Organizing Rebellion Symposium: A Multi-Faceted Study of Armed Groups and International Law

[Katharine Fortin is a lecturer at the Netherlands Institute of Human Rights, Utrecht University, and teaches human rights law and international humanitarian law. She is the founder and co-editor of the Armed Groups and International Law blog. This is the latest post in the co-hosted symposium with Armed Groups and International Law on Organizing Rebellion.]

It was hard to decide which parts of Tilman’s excellent book Organizing Rebellion to address in this post, as I have written on armed groups from the three perspectives that he covers – IHL, IHRL and ICL – in my book on armed groups and human rights law and elsewhere (see e.g. here and here). I am going to restrict my comments to Parts I and II of the book, which deal with IHL and IHRL. I will first examine his suggested approach to the organisation requirement under IHL, second make some comments on his handling of decentralized groups and third raise a number of questions regarding Part II on IHRL. I will end by identifying a few questions relating to non-State armed groups which are not explicitly explored in the book, but on which I would be interested to hear Tilman’s thoughts.

Organisation requirement in the spotlight

Part I of Tilman’s book puts the organisation requirement from the IHL threshold test in the spotlight, making it irrefutably clear how important it is when determining whether IHL applies to a given situation. It correctly puts to paid to the notion that the belligerency-inspired criteria provided in the commentary to the Geneva Conventions were ever seen as anything more than ‘convenient criteria’ by the drafters. It also rightly emphasizes that the criteria for organisation provided by the ICTY cannot ever be abandoned – but should not be treated in an overly legalistic approach.

In setting out this latter point, Tilman interestingly suggests there might be value in replacing the oft-cited five criteria for the organisation requirement set out in the Boskoski judgment with a looser set of three criteria: (i) that the group is a collective entity (ii) with the ability to ensure respect for fundamental humanitarian norms and (iii) with the capacity to engage in sufficiently intense violence. I can see that there are some advantages of his proposed approach. Most importantly, it encourages a more holistic approach to the organisation requirement that gives better visibility to its key raison d’etre. It highlights that the purpose of the test is to check that the entity is more than a loose aggregate of individuals and has crystallised into a single legal entity capable of being the duty-bearer of international humanitarian law rules. Although this purpose has not been explicitly made clear by the courts or tribunals, it is clearly the main rationale behind the organisation requirement – and the key reason why it can never be abandoned, in favour of a single intensity requirement. Adopting a test for the organisation requirement that gives more prominence to the conceptual underpinnings of the test is helpful, as achieving a better understanding of the point and relevance of the indicators set out in Boskoski test – will be a key tool in understanding how they can be applied in a flexible manner.

A disadvantage of the proposed three point test is that it conflates the different thresholds of ‘organisation’ required for CA3 and APII. It is questionable whether this is helpful or appropriate, especially considering that Tilman himself acknowledges that the amount of organisation required for an APII conflict remains marginally higher than a CA3 conflict. I also can’t help but wonder whether replacing the 5 criteria of the Boskoski judgment with a three criteria test might not simply just add an extra layer of headline indicators under which the Boskoski indicators will then re-grouped?

Decentralised or networked groups?

I fully agree with Tilman’s contention that more attention needs to be given to understanding how the organisation criteria should be applied to decentralized or networked groups. I have long been curious to see a detailed case study of this question, particularly with regard to groups adopting a deliberate cellular structure. Tilman only very briefly addresses these kinds of groups in his analysis. Instead, he gives more attention to the related (and equally important) question of how to assess unintentional fragmentation or decentralisation occurring before or after the threshold of IHL has been met. In other words, he mainly analyses armed groups that are fragmented and disorganized through circumstance, rather than design.

In addressing these types of armed groups he identifies two scenarios. Firstly, he suggests that a small but sufficiently organized group may emerge out of the disorganisation, that allows the NIAC threshold to be met, with the effect that IHL will apply to the whole territory. In this instance, he argues that there will be three consequences: IHL will govern (i) the relationship between the two initial parties (ii) the relationship between the initial parties and other groups acting in support of one of the parties and (iii) all other acts of individuals that qualify as a direct participation in hostilities, or are otherwise related to the conflict. Of these three consequences, the third is the most unexplored in international law and to me, the most intriguing. For a start, it shows how important it is to attain more clarity about how the nexus test should be applied in situations where an armed group controls territory, as it is clearly crucial to determining the parameters of IHL and individual criminal responsibility for war crimes.  It also raises interesting questions about the individual criminal responsibility of members of armed groups versus non-members e.g. are individuals who are DPH-ing bound by the same IHL norms, as members of a non-State armed group?

