Military forces involved in an armed conflict will routinely face attacks or threats of attack, and when the attacker is a lawful target under loac, the use of force in response to that threat or attack is regulated by loac. It would therefore be expected that the use of force in response to such threats or attacks would be a separate ‘use of force’ category or roe in nato roe doctrine. However, as explained in Section 3.3, which sets out the use of force in response to attacks or imminent attacks by opposing forces as a roe ‘use of force’ category, the MC 362/11 does not contain an express roe for authorising the use of force in these circumstances. This is likely a consequence of the change in roe methodology and the tendency to require all use of force to be authorised by roe or lawful as self-defence even for operations involving participation in an armed conflict. Unfortunately, due to the emphasis on self-defence as the regulating concept for all defensive uses of force, there has been little focus on developing this ‘use of force’ category. The existence of a ‘use of force’ category beyond pre-planned use of force and self-defence is mainly recognised in joint targeting doctrine, where it is emphasised that the targeting procedures do not apply to contact situations, which includes self-defence and combat engagement.2 Furthermore, the differing approaches among nato States to whether the use of force in response to an attack is regulated by self-defence only or may also be regulated by loac, can make it difficult to achieve agreement on how to regulate this ‘use of force’ category in nato roe when the doctrine is updated.
The lack of express roe for responding to attacks or imminent attack by opposing forces may be compensated for by reading an authorisation to respond to threats or attacks into other roe series or creating an appropriate roe in, for instance, the series authorising the use of force in designated operations. For instance, if the roe authorising attack on forces ‘declared hostile’ is available and the threat is considered to originate from persons in that category, this roe may be sufficient. It is even possible to consider the use of force authorised through the operational concept of self-defence set out in Section 8.3.2. This seems to be the most common practice today. However, as explained in Section 3.3, the lack of a clear standard for formulating this authorisation is unfortunate. Without a standardised manner in which force is authorised in these situations, the parameters for the use of force become less predictable and potentially unclear. This is problematic for forces whose lives depend on that authority; for the commander in need of exerting command and control over such a use of force; and, of course, civilians potentially placed at risk by the potential for misuse or excessive use of force. This Section will therefore deal with the use of force in response to attack as a separate category, even though some of the acts it covers are also discussed in Section 8.3.
Because the use of force in response to threats or attacks from the opposing forces is to be expected and forms part of the ordinary war fighting, it is generally regulated by loac. As has been emphasised, loac regulates the use of force in both offence and defence.3 A basic example of the use of defensive force is the use of force in response to an attack by the opposing forces.4 As explained in Chapter 5, the use of force must be directed at lawful targets, those who plan and carry out the attack must take all feasible precautions to verify that the attack is lawful and to avoid causing incidental harm to protected civilians and civilian objects, and, where such incidental harm is likely to result, the attack must be suspended if it is expected to cause excessive civilian harm in relation to the military advantage anticipated to be gained from the attack.5
The practical challenge with these rules is first and foremost that when military forces are under attack, they are often not in the position to make an assessment of the status of the attacking forces. As Corn explains, “[i]n the context of combat operations, the line between actions that constitute pure self-defense, such as a unit responding to a hostile act of civilians, and a unit performing the doctrinal military mission of conducting a hasty defense against a belligerent attack, is often blurred”.6 They may not have the resources to both defend themselves and examine the source of the threat, or it may be practically impossible to identify the attackers without getting killed in the process. Perhaps more importantly, it may be very difficult to collect information about civilians or civilian objects in the vicinity of the attackers in order to make the proper proportionality assessment. For these reasons, some States train their troops that they will be authorised to use force that is necessary and proportionate to defend themselves when attacked, so that they are not required to go through the loac analysis.7 This is an important reason why self-defence is presented as the justification for the use of force more often than would be expected in military operations during armed conflict. As explained, other reasons are the combined effects of restrictive roe causing military forces to look to alternative authorisations for the use of force, and self-defence being the only authorised exception to roe.
This trend begs two questions. First, is loac actually that difficult to apply when military forces are under attack? Second, what is the practical distinction between authorising the use of force that is necessary and proportionate and applying loac in situations of attack? Considering that loac was designed to apply to combat situations, it should be practically possible to apply to the situation where military forces are under attack. As explained in Section 5.4, the requirement is to do everything feasible under the circumstances, meaning that the required standard is adapted to the realities military forces face. If it is impossible to do more to verify the status of the attackers or the presence or absence of civilians, it is not required. In the case of someone attacking nato forces with military-type weapons such as rifles, rockets or grenades, it will in most cases be reasonable to assume that they are military forces or an organised armed group participating in the armed conflict. Their behaviour may in other words be sufficient to identify them as opposing forces.
