From treaty to custom: Shifting paths in the recent development of international humanitarian law

Abstract

From 1864 to the 1970s, international humanitarian law (IHL) changed through the path of formal treaty revision. Since 1977, however, purported changes to IHL have come not from treaty making but from interpretation, particularly through claims about the attainment of customary status by existing treaty rules. This article explains this shift as the result of the attitudes and choices of key IHL stakeholders under the changed conditions of post-Second World War multilateralism. It argues that the turn toward customary law claims-making was a reaction to the negotiation politics and contested outcomes of the 1977 Additional Protocols (APs) to the Geneva Conventions. After 1977, leading actors looked to custom as a means of arresting or encouraging legal change. The resulting, much-expanded IHL has proved influential and authoritative, even if its precise degree of acceptance by states remains unclear.

1. Introduction

Since its origins in the mid-nineteenth century, modern international humanitarian law (IHL)¹ changed over time in periodic episodes of state-sanctioned multilateral treatymaking.² However, after the adoption of two Additional Protocols (APs) to the 1949 Geneva Conventions more than four decades ago, states have shied away from new IHL treatymaking, with only one minor IHL instrument emerging: the Third AP (2005) establishing the ‘Red Crystal’ as an official emblem of the movement. A few prominent treaties have been created in areas related to IHL, including the Rome Statute of the International Criminal Court (ICC) and various weapons bans (landmines, cluster bombs, nuclear arms), but these are not normally taken as revisions to the core of IHL.

Despite stagnant treaty revisions, the development of IHL has not slowed down. Important recent changes to IHL have emerged through processes and practices of interpretation, specially interpretation via claims-making regarding the customary status attained by certain treaty rules.³ The extension of IHL through customary law claims-making is such that scholars consider it to have ‘revolutionized’ critical areas of the regime, notably the humanitarian regulation of non-international armed conflict (NIAC).⁴ According to Droege and Giorgiou, in the last 45 years IHL has undergone a ‘phenomenal transformation’ that has rendered it ‘literally unrecognizable’.⁵ This article takes up the question: What explains this change?

Based on archival and other primary records, secondary sources, and interviews, I explain this shift as the result of the purposeful attitudes and choices of key actors under the changed political and institutional conditions of post-Second World War multilateralism. More concretely, I argue that the turn to interpretation via customary law claims-making was largely a reaction to the negotiation politics and contested outcomes of the APs in 1974–1977. Soon after those negotiations, due to the perceived flaws of the APs, the perils of the universal multilateral process that produced them, and problems with their ratification, certain states and influential non-state actors looked to interpretation via customary law claims-making as a potential means of arresting or promoting legal change, respectively. Although initial state maneuverings regarding customary law in the 1980s were short-lived, non-state entrepreneurs began experimenting with it themselves in the 1990s, harnessing new institutions (international criminal tribunals) and traditional methods of normative development (expert consultations steered by the International Committee of the Red Cross, henceforth ICRC) in new ways, amid the discursive openings afforded by a renewed atrocity background. The resulting, much-expanded IHL has proved influential, even if its degree of acceptance by states is difficult to gauge and remains subject to debate.

I proceed as follows: Section 2 outlines the conventional path of change in IHL and examines the reasons for the eventual shift, particularly the transformative effects of decolonization upon traditional IHL-making. Section 3 argues that three influential IHL stakeholders (the United States; legal scholars Antonio Cassese and Theodor Meron; and the ICRC), all disappointed with the APs for different reasons, pivoted to the terrain of customary law as an alternative way to shape IHL’s evolution. Section 4 discusses how Cassese and Meron through their work as scholars and international judges, and the ICRC via its Customary Law Study, deepened that turn. In a fifth, concluding section, I reflect on the consequences of this shift, including states’ relative silence on the ICRC’s Customary Law Study.

2. The traditional pathway

From 1864 to 1977, IHL was normally created and revised via multilateral diplomatic conferences of government ‘plenipotentiaries’. States were the central protagonists in these processes through their power to negotiate, adopt, and ratify treaties, yet with some exceptions⁶ the process behind treaty revision always relied on the ICRC’s initiative, expertise, and organizational work. Time and again (in 1864, 1929, 1949, and the 1970s) the ICRC, together with Switzerland and scores of diplomats, practitioners, and scholars, promoted legal change by conducting extensive research, consultation, and drafting. Historically, IHL revision episodes did not simply aim at improving existing law in light of recent atrocity; they were also ‘forward-looking’, reflecting expectations of, and preparations for, future war-fighting.⁷

