Image credit: Metziker, “Norwegian Special Forces,” Flickr, Attribution-NonCommercial 2.0 Generic (CC BY-NC 2.0) license.
Note: A PDF of this section is available here [link], and more information about this HLS PILAC initiative as well as the full version of the Legal Briefing are available here [link].
Introduction
Even though NIACs are the most common type of contemporary armed conflict,[534] there are, as we illustrated in Section 5, fewer IHL provisions and rules concerning how NIACs end compared to IACs. In this section, drawing from existing international law and scholarly arguments, we postulate and scrutinize four theories on the vanishing point of the application of the international-legal framework of armed conflict in relation to NIAC:- The two-way-ratchet theory;
- The no-more-combat-measures theory;
- The no-reasonable-risk-of-resumption theory; and
- The state-of-war-throwback theory.
A prefatory point is in order. We do not include a standalone theory covering the factual situation where IHL of NIAC ceases to apply to the situation once it transforms into an IAC, at which point IHL of IAC immediately applies.[535] The transformation of a NIAC into an IAC is possible, for instance, where the acts of a non-state organized armed group fighting a state’s armed forces become attributable to another state.[536] IHL applies throughout. Thus, the exceptional state of affairs, with its attendant extraordinary measures that are tolerated by IHL, continues. The key classification question—with implications for the end of NIAC—is whether those powers are governed under the IHL of NIAC or the IHL of IAC.
Four Theories
The Two-Way-Ratchet Theory
According to this theory, a NIAC ends and the international-legal framework of armed conflict ceases to apply in relation to it as soon as at least one of the constituent elements of the NIAC ceases to exist. As noted in Section 5, jurisprudence of the ICTY and the ICC indicate that the constituent elements of a NIAC are (1) a sufficient level of organization of the non-state armed group or groups (the state is assumed to be sufficiently organized) and (2) a sufficient intensity of violence.[537] Thus, under the two-way-ratchet theory, the NIAC would end—and the international-legal framework of armed conflict would cease to apply in relation to it—as soon as the level of organization of the non-state armed group (or, as relevant, groups) or the intensity of violence falls below the threshold necessary for the NIAC to have existed in the first place. Despite being pegged to the ICTY’s formulation of the constituent elements necessary to determine the existence of a NIAC, the two-way-ratchet theory rejects the approach of an ICTY Trial Chamber concerning the end of armed conflict. (Recall that, in Haradinaj, an ICTY Trial Chamber held that until a peaceful settlement is achieved, “there is no need for the Trial Chamber to explore the oscillating intensity of the armed conflict….”[538])Some international-law scholars have endorsed the general idea underlying the two-way-ratchet theory, if by different names.[539] Marko Milanovic, for instance, articulates a general principle: “unless there is a good reason of text, principle or policy that warrants an exception, the application of IHL will cease once the conditions that triggered its application in the first place no longer exist.”[540]
In situations where the intensity of the violence or the organization of the armed group fluctuates (or where both oscillate), the two-way-ratchet theory may lead, in practice, to a revolving door of applicability and non-applicability of IHL. As noted above, an ICTY Trial Chamber warned against the dangers of such a revolving door, which could, in the Chamber’s view, lead to a “considerable degree of legal uncertainty and confusion.”[541]
Seeking in part to mitigate such uncertainty and confusion, some states have adopted rules of engagement (ROEs). The Colombian Ministry of Defence, for instance, adopted an Operational Law Manual that was published in 2009.[542] That Manual concludes that the situation in Colombia involved the parallel application of IHL and human rights law, and introduces concrete ROEs concerning the Colombian situation.[543] The Operational Law Manual’s authors developed the ROEs based on two sets of cards: the so-called “blue card is meant to deal with law enforcement operations,” and the so-called “red card addresses combat operations, i.e. when force is used against military objectives.”