About the author(s):
Nathan Derejko
Dr Nathan Derejko is the Mauro Chair in Human Rights and Social Justice and Assistant Professor of Law in the Faculty of Law at the University of Manitoba. Previously, Nathan was the Director of the Masters in Human Rights program at University College London (UK), Director of the Human Rights Clinic and Lecturer in Human Rights Law at the University of Essex (UK), and a Visiting Lecturer at the International Institute of International Humanitarian Law in Sanremo Italy.
Nathan’s research and teaching interests span three interrelated fields of international law: international humanitarian law, international human rights law, and the jus ad bellum. Nathan has a particular interest in the protection of human rights during armed conflict, counter-terrorism and human rights, and the law and practice of non-international armed conflict. Nathan’s research has been cited both within and beyond academia, including by the United Nations, the International Committee of the Red Cross, the New York City Bar Association, and has been covered by various media outlets, including the CBC, BBC, and Al-Jazeera English.
First, I would like to extend my sincere thanks to Gloria Gaggioli, Chiara Redaelli, Jelena Pejic, Katharine Fortin, René Provost, and Martha Bradley for taking the time to read the book and engage with its arguments. I also want to express my sincere thanks to the editorial teams at both Articles of War and Armed Groups and International Law for organising and facilitating this symposium, it’s been an intellectually enriching experience.
The contributions to this symposium were exceptionally insightful, offering a valuable opportunity to reflect on my own thinking and perspectives on the challenges and possibilities surrounding the identification of non-international armed conflict (NIAC). The authors raised compelling insights and arguments, several of which address complex questions that do not lend themselves to straightforward answers. Indeed, my thinking on many of these issues continues to develop. Due to space constraints, I will limit my comments to one or two key issues raised by each author.
Foundational Unresolved (?) NIAC Issues: Actors and Geography
Gloria Gaggioli provides a thought-provoking analysis of Chapter 1, regarding the material scope of NIAC, situating both the book’s arguments and her comments within the discourse of a singular NIAC definition that offers “simplified conceptual architecture” and a “streamlined approach” to NIAC identification. Gaggioli correctly notes that a distinction between Common Article 3 of the Geneva Conventions (CA 3) and Additional Protocol II to the Geneva Conventions (AP II) persists and two issues remain unresolved: the non-applicability of AP II to conflicts solely between organized armed groups (OAG); and its non-applicability to extraterritorial NIACs. Indeed, the book argues that these are the principal shortcomings of the material scope of AP II.
The shortfall of conventional international humanitarian law (IHL) in regulating NIACs exclusively between two or more OAG extends beyond AP II. Common Article 3 is notably silent on the composition of battlefield actors necessary for its activation, employing only the term “each Party.” The drafting history indicates that CA 3’s drafters did not discuss the possibility of a NIAC occurring between two or more OAGs absent the involvement of the territorial State, nor did the original International Committee of the Red Cross (ICRC) Commentaries favorably cater to the possibility. While the ICRC attempted to address this during the drafting of AP II, this lacuna in conventional IHL was not filled until the International Criminal Tribunal for former Yugoslavia’s Tadi? decision, which explicitly encompassed NIACs between two or more OAGs without State involvement. Today, the Tadi? Test for NIAC is the most authoritative method for determining the existence of a NIAC in practice and is widely considered as reflecting customary international law. According to the decision, “an armed conflict exists whenever there is a resort to … protracted armed violence between governmental authorities and organized armed groups or between such groups within a State.”
Regarding extraterritorial NIACs, again CA 3 is somewhat ambiguous about its geographical possibilities and the precise wording of the Tadi? definition, using the phrase “within a State,” does not lend itself favorably to extraterritorial interpretations. In practical terms, the legal possibility of extra-territorial NIACs was arguably made possible by the inclusion of the word “one” as opposed to “a” in the CA 3 and the universal ratification of the Geneva Conventions of 1949 and customary status of CA 3.
