Much has been written about the relationship between international humanitarian law (humanitarian law or IHL) and international human rights law (human rights or IHRL), their separate origins, and their convergence and mutual influence. Their fields of application, the rights protected and the respective implementation mechanisms have been compared. The International Court of Justice (ICJ) and various human rights bodies have explored the extent to which humanitarian law is the lex specialis compared with human rights. What has provoked less jurisprudence is that on some issues human rights constitute the lex specialis. The meaning of the lex specialis concept and how the lex specialis can be identified has been studied by the International Law Commission (ILC)Footnote 1 and scholars in general, as has specifically the relationship between the two branches.Footnote 2 A systematic analysis to determine which of the two branches constitutes the lex specialis on issues covered by both is still lacking, but can certainly not be the aim of this article.
In practice, the aforementioned theoretical questions do not matter for most problems actually affecting victims of armed conflicts. This is particularly true in international armed conflicts, for which humanitarian law treaties are best developed. First, the rules of both branches have the same addressees: states. Second, for many situations it is difficult to argue that human rights apply at all, because the victims cannot be considered as being in the territory or under the jurisdiction of the state attacking them. Third, with regard to many issues, one or other of the two branches simply contains no rules. There is nothing in humanitarian law about the freedom of the press in occupied territories, and human rights law says nothing about whether and how combatants have to distinguish themselves from the civilian population. Fourth, on most other issues the two branches lead to the same results, one or the other providing more details.
In non-international armed conflicts, too, both branches mostly lead to the same results. The treatment of persons detained or otherwise in the power of a state is prescribed in a very similar way. The judicial guarantees for persons undergoing trial are likewise very similar, but they are better developed in human rights. The jurisprudence of the European Court of Human Rights (ECtHR) –which never explicitly refers to humanitarian law – concerning deliberate or indiscriminate attacks against civilians in Chechnya and eastern Turkey shows that even on such a typical humanitarian law subject as precautionary measures, which have to be taken for the benefit of the civilian population when attacking military objectives, human rights can lead to the same result as humanitarian law.Footnote 3
However, on two questions which are crucial in practice, not only is the relationship between the two branches unclear but also the answer of humanitarian law alone. First, may a member of an armed group, as according to humanitarian law applicable to international armed conflict, be attacked (and therefore be killed) as long as he or she does not surrender or is not otherwise hors de combat, or is this, as in human rights, admissible only exceptionally and when an arrest is not feasible? Second, may a captured member of an armed force or group be detained similarly to a prisoner of war in international armed conflicts, until the end of active hostilities, and without any individual decision, or must the captured person, as prescribed by human rights, have an opportunity to challenge his or her detention before a judge?
These two questions have gained prominence in recent years in relation to some components of the ‘war on terror’, which have been classified by the US Supreme Court as non-international armed conflict.Footnote 4 Since 11 September 2001 the answer of the US administration has been that it may kill (e.g. in Yemen) and detain (e.g. in Guantánamo) ‘unlawful enemy combatants’ according to the same standards as the humanitarian law of international armed conflict prescribes for combatants (but without their benefiting from the protection offered to ‘lawful’ combatants). Critics object that most of these persons may be killed or detained only in accordance with the much more restrictive human rights rules. We shall try to provide answers independently of the particular nature of the ‘war on terror’, which would add two problems already discussed sufficiently elsewhere: the extraterritorial application of human rights; and whether all, some or any components of the ‘war on terror’ are armed conflicts at all and, if so, whether they are international in character or not.
While we shall deal with these questions in terms of traditional non-international armed conflicts between a government and rebel forces, we think that our answers constitute a starting point for replies to the same questions in ‘transnational armed conflicts’. The answers are made more difficult by several factors that sometimes point in different directions.
First, the treaty rules of the humanitarian law of non-international armed conflicts are more rudimentary than those applicable to international armed conflicts. Under the lex specialis principle, this would normally allow greater scope for human rights. In the last twenty years, however, the jurisprudence of international criminal tribunals, the influence of human rights and even some treaty rules adopted by states have instead brought the law of non-international armed conflicts closer to the law of international armed conflicts, and some even suggest eliminating the difference altogether. In the many fields where the treaty rules still differ, the convergence has been rationalized by the claim that under customary international law the differences between the two categories of conflicts have gradually disappeared. This development reached its provisional peak with the publication of the ICRC study Customary International Humanitarian Law (ICRC Study) which claims, after ten years of research on ‘state practice’ (in the form of official declarations rather than actual behaviour), that 136 (and arguably even 141) of 161 rules of customary humanitarian law, many of which parallel rules of Protocol I applicable as treaty law to international armed conflicts, apply equally to non-international armed conflicts.Footnote 5 Our discussion is further complicated by the fact that some of the practice upon which those customary rules are based is that of human rights bodies applying human rights law.
Second, even if the lex specialis principle were to provide clear answers for cases in which the two branches of law respond to the same question with different rules (and it does not), it could only operate if the answer provided by each of the branches were clear. We shall show that, at least in humanitarian law, such clarity is not found in response to the two questions under discussion here. As for human rights, the answers are frequently based on general treaties without universal ratification or on regional treaties – while the exact substance of customary human rights is at least as controversial as that of customary humanitarian law. Often those answers are also based on the practice of bodies which cannot take binding decisions and sometimes on soft-law instruments whose binding character is controversial. In addition, human rights limitations are often very flexible, inter alia because of vague limitation clauses which allow them to take the specific nature of each case into account. As there are only very few cases in which human rights mechanisms have resolved our questions in actual non-international armed conflicts, we must often base our answers on precedents arising outside armed conflicts, without any certainty that the relevant mechanism would have reached a similar decision if the case had arisen in an armed conflict.
Third, another factor in non-international armed conflicts which renders our discussion particularly complex (and is very neglected in scholarly writingsFootnote 6 and even in the ICRC Study) is that the humanitarian law of non-international armed conflict is, as Article 3 common to the Geneva Conventions points out, equally binding for ‘each party to the conflict’ – that is, for the non-state armed group just as much as the government side.Footnote 7 This raises the question whether human rights are equally addressed to armed groups or whether, by virtue of the operation of the lex specialis principle, the answer to our questions is not the same for the government and for its opponent.