While the ICRC’s authority has long been and remains central to IHL’s development,⁸ transformations in interstate multilateralism have altered the law’s modal pathway of change since the 1970s.⁹ Multilateral processes of IHL treatymaking have always been microcosms of the broader international politics of their time,¹⁰ mirroring the complications of the international society that produced them. Until 1949 IHL had remained largely a Western-dominated affair,¹¹ with major powers such as Britain, France, and the United States leading diplomatic majorities and playing a gatekeeping role during codification.¹² However, as international society gradually transformed in the twentieth century, but especially with decolonization and the Cold War after 1945, IHL multilateralism too began experiencing gradual change sparked by the ideas, lived experiences, and growing diplomatic protagonism and voting power of dozens of decolonized (and Socialist) states.

Previously excluded from (or sidelined during) lawmaking, non-western state participants approached postwar IHL codification vigorously, promoting transformative ideas to redress its received biases and the violence it had historically enabled upon them.¹³ In a singularly noteworthy change, in 1949 Socialist and Latin American states made a key contribution in supporting an end to IHL’s silence on internal conflict, a legal gap that had long benefitted imperial powers by licensing atrocity in their colonial territories. Although the reasons behind Soviet support of this particular innovation were tied to Cold War politicking (as the USSR did not intend to accept IHL’s application within its own empire), by pushing European empires and other powerful sceptics to compromise and accept rules for internal conflict, Socialist and Latin American states helped to forge the ‘revolutionary’ Common Article 3 (CA3) to the Geneva Conventions.¹⁴

Despite there being other important Socialist and non-Western state contributions in 1949,¹⁵ more crucial for the eventual shift to customary law analysed in this article was the round of IHL revisions in the 1970s. At the 1974–1977 Diplomatic Conference for the Reaffirmation and Development of International Humanitarian Law (CDDH hereafter, for its French name) decolonization and Cold War politics combined decisively to endorse legal innovations deeply cherished by anti-colonial Third World-led negotiating majorities.¹⁶ Chief among such formal changes were the extension of IHL to ‘armed conflicts in which peoples are fighting against colonial domination and alien occupation and against racist regimes in the exercise of their right of self-determination’ and the affordance of POW protection to captured ‘freedom fighters’, both achieved through Additional Protocol I (API). These innovations legally equated wars of national liberation to interstate conflicts, legitimating the former through the application of IHL. African and Asian states, alongside a dozen liberation movements present at the CDDH, together fought to explicitly steer the law in an anti-colonial and anti-racist direction to at least partially level a legal playing field theretofore slanted to favour the West. In practical terms, these ideas stood to increase the odds of national liberation success by committing imperial powers to observe some war-making restraints, as well as judicial guarantees and humanitarian protections towards colonial rebels.

Beyond matters of national liberation, API also brought a host of other crucial novelties supported by African and Asian states.¹⁷ These included the detailed codification of rules to protect civilians and civilian objects from the dangers of combat,¹⁸ the criminalization of IHL violations, rules on mercenaries, and the prohibition of reprisals against civilians.

A second treaty instrument, Additional Protocol II (APII), was negotiated in 1974–1977 to expand the humanitarian regulation of NIAC contained in CA3. Although reducing atrocity across a range of situations of internal violence had been one central reason for the launch of the revisions’ episode, the rules ultimately enshrined in APII only partially satisfied humanitarians’ ambitions. On the one hand, APII augmented CA3 in core areas (it introduced basic rules of civilian protection and guarantees for victims of internal armed conflict and medical and religious personnel), yet those rules’ actual import was limited by design through the insertion of a high threshold of application.

Although formally both APs were adopted ‘by consensus’ at the close of the diplomatic conference in June 1977, the compromises they crystallized were in fact deeply contentious, the product of oft-elusive efforts at collective agreement in an increasingly politicized world.¹⁹ Of all, it was probably the group of Afro-Asian states that walked away most content from the CDDH, due to their successful (partial) decolonization of IHL through the revindication of liberation war, API’s intricate regulation of combat operations, and the safeguarding of their sovereignty in the context of NIAC.

3. Shifting the path of change: The APs and their discontents

The APs revisions process demonstrated how an altered balance of influence amid postwar universal multilateralism could impact IHL’s trajectory. In theoretical terms, after 1949 IHL treatymaking gradually underwent a process of ‘power-outcome decoupling’,²⁰ i.e., a growing mismatch between the views of the predominantly Western actors that had until then constructed a largely Eurocentric body of law, and the outcomes of universal lawmaking conferences operating under sovereign equality (one state, one vote) in a postcolonial international society. By the 1970s the ‘non-West’ had indeed become a central protagonist of IHL-lawmaking.

