Revisiting Customary IHL Series – Getting it right? Challenges and Shortcomings in the Identification of Customary IHL

Editors’ note: This post is part of a series relating to the ICRC’s Customary International Humanitarian Law Study, featured across Articles of War and Völkerrechtsblog. The introductory post is available here. The series highlights presentations delivered at the young researchers’ workshop, Customary IHL: Revisiting the ICRC’s Study at 20, hosted by the Institute for International Peace and Security Law (University of Cologne) and the Institute for International Law of Peace and Armed Conflict (Ruhr University Bochum) on September 18-19, 2025. 

Most contemporary battlefields seem to be characterized by two sets of “facts.” First, the hostilities are usually accompanied by gross violations of both human rights law and the law of war (see here and here). Second, although the objective standard of protection is supposed to be found in international law, the scope of legal protections or the content of a rule benefitting those not taking part in the hostilities is heavily influenced by the narrative of the conflict parties. This is particularly true concerning custom, with its formation often lying under a “fog of law.”

While the International Committee of the Red Cross (ICRC) 2005 Study on Customary International Humanitarian Law is widely regarded a landmark in the identification of customary law, its approach was not free of criticism. It could be argued that, although the Study’s practice part is now updated online, it suffers from a classic dilemma of the law of armed conflict (LOAC): the determination of rules can only reflect issues of the past, while failing to properly address realities of active hostilities.

Academia largely relies on the authority of the Study; however, State responses have become increasingly sparse over the years. As Milanovic and Sivakumaran observe, the absence of any concerted pushback by States has led to increased acceptance of the Study. Nevertheless, a decrease in public dissent does not necessarily confer agreement with the Study’s conclusions and methods. In other words, the Study is not a “self-fulfilling” prophecy. Irrespective of how the first edition was received, this post regards matters from the opposite direction. It revisits how custom is shaped and considers potential paths towards an even more accurate assessment of how a renewed Study could assist in identifying customary rules.

Generally speaking, multiple challenges arise in the identification of LOAC State practice and opinio juris. From questions on how to handle asymmetric legal subjects, such as non-State actors, to new theaters of war and the use of modern technologies, uncertainties are plentiful. Customary law can only be robust enough to adapt to the challenges of modern conflicts if its identification is rebalanced to account for them. This post argues that the methodology for identifying customary humanitarian law requires both a deepening and a widening of the scope. It will do so by first setting out the role of custom and then argue for an approach to State practice that is closer to the actual ground where armed conflicts take place as well as the participants fighting in it.

Why Custom Carries Real Weight

Arguably, treaty law does not cover every battlefield or bind every belligerent. Customary international law as defined in Article 38(1)b of the International Court of Justice Statute (ICJ) fills critical gaps in international humanitarian law and is able to adapt to a multitude of conflict situations. While the Study was a remarkable collection of rules applicable in armed conflict at its time of drafting, its authority must still be questioned. “The Study says so” cannot be the end of analysis.

Much debate over an adequate approach to the identification of custom has been held in the past, culminating in proposals by the International Law Commission (ILC) in 2018. The ILC’s Draft Conclusions on Identification of Customary International Law restated the two-element test in its second conclusion: sufficiently widespread and representative practice; plus opinio juris. It places emphasis on what States actually do and whether to recognize a specific obligation or a right to act (see Commentary to the second conclusion). The ICJ—ambiguous as it has been in the past about concrete standards—has pressed in theory the same approach since North Sea Continental Shelf and Paramilitary Activities. It seems these criteria (State practice and opinio juris) must be applied with great rigor. Any claim about custom stands or falls on those elements.

And yet, while the basic framework might be clear, the weight that is put on the means of identification is not.

What the Study Did and the Core Critiques

The Study presents an excellent case to explore methods in identifying custom as it assessed a vast record of national manuals, legislation, judicial decisions, and statements in international fora. It remains a practical reference and provides a common language for judge advocates, investigators, and analysts.

From the beginning, the methodology applied within the Study drew sustained criticism, most prominently from the United States. In an early response, U.S. State and Defense Department lawyers praised the Study’s ambition but argued that some asserted rules rested on thin or misread practice, such as over-credited military manuals and UN resolutions, and treaty norms which slid into custom without enough practice or accompanying opinio juris. Although this critique has aged, it remains relevant to this day (see criticism on the ICRC’s recent Commentary to the Third Geneva Convention) and emblematic for the Study. In this sense, the ICRC’s use of supposed subsequent practice under Article 31 (3) b of the Vienna Convention on the Law of Treaties blurs the line between general practice and specific practice of a party to a treaty. Jean-Marie Henckaerts famously rebutted this critique, defending the use of manuals and warning against overextending doctrines like “specially affected States”.

