Methodology of establishing custom

The standard definition of customary international law sees it as arising from the confluence of two elements, State practice and opinio juris – or, in the words of Article 38(1)(b) of the Statute of the International Court of Justice (ICJ), “evidence of a general practice accepted as law”. The existence and correctness of various methodologies for establishing custom have been a perennial topic in legal scholarship, from many different viewpoints, e.g. those challenging or defending positivist orthodoxy, or those engaging in normative theories as to what courts and other actors should be doing as opposed to descriptive theories as to what they actually are doing. That bastion of orthodoxy, the International Law Commission (ILC), has had a notable recent foray into this set of issues,Footnote ⁵ but again academics have discussed them endlessly.

On the descriptive front, which is of greater interest to us here, Stefan Talmon's analysis of the jurisprudence of the ICJ is instructive, showing that the Court takes three different approaches to the identification of a rule of customary international law: an inductive approach, a deductive approach, and assertion.Footnote ⁶ However, these differing approaches are by no means limited to the ICJ.

The inductive approach refers to a process in which “a general custom is derived from specific instances of State practice”, with opinio juris being a secondary consideration.Footnote ⁷ It is a process which goes “from the specific to the general”.Footnote ⁸ The inductive approach arguably fits best with the Article 38(1)(b) reference to “general practice accepted as law”. However, only rarely is the practice and opinio juris of all States, or almost all States, considered. Instead, a rule of customary international law is usually identified from the practice of a small group of States.

At other times, a rule of custom is not established through an inductive process but through a deductive one. The deductive approach refers to a “process that begins with general statements of rules rather than particular instances of practice”.Footnote ⁹ In this way, deduction “is a process of going from the general to the specific”.Footnote ¹⁰ The ICJ, for example, has described the “rule of state immunity” as “deriv[ing] from the principle of sovereign equality of States”.Footnote ¹¹

But, as Talmon has argued, “[t]he main method employed by the Court is not induction or deduction but assertion. In the large majority of cases, the Court does not offer any (inductive or deductive) reasoning but simply asserts the law as it sees fit.”Footnote ¹² Although Talmon makes the point with reference to the ICJ, this again is true also of other bodies, judicial or not.

The key issue for our purposes, however, is in the nature of the assertion – is it bare, or is it supported by citation to authority? A bare assertion, with no citation to some other source, can be made either with regard to a legal proposition that is so uncontroversial that no further discussion is really necessary (e.g. “treaties are binding only on their parties”), or to a more controversial proposition that the decision-maker seeks to establish by virtue solely of its own authority. A supported assertion, by contrast, seeks to both amplify the normativity of the claim being made by invoking the authority of some other source, and spare the decision-maker of the need to do the inductive or deductive work of establishing custom independently.

Thus, for example, when scholars or courts say that “it is a rule of customary international law that the conduct of State organs is attributable to a State”, they will normally not do any independent inductive or deductive analysis themselves. Instead, they will simply cite Article 4 of the ILC Articles on State Responsibility, relying on the ILC's authority for that proposition and effectively outsourcing the work of establishing custom to the ILC.

Or, to give a more elaborate example, consider the jurisprudence that determined that Article 3 common to the four Geneva Conventions of 1949 reflected customary international law. In Nicaragua, the ICJ found that common Article 3 reflected customary international law and was applicable in international armed conflicts and non-international armed conflicts alike.Footnote ¹³ The ICJ's holding on this matter consisted of a bare assertion – it set out no inductive or deductive examination, and cited no other authority.

In subsequent judgments, other international courts and tribunals have taken a similar position. But instead of undertaking their own analysis of whether the rules in common Article 3 reflect customary international law, or whether the scope of the customary rule extends to international armed conflicts, they have chosen to rely on the ICJ's finding in Nicaragua or on subsequent cases that themselves relied on Nicaragua. In the Tadić Decision on Interlocutory Appeal on Jurisdiction, the International Criminal Tribunal for the former Yugoslavia (ICTY) Appeals Chamber relied on Nicaragua for the proposition that common Article 3 has passed into customary international law.Footnote ¹⁴ The Trial Chamber in Tadić then relied on the holding of the Appeals Chamber.Footnote ¹⁵ The International Criminal Tribunal for Rwanda (ICTR) Appeals Chamber in the Akayesu case relied on the Tadić Decision on Interlocutory Appeal and the Tadić Trial Judgment to reach the same conclusion.Footnote ¹⁶ The Special Court for Sierra Leone (SCSL) relied on the decisions in Tadić and Akayesu as well as the ICTY Appeals Chamber judgment in the Delalic case.Footnote ¹⁷ All of these were assertions of the customary status and scope of the rule supported by citations to authority, ultimately leading down a chain of citations to Nicaragua.

There is nothing particularly objectionable about such reliance (at least if the cited authority actually stands for the proposition for which it is being cited). Indeed, it would be more surprising if each court undertook its own analysis from scratch rather than utilizing the holdings of its peers. This reliance on authority to establish custom might be total, i.e. without any additional investigation on the part of the court or tribunal in question, or partial, that is to say accompanied by some further investigation. Either way, the reliance on authority reflects considerations of both judicial economy and judicial comity. Judge Peter Tomka, then President of the ICJ, observed that:

We are particularly interested in precisely which bodies are relied upon for their holdings of customary international law. Judge Tomka refers to “States and bodies like the International Law Commission”. The ICJ has historically tended to cite other international courts less frequently than other international courts have cited the ICJ, preferring instead to rely on its own authority. This is probably partly due to the ICJ's self-regard as the apex court of the international system, partly to avoid criticisms that its sources are selective or biased, and partly due to tradition and institutional inertia.

The example we discussed above relating to common Article 3 concerns findings of international courts. In this article, we will explore how the ICRC's Study has been used by States, international and domestic courts and tribunals, and other relevant actors, i.e. how the Study is regarded by other influential actors within the system for the purpose of establishing a rule of customary IHL. In doing so we are not making any kind of normative claim that no other actors possess such influence, nor that the influence of all of these actors is equal – far from it. The sample of the real-world reliance on the Study that we have produced is inevitably limited. That said, our sense is that the sample is sufficiently representative of how the various actors in the international legal system perceive and use the Study so that we can draw reliable conclusions.