Specific challenges emerge when both CA 3 and AP II are applicable within the same theatre, possibly creating a fragmented legal regime dependent on actor and geographical location. The book examines a number of practical strategies to navigate this fragmented landscape, some of which are relevant to extraterritorial NIACs or extraterritorial hostilities of an otherwise internal NIAC. However, the book approaches these challenges by questioning the continued relevance of the CA 3 and AP II distinction in light of the Tadi? definition and corresponding Tadi? test for NIAC.
As Gaggioli notes, the book argues that Tadi? has essentially harmonized the activation thresholds and largely, albeit not exclusively, consolidated the rules applicable to CA 3 and AP II NIACs. Consequently, and in terms of conflict identification, AP II’s territorial restrictions are rendered somewhat redundant by the geographical possibilities of CA 3. In terms of conflict classification, the applicability of AP II activates its few remaining non-customary provisions, including the general amnesty provision (art. 6(5)), the various injunctions on the death penalty (art. 6(4)), and the due diligence requirement following the release of persons deprived of their liberty (art. 5(4)). These are important provisions and further research into the customary status of these rules could further harmonize the NIAC framework. However, complete elimination of the legal and operational differences between CA 3 and AP II remains unlikely.
How CA 3 and AP II Work Together
Because CA 3 and AP II emerged at different periods and in response to different factual and legal challenges, their respective legal architecture differs considerably, with respect to both the material concept of NIAC and substantive rules. For example, the so-called “lower threshold” of NIAC contained in AP II, internal disturbances and tensions, has incontrovertibly been extended to define the lower threshold of CA 3.
The invaluable guidance of AP II with respect to IHL’s temporal scope of application during NIAC naturally extends to CA 3 NIACs without controversy. Determining the personal scope of IHL requires drawing from both instruments, as CA 3 lacks the terms “civilian” and “civilian population,” while AP II omits “each Party” and makes no reference to individuals hors de combat.
In terms of regulating the conduct and consequences of NIAC, CA 3 includes valuable provisions that are absent from AP II, such as the provisions regarding ICRC initiative and special agreements, the latter being notably absent from the ICRC Customary IHL Study. In some areas, the content of AP II will inform the interpretation of CA 3’s laconic provisions, particularly with regard to the content and scope of humane treatment. Accordingly, the book argues that is not possible to effectively identify or adequately regulate contemporary NIACs without recourse to the specific provisions and supporting legal architecture of CA 3, AP II, and now Tadi?.
The NIAC Aggregated Intensity Threshold
Chiara Redaelli provides a sophisticated discussion on the role and relevance of the aggregated intensity approach when determining the threshold of NIAC in Chapter 2 of the book, applying this analysis to criminal armed violence to demonstrate its legal and operational utility. Redaelli offers a nuanced framework that restricts aggregation to the State-directed component of prevailing violence, thereby excluding intergroup violence, and further refines the concept of a “common enemy.” Specifically, Redaelli asserts that “what must be common must not only be the adversary, but the combined, state-directed pressure that makes ordinary policing untenable.” Crucially, Redaelli clarifies that aggregation functions solely as a method for evaluating intensity and cannot supplant the separate inquiry into organization; each group must satisfy the organizational requirement independently.
As Redaelli notes, this formulation of aggregated intensity represents a demanding test, rendering its application exceptional. The underlying logic seems to be that when multiple sufficiently organized criminal organizations cumulatively exhaust the law enforcement regime, the intensity threshold of NIAC should be considered satisfied. This argument is compelling, particularly regarding the cumulative exhaustion of the law enforcement regime: if the crux of the intensity equation is the operational capacity of the law enforcement regime, what matters is the overall disruption of that capacity rather than the number of actors causing it. The critical caveat, however, is whether these seemingly independent actors also satisfy the organizational requirement on an individual basis. It is here that at least two operational challenges may emerge.