Fourth, while state practice concerning our two questions in international armed conflicts is fairly uniform (though sometimes blurred by controversies about how a certain conflict should be classified), it is clearly contradictory in non-international armed conflicts. In confrontations with rebel groups, some states let human rights prevail, some apply by analogy the rules of humanitarian law governing international armed conflicts and some a mix of the two. As for the international supervisory mechanisms, their solutions also differ and it is not always clear whether their answer is based on an appreciation of the law in its entirety or limited, due to the subject matter of their jurisdiction, to the application of only one of the two branches under discussion here.
The lex specialis principle and its meaning
In this article it is assumed that human rights apply in armed conflicts. Some states disagree, but they have never specifically done so with regard to non-international armed conflicts on their own territory. As for humanitarian law, it is designed specifically to regulate armed conflicts.
The problems arising from this simultaneous applicability of the said two branches of law must be solved by reference to the principle ‘lex specialis derogat legi generali’.Footnote 8 This principle seeks to establish, through an objective standard corresponding to the regulated subject matter, a preferential order for two rules that apply to the same problem but regulate it differently. The reasons for preferring the more special ruleFootnote 9 are that it is closer to the particular subject matter and takes better account of the uniqueness of the context. It also better represents states' intentions on how to regulate the given problem. Theoretically, a situation should be regulated by applying the most just rule, but in order to avoid evaluations that are too subjective, it is preferable to refer to a more objective standard that still reflects justice.Footnote 10 The principle does not indicate an inherent quality in one branch of law, such as humanitarian law, or of one of its rules. Rather, it determines which rule prevails over another in a particular situation.Footnote 11 Each case must be analysed individually.Footnote 12
Several factors must be weighed to determine which rule, in relation to a certain problem, is special. When the legal consequences of two norms regulating the same situation are mutually exclusive, speciality in the sense of logic implies that the norm that applies to certain facts must give way to the norm that applies to those same facts as well as to an additional fact present in that situation. Between two applicable rules, the one which has the larger ‘common contact surface area’Footnote 13 with the situation applies. The norm with the scope of application that enters completely into that of the other norm must prevail, otherwise it would never apply.Footnote 14 It is the norm with the more precise or narrower material and/or personal scope of application that prevails.Footnote 15 Precision requires that the norm explicitly addressing a problem prevails over the one that addresses it implicitly, the one providing the advantage of detail prevails over the other's generality,Footnote 16 and the more restrictive norm over the one covering the entire problem but in a less exacting manner.Footnote 17
A less formal – and also less objective – factor for determining which of two rules applies is the conformity of the solution to the systemic objectives of the law.Footnote 18 To characterize this solution as ‘lex specialis’ is perhaps a misuse of language. The systemic order of international law is a normative postulate founded on value judgements.Footnote 19 Some consider that in reality the decision-maker first determines which rule is more just and then characterizes it as lex specialis. Footnote 20 In particular, when formal standards do not indicate a clear result, the teleological criterion must weigh in, even though it allows for personal preferences.Footnote 21
In our opinion, the more the formal standard points to a clear conclusion in relation to a certain problem the less necessary it becomes to assess the systemic objective. The inverse is also true – the more clearly the objective can be seen, the easier it is to distance oneself from the formal standard.
Once the lex specialis is determined, the lex generalis still remains present in the background. It must be taken into account when interpreting the lex specialis;Footnote 22 an interpretation of the lex specialis that creates a conflict with the lex generalis must be avoided as far as possible and an attempt made instead to harmonize the two norms.Footnote 23
According to doctrine, the principle appears to refer implicitly to the antinomies between conventional – that is, treaty-based – rules. Whether the principle also applies to the relationship between two customary rules is less clear. In a traditional understanding of customary law this is theoretically not the case. The customary rule applicable to a certain problem derives from the practice and opinio juris of states in relation to that problem. In relation to the same problem, there cannot be a customary ‘human right’ and a different customary ‘humanitarian rule’. The focus is always placed on the practice and the opinio juris manifested in relation to problems as similar as possible to the one to be resolved. This appears to be the approach adopted by the ICRC, which refers in its Study to a vast array of practice in human rights, both within and outside armed conflicts.Footnote 24 In practice, however, when looking for a customary rule one generally consults a text, be it a treaty or other instrument codifying customary law, a text that instigated the development of a customary ruleFootnote 25 or even a doctrinal text. It may then be found that one specific problem is covered by two contradictory texts, both deduced from state practice. The choice between these two texts is, in our opinion, governed by the same principles as the choice between two treaty-based rules. If the state practice clarifying which of the two rules prevails in the given situation is not dense enough to be conclusive, the usual methods must be used to discover which of the two rules, derived from the practice analysed from different perspectives, constitutes the lex specialis.
When may an enemy fighter be killed?
The traditional answer of the humanitarian law of international armed conflicts
In international armed conflicts, members of armed forces belonging to a party to the conflict are qualified as ‘combatants’. Combatants may be attacked at any time until they surrender or are otherwise hors de combat, and not only while actually threatening the enemy. Combatants are part of the military potential of the enemy and it is therefore always lawful to attack them for the purpose of weakening that potential. The traditional understanding is that no rule restricts the use of force against combatants only to those circumstances in which they cannot be captured. Within humanitarian law this view has been challenged, citing both the principle of military necessity as a restriction on all violenceFootnote 26 and the prohibition of treacherous killings.Footnote 27 However, neither of these views has been translated into actual battlefield instructions, even less into actual battlefield behaviour.Footnote 28
As for the proportionality requirement, it applies to attacks directed at legitimate targets, but only to protect civilians from their possible incidental effects.Footnote 29 Attacks against combatants are not subject to a proportionality evaluation of the harm inflicted on the combatant and the military advantage derived from the attack.