Degrees and kinds of authority

There are different senses to the word “authority”.Footnote ¹⁹ It can, for example, convey the general notion that a person can oblige others to do something, regardless of whether that course of action is right or wrong, i.e. the obligation exists independently of its content – the key question there being whether such a claim to authority can ever be legitimate and justified. It can also refer to an essential quality of any legal system, as most notably in the work of Joseph Raz, to the power of law to direct behaviour to the exclusion of other reasons.Footnote ²⁰ Also it can convey the idea that some persons are epistemic or theoretical authorities, in the sense that they should be trusted with certain matters because they possess knowledge and expertise about them which others do not – for example, that an electrician should be trusted on repairing a refrigerator.Footnote ²¹

To lawyers trained in the common law tradition, the word “authority” can also have the more mundane meaning of the sources which they rely on (and cite) in their briefs, judgments and other instruments. Law is an argumentative practice, but also an authoritative one, in which there is more reliance than in most other fields on the nature of the source of a proposition or an argument than on its content or correctness, which makes sense as to why common lawyers refer to their cited sources as “authorities”.Footnote ²² A standard distinction in that regard is between authorities that are binding (as the decision of a higher court would be for a lower court in a precedent-based jurisdiction) and those that are merely persuasive (such as the judgments of hierarchically equal courts or the work of academics). As Schauer explains, however, the notion of a “persuasive authority” is oxymoronic. If the reasons given by the source are persuasive, then authority has nothing to do with the process of persuasion. If the reasons given are not persuasive, however, yet the source is still relied on, even despite any substantive disagreement with its views, then it is content-independent authority and not persuasion that does all the work.Footnote ²³

Schauer thus correctly observes that the notion of “persuasive” authority really refers to the fact that reliance on that type of authority is entirely optional – the courts (or whoever else) can choose whom to cite and rely on.Footnote ²⁴ In addition, they can exercise this discretion for several purposes.Footnote ²⁵ First, this discretion can be exercised for genuine persuasion, where the source is cited because of the rigour and correctness of its reasoning. Thus, if a court conducted its own independent examination of whether a particular rule formed part of customary IHL, cited the ICRC Study in support, and in doing so looked in detail at the practice on which the Study based its conclusion about the specific rule, we could say that the court was genuinely persuaded by the Study. Second, this discretion can be exercised for deference to authority. If a court simply asserted that a rule was one of customary IHL and cited the Study to that effect, without actually independently verifying that the Study's conclusion was correct, that court would be deferring to the expertise, status and mandate of the ICRC, i.e. it would be treating it (and the Study as its product) as an authority. This is essentially no different from, say the two of us, as lawyers by training, choosing to believe in the existence of anthropogenic climate change by trusting the conclusions of climate scientists on this point, which we as non-experts have no way of independently verifying. Third, this discretion can be exercised as a reflection of the law's inherent conservativism, a denial of novelty, a signal that the court did not just make its conclusions up.Footnote ²⁶ While the first and second reasons for reliance on authority exclude each other, the third can co-exist with them in parallel.

In making these choices about citations, courts potentially enhance the persuasiveness and authority of their own decisions, as assessed by their primary audiences. However, through citation they also equally reaffirm the authority of the sources they approvingly rely on. Authority is reinforced through habit and repetition, through practice, in an “informal, evolving, and scalar process by which some sources become progressively more and more authoritative as they are increasingly used and accepted”.Footnote ²⁷ Thus, the more that international and domestic courts, and other various influential actors in the global legal system, treat an instrument such as the ICRC Study as authoritative, the greater its authority becomes, and the more likely it is for others to start regarding it as authoritative and cite it in their own decisions.Footnote ²⁸

Perhaps the best contemporary example of such a positive feedback loop is the ILC's Articles on State Responsibility. Upon their adoption in 2001, David Caron famously and correctly predicted a paradox between their form and authority.Footnote ²⁹ Despite the fact that the Articles are not in any way formally binding, and have by the ILC's own admission included a measure of progressive development, they have been cited and relied on by all international courts, including the ICJ.Footnote ³⁰ They are one of the ILC's most successful codification projects, and have succeeded in transforming how international lawyers think – and are trained to think – about concepts of State responsibility.Footnote ³¹ The success of the Articles is due to many factors, including the substantive need for such a product in this particular area, the nature of the ILC's codification process, the fact that the ILC is a State-empowered entity and that States had substantial input in their making.Footnote ³² Judge Tomka thus explained the ICJ's reliance on the ILC as follows:

Not all ILC products have, however, been able or will be able to trigger a positive feedback loop of authority.Footnote ³⁴ The question for us here is whether the ICRC's Study has been able or will be able to do so, bearing in mind that many of the same considerations that warranted the acceptance of the ILC's authority are relevant for the Study as well.

Success or failure in building authority

For the Study to successfully build authority in the international system, it needs to be useful within the parameters of that system. So, this is precisely what the Study sought to do, by filling gaps which as a purely pragmatic matter needed to be filled in contemporary IHL. Simply put, a codification project, i.e. one that is meant to restate existing law without precluding developments in that law, is more likely to succeed if influential actors within the system believe such a project to be necessary. How does the Study achieve this? First, this can be done by reducing largely unwritten custom to text, to rules that are more certain in their content, that can be interpreted and applied in the same way that a treaty rule can be. In doing so the Study inevitably contains a measure of progressive development (the progressiveness of which may well be contested), and reasonable people might disagree about how any given rule is drafted – this is just par for the course. In other words, when reducing customary rules to text, and doing so comprehensively in a sub-field of international law, it is impossible to completely keep separate the codification of existing rules from their development. Second, this can be done by making these rules universally applicable on account of their customary character, thus transcending the difficulties that some treaties that have universal or near-universal acceptance (such as the Geneva Conventions) do not comprehensively cover all of IHL (e.g. do not deal with the conduct of hostilities), whereas other treaties (such as the Additional Protocols) for various reasons have major gaps in their ratification – what Yoram Dinstein has called the “Great Schism”.Footnote ³⁵ Finally, the deepest of all gaps that the Study fills is the regulation of non-international armed conflicts, by identifying the vast majority of its rules as applicable to both types of armed conflict, even if many of them under treaty law apply only to international armed conflicts.Footnote ³⁶

