[Rogier Bartels is a Legal Officer (Chambers) at the International Criminal Court and a research-fellow at the Netherlands Defence Academy. The views below are the author’s alone.]
The first part of this post discussed that a non-international armed conflict (NIAC) ends when the NIAC-criteria (a certain level of organisation of the parties groups, and a certain intensity of the armed violence) are no longer both present.
At the ICTY, the various trial chambers seized of cases concerning Kosovo and Macedonia had to consider the lower threshold for the start of (or continued existence of) a NIAC. The Boškoski and Tarculovski Trial Chamber, for example, gave a detailed overview of the indicators used so far and reviewed how the relevant elements of Common Article 3 recognised in Tadić (organisation and intensity; see Tadić TJ, para. 562) are to be understood. Its findings were confirmed by the Appeals Chamber (see Boškoski and Tarculovski AJ, paras 19-24). Certain “factors”, and a number of “indicators” thereof, were identified that need to be taken into account when assessing the organisation and intensity criteria. These factors have since been adopted by the Lubanga Trial Chamber in the first ICC judgment (paras 537-538).
If agreed that a NIAC ends when the criteria of “intensity” and “organisation” no longer exist, using these factors and indicators identified in the case law, could be helpful in determining such an ending. Naturally, not all indicators are of assistance. Most notably, the indicator of the existence of (attempts to broker) ceasefire agreements shows that parties considered that there was an armed conflict took place (at the time of the alleged crimes), but obviously does not answer the question whether the conflict continued or ended after such agreements.
Other indicators cannot easily be applied ‘in reverse’. A reversed examination of “the extent of destruction”, for example, would be difficult, as it is hard to assess whether damage diminishes if only few buildings are left standing or if few potential targets remain. The lack of (new or ongoing) damage may well be due to these circumstances, rather than result from the end of the conflict. Nevertheless, an indicator merely serves to ‘indicate’ the existence of an NIAC, and has to be seen in relation to the other indicators: if few military objects remain and a prolonged period occurs during which no targets are attacked, this may well be a sign that the conflict has ended.
In addition, some indicators could be adapted. Instead of looking at the (type of) weapons used, an indicator could be the effectiveness of a disarmament programme: the type and amount of weapons handed in vis-à-vis the initial number of fighters or the approximate type and number of weapons initially deployed. For the indicator of refugee flows from combat zones, one could look, rather than at the number of civilians fleeing an area, at the number of civilians returning home, i.e. considering their pre-conflict place of residence safe enough to return to. (That is not to say that a conflict could never be considered as ended when refugees and or IDPs do not return to their homes as this may be caused by other factors, such as, a changed ethnic composition of the area concerned, lack of cooperation by the government and/or measures implemented by the victorious party).
When peace agreements, as suggested by Tadić, are considered to be the end of NIACs, the focus appears to be laid on the intensity requirement. The discussion regarding the start of the Syrian NIAC (see here for an overview), however, has highlighted the (greater) importance of the organisational requirement. Between the two NIAC-criteria, organisation is the most relevant for the assessment of the end of such conflicts. The decline in organisation of one or more of the parties to the conflict can result in a security vacuum when the controlling regime (i.e. the state or the rebel force) gives way and the resulting (state) apparatus is not (yet) able to provide for effective security. Also, the opposing party will mainly target the organisational structure of an armed group. Whilst targeting the leadership was relatively uncommon in IACs, it has been the main goal in NIACs. It appears also the most effective way to bring about the end of such a conflict. See, for example, the killing of LTTE leader Prabhakaran in 2009, the effects of air strikes killing commanders of the FARC, and the (drone) attacks by United States on the Al-Qaeda leadership. Furthermore, intensity or ‘protractedness’ is hard to pinpoint on a specific moment, because some time element – despite claims to the contrary (see, e.g., the ICTY’s Delalić et al. TJ, para 184 and Kordić and Cerkez AJ, para. 341) – is still inherent in this requirement. Moreover, small break-away fractions of an armed group could continue to carry out attacks, or sectarian violence could go on after – or perhaps result from – the disappearance of the organisational structure of one or more of the fighting parties. Take, for example, the situation in Libya in the period after the defeat of the Gaddafi regime and the forming of the new government by the rebels.
My submission that NIACs end when the level of violence and/or organisation drops below a certain lower threshold, has consequences for the application of IHL and consequently for the protection afforded by IHL. It may be feared that it would lead to “legal uncertainty and confusion” (compare Gotovina et al TJ, para. 1694). In practice, however, having an end-threshold should not create a gap in protection, hence no uncertainty – or at least no more uncertainty than as to the start of the application of IHL at the beginning of a NIAC. Using the lower threshold for the application of IHL ‘in reverse’ in order to determine the end of a NIAC may actually allow for a smoother transition between the law governing the use of force during armed conflict (conduct of hostilities paradigm) and the law governing force outside situations of armed conflict (law enforcement paradigm). It makes sense to gradually move towards a law enforcement approach in the end stages of a NIAC. When the intensity of the fighting has decreased, and/or organisational structure of concerning groups has broken down, to such an extent that it no longer reaches the lower threshold, persons belonging to a (partly or fully broken down) group, would not be “directly participating in hostilities” in the traditional sense, but rather find themselves in a situation where the opposing party controls the territory they are in. As advocated elsewhere (albeit received with much criticism; see here for an overview), the opposing party should then apply the human rights/law enforcement approach when taking action against these persons. If it is unclear whether or not a situation of armed conflict continues to exist, the attacking party should err on the safe side and apply the least amount of force necessary (i.e. in line with law enforcement type of proportionality). This also follows from a moral as well as practical point of view: if the conflict is ending, what would be the benefit of and why would one want to continue to kill the opponents, rather than to start thinking about a process that would bring a lasting peace after the conflict?
The breakdown of the organisational structure of an armed group (which will, amongst other things, be indicated by the inability to carry out military operations) should result in the cessation of the “continuous combat function” of members of that group, thereby limiting the right to target the persons concerned. For those advocating for the so-called “membership approach”, no problem arises either: an even further breakdown of the group’s organisational structure would result in the concerning persons ceasing to be ‘members’; and thus targetable. After all, there needs to be a group or organisation in order for someone to be a member of it.
To sum up, it is my hypothesis that NIACs do not necessarily end only by virtue of a peace settlement being reached, but rather by the more factual circumstance of the level of “organisation” and “intensity” falling below the threshold set for the application of IHL. To assess when NIACs end, one could resort to using the factors and indicators for determining the lower threshold for the start of such conflicts, as identified by the ICTY in its voluminous case law. However, they are to be applied on a case-by-case basis, as not all of them are adaptable to the specific circumstances in which some conflicts take place.