Soldier Self-Defense Symposium: Self-Defense in Armed Conflicts–The Babel Tower Phenomenon

[Gloria Gaggioli is a Swiss National Science Foundation (SNSF) Professor at the Law Faculty of the University of Geneva as well as Lecturer at the Geneva Academy of International Humanitarian Law and Human Rights and at the University of Neuchâtel (Switzerland). She is specialized in international humanitarian law and human rights law. Prior to joining the University of Geneva, she served as Legal Adviser in the legal division of the International Committee of the Red Cross (ICRC). This post is part of our week-long symposium on soldier self-defense and international law.]

From Syria to Gaza, from Afghanistan to Somalia, self-defence is constantly mentioned in relation to armed conflict situations. Not only States refer to self-defence to justify the use of force at the strategic, operational and tactical levels, but also other actors such as the United Nations in relation to Peace Operations or Private Military and Security Companies. Even individual soldiers are tempted – when facing trial – to justify their resort to force by reference to self-defence instead of arguing that their use of force was lawful under international humanitarian law (IHL), the body of international law designed to regulate hostilities in armed conflict. (See e.g. Henderson & Cavanagh) This practice gives rise to a “Babel Tower Phenomenon” whereby everyone refers to self-defence in armed conflicts but each one understands it in a different way.

Despite this omnipresence of self-defence in current military discourses, IHL experts are often puzzled by this confusing concept, which finds generally no proper legal basis in IHL. They tend to dismiss it as a national or military concept, which has no bearing on international law. As a matter of fact, though, self-defence is solidly entrenched in military training and practice. Depending on the scope it is being given, it may collide with, or even progressively be used as a substitute for IHL. The way self-defence is being applied in the field may also have an important humanitarian impact (see e.g. Erica Gaston).

This symposium on Soldier Self-defence is therefore a welcome endeavour as one of the rare attempts to shed further light on self-defence in armed conflict situations. As well illustrated by the fascinating posts on States’ and United Nation’s perspectives, self-defence is a multi-faceted or multi-layered concept that is being interpreted and applied very differently by various actors. In the present blog, I will highlight three main differences in relation to self-defence for military personnel, as regards: 1) its legal source(s), 2) its content and 3) its effect on military operations (I develop on these aspects in a chapter I authored in the recently published Complex Battlespaces: The Law of Armed Conflict and the Dynamics of Modern Warfare, edited by Winston S. Williams and Christopher M. Ford).

1. Source of soldier’s self-defence: Jus ad bellum? Criminal law? Human Rights Law?

The legal source of soldier’s self-defence remains shrouded in doubts and States have diverging views, if any, on the matter. One can discern two broad approaches in operational law: one tends to locate self-defence within jus ad bellum and the other within domestic criminal law. I will then present a third – additional – source for soldier’s self-defence, i.e. international human rights law.

a. Jus ad bellum approach

A first approach consists in taking State or jus ad bellum self-defence as a starting point and then deriving self-defence rights at other sub-levels, most importantly at the level of the military unit and at the level of the individual soldier as a State representative. This is typically the approach adopted in the US Standing Rules of Engagement (SROE). An obvious problem in deriving unit and personal self-defence from jus ad bellum is the risk of conflation between jus ad bellum and jus in bello. Randall Bagwell cleverly suggests that these self-defence concepts under the SROE should only be invoked when a conflict begins and not thereafter as they emanate from jus ad bellum. Unfortunately, US military practice indicates, on the contrary, that the SROE concept of self-defence is regularly asserted as an authority to use force during armed conflicts in parallel, or even instead of, IHL. Arguably, this US approach has also influenced the practice of the United Nations, which – as highlighted by Bruce Oswald – refers to the “inherent right of United Nations forces to take appropriate measures for self-defence” (emphasis added) in order to authorize the use of force beyond what domestic law would usually authorize. In my view, the adjective “inherent” is an implicit reference to Article 51 of the United Nations Charter and therefore locate self-defence of UN Peace Forces within a jus ad bellum framework. The San Remo Rules of Engagement Handbook (2009) follows more or less the US approach, which is described as the “more generally accepted view” of self-defence.

b. Criminal Law approach

Another, radically opposed approach, is to derive soldiers’ self-defence from domestic criminal law. Even unit self-defence would be seen as nothing more than a collective form of personal self-defence or else a special form of defence of others as generally recognized by domestic law. This is the view adopted by Norway according to Camilla Cooper and Sigrid Redse Johansen. This restrictive view is seducing, however it fails to recognize that soldiers, as State agents, may/are expected to use force beyond what any private individual could do under domestic criminal law, in order for instance to protect individuals under its jurisdiction from criminal acts (e.g. pursue and arrest an alleged criminal).