However, in order to make these contributions – to truly succeed as a codification project – the Study's status must become elevated from that of a mere academic work to a higher degree of authority. Furthermore, that claim to authority is multifaceted. First, in a system, such as the international one, in which States are regarded as the primary lawmakers, the Study's authority is enhanced by its link to States. That link exists at a number of levels, including regarding the Study's inception. States have empowered the ICRC as an institution with authority regarding IHL generally. This is apparent from the Geneva Conventions as well as the Statutes of the Red Cross and Red Crescent.Footnote ³⁷ States also mandated the ICRC to conduct the Study project specifically. Recall that in 1995, the International Conference of the Red Cross and Red Crescent recommended that:

The recommendation was originally made by the Intergovernmental Group of Experts for the Protection of War Victims, which had met earlier that year.Footnote ³⁹ That Group, which consisted of “experts, representing 107 States and 28 governmental and non-governmental organizations”, had prepared a series of recommendations on enhancing respect for the law.Footnote ⁴⁰

The Study's second claim to authority is epistemic. It is authoritative because it is the ICRC, with its 150-plus years of expertise in IHL that stands behind it. The Study's claim to authority is based not only on the expertise of the ICRC itself, but also on the participation of many independent experts in the Study's process of preparation and review and on the exhaustiveness and rigour of that process. The ICRC also engaged in consultations with academic and governmental experts.Footnote ⁴¹

The Study's claim to epistemic authority is linked to persuasion. Through clear commentaries on the rules and the massive, simply unprecedented amount of practice assembled, the Study seeks to convince its readers that its conclusions are verifiable, and thus correct.Footnote ⁴² This is enhanced by the rigorous way in which the Study was carried out. The volume on practice, here both an exercise in legal rigour and a conscious effort to prospectively enhance the authority of the Study, is tangible evidence of its authors’ expertise and the amount of effort invested. The practice part of the Study is thus not only performative in the “we didn't make this up” sense; it is also a signal that the ICRC's expertise in this domain has few rivals. Indeed, had the rules part of the Study been published with the exact same content but without the practice part, the Study would inevitably have been greatly diminished in authority. This is true even if it turned out that few people today read the practice database and verify that the ICRC's conclusions are correct – its mere existence enhances the Study's authority.

Finally, the Study's authority rests on its subsequent approval by the influential actors of the international system, principal amongst which are States.Footnote ⁴³ How the Study is received can greatly affect its authority. It can enhance it or diminish it. The more the approval builds up, the more authoritative the Study becomes, and the more likely it becomes that others will see it as authoritative in turn. It is precisely these reactions to the Study that we wish to measure.

State reactions

The critical time point for the Study's authority was immediately upon its publication, but, as we have seen, authority builds up in an iterative process of long duration.Footnote ⁵² Had the Study immediately been met with a concerted negative response, its authority would have been nipped in the bud. Indeed, as the Study came out, it attracted criticism from a number of academics and government and military lawyers, mainly from powerful Western States, either writing officially or, more frequently, in an individual capacity. The criticisms of the Study tended to involve a mix of two types of arguments: first, that the ICRC's methodology was insufficiently rigorous; and second, that the ICRC got specific rules wrong.Footnote ⁵³ The most notable example – and a potentially mortal one for the Study's authority – was an official letter sent to the ICRC by John B. Bellinger III and William J. Haynes II, then the top lawyers in the US State Department and Defense Department, respectively, in which they conveyed the US government's “initial reactions” to the Study, accompanied by an extensive annex analysing four specific rules in detail.Footnote ⁵⁴

The US letter was respectful in tone but harsh in content. It reads like a very bad review, as if the Study was brimming with methodological and substantive flaws. The Study is criticized for frequently failing to rigorously assess State practice; for relying on inappropriate practice and generally giving excessive weight to the practice of non-State entities; for failing to pay due regard to the views of specially affected States (which the United States sees itself as being across the board on account of being involved in a great number of armed conflicts); for inappropriately conflating practice and opinio juris; and for oversimplifying “rules that are complex and nuanced”. Therefore, the United States considered that “the Study's methodological flaws undermine the ability of States to rely, without further independent analysis, on the rules the Study proposes”.Footnote ⁵⁵

This was, in a word, a total denial of the Study's authority. Curiously, however, the critique in the illustrative annex to the US letter, which addresses four rules in detail, is often reasonable but hardly devastating. Nor were the rules that the United States chose as examples (and, in particular, its disagreements with the Study's authors) genuinely pivotal to the structure of customary IHL as set out in the Study.Footnote ⁵⁶ They do not, for example, challenge the Study's main contributions, such as the generalizability of the conduct of hostilities rules or the applicability of most rules to non-international conflicts. The response also served as a placeholder, with the letter promising a follow-up by saying that its response was “initial” and that the United States will continue its review and “expect to provide additional comments or otherwise make our views known in due course”.

The authors of the Study of course felt a need to respond to the methodological criticism, themselves dealing with the four controversial rules in the annex only summarily.Footnote ⁵⁷ The promised US follow-up never came – at least not publicly – although the United States continues to critique the Study.Footnote ⁵⁸ Nor did other States, including the closest allies of the United States, make remotely similar comments in public, after the Study's publication or since. For example, an unnamed legal adviser of the UK Foreign and Commonwealth Office criticized aspects of the Study's methodology, and pointed to Rules 4 and 45 as examples of rules over which there were doubts. However, the reservations were a little over one page in length and rather general in nature. The statement concluded by noting that:

This was not, either in content or tone, the total attack on the Study's authority as in the US letter.Footnote ⁶⁰ For its part, Israel has been more critical of the Study, stating that:

Nonetheless, there has not been an accumulation of similarly adverse reactions by other States.

