The Ituri Conundrum: Qualifying Conflicts between an Occupying Power and an Autonomous Non-State Actor

Last week, Trial Chamber VI of the International Criminal Court (ICC) issued the long-awaited judgment in the Ntaganda case. The judges found the defendant guilty on all 18 counts, including the ICC’s first ever conviction for sexual slavery. Although the Chamber is yet to resolve matters related to sentencing and reparations, the decision marks an important milestone in the proceedings, which began with an arrest warrant issued back in August 2006 (Mr Ntaganda surrendered himself to the ICC in March 2013).

Readers of this blog will be familiar with the case as well as with some of the controversies surrounding its progress. In brief, Bosco Ntaganda was the Deputy Chief of Staff of the Patriotic Forces for the Liberation of Congo (FPLC), the armed wing of the Union of Congolese Patriots (UPC). The UPC/FPLC was one of the armed groups involved in the so-called Ituri conflict, which took place between 1999 and 2003 in the Ituri region in the north-eastern Democratic Republic of the Congo (DRC). Before the ICC, Mr Ntaganda was charged with 13 counts of war crimes and five counts of crimes against humanity, all allegedly committed in Ituri between 2002 and 2003.

The judgment, which fills over 500 pages, no doubt deserves careful scrutiny before any general pronouncements can be made as to its overall quality and rigour. Instead of analysing the judgment as a whole, this post focuses on a narrow question related to the Chamber’s legal qualification of the conflict in Ituri at the material time (discussed in paras 699–730 of the judgment). In particular, I am going to look at how international humanitarian law (IHL) qualifies conflicts between an occupying power and an autonomous non-State actor. The analysis builds on my research into complex conflict situations, which was published as part of my recent book on Internationalized Armed Conflicts in International Law (OUP 2018, especially chapter 3).

The situation in Ituri between 2002 and 2003 was notoriously convoluted, with a contemporary Human Rights Watch report describing it as “one of the world’s most complex conflict areas” (at page 14). At various times, the hostilities involved several States including the DRC, Uganda, and Rwanda, as well as a plethora of non-State actors, including Mr Ntaganda’s own UPC/FPLC. Crucially, this factual pattern included an unusual combination of an outside occupying power (in this case, Uganda) pitted against an autonomous non-State armed group (here the UPC/FPLC), which poses a difficult challenge for the rules on conflict qualification under IHL.

Why does it matter?

Qualifying situations of armed violence essentially presents a choice between three different options. Every such situation may qualify as (1) an international armed conflict (IAC), (2) a non-international armed conflict (NIAC), or (3) a situation to which IHL does not apply. In the context of international criminal trials such as the Ntaganda proceedings, a finding of either (1) or (2) is a necessary precondition for any war crime conviction. This is because war crimes are a category of (serious) violations of IHL and thus if IHL does not apply to a particular situation, then any conduct committed in that situation, however heinous, cannot constitute a war crime.

Additionally, the distinction between (1) and (2) is much more than semantic. Although there has been some convergence of the rules regulating these two types of conflict, the IAC/NIAC dichotomy remains a defining feature of IHL (for a more detailed discussion, see pages 9–23 of the introduction to my book). Thus, for example, notions of combatant status or belligerent occupation are applicable as a matter of law only in IACs. Similarly, since Article 8 of the ICC Statute has different lists of war crimes for IACs and NIACs, the applicability and elements of a given war crime may differ based on the legal qualification of the conflict at hand—as was the case in Ntaganda.

Specifically, the charges against Mr Ntaganda included the war crime of enlisting and using children (counts 14–16). With respect to NIACs, the relevant definition speaks of conscription or enlistment of children “into armed forces or groups” (Article 8(2)(e)(vii) ICC Statute). By contrast, the corresponding definition applicable in IACs refers to “national armed forces” (Article 8(2)(b)(xxvi) ICC Statute). The defence seized on this difference and argued that it presented one of the key reasons why conflict qualification mattered in the Ntaganda case. The defence counsel expressly argued (at page 60) that if the situation in Ituri qualified as an IAC, “Mr Ntaganda cannot be found guilty of crimes of enlistment, conscription and use of child soldiers because the FPLC is not a national armed force” (emphasis mine).

Occupying power vs. non-State actor

It will be obvious at this point that the defence argument did not succeed at trial. The Trial Chamber held that all hostilities involving the UPC/FPLC at the material time qualified as a NIAC, irrespective of who its adversary was in any given context (see especially paras 725 and 730 of the judgment). Consequently, enlistment or conscription of children into the UPC/FPLC met the disputed element of the said war crime as the UPC/FPLC was found to “constitute an armed group for the purpose of Article 8(2)(e)(vii)” (at para. 1117; emphasis mine).

To an extent, the finding of a NIAC is unsurprising. After all, the UPC/FPLC was a non-State actor mostly engaged in fighting other non-State actors in the territory of a single State, the DRC, who did not actively oppose it (cf. para. 728) )—and it is uncontroversial that confrontations “between such groups within a State” qualify as NIACs (Tadić Jurisdiction Appeal Decision, para. 70), provided the requirements of organization and intensity are met (eg Limaj Trial Judgment, para. 84). Similarly, while the UPC/FPLC did receive some outside support—most prominently from Rwanda—the Chamber held that this assistance did not reach the level of overall control required for the existence of an IAC involving the outside State (at paras 729–30). However, one element of the complex fact pattern seemed to pull in the opposite direction, suggesting that at least some of the hostilities in Ituri could have qualified as an IAC.

