A Swedish District Court (SD Court) has recently ruled that non-state armed groups have the capacity under international law to establish courts and carry out penal sentences, but only under certain circumstances. While the issue has been widely debated by legal scholars over the past decade (Somer, Sivakumaran, Hakimi), this may very well be the first time that any domestic or international court has made a definitive ruling.
The implications at stake are as clear as the facts of the case. A member of an armed group admits to executing enemy detainees, but argues his actions were lawful as he was carrying out a sentence to punish war criminals as a result of a fair trail of a legitimate (but non-state) court. Notwithstanding the veracity of the claim, does this act amount to summary execution or the execution of justice?
International Humanitarian Law (IHL) prohibits the passing of sentences without fair trail guarantees for acts or omissions related to an armed conflict. For armed groups, this poses two existential challenges to the establishment of criminal courts. First, common article 3 to the Geneva Conventions requires courts to be ‘regularly constituted’. Second, the due process principle of legality (nullum crimen sine lege) requires that criminal offenses be established ‘under the law’.
The SD Court quite remarkably rules that armed group may establish courts in principle, but then seemingly aware of the vast public policy implications of this decision, attempts to rein it in by imposing conditions on armed group trials that seem more attuned to the court’s policy concerns than sound legal reasoning.
On the 1st of May 2012, the defendant, Haisam Sakhanh, joined a Syrian armed opposition faction known as ‘Firqat Suleiman el-muqatila’. Three days later, eleven suspected members of the Syrian regime armed forces were captured. A Free Syrian Army Military Council ordered the release of two and the detention of the other nine. A Judicial Council was then convened. It consisted of three members who had previously served as judges in the regime prior to defecting to the FSA. Sakhanh was not present but heard from others that seven of the alleged perpetrators were found guilty of rape and murder of civilians and sentenced to death under existing Syrian law, while two others were acquitted. Sakhanh was ordered by his commander to carry out the death sentence. He argues that he agreed to do so under his presumption that the order followed the pronouncement of a sentence following a fair trial. Approximately 36 hours had elapsed between the time of capture and the carrying out of the sentence.
The SD Court first determines that the act took place in the context of a non-international armed conflict (NIAC), and then turns to the question of whether an armed group may establish courts. After accepting that the term ‘regularly constituted court’ from common article 3 may give the impression that only a state can establish courts, the SD Court finds that reference to the Additional Protocols and their commentaries indicates that the focus has shifted from how a court is established to whether it upholds fundamental procedural guarantees of impartiality and independence. This view is supported, particularly in the criminal law context, by reference to the Elements of Crime of the International Criminal Court.
The SD Court then rules that since IHL requires armed groups to refrain from inhumane acts such as murder and torture, it also makes demands on them to maintain discipline in their own ranks. Therefore, an armed group must be able to establish courts, but the legal capacity to do so is limited to i) uphold discipline in the actions of its own armed forces and ii) uphold law and order on a given territory under the condition that the court is staffed by personnel who were appointed as judges or officials in the judiciary prior to the outbreak of conflict, AND that the court applies the law which was in effect before the conflict, or at least does not differ substantially in a stricter direction from the law that existed before the conflict. Of course any trial must fulfil due process standards.
One can certainly understand the policy reasons for the SD Court’s decision. Amid a chorus of calls for all parties to conflict to ensure respect for IHL and accountability for war crimes committed within an armed group’s ranks, the SD Court likely did not want to be seen as closing this door to armed group internal accountability. As that door probably looked an awful lot like a floodgate, the SD Court likely felt compelled to manage the flow by imposing strict limitations. First, the ruling maintains the core integrity of the state’s monopoly on justice, and more generally pushes back against an armed group’s effort to gain legitimacy by displacing the state. Second, it mitigates against what one UN independent expert has called “a veneer of legality to what would better be termed vigilantism”. Third, it reduces the chances that unqualified or partial judges will be appointed. Fourth, it ensures that populations will not be subjected to partisan laws, or for example religious laws that deviate from the state’s legal foundations (although the tables could be turned in a conflict where a theocratic state faces secular opposition). Finally, it ensures that populations will not find themselves under two competing sets of law with which it is impossible to comply—for example, if an armed group imposes conscription, to refuse would run contrary to rebel law, while to comply would run foul of government law prohibiting insurrection.