On the contrary, certain other States responded to the Study positively. Malaysia noted that “[p]raise was … due to ICRC for the publication of the study entitled Customary International Humanitarian Law.”Footnote ⁶² Sweden, on behalf of the Nordic countries, “welcomed the ICRC study and hoped that States would disseminate it as widely as possible”.Footnote ⁶³ Australia, on behalf of the CANZ group, noted that the Study “was already proving to be an important resource for States”.Footnote ⁶⁴ France opined that “[t]he comprehensive study by ICRC of customary international humanitarian law deserved careful study by Member States”,Footnote ⁶⁵ although on a later occasion it noted cautiously that while “the study constitutes a useful doctrinal work, it could not be used as such against States”.Footnote ⁶⁶ More neutrally, Tunisia indicated that it “was following with interest the debate inspired by the 2005 publication of the ICRC study”.Footnote ⁶⁷ And so, in the aftermath of the Study's publication the pushback against it, such as it was, does not seem to have continued, except sporadically and by a small number of States.Footnote ⁶⁸

Indeed, over the years, we have seen not a pushback but an embrace of the Study, at least on the part of several States. A number of States have cited the Study in a variety of public statements and documents. This includes Armenia,Footnote ⁶⁹ Azerbaijan, Belgium,Footnote ⁷⁰ the Democratic Republic of the Congo,Footnote ⁷¹ Germany,Footnote ⁷² Greece,Footnote ⁷³ Malaysia, the Netherlands, SwedenFootnote ⁷⁴ and Switzerland.Footnote ⁷⁵

Azerbaijan has referred to the Study extensively over the years in various letters to the UN Secretary-General and describes the Study as “authoritative”.Footnote ⁷⁶ Malaysia and the Netherlands have cited the rules numerous times and both have treated the rules akin to a binding instrument.Footnote ⁷⁷ Switzerland has stated that the Study, together with the Rome Statute, “provide indications of the current state of international humanitarian law”,Footnote ⁷⁸ and that:

For its part, in one document Israel has taken a more cautious tone, observing that, “[l]ike many other States, Israel does not agree that all of the ‘rules’ stated in the ICRC CIL Study reflect customary international law”,Footnote ⁸⁰ but Israel does refer to a few of the Study's rules and cites mainly the practice volume. Similarly, while the UK government expressed reservations about the Study, when it had to make formal submissions to the Baha Mousa Inquiry the lawyers representing the Ministry of Defence expressly noted that the government accepted that Rules 47, 87, 90, 91, 99, 118, 121, 122, 127, 128B and 142 as articulated in the Study did reflect customary IHL and relied on the Study as the primary authority for those propositions.Footnote ⁸¹

Other States, or parts thereof, refer to the Study in their military manuals. Colombia, Denmark and New Zealand cite the Study extensively in their manuals; and Spain mentions the Study.Footnote ⁸² There are some 242 references to the Study in the New Zealand manual and, in places, the manual treats the Study akin to a legislative text. The manual explains its use of the Study as follows:

Colombia also refers to the rules of the Study throughout but adds a disclaimer that citation does not constitute recognition of the customary rule.Footnote ⁸⁴ Along similar lines, Denmark refers to the Study extensively – with some 227 substantive references – and likewise approaches it in places like a legislative text. It prefaces its use with a one-page discussion of the Study. The manual recalls that the ICRC “worked for a decade to identify customary law in the field of IHL” (a reference to its rigour and epistemic authority), recounts the criticisms of the Study and concludes its discussion by noting that:

Other manuals demonstrate greater reservations regarding the Study's authority. For Germany, at the time of publication of its military manual (2013), “[i]t remains yet to be seen whether it [the Study] will come to be regarded as a reliable compilation of customary international humanitarian law.”Footnote ⁸⁶ The manual of Argentina, which was published after the publication of the Study, does not cite it. For its part, the US Department of Defense Law of War Manual does not cite the Study in its analysis of the various rules of the law of armed conflict, but cites it once at the very end of the Manual to reiterate US criticisms. It does, however, cite the US formal response to the Study at various other points in the Manual.Footnote ⁸⁷ The absence of citations to the Study is clear evidence of disapproval, in light of the specific context. It expresses the view of the Department of Defense, but it is difficult to assess to what extent that disapproval carries across the many layers of the vast US bureaucracy and armed forces. (This comment is of course valid mutatis mutandis for all States and their expressions of approval or disapproval of the Study.)

It is near-certain that the military and government lawyers of many more States use the Study in their day-to-day work behind closed doors. This has been our anecdotal experience from interacting regularly with such lawyers.

In sum, we can see that certain States have embraced the Study, treating it as an authoritative text and relying on it to a significant degree, others use it routinely in various contexts, while a few States are more ambivalent. The United States, or at least the Department of Defense, remains negatively disposed. It is clear, however, that there has not been any concerted pushback on the part of States generally – the Bellinger/Haynes letter did not generate a trend in that regard. While some States are not enthusiastic about the Study, there is no organized attempt to mobilize rejection of the Study among States. This inevitably leaves greater space for the reactions of other leading actors to affect the Study's authority. We turn first to international and then domestic courts.

International courts and tribunals

The Study has been cited by almost all major international and regional courts and tribunals. It has been cited in at least fifty judgments, eighteen separate opinions, and one arbitral award. In many of these outputs, the Study was cited multiple times – the sixty-nine decisions amount to 162 citation records in our spreadsheet.