This conundrum arises out of Uganda’s presence as an occupying power in the Ituri district at the time. As of 2019, this is a well-established fact arrived at through extensive international litigation not limited to the ICC. In particular, in 2005, the International Court of Justice held that “Uganda was the occupying Power in Ituri” (Armed Activities on the Territory of the Congo judgment at para. 178). Various later decisions of the ICC have then confirmed this interpretation (Lubanga Confirmation of Charges decision at para. 220; Katanga Trial Judgment at para. 1205; Ngudjolo Trial Judgment at para. 494). Accordingly, the Ntaganda Trial Chamber also acknowledged Uganda’s status as a belligerent occupier, but in unpicking its implications, it got itself into some murky waters (at para. 728; all emphases mine):

On one side, by virtue of its presence as an occupying force in Ituri, and as it was fighting in the DRC’s territory without its consent, the UPDF [= Uganda People’s Defence Forces] was engaged in an international armed conflict with the DRC. On the other side, with regard to the clashes between the UPDF and the UPC/FPLC, the Chamber finds that, without evidence to the effect that the UPC/FPLC was fighting under the overall control or on behalf of a State, and given that it cannot be established that said fighting took place within the area of the DRC under effective control of the UPDF, the fighting constituted a non-international armed conflict. In any event, had the fighting taken place within the area considered as occupied by the UPDF, the law of occupation would only apply to the UPDF, and not to the UPC/FPLC.

At its core, the relevant legal question considered in this long passage can be expressed as follows: What is the correct legal qualification of a conflict between an occupying power (such as the UPDF in this case) and a non-State armed group (such as the UPC/FPLC) which is not under the overall control of another State and is operating from the territory of the occupied State (here, the DRC)?

Unravelling the conundrum

In my view, the Chamber’s overall conclusion is correct. The notion of an IAC presupposes the existence of a confrontation (or, as per the Tadić Jurisdiction Appeal Decision, “a resort to armed force”) between two States. Thus, if a non-State actor fighting against a State is operationally autonomous (more on this term here at pages 99–104) in respect to any other third State, it would be wrong to describe the non-State/State conflict pair as an IAC. The belligerent relationship between an armed group such as the UPC/FPLC and a State such as Uganda is thus subject to the law of NIAC.

There is no reason to modify this conclusion even if the belligerent State (= Uganda) happens to be in occupation of a territory belonging to another State (= the DRC). To be sure, due to the operation of Common Article 2(2) to the Geneva Conventions, the establishment of occupation triggers the application of the law of IAC as between the occupying and the occupied States (= between Uganda and the DRC). It also means that, from that point onwards, the law of occupation protects persons who find themselves in the hands of the occupying power (Article 4 Geneva Convention IV). Some of them may even be members of the non-State group in question (= UPC/FPLC). However, the occupying power is bound by the law of occupation vis-à-vis these persons on account of its conflictual relationship with the occupied State—and not because of their association with the non-State party to the separate conflict between the occupier and that party. Accordingly, the confrontation between the occupying power and the autonomous non-State group is governed by the law of NIAC, while the duties of the occupying power towards the occupied population are subject to the law of IAC pertaining to belligerent occupation.

Admittedly, it is sometimes said that the object and purpose of IHL militates in favour of classifying the entirety of a complex situation of this kind as an IAC because the law of IAC, the argument goes, provides a more extensive range of protections than the law of NIAC (eg Cassese, International Law (2nd edn, OUP 2005) 420). However, as the strategy of the defence team in this case well illustrates, there are issues (such as the prohibition on enlistment of child soldiers) with respect to which it is the law of NIAC that is in fact more protective. In his excellent book The Law of Non-International Armed Conflict (OUP 2012), Sandesh Sivakumaran has identified other such examples, including the prohibition on deportation and the protection of works containing dangerous forces (at pages 68–69). In addition, the presumption in favour of a more protective regime seems to be at odds with the criminal law maxim in dubio pro reo, which requires that any doubts should be resolved in a manner favourable to the defendant (eg Bartels (2017) 20 YIHL 37, 60–61). All in all, the teleological approach may offer some help, but it should not be a substitute for careful analysis.

In my view, the Trial Chamber was thus correct to conclude that Mr Ntaganda’s conduct had to be measured against the relevant rules of the law of NIAC. By contrast, I disagree with the formulation of the penultimate sentence of the quoted paragraph insofar as it may be taken to mean that if the relevant fighting had occurred “within the area of the DRC under effective control of the UPDF”, it would not have “constituted a non-international armed conflict”, ie, presumably it would have amounted to an IAC. To my mind, the issue of effective control over territory is relevant for the geographical scope of the law of occupation (which the Chamber itself seems to acknowledge in the final sentence of the paragraph). However, as I argued above, the existence of a belligerent occupation only affects the obligations of the occupying power, but it does not change the legal qualification of its conflict with an autonomous non-State armed group. Consequently, whether the fighting between them takes place inside or outside the occupied territory is immaterial for its classification under IHL.

The answer to the “Ituri conundrum” therefore lies in the careful untangling of the individual strands of a complex conflict situation like the one at the core of the Ntaganda case. The presence of an occupying force in a foreign territory triggers the law of belligerent occupation for the tripartite relationship between the occupier, the occupied State, and the occupied population. However, the belligerent relationship between the occupying power and an operationally autonomous non-State armed group remains governed by the law of NIAC, irrespective of the precise location of the attendant fighting.