The unclear answer of the treaty rules of humanitarian law applicable to non-international armed conflicts
In contradistinction to international armed conflicts, it is not clear under the treaty law of non-international armed conflicts when an enemy fighter may be killed (we use the term ‘fighter’ in this article for a member of an armed group with a fighting function and for members of government armed forces). Neither Article 3 common to the Geneva Conventions nor their Protocol II refers to ‘combatants’, because states did not wish to confer the right to participate in hostilities and its corresponding combatant immunity on anyone in non-international armed conflicts. The relevant provisions prohibit ‘violence to life and person, in particular murder’ directed against ‘persons taking no active part in hostilities’, including those who have ceased to take part in hostilities.Footnote 30 Specifically addressing the conduct of hostilities, Article 13 of Protocol II prohibits attacks against civilians ‘unless and for such time as they take a direct part in hostilities’.Footnote 31
It may be deduced from these rules and from the absence of any mention of ‘combatants’ that in a non-international armed conflict everyone is a civilian, and that no one may be attacked unless he or she directly participates in hostilities. However, the ICRC Commentary on Protocol II states that ‘[t]hose belonging to armed forces or armed groups may be attacked at any time.’Footnote 32 Otherwise it would, first of all, indeed be astonishing that Article 13 uses the term ‘civilian’ instead of a broader term such as ‘person’.Footnote 33 Second, if everyone is a civilian, the fundamental principle of distinction becomes meaningless and impossible to apply.Footnote 34 Third, Common Article 3 confers its protection on ‘[p]ersons taking no active part in hostilities, including members of armed forces who have laid down their arms and those [otherwise] placed hors de combat’. The latter phrase suggests that the mere fact of no longer taking active part in hostilities is not enough per se for such members of armed forcesFootnote 35 to be immune from attack: they must take additional steps and actively disengage. Fourth, on a more practical level, to prohibit attacks by government forces on clearly identified fighters unless engaged by those fighters is militarily unrealistic, as it would oblige the government forces to act purely reactively while facilitating hit-and-run operations by the rebel group. The objection that such rebels may be arrested anyway (and their resistance, if any, to arrest would be direct participation in hostilities) is convincing only if the rebels do not control territory.
There are two ways to conceptualize the conclusion that fighters may be attacked at any time until they disengage from the armed group. First, ‘direct participation in hostilities’ can be understood as encompassing the mere fact of remaining a member of the groupFootnote 36 or of retaining a fighting function.Footnote 37 Second, fighters can be considered not to be ‘civilians’ (who are entitled to protection against attacks unless and for such time as they directly participate in hostilities).Footnote 38 However, both interpretations raise difficult questions in practice. How are government forces to determine membership of an armed group so long as the individual in question commits no hostile acts? How can membership of the armed group be distinguished from simple affiliation with a party to conflict for which the group is fighting – in other words, membership in the political, educational or humanitarian wing of a rebel movement? One of the most convincing avenues envisaged is to allow attacks only against a person who either actually directly participates in hostilities or has a function within the armed groups to commit acts that constitute direct participation in the hostilities.Footnote 39
Customary humanitarian law provides no answer
According to the ICRC Study, in both international and non-international armed conflicts ‘[a]ttacks may only be directed against combatants’,Footnote 40 while ‘civilians are protected against attack unless and for such time as they take a direct part in hostilities’.Footnote 41 ‘Civilians are persons who are not members of the armed forces.’Footnote 42 The commentary clarifies, however, that the term ‘combatant’ in non-international armed conflicts simply ‘indicat[es] persons who do not enjoy the protection against attacks accorded to civilians’.Footnote 43 ‘While State armed forces may be considered combatants … practice is not clear as to the situation of members of armed opposition groups’.Footnote 44 Furthermore, the authors add, ‘[p]ractice is ambiguous as to whether members of armed opposition groups are considered members of armed forces or civilians’.Footnote 45 If they are the latter, an imbalance between such groups and government armed forces could be avoided by considering them to take a direct part in hostilities continuously.Footnote 46 Customary law is therefore as ambiguous as the treaty provisions on the crucial question of whether fighters in non-international armed conflicts may be attacked in the same way as combatants in international armed conflicts.
Arguments for and against an analogous application of the rule applicable in international armed conflicts
As mentioned above, the general tendency is to bring the law of non-international armed conflicts closer to that of international armed conflicts; this also has the positive side effect that controversies on whether a given conflict is international or non-international, and on what law to apply in conflicts of a mixed nature, are rendered moot. Even those who remain sceptical as to whether state practice has truly eliminated the difference to the extent claimed in the ICRC Study suggest that questions not answered by the law of non-international armed conflicts must be dealt with by analogy to the law of international armed conflicts, except in cases where the very nature of non-international armed conflicts does not allow for such an analogy (e.g. concerning combatant immunity from prosecution and the concept of occupied territories).Footnote 47 There is, moreover, no real difference between a non-international armed conflict such as the fighting between Sri Lankan government forces and the LTTE in northern Sri Lanka in 2008 and the international armed conflict between Eritrea and Ethiopia. To require soldiers in non-international (but not in international) conflicts to capture enemies whenever feasible rather than kill them is unrealistic on the battlefield. In addition, the decision when an enemy may be shot at must be taken in a split second by every soldier on the ground, and cannot be left to commanders and courts (as can the decision, discussed later, to intern a person). Clear instructions must exist. Whenever possible, the training of soldiers must be the same for both international and non-international armed conflicts in order to create reflexes that work in the stress of battle.
On the other hand, strong arguments question the appropriateness of applying the same rules as in international armed conflicts. Many non-international armed conflicts are fought against or between groups that are not well structured. It is much more difficult to determine who belongs to an armed group than who belongs to government armed forces. The positive humanitarian law of non-international armed conflicts does not even explicitly prescribe, as does the law of international armed conflicts, that fighters must distinguish themselves from the civilian population. Individuals join and quit armed groups in an informal way, whereas members of government armed forces are formally incorporated and formally dismissed. As armed groups are inevitably illegal, they will do their best not to be recognized as such and to conceal their militant nature. Claiming that fighters may be shot at on sight may therefore put many civilians in danger,Footnote 48 whether they are sympathizers of the group, members of the ‘political wing’, belong to the same ethnic group or simply happen to be in the wrong place at the wrong time. And while a clear distinction exists, in international armed conflicts, between the conduct of hostilities by combatants against combatants and law enforcement vis-à-vis civilians by the police, there is no equivalent clear distinction in non-international armed conflicts. Indeed, in a non-international armed conflict, insurgence always constitutes a crime under domestic law. Criminals should be dealt with by courts and may not be ‘punished’ by instant extrajudicial execution.