Yet there are also compelling policy considerations against such existential limitations to armed group courts. First and foremost, it will be difficult for those engaging armed groups to sell this outcome. The state apparatus can capture and try rebel fighters for war crimes or even for merely participating in hostilities (as PoW status does not exist in NIAC), but armed group ‘authorities’ will be considered international war criminals if they do the same. And armed groups are often reluctant to apply the rules of their enemy, especially if the conflict is ideologically based. Second, the capacity of an armed group to administer justice depends on the somewhat arbitrary factor of whether it has access to regime judges (rather than whether judges are independent, impartial and competent). It may also put judges in harms way, as rogue regimes may prefer to see judges indisposed of rather than legitimizing opposition courts. Third, the assumption is that the state penal laws respects and protects human rights, which is often not the case and may be a root of the conflict (see Hungary statement below). Imagine if women are not protected from all forms of sexual violence under the law of the state? Fourth, the international community has called on armed groups to bring perpetrators of international crimes to justice (p. 5). In theory there are other options for armed groups such as handing over suspects to 3rd states, but in practice it is rare to find any takers. Fourth, armed group fighters accused by their own courts should be entitled to the same level of due process protection as others (par. 547). To rule that they may face sanction by a court from which others are immune is to diminish their protection before the law.
From a legal standpoint, the SD Court’s analysis was a mixed bag. It did well in ruling that the archaic ‘regularly constituted court’ requirement of common article 3 has been replaced by the ‘independent and impartial’ criteria.
However, on the issue of an armed group’s capacity to enact laws, the judgment seems to be derived from an erroneous—or at least incomplete—consideration of the term ‘law’ in the context of NIAC. The SD Court reasoned that, “it clearly follows from [the international customary law of legality] that it is only states, which in accordance with their respective constitutional rules, may impose criminal sanctions” (par. 30, my translation). This assertion may be true under human rights law in times of peace, where the state is the only game in town, but it does not hold up in the IHL of NIAC, where competing authority is the name of the game, and where the meaning of ‘law’ is all but clear. The ICRC commentaries to common article 3 (par. 692) and Additional Protocol II (par. 4605) contemplate the possibility of parallel state and armed group laws. In his seminal work on the law of NIAC (p. 561), Sivakumaran identifies the UK, Hungary and the USSR (arguably) as states that accept ‘law’ to also include armed group law. Hungary, lending credence to the policy argument against the SD Court ruling above, has stated that armed group law “may be more in harmony with the demands of the time and more humanitarian than the laws in force at the beginning of the conflict”.
Another problematic aspect of the judgment is that the SD Court provides no justification for the requirement that judges must have been part of the state apparatus prior to the outbreak of hostilities.
The decision also has implications for the legal basis of detention in NIAC in light of recent jurisprudence. Surely it would be absurd if armed groups are granted the legal capacity to hold trials under certain circumstances but lack a legal basis to detain.
The SD Court should be commended for taking up the challenge head on. It could have simply ruled that due process cannot be achieved in 36 hours and ignored the existential issue of whether an armed group can establish courts (interestingly, in the case at hand, it seems that the armed group court allegedly meets the existential requirements as it purports to be composed of judges who defected and who applied Syrian law). The fact that there is now a decision for other domestic and international courts to consider and legal scholars to debate is commendable.
An alternative approach could be to use an analogy to the law of occupation as a starting point. According to articles 64-67 of the 4th Geneva Convention, an occupying power: i) is barred from promulgating new laws other than to comply with IHL (and arguably human rights standards), to maintain orderly government, or to ensure the security of the occupying power; ii) may establish courts to consider the penal provisions it enacted; and iii) must uphold the principle of legality by publicising new laws in the language of the inhabitants and not applying them retroactively. Of course there are many reasons to argue that armed group control of territory is not analogous to occupation by another state. Still, a substantive discussion on such questions should be welcomed.
In an ideal world, armed groups would not exist and would not establish courts. But armed conflict is the antithesis of an ideal world. If legal norms are to have a chance of mitigating the harm of armed conflict, they must be non-prejudicial and realistic. Currently, an expert process established through the University of Manchester International Law Centre and the Syrian Legal Development Programme is working to develop guidelines for fair trial standards of armed group courts. While not a legal exercise, the aim is to provide practical guidance for armed groups and those engaging with them. This is good strategy even for those who prefer not to see a proliferation of armed group courts. The more armed groups are aware of the sophisticated processes they must have in place, the harder it will be for them to claim their own sub-par courts as fair. The SD Court judgment, which looks to be heading to appeal, makes the need for such an initiative all the more clear.