In-between these two extremes, there are a number of variations and the jus ad bellum/criminal law approaches are sometimes mixed up or at least combined. For instance, Hans Boddens Hosang and Terry Gill present a mixed approach for the Netherlands, which is to derive unit self-defence from State self-defence, while at the same time basing personal self-defence on domestic criminal law. Interestingly, these experts nevertheless contend that “[s]elf-defence under Netherlands criminal (case) law has a number of aspects which are not readily applicable in the military operational context”. In other words, if criminal law is seen as the source of personal self-defence, it is nevertheless adapted or applied differently – with due regard to ROE – within military operations; thus suggesting that self-defence of military personnel is not applied in the exact same way as for any other citizen.

c. International Human Rights Law as an Additional Source

In my view, there is an additional international legal basis (and constraint) on soldier’s self-defence that is often overlooked and that is to be found in international human rights law (IHRL). (On personal self-defence under human rights law, see Jan Arno Hessbruegge). Under IHRL, self-defence is one of the circumstances under which State agents may use lethal force and thus deprive an individual from his/her life. This is evidenced by the UN Basic Principles on the Use of Force and Firearms (Principle 9) and by Article 2, paragraph 2 a) of the European Convention on Human Rights notably. To the extent that IHRL applies in armed conflict situations as well as extraterritorially (both controversial issues), this human rights notion of self-defence is certainly relevant in armed conflict situations.

Self-defence is thus part and parcel of the use of force in law enforcement by State agents. Law enforcement is broader than self-defence though, because – unlike private individuals – State agents may use force (including potentially lethal force depending on the circumstances) to effect a lawful arrest, to prevent escapes or to cope with a riot situation for instance.

It is generally admitted today that certain situations of violence in armed conflicts are to be dealt with under a law enforcement paradigm, as developed under IHRL, rather than under an IHL conduct of hostilities paradigm. (See ICRC Use of Force Report 2013 and ICRC Challenges Report 2015). Although the interplay between the conduct of hostilities and law enforcement paradigms is complex and a matter of ongoing legal debate, it is nevertheless quite clear that situations such as civilian unrest (e.g. riots) or the maintenance of law and order in detention (e.g. to prevent escapes or to contain violent detainees) has to be dealt with under law enforcement rules. Similarly, when a serviceman is attacked by a civilian for reasons that are unrelated to the armed conflict situation (no nexus), he/she may only use force in accordance with law enforcement (which includes self-defence). The same is true when a serviceman is under an imminent threat of attack and has a doubt as to whether the aggressor is a legitimate target under IHL (I develop on this point here).

Like jus ad bellum self-defence, human rights self-defence relates to State responsibility, not individual criminal responsibility. In other words, if a State agent killed an individual in self-defence or to prevent him/her from committing a particularly serious crime involving grave threat to life, the State it represents will not be held responsible for a violation of the right to life, provided the use of force was otherwise consonant with law enforcement rules (see below regarding necessity, proportionality and precautions). But unlike jus ad bellum self-defence, human rights self-defence, including defence of others, is situated at the personal level, not at the State level. This is to say that human rights self-defence and defence of others come into play when a State agent (an individual or a group of individuals), not a State as an abstract entity, is confronted with an imminent threat of death or serious injury to himself or herself, or others.

From an operational law perspective, the use of force in law enforcement may either be included in the rules of engagement pertaining to mission accomplishment or at least partly in the rules of engagement pertaining to self-defence. I submit that the tendency to give overbroad definitions of self-defence in armed conflicts would be better contained if the use of force in law enforcement, in accordance with IHRL, would be given the place it deserves in armed conflicts.

