A Government-Based Framework for Conflict Classification

Note - Volume 101 - Issue 1



In the early morning of Thursday, February 24, 2022, explosions rocked several cities in eastern Ukraine.[1] Russia rained down missiles on Ukrainian cities, heralding the beginning of Russia’s full-scale invasion.[2] This would be the largest interstate conflict in Europe since World War II.[3] Within the war’s first day, Russian soldiers had advanced into Ukrainian territory from the east and landed by sea to the south.[4] By the end of the first month, however, Russia had largely failed in its aims to capture Ukraine’s largest cities.[5] And six months later, although Russia had seized swaths of Ukrainian territory, Ukraine’s lightning-fast counteroffensive in the Ukrainian northeast retook thousands of square miles.[6] At the time of this writing, the war is still ongoing.

Issues of international law have infused public discussion of the Russia–Ukraine war with a distinctively legal tenor. Allegations of war crimes, often in connection with specific attacks perpetrated by Russia, invoked the targeting rules of international humanitarian law (also known as the law of armed conflict).[7] Commentators augmented reports of these accusations with details of the existing (or, as some pointedly remarked, inadequate) legal mechanisms under which war crimes could be prosecuted.[8] Similarly, responding to intelligence that Russia would conscript Ukrainian civilians under occupation, the United Kingdom Ministry of Defence tweeted that any such conscription would be illegal—providing a specific citation to Article 51 of the Fourth Geneva Convention.[9] The particular legal meaning of “genocide”—which both Russia and Ukraine have accused the other of perpetrating—was highlighted by American officials’ careful statements on the subject: “We have seen war crimes,” National Security Advisor Jake Sullivan said, going on to specify that “[w]e have not yet seen a level of systematic deprivation of life of the Ukrainian people to rise to the level of genocide.”[10]

Notably, the centrality of legal norms to the conflict have not subsisted solely in public statements and news commentary; international legal mechanisms reacted, by the standard of those bodies, quickly. With the support of 141 of its 193 members, and amid a standing ovation, the United Nations General Assembly (UNGA) adopted a resolution condemning Russia’s invasion of Ukraine.[11] Russia was forced to veto a similar condemnation introduced in the United Nations Security Council (UNSC); that resolution attracted affirmative votes by eleven of the fifteen UNSC members.[12] A March 16 order from the International Court of Justice, in a proceeding initiated by Ukraine in response to Russian allegations of genocide, admonished that Russia must “suspend the military operations that it commenced . . . in the territory of Ukraine.”[13] The prosecutor of the International Criminal Court’s (ICC) announcement that he would actively investigate alleged war crimes in Ukraine added to the worldwide cavalcade of outrage against Russia’s aggression.[14]

However, the historic spotlight the Russia–Ukraine war has shone on international humanitarian law has also highlighted some of its weak spots. For one, as already alluded to, any prosecution of Russian military or civilian leaders for war crimes is unlikely.[15] While a body like the ICC—whose consent-based jurisdiction over war crimes perpetrated in the territory of Ukraine is all but certain[16]—could bring a criminal prosecution, the execution of any arrest warrants on Russian territory is unimaginable.[17] These sorts of enforcement problems are perennial in international criminal law.[18]

Another weakness, the topic of this Note, goes to the basic question of what set of laws applies to a given conflict. That weakness is illustrated in the following hypothetical. Imagine that Russia achieves its reported initial goal of capturing Kyiv and overthrowing the Ukrainian government, resulting in the displacement of President Volodymyr Zelenskyy.[19] In his stead, Russia installs a preferred pro-Russian Ukrainian official who claims to be the new President; Russia grants the pro-Russian official nominal control over captured Ukrainian territory.[20] A Ukrainian military leader, still in command of Ukraine’s military forces and proclaiming to act on behalf of Ukraine, refuses to recognize the pro-Russian government and continues the ongoing armed conflict with Russia and the pro-Russian government. In that case, how should the conflict be characterized? As one between two states: Ukraine and Russia? As one between a state (Russia) and a nonstate military actor (the forces of Ukraine’s military leader)?

The treaty-based law of conflict classification, however, provides no simple answer. Under international humanitarian law, there are two kinds of armed conflict: international armed conflicts (IACs) and non-international armed conflicts (NIACs).[21] This classification regime is essentially status based, principally concerned with whether the parties involved are states.[22] However, this status-based regime aligns increasingly poorly with the realities of modern conflicts.[23] This misalignment of doctrine and reality gives interested parties the room for political maneuvering, allowing the parties to shirk their legal obligations under the cover of uncertainty.[24]

Moreover, a conflict’s classification carries enormous legal consequences, since “the rights and duties of those engaged in armed conflicts depend to a considerable extent on the exact qualification of the situation at hand.”[25] These differences have both individual and nationwide effects. For instance, combatants engaged in an IAC are entitled to immunity from prosecution for their warlike acts; nonstate combatants in NIACs, on the other hand, may be prosecuted for exactly the same conduct.[26] Detention of combatants and civilians, an issue with a nationwide impact, is similarly bifurcated. Parties to an IAC must comply with an extensive regulatory regime while parties to a NIAC are not so restricted.[27]

Given the extent to which conflict classification forms the foundations of the law governing armed conflict, it is imperative that its premises subsist. However, as this Note will explain, the law of conflict classification suffers from its exclusive focus on states. Instead, governments should be acknowledged as the true parties to armed conflicts. But following this argument while retaining the bifurcated regime—and thereby refocusing conflict classification on governments rather than states—would create ambiguities that prevent international humanitarian law from achieving its goals. Therefore, the reasons compelling a government-based framework also ultimately support abolishing the distinction between the different categories of armed conflict. This conclusion aligns with existing scholarship advocating abolition, but rests upon distinct reasoning in reaching it.[28]

Part I explains the law of conflict classification by exploring the foundations of the bifurcated conflict classification regime. Part II examines how conflict classification has developed since the accession of the governing treaties; it dives into a case study, walks through reactions to changes in modern conflicts by the legal community, and evaluates reasons why conflict classification remains bifurcated. Part III charts a new course by arguing that, to better reflect reality, conflict classification should center on governments rather than states. Finally, Part IV explores the consequences of a government-based conflict classification regime. Principally, a government-based conflict classification regime makes questions about the relationship between a government and a state central. However, focusing on that relationship also opens the door to political mischief. Therefore, the consequences of a government-based conflict classification regime, which better reflects the realities of modern conflicts, provide another reason why the distinction between IACs and NIACs should be eliminated.

I. The Law of Conflict Classification

This Part examines the current bifurcated conflict classification regime. Subpart I(A) provides background and introduces conflict classification under the Geneva Conventions and their Additional Protocols. Subpart I(B) walks through the conflict classification regime as provided in those treaties, identifying the consistently differential treatment of IACs and NIACs.

A. Foundations of the Bifurcated Regime

The law of armed conflict is founded on the Geneva Conventions.[29] The modern Conventions were negotiated and promulgated in the wake of World War II in response to the inadequacy of preexisting legal frameworks.[30] The original four Conventions provided for protection of armed forces in the field, of armed forces at sea, of prisoners of war, and of civilians.[31] By the end of the initial six-month signing period in February 1950, sixty-one states had signed these Conventions, including China, France, the United States, and the U.S.S.R.[32] Since then, every remaining state has ratified the Conventions, rendering them universally applicable.[33]

The four original Conventions have been augmented by two (relevant) Additional Protocols to the Geneva Conventions.[34] The first two Additional Protocols were adopted in 1977 to address developments in armed conflicts since the Conventions’ accession.[35] As explored throughout this Part, the Conventions and the Additional Protocols are unified in their differential treatment of IACs and NIACs.[36]

To understand the current legal regime, it is necessary to understand the structure of the Conventions. All the Conventions begin with the same two articles: Common Article 2 (CA2) and Common Article 3 (CA3).[37] Those articles provide the two different triggering conditions for applying the law articulated in the Conventions: conflicts triggering CA2 are IACs while conflicts triggering CA3 are NIACs.[38] After CA2 and CA3, each Convention has dozens of articles containing extensive rules regulating the relevant conduct.[39] But first, to trigger the Conventions under either article, there is a threshold question of whether there is an “armed conflict” within the meaning of CA2 and CA3.[40]

While armed conflict is not defined in the Conventions, the term has been interpreted by influential international organizations and legal bodies.[41] The International Criminal Tribunal for the former Yugoslavia (ICTY) ruled that “an armed conflict exists whenever there is a resort to armed force between States or protracted armed violence between governmental authorities and organized armed groups or between such groups within a State.”[42] The International Committee of the Red Cross (ICRC)—in its authoritative commentaries on the Conventions[43]—acknowledged that the ICTY’s definition is the generally accepted starting point for understanding the meaning of armed conflict.[44] As will be explored in more detail, armed conflicts are easier to identify in the context of IACs than they are in the context of NIACs.[45]

Once there is an armed conflict, the content of the governing law depends on whether the conflict triggered CA2 or CA3.[46] The difference between the two regimes is vast.[47] If the conflict triggers CA2, it is an international armed conflict and many hundreds of articles from the Conventions and Additional Protocol I (API) apply.[48] The application of those treaties means that IACs are subject to an elaborate body of rules.[49] On the other hand, if the conflict triggers CA3, it is a non-international armed conflict.[50] Within the Conventions, CA3 itself contains the only rules governing NIACs; Additional Protocol II (APII) adds about twenty substantive provisions, offering some further modest rules.[51] Accordingly, “[i]t is essential to distinguish between international and non-international armed conflicts.”[52]

B. The Geneva Conventions and the Additional Protocols

The regime established by the Conventions and the Additional Protocols distinguishes sharply between IACs and NIACs. Common Article 2 and Common Article 3 originated this distinction. While the Additional Protocols blur the line between IACs and NIACs, they nonetheless retain the distinction. Because of the enormous differences in the substantive protections afforded in IACs and NIACs, a conflict’s classification under these provisions carries wide-reaching consequences.

1. International Armed Conflicts

.—International armed conflicts are those described by CA2 and API. As mentioned, IACs are subject to the “full corpus of the laws of war.”[53] The IAC framework subjects the involved parties to an extensive set of regulations, which affect everything from high-level strategy decisions to the minute details of captive detention. For example, the Fourth Geneva Convention prohibits parties from employing reprisals against civilians as a military strategy.[54] On the other side of the spectrum, the Third Geneva Convention requires that parties detaining prisoners of war allow the prisoners to use tobacco.[55] Additional Protocol I added to the set of laws governing IACs by, among other things, prohibiting indiscriminate attacks.[56]

Still, while CA2 and API both govern IACs, they go about defining those conflicts differently. Under CA2, whether an armed conflict is an IAC is an essentially status-based inquiry asking whether the parties involved are states. Under API, certain conflicts can be IACs even though two states are not involved.

a. Common Article 2.

