Temporalities of Protection and the ‘End’ of Non-International Armed Conflict

About the author(s):

Nathan Derejko

Dr Nathan Derejko is a Lecturer in the School of Law and Human Rights Centre at the University of Essex. Previously, Nathan was Director of the Human Rights Centre Clinic at the University of Essex, Director of the Masters in Human Rights Programme at University College London, 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: International humanitarian law, international human rights law, and the use of force under the jus ad bellum. He has a particular interest in the applicability and application of human rights during armed conflict, counter-terrorism and human rights, and the law and practice of non-international armed conflicts. Nathan is also actively engaged outside of academia, and has previously worked with a number of human rights and humanitarian organisations in Canada, Europe, East Africa and the Middle East, and continues to deliver advanced training sessions for military personnel and government agencies, support strategic litigation efforts, and serves as an expert member for various NGO committees. Nathan is currently on the Defend Digital Me Advisory Council, an NGO focused on children’s rights to privacy and family life. Nathan Tweets at @NathanDerejko

The looming threat of a ‘forever war’, characterised by the so-called ‘Global Battlefield’ and the perpetual applicability of international humanitarian law (IHL), has thrust the question of when and how non-international armed conflicts (NIAC) end to the forefront of international concern and debate.

In both practical and legal terms, identifying the end of a NIAC is notoriously difficult. There are several reasons for this, but three in particular are important to highlight. The first is the complex spectrum of social, political and economic factors that underpin, propel, protract and ultimately bring NIACs to an end. Indeed, history is replete with NIACs spanning several years and, in some cases, several decades. The second is IHL’s virtual silence regarding its temporal scope of application during NIAC. While conventional IHL speaks of the ‘end of hostilities’ and the ‘end of the armed conflict’, it stops short of providing any guidance on the precise meaning and scope of these expressions, or the relationship between them. The third is the broad range of stakeholders that declare the end of a NIAC, including governments, military advisors, international organisations, courts and tribunals, humanitarian actors, and human rights defenders. The national interests or organisational mandates that motivate these stakeholders to declare the end of a NIAC invariably influence their determinations in accordance with the specific legal or policy consequences of IHL’s continuation or termination (p165).

In a recent article, A Forever War? Rethinking the Temporal Scope of Non-International Armed Conflict, I undertake a forensic examination of IHL’s threshold of termination during NIAC. The article first explores the temporal architecture of Common Article 3 (CA3) and Additional Protocol II (APII) to determine what, if anything, IHL has to say about its threshold of termination. From here, it critically examines two of the leading approaches for determining IHL’s threshold of termination during NIAC: the ‘peaceful settlement’ approach advanced in the jurisprudence of International Criminal Law (ICL) and the ‘lasting pacification’ approach advanced by the International Committee of the Red Cross (ICRC). While both of these approaches possess advantages and limitations, I argue that neither produce entirely satisfactory results for determining IHL’s threshold of termination during NIAC. For these reasons, the article develops and proposes an alternative ‘functional approach’, which addresses both the silence and shortcomings of existing law and doctrine, while upholding the very object and purpose of IHL during NIACs.

Rather than repeating this analysis, the purpose of this post is  further to explore the logic that underpins and informs the foundational argument in the article: the need to bifurcate IHL’s threshold of termination during NIAC in accordance with the object and purpose of IHL itself. 

Striking the Right Balance Between Premature and Overdue

The underlying challenge in determining IHL’s threshold of termination during NIAC is striking a harmonious balance between a premature and an overdue termination of IHL. In practice, this balance is often skewed by our understanding of how and when NIACs end. Most, but not all (para 70) theories on when and how NIACs end focus on whether hostilities have ceased altogether (or more)(p10), or have been sufficiently reduced in terms of frequency and intensity (somewhere)(p310) below the threshold of activation (p180). While there are good legal and logical reasons for this focus, the rules of IHL applicable during NIAC are not limited to regulating the conduct of hostilities between the Parties but are equally concerned with providing legal protection for individuals who do not, or no longer, participate in hostilities.

Broadly speaking, the framework of IHL applicable during NIACs can be divided between two general sets of rules: the ‘hostilities regime’ and the ‘protections regime’. The ‘hostilities regime’ refers to the IHL rules that regulate the conduct of hostilities between the Parties to a NIAC, the majority of which are found in customary IHL. The ‘protections regime’ refers to the conventional and customary rules of IHL applicable to individuals who do not or no longer directly participate in hostilities, including civilians and members of armed forces who have laid down their arms or are otherwise placed hors de combat. 

Given the distinct function each regime performs during NIAC, their temporal scope of application inescapably varies. For example, while the cessation of, or significant reduction in, hostilities certainly call into question the factual necessity for the rules regulating hostilities, they do not challenge the factual necessity of the protections for individuals deprived of their liberty, subject to criminal procedure, displaced for reasons related to the NIAC, or for individuals who find themselves living in territory under the control of armed groups. In other words, while the hostilities may have ended, the need for IHL’s protections may nonetheless remain.