2. Content: hostile act/demonstrated hostile intent? Unlawful attack? Necessity and proportionality? Imminence?

Depending on what the source of soldier’s self-defence is, its content or, in other words, the principles that govern it, will diverge significantly.

a) The US SROE, for instance, consider that the triggering act for self-defence is a hostile act, i.e. an attack “or other use of force against the United States, U.S. forces, or other designated persons or property”, or a demonstrated hostile intent, i.e. “the threat of imminent use of force against the United States, U.S. forces, or other designated persons or property”. This same triggering act applies to unit, individual, national, and collective self-defence. Imminence is also defined identically for all levels of self-defence and is understood broadly as “not necessarily mean[ing] immediate or instantaneous”. Self-defence in the SROE is further constrained by the principles of necessity and proportionality, but it is unclear how imminence, necessity and proportionality need to be interpreted or understood at the level of the unit or serviceman. For instance, is an attack imminent under this understanding of self-defence if it is going to happen two days or one month later (e.g. in relation to the placing of an IED)? Is it proportionate under this concept of self-defence to launch a grenade against a person demonstrating a hostile intent if this is the only mean available, knowing that this will cause the incidental killing of numerous bystanders (and thus possibly be considered as disproportionate from an IHL perspective; and even more so from a IHRL perspective)? Uncertainties already exist in relation to the principles governing national self-defence under Article 51 of the UN Charter (see e.g. on proportionality David Kretzmer), but they are further magnified when applied at the lower levels of the unit/individual serviceman within an armed conflict.

b) If one takes the opposite view that individual, and possibly unit self-defence as well as defence of others, derive from the criminal law notion of self-defence, then a prerequisite is that the triggering act must be an “unlawful act”. (Note that countries adopting this view, may nevertheless use the notions of “hostile act/hostile intent” in relation to mission accomplishment, not self-defence, which further adds to the confusion surrounding these concepts). Then, the question is how to determine the lawfulness/unlawfulness of the triggering act. As I develop elsewhere, I fully agree with both Randall Bagwell and Camilla Cooper & Sigrid Redse Johansen that, within an armed conflict, the lawful/unlawful character of the triggering act needs to be mainly assessed under IHL. This means that the overlap between self-defence and IHL is greatly reduced compared to the “jus ad bellum approach” under (a). For instance, the use of force by an enemy combatant in an international armed conflict is lawful and therefore the use of force in response would only be governed by IHL; the criminal law concept of self-defence cannot justify the use of force in response. It is however unclear whether the same is true in relation to the use of force by a civilian directly participating in hostilities or a fighter in a non-international armed conflict who do not have combatant immunity, i.e. the right to take part in hostilities under IHL. While such use of force is not unlawful per se under IHL, it would nevertheless most probably be unlawful under the domestic law of the defending party. To the extent that domestic law has also to be taken into account to determine the lawfulness/unlawfulness of the triggering act, there is room for overlap between IHL and criminal law self-defence. While IHL could for instance be invoked as a main argument in such cases, criminal law self-defence could still be argued in the alternative. The interplay between criminal law notions of self-defence and IHL is therefore complex and the application of the lex specialis maxim in this context – to prefer IHL over self-defence in a trial – is doubtful in my view.

It shall be added that domestic laws differ considerably from one another in relation to the criminal law concept of self-defence. Some see self-defence as a justification, others as an excuse. Some accept self-defence pleas to protect objects, others do not. But usually domestic laws include a requirement of necessity and proportionality or reasonableness or else justifiability. Imminence may be mentioned explicitly or seen as encompassed in necessity. These principles are obviously very different from, and much more stringent than, their jus ad bellum counterparts. For instance, the means employed in the response will clearly be taken into account to assess proportionality. Additional considerations, such as the duty to retreat, can be further weighed in.

Although, in principle, domestic law notions of self-defence are supposed to be consonant with IHRL, in practice they tend to be interpreted and applied more flexibly. For instance, while the use of deadly force to defend property would generally be considered disproportionate from a IHRL perspective, a number of States do accept it. The necessity, proportionality, reasonableness or justifiability principles may also at times be interpreted more generously by domestic courts than what international law would require. (But see Armani Da Silva v. United Kingdom (2016) where the European Court of Human Rights astonishingly accepts a purely subjective test to assess “absolute necessity”/reasonableness in a context of putative self-defence where a mistaken suicide bomber had been shot at by a British policeman).

c) Lastly, if one considers that soldier’s self-defence is a subset of the human rights law enforcement paradigm, the customary principles governing it are quite clear – and generally identical for all States. The use of lethal force in self-defence must respond to an unlawful and prior or imminent threat to life or limb, i.e. “a matter of seconds not hours” to quote Christof Heyns, former Special Rapporteur on Extrajudicial, Summary or Arbitrary Execution. It must be absolutely necessary and strictly proportionate. This entails notably that the use of force must be the last resort (which involves de-escalation if possible and a differentiated use of force) and that incidental loss of lives must be avoided as far as possible. Every precaution must be taken in the planning and control of operations in order to avoid the use of force by State agents in self-defence or otherwise. The human rights law enforcement paradigm also entails obligations before and after the use of force, such as the obligation of States to appropriately train and equip their agents and the obligation to investigate credible allegations of a violation of the right to life. (For a summary of the principles governing the use of force in law enforcement, see e.g. ICRC Factsheet 2015).