Common Article 2 sets the baseline for treaty-based conflict classification, creating a status-based inquiry. It provides that the Conventions’ provisions “shall apply to all cases of declared war or of any other armed conflict which may arise between two or more of the High Contracting Parties, even if the state of war is not recognized by one of them.”[57] Since the term High Contracting Parties refers to the states that have signed the Convention (at this point, all existing states[58]), under CA2 “an international armed conflict is essentially an inter-state conflict.”[59]

Common Article 2 addresses two situations: declared war and armed conflict. The application of the Conventions beyond instances of declared war was a functionalist move intended to prevent states from circumventing legal norms.[60] Before CA2, the application of any legal norms to conflict situations required a formal declaration of war.[61] Accordingly, a state needed only to refuse to issue a formal declaration to be able to act with impunity in pursuit of its military objectives.[62] To address that weakness, CA2 provides that the Conventions are also triggered by armed conflicts between states.

Therefore, it is unsurprising that, in the context of IACs, the term armed conflict has been defined in terms of the de facto relationships between states rather than in terms of formal pronouncements. As mentioned, the ICTY’s definition of the term has been widely accepted;[63] under that definition, and as relevant to armed conflicts within the meaning of CA2, “an armed conflict exists whenever there is a resort to armed force between States.”[64] This formulation shifts focus away from the nature of the force used—indeed, an armed conflict exists whenever there is a resort to armed force—and toward the status of the involved parties.[65]

The ICRC is in accord, and its commentaries have stressed that the status of the parties to the conflict is the sine qua non of triggering CA2. It emphasized that “the identity of the actors involved in the hostilities—States—will therefore define the international character of the armed conflict”; in other words, “statehood remains the baseline against which the existence of an armed conflict . . . will be measured.”[66] Ultimately, this means that the application of the Conventions under CA2 collapses into whether the parties effectuating, and suffering uses of, armed force are states.

b. Additional Protocol I

Additional Protocol I veers sharply away from CA2’s state-centered considerations, introducing different and substantially subjective considerations. Reacting to the liberation struggles of the years following World War II, API internationalized conflicts that would previously have been categorized as non-international.[67] Accordingly, Article 1(4) of API provides that international armed conflicts include those “in which peoples are fighting against colonial domination and alien occupation and against racist régimes in the exercise of their right of self-determination.”[68]

This expansion of the definition of IACs was controversial.[69] Some criticized the provision as a politicization of the putatively de facto questions involved in conflict classification.[70] Responses by states ranged from opting out of the Article by reservation to refusing to ratify API—the United States took the latter route.[71] Underlying the United States’ objection, and evident from the difficult-to-define nature of the language itself, is that Article 1(4) shifts the inquiry away from the status of the parties and to questions about their motivations.[72] One reason many states may have accepted this shift is because of the perception that the relevant conflicts were “functionally distinguishable from the traditional civil war type non-international armed conflict”; these conflicts involved challenging authority that was divorced in some way from the sovereignty of the state.[73]

2. Non-International Armed Conflicts

Non-international armed conflicts are a semi-residual category defined by CA3 and APII. Unlike the fully reticulated legal regime governing IACs, the law governing NIACs is quite modest.[74] To understand the implication of classifying a conflict as a NIAC, it is again instructive to clarify the structure of CA3 and APII.

Common Article 3 has a structure unlike any other conflict classification provision. Whereas triggering CA2 causes the application of the hundreds of Conventions articles,[75] triggering CA3 only causes the application of the legal norms articulated within CA3 itself.[76] Accordingly, CA3 is often referred to as a convention in miniature because it contains both triggering conditions and consequent legal rules.[77] Those rules require the parties to respect “elementary considerations of humanity,”[78] such as refraining from torturing civilians.[79]

Additional Protocol II, like API and the Conventions vis-à-vis CA2, contains a triggering provision, Article 1, separate from its substantive provisions.[80] Those substantive provisions, as described by President Reagan in submitting the treaty to the Senate for ratification, are “an expansion of the fundamental humanitarian provisions contained in the 1949 Geneva Conventions with respect to non-international armed conflicts.”[81] For example, APII explicitly forbids the enslavement of civilian populations.[82] Despite this expansion, the protections afforded by APII’s substantive law are modest.[83]

Besides these substantive differences, CA3 and APII are also different in the scope of their application. Common Article 3 provides a two-step inquiry involving both a status-based question and a question of fact. Additional Protocol II takes a different approach, extending its protections to a narrower set of armed conflicts.

a. Common Article 3

Whereas conflict classification under CA2 is principally concerned with the status of the parties involved, conflict classification under CA3 is about evaluating the characteristics of the conflict. It provides that its substantive provisions shall apply “[i]n the case of armed conflict not of an international character occurring in the territory of one of the High Contracting Parties.”[84] The universal ratification of the 1949 Conventions means that CA3 applies to conflicts occurring in any state’s territory that meet the remaining requirements.[85]

Therefore, the major interpretive question is the meaning of armed conflict not of an international character. The threshold observation is that the relevant armed conflict is not of an international character, meaning that it is not an IAC and, therefore, that it is not a conflict between states. As for whether it is an armed conflict within the meaning of CA3, the inquiry is whether the conflict consists of “protracted armed violence between governmental authorities and organized armed groups or between such groups within a State.”[86] This inquiry consists of two requirements: minimum organization and intensity.[87]

The requirements of minimum organization and intensity are inherently fact-bound questions. To meet the organization requirement, the party must “be militarily organized, the indicators of which include the presence of a command structure, the ability to determine a unified military strategy . . . , as well as the capability to comply with [international humanitarian law].”[88] The intensity requirement essentially asks whether the violence is of a military, rather than police, nature.[89] Relevant considerations include the number of individual confrontations, the type of weapons used, the number of forces participating in the fighting, and the number of casualties.[90] These requirements were intended to exclude from the category of armed conflict violence understood to be internal strife or civil disturbance.[91]

Beyond meeting these minimum qualifications, the concept of armed conflict under CA3 sweeps broadly. It is triggered by conflicts between governmental armed forces and armed groups; conflicts between several armed groups; conflicts occurring solely within the territory of one state; and conflicts that cross international borders.[92]

b. Additional Protocol II

While APII also addresses regulating NIACs, its triggering conditions are different than those in CA3. The substantive provisions of APII apply to those armed conflicts not governed by API, where:

[Such conflicts] take place in the territory of a High Contracting Party between its armed forces and dissident armed forces or other organized armed groups which, under responsible command, exercise such control over a part of its territory as to enable them to carry out sustained and concerted military operations and to implement this Protocol.[93]

As of 2022, 169 states have become party to APII, and 3 states have signed the treaty without ratifying it—the United States falls into the latter category.[94]

While APII uses the same interpretive framework as CA3 for identifying armed conflicts, the threshold for triggering APII is higher than that for triggering CA3.[95] Additional Protocol II imposes a more rigorous threshold in two ways. First, it applies only where government forces are involved in the conflict, excluding conflicts solely between nongovernment armed groups.[96] Second, it is limited to situations where the nongovernment armed group has control over a sufficient part of the opposing state’s territory such that it can organize itself and project force in a particular way.[97]

II. Bifurcated Conflict Classification Post-Treaties

Since the accession of the Conventions, the reaction of the international legal landscape to developments in armed conflicts has left the state of conflict classification more unsettled. Simply put, contemporary armed conflicts are complex and difficult to classify.[98] This increasing complexity has contributed to the “gradual convergence” between IACs and NIACs in the law of conflict classification.[99] Subpart II(A) provides a case study of the hostilities in Afghanistan since 1996 to demonstrate how modern conflicts strain the existing law of conflict classification, allowing involved parties to circumvent international humanitarian law. Subpart II(B) explores how the law of conflict classification reacted to these issues. Subpart II(C) takes a step back and asks why the distinction between IACs and NIACs persists, addressing arguments in favor of maintaining the distinction.

A. Developments in Armed Conflict: A Case Study

Modern conflicts have become increasingly complex, and the existing law of conflict classification has strained to account for them. This lack of clarity in the legal regime allows involved parties to deny the application of international humanitarian law.

A case study is instructive. Consider the conflict between various actors—ranging from states to nonstate actors to actors seemingly falling somewhere in between—occurring in the territory of Afghanistan between 1996 and the present.[100] Before 2001, two governments fought a civil war over the governance of Afghanistan: the Taliban, recognized by Pakistan, Saudi Arabia, and the UAE; and the Northern Alliance, recognized by several Western states.[101] After 9/11, in order to pursue Al-Qaeda, the United States and its Coalition partners began conducting military operations in Afghanistan, including against the Taliban.[102]

Then, a series of developments complicated the picture further. In November 2001, a United Nations-endorsed meeting of non-Taliban Afghan factions agreed to install Hamid Karzai as the head of an interim Afghan government.[103] A UNSC resolution, responding to consent by the interim government, then authorized a multinational security mission (the International Security Assistance Force, or ISAF) to use force in Afghanistan.[104] The ISAF fought against the Taliban through most of the 2000s.[105] Skipping forward over a decade, the status of the involved parties changed again when the Taliban retook control of Afghanistan in August 2021.[106] With the United States’ removal of all military forces from Afghanistan,[107] there is no longer armed force consistently used against the Taliban. Still, the contemporary status of the conflict[108] remains relevant to understanding the conflict’s residual issues, for example, the law governing remaining prisoners detained by both parties.[109]

Accordingly, there are four stages of conflict in the simplified and abbreviated history of modern Afghanistan that all defy easy categorization under the treaty-based regime. First, there was the conflict between the Taliban and the Northern Alliance—both of which claimed to be the proper government of Afghanistan.[110] This conflict was evidently a NIAC[111] since it only involved, at most, one state. Still, it is not clear whether it would fall within the APII regime because it is possible that, due to their conflicting claims of authority, neither party constituted the armed forces of Afghanistan within the meaning of APII.[112]

Second, there was the conflict between the Taliban on one side and the United States and its Coalition partners on the other. The classification of this conflict turns on whether the United States’ conflict with the Taliban constituted a conflict with Afghanistan.[113] If so, then the conflict was an IAC; if not, then it was a NIAC.[114] Of course, this completely skates over the difficult question: how to tell if a conflict with the Taliban constituted a conflict with Afghanistan. Precisely how difficult that question is to answer, as well as the implications of that difficulty, is addressed more fulsomely in subpart IV(A).[115] Here, it is sufficient to acknowledge that the proper classification of the conflict between the United States and the Taliban is hardly straightforward.

