Image credit: Sasha Maksymenko, “Ukraine Sep[a]ratists Crisis Slovyansk ATO,” 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

This section sketches some of the diverse contemporary situations that might pose challenges to ending, and to discerning the end of, armed conflict under the relevant international-legal framework of armed conflict. Examples are broken down along the following conceptual lines:
  • Conflict classification and party identification;
  • Non-recognition of armed conflict;
  • Status of and adherence to agreements between adverse parties;
  • Long-term enmity marked by intermittent violence; and
  • State responses to terrorism.

Conflict-Classification and Party-Identification Challenges

To varying degrees and in diverse manifestations, recent or ongoing armed conflicts underscore some of the difficulties in precisely delineating how many armed conflicts simultaneously exist, the international-legal classification of those conflicts, the parties to those conflicts, and the international responsibility of relevant states in relation to those conflicts. In turn, those often-interconnected complications may make it more difficult to accurately evaluate when the relevant conflict has terminated and when the accompanying international-legal framework of armed conflict no longer applies in relation to it.

A recurrent threshold challenge is obtaining—amid ongoing hostilities or even after they cease—suitably reliable and sufficiently comprehensive information to make a legal analysis. For instance, unraveling the complexities of the Second Congo War has proved particularly challenging.[329] During many contemporary conflicts, relationships between parties shift during the conflict, as in the DRC, further complicating fact-finding.[330]

The existence of simultaneous overlapping or parallel armed conflicts—as well as challenges in establishing control by one party over another for purposes of conflict classification or of attribution of actions and responsibilities—might make it difficult to detect the end of a particular conflict. In the Second Congo War, Louise Arimatsu notes, “[n]ot only were there an extraordinary number of States and non-state armed groups engaged in the fighting but there were multiple international and non-international conflicts being fought concurrently on the Congo’s vast territory.”[331] Similar complications have arisen in Afghanistan,[332] Iraq,[333] Lebanon,[334] South Ossetia,[335] Syria,[336] and Ukraine,[337] among others.

At least three sets of interconnected legal and factual issues may emerge.[338] The first concerns part of the applicability of IHL ratione personae—namely, who constitutes a party to the relevant armed conflict. The second issue concerns the scope of application of IHL ratione materiae—in short, whether the legal framework of IAC, of NIAC, or of a combination thereof (due to the existence of multiple, simultaneous armed conflicts) applies. And the third issue concerns establishing the responsibility of a state or an international organization for actions carried out by a non-state actor, such as a non-state organized armed group.

The precise contours of all three issues lack an agreed consensus. Debates over classification and attribution thus further complicate discerning an end-point of armed conflict.

Non-Recognition of Armed Conflict

Before it can end, an armed conflict must first exist. But despite the attempt to move, since the adoption of the Geneva Conventions in 1949, toward more factually-oriented criteria to establish the existence of an armed conflict, lack of recognition that a situation amounts to an armed conflict remains a challenge in some contemporary situations.

For instance, one or more of the purported parties to an armed conflict might refuse to acknowledge the existence of the conflict. Some scholars, for instance, argue that a NIAC existed in relation to Northern Ireland as of 1972, even though the U.K. government adopted a policy of non-recognition.[339] At various times, hostilities between Turkey and military components of the Kurdistan Workers’ Party (PKK) arguably met the criteria for the existence of a NIAC,[340] yet Turkey appears to have maintained a posture of non-recognition of the purported armed conflict.

Other than perhaps concerning the doctrine of the state of war in a formal sense, it is not clear what international-legal significance the political stance of the parties vis-à-vis the existence of a conflict may have, as opposed to an objective factual determination. Still, non-recognition by a party to the conflict or a third party may have practical implications.

Consider, for instance, that a human-rights body or court might not take cognizance of the existence of a situation of armed conflict even where that conflict pertains to a matter within that body’s or court’s competence. Thus, despite the ICJ’s approach concerning the applicability of IHL,[341] a Court might limit its judicial review of a situation of armed conflict to matters arising only from a relevant constitutive IHRL instrument, not (also) IHL.