The takeaway for users is straightforward. The Study is a starting point, not an evidentiary shortcut. For any specific rule, the custom case still needs to be made with clearly separated files for State practice and for opinio juris.

State Practice: Balancing Paper Policy with Practice Under Fire

The Study leans on sources that leave a paper trail: manuals; legislation; statements at the UN; and case law. While that approach is transparent and replicable, it risks skewing toward well-resourced States and underrepresenting operational behavior that shows what actually happens in combat. It also falls short of properly taking into account situations where access to information is limited or particularly biased. This is not to say that physical evidence was ignored. Yet, for rules that govern the conduct of hostilities (i.e., distinction, proportionality, feasible precautions, protection of works containing dangerous forces) operational evidence should carry significant weight.

While the use of open-source evidence for international custom has been sparse and largely limited to questions of evidence for international criminal investigations, technological advances may broaden the scope of options to assess battlefield practice. After-action reviews, incident assessments, and national inquiries (all of which are already relevant to domestic jurisprudence) reveal how commanders interpret and apply rules under pressure. Often handbooks or doctrine differ from practice (e.g., what is feasible in concrete battle damage assessments?). Placing emphasis on such evidence might portray States’ actions more accurately than the aforementioned documents could. While manuals and resolutions remain important on this view, they would function as corroboration rather than substitutes for evidence of custom.

While the Study gives great analytical context, the devil lies in the detail of the methodology for identifying custom. Weight might also be put on silence or the absence of protest. While this is a general matter of international Law, it does apply to the methodology of the Study as well. A prohibition often manifests through consistent non-use. For instance, few States would issue press releases celebrating that they did not target hospitals, while their sustained record of restraint across operations could be regarded as relevant practice. The challenge here would be to document silence with enough granularity to show a pattern. As the ILC in its Draft Conclusions has put it concerning silence as evidence of State practice, “The significance of a State’s failure to protest will depend upon all the circumstances, but may be particularly significant where concrete action has been taken, of which that State is aware and which has an immediate negative impact on its interests” (p. 128).

An additional blind spot is linguistic and regional coverage. At its outset, the Study consolidated rules from predominantly Western sources. If a methodology privileges anglophone materials, small States will be undercounted. The result is not only incomplete data; it is a perceived legitimacy problem when the loudest voices define the “global” picture.

Opinio Juris: Separating Law from Convenience

Practice alone does not make custom. The second element—acting out of a sense of legal obligation or legal right—is harder to demonstrate, especially when States prefer strategic ambiguity. While providing no absolute threshold for the distillation of opinio juris, the abstention from prohibited acts and the absence of protest do not always support the existence of a rule of customary law (as Dr. Henckaerts rightly notes). It is methodologically difficult to tell when a State’s restraint or facilitation evidence opinio juris rather than political feasibility, courtesy, or pressure. Without the use of words, it remains unclear whether States believe they are under an obligation (see e.g. Israel’s treatment of humanitarian aid to Gaza). The issue gains further relevance when observing rules that run parallel in treaty and custom, as we risk the standard falling short of meeting the protection needs arising from (especially non-international) conflicts. In many cases the customary rules exeed the treaty norm.

Here, the ILC’s Draft Conclusions also help in two ways, as they offer operational filters in insisting on explicit legal framing and ignoring regularities that look like convenience or diplomacy. In this sense, silence could first be regarded as evidence of opinio juris when the State is well positioned to react, and its inaction reasonably signals legal acceptance. Second, opinio juris has many evidentiary windows including: explanations of vote; official legal memoranda; national court decisions; and public legal positions given by ministries and militaries. Not all public statements are equal. Political postures often speak to policy preferences rather than legal obligation.

In this regard, classic guideposts remain relevant. The ICJ in North Sea Continental Shelf long cautioned against a too lenient approach and treating convenience as law (para. 77); it is also worth taking note here of the ICJ’s Paramilitary Activities decision, recognizing the fact that persistent legal framing across time helps crystallize custom (para. 186; paras. 188–90). The more States justify conduct with legal language—and accept constraints in hard cases—the stronger the evidence for opinio juris becomes. This is illustrated by the Gaza humanitarian-aid deliveries. Israel publicly frames facilitation as being conducted “in full accordance with international law,” with contemporaneous legal and diplomatic pressure pulling in the same direction (e.g. UNSC Res. 2720 (2023) or the ICJ’s provisional-measures of 28 March and 24 May 2024), while no such compliance can be observed on the ground.

Non-State Armed Groups: Invisible in Doctrine, Decisive in Reality

While the Study itself is primarily focused on the identification of custom as evidenced by States, it fails to recognize other actors beyond an indirect involvement (p. xlii and here).