As expected, the Study has been cited most frequently by the international criminal courts and tribunals. The ICTY has cited the Study in sixteen judgments and four separate opinions, the ICTR in one judgment, the MICT in one judgment, the ICC in six judgments and two separate opinions, the ECCC in two judgments, and the SCSL in four judgments. If we were to go beyond judgments, we would find that the Study has been cited on many more occasions, such as in decisions on interlocutory appealsFootnote ⁸⁸ and the confirmation of charges,Footnote ⁸⁹ including by the newest tribunal, the KSC.Footnote ⁹⁰

The Study has also been cited regularly by the regional human rights courts, except for the African Court which has simply not had the opportunity to pronounce on issues of customary IHL. It is cited in nine ECtHR judgments and seven separate opinions as well as in eleven IACtHR judgments and two separate opinions. The Study has thus been cited in more judgments of the ECtHR and IACtHR than the ICTR and ICC. Although perhaps surprising, the citation frequency can be explained by the number of cases involving an armed conflict that are brought before the regional human rights courts, in which they sometimes apply IHL while interpreting the human rights treaties over which they have jurisdiction, as well as the relatively few judgments (as distinct from decisions) delivered by the ICC. Insofar as the ICTR is concerned, relatively few instances of war crimes were adjudicated, the focus of that tribunal tending to be on genocide and crimes against humanity.

Moving beyond the international criminal tribunals and the regional human rights courts, the Study has been cited in three separate opinions to ICJ judgments, all by Judge Cançado Trindade, but not in any judgment of the Court itself.Footnote ⁹¹ It has been cited in one award of the Eritrea–Ethiopia Claims Commission. There were no citations in judgments of the STL, ITLOS, or the African Court of Human and Peoples’ Rights. This can most likely be explained by the subject matter and the small number of relevant judgments handed down by these bodies. We carefully considered such absences of citation to determine whether the relevant court is silently expressing disapproval of the Study by failing to cite it when such a citation could reasonably be expected, in particular when the court is relying on any rule of customary IHL. We could not find any such instances – for example, the ICJ cases in which Judge Cançado Trindade cited the Study, but the majority did not, were not really dealing with IHL in detail.

Of particular interest to us is how the Study has been used. In all but two instances, the tribunal that cited the Study either agreed with the Study or was neutral – to reiterate, of 162 total citations in sixty-nine decisions there are only two instances of disagreement. In seventy-five citations, the Study is the primary or sole authority for the proposition for which it is cited (most often, but not always, the content of a customary rule). In ninety-nine citations, the court is expressly relying on the Study to establish the existence and content of a customary rule or some broader normative proposition, without conducting any investigation of its own into the customary status of a rule, i.e. without independently evaluating State practice and opinio juris supporting that rule. These are the clearest cases of the Study being relied on as an authority. There are only nine examples where the court is conducting some kind of independent assessment, but most often this examination is cursory and relies on the practice assembled in the Study as a source. Some patterns clearly emerge, notably that over time the Study is being regarded as more authoritative. However, let us first address the two instances of disagreement with the Study's findings.

The first such instance is that of an award by the Eritrea–Ethiopia Claims Commission. The Commission found that:

In reaching this Conclusion, the Commission referred to the Study in a footnote:

As can be seen, the Commission goes out of its way to commend the Study, perhaps owing to its disagreement. The Commission notes the Study “with appreciation” and describes it as “exhaustive”. At the same time, the Commission states that it “need not, and does not, endorse the study's broader conclusion” on the issue in question. Whereas Rule 54 of the Study provides that “[a]ttacking, destroying, removing or rendering useless objects indispensable to the survival of the civilian population is prohibited”, the Commission limits the customary prohibition to attacks “for the specific purpose of denying them for their sustenance value to the civilian population or to the adverse Party”. In other words, the Commission regards the customary rule to be narrower than as defined in the Study, limited by the requirement of a specific purpose.

Of note, this instance of disagreement took place in 2005, the very year the Study was published and thus before it could accrue any greater authority over time. The disagreement is expressed without any contrary analysis of State practice or opinio juris, by way of assertion. The “does not” expression of disapproval is particularly curious when a “need not” would have sufficed. One possible explanation for the Commission's rather firm disagreement – but hardly a conclusive one – is that one of its members was George Aldrich, who wrote a very critical review of the Study that same year in the British Yearbook of International Law, in which he did expressly deal with Rule 54 of the Study and its allegedly incorrect encapsulation of custom.Footnote ⁹⁴

The second instance of disagreement is that of the ICTY Đorđević Trial Judgment.Footnote ⁹⁵ There the Trial Chamber “recalls the principle of international humanitarian law that in case of doubt whether a person is a civilian, that person shall be presumed to be a civilian”. The footnote to that sentence reads:

The Trial Chamber thus finds a rule of customary IHL that the Study did not endorse in either type of armed conflict, due to substantial controversy surrounding the presumption in Article 50(1) of Additional Protocol I. Furthermore, the Chamber does so in a purely assertive mode. There is no detailed examination of literature, State practice or the actual extent of the difference between its approach and the Study – the Chamber simply preferred to say that the treaty provision codified custom, relying on the Study even while gently disagreeing with it.

If we move beyond instances of agreement or disagreement and dig deeper, we find that different entities have used the Study in different ways. Consider the ICTY. In the period immediately after the Study was published (2005–2008), with some exceptions, the ICTY tended to cite the Study for its compilation of practice. It not infrequently referred to the military manuals and domestic legislation compiled in Volume II of the Study. For example:

The ICTY also cited the Study as an academic authority,Footnote ⁹⁷ and alongside teachings of publicists.Footnote ⁹⁸ Individual judges relied on the Study also for the methodology of determining customary international law.Footnote ⁹⁹

Over time, however, the approach of the ICTY to the Study changed. By 2011, the Study was being used in a more authoritative manner that transcends “mere” academic authority or a collection of practice. The ICTY was citing the Study for a variety of statements and propositions.Footnote ¹⁰⁰ Also it relied on the Study as the sole authority for the customary status of particular rules without any independent assessment. For example, the Popović appeal judgment reads:

As can be seen from the passage, the ICTY also cited the Study alongside treaty provisions, placing it at par with binding texts.