Secondly, Tilman indicates that independently formed smaller groups/ units may merge to become one party to a NIAC. He explains that this merging can happen in 3 scenarios (a) if different armed groups that are not party to a NIAC join forces and subsequently meet the NIAC threshold jointly (b) if an organized armed group is party to a NIAC, and another armed group joins it and (c) where different existing parties merge into one. In addressing these scenarios, Tilman scrutinizes (and critiques) two suggested approaches to determine what link must exist between discrete groups in order to identify them as one party to a NIAC: (i) the belonging test from Article 4 of GC3 and (ii) the overall control test. While these solutions are interesting (and I agree with him that the overall control test seems more convincing), I am not convinced that either of these tests can provide a satisfactory solution to the first scenario  – because they both rely upon the prior existence of a legal entity.  Surely at least one armed group meeting the organisation requirement already needs to exist, before you can prove that a second armed group belongs to it or is under the control of it?

As a result, I’m not sure that either of these tests help address the first identified scenario i.e. the situation when different armed groups that are not party to a NIAC (i.e. that presumably do not meet the organisation requirement) join forces and subsequently meet the NIAC threshold jointly. Indeed, in the Katanga case, where the Chamber was presented with this problem, it seemsthat rather than applying anything akin to the overall control test or the slightly higher threshold Tilman suggests might be more appropriate, the chamber simply conducted a straight forward but flexible application of the organisation test set out in Boskoski i.e. analysing whether the loose aggregate of discrete militias and units had gained sufficient cohesion and inter-dependency to be seen as a single entity of international law.

Armed groups and human rights law

On the topic of armed groups and international human rights law, I think Tilman and I agree on a lot. I am not entirely convinced by his categorization of armed groups into (i) groups exercising quasi-governmental authority in defined territory (ii) groups exercising de facto control over territory and population and (iii) groups not exercising control over territory and population. I would tend to see more value in conflating the first two categories (which seem rather similar already), creating a second category of groups not exercising sustained control over territory and population and a third category of groups operating outside the context of armed conflict. Perhaps I’m biased – full disclosure: this is (more or less) how I organized my own analysis.

Tilman’s third category is of course the most interesting in terms of legal justification and I think he does an especially good job at highlighting the relevance of State failure in these scenarios. While I am open to the notion that armed groups may in exceptional circumstances have human rights obligations outside an armed conflict situation, I have taken the view that extra scrutiny will need to be given to the organisation requirement in these circumstances together with what I term the ‘international requirement’ (equivalent of the intensity requirement). It is for this reason that I continue to have difficulties with the finding of the COI in Syria in February 2012 that the Free Syrian Army was organized enough to have human rights obligations, at a moment in time when it was deemed not to be organized enough to have obligations under common Article 3 to the Geneva Conventions.

Concluding thoughts and comments

If I could ask Tilman one question about his research, I would ask him about the civilian wing of an armed group and what role he sees it to play in the organisation requirement, if any – under IHL and then IHRL. It is clear that armed groups are not required to have a civilian wing to meet the ‘organisation requirement’ in CA3, but if they do have one (and many do) – would that part of the armed group be included in the organisation that constitutes the ‘party’ referred to in common Article 3? I would then continue to ask the same question with regard APII – and then again, go on to ask the same question with regard IHRL? These questions are interesting for at least three reasons. Firstly, they lead to a greater understanding of the legal personality of armed groups, secondly they may bring clarity to who is bound by the various primary rules and thirdly, they are relevant for the development of secondary rules for armed groups.  To illustrate the last point: if an armed group at the end of a CA3 conflict has demobilised so that it now only constitutes a civilian ‘branch’ (e.g. a political party) – can that legal entity be held responsible for the acts of its previous military incarnation that was bound by CA3? The same sort of question can be asked for an armed group that starts as a guerilla organisation – then becomes very organized and controls huge swathes of territory – like ISIS – and subsequently ends up as a guerilla organisation again? Is the remnant smaller ‘organisation’ (now bound only by CA3) responsible for the acts of the previous larger incarnation of the group (previously bound by APII and also IHRL)? These are difficult questions, but I think if we accept that an armed group’s legal obligations fluctuate along with its fluctuating capacities over time (i.e. CA3,APII, IHRL)  – they need to be asked.

It has been a pleasure to re-read Tilman’s book on armed groups, which is an exceptionally valuable addition to the literature and a much recommended read for academics and practitioners in the field. It’s a rare and fascinating experience to encounter work done by a researcher who has walked so many of the same thought paths as you, studying similar materials and puzzling over similar and connected questions!