Although acts of aggression by the Russian Federation, Rwanda, Azerbaijan, or arguably the United States are pertinent examples of a recent shift, most contemporary conflicts still are non-international in character. While there are differences on the concrete classification, over 120 armed conflicts are counted globally, involving more than 60 States and 120 non-State armed groups. Organized armed groups operating in areas of limited Statehood administer territory, detain individuals, or in other words, engage in State-like behavior. Some even tax the civilian population or operate with structures resembling administration. They may engage in hostilities against State and non-State belligerents alike. Their conduct shapes civilian risk and battlefield behavior like State conduct would.

Yet, traditional doctrine treats only States as “makers of international law,” meaning armed group practice does not “create” custom (a notable exception can be found in a dictum by the International Criminal Tribunal for the former Yugoslavia in its Tadić decision noting that the formation of the relevant customary rules can stem from the behavior of insurgents (paras. 103-08; for a substantial critique, see here)).

That formal division translates to a real protective gap. Additionally, in a theoretical sense, Professors Anthea Roberts and Sandesh Sivakumaran argue that non-State actors do engage in norm articulation. They develop codes of conduct and engage in unilateral commitments that influence compliance without conferring State-like status (“Law-Making by non-State actors”). In other words, their practice, covered by a legal commitment, matters. The ICRC’s decades-long dialogue with armed groups reflects the same reality. Their behavior decides whether rules will have an impact in non-international armed conflicts.

Non-State practice must be interpreted as counting for custom. Recognizing this would not necessarily require redefining sources of law. It would only require addressing their legal personality within the framework of LOAC (Widdig, p. 154 onwards). It could mean treating armed-group codes, training materials, and public commitments as interpretive context for how rules function in practice. Continuing to discard these materials will exclude the formation of legal protections in conflicts, where it matters most. Their practice must be considered according to their legal status. This is not new. From a perspective of legal subjectivity, they are in a minority. International law can react to these situations by acceping, as a fiction, that certain actors are in need of more flexibility in terms of legal standard (e.g. national liberation movements, stabilized de facto regimes, etc.)

The Risks of Standing Still and What It Means for the Study

In view of the above, a static approach to identifying custom risks deflating protective safeguards; therefore, a new study on custom might have to involve new methods. Otherwise three specific dangers come to mind. First, protections can erode by drift. Sparse public positions, combined with strategic ambiguity can be misread as acceptance of harmful conduct. Second, anglophone bias leads to skew. If methodology privileges well-documented major powers, “global” custom starts to look like the practice of a few capitals. Third, non-international armed conflicts remain under-served. The practice of organized armed groups retains relevance in most modern conflicts.

A stronger approach anchors each claim and shows the evidence for both. For conduct-of-hostilities rules, operational materials deserve pride of place: after-action reports; investigation findings; published targeting guidance; and domestic judgments. Manuals and UN resolutions remain useful as corroboration.

Capturing abstention matters, especially for prohibitions. Patterned non-use across comparable operations can be documented through incident databases, official inquiries, and parliamentary reports as well as open source intelligence. Mapping the field beyond major militaries is essential.

While the Study faces some challenges regarding the incorporation of modern evidence and a need of a more objective interpretation, it remains a well drafted baseline. The challenge is methodological renewal rather than replacement. This could include rebalancing the evidence toward operations, clarifying opinio juris with primary legal sources, and systematically including practice by smaller States. Bringing organized armed group conduct into view as interpretive context in non-international armed conflicts could serve to add legitimacy. Those adjustments could replenish the Study’s strengths while re-aligning it with the issues practitioners actually face.

No matter the challenges, customary international law will remain central to the corpus of humanitarian protection. Its credibility depends less on whether the rules exist on the books and more on whether our methods of identification of it capture the full reality of how conflicts are fought and justified today. Compliance requires, as a necessary condition, trust in a legitimate norm-making process that includes the key actors. Particularly in domains where the law remains unsettled or is still emerging, rigorous observance of the appropriate legal methodology and the inclusion of the relevant (legal) actors are indispensable.

***

Vincent Widdig is a lecturer in International Law and Research Associate at the Leibniz University in Hannover (Germany).

The views expressed are those of the author, and do not necessarily reflect the official position of the United States Military Academy, Department of the Army, or Department of Defense. 

Articles of War is a forum for professionals to share opinions and cultivate ideas. Articles of War does not screen articles to fit a particular editorial agenda, nor endorse or advocate material that is published. Authorship does not indicate affiliation with Articles of War, the Lieber Institute, or the United States Military Academy West Point.

Photo credit: Aleix Ventayol via Unsplash