[544] The battalion commander decides whether to use the blue card or the red card.[545] In a situation of belligerent occupation, the Occupying Power is obliged to restore and maintain public order and safety, as far as possible.[546] That obligation “has long made the Occupying Power responsible for policing the inhabitants of the occupied territory.”[547] The maintenance of such security “involves an integration, and at times an overlap, of effort between law enforcement and military forces.”[548] Some Occupying Powers adopt ROEs to regulate that effort. Those ROEs are rarely made public, though parts of one were disclosed in a 2013 Israeli Military Court District judgment convicting, after a plea bargain, a soldier of “causing death by negligence ... by carrying out firing against the rules of engagement” in the West Bank.[549]
Another possible downside to the two-way-ratchet theory is that for many NIACs, it will be difficult to determine contemporaneously whether the violence continues to be sufficiently intense[550] or whether the relevant non-state armed group continues to be sufficiently organized (or both). In Afghanistan, for instance, fluctuations in violence have seemed at times to correlate with the changing seasons.[551] The theory may also be frustrated—or, at least, unevenly implemented—if its application turns on the undisclosed intelligence assessments of a state party. For example, a state party may determine, according to secret intelligence analysis, that a non-state armed group has functionally disbanded and, through that analysis, ascertain the end of the conflict. Meanwhile, for those without access to that analysis, the conflict might appear to continue.
Bartels notes that some of the “indicative factors” of the existence of a NIAC articulated by the ICTY (which underlie the two-way-ratchet theory) may be difficult to apply “in reverse.”[552] The attention of the U.N. Security Council to the situation, for example, may vary or recede.[553] Examining ongoing damage might be difficult because “[a] drop in the number of strikes carried out, or in the case of an air campaign, the number of sorties flown, could be the result of a decreasing number of military objects that can be legitimately targeted, rather than the result of a diminishing intensity.”[554] Bartels suggests not only detecting the absence of the “indicative factors” enumerated by the ICTY but also seeking signs of the close of the NIAC. One such indicator that he raises is the number of civilians returning home (while recognizing that other factors might cause refugees or internally dispersed persons not to return).[555] Another such indicator that he raises is the effectiveness of the disarmament process (as measured in the number, type, and amount of weapons handed in compared, for example, to the earlier numbers of fighters or the type and make of weapons used in the hostilities).[556]
The No-More-Combat-Measures Theory
According to this theory, a NIAC ends and the international-legal framework of armed conflict ceases to apply in relation to it upon the general close of military operations as characterized by the cessation of actions of the armed forces with a view to combat. This theory is rooted in the “general close of military operations” formulation, which was established in relation to IAC in Articles 6(2)–(3) of GC IV and 3(b) of AP I and which has been addressed (in relation to IACs) by some scholars.[557]As noted in Section 4, the ICRC defines such “military operations” (in the sense of the antecedent IHL-of-IAC treaty provisions) as “the movements, manoeuvres and actions of any sort, carried out by the armed forces with a view to combat.”[558] If that definition is adopted in the context of a NIAC, then, under the no-more-combat-measures theory, military operations (even in the absence of active hostilities) that have a continuing nexus with the NIAC—such as redeploying forces along the front to build up military capacity or mobilizing or deploying forces for defensive or offensive purposes—will justify the maintenance of the classification of the situation as a NIAC.[559]
Under this theory, once actions carried out by one of the parties to the NIAC with a view to combat cease, the NIAC will terminate, at least with respect to that party. In principle, that NIAC could continue as between other parties (assuming that there were and continue to exist two or more other parties to the NIAC) so long as those other parties have not ceased actions by their respective armed forces with a view to combat.