First, although Redaelli correctly treats intensity and organization as strictly distinct, the two requirements are also deeply interrelated in practice. Existing case law reveals that the question of whether an armed group can be considered “organized” for the purposes of IHL is routinely determined by examining, inter alia, the armed group’s demonstrated ability to engage in sufficiently intense armed violence. Indeed, given the inherent challenges in assessing the internal structures and organizational capacity of armed groups, some commentators suggest the organizational requirement can be “inferred primarily from the group’s ability to engage in intense armed violence.” Without a parallel evaluation of the intensity of a specific criminal organization’s operations, determining if it satisfies the organizational requirement will likely prove even more challenging in situations where the intensity of violence is aggregated. Moreover, this approach still requires a case-by-case analysis of the organizational requirement, a process that some proponents of aggregated intensity argue is impractical.
Second, additional factual and operational challenges emerge when a criminal group directs violence against the State, thereby factually contributing to the cumulative exhaustion of law enforcement regime, although does not satisfy the organizational requirement. Pursuant to Redaelli’s analysis, violence from such groups must be excluded from aggregation. In practice, this produces a fragmented legal landscape where certain criminal organizations operating against a common enemy in the same theatre qualify as Parties to a NIAC, while adjacent unorganized groups do not. Although Redaelli correctly notes that this patchwork stems from the organizational requirement itself, it remains unclear how insufficiently organized violence can be factually and legally separated from the overall intensity and thus the cumulative exhaustion of the law enforcement regime.
These legal and operational challenges invite broader reflection on the overarching object and purpose of the aggregated intensity approach. While the book does not outright reject the aggregation of armed violence, it does restrict it to specific circumstances where a constellation of armed groups can constitute a single Party to a NIAC. In these scenarios the linchpin is “belonging” to a Party to a NIAC, and if and when determined, both the intensity and organizational requirements can be aggregated.
Classification of Battlefield Actors
Jelena Pejic combines a detailed thematic mapping with a systemic evaluation of Chapter 3’s framework for the classification of battlefield actors, offering incisive commentary on its legal and operational utility as well as factual and practical challenges. Pejic correctly highlights that the chapter does not provide a factual or evidentiary methodology for identifying operational level command and control on the battlefield, and thus by extension “belonging” to a Party to a NIAC. To address this, Pejic outlines several factors that may be relevant for assessing belonging, including joint planning structures, joint operations, shared communications, compliance with ceasefires, territorial divisions of responsibility, weapon flows, personnel transfers, and financial dependence. Additional factors may also be drawn from established indicators of organization, particularly those factors indicative of the existence of responsible command under IHL. In practical terms, identifying belonging on the battlefields of NIAC requires identifying factors indicative of operational level command and control between battlefield actors. Three issues to keep in mind. Assessing Battlefield Evidence of Belonging
First, while a range of ‘belonging indicators’ will be relevant for both state and non-state actors, their factual manifestation on the battlefield may differ. This is because command and control can manifests differently depending on actors. State forces almost exclusively use vertically integrated command and control structures, with authority and orders flowing from top to bottom. In contrast, non-State armed groups increasingly operate under decentralized or collective command and control systems, often using a combination of hierarchal, horizontal, and circular structures rather than strictly hierarchical ones. Irrespective of the blueprints, what matters is the existence of a de facto relationship between a Party to a NIAC and an armed group(s) where the former exercises a level of authority and control over the latter that is equivalent to operational level command and control in military doctrine.
Second, it is important to distinguish between conceptual or normative indicators of ‘belonging’ and the factual evidence used to establish them. For example, without controversy a unified chain of command would be an indicator of belonging but what remains to be determined is the battlefield evidence that indicates the existence of a unified chain of command. Such battlefield evidence may include, operational directives and procedures, unified internal codes of conduct and disciplinary system, standardized military uniforms and identification cards, or serial-numbered weaponry linked directly to individual personnel records. Identifying evidence of a unified chain of command may prove more straightforward for State armed forces than for non-State armed forces. For non-State forces, evidentiary proof will likely emerge incrementally through signals intelligence, or the recovery of unit rosters, payroll logs, equipment records, tasking sheets, or casualty lists from captured positions over the course of a NIAC.