The arguments in favour of a different rule for non-international armed conflicts obviously apply mainly to armed groups and could therefore be seen as requiring a distinction between government forces and armed groups rather than between international and non-international armed conflicts. The problem is that to have the slightest chance of being respected, humanitarian law must be the same for both sides. The principle of the equality of the belligerents before humanitarian law also applies in non-international armed conflicts.Footnote 49
Rules on the peacetime use of force under human rights law
Human rights treaties prohibit the arbitrary deprivation of life. Most instruments, however, do not specify when a killing is arbitrary. Only the European Convention on Human Rights specifies that the deprivation of life will not be deemed arbitrary when it results from the use of force ‘absolutely necessary:
(a) in defence of any person from unlawful violence;
(b) in order to effect a lawful arrest or to prevent the escape of a person lawfully detained;
(c) in action lawfully taken for the purpose of quelling a riot or insurrection’.Footnote 50
In its case law outside armed conflicts, the European Court of Human Rights has admitted the lawfulness of killing a person the authorities genuinely thought was about to detonate a bomb, but found that the insufficient planning of the operation violated the right to life.Footnote 51 With regard to arrest, the Court has found that the life of a fugitive may not be put at risk for the purpose of arresting him if he poses no threat to life and is not suspected of a violent offence, even if he cannot be arrested otherwise.Footnote 52
By and large, other universal and regional human rights bodies take the same approach.Footnote 53 The UN Basic Principles on the Use of Force and Firearms by Law Enforcement Officials provide an authoritative interpretation of the principles that authorities must respect when using force, so as not to infringe the right to life. Those principles limit the use of firearms to cases of
self-defence or defence of others against the imminent threat of death or serious injury, to prevent the perpetration of a particularly serious crime involving grave threat to life, to arrest a person presenting such a danger and resisting their authority, or to prevent his or her escape, and only when less extreme means are insufficient to achieve these objectives.
The intentional lethal use of firearms is only admissible ‘when strictly unavoidable in order to protect life’. In addition, law enforcement officials
shall … give a clear warning of their intent to use firearms, with sufficient time for the warning to be observed, unless to do so would unduly place the law enforcement officials at risk or would create a risk of death or serious harm to other persons, or would be clearly inappropriate or pointless in the circumstances of the incident.Footnote 54
It must, however, be stressed that the Basic Principles are addressed to officers ‘who exercise police powers, especially the powers of arrest or detention’. Military authorities are included, but only if they exercise police powers,Footnote 55 which could be interpreted as meaning e contrario that the rules are not binding for military authorities engaged in the conduct of hostilities. If human rights are to provide an answer as to when a fighter may be killed, it would thus be imperative to know when military authorities, in a situation of armed conflict, are or should be exercising police powers.
The few precedents of human rights bodies in armed conflicts
Theoretically, human rights are the same in international and non-international armed conflicts and outside armed conflicts. The right to life is furthermore not subject to derogations except, under the European Convention, in respect of ‘lawful acts of war’.Footnote 56 The classic instance in which a human rights body has assessed the right to life in the context of an armed conflict is the Tablada case, concerning a group of fighters who attacked an army base in Argentina. The Inter-American Commission on Human Rights held that ‘civilians … who attacked the Tablada base … whether singly or as a member of a group thereby … are subject to direct individualized attack to the same extent as combatants’ and lose the benefit of the proportionality principle and of precautionary measures.Footnote 57 It then exclusively applied humanitarian law (applicable to international armed conflicts) to those attackers. Only civilian bystanders and attackers who surrendered were considered to be entitled to benefit from the right to life. The Commission did not raise the issue whether the fighters should have been arrested whenever possible, rather than killed.
In the Guerrero case, the Human Rights Committee found Colombia to have arbitrarily deprived persons of their right to life who were suspected – but not proved, even by the subsequent inquiry – to be kidnappers and members of a ‘guerrilla organization’. The police waited for the suspected kidnappers in the house where they had believed the victim of a kidnapping to be held, but which they found empty. When the suspected kidnappers arrived, they were shot without warning, without being given an opportunity to surrender and despite the fact that none of the kidnappers had fired a shot, but simply tried to flee.Footnote 58
A national human rights body, the Nepali Human Rights Commission, deemed that the Nepali security forces had seriously violated ‘humanitarian law’ when they shot at a group of armed Maoists forcing students and teachers at a local school to follow a ‘cultural programme’. Six Maoists and four students were killed and five students were injured. The Commission stressed that there was no firing by the Maoists when the army encircled them and that the Maoists were only lightly armed. In the Commission's view, after giving them a warning the army could easily have arrested them. It is not clear whether the Commission found a violation only because of the killed and injured students, or whether it considered that shooting at the armed Maoists without an attempt to arrest them also constituted a violation of ‘humanitarian law’.Footnote 59
The jurisprudence of the European Court of Human Rights in cases involving the right to life in the non-international armed conflict in Chechnya includes statements which appear to require that in the planning and execution even of a lawful action against fighters, any risk to life and the use of lethal force must be minimized.Footnote 60 While these statements were not limited to protection of the lives of civilians, the actual victims in the case were civilians. In all other cases in which human rights bodies and the ICJ applied the right to life to armed conflicts not of an international character, the persons killed were either hors de combat or not alleged to have been fighters.Footnote 61 Although fighters are very often killed, for example bombed, while they are not hors de combat, no such case has been brought before an international human rights monitoring body. Some observers have deduced from the absence of any such case law that such killings do not violate the right to life, it being ‘unthinkable’ that a case would be brought before the Inter-American system by a surviving relative of an FARC member.Footnote 62
The limited body of case law thus gives no conclusive answer as to what human rights law requires of government authorities using force against fighters.
Possible solutions under the lex specialis principle
First, we would like to emphasize that there is a good deal of common ground between the two branches of law. In a ‘battlefield-like’ situation, arrest is virtually always impossible without exposing the government forces to disproportionate danger. A fighter presents a great threat to life, even if that threat consists of attacks against armed forces. The immediacy of that threat might be based not only on what the targeted fighter is expected to do, but also on his or her previous behaviour.Footnote 63 In such situations lethal force could therefore be used, even under human rights law. On the other hand, the life of a fighter who is hors de combat is equally protected by both branches.
It is where the solutions found in the two branches actually contradict each other that the need to determine the applicable rule by reference to the lex specialis principle arises. The quintessential example of such a contradiction is that of a guerrilla leader shopping in a supermarket in the government-controlled capital of the country. Many interpret humanitarian law as permitting authorities to shoot to kill, since he is a fighter, but this is controversial. Human rights law would clearly say that he must be arrested and a graduated use of force must be employed, but this conclusion is based on precedents which arose in peacetime and human rights are always more flexible according to the situation.
