepted at an earlier stage without all that much formal investigation, in order thus to keep cases out of the public eye.

4 Self-defence and Rules of Engagement in Dutch Law

As mentioned in the Introduction, on 16 June 2007, Taliban forces attacked the Chora district of Uruzgan province and primarily Dutch ISAF troops employed artillery shelling and airstrikes to regain control over the district, resulting in numerous civilian casualties. Following investigations into the battle over Chora by the Dutch Government, UN Assistance Mission in Afghanistan, Afghanistan Independent Human Rights Commission and others, various reports suggest evidence that the acts may have amounted to violations of Common Article 3 of the Geneva Conventions and questioned whether less damaging methods could have been used. However, they generally concluded that the actions were in accordance with IHL, and the Dutch Ministers of Foreign Affairs, Defence and Cooperation Development stated that defending Chora fell within ISAF’s mandate and was permitted based on the right to self-defence and the RoE.³⁶ In this way, self-defence was advanced as justification for the acts committed by the Dutch forces and mentioning the right to self-defence might have caused confusion as to which level of self-defence, as described above, was meant. With respect to national self-defence, it is accepted that the Netherlands, as contributing nation to NATO, was acting in collective self-defence through permission of the Security Council.³⁷ In order to understand such claim as individual self-defence, one must understand it within the Dutch legal system.

Dutch authorities have exclusive criminal jurisdiction over Dutch ISAF personnel and these are therefore immune from arrest or detention by Afghan authorities.³⁸ This is determined in the Military Technical Agreement signed by the Afghan Interim Administration and ISAF in 2002.³⁹ Moreover, when making status of forces agreements, the Minister of Defence ensures that the Netherlands has exclusive criminal jurisdiction over its soldiers. This jurisdiction is also established in Dutch legislation, and thus the Netherlands has jurisdiction over all criminal acts of Dutch soldiers, wherever they take place.⁴⁰

In the Netherlands, the International Crimes Act,⁴¹ the Dutch Military Criminal Code,⁴² and the Dutch Criminal Code⁴³ make punishable the unlawful use of force by soldiers. When Dutch military forces operate in an international coalition such as ISAF, the legitimacy of the use of force is also dependent on the mandate set by the coalition or organisation. ISAF has a mandate to use force under Chapter VII of the UN Charter and thus also falls under the authority of the UN Security Council. As mentioned earlier, in 2003, NATO took command of ISAF and set the RoE for ISAF. When an unlawful use of force is suspected, there are different ways in which self-defence, independently or as stated in RoE, can function as a justification or excuse for the violation under Dutch law. Section 1 of the Dutch Criminal Code consists of general provisions including the grounds excluding criminal responsibility, which also apply to offences in the Dutch Military Criminal Code, as long as the law does not provide otherwise.⁴⁴ At the most basic level, a Dutch soldier, like anybody else, is always able to claim his individual right to self-defence based on Article 41⁴⁵ of the Dutch Criminal Code, as reflected in the self-defence clause in the RoE.

Another ground excluding criminal responsibility can be found in Article 42⁴⁶ of the Dutch Criminal Code, which justifies an act when it is reasonably necessary in order to execute a legal duty. Similarly, if RoE were considered an official order from a Dutch authority,⁴⁷ these could also lead to the exclusion of criminal responsibility pursuant to Article 43⁴⁸ of the Dutch Criminal Code, if the act was reasonably necessary to carry out that official order.⁴⁹ There was extensive discussion in the Netherlands and there remains disagreement on the status of RoE and tactical directives issued by foreign military authorities, and whether they are considered legal provisions, legal orders or something else.⁵⁰ The suggestion was raised to allow foreign military authorities to gain the same legal position as a Dutch commanding officer that decides on the RoE, through Article 75(a)⁵¹ of the Dutch Military Criminal Code.⁵²

In one case, the Court of Appeals in Arnhem decided that the applicable RoE fell “within the scope of the term ‘military order’ pursuant to Article 135”⁵³ of the Dutch Military Criminal Code.⁵⁴ However, it did not decide that all RoE be considered legal orders.⁵⁵ In the case before us, the RoE in force for ISAF at the time would have to be analysed and a judge would have to decide whether these meet the demands of Article 135. If they would, then violating the RoE would be unlawful but might be justified by the self-defence clause therein.

Article 38⁵⁶ of the Dutch Military Criminal Code offers a solution to the issues presented by the previously discussed articles. When combining this article with Article 71,⁵⁷ an otherwise unlawful act committed during an armed conflict that is in conformity with IHL and within the limits of a soldier’s competence cannot be punished. Since ISAF was operating during an armed conflict as recognised by the government of the Netherlands, this article is of great significance. It can be used as a defence to exclude criminal responsibility, and it thereby offers protection for soldiers during armed conflicts, in which circumstances are different than normal and urgent decisions may have to be made without time to discuss alternative options. In the event that the RoE would permit a more extensive use of force than permitted by IHL, IHL would serve as the upper limit, as decided in this provision. Paragraph 2 of Article 38 is the most significant provision. This paragraph came into force in November 2010 and has retroactive effect on the basis of Article 1 of the Dutch Criminal Code.⁵⁸ It excludes punishment when a soldier uses force in the legitimate exercise of his tasks and competency, and in compliance with the rules set for the exercise of his tasks. This article includes orders and instructions, especially the RoE, aide-mémoire and soldiers card issued by foreign military authorities.⁵⁹ This provision also covers situations that are not recognised as an armed conflict and are thus not covered by para 1.⁶⁰ This means that RoE can be used to exclude criminal responsibility when they have been properly followed. Jurisprudence with respect to this article still needs to develop; however, the need for a stricter rule than how it currently stands might increase.

It seems that Article 38(2) underlies the decision of the public prosecutor not to prosecute Dutch forces for the incident in Chora. Although some of the investigations and reports suggest that there is evidence that the acts may have amounted to violations of Common Article 3 of the Geneva Conventions,⁶¹ the Office of the Public Prosecutor decided to terminate its investigation. It concluded that the use of force was exercised in accordance with IHL, the applicable RoE and the right to self-defence.⁶² The ministers also stated that the use of airplanes and helicopters was in accordance with ISAF rules and procedures.

In all these military cases, there remains a problem regarding prosecution. Decisions to prosecute are still very much a political choice, because the public prosecutor makes this decision together with the minister of defence. In cases where the involved military authority is against a criminal procedure, because it may be in conflict with the interests of the armed forces, the minister of defence resolves this disagreement with the minister of justice. Specific cases might not be dealt with due to various interests involved, especially the protection of a country’s reputation and preventing bad publicity.

5 Conclusion

Having looked at the rules of engagement and self-defence as a justification for the battle over Chora, one can see that this claim could have been based on various grounds. National self-defence is undoubtedly not applicable in this case. Unit self-defence on the other hand, can be argued, but due to the lack of a legal basis, would be unlikely to succeed. Lastly, individual self-defence could have been argued, but due to the fact that the acts occurred in an armed conflict, allows international humanitarian law to be the governing set of rules. The significance of self-defence clauses in RoE lies in the fact that they simply repeat the inherent right of soldiers to protect themselves and thereby reflects domestic rules on self-defence. In the Netherlands as in other countries, self-defence is recognised in its Criminal Code and is not notably different to other countries. The status of the RoE, however, has developed within the Dutch legal system and has been given a fundamental legal status in 2010. This step was finally taken after heavy discussion on their foundation and role in the Dutch legal system. As a result of the newly implemented article, the rules of engagement, even those issued by foreign military authorities, are now recognised as a legitimate body of rules to be followed by Dutch soldiers, even though rules of engagement are changeable.