The No-Reasonable-Risk-of-Resumption Theory
According to this theory, a NIAC ends and the international-legal framework of armed conflict ceases to apply in relation to it where there is no reasonable risk of hostilities resuming. This theory pegs the continued applicability of IHL of NIAC on a “reasonableness” assessment—admittedly vague as that standard might be—of the threat of further hostilities.In a sense, the no-reasonable-risk-of-resumption theory imposes a lower standard than the two-way-ratchet theory. That is because the no-reasonable-risk-of-resumption theory does not turn on either the organization criterion or the intensity criterion falling below the thresholds necessary for the NIAC to exist in the first place. The no-reasonable-risk-of-resumption theory recognizes that the intensity of the conflict might oscillate, but the theory places a premium on the clarity of the applicable legal framework. On the upside, the operation of this theory is designed to make it harder for parties to argue that no law applies (which may be especially important where, in practice, the application of IHRL is contested) or that it was not apparent whether at least IHL governed a party’s conduct. On the downside, this theory risks extending the application of IHL, including the relevant extraordinary measures that IHL tolerates. In practice, as with the two-way-ratchet theory, the no-reasonable-risk-of-resumption theory may also be frustrated—or, at least, unevenly implemented—where its application turns on the undisclosed intelligence assessments of a party.
The no-reasonable-risk-of-resumption theory captures part of the rationale underlying the U.S.’s “tipping point” formulation concerning the possible end of the War on Terror. (Recall that that basic theory is that, “[a]t a certain point, the United States will degrade and dismantle the operational capacity and supporting networks of terrorist organizations like al-Qa’ida to such an extent that they will have been effectively destroyed and will no longer be able to attempt or launch a strategic attack against the United States.” At that “tipping point,”[560] “there will no longer be an ongoing armed conflict between the United States and those forces.”[561]) Yet, the no-reasonable-risk-of-resumption theory requires that, for the IHL of NIAC to continue to be applicable, there must be a reasonable risk of resumption of hostilities—not (only) on the capacity of a terrorist organization to launch a strategic attack against the U.S. mixed with will to do so.
Legal experts have outlined some of the basic contours of the no-reasonable-risk-of-resumption theory. Derek Jinks, for instance, suggests that the end of an armed conflict arrives when “there is no probability of a resumption of hostilities in the near future.”[562] In the view of the ICRC, “a lasting cessation of armed confrontations without real risk of resumption” will “undoubtedly constitute the end of a NIAC.”[563] Further, the ICRC states that such a cessation would also “equate to a peaceful settlement of the conflict, even without the conclusion or unilateral pronouncement of a formal act such as a ceasefire, armistice or peace agreement.”[564] More generally, the ICRC argues that “[i]t is impossible to state in the abstract how much time without armed confrontations needs to pass to be able conclude with an acceptable degree of certainty that the situation has stabilized and equates to a peaceful settlement.”[565] According to the ICRC, “it is not yet possible to conclude that a situation has stabilized, and a longer period of observation will be necessary,” where, for instance, a “Party may … decide to temporarily suspend hostilities, or the historical pattern of the conflict may be an alternation between cessation and resumption of armed confrontations.”[566] At the same time, “[h]owever, the lasting absence of armed confrontations between the original Parties to the conflict may indicate – depending on the prevailing facts – the end of that [NIAC], even though there might still be minor isolated or sporadic acts of violence.”[567] The example elements that the ICRC identifies “that may indicate that a situation has sufficiently stabilized to consider that a [NIAC] has ended” include the following:
- The effective implementation of a peace agreement or ceasefire;
- Declarations by the parties, not contradicted by the facts on the ground, that they definitely renounce all violence;
- The dismantling of government special units created for the conflict;
- The implementation of disarmament, demobilization and/or reintegration programs;
- The increasing duration of the period without hostilities; and
- The lifting of a state of emergency or other restrictive measures.[568]
At base, this ICRC approach stresses “a full appraisal of all the available facts,” while recognizing that “such predictions can never be made with absolute certainty” and that “[i]t is not a perfect science.”[570] Ultimately, “in the view of the ICRC, it is preferable not to be too hasty and thereby risk a ‘revolving door’ classification of a conflict which might lead to legal uncertainty and confusion.”[571]