In our view, some situations are more characteristic of those for which the humanitarian law rule was established, and others are characterized more by facts for which human rights were typically established. There is a sliding scaleFootnote 64 between the lone FARC member in a supermarket in Bogotá and the soldier in Franco's forces during the Battle of the Ebro. It is impossible and unnecessary to provide a ‘one size fits all’ answer; as shown above, the lex specialis principle does not determine priorities between two rules in the abstract, but offers a solution to a concrete case in which competing rules lead to different results. The famous ICJ dictum that ‘[t]he test of what constitutes an arbitrary deprivation of life … must be determined by the applicable lex specialis, namely the law applicable in armed conflicts’Footnote 65 should not be misunderstood. It has to be read in the context of the opinionFootnote 66 in which the ICJ had to determine the legality in abstracto of the use of a certain weapon.
Such a flexible solution, in which the actual behaviour required depends upon the situation, is dangerous – especially in our context, where it has to be applied by every soldier and leads to irreversible results. It is therefore indispensable to determine factors which give precedence either to the rule derived from the humanitarian law of international armed conflicts or to human rights.
The existence and extent of government control over the placeFootnote 67 where the killing occurs points to human rights as the lex specialis.Footnote 68 For government forces acting on their own territory, control over the place where the attack takes place is not a requirement for human rights to apply,Footnote 69 but simply a factor causing human rights to prevail over humanitarian law. The latter was designed to regulate hostilities against forces on or beyond the front line – that is, in a place not under the control of those who attack them, whereas law enforcement concerns persons under the jurisdiction of the enforcers. In traditional conflict situations the question here would be how far away the situation is from the battlefield,Footnote 70 although fewer and fewer contemporary conflicts are characterized by front lines and battlefields. What, then, constitutes sufficient control to warrant human rights predominating as the lex specialis? A government cannot simply argue that the presence of a solitary rebel or even a group of rebels on a stable part of its territory indicates that in fact it is not fully in control of the place, and therefore act under humanitarian law as lex specialis. The question is rather one of degree. If a government could effect an arrest (of individuals or groups) without being overly concerned about interference by other rebels in that operation, then it has sufficient control over the place to make human rights prevail as lex specialis.
This criterion of government control leaves the solution slightly more open in an area within the territory of the state whose government is fighting the rebels,Footnote 71 but which is neither under firm rebel nor under firm government control (such as regions of central Peru at the time of the Sendero Luminoso insurgency).Footnote 72 Even where the strict requirements of necessity under human rights law are not fulfilled (if they are, both branches of law lead to the same result), the impossibility of arresting the fighter,Footnote 73 the danger inherent in an attempt to do so,Footnote 74 the danger the fighter represents for government forces and civilians and the immediacy of this dangerFootnote 75 may cause the conclusion to be reached that humanitarian law is the lex specialis in that situation. These factors are interlinked with the elements of control described above. In addition, where neither party has clear geographical control, the higher the degree of certainty that the target is actually a fighter, the easier it is in our view for the humanitarian law approach to be regarded as lex specialis.Footnote 76 Attacks are lawful against persons who are actually fighters, while law enforcement is by definition directed against suspects.
Even where human rights prevail as the lex specialis in the context of armed conflict, humanitarian law remains present in the background and relaxes the human rights requirements of proportionality and warning once an attempt to arrest has been made unsuccessfully or is not feasible. By the same token, where humanitarian law prevails, human rights likewise remain present and require that an inquiry be conducted whenever a person has been deliberately killed.Footnote 77
If such a flexible approach to determining the lex specialis is accepted, the question arises whether it is valid for members of both armed groups and government forces. The applicable humanitarian law stipulates the equality of both parties, but they are not equal in terms of human rights obligations.Footnote 78 Indeed, it is controversial whether the latter apply to non-state actors.Footnote 79 Even if they are, they can only require from them certain conduct towards persons in an area under their control, since such actors have no jurisdiction over a territory.Footnote 80 In practice, government soldiers do not shop in supermarkets in rebel-controlled areas. Moreover, a government has the alternative of law enforcement and of applying domestic criminal law and is therefore able to plan an operation that maximizes the possibility of arresting persons.Footnote 81 Conversely, the question whether armed groups may enforce government legislation or legislate to make government action illegal is controversial. Requiring rebels to detain rather than kill could in some respects increase the asymmetry of conflicts, since detention by rebels is often seen as ‘hostage taking’ and demands are frequently made of them to release government forces they have captured.
It is therefore not unreasonable to consider armed groups as bound only by humanitarian law and domestic law (which in any case renders any killing by them unlawful), whereas government forces are bound by both humanitarian law and human rights, the latter prevailing in some situations and to a certain extent as lex specialis. The fact that rebels do not have human rights obligations limiting attacks on security forces does not mean that there are no humanitarian law limits on such attacks. While police forces cannot be considered to be civilians (as in international armed conflicts) if engaged in law enforcement operations to search for and arrest rebels,Footnote 82 attacks on police units not involved in a non-international armed conflict but performing normal peacetime police activities would violate the prohibition on attacking civilians.Footnote 83
The main problem with our solution is whether it is practicable in actual armed conflicts. If the permissible conduct varies, depending on the specific situation, how can a soldier know which rules to apply? This problem can be solved only by precise instructions and rules of engagement for each and every operation and sortie. In addition, guidelines might be developed at international level in discussions between humanitarian law experts, human rights experts, law enforcement practitioners and military representatives. Logically, (former) fighters should also be involved – especially if the guidelines also cover the conduct of such groups – to ensure that they can be applied in practice.
On which basis and following which procedures may a member of armed forces or groups be interned in non-international armed conflict?
In peacetime and during armed conflict, persons may be detained pending trial or after conviction for a crime. For the individual non-state actor, participation in a conflict constitutes a crime under the domestic law of the state affected by that conflict. A more specific feature of armed conflicts is that enemies may also be interned without criminal charge as a preventive security measure. In international armed conflicts this is the essence of prisoner-of-war status for combatants and internment for civilians.
However, the lack of clarity, particularly as to the legal safeguards to be applied by non-state actors in non-international armed conflict, gives cause for concern. The present section focuses on procedural regulation of this particular form of deprivation of liberty – internment.Footnote 84 By analysing the two branches of law and their complementary relationship, an attempt will be made to work out the basis and procedures for possible internment of a member of government armed forces or a member of an armed group (hereinafter: participant in hostilities) in relation to a non-international armed conflict.