The next question for the military forces will be whether the information they have (or lack) makes it reasonable to use force, and therefore potentially risk causing harm to civilians. The determination will be entirely context-dependent. However, some general remarks are still possible. For instance, if their situational awareness and the general knowledge of the pattern of life in the area indicates that the risk of civilian presence is high, an assessment should be made as to whether the use of force is necessary or whether it can be avoided. If a decision not to use force results in unacceptable risk to one’s own forces or the target is deemed sufficiently important to justify the potential civilian losses, the continued use of force will be lawful under loac.
Legally, loac provides military forces with the necessary authorities to deal with actual attacks or threats of attack. There are two likely reasons why personnel are trained to use the force that is necessary and proportionate in such situations. First, the loac rules are very detailed and therefore complex to teach and apply. Second, it means that the ‘use of force’ training will be applicable to all defence situations. As a result of the restrictive ‘use of force’ authorities under the law enforcement paradigm, the outcome of applying those principles in a loac is likely to be the same.8 If anything, the discussion in Section 8.2 indicates that the use of force in self-defence in general permits less force than loac. Because any use of force not permitted by loac must be a measure of last resort, it is unlikely that the use of the general or peacetime concepts of necessity and proportionality would entail a violation of loac. When nato forces are attacked or about to be attacked, the main difference between the use of force permitted by loac and self-defence would be the degree to which the impact on innocent parties or civilians is considered;9 however, military forces are trained to always take the risk of harm to civilians into account. Furthermore, because such troops in contact (tic) situations are not pre-planned, the extent to which precautions may be taken in the preparation for the attack in order to verify the presence and hence risk to civilians and civilian objects, is limited.10 As a result, there is little reason to go into the details of loac when training tactical level military forces in how to respond to attacks by opposing forces.
Importantly, this does not mean that loac training in general is not required. Tactical level forces may face threats that amount to dph but do not meet the required threshold of self-defence. The authority to respond to such threats will be regulated by the nato hostile act and/or hostile intent roe. As will be further explained in Chapter 11, it is essential that military forces participating in operations where the nato hostile act and hostile intent roe are implemented are trained both in how these roe should be interpreted and applied, and how they relate to other ‘use of force’ categories.
The following example may be used to illustrate some of the complexities of the use of force in response to attacks or imminent attacks by opposing forces. A platoon involved in an armed conflict is under fire at a forward operating base (fob). They know that there may be civilians in the area, but they are not certain where. If they were to stand up and see exactly where to return the fire, they would likely be killed. They are thus aiming in the direction they are being shot from. At first glance, it may be difficult to see how the loac requirements can be met. The platoon cannot aim specifically at the individual members of the attacking forces, and unable to see where the civilians are, the platoon cannot avoid harming them. Self-defence may therefore appear to be the better legal authority for the defensive acts.
While this is an area where loac application is complicated, it is nonetheless possible to apply. First, the use of force may be sufficiently aimed at the opponents even if each shot is not directed at a specific individual. The requirement to do everything feasible to verify that attacks are only directed at lawful targets is therefore met.11 The second issue is the presence of civilians. Although civilians will usually attempt to get away from an area where there is ongoing fighting, this may not be possible. They may be forced to stay as involuntary human shields or they could be otherwise physically prevented from leaving the area. As a result, the platoon must assume that there is a risk of collateral damage. However, due to the heavy incoming fire, the troops are unable to verify the exact presence or absence of civilians or to reduce the potential harm caused to civilian persons or objects without themselves being killed. loac requires that all feasible precautions are taken to minimise collateral damage, and to refrain from launching attacks expected to cause excessive civilian harm.12 In the current case, the extent of civilian harm is not known or expected to be excessive to the military advantage of saving the lives of the platoon members, and further precautions are not feasible. Although the failure of the attackers to distance themselves from their civilian population places the latter at risk, it does not prevent the platoon from defending themselves.13
Even if the decision is made to simplify the training of tactical-level forces to focus on necessity and proportionality, and the requirement to take all feasible precautions to avoid collateral damage, it should be remembered that any assessment of the lawfulness of the use of force after the fact should be based on the applicable legal regime. During armed conflict, this is most likely to be loac. The decision to simplify the requirements for the use of force does not alter which legal regime actually applies. If the nato forces are assessed to have used force beyond what is justified by the general principles of necessity and proportionality, it may still be lawful under loac. In particular, there is no requirement under loac to use lethal force as a measure of last resort. Attacking opponents who are no longer a threat (without being hors de combat) could be a violation of an order to only use necessary and proportionate force, but it would not be unlawful killing.