The evolution of the Tribunal's approach was not entirely linear. The ICTY did treat the Study with elevated authority even shortly after its publication. In the 2006 Hadžihasanović trial judgment, the Chamber observed that the Study was “considered an authoritative source on the subject”Footnote ¹⁰² and in the 2007 Martić case, the Trial Chamber cited the Study alongside treaty provisions and judicial decisions.Footnote ¹⁰³ Nonetheless, there is a noticeable shift in approach in the years immediately after the Study was published as compared with some years later.

For its part, the ICC has used the Study in different ways, sometimes in the course of the same judgment. At times, the Study is cited alongside treaty provisions,Footnote ¹⁰⁴ indicating that it is on a par with formally binding instruments, or it is cited alongside judicial decisions.Footnote ¹⁰⁵ Rules of the Study are also cited as accurate reflections of customary international law. For example, in the Ntaganda trial judgment, the Chamber stated that:

Likewise, in the Ntaganda appeal judgment, the Appeals Chamber stated that “[a] similar prohibition exists under customary law and is set out in rule 129(B) of the ICRC's compilation of customary rules of international humanitarian law.”Footnote ¹⁰⁷

On occasion, though, the Study is cited in a footnote alongside academic articles,Footnote ¹⁰⁸ suggesting that the Study is being treated as akin to teachings of publicists. Also, the language used when describing the Study can be more tentative, suggesting a lesser degree of authority. Thus, in the Ntaganda trial judgment, the Chamber stated that “It has also been considered as a rule of customary IHL, applicable in both international and non-international [armed conflict] by the ICRC: see Rule 15 of the ICRC Study on Customary IHL, and underlying practice.”Footnote ¹⁰⁹ Aside from these isolated instances, the Study is cited as authoritative and no independent analyses of its conclusions on customary international law are set out in the judgments.

The IACtHR uniformly affords the Study a high degree of authority. The way in which it is cited suggests that it is seen as comparable to a legislative text. For example, in the Case of the Massacres of El Mozote and Nearby Places v. El Salvador, the Court stated:

The Study is thus treated as an, even “the”, authoritative statement of customary IHL.

This approach is operationalized in the jurisprudence of the IACtHR. For example, throughout the judgment in the Case of the Santo Domingo Massacre v. Colombia, the Court refers to the Study as the sole authority for a variety of propositions of customary IHL and does not undertake its own independent analysis.Footnote ¹¹¹ The same is true of the Case of Vásquez Durand et al. v. Ecuador.Footnote ¹¹² In the Case of the Afro-descendant Communities Displaced from the Cacarica River Basin (Operation Genesis) v. Colombia, the Court states:

The “Rules” identified in the Study are treated as if they are rules of international law akin to treaty provisions.Footnote ¹¹⁴

The approach of the ECtHR is somewhat closer to that of the ICTY. In recent years, the ECtHR has treated the Study as highly authoritative. However, this was not always the case. In the period shortly after the Study was published, the Study was cited in a tentative manner. In the 2008 case of Korbely v. Hungary, the Grand Chamber noted that:

In the same paragraph, the Chamber notes that “the ICRC's study on customary international humanitarian law (2005) proposes the following rule …”.Footnote ¹¹⁶ The way in which the Study is cited suggests that the views are those of the ICRC and the Grand Chamber does not (necessarily) adopt them. In the 2010 Van Anraat v. Netherlands case, the Court cited the Study for its practice set out in Volume II.Footnote ¹¹⁷

By 2013, after some years had passed and there was time for the Study to become embedded, the ECtHR started to treat the Study in a more authoritative manner. In Janowiec and Others v. Russia, the Grand Chamber recounted that “[u]nder customary international humanitarian law, States have an obligation ‘to investigate war crimes allegedly committed by their nationals or armed forces, or on their territory, and, if appropriate, prosecute the suspects’”Footnote ¹¹⁸ and refers to the Study, the four Geneva Conventions, and various General Assembly resolutions in a footnote in support of that proposition.Footnote ¹¹⁹ In the 2015 Chiragov and Sargsyan cases, the Grand Chamber referred to Rule 132 of the Study as the sole authority in support of the proposition that:

The Rule is mentioned in brackets in the main text. In neither of those cases is there an independent assessment of the customary status of the rule.

By 2017, and similarly to the approach taken by the IACtHR, the Study was being cited as if it were a legislative text. For example, in the 2017 Tagayeva case, the Court observed:

This approach was subsequently followed in key Grand Chamber cases such as Georgia v. Russia (II) and Hanan v. Germany.Footnote ¹²² As before, the Study is the sole authority for the customary status of particular propositions and at no time is independent analysis set out in the judgment.

Domestic courts

Turning to domestic courts,Footnote ¹²³ the Study has been cited in at least 141 judgments. These are: nine judgments from courts in the UK, including two from the Supreme Court; seven US judgments, including one from the Supreme Court, as well as one concurring opinion; and seven Israeli Supreme Court judgments and one from the Military Court of Appeals. It has been cited in one Peruvian judgment, three Swedish judgments, three Dutch judgments, and six German judgments, including one from the Federal Court of Justice and two from the Federal Constitutional Court. It has been cited in at least twenty-six judgments of the Court of Bosnia and Herzegovina; and in seventy-eight Colombian judgments, including eight judgments of the Colombian Constitutional Court and sixty-nine judgments of the Jurisdiction for Peace of which thirty-eight were in separate opinions of Sandra Gamboa Rubiano. The sample is far from comprehensive and is heavily skewed towards those decisions contained in the databases we consulted. A substantial number of other citations would almost certainly be uncovered by a sufficiently large research team with appropriate legal and linguistic expertise.