The traditional answer of humanitarian law applicable to international armed conflict
Treaty-based humanitarian law applicable to international armed conflict provides rules procedurally regulating internment. The Third Geneva Convention permits the detaining power to ‘subject prisoners of war to internment’,Footnote 85 and the Fourth Geneva Convention allows for internment of a person ‘if the security of the Detaining Power makes it absolutely necessary’Footnote 86 or ‘for imperative reasons of security’.Footnote 87
Having stipulated when internment may occur, humanitarian law also indicates when it must cease. Some prisoners of war must be repatriated during armed conflict for medical reasons,Footnote 88 and all must be released and repatriated, without delay, after the cessation of active hostilities.Footnote 89 A civilian must be released ‘as soon as the reasons which necessitated his internment no longer exist’.Footnote 90 For those civilians not released during the armed conflict, it states that their ‘[i]nternment shall cease as soon as possible after the close of hostilities’.Footnote 91
Treaty-based humanitarian law applicable to international armed conflict recognizes that in the case of civilians, who, unlike prisoners of war, cannot be automatically interned for the duration of an armed conflict, compliance with the legal basis for internment requires an assessment to determine whether an individual continues to pose a threat to security and may remain interned, or no longer poses a threat and must be released. Accordingly the Fourth Convention lays down procedures for reviewing the internment of civilians, whether they are aliens in the territory of a party to the conflict or interned in occupied territory, designating the type of review body – either a court or administrative board – and providing for appeal and periodic review.Footnote 92 Protocol I introduces an additional safeguard into the process: the person interned is to ‘be informed promptly, in a language he understands, of the reasons why these measures were taken’.Footnote 93 Finally, it should be noted that unlawful confinement is a grave breach of the Fourth Convention.Footnote 94
The uncertain answer of treaty-based humanitarian law applicable to non-international armed conflict
In non-international armed conflict, treaty-based humanitarian law prescribes how persons deprived of their liberty for reasons related to the armed conflict must be treated, and lays down judicial guarantees for those undergoing prosecution for offences also related thereto, but it does not clarify for which reasons and by which procedures a person may be interned. Yet the drafters of Protocol II recognized the possibility of internment in non-international armed conflicts, as shown by the specific reference to internment in Articles 5 and 6 thereof.Footnote 95
Customary humanitarian law
Customary humanitarian law prohibits the arbitrary deprivation of liberty in both international and non-international armed conflicts.Footnote 96 But how is this prohibition to be understood with regard to internment in non-international armed conflicts – that is, what is the basis for it and with which procedures must there be compliance so that such internment does not amount to arbitrary deprivation of liberty? The non-binding commentaryFootnote 97 in the ICRC Study interprets this rule by significant reference to human rights. Applying the two criteria of the principle of legality, it states that the basis for internment must be previously established by law and stipulates two procedural requirements: (i) an ‘obligation to inform a person who is arrested of the reasons for arrest’; and (ii) an ‘obligation to provide a person deprived of liberty with an opportunity to challenge the lawfulness of detention’, described as the ‘so-called writ of habeas corpus’.Footnote 98
Human rights rules on the procedural regulation of security detention
Human rights provisions regulating the deprivation of liberty can be found in a variety of different treaties. Treaty-based human rights stipulate that a person may be deprived of liberty only ‘on such grounds and in accordance with such procedure as are established by law’.Footnote 99 All treaties prohibit arbitrary arrest or detention,Footnote 100 but only the European Convention specifically lists the bases for depriving a person of his or her liberty.Footnote 101 Treaty-based law articulates two procedures with which compliance is obligatory for a person to be lawfully deprived of his or her liberty when not specifically charged with a crime. First, an arrested person must be promptly informed of the reasons for arrest.Footnote 102 Second, any person deprived of liberty ‘shall be entitled to take proceedings before a court, in order that that court may decide without delay on the lawfulness of his detention and order his release if the detention is not lawful’.Footnote 103
Unlike the four Geneva Conventions, however, none of these treaties is universally ratified, most are regional instruments, and in certain circumstances derogations from the provisions relevant to security detention are possible.Footnote 104 It would therefore be helpful to have an assessment of the status of universally applicable customary human rights, as well as whether and to what extent such rules are derogable. The same applies to the judicial review requirement, the so-called writ of habeas corpus, because even though it or at least some of its aspects are widely claimed, by among others the Human Rights Committee, to be non-derogable, this is not stipulated in treaty-based human rights law.Footnote 105
The few precedents of human rights bodies in armed conflicts
Human rights bodies have elaborated upon the content of human rights through their jurisprudence, which does not exist to the same extent for humanitarian law as there are fewer expert bodies mandated to apply it.Footnote 106 Thus recourse to human rights is appealing when interpreting humanitarian law, but it must be borne in mind that such jurisprudence is often not binding or not binding universally. In addition, precedents of human rights bodies specifically addressing both aspects, the procedural regulation of security detention and in relation to armed conflict, are rare.
The European Court of Human Rights has upheld state security detention during situations described as ‘terrorist campaign[s]’;Footnote 107 it remains to be seen how the Court would assess whether the measures adopted by the state – after derogation – do or do not go beyond ‘the extent strictly required by the exigencies’Footnote 108 of a situation characterized as armed conflict. In relation to the ‘war on terror’ and detention at Guantánamo Bay, the Human Rights Chamber for Bosnia and Herzegovina found that the state had breached the European Convention ‘by handing over [persons] into illegal detention by US forces’, as information on the basis for their detention was neither sought nor received.Footnote 109
Human rights bodies of the Inter-American system have addressed issues relating to armed conflict, but few refer to security detention. In Coard v. US, the Inter-American Commission found detention without any review, during an international armed conflict, ‘incompatible with the terms of the Declaration as understood with reference to art. 78 of Geneva Convention IV’.Footnote 110 The Commission stated that,
pursuant to the terms of Geneva Convention IV and the Declaration, [the necessary review of detention] could have been accomplished through the establishment of an expeditious judicial or board (quasi-judicial) review process carried out by US agents with the power to order the production of the person concerned, and release in event the detention contravened applicable norms or was otherwise unjustified.Footnote 111
In the said case there was no discussion, however, of applicable rules for non-international armed conflict. In 2002 the Commission – in the context of the United States considering ‘itself to be at war with an international network of terrorists’ – adopted precautionary measures requesting the United States ‘to take urgent measures necessary to have the legal status of the detainees at Guantánamo Bay determined by a competent tribunal’ rather than by a political authority, in order to ensure that the detainees ‘are afforded the legal protections commensurate with the status that they are found to possess, which may in no case fall below the minimum standards of non-derogable rights’.Footnote 112
Application of human rights as lex specialis
When comparing the treaty-based provisions regulating internment in both branches of law, human rights are more elaborate than the sparse treaty-based humanitarian law rules applicable to non-international armed conflict. However, this is not the case when comparing human rights law with humanitarian law applicable to civilians in international armed conflict.Footnote 113 For example, humanitarian law explicitly guarantees, for civilian internees, a right to appeal and a time frame for periodic review if not released; treaty-based human rights do not specify this with regard to security detention. Moreover, the relevant human rights provisions are derogable, at least under the formal terms of the treaties; these provisions of humanitarian law, at least in occupied territories, are not derogable.Footnote 114 But human rights law does stipulate that the body reviewing the deprivation of liberty must be judicial, whereas humanitarian law provides for either a court or an administrative board in the case of civilians but grants no review at all for prisoners of war, except in determining prisoner-of-war status.Footnote 115
As humanitarian law applicable to non-international armed conflict is silent on the procedural regulation of internment, it would seem clear that in accordance with the lex specialis principle as a maxim of logic, human rights should step in to fill the gap. The ICRC Study appears to adopt this approach when it interprets the humanitarian law rule prohibiting the arbitrary deprivation of liberty through the lens of human rights.Footnote 116 The fact that internees – unlike persons to be attacked – are always under the control of those who detain them, reinforces the human rights approach. In this formalistic application of the lex specialis principle, the only exception where humanitarian law would take precedence is when the parties to a non-international armed conflict agree theretoFootnote 117 or when the government unilaterally concedes prisoner-of-war status to captured fighters. Despite the appeal of this formal lex specialis approach, difficulties remain.