Third, there was the conflict between the ISAF and the Taliban. The ISAF was comprised of contingents of military forces contributed by states, meaning that states were on one side of the conflict.[116] However, the status of the Taliban again belies easy identification. At some point, the Taliban ceased to represent Afghanistan.[117] However, international recognition of the Karzai government as the government of Afghanistan occurred prior to the deployment of the ISAF forces that displaced the Taliban government.[118] Therefore, there was a period of conflict between the ISAF and the Taliban occurring at the height of the latter’s ability to plausibly assert that it represented Afghanistan.[119] Compounding the complexity, it was the Karzai government, purporting to act on behalf of Afghanistan, that consented to the intervention of the ISAF.[120] To restate the situation: the conflict was between (1) states acting with the consent of a new but internationally recognized government purporting to represent Afghanistan, and (2) a preexisting government with de facto control over most of the territory of Afghanistan[121] but that, at some nebulous point, ceased to represent Afghanistan. What a mess.

Fourth and finally, there is the conflict between the United States and the Taliban since it retook the country in August 2021. After the point at which the Taliban no longer represented Afghanistan, the conflict was a NIAC.[122] However, with the Taliban takeover of Afghanistan in August 2021,[123] the government-displacement issues that complicated classifying the conflict between the ISAF and the Taliban now arise in reverse: at some point, the Taliban became the government of Afghanistan again.[124] So, how to classify a conflict where a previously nonstate actor comes to represent the governing authority of a state? And, as before, how to tell when an actor constitutes the governing authority of a state?

The parties involved in these conflicts seized on these ambiguities to deny the application of the law of armed conflict. In the second conflict, between the United States and the Taliban, although the United States seemed to impliedly accept that the conflict was an IAC, it initially denied the applicability of the Geneva Conventions.[125] In the third conflict, between the ISAF and the Taliban, views on the conflict differed: while the United States appeared to acknowledge it as a NIAC[126] (although, without articulating the point at which the conflict changed from an IAC to a NIAC[127]), Germany denied that there was an armed conflict at all.[128]

The complexities involved in the case study of Afghanistan may be particularly acute, but they are not exceptional in terms of modern conflicts. Hostilities in the Democratic Republic of the Congo,[129] Lebanon,[130] and Ecuador[131] have all raised questions about the proper classification of conflicts involving many actors with opaque statuses and mutable relationships. And this trend does not seem to be slowing down; armed conflict is “likely to continue to be messy.”[132] Given the enormous consequences of proper conflict classification, the onus is on the law of armed conflict to keep up.

B. Developments in the Law and Commentaries

As a reaction, in part, to the complexity of modern armed conflicts, the law of conflict classification has undergone a convergence.[133] This erosion of the distinction between IACs and NIACs emerged from distinct quarters. The law itself—formal treaties, judicial decisions, and military policies—is one source of the erosion. Simultaneously, a supporting body of scholarship has questioned the distinction, suggesting that the law of armed conflict needs to be revamped with some tertium quid or by the elimination of the distinction altogether. And yet other scholars have brought new analytic tools to bear to classify complex conflicts within the existing bifurcated regime.

The blurred line between IACs and NIACs is a hard-nosed fact of legal practice. The Additional Protocols themselves are arguably the progenitors of the blurred distinction, as they changed some previously non-international conflicts into IACs and created a new, higher standard for NIACs.[134] However, the “process of confluence” between the two classifications is likely best understood as kick-started by the highly influential decision on jurisdiction by the ICTY in Prosecutor v. Tadić.[135] In that case, the ICTY held that “customary rules governing [NIACs] comprise many principles traditionally thought to regulate international conflicts only.”[136] A subsequent ICRC study found that approximately 90% of the international humanitarian law rules it identified apply in both IACs and NIACs.[137] Following this trend, some treaties relating to armed conflict bit the bullet and forewent the distinction entirely.[138] And, perhaps in response to the same currents, some national militaries—including that of the United States—have expressly stated that they apply the same international humanitarian law rules regardless of the proper classification of the conflict.[139]

International humanitarian law scholars have questioned the bifurcation of IACs and NIACs. A few scholars suggest adding a new category of conflict.[140] Professor Geoffrey Corn, for example, responding to the limitations of the bifurcated regime’s application to conflicts in Afghanistan and Lebanon, advocates for a third category of conflict styled as “transnational armed conflict.”[141] Corn defines a transnational armed conflict as characterized by “the de facto existence of armed conflict”; he argues that the category is necessary because it “is not limited [in its application] by either the non-state status of a party to the conflict or the geographic scope of the conflict.”[142] A larger set of scholars prefer eliminating the distinction altogether.[143] Professor Steven Ratner identifies the differential treatment of atrocities occurring in IACs and NIACs as one of the “arbitrary schisms” of international criminal law.[144] In accord, the argument goes that, given the humanitarian concerns that arise equally in both classes of conflict, “the humanitarian aims of international humanitarian law are best fulfilled by the abolition of the distinction.”[145]

Yet other scholars work to account for modern complex conflicts by deploying new concepts within the bounds of the existing bifurcated regime. Responding to the fuzzy borderline between IACs and NIACs, one prominent contribution has been work on the so-called “internationalization” and “de-internationalization” of armed conflicts.[146] These concepts describe situations where a conflict that began as an IAC transforms into a NIAC or vice versa.[147] The transformation of a conflict from one category to another carries with it legal implications that do not match cleanly with traditional law-of-armed-conflict doctrines.[148] For example, the law of occupation is designed to govern the presumed relationship between the occupying state and the displaced state;[149] but how does the law of IACs react when the occupying power was initially a nonstate actor? Kubo Mačák’s important book on internationalized NIACs carefully thinks through the way in which these complex legal situations can be understood within the existing bifurcated law of conflict classification.

This latter strand of analytic scholarship comports with the reality that the distinction between IACs and NIACs persists. Even the Additional Protocols, the arguable progenitor of the convergence of IACs and NIACs, “underscored that the two types of conflict remained governed by separate legal frameworks.”[150] The Rome Statute, the multilateral treaty that established the International Criminal Court,[151] preserved the distinction between IACs and NIACs in its provisions on war crimes.[152] Even the scholarship rebelling against the bifurcation of conflicts takes as its premise that the current law of conflict classification is bifurcated. Accordingly, although convergence may accurately describe the current trajectory of the law, it would be naïve to expect that trajectory to terminate at unification. As the next subpart explores, disagreements ranging from the conceptual to the concrete explain the reason why “the distinction is here to stay.”[153]

C. Why Bifurcate Conflict Classification?

Given the complications inherent to the treaty-based conflict classification scheme, why bifurcate conflict classification at all? Why not instead have any armed conflict trigger the law that currently applies to IACs? While some arguments supporting the distinction are decidedly weak, others must be taken seriously.

On one level, the distinction between IACs and NIACs is merely a historical accident—the result of the negotiation process that created the Conventions.[154] The negotiating governments wanted to remain generally unconstrained in their ability to address domestic conflicts as they saw fit.[155] Indeed, the parties that negotiated the Conventions might well object that the bifurcated regime was a concession and that the creation of the category of NIACs and its applicable law constituted a major development in the law of armed conflict.[156] As a response to the Spanish and Greek civil wars, CA3 was designed to capture the kinds of internal conflict that the international community had seen in the first half of the twentieth century.[157]

Still, wholly historical explanations fail to explain why the distinction between IACs and NIACs persists.[158] From the standpoint of the Conventions, there are two characteristically twentieth-century rationales used to explain the distinction: a sovereignty rationale and a security rationale. The sovereignty rationale posits that equating international and non-international conflicts would undermine state sovereignty.[159] This rationale was articulated by the United States in the debates surrounding the drafting of the Conventions; the United States argued that “every government has a right to put down rebellion within its borders and to punish insurgents in accordance with its penal laws.”[160] However, this rationale is weak because high-flown conceptions of sovereignty have little to do with the objective of the law of armed conflict: atrocity prevention.[161] Moreover, it is unclear why sovereignty interests justify treating IACs and NIACs differently. For example, when one state invades another, the invaded state’s sovereignty interests seem impinged to the same—or even a greater—extent than if the armed conflict were with nonstate insurgents. And if the justification subsists in an interest in denying legitimacy to insurgents,[162] there still needs to be a reason why that interest outweighs the humanitarian interests of conflict regulation.

The second Convention-era rationale is a security rationale. According to this argument, treating NIACs the same as IACs “would not only encourage secessionist movements, by giving them a status under international law, but . . . would restrain the hand of the State in seeking to put down rebellions.”[163] Although this rationale is more plausible, it lacks force in its basic form. First, it is doubtful that the attachment of international humanitarian law rules is a meaningful contributor to the decision of an armed group to project force. Non-international armed conflicts are initiated by nonstate groups for many economic, political, and sociological reasons that have little to do with the group’s status in international law;[164] in these circumstances, providing a legal status is unlikely to act as a catalyst, just as refusing a legal status is unlikely to act as anything but a de minimis deterrent.[165] Second, this rationale also fails to limit itself to NIACs. In other words, why do security interests make it acceptable to “restrain” a state’s hand in IACs but not in NIACs?[166] And, weighing against any differential treatment based on the security rationale, the humanitarian concerns underlying the IAC rules are fully present in NIACs. The security rationale therefore provides only middling support for the distinction between IACs and NIACs.

The strongest argument in favor of the distinction involves the doctrine of combatant immunity,[167] which prohibits combatants in IACs from being prosecuted for their warlike acts.[168] Doctrinally, this immunity comes from a combatant’s status as a prisoner of war.[169] However, because prisoner-of-war status is only afforded to combatants who “fight on behalf of a state” in an IAC, individuals who fight on behalf of a nonstate entity in a NIAC are denied combatant immunity.[170] Accordingly, applying the same set of rules to both IACs and NIACs would ostensibly afford prisoner of war status and the concomitant combatant immunity to fighters in every conflict, regardless of its classification. This is a red line for states.[171] States want to “preserve [their] prerogative” to punish nonstate actors for merely participating in hostilities.[172]

This concern is serious because it identifies a tension between the concept of armed conflict and the interest of states in prosecuting what would otherwise be criminal acts. Granting combatant immunity to all participants in armed conflicts would create an incentive for states to deny the existence of a putative armed conflict. For example, if a state believes a fighter’s cause is illegitimate and wants to prosecute that fighter, it would have to deny that the fighter was engaged in an armed conflict. Given that states have a strong interest in prosecuting acts they consider criminal—especially acts that would otherwise constitute treason—the incentive would be forceful and sympathetic.[173] This sort of tension between political objectives and doctrine is exactly what led the United States to deny that any international humanitarian law protections applied to its adversaries in the War on Terror.[174] An incentive to dilute the concept of armed conflict is counterproductive to achieving humanitarian goals.

While the problem of combatant immunity may do a better job of explaining why the distinction persists, it says less about whether the distinction is useful. If the distinction between IACs and NIACs should be rejected, as the remainder of this Note will argue, the extent of combatant immunity represents a problem to be solved, not a reason to maintain an otherwise defunct distinction.

III. The Relationship Between States and Governments

The preceding analysis used two terms imprecisely that, at this point, are important to distinguish between: states and governments. As previously emphasized, the treaty-based law of conflict classification principally involves an inquiry about the status of the involved parties—more specifically, whether the involved parties are states.[175] However, this state-based focus maps poorly onto reality, causing complications for conflict classification. For example, the doctrinal focus on states to the exclusion of governments lurked behind the complications arising in the case study of Afghanistan.[176] Accordingly, disaggregating states and governments opens a path forward for the law of conflict classification.

Governments are not the same things as states; it is well-established in international law that governments and states are related but distinct legal entities.[177] Distinguishing between them plays an important part in developing an approach to conflict classification that comports more closely with reality.

Still, in international law, states and governments are closely intertwined, making them difficult to disaggregate. Take this pair of statements, for example, by a scholar taking a distinctly international-law-as-science approach to the law of recognition: (1) ”The first condition of statehood is that there must exist a government actually independent of that of any other State”;[178] (2) ”An authority cannot be recognized, de jure, as a government without being recognized as the government of a State.”[179] In other words: for there to be a state, there must first be a government, and for there to be a recognized government, it must first be said to represent a state. Yet rather than indicating a contradiction, these statements highlight that the concepts are tightly bundled in some areas of international law. Even the most sophisticated commentators will occasionally conflate the interests of states and their governments.[180]

Accordingly, it is helpful to define governments and states in sharp relief from one another. As compared to governments, states are characterized by a consistency of identity.[181] That identity amounts to an “international legal personality”[182] that enables the state to have “normal diplomatic and economic relations, to join international organizations, and to sign international treaties and agreements.”[183] These endeavors truly impact the state, since they change the nature of the rights and obligations of the state vis-à-vis the international community regardless of what government rules.[184] On the other hand, the state is not the one undertaking these endeavors—governments are. Rather than being international legal persons, governments are the agents of states.[185] As agents, governments represent and act on behalf of states in conducting their international affairs.[186] The persons conducting trade, sitting in on UNGA meetings, and putting their names on treaty documents are all governmental representatives. To reiterate the distinction, governments are fundamentally de facto entities, whereas states are somewhere closer to the border between de jure and de facto entities.[187]

Given these distinctions, the law of armed conflict is a particularly poor fit for a focus on states over governments. In international treaty law, for example, considering states to be the relevant players makes sense—treaty obligations may best be understood as attaching to the stable identity of the state rather than the contingent identity of the government.[188] Not so in the case of armed conflicts. In armed conflicts, the important questions are raised by on-the-ground actions—firing a missile, capturing a combatant, and launching a cyber offensive, for example. These are not the high-flown policies of states but the imperfect realities of governments. It is for that reason that nearly all inquiries in international humanitarian law turn on the facts on the ground.[189]

So why does the law of conflict classification focus on the statehood of the parties? As described in Part I, under CA2 whether an armed conflict is an IAC depends completely on whether the parties involved are states.[190] Similarly, the first question in identifying a NIAC under CA3 is asking whether it is a conflict not of an international character—in essence, whether the parties are states.[191] Yet, as described, “[s]tates are abstract entities unable to act on their own”; conversely, governments, with a monopoly on “speaking and acting for their state,” are the effectual actors on the international plane.[192] Thus, the law of conflict classification’s outcome-determinative focus on the statehood of the parties neglects the reality that, at bottom, the relevant parties are governments, not states.

A closer alignment between reality and legal doctrine would clarify the legal obligations owed by parties to an armed conflict. Clear legal obligations ensure that sophisticated parties cannot skirt their responsibilities and provide insufficient protection to those affected by armed conflicts.[193] Therefore, to better align with the reality that governments are the de facto parties to armed conflict, the law of conflict classification should be recentered on governments rather than states.

IV. Government-Centered Conflict Classification

A government-centered framework for conflict classification would align better with the reality of armed conflict, obviating the mismatch between legal doctrine and the reality of modern armed conflicts. However, although recentering on governments would be consistent with the existing bifurcated conflict classification regime, maintaining the bifurcation would put center stage the intractable problem of recognizing governments. Subpart IV(A) lays the groundwork for conceptualizing conflict classifications in terms of governments, explaining why the recognition of governments becomes a central consideration in a government-centered framework; it argues that the impossibilities associated with recognition counsel abandoning conflict bifurcation altogether. Finally, subpart IV(B) considers some further implications of a government-centered framework.

A. Recentering on Governments: Recognition and the Elimination of the Bifurcation

A government-centered conflict classification regime would change the basic inquiry of conflict classification. That change, however, would also make questions of recognition central, an untenable prospect for the law of armed conflict. Thus, a focus on governments provides another independent reason to eliminate the distinction between IACs and NIACs.

A government-centered approach to conflict classification would focus classification on the question of whether a government represents a state. As previously explained, whether an armed conflict is an IAC under CA2 depends on whether it is between two different states.[194] In a government-centered regime, the decisive question is instead whether the two governments involved represent two different states.[195] If the governments involved represent different states, then the conflict is between different states and is thus an IAC; if not, and the conflict rises to the level of an armed conflict within the meaning of the Geneva Conventions, then the conflict is a NIAC.[196]

While this move may seem minor, changing the basic inquiry of conflict classification to focus on the relationship between a government and a state raises considerable difficulties. Most immediately, how to tell whether a government represents a state. That is, in the ICRC’s words, “a thorny issue.”[197] And while the ICRC’s authoritative commentaries do provide guidance, its approach redounds to the generally one-step-forward-two-steps-back character of bifurcated conflict classification analysis. Here, as in most places, the ICRC emphasizes the need for fact-based determinants.[198] Its conclusion is that

[t]he very fact that the said government is effective and in control of most of the territory of the State concerned means that it is the de facto government and its actions have to be treated as the actions of the State it represents with all the consequences this entails for determining the existence of an international armed conflict.[199]

This comment raises as many questions as it answers. First, how literally should practitioners understand “most of the territory of the State”? If the answer is very literally, such that state representation depends on which government controls at least 51% of a state’s territory, the approach is arbitrary—why does majority control accord the government any better claim to legitimately represent the state?[200] Yet anything less than that hyper-literal reading is also unsatisfactory, as it means that multiple governments could de facto represent a state if they each had sufficiently extensive control. Second, the claim that a qualifying government’s actions “have to be treated as the actions of the State it represents” is hardly self-evident. Indeed, that claim implicates the significance of a government’s recognition of another government as representing a state. Questions of recognition clash with the ethos of the law of armed conflict.

Regardless of one’s understanding of the precise nature of recognition, its political character is itself incompatible with the law of armed conflict. In international law, recognition means “acknowledgement of the existence of an entity or situation indicating that the full legal consequences of that existence will be respected.”[201] The ICRC’s statement above suggests that this acknowledgment is an obligation arising out of a government’s effective control over territory. That de facto understanding belies the decidedly political reality of recognition in practice. In practice, many established governments have refused to recognize that a government represents a state.[202] The conflict in Afghanistan provides an example here. While the Taliban continued to exercise effective control over the majority of Afghanistan for some time after the universal recognition of the Karzai government, “there is a clear international consensus that after [the establishment of the Karzai government], the situation in Afghanistan transformed into a NIAC.”[203] In other words, the international community’s understanding of the nature of the armed conflict turned more on nations’ acts of recognizing the Karzai government than on the reality of the extent of the Taliban’s control.

Admittedly, many scholars resist, with varying degrees of intensity, the characterization of recognition—at least to the extent that it plays a role in conflict classification—as an ultimately political act. On one view, recognition involves both a subjective (i.e., political) and an objective (i.e., fact-based) component.[204] On another, “the better approach is to inquire instead whether the new government has acquired such level of effectiveness that its status as a government cannot be denied any longer.”[205] Several others speak in absolute terms about the outcome-determinativeness of facts regardless of the formal act of recognition.[206] In an important respect, these scholars are all correct—it is important to deny that the application of the full corpus of legal protections depends on the political act of recognition.

Still, sufficient mischief is wrought by the introduction of ambiguity inherent to any role for recognition in conflict classification. Clear legal rules ensure that governments cannot deny the application of humanitarian laws simply because they believe the laws to be politically inexpedient.[207] Yet, insofar as recognition plays a role in conflict classification, governments can do just that. The argument runs: We deny that government X represents state Y, and so we deny that our conflict with government X is with another state and therefore deny that the conflict is an IAC. Indeed, that argument resonates, rightly or wrongly, with discussion of a variety of disfavored governments.[208] The ability to deny the application of fulsome legal protections based on a plausible refusal to undertake a political act erodes the consensus building necessary for effective atrocity prevention.[209] This problem is even more acute because it occurs at the very outset of an analysis of possible law-of-armed-conflict violations.

Accordingly, the arguments compelling a government-centered approach to conflict classification also ultimately provide support for abolishing the distinction between IACs and NIACs. To summarize, a government-focused approach better reflects the reality of modern armed conflicts. It also clarifies complex classification problems, focusing analysis on whether the governments in a conflict represent two different states. A government-centered approach, however, also makes central the question of when a government does, in fact, represent a state. While that question may be considerably fact-based, it also involves a non-negligible political aspect. Yet all this complication can be avoided if the full corpus of laws applied to all armed conflict. Without the need to distinguish between IACs and NIACs, the only inquiry would be whether the relevant conflict is an armed conflict, an inquiry that is both grounded in reality and determinable by facts.

Insofar as the distinction between IACs and NIACs is here to stay, as the case seems to be,[210] it is important to theorize the relationship between governments and states more clearly. It is concededly the case that the governments of most states are stable, and so determining what government represents that state is often a trivial analysis. But that does not change the fact that the first step of conflict classification analysis should be answering the question of representation, with all its concomitant consequences. This approach clarifies complex conflicts; it also raises problems of its own.