There are no cases that we could find of outright disagreement with the Study. In one instance, a court expressed some doubt as to the customary status of Rule 99 of the Study, but did so only by way of assertion without examining any practice or opinio juris, and indicated that it did not need to decide the point.Footnote ¹²⁴

The courts in different domestic systems take different approaches to the authority of the Study. Judgments of the courts of England and Wales tend to treat the Study as a “mere” academic authority. For example, in the Court of Appeal judgment in the Serdar Mohammed case, the Study is discussed under the heading “Academic commentaries”,Footnote ¹²⁵ it is cited in part for the “dominant approach in the international humanitarian law literature”,Footnote ¹²⁶ and the introduction to the Study by President Jakob Kellenberger is quoted for the proposition that

The Serdar Mohammed litigation, in which most of these citations appear, is, however, a peculiar example because it examined the question whether IHL expressly authorized detention in non-international armed conflicts which is not, as such, directly dealt with in the Study but was the subject of academic inquiry.

The position of US courts is mixed. In Hamdan, the Supreme Court used the Study to interpret the phrase “regularly constituted court” in common Article 3,Footnote ¹²⁸ an approach that was followed by the Fourth Circuit Court of Appeals in Hamidullin.Footnote ¹²⁹ Indeed, in the latter case, the Court of Appeals noted that “[a]lthough non-binding, the ICRC's interpretation of the Geneva Conventions has been treated as persuasive by the Supreme Court.”Footnote ¹³⁰ The mainstreaming of the Study by the Supreme Court in Hamdan clearly enabled further citations by the lower courts. In US district courts, the Study has been used as sole authority for the customary status of particular rules.Footnote ¹³¹ The US Court of Military Commissions Review has used the Study in an explanatory sense, for example, as authority for the proposition that the conventional law of non-international armed conflict is rudimentary.Footnote ¹³²

By contrast, Israeli courts consistently treat the Study with greater authority. In Ahmed v. Prime Minister, the Israeli Supreme Court notes that:

The Study is treated as an authoritative statement of customary international law on point. Also, in Public Committee Against Torture in Israel v. Government of Israel et al., the Study is cited numerous times and at length.Footnote ¹³⁴ Again, the way in which it is cited is instructive. For example, the Court notes that:

Some UN commissions of inquiry (discussed further below), when citing the Study, refer to the favourable citation of the Study on the part of the High Court of Justice.Footnote ¹³⁶ This is evidence of the snowballing effect we discussed above.

Along similar lines, the State Court of Bosnia and Herzegovina treats the Study as highly authoritative.Footnote ¹³⁷ So do certain Swedish cases. It is worth quoting from Public Prosecutor (on behalf of Behram (Hussein) and ors) v. Arklöf (Jackie) at length. The Court observes:

As is evident from the passage, the rules are seen as a reflection of customary international law. The Court also refers to the collaboration with legal experts and to the Study's epistemic authority.

In sum, there is no doubt that the domestic courts we have surveyed treat the Study as an authoritative instrument, but they do so variably. None of them engage in any real independent evaluation of custom, but then again few of the domestic cases actually dealt with the customary status of a specific rule. Many of the citations of the Study are tangential or generally about what IHL requires. Citations of specific rules tend to be to those with which few, if any, would disagree.

UN commissions of inquiry

Citation of the Study is by no means limited to decisions of courts and tribunals. The Study has been cited by a variety of UN commissions of inquiry, a term we use to include commissions, panels of experts, and fact-finding missions. The Study has been cited by inter alia the Panel of Experts on Yemen, the International Commission of Inquiry on the Central African Republic, the Commission on Human Rights in South Sudan, the Independent International Commission of Inquiry on the Syrian Arab Republic, the Independent International Commission of Inquiry on the Protests in the Occupied Palestinian Territory, the Independent International Fact-Finding Mission on Myanmar, the Independent Commission of Inquiry established pursuant to Human Rights Council Resolution S-21/1, the International Commission of Inquiry on Libya, the Fact-Finding Mission on the Gaza Conflict, the Commission of Inquiry on Lebanon, the Panel of Experts on Accountability in Sri Lanka, the Panel of Experts on Yemen, and the Panel of Experts on Sudan.

Some commissions utilize the Study extensively.Footnote ¹³⁹ Commissions have variously described the Study as “authoritative”,Footnote ¹⁴⁰ “[o]ne repository of the principles of customary IHL”,Footnote ¹⁴¹ and “as indicative of the existence of customary norms”.Footnote ¹⁴² They treat the Study in a similar manner to the IACtHR. Particular rules of the Study are cited alongside treaty provisions,Footnote ¹⁴³ suggesting that they are at par with binding instruments; as sole authority for the statement that a particular proposition reflects customary international law;Footnote ¹⁴⁴ and the volumes as a whole are cited as reflective of customary IHL.Footnote ¹⁴⁵

Only rarely is there comment on the status of the Study. The Report of the Secretary-General's Panel of Experts on Accountability in Sri Lanka is exceptional in this regard. The report notes that:

International Law Commission

Engagement with the Study on the part of the ILC is mixed. The ILC does not refer to the Study as such in its commentaries to the draft conclusions on identification of customary international law.Footnote ¹⁴⁷ It does, however, refer to the Study in its commentaries to the Draft conclusions on subsequent agreements and subsequent practice in relation to the interpretation of treaties, the Draft articles on Prevention and Punishment of Crimes Against Humanity and the Draft principles on protection of the environment in relation to armed conflicts (the latter adopted on first reading).