Difficulties involved in applying human rights
The human rights requirement that internment be subject to judicial review – if no derogation is made, or cannot be made as the rule is found to be customarily non-derogable – demonstrates some of these difficulties. One is whether it is realistic to expect states and non-state actors, possibly interning thousands of people, to bring all internees before a judge without delay during armed conflict. If it is not, such an obligation risks severely hampering the effective conduct of war and could thus lead to less compliance with the rules in the long term – for example, summary executions disguised as battlefield killings.
A second concern stems from the differences between state and non-state actors, which have equal obligations under humanitarian law but not under human rights. Even if the question of whether non-state actors are bound by human rights were settled, how is the non-state actor to meet the obligation of judicial review? The question of whether a non-state actor may establish a court remains controversial.Footnote 118 The requirements that there be a legal basis and procedures established by law for internment give rise to the same concern. While human rights themselves set at least two procedural requirements, neither they nor humanitarian law applicable to non-international armed conflict provide a specific legal basis for internment. A state can do so in its domestic law, but how is the non-state actor to establish that legal basis?
Even if human rights are found to be binding on some non-state actors – that is, those ‘capable of fulfilling the conditions to exercise those rights’,Footnote 119 the difficulties may not be fully resolved and additional ones may arise. For example, could a non-state actor derogate from human rights? What is the standard to determine which non-state actor qualifies to do so and which does not? And this approach would not, of course, resolve the difficulties remaining with regard to those non-state actors which do not ‘qualify’.
Application of human rights seems to make it impossible for one party to an armed conflict – the non-state actor – to intern legally. Parties to armed conflicts intern persons, hindering them from continuing to bear arms, so as to gain the military advantage. If the non-state actor cannot legally intern, it is left with little option – for it is a serious violation of humanitarian law to deny quarterFootnote 120 – but to release the captured enemy fighters. However, if rules applicable to armed conflict make efficient fighting impossible, they will not be respected, undermining any protection the law provides.Footnote 121
Apply the humanitarian law of international armed conflicts by analogy?
In view of the complications generated by applying human rights, perhaps the correct solution requires applying the humanitarian law of international armed conflict to non-international armed conflicts by analogy.Footnote 122 The humanitarian law applicable to non-international armed conflict indicates that internment occurs in such conflicts, but contains no indication of how it is to be regulated. Such regulation is necessary so that internment can take place in practice. While the provisions regulating internment in international armed conflict are set out according to protected person categories, ‘no fundamental difference between the regimes applicable to the two situations prohibits the application of those same’ provisions, as long as the rules are applied according to the person's function rather than status.Footnote 123 Such a legal analogy is also consistent with the noted gradual erosion of the distinction between the law applicable in international and non-international armed conflicts.
Applied by analogy, the humanitarian law of international armed conflict allows for internment and provides for review procedures. It leaves open the option for the internment review to be carried out by an administrative board or a court. Given the organizational criteria a non-state actor must meet to be considered a party to an armed conflict,Footnote 124 any non-state actor should be able to fulfil these obligations, particularly as it can opt for an administrative review procedure rather than review by a court.
The humanitarian law of international armed conflict applies if ‘the government of the State affected by the non-international armed conflict [has] claimed for itself belligerent rights’.Footnote 125 While ‘the recognition of belligerency has fallen into disuse as a legal concept’,Footnote 126 application by analogy appears to continue to make sense; Common Article 3 moreover urges parties to non-international armed conflicts to agree on such application. The ICRC Study in fact indicates application by analogy in such conflicts of the standards of the Fourth Geneva Convention to civilians and those of the Third Geneva Convention to persons designated as ‘combatants’.Footnote 127 Making this distinction between ‘combatants’ and civilians in non-international armed conflict would also be consistent with current discussions by experts on use of the ‘membership approach’ to interpret ‘direct participation in hostilities’ in such conflict.Footnote 128 As analogous application does not confer any status, there would still, for example, be no combatant immunity. The Geneva Conventions, while allowing for internment, would therefore not prevent the repression of acts prohibited by domestic law. Nor would application by analogy of the Third Convention to members of armed forces and groups, as suggested by the ICRC Study, entitle them to any review procedure; such procedural regulation could only be found in the Fourth Convention.
Does the analogous application of the law of international armed conflicts sufficiently consider, however, the fundamental distinction between that law and the law of non-international armed conflict – that is, that the rules applicable to international armed conflict generallyFootnote 129 apply only to protected person categories, such as prisoners of war or enemy civilians, and that no such categories exist in non-international armed conflict? Even if the distinction in non-international armed conflict could be made by function rather than status, on which criteria should the assessment of a civilian or ‘combatant’ be based? Should ‘combatants’ be measured against the criteria in Article 4 of the Third Convention or Article 44 of Protocol I, or perhaps through the ‘membership approach’?Footnote 130 Would Article 5-type tribunalsFootnote 131 need to be instituted in non-international armed conflicts to make the determination?