The State-of-War-Throwback Theory
According to this theory, a NIAC ends and the international-legal framework of armed conflict ceases to apply in relation to it upon the achievement of a peaceful settlement between the formerly-warring parties. This theory is rooted, in part, in an element of the traditional state-of-war doctrine, whereby a state of war in the legal sense would continue to exist until the conclusion of a war-terminating instrument, such as a treaty of peace.[572] The state-of-war-throwback theory presumes that the parties to a NIAC are capable in principle of agreeing to end the conflict and in practice of exercising sufficient control over their relevant components in order to effectively implement that agreement. Since most non-state organized armed groups are, in general, not considered capable of contracting into treaties,[573] the focus in this theory is on achieving a peaceful “settlement,” such as in the form of a durable agreement between the parties.The primary international-legal root of this theory is the end-of-application-of-IHL-to-NIAC formulation—“until … a peaceful settlement is achieved”—that was laid down by the ICTY Appeals Chamber in Tadić and that was endorsed recently by an ICC Trial Chamber in Bemba.[574] The notion of such a “peaceful settlement” seems to concern not only putting the factual situation on a permanent footing but also a more formal, subjective element, akin, perhaps, in certain respects, to a treaty of peace under the traditional state-of-war doctrine. Interpreted with these subjective elements in mind, the “peaceful settlement” formulation is at variance with the fact-centered approach that arose with the notion of “armed conflict”—including NIAC—in the Geneva Conventions of 1949.[575]
An advantage of the more formal, subjective approach of the state-of-war-throwback theory is that it may dissipate some of the concerns that arise with purely fact-based approaches. (For its part, the ICRC seems to interpret the “peaceful settlement” standard in the Tadić formulation as not requiring an explicit, formal, subjective agreement, as such.[576]) A disadvantage is that a government may refuse to recognize a non-state organized armed group (or vice versa), such that a party rejects even the possibility of entering a formal agreement with an adverse side. Nor is it clear if the state-of-war-throwback theory encompasses a complete and definitive surrender by one of the parties.
[535]. Nor do we address the termination of IHL protections in relation to a person deprived or restricted of liberty even after the conflict has ended; see supra notes 287–88 and corresponding text.
[536]. See Ferraro, Foreign Intervention, supra note 32.
[537]. See supra Section 5. As to whether the violence must also be protracted, in the sense of a minimally long duration, see id.
[538]. Haradinaj, Trial Judgment, supra note 312, at ¶ 100.
[539]. In a similar vein, expounding on an idea from Jeh Johnson and in the context of debates around U.S. legal domestic authorizations for use of force, Bill French and John Bradshaw enumerate three “tipping points.” Bill French with John Bradshaw, Ending the Endless War: An Incremental Approach to Repealing the 2001 AUMF, National Security Network, Aug. 2014, https://www.justsecurity.org/wp-content/uploads/2014/08/ENDING-THE-ENDLESS-WAR_FINAL.pdf <https://perma.cc/4PCH-K4DZ>. The first tipping point (coined the strategic-attack tipping point) is where an enemy belligerent has been “sufficiently degraded so that it can no longer threaten strategic attack beyond the capacity of law enforcement to manage.” Id. at 12. The second tipping point (termed the regional-attack tipping point) would arise “when [organizations with regional reach] have been degraded to the point of not being able to participate effectively in hostilities against U.S. assets in a given region or area.” Id. at 12. And the third tipping point (called the no-longer-a-co-belligerent tipping point) would arise when a so-called “co-belligerent”—which is not, however, a legally recognized concept in the IHL of NIAC—of the primary enemy belligerent “has left the fight or, more likely, the organization that it was co-belligerent with has been defeated.” Id. at 13.
[540]. Milanovic, End of IHL Application, supra note 13, at 170.
[541]. See supra note 67 and the corresponding text. That part of that ICTY Trial Chamber’s reasoning concerned an international armed conflict, and, as part of that analysis, that Chamber expressly rejected the defense’s submission that a significant decrease in the level of intensity of violence or in the level of organization of one of the parties (or both) could terminate the applicability of IHL. Gotovina, Trial Judgement, supra note 67, at ¶ 1694.