B. Further Implications of a Government-Centered Approach

A government-centered approach raises additional implications touching on other aspects of the law of armed conflict. First, it challenges how combatant immunity is currently afforded. Second, it raises questions about the entities that should properly be the parties to international humanitarian law treaties. Although these issues are of tremendous practical importance, they will only be gestured towards here.

1. Combatant Immunity

In a government-centered approach, it may be fair to extend combatant immunity to those who fight on behalf of a government rather than just to those who fight on behalf of a state. The motivating principle of combatant immunity is that it incentivizes compliance with the law of armed conflict by simultaneously legitimizing and circumscribing the acts that may be permissibly undertaken during armed conflict.[211] The idea is that the immunity provides latitude for fighters to achieve legitimate military objectives while threatening sanctions for violations of legal norms.[212] This incentives-based conception applies with equal force to those who fight on behalf of governments, regardless of whether that government represents a state. Moreover, the idea of a government itself presupposes effective control of territory,[213] meaning that a government necessarily has the organization to comply with humanitarian principles.[214] Thus, in a government-focused framework, which deflates the role of sovereignty, there appears to be little reason to refuse to extend immunity to those who satisfy the criterion of fighting on behalf of a government.

These brief comments fit into a larger debate about the proper recipients of combatant immunity. As was previously explored, the exclusive allowance of combatant immunity to those who fight on behalf of states remains a substantial obstacle to the complete convergence of IACs and NIACs.[215] Meanwhile, some scholars have questioned the rationales supporting the idea that combatant immunity should only extend to those fighting on behalf of a state.[216] A government-centered approach fits more closely with those advocating for wider extensions of immunity.

2. Nonstate Accession to Treaties

Even more fundamentally, a government-centered approach raises questions about who should be the parties to agreements creating law-of-armed-conflict obligations. While it is true that states’ steadier identities make them a more natural repository of legal obligations,[217] governments bear the ultimate responsibility for legal compliance. Accordingly, with a government-centered approach, it may be more consistent to focus on the legal obligations of that government. A government could acquire those legal obligations through its relationship to a state or through efforts focused on binding the government itself.[218] Both means have further implications. If the relationship between the government and the state becomes the focus, should all governments purporting to represent a state be considered bound by that state’s legal obligations? If governments are best bound individually, what method is best to accomplish that goal and achieve accountability?[219] Displacing states as the sole party to legal instruments would raise difficult questions but also opportunities for doctrinal innovation.


While conflict classification is fundamental to the application of the law of armed conflict, the law of conflict classification is currently in a state of flux. Although the treaty regime upon which the law of conflict classification is based divides the universe of conflicts in half—into IACs and NIACs—the reality of modern conflicts is much more complicated. Courts, treaties, and scholars have responded in a variety of ways. Some have challenged the IAC/NIAC bifurcation, some have embraced it, and others have charted middle paths. Altogether, the trend has been towards a convergence in conflict classification. Yet, the distinction between the two classes remains.

One source of confusion in conflict classification is the exclusive focus on the statehood of parties to a conflict. To better match the reality that governments are the active entities able to participate in conflicts, conflict classification should focus on identifying governments involved in a conflict rather than focusing on states. This government-centered approach would cohere with the current bifurcated conflict classification regime—the decisive inquiry would be whether the governments involved represent two different states. That inquiry, however, raises the question of how to tell when a government properly represents a state; this is the question of recognition. The irreducibly political aspect of recognition creates ambiguities of the sort that international humanitarian law rejects. Accordingly, a government-centered framework for conflict classification provides another reason to eliminate the distinction between IACs and NIACs.