In the commentaries to the draft conclusions on subsequent agreements and subsequent practice, the Study is used for its commentary to one rule and its compilation of practice.Footnote ¹⁴⁸ In the commentaries to the Draft articles on crimes against humanity, the ILC refers to a rule of the Study for a reflection of customary international law on a particular issue.Footnote ¹⁴⁹ In the commentaries to the Draft articles on protection of the environment, as adopted on first reading, the ILC refers to the Study extensively and uses it in a myriad of ways. Part of one draft article is said to be “based on the first paragraph of rule 43 of the ICRC Study”Footnote ¹⁵⁰ and, at times, the Study is used as the sole authority for the proposition that a particular rule is of customary status.Footnote ¹⁵¹ The Study is also used as an academic authority;Footnote ¹⁵² and for the practice it assembled.Footnote ¹⁵³ Overall, the Study is used circumspectly, with the ILC noting in numerous places that “[t]he ICRC study on customary law considers that this constitutes a rule under customary international law”,Footnote ¹⁵⁴ rather than “this is a rule of customary international law” with reference to the Study in a footnote. States at the Sixth Committee had also drawn attention to the Study when discussing the topic.Footnote ¹⁵⁵

The Study also features in reports of individual special rapporteurs;Footnote ¹⁵⁶ the approach they take to the Study is also mixed. The Special Rapporteur on formation and evidence of customary international law, Michael Wood, discusses the Study in his first report, but does so in neutral terms, whilst noting also the US and UK reactions to the Study.Footnote ¹⁵⁷ By contrast, the first Special Rapporteur on the protection of the environment in relation to armed conflicts, Marie Jacobsson, discusses the Study in more positive terms. She notes:

When the Special Rapporteur cites the Study for particular propositions, she too notes somewhat cautiously that the “ICRC considers that State practice establishes this rule as a norm of customary international law”.Footnote ¹⁵⁹

Other UN bodies

A variety of other UN bodies also cite the Study. These include the UN Secretary-General,Footnote ¹⁶⁰ special representatives of the UN Secretary-General,Footnote ¹⁶¹ special procedures of the UN Human Rights Council,Footnote ¹⁶² the Human Rights Council Advisory CommitteeFootnote ¹⁶³ and the Office of the United Nations High Commissioner for Human Rights (OHCHR).Footnote ¹⁶⁴ Almost without exception, they treat the Study as akin to legislative texts. They frequently cite the rules of the Study as sole authority for the customary status of a particular proposition, and they do not carry out any independent analysis. Of particular note, the UN Secretary-General has observed that the Study has “made a significant contribution to the process of identifying fundamental standards of humanity by clarifying, in particular, international humanitarian law rules applicable in non-international armed conflict”.Footnote ¹⁶⁵

Academics

We did not wish to conduct an extensive analysis of how the Study is being used by academics – our focus was on actors that are themselves regarded as more authoritative than “mere” scholars. However, the same pattern we have observed for other actors holds here too. The Study is frequently relied on by academics as the sole or primary authority for propositions of customary IHL, most often without independent analysis of State practice and opinio juris. Further, while in the years immediately following the Study's publication, scholarship citing the Study did so mainly for the purpose of providing critical evaluations thereof,Footnote ¹⁶⁶ few people have done so in the years since. Today the Study is simply being used routinely, as “the” standard reference work for the content of customary IHL. Academic use of the Study is on a general upward trend, as can be seen from the graph in Figure 1, peaking at 186 annual citations in 2019.Footnote ¹⁶⁷

Figure 1. Academic citations to the Study by year (Google Scholar).

Textbooks published in recent years also discuss and utilize the Study throughout their pages.Footnote ¹⁶⁸ The approach of textbooks is particularly important owing to their role in the education of future generations of international humanitarian lawyers. If future lawyers learn when studying the subject that the Study is authoritative or that it reflects customary IHL, they are more likely to adopt that same position when in practice.

Sassòli discusses the Study in a section on “customary law”. He notes that “[t]he ICRC Customary Law Study greatly facilitates the identification of official State practice and the resulting customary rules”Footnote ¹⁶⁹ and the Study is referenced in footnotes throughout the work. The Study is also discussed at length in educational works published by the ICRC written by Sassòli, Bouvier, Quintin and Grignon, as is the US response to the Study;Footnote ¹⁷⁰ Melzer notes that “[t]he ICRC's extensive study on customary IHL is also a widely recognized source of reference in this respect.”Footnote ¹⁷¹ Melzer also notes that “the ICRC's study as such is not binding. However, it carries the authority of an organization specifically mandated by the international community ‘to work for the understanding and dissemination of knowledge of international humanitarian law’”,Footnote ¹⁷² a reference to the epistemic authority of the Study. Kolb and Hyde note that guidance on the content of customary IHL “has to be sought” from the Study, because it “provides a thorough examination of the subject and sets out the norms, outside the universally accepted Geneva Convention of 1949, that can be considered to be part of custom”, and rely on the Study throughout their work.Footnote ¹⁷³ The Fleck volume includes “CIHL” in its list of abbreviations as shorthand for the Study and the section on sources contains a footnote which reads: “[s]ee generally on the subject of rules of international humanitarian law as customary international law, CIHL”.Footnote ¹⁷⁴ The Study is cited regularly throughout the work, for example, in the chapter on the law of non-international armed conflict. The Saul and Akande collection discusses the Study in the chapter on history and sources, which was written by one of the authors of the Study, Jean-Marie Henckaerts.Footnote ¹⁷⁵ The Study is also cited regularly throughout the volume, often as a reflection of the state of customary international law.Footnote ¹⁷⁶

Crawford and Pert are more cautious in their use of the Study. In their textbook, they discuss the Study in the section on custom, mention the critiques of the Study, and conclude that:

Similarly, Solis notes that the ICRC should be treated as a “respected corporate publicist” and that its Study “should not be overlooked”, and does in fact proceed to repeatedly cite the Study in his textbook while generally treating it like an academic work.Footnote ¹⁷⁸ For their part, Tsagourias and Morrison only cite the Study on two occasions, although they do so approvingly, without any discussion of its status.Footnote ¹⁷⁹ Dinstein mentions the Study once in passing in the main text, while noting US disapproval, and otherwise uses it in footnotes like any other academic work.Footnote ¹⁸⁰ Corn et al. similarly note US disapproval of the Study, in a work focused primarily on US practitioners.Footnote ¹⁸¹

Overall, there is clear acceptance of the value of the Study in the textbooks we surveyed, although the degree of authority attributed to the Study is variable. Some authors use the Study as if it were a binding instrument, similarly to the various courts we examined above; others clearly assign it weight over and above “mere” academic scholarship, while a smaller group treats the Study purely as a reference work. All in all, the Study appears to be substantially embedded into the instruction of IHL, which is likely to further enhance its authority as time goes by.