[542]. See ICRC, Use of Force, supra note 3, at 81.
[543]. Id. On the interplay between IHL and IHRL in the context of Colombia, see Pablo Kalmanovitz, Ius Post Bellum and the Imperative to Supersede IHL, 110 Am. J. Int’l L. Unbound 193 (2016).
[544]. ICRC, Use of Force, supra note 3, at 46.
[545]. Id. at 46 n.116.
[546]. With respect to a situation of belligerent occupation, Article 43 of the Hague Regulations lays down that “[t]he authority of the legitimate power having in fact passed into the hands of the occupant, the latter shall take all the measures in his power to restore, and ensure, as far as possible, public order and safety, while respecting, unless absolutely prevented, the laws in force in the country.” (Note that the official French text of Article 43 of the Hague Regulations refers to “l’ordre et la vie publics.”) See also Article 64(3) GC IV (providing that “[t]he Occupying Power may ... subject the population of the occupied territory to provisions which are essential to enable the Occupying Power to fulfill its obligations ... to maintain the orderly government of the Territory....” And, as Kenneth Watkin explains, GC IV further “requires that ‘protected persons’ are protected against all acts or threats of violence, and establishes rules governing the maintenance of laws, courts, internment, and so forth.” Kenneth Watkin, Use of force during occupation: law enforcement and conduct of hostilities, 94 Int’l Rev. Red Cross 267, 270 n.13 (2012) [hereinafter, “Watkin, Use of force”] (citing to Articles 27 and 47 et seq. GC IV). Watkin explains that “neither the Hague Regulations, nor the Geneva Conventions, nor Additional Protocol I refer directly to policing, although such activity is an inherent part of the detention, internment, and prosecution of criminals or security detainees authorized by humanitarian law.” Id. (citing to Articles 64–78 GC IV).
[547]. Watkin, Use of force, supra note 546, at 314.
[548]. Id.
[549]. See Amnesty International, Trigger-Happy: Israel’s Use of Excessive Force in the West Bank, MDE 15/002/2014, Feb. 2014, p. 13 (stating that “[t]his directive set out rules to be followed by all army soldiers deployed in the West Bank and in the zone around the fence/wall. These specify that soldiers must avoid and refrain from harming ‘non-combatant’ Palestinian civilians, particularly women and children, and instruct soldiers that they must use their weapons only as a last resort; the directive states that the ‘necessity of firing’ is to be examined at every stage, and, as far as possible, directly by the commander who is in charge or according to his order”) (citations omitted). On the status and approach under Israeli law concerning the de jure and de facto applicability (or not) of certain IHL provisions in relation to the West Bank, see, e.g., Dvir Saar and Ben Wahlhaus, Preventive Detention for National Security Purposes – The Three Facets of the Israeli Experience, in Detention of Non-state Actors Engaged in Hostilities 240–41 n.224 (Gregory Rose and Bruce Oswald eds., 2016).
[550]. And, if relevant, protracted; see supra Section 5.
[551]. See Alice Speri, It’s Spring in Afghanistan, Time for Taliban Fighting Season, Vice News, May 12, 2014, https://news.vice.com/article/its-spring-in-afghanistan-time-for-taliban-fighting-season <https://perma.cc/SUT5-54PV> (quoting a security analyst focusing on Pakistan and Afghanistan as stating that “[t]he Taliban launches a spring offensive each year around this time and the reason for that is that the weather is changing”).
[552]. Bartels, When NIACs End, supra note 13, at 309.
[553]. Id.
[554]. Id. at 309 n.68.
[555]. Id. at 309.
[556]. Id.
[557]. See, e.g., Grignon, L’applicabilité Temporelle, supra note 13, at 275–82.
[558]. Commentary on the APs, supra note 68, at 67.