That, for example, was the approach of the European Court of Human Rights in relation to recent eruptions of violence in Chechnya.[342] A formal Russian Counterterror Operation (KTO) was authorized by ukaz (Presidential decree) No. 1255c of September 23, 1999; the KTO was ultimately lifted on April 16, 2009 by order of the President. The ukaz authorized operations by the Unified Forces Group (made up of elements of several different Russian agencies) in the “North-Caucuses region of the Russian Federation” in accordance with the federal law “On the Suppression of Terrorism.” In Isayeva v. Russia, the European Court of Human Rights considered “that using … [certain bombs and other non-guided heavy combat weapons] in a populated area, outside wartime and without prior evacuation of the civilians, is impossible to reconcile with the degree of caution expected from a law-enforcement body in a democratic society.”[343] Still, the Court concluded that “[t]he operation in question ... has to be judged against a normal legal background” because “[n]o martial law and no state of emergency ha[d] been declared in Chechnya, and no derogation ha[d] been made under Article 15 of the [European] Convention [for the Protection of Human Rights and Fundamental Freedoms].”[344]

Outside the traditional battlespace, in some domains—not least in the realm of cyber operations—there is vanishingly little consensus on what may give rise to an armed conflict in the first place,[345] let alone what should and would mark its end. We expect that this concern is likely to gain greater importance in the near future.

Status of and Adherence to Agreements between Adverse Parties

At least two relevant challenges might arise in relation to agreements between adverse parties to an armed conflict. First, as noted above, in relation to IAC, there appears to be a lack of consensus on the legal status of armistices under international law, especially whether an armistice may denote the termination of a war.[346] And second, as also noted above, in light of the approach that a NIAC is said to come into existence based on the fact of a sufficiently organized armed group (or groups) and sufficiently intense hostilities, the adoption of a peace agreement might not, of itself, definitively extinguish the existence of the NIAC and the applicability of the international-legal framework pertaining to it.[347] One example is the August 2015 peace agreement in South Sudan, which, as of February 2017, has still not brought an end to armed conflict.[348] Another is the Arusha Peace and Reconciliation Agreement for Burundi of August 28, 2000, which did not prevent the resumption of fighting; instead, hostilities only ended with a ceasefire agreement three years later.[349]

Long-term Enmity marked by Intermittent Violence

Situations of long-term enmity marked by intermittent violence might pose challenges to discerning the end of armed conflict as well. Consider two situations, both of which raise considerations under the jus ad bellum and IHL.

First, Israel, on one side, and Iraq (along with other Arab states), on the other, engaged in hostilities beginning in 1948. Iraq took part in hostilities against Israel in 1967, and then again in 1973; Israel bombed a nuclear reactor in Iraq in 1981; and Iraq launched missiles against Israel in 1991. A question is whether the period from 1948 to the present (or, at least, until 1991) may be characterized under international law as a continuous IAC between Israel and Iraq.[350]

Second, consider whether the situation concerning Iraq and the U.S. (and, at times, certain U.S. allies) from 1991 to 2004 (or a period therein) may be categorized as a continuous IAC. The international-legal analysis[351] turns in part on whether the relevant U.N. Security Council decisions concerning the ceasefire in Iraq pertaining to the Persian Gulf War terminated the armed conflict[352] or suspended hostilities without terminating the armed conflict.[353]

State Responses to Terrorism

Various states’ responses to terrorist threats also pose challenges to ascertaining the end of armed conflict. That is in no small part because some of those responses blend belligerent rights traditionally associated with war (the contemporary IHL and ad bellum frameworks) with sovereign rights traditionally associated with criminal law-enforcement (part of the contemporary IHRL framework). Borrowing elements from these fields of law, such a hybrid approach might result in a “quasi-permanent condition of juridical twilight, a state of neither peace nor war.”[354] As we detail in the next section with the example of the U.S.’s War on Terror, these admixtures have at times created confusion around what does and does not count as armed conflicts, who is considered a party to conflicts, how far conflicts extend geographically, and how long conflicts last (as well, therefore, of when the corresponding authorities, rights, and obligations terminate).

[329]. See Louise Arimatsu, The Democratic Republic of the Congo 1993–2010, in International Law and Classification of Conflict 176 (Elizabeth Wilmshurst ed., 2012) (footnotes omitted).