Apply the Fourth Convention rather than the Third Convention?
If the Third Geneva Convention is applied by analogy, a participant in hostilities could be detained without any individual determination for the whole duration of the conflict.Footnote 132 However, as seen above, it is much more difficult in non-international than in international armed conflicts to determine who is actually a fighter. Such a determination must therefore be made on an individual basis. It is also much harder to determine the actual end of hostilities than in an international armed conflict between states that may conclude a ceasefire or surrender. All this may support application, if at all, of the law of international armed conflict to non-international armed conflict by analogy to the Fourth Convention alone, as there are no combatants and hence no concomitant prisoner-of-war status in non-international armed conflict. Analogy with the Fourth Convention could be founded on a determination of the lex specialis according to the overall systemic purposes of the international legal order. In non-international armed conflict it would avoid internment of persons without review for the duration of the conflict.
Apply both branches of law in parallel?
Can an analogy legally be made to humanitarian law applicable to international armed conflict when doing so displaces specific human rights rules and thus directly overrides the formal lex specialis maxim? As human rights addressing security detention do exist, such an analogy on this issue may be unlike analogies made, for example, to the definition of ‘military objectives’, a term unique to humanitarian law. Furthermore, analogy to the humanitarian law of international armed conflict may not avoid the difficulties raised by human rights, as these may still apply. For example, if humanitarian law procedures applicable to international armed conflict that regulate internment of civilians are inadequate, human rights may be required to fill the gap.Footnote 133
Consideration of the manner in which human rights complement humanitarian law may help to find a solution for some of the difficulties encountered when applying human rights in non-international armed conflicts, particularly to non-state actors.
[I]f IHRL is not used to interpret an IHL rule, but instead IHRL [is applied] … ‘[parallel] to’ IHL … IHL would apply to parties to the conflict, State and non-State actors, and IHRL would continue to apply to State actors, as it was traditionally designed to do. This avoids the problematic application to non-State actors, and, yet, mandates States to continue to meet their international legal obligations. One may claim this is unfair, as States would need to abide by additional obligations than non-State actors in a non-international armed conflict. This is true at the international law level, where States are traditionally the legal actors, but it would only be true of the rules to which the States obligated themselves. Also, it must not be forgotten that non-State actors remain bound by domestic law.Footnote 134
This approach remains consistent with the understanding that both bodies of law are complementary and implemented through the maxim of lex specialis, and it respects the maxim of lex posterior, as most human rights developed after the main humanitarian law provisions discussed here. Also, the concern about impracticable obligations arising, such as a judicial review of internment without delay in all cases, remains limited, as human rights treaties include derogation clauses providing states with a regulated ‘way out’ in such situations.
However, reliance solely on human rights to compensate for the lack of procedural regulation of internment in non-international armed conflict may offer only a minimal solution, for unless supplemented by customary lawFootnote 135 it clarifies the obligations of states only if the state is party to the human rights treaty, and it is limited by any derogation made. In addition (but this issue is not covered in the present article) the extent to which human rights apply extraterritorially is controversial, which weakens the solution relying on the application of human rights law to non-international armed conflicts abroad. Thus human rights may provide no regulation of state internment. As for the non-state actors – if they are bound at all, they will face conceptual difficulties in complying.
So perhaps the solution is not either to apply human rights or to draw an analogy with the humanitarian law of international armed conflict, but to harmonize appropriately both approaches. The recognized relationship between the two branches of law, the objectives of both, as well as the differences between states and non-state actors, may in fact call for this complementary approach.
Application to non-international armed conflicts of the rules of the humanitarian law of international armed conflicts on civilians provides procedures for internment that are binding on both the state and the non-state actor, as well as places a constraint on the basis for internment. These obligations apply equally to all parties to the conflict. Such application does not preclude the corresponding parallel application of human rights. Thus, if application of the latter is required to provide further details about the regulation, for example of civilian internment,Footnote 136 consistent with the maxim of lex specialis, human rights may step in and clarify state obligations, particularly with regard to judicial review. While derogation from human rights may still occur, analogous application of the humanitarian law of international armed conflicts – from which no derogation is possibleFootnote 137 – would set a baseline below which no derogation from the human rights rules could go, as humanitarian law provisions already apply in situations most threatening to a nation. This complementary approach would ensure procedural protection for interned non-state actors and would bring much needed clarity to the non-state actor's obligations, thus also procedurally protecting members of the state's armed forces captured in non-international armed conflict. The practicality of this approach, however, does not make it legally binding and the concerns mentioned above remain, particularly as to the appropriateness of drawing an analogy to the humanitarian law of international armed conflict and – within that law – to the Fourth Convention rather than the Third Convention.
To sum up, unfortunately no ideal, straightforward solution exists. While the application of human rights is more satisfactory under the formal lex specialis principle, it is unrealistic, particularly for the non-state armed group. Moreover, the content of the applicable human rights is controversial. To apply by analogy the regulation provided for civilians by the humanitarian law of international armed conflicts may therefore correspond better to the overall systemic purposes of both branches. In practice, the only difference between the two approaches is that the humanitarian law approach is satisfied by an administrative review, while the human rights approach stipulates judicial review. Most people detained in relation to a non-international armed conflict would already greatly appreciate having any opportunity at all for an independent and impartial review of their internment. The establishment of guidelines to assess the relationship between humanitarian law and human rights law in a given situation could be helpful, particularly for the practitioner. Such guidelines would help to ensure consistent solutions that do not compromise the two branches' specific objectives and purposes.Footnote 138
Conclusion
On both issues discussed in this article, the result of a formal application of the lex specialis principle is not entirely satisfactory. The sliding scale for the respective application of human rights and international humanitarian law to attacks against fighters in non-international armed conflicts leads to much uncertainty in borderline situations and establishes different obligations for the government and the rebels. To give captured fighters the benefit of habeas corpus as defined by human rights may be unrealistic and even counterproductive in many conflicts and raises serious conceptual problems when required from armed groups. While these limits have to be accepted with regard to the use of force and must be mitigated by instructions and guidelines, they may be overcome in the procedural regulation of internment by applying by analogy the regime foreseen in the humanitarian law of international armed conflicts for civilians. In both cases, humanitarian law applies as a minimum whenever human rights, for whatever reason, do not apply.