As previously explained,14 the scope of application of self-defence and hence the relationship between loac and self-defence will differ depending on the type of the conflict. During niacs, the opposing forces will not have the right to participate in hostilities. As a result, the initial self-defence requirement of unlawful or unjustified attacks is met, and self-defence would legally be applicable. However, because loac is better suited to war fighting than self-defence, self-defence is most relevant in the assessment of the legality of the use of force where there is uncertainty regarding the opponent’s status as lawful target under loac.15
nato, Military Decision on MC 362/1 – NATO Rules of Engagement [hereinafter MC 362/1], 30 June 2003. The MC 362/1 is nato unclassified, however, permission to use parts of the document for the purposes of this research is granted by the nato Military Commission in document imstam(O&P)-0006-2018 (copy on file with the author).
As explained in Section 3.1, combat engagement encompasses actions in response to both threats or attacks not giving rise to self-defence; in other words, hostile intent not constituting an imminent attack, hostile acts not constituting an actual attack, and imminent or actual attacks from opposing forces in an armed conflict.
Protocol Additional to the Geneva Conventions of 12 August 1949, and relating to the Protection of victims of International Armed Conflicts (Protocol I) [1977], printed in Dietrich Schindler and Jiri Toman, The Laws of Armed Conflicts (Martinus Nijhoff Publisher, 1988) pp. 711ff, Article 49(1). See also Section 5.1.3.
The commentary to AP I Article 49(1) stresses that defensive force particularly includes ‘counter-attacks’. Yves Sandoz, Christopher Swinarski, and Bruno Zimmermann (eds.), Commentary on the additional protocols of 8 June 1977 to the Geneva Conventions of 12 August 1949 [hereinafter: AP Commentary] (icrc, Martinus Nijhoff Publishers, Geneva, 1987) para. 1880.
Gary P. Corn, ‘Developing Rules of Engagement: Operationalizing law, policy and military imperatives at the strategic level’, in Geoffrey S. Corn, Rachel E. VanLandingham and Shane R. Reeves (eds.), U.S. Military Operations – Law, Policy and Practice (Oxford University Press, Oxford, 2016) p. 244.
For example, a soldier’s card issued to British forces involved in the occupation phase of Operation Iraqi Freedom, “Card Alpha”, stipulates that “[t]his guidance does not affect your inherent right of self-defence. However, in all situations you are to use no more force than absolutely necessary”. This clearly imposes a law enforcement ‘use of force’ standard to a situation regulated by loac. The card was classified as restricted, but is published in: UK, The Iraq Fatality Investigations, Report into the death of Ali Salam Naser by Inspector Sir Georg Newman, Presented to Parliament by the Secretary of State for Defence by Command of Her Majesty, March 2017, Document CM 9410, (
See also Gary P. Corn, ‘Should the Best Offense Ever be a Good Defense? The Public Authority to Use Force in Military Operations: Recalibrating the Use of Force rules in the Standing Rules of Engagement’, 49 Vanderbilt Journal of Transnational Law 1 (2016), p. 46.
See discussion in Section 8.2.7 in relation to the use of force in self-defence and effects on innocent bystanders.
See Section 5.5.
AP I (n 3) Article 51(2)(i).
ibid, Article 51(2)(ii-iii).
As Merriam explains in the context of target verification, the tendency of non-state actors not to distinguish themselves from the civilian population will also affect what may reasonably be required of an attacker with regard to verification. While the failure of the opposing forces to comply with the principle of distinction would require the attacking forces to take further measures to gather information and verify the opponents, it may also “explain or excuse a faulty targeting decision based on that information”, because the routine failure to comply with the principle of distinction will cause the attacker to make mistakes. John J. Merriam, ‘Affirmative Target Identification: Operationalizing the Principle of Distinction for U.S. Warfighters’, 56 Virginia Journal of International Law 83 (2016), p. 130.
See Sections 6.3.2.1 and 8.2.3.2.
See discussion in Section 8.2.3.2.