  1. .Tim Lister, Tara John & Paul P. Murphy, Here’s What We Know About How Russia’s Invasion of Ukraine Unfolded, CNN, https://www.cnn.com/2022/02/24/europe/ukraine-russia-attack-timeline-intl/index.html [https://perma.cc/QWL4-L9PS] (Feb. 24, 2022, 11:42 AM); Timeline: The Events Leading Up to Russia’s Invasion of Ukraine, Reuters (Mar. 1, 2022, 3:03 AM), https://www.reuters.com/world/europe/events-leading-up-russias-invasion-ukraine-2022-02-28/ [https://perma.cc/8QEG-YJLR].
  2. .Lister et al., supra note 1.
  3. .Vast Russian Military Convoy Closes In on Ukraine Capital, Al Jazeera (Mar. 1, 2022), https://www.aljazeera.com/news/2022/3/1/russia-amasses-65km-army-convoy-near-ukraine-capital [https://perma.cc/6HZ2-69QM].
  4. .Panic, Fear and Stoicism in Ukraine as Russia Invades, Al Jazeera (Feb. 24, 2022, 2:13 PM), https://www.aljazeera.com/news/2022/2/24/ukrainian-civilians-react-as-russian-forces-invade [https://perma.cc/422J-DFQX].
  5. .Carlotta Gall, As the War in Ukraine Moves into Its Second Month, Fears Grow of Mariupol’s Fall to Russia, N.Y. Times (Mar. 27, 2022), https://www.nytimes.com/2022/03/27‌/world/europe/russia-ukraine-war-combat.html [https://‌perma.cc/5ANR-6WH5].
  6. .See Ivan Nechepurenko & Eric Nagourney, Russia Signals That It May Want a Bigger Chunk of Ukraine, N.Y. Times (July 20, 2022), https://www.nytimes.com/2022/07/20/world/europe/putin-ukraine-invasion-russia-war.html [https://perma.cc/A6CW-9FEF] (describing Russian seizure of Ukrainian territory); David Leonhardt, Ukraine on the March, N.Y. Times (Sept. 13, 2022), https://www.nytimes.com/2022/09/13/briefing/ukraine-counteroffensive-kharkiv-war.html [https://‌perma.cc/X5TH-KASH] (describing the Ukrainian counterattack).
  7. .See Michael Crowley, U.S. Calls Putin a ‘War Criminal,’ but Consequences Are Unclear, N.Y. Times (Mar. 17, 2022), https://www.nytimes.com/2022/03/17/us/politics/russia-ukraine-war-crimes.html [https://perma.cc/E9NE-XCS4] (reporting Secretary of State Antony Blinken’s statement that “[i]ntentionally targeting civilians is a war crime” and summarizing Professor Oona Hathaway’s articulation of the relevant legal standards); Carole Landry, The Horror in Bucha, N.Y. Times (Apr. 4, 2022), https://www.nytimes.com/2022/04/04/briefing/russia-ukraine-war-briefing-bucha-warcrimes.html [https://perma.cc/MCL7-9ZTB] (connecting killings of Ukrainian civilians in occupied areas with human rights organizations’ accusations of war crimes by Russian forces); What Is International Humanitarian Law?, Int’l Comm. Red Cross (Apr. 6, 2022), https://www.icrc.org/en/document/what-international-humanitarian-law [https://perma.cc/Z9FE-KMPL] (identifying “the law of armed conflict” as a synonym for international humanitarian law).
  8. .E.g., Landry, supra note 7; Dominic Casciani, What Is a War Crime and Could Putin Be Prosecuted Over Ukraine?, BBC (July 7, 2022), https://www.bbc.com/news/world-60690688 [https://perma.cc/G3BP-2TA8].
  9. .Ministry of Defence (@DefenceHQ), Twitter (Apr. 23, 2022, 3:00 PM), https://twitter.com/DefenceHQ/status/1517956476211503105 [https://perma.cc/45EP-LA99].
  10. .See Bohdan Vitvitsky, At What Point Do Russian War Crimes in Ukraine Qualify as Genocide?, Atl. Council (Apr. 11, 2022), https://www.atlanticcouncil.org/blogs/ukrainealert/at-what-point-do-russian-war-crimes-in-ukraine-qualify-as-genocide [https://perma.cc/ET45-SGXH] (“Much depends on the issue of intent.”); Max Fisher, Putin’s Baseless Claims of Genocide Hint at More Than War, N.Y. Times (Feb. 19, 2022), https://www.nytimes.com/2022/02/19/‌world/europe/putin-ukraine-genocide.html [https://‌perma.cc/UV53-J85N] (describing Russian President Vladimir Putin’s baseless prewar claims that Ukraine was committing genocide); Ukraine’s President Accuses Russia of Committing ‘Genocide,Al Jazeera (Apr. 3, 2022), https://www.aljazeera.com/news/2022/4/3/ukraines-president-calls-civilian-killings-by-russia-genocide [https://perma.cc/MMU9-EFTR] (quoting Ukrainian President Volodymyr Zelenskyy as saying “[i]ndeed, this is genocide” in response to Russian actions in Ukraine).
  11. .Farnaz Fassihi, The U.N. General Assembly Passes a Resolution Strongly Condemning Russia’s Invasion, N.Y. Times (March 2, 2022), https://www.nytimes.com/2022/03/02/world/‌europe/russia-un-invasion-condemn.html [https://perma.cc/L3D7-RVTN].
  12. .Michelle Nichols & Humeyra Pamuk, Russia Vetoes U.N. Security Action of Ukraine as China Abstains, Reuters (Feb. 25, 2022, 9:13 PM), https://www.reuters.com/world/russia-vetoes-un-security-action-ukraine-china-abstains-2022-02-25 [https://perma.cc/E2XP-EKV5].
  13. .Allegations of Genocide Under the Convention on the Prevention and Punishment of the Crime of Genocide (Ukr. v. Russ.), Order, ¶ 86 (Mar. 16, 2022), https://www.icj-cij.org/public/files/case-related/182/182-20220316-ORD-01-00-EN.pdf [https://perma.cc/T8GH-NQUM].
  14. .See Marlise Simons, The International Criminal Court Prosecutor Fast-Tracks an Investigation of Possible War Crimes in Ukraine, N.Y. Times (Mar. 3, 2022), https://www.nytimes.com/2022/03/03/world/europe/war-crimes-russia-ukraine-icc.html [https://‌perma.cc/GNV4-WRUT] (reporting prosecutor’s announcement); Roger Cohen, A Surge of Unifying Moral Outrage Over Russia’s War, N.Y. Times, https://www.nytimes.com/2022/03/01/‌world/europe/zelensky-ukraine-war-outrage.html [https://‌perma.cc/RQY2-VL8G] (Mar. 3, 2022) (highlighting global outrage).
  15. .Max Fisher, Why Calls for War Crimes Justice Over Ukraine Face Long Odds, N.Y. Times (Apr. 10, 2022), https://www.nytimes.com/2022/04/10/world/europe/russia-ukraine-war-crimes.html [https://perma.cc/XAX5-YMXC].
  16. .See Jaime Lopez & Brady Worthington, The ICC Investigates the Situation in Ukraine: Jurisdiction and Potential Implications, Lawfare (Mar. 10, 2022, 10:08 AM), https://www.lawfareblog.com/icc-investigates-situation-ukraine-jurisdiction-and-potential-implications [https://perma.cc/5ZVJ-3MHL] (“In light of Ukraine’s acceptance of jurisdiction, the necessary conditions for the ICC’s exercise of jurisdiction are indeed present.”).
  17. .Cf. Fisher, supra note 15 (describing how ICC arrest warrants could not be executed against Sudan’s ex-leader Omar al-Bashir in the countries that continued to host him).
  18. .See id. (identifying several difficulties involved in prosecutions for war crimes).
  19. .See W.J. Hennigan, Russia Has ‘Every Intention of Decapitating’ Ukraine’s Government: Senior U.S. Defense Official, Time (Feb. 24, 2022, 5:04 PM), https://time.com/6151024/russian-invasion-ukraine-kyiv-government-us-defense [https://perma.cc/L98C-7D62] (identifying a U.S. defense official who stated that Russia had “every intention of decapitating the government and installing their own method of governance”).
  20. .See Michael Holden, UK Accuses Kremlin of Trying to Install Pro-Russian Leader in Ukraine, Reuters (Jan. 22, 2022, 8:48 PM), https://www.reuters.com/world/uk/uk-accuses-kremlin-trying-install-pro-russian-leader-ukraine-2022-01-22/ [https://perma.cc/6TPX-FJJ7] (reporting U.K. assessments that Russia intended to install a pro-Russian leader in Ukraine).
  21. .See infra subpart I(B).
  22. .See infra subpart I(B).
  23. .See Steven Haines, The Nature of War and the Character of Contemporary Armed Conflict, in International Law and the Classification of Conflicts 9, 9 (Elizabeth Wilmshurst ed., 2012) (“Contemporary forms of hostilities are less frequently conflicts between States than a variety of armed struggles involving not only States but a growing number of organized armed groups motivated by a wide range of interests.”).
  24. .See infra notes 124–128 and accompanying text; see also Hans-Peter Gasser, Internationalized Non-International Armed Conflicts: Case Studies of Afghanistan, Kampuchea, and Lebanon, 33 Am. U. L. Rev. 145, 157–58 (1983) (explaining that difficulties involved with classifying certain conflicts allow states to deny the existence of the conflicts).
  25. .Kubo Mačák, Internationalized Armed Conflicts in International Law 23 (Suzannah Linton, Robert Cryer & Salvatore Zappalà eds., 2018).
  26. .Geoffrey S. Corn, Thinking the Unthinkable: Has the Time Come to Offer Combatant Immunity to Non-State Actors?, 22 Stan. L. & Pol’y Rev. 253, 254 (2011).
  27. .See Jelena Pejic, Conflict Classification and the Law Applicable to Detention and the Use of Force, in International Law and the Classification of Conflicts, supra note 23, at 80, 86–90 (detailing the procedural protections applicable in IACs and recognizing that there are no procedural protections for detainees in NIACs).
  28. .See infra notes 143–45 and accompanying text.
  29. .See Dapo Akande, Classification of Armed Conflicts: Relevant Legal Concepts, in International Law and the Classification of Conflicts, supra note 23, at 32, 33–34 (describing the accession of the Geneva Conventions as the law regulating conflict classification).
  30. .Laura Lopez, Uncivil Wars: The Challenge of Applying International Humanitarian Law to Internal Armed Conflicts, 69 N.Y.U. L. Rev. 916, 921–22 (1994).
  31. .Id. at 924.
  32. .Jean S. Pictet, The New Geneva Conventions for the Protections of War Victims, 45 Am. J. Int’l L. 462, 468 (1951).
  33. .The Geneva Conventions of 1949 and Their Additional Protocols, Int’l Comm. Red Cross (Jan. 1, 2014), https://www.icrc.org/en/document/geneva-conventions-1949-additional-protocols [https://perma.cc/4X8P-JHUE].
  34. .Akande, supra note 29, at 34.
  35. .See The Geneva Conventions of 1949 and Their Additional Protocols, supra note 33 (identifying the Additional Protocols as a response to the development of conflicts in the two decades since the Conventions were adopted).
  36. .Akande, supra note 29, at 33–34.
  37. .Geoffrey S. Corn, Hamdan, Lebanon, and the Regulation of Hostilities: The Need to Recognize a Hybrid Category of Armed Conflict, 40 Vand. J. Transnat’l L. 295, 300 (2007).
  38. .Id. at 300–01.
  39. .See, e.g., Geneva Convention Relative to the Treatment of Prisoners of War art. 4–132, Aug. 12, 1949, 6 U.S.T. 3316, 75 U.N.T.S. 135 [hereinafter POW Convention] (providing the extensive rules regulating the treatment of prisoners of war).
  40. .Corn, supra note 37, at 302.
  41. .See id. (pointing out that armed conflict is a term “undefined by the express language of either Common Article 2 or 3”).
  42. .Prosecutor v. Tadić, Case No. IT-94-1-I, Decision on Defence Motion for Interlocutory Appeal on Jurisdiction, ¶ 70 (Int’l Crim. Trib. for the Former Yugoslavia Oct. 2, 1995).
  43. .See Corn, supra note 37, at 302 (describing the ICRC commentaries as “the primary interpretative aid” for understanding CA2 and CA3).
  44. .Int’l Comm. of the Red Cross, Commentary on the Third Geneva Convention 92–93 (2021) [hereinafter Third Convention Commentary].
  45. .Compare id. at 92 (“Armed conflicts . . . are those which . . . occur when one or more States have recourse to armed force against another State, regardless of the reasons for or the intensity of the confrontation.”), with id. at 157 (“A situation of violence that crosses the threshold of an ‘armed conflict not of an international character’ is a situation in which organized Parties confront one another with violence of a certain degree of intensity. It is a determination made based on the facts.”).
  46. .Corn, supra note 37, at 300–01.
  47. .Akande, supra note 29, at 34.
  48. .Id.
  49. .Id.
  50. .Id. at 35.
  51. .Id.
  52. .Id. at 34.
  53. .Corn, supra note 37, at 300.
  54. .Geneva Convention Relative to the Protection of Civilian Persons in Time of War art. 33, Aug. 12, 1949, 6 U.S.T. 3516, 75 U.N.T.S. 287 (“Reprisals against protected persons and their property are prohibited.”).
  55. .POW Convention, supra note 39, 6 U.S.T. at 3340, 75 U.N.T.S. at 158.
  56. .Jakob Kellenberger, Humanitarian Law: More Effective 25 Years Later, N.Y. Times (June 8, 2002), https://www.nytimes.com/2002/06/08/opinion/IHT-humanitarian-law-more-effective-25-years-later.html [https://perma.cc/6LVX-MAS2].
  57. .POW Convention, supra note 39, 6 U.S.T. at 3318, 75 U.N.T.S. at 136. As mentioned, the texts of CA2 and CA3 are identical across all the Conventions. See supra, note 37 and accompanying text. I cite to the text of the POW Convention because its accompanying ICRC commentary is the most recent—it was published in 2021. See Third Convention Commentary, supra note 44, at xviii (reflecting on the role of the ICRC commentaries and on the commentaries to the First and Second Geneva Conventions, which were published in 2016 and 2017 respectively).