It is within this operational context that striking a harmonious balance between a premature and overdue termination of IHL’s applicability during NIAC is challenging, if not impossible. For example, if IHL’s threshold of termination during NIAC is determined by an exclusive focus on the cessation of, or significant reduction in, hostilities, this would effectively disregard the temporalities of IHL’s protections and, as a result, (very) likely result in a premature termination of IHL’s protections for individuals who do not, or no longer, participate in hostilities. At the same time however, if IHL’s threshold of termination during NIAC is determined by focusing on the continued need for its protections, this may result in the continued applicability of the rules regulating the conduct of hostilities long after hostilities have been significantly reduced, or even ceased altogether. Considering that a number of IHL’s protections can remain applicable for months or even years after the cessation of hostilities, such  an approach renders IHL vulnerable to abuse by way of providing a thin veneer of legality for conduct that would otherwise be unlawful under the law enforcement regime.

Evidently, the real challenge is the prevailing ‘all or nothing approach’ to IHL’s threshold of termination during NIAC. Moreover, this ‘all or nothing approach’ seems to be the product of International Criminal Law (ICL) interpretations of IHL (para70), rather than a strict interpretation of IHL itself. Indeed, a number of the provisions of APII contain distinct temporal scopes of application that are entirely divorced from notions of ‘end of hostilities’ or ‘end of the conflict’, as is evident in the protections afforded to persons deprived of their liberty, or whose liberty has been restricted. In addition, a number of customary IHL obligations may only be activated following the cessation of hostilities, which in effect are the “post-conflict obligations” of IHL (p24), such as the obligation to identify, remove or neutralise landmines;  the various obligations with respect to the missing or dead; and the obligation to investigate and prosecute alleged IHL violations, and provide any necessary reparations for such violations. As a result, any endeavour to identify a single point in time to terminate the applicability of IHL in toto pays scant regard to the varying temporalities of IHL’s protections and invariablyresults in the over-extension of IHL to factual circumstances that no longer warrant its application, or by the premature termination of its applicability before comprehensive protection is restored under international human rights law. Considering the object and purpose of IHL during NIAC is regulating not only the conduct of hostilities, but also the the consequences of hostilities, a more nuanced approach is necessary.

A Functional Approach to IHL’s Threshold of Termination During NIAC

Given the distinct function the hostilities and protections regimes perform during NIAC, the factual necessity that underpins and informs their temporal scope of application will rarely, if ever, terminate simultaneously. For this reason, a bifurcated approach whereby the threshold of termination of the hostilities and protections regimes are determined separately, is both necessary and appropriate. This functional approach, considered as such, is both legally and operationally sound. Legally sound insofar that it reflects and respects the varying temporalities inherent to the protections of APII, and is predicated on the idea that the temporal scope of IHL must be interpreted in a manner that gives effect to its object and purpose. In operational terms, this functional approach equally guards against a potential premature or overdue termination of IHL during NIAC, and therefore also aligns with the object and purpose of IHL during NIAC.

So how does this work in practice? At its most basic, the temporal scope of the hostilities regime should be interpreted in a manner that gives effect to the function that it serves: regulating the conduct and consequences of hostilities during NIAC. As a general rule, if hostilities are ongoing, the rules designed to regulate hostilities are applicable. Importantly however, the termination of the hostilities regime should not require the complete cessation of hostilities but, rather, a sufficient and factual reduction in their intensity and frequency to the point where the residual armed violence can be factually described as isolated or sporadic, and therefore once again regulated by the law enforcement regime. This determination requires a careful and comprehensive assessment of the prevailing factual circumstances in order to distinguish between a mere lull in hostilities or the temporary suspension of hostilities, from a factual reduction in hostilities that warrants the termination of the hostilities regime. Importantly, the termination of the hostilities regime will have no effect on the applicability of the protections regime.

Along similar lines, the temporal scope of IHL’s protections regime must be construed in a manner that gives effect to the function that it serves during NIAC: to provide legal protection to the victims of NIAC. As a general rule, IHL protections remain applicable until the objective conditions that gave rise to their activation no longer exist, or until such time that individuals benefit from equal or more favourable legal protection under international human rights law. This approach creates a balance between the temporal scope of IHL protections, and the role and relevance of international human rights law during NIAC. This determination can only be made on a case-by-case basis, and through a careful assessment of the prevailing factual circumstances and reference to the specific protections under consideration. As a result, the temporal scope of the various protections during NIAC may in fact vary in accordance with the prevailing factual circumstances.

Conclusion

To sum up, what I have attempted to demonstrate is that neither the ‘end of hostilities’ nor the ‘end of NIAC’ will necessarily bring about the termination of IHL in toto during NIAC. This is because IHL is a pragmatic and functional legal regime comprised of a diverse range of obligations and protections applicable to specific factual circumstances. It is these specific factual circumstances, coupled with the object and purpose of the corresponding provisions of IHL, that will determine the temporal scope of IHL during NIAC. Any other approach would be contrary to the very object and purpose of IHL itself.

(Visited 6 times, 4 visits today)

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: