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

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

Published on July 15, 2019        Author: 
Follow by Email

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.

Print Friendly, PDF & Email

7 Responses

  1. Excellent post, and I completely agree with your conclusion, but I’m a bit thrown by the statement that “[i]n 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.” In Ntaganda, resolving doubts in favour of finding the conflict a NIAC harmed the defendant, as you yourself note elsewhere in the post.

  2. Dapo Akande Dapo Akande

    Hi Kubo,

    Many thanks for the great post. I do have a question though. Your post does not deal with the central point of the relationship between the law of occupation and NIAC law in situations where the occupying power has to apply both. How should the occupier treat people protected by the law of occupation but who may also be members of an organized armed group involved in a NIAC with that occupying power. To my mind, saying that the occupy power has obligations under both bodies of law, but to different entities (one set of obligations to the occupied state and another set to the non-state armed group), does not quite answer the question of how both bodies of law are to apply in a situation where the fact scenario raises both issues. For one thing, arguably both obligations are owed erga omnes and thus both are owed to every other state in the world. In any event, it will be difficult if not impossible in practice to separate out the two sets of obligations.

    Consider the issue of detention. An occupying power detains members of an organized armed group. What law applies? If it is just a NIAC then the occupying power has no obligations to provide for periodic review. However, since it is a situation of occupation then GCIV would apply and imposes those obligations of review in relation to all protected persons detained by the occupier for security reasons. Or does GCIV no longer apply just because those protected persons have formed themselves into an armed group? Does that fact free the occupier from obligations that would arise if they were not an armed group?

  3. Liron A. Libman

    Thanks, Kubo, for an interesting piece.
    Another thing interesting, in this context, is that the Court choose the “overall control” test to attribute state responsibility from the ICTY Tadic case. It does mention in a footnote (2253) the opposing “effective control” test of the ICJ in the Nicaragua case, but not the more recent ICJ case Application of the Convention on the Prevention and Punishment of the Crime of Genocide (Bosnia and Herzegovina v. Serbia and Montenegro)(2007) rejecting the overall control test.

  4. Kevin: Thank you for your kind words about the post. To respond to your question, I cited Rogier Bartels’s piece on this point mainly to demonstrate the difficulty in identifying one body of law which would supposedly prevail in abstracto due to some overarching (for example, teleological) considerations. As I will also detail below in response to Dapo’s question, in my view, a lot actually depends on the specific rules at play and the precise context in which a particular legal question is raised. I agree with you that – at least with respect to the war crime of enlisting and using children – the finding of a NIAC did not help the defendant in Ntaganda. However, I take it the trial chamber did not consider the qualification question to have been encumbered by doubt (as the conduct of the defendant only related to conflict pairs which would have in their view in any event qualified as NIACs), so the issue of in dubio pro reo likely did not arise in this regard.

  5. Dapo: This is an excellent question and one I probably still need to think about a bit more. My current take on it is that the answer likely depends on the context in which the question is asked. If the context is one of criminal proceedings and the dual qualification presents reasonable doubts (which, as mentioned above, arguably wasn’t the case on the specific facts in Ntaganda), then this uncertainty militates towards an interpretation more favourable for the defendant, i.e. their conduct should be measured against the less onerous legal regime (in your example that would be the law of NIAC). By contrast, if the context is one of ongoing military operations and a legal advisor belonging to the occupying forces is asked a question along these lines, then the overall protective aim of the law of occupation as well as general prudence in providing sound and responsible legal advice would point in favour of the more protective legal regime (which in your example would the GC IV regime belonging to the law of IAC). But reasonable people may disagree—after all, this very issue divided the experts consulted by the ICRC in its project on Occupation and Other Forms of Administration of Foreign Territory, some of whom claimed the law of NIAC prevails as a matter of lex specialis, whereas others considered that the law of occupation prevails as long as the detained personnel meets the Article 4 GC IV criteria (at pages 127–28)—and I will be curious to hear what others think.

  6. Thanks Kubo for a very interesting and insightful analysis. I have some fundamental questions:
    1. Is it necessary for an International Tribunal to rely upon the test evolved by any other international tribunal (referring to the overall and effective control test)or can it evolve its own criteria to determine the extent of control ? And, if the answer is yes then, What is the probative value to be attached (to that determination of the extent of control) by the other International Tribunal?
    2. In case the occupying power is an armed group and if it is fighting with another armed group on that territory, then What will be the applicable law ?

  7. Aditya: Thank you for your comment and for your very good questions.

    1. It is not “necessary“ to do so in the sense that a tribunal would be bound by the jurisprudence of another tribunal on the same topic. However, maintaining consistency in international case-law is certainly desirable as a general aim and a counterweight to the fears of fragmentation of international law. As far as the ICC itself is concerned, one of its trial chambers has expressly noted that it may “be required to refer to the jurisprudence of the ad hoc tribunals and other courts” to resolve matters on which the ICC Statute is not conclusive (Katanga Trial Judgment, para. 47). Insofar as your question refers specifically to the appropriate test of control to determine whether a NIAC has become internationalized, the “overall control” test introduced by the ICTY has now indeed been endorsed by several trial chambers of the ICC (see, eg, Lubanga Trial Judgment, para. 541; Katanga Trial Judgment, para. 1178; Bemba Trial Judgment, para. 130; and now the Ntaganda Trial Judgment, para. 727).

    2. In my view, a fully autonomous non-state armed group cannot be “the occupying power”, because its autonomy implies that it can only be involved in conflict pairs that are qualified as NIACs, and the concept of belligerent occupation is entirely absent from the law of NIAC. By contrast, if a group’s autonomy has been sacrificed entirely in favour of an outside state, it is that state that qualifies as the occupying power, and not the group under its control. There is a possible though unusual middle situation, which I discussed in some detail in my book on conflict internationalization (see especially pages 229–30). The key passage in this respect is the following (footnotes omitted):

    “In situations in which the non-state party maintains its operational autonomy in full, the conflict between itself and the territorial state remains non-international in nature in spite of the involvement of the third state. Therefore, the law of belligerent occupation would not apply to the territory under its control at all. At the other end of the spectrum, if the entirety of the non-state party’s operational autonomy had been sacrificed in favour of the intervening (and thus controlling) state, it is this state that should be considered to be the occupying power. The non-state actor acts as the agent of the intervening state in the sense of Article 29 GC IV and Article 8 ILC ASR.

    In the middle between these two extreme types of situations lie conflicts which should be considered internationalized due to the sufficient control of the outside party, but which do not yet trigger the responsibility of the intervening state for the conduct of the armed group in question. In other words, these are the situations, in which the non-state actor controls a part of the home state’s territory, but is not itself subject to ‘effective control’ of the outside state. In this middle-type of situation, which seemed to have been the case in some of the ex-Yugoslavian conflicts, for some time during the 2011 Libyan conflict, and possibly in the Abkhaz-controlled territory of Georgia, the semi-autonomous non-state actor should be seen as the occupying power.”

    In addition, I am also of the view that national liberation movements in Article 1(4) AP I type of conflicts and non-state armed groups whose belligerency was recognized by the territorial state may in principle become occupying powers, because the conflicts to which such entities are parties qualify as IACs. This position is explained in more detail on pages 226–28 of the book.

Leave a Reply

Your email address will not be published. Required fields are marked *