  58. .See supra note 33 and accompanying text.
  59. .Akande, supra note 29, at 39.
  60. .See id. at 39–40 (explaining the differing rules of application).
  61. .Id. at 39.
  62. .See id. at 39–40 (“[W]here the parties failed to consider themselves at war they were able to escape the application of the laws of war.”).
  63. .See supra notes 42–44 and accompanying text.
  64. .Prosecutor v. Tadić, Case No. IT-94-1-I, Decision on Defence Motion for Interlocutory Appeal on Jurisdiction, ¶ 70 (Int’l Crim. Trib. for the Former Yugoslavia Oct. 2, 1995) (emphasis added); see also Akande, supra note 29, at 40–41 (focusing on this component of the ICTY’s statement in interpreting the term armed conflict in CA2).
  65. .See Akande, supra note 29, at 41 (“Almost any use of armed force by one State against another will bring into effect an international armed conflict . . . .”).
  66. .Third Convention Commentary, supra note 44, at 93.
  67. .See Akande, supra note 29, at 49 (identifying that API responded “to the desire, mainly of developing countries, for legitimation of those engaged in liberation struggles”); Corn, supra note 26, at 269 (explaining that API internationalized three kinds of previously non-international armed conflicts).
  68. .Protocol Additional to the Geneva Conventions of 12 August 1949, and Relating to the Protection of Victims of International Armed Conflicts (Protocol I) art. 1(4), June 8, 1977, 1125 U.N.T.S. 3.
  69. .Corn, supra note 26, at 269.
  70. .Id. at 270.
  71. .Id.
  72. .See id. at 272 (identifying the U.S. view that “allowing the motivation for armed struggle to dictate the status of a conflict contradicted the emphasis on de facto conflict character that defined the Common Article 2/3 law-triggering paradigm”).
  73. .Id.
  74. .See supra note 51 and accompanying text.
  75. .See supra note 48 and accompanying text.
  76. .See POW Convention, supra note 39, 6 U.S.T. at 3318, 75 U.N.T.S. at 136 (pronouncing that in NIACs, the parties to the conflict “shall be bound to apply, as a minimum, the following provisions” contained within CA3).
  77. .See Mačák, supra note 25, at 18–19 (commenting on the convention in miniature moniker).
  78. .Id. at 19.
  79. .See POW Convention, supra note 39, 6 U.S.T. at 3318, 3320, 75 U.N.T.S. at 136, 138 (forbidding the use of torture against “[p]ersons taking no active part in the hostilities”).
  80. .See Protocol Additional to the Geneva Conventions of 12 August 1949, and Relating to the Protection of Victims of Non-International Armed Conflicts (Protocol II) art. 1, June 8, 1977, 1125 U.N.T.S. 609 [hereinafter Additional Protocol II] (providing the triggering condition for the substantive protections of APII).
  81. .United States: Message from the President Transmitting Protocol II Additional to the 1949 Geneva Conventions, Relating to the Protection of Victims of Noninternational Armed Conflicts, 26 I.L.M. 561, 562 (1987).
  82. .Additional Protocol II, supra note 80, at 612 (forbidding “[s]lavery and the slave trade in all their forms” against “[a]ll persons who do not take a direct part . . . in hostilities”).
  83. .See supra note 51 and accompanying text.
  84. .POW Convention, supra note 39, 6 U.S.T. at 3318, 75 U.N.T.S. at 136.
  85. .See supra note 33 and accompanying text.
  86. .Prosecutor v. Tadić, Case No. IT-94-1-I, Decision on Defence Motion for Interlocutory Appeal on Jurisdiction, ¶ 70 (Int’l Crim. Trib. for the Former Yugoslavia Oct. 2, 1995); see supra notes 42–44 and accompanying text.
  87. .Mačák, supra note 25, at 19.
  88. .Id.
  89. .See id. (“[T]he hostilities must surpass a certain level of intensity, for instance when the police forces of the state in question are no longer capable of dealing with the insurrection, and therefore the army has to be mobilized in order to defeat the insurgents.”).
  90. .Akande, supra note 29, at 53.
  91. .Id. at 50 (“Where a situation of violence is regarded merely as one of internal strife of civil disturbance, international law considers that it does not reach the threshold of ‘armed conflict’ . . . .”).
  92. .Mačák, supra note 25, at 20.
  93. .Additional Protocol II, supra note 80, at 611.
  94. .See Protocol Additional to the Geneva Conventions of 12 August 1949, and Relating to the Protection of Victims of Non-International Armed Conflicts (Protocol II), 8 June 1977, Int’l Comm. Red Cross, https://ihl-databases.icrc.org/applic/ihl/ihl.nsf/INTRO/475?OpenDocument [https://perma.cc/2J97-WWS2] (identifying that 169 states are party to the treaty and 3 states are signatories).
  95. .See Akande, supra note 29, at 54 (stating that APII “does not apply to situations of internal disturbance and tensions such as riots, isolated and sporadic acts of violence (the threshold for ‘armed conflict’)” and that its threshold for application is higher than that for CA3).
  96. .Id.
  97. .See Mačák, supra note 25, at 21 (“[Article 1(1)] requires the insurgent group to be in control of a discernible part of the state’s territory . . . .”).
  98. .Haines, supra note 23, at 29.
  99. .See Mačák, supra note 25, at 22–23 (linking this trend of convergence to the “reality of NIACs”).
  100. .See Françoise J. Hampson, Afghanistan 2001–2010, in International Law and the Classification of Conflicts, supra note 23, at 242, 278–79 (providing a timeline overview of the conflicts in Afghanistan).
  101. .See id. at 242–43 (describing the conflict between the Taliban and the Northern Alliance, headed by Ahmad Shah Massoud).
  102. .See id. at 243 (“[T]he US and its Coalition partners had to fight the [Taliban] in order to be able to reach Al-Qaeda, their real goal.”).
  103. .Id. at 279.
  104. .Id. at 244.
  105. .See id. at 253 (chronicling the conflict between the ISAF and the Taliban through 2008).
  106. .See Charlie Savage, U.S. Eases Sanctions to Allow Routine Transactions with Afghan Government, N.Y. Times (Feb. 25, 2022), https://www.nytimes.com/2022/02/25/us/politics/us-sanctions-afghanistan.html [https://perma.cc/XQE7-47SG] (commenting that the Taliban takeover occurred in August 2021).
  107. .David Zucchino, Shifting to Governing, Taliban Will Name Supreme Afghan Leader, N.Y. Times, https://www.nytimes.com/2021/09/01/world/asia/afghanistan-taliban-government-leader.html [https://perma.cc/7REW-2JMK] (Sept. 21, 2021).
  108. .The continued existence of an armed conflict is a separate question that raises different issues. Because the Taliban takeover did not necessarily coincide with a cessation of hostilities with the United States, my analysis focuses on the complexities raised by the changing status of the parties under the law of armed conflict by assuming that an armed conflict continues.
  109. .See, e.g., Michael Ames, A New Video Shows a Missing American Hostage Pleading for Help in Taliban Custody, New Yorker (Apr. 1, 2022), https://www.newyorker.com/news/news-desk/a-new-video-shows-a-missing-american-hostage-pleading-for-help-in-taliban-custody [https:‌//perma.cc/65Y3-ZPNN] (conveying the story of Mark Frerichs, an American Navy veteran who had been detained by the Taliban). Mr. Frerichs was released from Taliban custody in September 2022. US–Taliban Prisoner Swap: Who Are Mark Frerichs, Bashir Noorzai, Al Jazeera (Sept. 19, 2022), https://www.aljazeera.com/news/2022/9/19/us-taliban-prisoner-swap-who-are-mark-frerichs-bashir-noorzai [https://perma.cc/ZZ3E-THAK].
  110. .See supra note 101 and accompanying text.
  111. .Hampson, supra note 100, at 252.
  112. .See supra note 93–96 and accompanying text. Afghanistan only became a signatory to the Additional Protocols in 2009, so the Additional Protocols would not have applied by their own terms prior to then. Hampson, supra note 100, at 243. Still, the customary international law equivalent of APII may have applied, which carries the same problems regarding conflict classification. See id. at 252 (asking whether the customary law equivalent of Additional Protocol II applied to the conflict).
  113. .See Hampson, supra note 100, at 245 & n.9 (looking at the Taliban’s claim to govern Afghanistan and the recognition thereof in the context of classifying the conflict).
  114. .See id. at 245 n.9, 251 (identifying the conflict as an IAC, but noting that a different view on the question of the relationship between the Taliban and Afghanistan could render the conflict a NIAC).
  115. .See infra notes 202–204 and accompanying text.
  116. .See Hampson, supra note 100, at 254 (describing involvement in ISAF by several states, including the United States, the United Kingdom, Canada, and Germany).
  117. .See id. at 256 (“[T]he Taliban no longer represent the armed forces of Afghanistan . . . .”).
  118. .See id. at 244 (detailing the meeting that assented to the Karzai government, followed by the meeting’s request for deployment of a security force).
  119. .Cf. id. at 245 (“[G]iven the priority which the Geneva Conventions attach to facts, it may be the case that the forces of the State are those forces in fact exercising authority over the majority of the territory . . . .”).
  120. .Id. at 244. Notice too that the Karzai government could only give consent to the ISAF if it, in fact, represented Afghanistan. Id. at 245. Françoise Hampson, whose thoughtful treatment of the conflicts in Afghanistan I heavily rely on in this subpart, dodges the inherent complexity of this issue, stating: “It will be assumed that, from its first deployment, ISAF was present in Afghanistan with the effective consent of the national authorities.” Id.
  121. .Id. at 245.
  122. .See id. at 256 (classifying the conflicts involving the United States against the Taliban after the point that “the Taliban no longer represent[ed] the armed forces of Afghanistan” as a NIAC).
  123. .See supra note 106 and accompanying text.
  124. .See Thomas Gibbons-Neff & Yaqoob Akbary, In Afghanistan, ‘Who Has the Guns Gets the Land’, N.Y. Times, https://www.nytimes.com/2021/12/03/world/asia/afghanistan-land-ownership-taliban.html [https://perma.cc/3QRS-7QFS] (Dec. 9, 2021) (characterizing the post-takeover Taliban as “the new government,” and noting that land disputes arising from their accession mirror disputes arising out of previous changes in government); Savage, supra note 106 (reporting that the Biden administration has relaxed sanctions against Afghan institutions, normalizing relations to some extent).
  125. .Hampson, supra note 100, at 249–50.
  126. .Id. at 255.
  127. .See id. at 251 (“It is not easy however to ascertain the date on which the Coalition took the view that the international armed conflict had come to an end and that a non-international armed conflict was in progress.”).
  128. .Id. at 255.
  129. .See Louise Arimatsu, The Democratic Republic of the Congo 1993–2010, in International Law and the Classification of Conflicts, supra note 23, at 146, 176–77 (reiterating a characterization of the Second Congo War as “legally complex” and identifying the factors that complicate the classification).
  130. .See Iain Scobbie, Lebanon 2006, in International Law and the Classification of Conflicts, supra note 23, at 387, 401 (describing the complications in conflict classification arising out of the relationship between Hezbollah and the Lebanese government with regard to military intervention by Israel).
  131. .See Felicity Szesnat & Annie R. Bird, Colombia, in International Law and the Classification of Conflicts, supra note 23, at 203, 236–37 (analyzing the classification of the conflict between Colombia and guerilla dissidents, in which Colombia used armed force on the territory of Ecuador without the latter’s consent).
  132. .Haines, supra note 23, at 30.
  133. .See supra notes 98–99 and accompanying text.
  134. .Mačák, supra note 25, at 22–23; see supra subpart I(B).
  135. .Prosecutor v. Tadić, Case No. IT-94-1-I, Decision on Defence Motion for Interlocutory Appeal on Jurisdiction (Int’l Crim. Trib. for the Former Yugoslavia Oct. 2, 1995); Mačák, supra note 25, at 241.
  136. .Mačák, supra note 25, at 22.
  137. .Id.
  138. .Id. at 21 (identifying that “[a] growing number of international conventions regulating the use of weapons apply equally to both types of conflict”).
  139. .See id. at 22 & n.160 (collecting military sources pronouncing the consistent application of the law of armed conflict, including from the United States and Germany).
  140. .See, e.g., Roy S. Schöndorf, Extra-State Armed Conflicts: Is There a Need for a New Legal Regime?, 37 N.Y.U. J. Int’l L. & Pol. 1, 7 (2004) (proposing “the conceptualization of extra-state armed conflicts as a separate category of armed conflicts”).