3. Effect: ex ante authority to use force or limited ex post justification? Restraint on military operations or license to kill?

The enormous differences in approaches outlined above may lead to see self-defence either as an ex ante authority to use force over and beyond IHL or as a limited ex post justification/excuse for the use of force by individual soldiers.

I submit that self-defence cannot and should not be an additional ex ante authority for the use of force in military operations. From an international law perspective, the degree and amount of force to be used within an armed conflict is either regulated by the IHL conduct of hostilities paradigm or the human rights law enforcement paradigm. Self-defence is not a third paradigm. It is part and parcel of the human rights law enforcement paradigm. It is therefore constrained by the principles of absolute necessity, strict proportionality and precautions as understood under human rights law.

Jus ad bellum self-defence, applied at the lower levels of units and individual servicemen cannot replace IHL. Beyond the initiation of an armed conflict, jus ad bellum self-defence is seen by most international lawyers as irrelevant because it has generally no direct impact on the degree and amount of force used at the tactical or operational level. Thus, for instance, when – in the midst of an armed conflict – a military unit is being attacked by the enemy, the lawfulness of the response is to be mainly assessed under the IHL principles of distinction, proportionality and precautions; not under the jus ad bellum concept of self-defence. From an IHL perspective, the principles of the conduct of hostilities apply irrespective of who is the aggressor versus aggressed party from a jus ad bellum perspective.

In parallel to the human rights notion of self-defence, which has an important role to play in contemporary military operations, there are the criminal law concepts of self-defence. These criminal law concepts of self-defence are certainly relevant and available to servicepersons, but they cannot provide an ex ante “authority to use force” to soldiers/military units. They may merely justify certain conducts ex post when individual soldiers are facing trial. In principle, there should be no need to resort to the criminal law defence of self-defence if the use of force was consonant with IHL. Thus, even ex post, self-defence pleas should be the exception rather than the rule for acts of war.

To conclude, I would like to highlight that the incredible confusion over the sources and content of self-defence coupled with an increasing reliance on this concept in contemporary military operations is not only dissatisfactory from a legal perspective as highlighted above, but is also worrying from an operational and humanitarian perspective. I will limit myself to three particular challenges in this regard:

• The “Babel Tower phenomenon” leads to interoperability issues in multinational operations because various stakeholders refer to self-defence but have different meanings/definitions in mind. Self-defence does not simplify military operations, but rather adds a useless layer of complexity.
• In terms of military process, portraying a situation as self-defence-related (or “troops-in-contact”) may authorize in certain countries to forego a formal collateral damage estimate and may have an impact in terms of weapons availability (e.g. opening up options for close air support); thus, potentially by-passing the IHL principles of proportionality and precautions. For instance, in relation to the erroneous attack conducted against the Kunduz Trauma Centre operated by Médecins Sans Frontières on 3 October 2015, it has been alleged that the fact that the US airstrike had been initially called in in self-defence by Afghan forces, allowed to bypass certain precautionary measures and notably to forego higher-up approval for the strike; thus, increasing the risk of error.
• In terms of perception, the alleged “inherent right to self-defence” conveys to military personnel a sense of legitimacy that is often understood as a license to kill irrespective of IHL limitations. Erica Gaston, based on an empirical study conducted in Afghanistan, shows that expansive views of self-defence “result in overbroad threat designations and wide latitude in the level of force permitted” and “can increase the risk of civilian casualties and disproportionate uses of force”.

In brief, although the increasing reliance on self-defence in contemporary armed conflicts, especially in counterinsurgency or stability operations, might have been intended to diminish civilian casualties, in practice, it might in some cases have the exact opposite effect. There are thus cogent reasons to oppose the progressive replacement of a stable, agreed-upon framework – IHL – with an uncertain/multifaceted concept of self-defence.