[330]. Id. (footnotes omitted).

[331]. Id. (footnotes omitted).

[332]. See, e.g., Hampson, Afghanistan, supra note 32.

[333]. See, e.g., Michael N. Schmitt, Iraq (2003 onwards), in International Law and the Classification of Conflicts (Elizabeth Wilmshurst ed., 2012).

[334]. See, e.g., International Committee of the Red Cross, International Humanitarian Law and the challenges of contemporary armed conflicts 10 (2011), 31IC/11/5.1.2; Iain Scobbie, Lebanon 2006, in International Law and the Classification of Conflicts (Elizabeth Wilmshurst ed., 2012).

[335]. See, e.g., Philip Leach, South Ossetia (2008), in International Law and the Classification of Conflicts (Elizabeth Wilmshurst ed., 2012).

[336]. See, e.g., Terry D. Gill, Classifying the Conflict in Syria, 92 Int’l L. Stud. 353 (2016).

[337]. See Shane R. Reeves and David Wallace, The Combatant Status of the “Little Green Men” and Other Participants in the Ukraine Conflict, 91 Int’l L. Stud. 361, 372–83 (2015).

[338]. See, e.g., Ferraro, Foreign Intervention, supra note 32.

[339]. See, e.g., Steven Haines, Northern Ireland 1968–1998, in International Law and the Classification of Conflicts (Elizabeth Wilmshurst ed., 2012).

[340]. See, e.g., Geneva Academy of International Humanitarian Law and Human Rights, Non-international armed conflict in Turkey, in Rule of Law in Armed Conflicts Project (last updated: Feb. 2, 2017), http://www.rulac.org/browse/conflicts/non-international-armed-conflict-in-turkey <https://perma.cc/2JZM-G2PF>.

[341]. See Armed Activities on the Territory of the Congo (Dem. Rep. Congo v. Uganda), 2005 I.C.J. 116, 242–45 (Dec. 19); Legal Consequences of the Construction of a Wall in the Occupied Palestinian Territory, Advisory Opinion, 2004 I.C.J. 136, 177–81; Legality of the Threat or Use of Nuclear Weapons, Advisory Opinion, 1996 I.C.J. 226, 240.

[342]. On the approach of the European Court of Human Rights to IHL, see, e.g., Samuel Hartridge, The European Court of Human Right’s [sic] Engagement with International Humanitarian Law, in Applying International Humanitarian Law in Judicial and Quasi-Judicial Bodies (Derek Jinks, Jackson N. Maogoto, and Solon Solomon eds., 2014).

[343]. Isayeva v. Russia, App. no. 57950/00, Eur. Ct. H. R., Judgment, Feb. 24, 2005, ¶ 191.

[344]. Id. (emphasis added).

[345]. For a recent overview, see Johann-Christoph Woltag, Cyber Warfare, in Max Planck Encyclopedia of Public International Law (2015).

[346]. See supra note 137.

[347]. See supra notes 321–22 and corresponding text.

[348]. See Lauren Ploch Blanchard, Conflict in South Sudan and the Challenges Ahead, Cong. Res. Serv., Sept. 22, 2016, at 5 (stating that, “[w]hile both sides publicly committed to implementing the [August 2015] peace agreement, progress stalled after it was signed. Major clashes between the two sides decreased, but armed conflict continued”).

[349]. See Venturini, Temporal Scope, supra note 13, at 62 n.78 (citation omitted).

[350]. See supra note 85 and corresponding text. The question implicates not only the application of IHL but also, among other things, whether, under the jus ad bellum, the relevant side needed a new valid basis to resort to the use of force in respect of each of the post-1948 instances outlined above.

[351]. For an overview, see Wolff Heintschel von Heinegg, Invasion of Iraq (2003), in Max Planck Encyclopedia of Public International Law (2015).

[352]. See Christine Gray, After the Ceasefire: Iraq, the Security Council and the Use of Force, 65 Brit. Y. B. Int’l L. 135 (1994).

[353]. See Dinstein, War, supra note 13, at 319–22.

[354]. Neff, War, supra note 80, at 394.


Indefinite War

Unsettled International Law on the End of Armed Conflict