  141. .See Corn, supra note 37, at 327 (pointing to the Supreme Court’s decision in Hamdan v. Rumsfeld, 548 U.S. 557 (2006), and to the conflict in Lebanon as the catalysts making it “necessary to consider [the] critical evolution” of a new category of armed conflict).
  142. .Id.
  143. .See Akande, supra note 29, at 39 & n.31 (citing scholars who argue in favor of abolishing the distinction); Eric Talbot Jensen, Future War, Future Law, 22 Minn. J. Int’l L. 282, 290–91, 291 n.38 (2013) (citing scholars for the proposition that the distinction has lost its usefulness).
  144. .Steven R. Ratner, The Schizophrenias of International Criminal Law, 33 Tex. Int’l L.J. 237, 238 (1998).
  145. .Akande, supra note 29, at 39.
  146. .Mačák, supra note 25, at 25–26, 105 (addressing the literature on the internationalization and de-internationalization of armed conflicts).
  147. .See id. at 2, 110–11 (conceptualizing the transformative process of internationalization and de-internationalization).
  148. .See id. at 2 (“[T]he intra-state origin of [internationalized non-international armed conflicts] provides for an uneasy match with many of the precepts of the law of IAC . . . .”).
  149. .See id. at 183 (“The triangular relationship at the heart of the law of belligerent occupation between the occupying power, the ousted power, and the occupied population cannot be easily transposed into the reality of a conflict occurring in the territory of a single state.”).
  150. .Id. at 22–23.
  151. .Chinyere Obasi, It’s Time to Ratify the Rome Statute. No, Really This Time., Harv. Pol. Rev. (Nov. 22, 2021), https://harvardpolitics.com/ratify-rome [https://perma.cc/ABL4-QEH2].
  152. .Mačák, supra note 25, at 23.
  153. .Id.
  154. .See Akande, supra note 29, at 37 (“[T]he distinction between international and non-international armed conflicts can be explained by reference to the history of the development of international law . . . .”).
  155. .See Mačák, supra note 25, at 17 (explaining that states were hesitant to change from a paradigm where they were “generally unconstrained when responding to domestic insurgencies”).
  156. .See Third Convention Commentary, supra note 44, at 145–46 (“Common Article 3 represented one of the first provisions of international law that dealt with what was at the time considered by States as being exclusively their domestic affair.”).
  157. .See id. at 150 (describing how attention to NIACs emerged out of the experiences of the Spanish Civil War and the Greek Civil War).
  158. .Akande, supra note 29, at 37.
  159. .See id. (“The main reason for the persistence of the distinction is the view by States, or some of them, that equating non-international and international armed conflicts would undermine State sovereignty . . . .”).
  160. .Raymund T. Yingling & Robert W. Ginnane, The Geneva Conventions of 1949, 46 Am. J. Int’l L. 393, 395 (1952).
  161. .See Corn, supra note 26, at 293 (identifying the contemporary focus of international humanitarian law as “the protection of the human person and all victims of war”).
  162. .Cf. id. at 279 (providing one explanation for the denial of combatant immunity to nonstate armed groups: denying them legitimacy).
  163. .Akande, supra note 29, at 37.
  164. .See, e.g., Saeed Bagheri, International Law and the War with Islamic State 8–10 (2021) (previewing the role of energy resources, religious beliefs, and the weakness of the Iraqi government in the rise of the Islamic State).
  165. .Cf. Corn, supra note 26, at 283 (arguing that “[i]t is difficult to imagine” that extending combatant immunity to nonstate actors “would provide a significant motivation for the recruiting efforts of these organizations”).
  166. .One answer could be the nonstate status of one of the parties. But this answer again raises the question of why the sovereign status of a party should make a difference in the substance of the applicable law.
  167. .See Corn, supra note 26, at 255 (identifying that the combatant immunity following from prisoner-of-war status is “perhaps the most significant exception” to the trend of the convergence of IACs and NIACs).
  168. .See supra note 26 and accompanying text.
  169. .Corn, supra note 26, at 256.
  170. .Id. at 254–56.
  171. .See id. at 255 (“States have been absolutely unwilling to extend [prisoner-of-war status] with its accordant lawful combatant immunity to non-state operatives.”).
  172. .Id.
  173. .See Akande, supra note 29, at 38 (“[I]f the principle of combatants’ immunity . . . were to be applied in non-international armed conflicts, States would be unable to criminalize acts which are traditionally regarded as treasonous.”).
  174. .See Corn, supra note 26, at 253 (describing how the United States deployed the concept of an “unlawful enemy combatant” to justify its decision to detain certain fighters while also denying the application of Convention protections).
  175. .See supra subpart I(B).
  176. .See supra subpart II(A).
  177. .See Hersch Lauterpacht, Recognition in International Law 87 (1947) (“It is a fundamental rule of international law that every independent State is entitled to be represented in the international sphere by a government . . . .”); Third Convention Commentary, supra note 44, at 95 (“The government is only one of the constitutive elements of the State . . . .”).
  178. .Lauterpacht, supra note 177, at 26.
  179. .Id. at 94–95.
  180. .See, e.g., Corn, supra note 26, at 281 (identifying internal dissident forces as a “threat to state sovereignty” because they seek to overthrow the “lawful government”).
  181. .Lauterpacht, supra note 177, at 92–93 (“[T]he State and its obligations remain the same notwithstanding constitutional or governmental changes, revolutionary or other.”).
  182. .M.J. Peterson, Recognition of Governments 20 (1997).
  183. .See Mikulas Fabry, Recognizing States 7 (2010) (indicating the consequences of a state’s recognition by and participation in the international community).
  184. .See Peterson, supra note 182, at 21 (positing that a government is bound to the rights and obligations of the state).
  185. .Id. at 20.
  186. .Id.
  187. .Compare id. at 21 (“[E]ven an unrecognized government that actually rules all or most of the territory of its state has certain competences that must be accepted . . . .”), with Thomas D. Grant, The Recognition of States 1 (1999) (describing the debate between the “constitutive” and “declarative” models of state recognition, where the former suggests that states are essentially de jure entities and the latter suggests that they are essentially de facto entities).
  188. .Cf. Peterson, supra note 182, at 21 (enumerating maxims for understanding states’ continuing international legal commitments).
  189. .See, e.g., Third Convention Commentary, supra note 44, at 157 (stating that whether violence crosses a threshold into armed conflict is “a determination made based on the facts”); Iain Scobbie, Gaza, in International Law and the Classification of Conflicts, supra note 23, at 280, 296 (“Traditionally, the test for the termination of an occupation was seen as a simple question of fact.”); Scobbie, supra note 130, at 405 (“Reliance solely on formal legal provision [for determining whether a nonstate actor’s behavior can be attributed to the state] is . . . inadequate and ‘a State cannot avoid responsibility for the conduct of a body which does in truth act as one of its organs . . . .’”); Mačák, supra note 25, at 117 (rejecting a theory of understanding the effectiveness of governments because “it is incompatible with the requirement that [international humanitarian law’s] scope of applicability must be determined on the basis of objective and factual criteria”).
  190. .See supra notes 65–66 and accompanying text.
  191. .See supra subsection I(B)(2)(a).
  192. .Peterson, supra note 182, at 20.
  193. .See supra notes 125–128 and accompanying text.
  194. .See supra notes 65–66 and accompanying text.
  195. .See Third Convention Commentary, supra note 44, at 97 (considering the issue of who gets to represent a state as impactful for classifying a conflict as an IAC).
  196. .See supra subsection I(B)(2)(a).
  197. .Third Convention Commentary, supra note 44, at 97.
  198. .See supra note 189 and accompanying text.
  199. .Third Convention Commentary, supra note 44, at 98.
  200. .How would this approach best understand three governments that each exercised effective control over 33% of a state? The fact that these questions seem silly highlights that the ICRC’s de facto approach involves the same sort of ambiguities and arbitrariness it was designed to erase.
  201. .Peterson, supra note 182, at 1.
  202. .See id. at 170–71 (reflecting on political usages of recognition).
  203. .Mačák, supra note 25, at 116–17.
  204. .Stefan Talmon, Recognition of Governments in International Law: With Particular Reference to Governments in Exile 14 (1998).
  205. .Mačák, supra note 25, at 117.
  206. .See, e.g., Third Convention Commentary, supra note 44, at 98 (“[I]t does not matter that a government failed to gain recognition by the international community at large.”); Elizabeth Wilmshurst, Conclusions, in International Law and the Classification of Conflicts, supra note 23, at 478, 483 (“[I]t is the facts rather than a subjective act of recognition alone which determines the category of armed violence.”).
  207. .See supra notes 24, 125–128 and accompanying text.
  208. .See, e.g., Abdulqawi A. Yusuf, Somalia’s Warlords: Preying on a Failed State, N.Y. Times (Jan. 21, 2004), https://www.nytimes.com/2004/01/21/opinion/somalias-warlords-preying-on-a-failed-state.html [https://perma.cc/QKM2-MH7J] (describing the ineffectiveness of a central government in Somalia and thereby arguing for external intervention); Eric McGlinchey, Running in Circles in Kyrgyzstan, N.Y. Times (Apr. 9, 2010), https://www.nytimes.com/2010/‌04/10/opinion/10mcglinchey.html [https://perma.cc/XM5L-HQSS] (asserting that Kyrgyzstan was “a failed state that need[ed] a couple of steady outside hands to help it succeed”).
  209. .See supra note 161 and accompanying text.
  210. .See supra notes 150–153 and accompanying text.
  211. .See Mačák, supra note 25, at 144 & n.24 (citing Johnson v. Eisentrager, 339 U.S. 763, 793 (1950) (Black, J., dissenting) (cautioning that “[i]t must be remembered that legitimate ‘acts of warfare,’ however murderous, do not justify criminal conviction”)); id. at 144 (“[T]he availability of combatant status contributes towards the general goal of securing compliance with [international humanitarian law].”).
  212. .See id. at 141–42 (identifying that combatant immunity helps balance the diametrically opposed first-order principles of international humanitarian law: “the principle of military necessity and the principle of humanity”).
  213. .See supra note 199 and accompanying text.
  214. .Besides this definitional point, it is also relevant that there are existing legal mechanisms ensuring that immunity would only extend to those with the capacity to comply with international humanitarian law. For example, the POW Convention requires that, for heightened prisoner-of-war protections to apply, the captured person must be a member of an organized and disciplined group. See POW Convention, supra note 39, 6 U.S.T. at 3320, 75 U.N.T.S. at 138 (describing protected persons, including those belonging to a party to the conflict, those professing allegiance to a government not recognized by the detaining power, and those belonging to a militia meeting a set of organizing conditions).
  215. .See supra notes 167–174 and accompanying text.
  216. .See generally Corn, supra note 26 (asking whether combatant immunity should be extended to nonstate actors); see also Mačák, supra note 25, at 162–63 (finding unpersuasive possible arguments for refusing to extend combatant immunity to fighters in internationalized non-international armed conflicts).
  217. .See supra note 188 and accompanying text.
  218. .For example, Geneva Call, a nongovernmental organization, engages with nonstate armed groups to encourage them to adopt humanitarian obligations. What We Do, Geneva Call, https://www.genevacall.org/what-we-do/ [https://perma.cc/C7LM-3SJ5].
  219. .See generally, Liesbeth Zegveld, Accountability of Armed Opposition Groups in International Law (2002) (examining accountability for nonstate armed groups and identifying gaps in current efforts to hold nonstate groups accountable for violations of the law of armed conflict).