[559]. For relevant analysis concerning IAC, see GC I 2016 Commentary, supra note 68, at ¶ 279.
[560]. See Johnson, How Will It End?, supra note 356.
[561]. Legal and Policy Frameworks Report, supra note 356, at 11–12 (citing to Johnson, How Will It End?, supra note 356); see also Preston, Legal Framework, supra note 360.
[562]. Jinks, The Temporal Scope, supra note 13, at 3 (stating that, “[g]iven the de facto ‘armed conflict’ regime of the Geneva Conventions, the general applicability of international humanitarian law terminates if active hostilities cease and there is no probability of a resumption of hostilities in the near future”. Emphasis added. This theory also draws on the interpretation in the U.K. Ministry of Defence’s The Joint Service Manual of the Law of Armed Conflict (2004) of the main IHL-of-IAC provision concerning the release of prisoners of war. As discussed above, under Article 118(1) of GC III, “[p]risoners of war shall be released and repatriated without delay after the cessation of hostilities.” According to the U.K. Ministry of Defence’s The Joint Service Manual on the Law of Armed Conflict (2004), that “‘[c]essation of active hostilities’ is a question of fact and does not depend on the existence of an armistice agreement. Active hostilities have ceased where there is no immediate expectation of their resumption.” That Manual also stipulates, in this connection, that “[c]essation is not affected by isolated and sporadic acts of violence.” Id.
[563]. GC I 2016 Commentary, supra note 68, at 491.
[564]. Id.
[565]. Id. at ¶ 492.
[566]. Id.
[567]. Id. at ¶ 494.
[568]. Id. at ¶ 495.
[569]. For an overview of the end stages of that conflict and of violations committed therein, see Report of the OHCHR Investigation on Sri Lanka (OISL), UN Doc. A/HRC/30/CRP/2 (Sept. 16, 2015).
[570]. GC I 2016 Commentary, supra note 68, at ¶ 496.
[571]. Id. (internal footnote citation omitted). In a certain sense, the ICRC also might be seen as following the general approach of the two-way-ratchet theory, although with some meaningful variations. For instance, the ICRC recognizes that a NIAC “can cease by the mere fact that one of the Parties ceases to exist. A complete military defeat of one of the Parties, the demobilization of a non-State Party, or any other dissolution of a Party means that the armed conflict has come to an end, even if there are isolated or sporadic acts of violence by remnants of the dissolved Party.” Id. at ¶ 489. In principle, however, the two-way-ratchet theory does not require the complete annihilation of the non-state organized armed group. On the other hand, the ICRC states that “it is not possible to conclude that a [NIAC] has ended solely on the grounds that the armed confrontations between the Parties have fallen below the intensity required for a conflict to exist in the first place.” Id. at ¶ 494 (citations omitted). In the view of the ICRC, “a temporary lull in the armed confrontations must not be taken as automatically ending the [NIAC].” Id. at ¶ 492. Rather, the ICRC prefers an assessment that is based on the factual circumstances and that takes “into account the often fluctuating nature of conflict to avoid prematurely concluding that a [NIAC] has come to an end.” Id. at ¶ 493.
[572]. See supra Section 4.
[573]. But see, e.g., Article 96(3) AP I; see also supra Section 5.
[574]. Bemba, Trial Judgment, supra note 273, at ¶ 141.
[575]. In this connection, as noted earlier, Kleffner avers that this Tadić marker “only holds true to the extent that the ‘peaceful settlement’ is not a matter of mere agreement, but is also an accurate description of the factual situation on the ground.” Kleffner, Peace Treaties, supra note 86, at ¶ 11.
[576]. GC I 2016 Commentary, supra note 68, at ¶¶ 478, 490 (citations omitted) (and noting that, “while the existence of such agreements may be taken into account when assessing all of the facts, they are neither necessary nor sufficient on their own to bring about the termination of the application of humanitarian law”). Id. at ¶ 490.
Indefinite War
Unsettled International Law on the End of Armed Conflict