Third, as Pejic notes, while some factors may clearly indicate belonging, others may be more of an indication of supporting a Party. The primary analytical challenge lays in evaluating scenarios where the evidence could support either conclusion. For example, joint operations, shared territorial zones, temporary personnel transfers, and joint fixed installations, may be indicative of ’belonging’ to or ‘supporting’ a Party and drawing a bright line distinction between the two may be difficult. Given the distinct legal consequences associated with each status, a totality-of-the-evidence approach is necessary to weigh the full spectrum of structural, operational, and material factors that indicate ‘belonging’ on the battlefields. Moreover, and similar to the indicators of intensity and organization, no single indicator would be sufficient on its own, cumulative assessment and corroborating evidence will likely prove influential if not determinative.
Personal Jurisdiction and the Ratione Personae Analysis
Katharine Fortin offers a critical analysis of the foundational premises and analytical frameworks informing Chapter 4, which examines the personal scope of IHL during NIAC. The framework rests on the distinction between active and passive ratione personae, delineated along the lines of bearers of obligations and beneficiaries of protection. Fortin questions whether this binary approach is legally accurate or helpful. As Fortin notes, the chapter highlights several exceptions to this binary alternative, whereby duty-bearers are simultaneously beneficiaries of protection, and vice versa. Within the conventional ratione personae architecture of CA 3 and AP II, the active and passive distinction provides a useful starting point for analysis of the personal scope of IHL during NIAC.
Chapter 4 builds on existing ratione personae analysis, and draws directly from the structural architecture of CA 3 and AP II, both of which are delineated along the lines of the armed forces of a Party to a NIAC, which are presumed to be directly participating in hostilities. The analysis looks at individuals who do not, or no longer, directly participate in hostilities, including civilians and individuals hors de combat. It addresses the concept of hors de combat, including its absence from the text of AP II, and its nuanced application during NIAC, its international criminal law implications, and its role in regulating intra-party violence and intra-party protection.
Ultimately, the focus must be on the question of civilian status for two key reasons. First, the concept plays a crucial role within IHL, particularly in relation to one of the “cardinal principles” of IHL, the principle of distinction. Second, conventional IHL does not provide a definition of “civilian” during NIAC, giving rise to significant debates regarding the identification of civilians during NIAC. As Fortin observes, the current legal categories of, and corresponding protections for, individuals during NIAC are wholly inadequate. Furthermore, the traditional scope of “civilian harm” in IHL is too narrow, and a deeper understanding of how civilians actually experience harm during NIAC is imperative for inclusive interpretations of existing IHL and the future development of IHL.
Further Questions of Geographical Scope of IHL during NIAC
René Provost examines the foundational premises, core paradigms, and analytical frameworks of the geographical scope of IHL during NIAC in Chapter 5. Provost situates the geographical application of IHL within the broader discourses of legal pluralism and legal geography, alongside the complex interplay of sovereignty, power, and State violence, both in the conceptualization of war and in the geographical pursuit of hostilities. As explained by Provost, these theoretical intersections demonstrate why arriving at a simple formula to determine IHL’s geographical scope during NIACs remains elusive. This difficulty is compounded by conventional IHL’s silence regarding its geographical scope during NIAC, as well as the advent of new military technologies on the battlefields of NIAC that enable parties to project force across unprecedented distances.
The few notable exceptions notwithstanding, a forensic examination IHL’s “DNA” reveals an absence of provisions designed to geographically contain the spread of armed conflict, including NIAC. On the contrary, the historical codification of the Law of Armed Conflict (LOAC) and IHL reflects an ever-expanding spatial reach that effectively shadows the spread of hostilities, from the provisions of the Lieber Code that apply to armies operating “in the field” and “along the front” to Additional Protocol I regulating attacks targeting military objectives situated within a “city, town, or village.” Contemporary scholars and practitioners echo this trajectory: the Tallinn Manual corroborates the applicability of IHL to military operations in cyberspace, as the Woomera Manual does for applicability of IHL to military operations in outer space. Just as IHL does not function to prohibit or end wars, nor was it designed to geographically restrict them.
Jus ad Bellum’s Independence from Jus in Bello
Provost also highlights the book’s observation of a paradox, which is that the jus ad bellum does not affect the applicability of the jus in bello. This conclusion rests upon the doctrine of the strict separation between the two legal regimes and the fundamental principle of the equal application of IHL. While the utility and limitations of this distinction remain subjects of debate, the doctrine of strict separation also contains the “parallel consequence that the application of IHL cannot legitimize any illegal use of force under jus ad bellum.” Indeed, the relationship between these two legal regimes was debated extensively during the drafting of AP I, culminating in its Preamble, which affirms that “nothing in this Protocol or in the Geneva Conventions of 12 August 1949 can be construed as legitimizing or authorizing any act of aggression or any other use of force inconsistent with the Charter of the United Nations.” This preambular recognition clarifies that jus in bello rules cannot be invoked to justify breaches of jus ad bellum or to circumvent its requirements. Furthermore, the explicit reference to the 1949 Geneva Conventions extends this logic to NIACs via CA3, at least for State parties, precluding IHL from serving as a “legitimizing blanket” for violations of other applicable bodies of international law.
The Temporal Scope of
Martha Bradley carefully engages with Chapter 6, which focuses on the temporal scope of IHL during NIAC, and especially if and how the terminal decline of a non-State Party to a NIAC can terminate the applicability of IHL. Through a detailed and compelling case study of the NIAC in Central African Republic (CAR), Bradley contrasts the functional approach advanced in Chapter 6 with the ICRC’s “lasting cessation” approach. She effectively maps the relevant criteria and argues that the ICRC’s approach sets the threshold for IHL’s termination excessively high. The cumulative requirements of “lasting cessation” and in particular the “no real risk of resumption” test will rarely be met in practice “where the root of the conflict is not sufficiently addressed, and the victims or parties are dissatisfied with how the conflict ended”
Determining the threshold for IHL’s termination is at least as complex as, if not more complex than, identifying its initial threshold of activation. Bradley’s forensic assessment of the factual conditions in CAR illustrates the utility of the functional approach proposed in Chapter 6. Because IHL comprises a range of obligations and protections designed for specific factual circumstances, the temporal scope of these rules inescapably varies. Consequently, the functional approach maintains that the hostilities regime and the protections regime must be evaluated independently when determining their respective threshold of termination.
Bradley’s application of this model provides an illustrative case where prevailing hostilities between opposing Parties have subdued to a degree that she argues warrants the termination of IHL. Importantly, Bradley distinguishes temporary lulls in hostilities and non-reciprocal armed violence directed exclusively against civilians from prevailing hostilities between the Parties. Just as unilateral violence against civilians cannot generate a NIAC ab initio, neither should it justify prolonging a NIAC in the absence of protected armed violence between opposing Parties. While the premature termination of IHL should be avoided, overdue termination is equally problematic. If residual, sporadic acts of violence can be adequately regulated by international human rights law (IHRL), the rules regulating the conduct of hostilities should cease to apply. Moreover, while the Appeals Chamber in Gotovina cautioned against a “revolving door” of alternating legal regimes, this dynamic is less detrimental than it appears. Instead, it also offers a practical mechanism that facilitates de-escalation by transitioning back to the law enforcement regime once armed violence falls below the activation threshold, while preserving the capacity to reactivate IHL should intense hostilities resume.
The absence of hostilities and the termination of the hostilities regime, however, should have no bearing on the applicability of the protections regime. Because protective rules serve distinct functions, their temporal scope of application diverges from conduct of hostilities rules. The key remaining question is whether the prevailing facts on the ground in CAR also warrant terminating the protections regime. This is a question of both fact and law. Specifically, when determining whether the factual conditions that gave rise to their activation no longer exist, also examine whether individuals will benefit from equal or more favorable legal protection under IHRL.
Conclusion
Ultimately, this book symposium demonstrates that the nuanced legal and factual challenges surrounding the identification of NIACs resist single or straightforward resolutions. Continued research is necessary to addresses these challenges inherent to the changing nature of NIAC. To this end, allow me to extend my sincere gratitude to the symposium’s contributors and hosting platforms for cultivating these rigorous and thought-provoking discussions.