I argue that that the turn toward interpretation via customary law claims-making by three influential types of stakeholders was linked to their disappointment with the AP’s negotiation process and outcomes. That said, it is essential to clarify that the states and non-state actors analysed below had different reasons to be dissatisfied with the APs. As the next section explains, while major Western states denounced the national liberation-related innovations as ‘political’, and disliked some of the new combat rules for their alleged lack of realism, both scholar-jurists and the ICRC tended to lament the underwhelming expansion of IHL rules for NIAC via APII. This diversity of concerns shaped these actors’ subsequent approaches to customary law claims-making. What they all seemed to share, however, was an understanding of the risks and complications of traditional treaty-making as a means of legal change.

3.1 Major Western states

In the 1970s as in 1949, the United States, the UK, and France again constituted the backbone of the Western state group involved in the ‘reaffirmation and development’ of IHL. Initially unenthusiastic about entering a new revisions episode at all, these states worked strenuously for years to ensure that the APs met their preferences across different areas.²¹ While by no means wholly unsuccessful across the range of issues considered in negotiations, the Third World’s successful legitimation of national liberation war and freedom fighters through API certainly represented the most politically salient battle for the conservative Western gatekeepers.²² Although desirable, fair, and necessary from the perspective of a decolonized world long subjected to imperial atrocity,²³ major Western states viewed these changes as noxious (introducing allegedly-dangerous notions of ‘just war’ into IHL, hence threatening to foment violence),²⁴ disingenuous (because they did not expect liberation movements to be willing or able to respect IHL), of little humanitarian use (because they would make the law applicable to a limited, and purportedly waning form of conflict), and politically insulting (by bringing pointed opprobrium on imperialism, racism, occupation).

National liberation-related features were not API’s sole ‘defects’ in Western eyes, however. Additional changes, especially the prohibition of reprisals against civilians, the regulation of armed attack to protect installations containing ‘dangerous forces’ (such as dams, dykes, and nuclear plants) and the environment, and potential IHL limitations on the use of nuclear weapons, all left Western states concerned with the new law and its production process.²⁵

Archival evidence supports this argument. In the case of the American delegation, while the initially striking a positive note (‘we succeeded beyond our expectations in codifying and developing the law’), the US post-negotiation report noted having ‘considerable misgivings’ regarding the revisions exercise, viewing it from the start as ‘more of a hazard than an opportunity’, hence with ‘caution and concern’. Why?

We had seen in other contexts the risk that conferences of one hundred or more countries would be dominated by a majority of developing countries, a majority that all too often seems to be led by radical States bearing grudges against the wealthy countries in general and against the United States in particular. These concerns were in fact justified …²⁶

Due to these difficulties, American delegates concluded that it was:

most important … in any broad multilateral conferences of the future to be in such a posture that we can get along without an agreement, to make this relative independence clear and credible, and to work closely with the other developed countries to ensure a cohesive approach.²⁷

Like their American colleagues, British delegates admitted they had approached IHL revisions in the 1970s:

with mixed feelings … the UK feared that [the conference] would be likely to … weight the scales in favour of the causes currently popular with the non-Western majority at the United Nations. Nevertheless we hoped that real humanitarian advances could be made.

Like US diplomats, British officials lamented that ‘both these hopes and these fears we realised’.²⁸ Their final confidential report explained that, on the one hand:

the two Protocols do mark … an important humanitarian advance in many fields, especially … the protection of the civilian population. On the other hand, the principle of equality of application has been breached, most notably through in the reclassification of certain types of national liberation conflicts as being international in character and in the treatment of captured irregular forces … the problem is whether the aim of protection of enemy civilians and civilian objects and the consequent restrictions on the freedom of military action has been pursued further than realism would dictate.²⁹

The French delegation’s report was more blunt and bitter: ‘After four years of dubious combat, the Conference ended with pressured consensus: humanitarian texts masking deep disagreement between participating states regarding the value and the content itself of the adopted law.’³⁰ French delegates denounced API’s ‘multiple, detailed obligations, albeit formulated in ambiguous and confusing ways’ and chastised APII which became a ‘lapidary expression of some general principles curiously placed alongside rare, more elaborate articles, [which are] testament of an older text whose core had disappeared’. ‘The reasons for this’, they explained, ‘are political’.³¹

As some of the above evidence suggests, Western gatekeeper states’ disappointment with universal multilateralism since the mid-1970s was not limited to the IHL or the APs’ experience; the US infamously declined to sign the UN Convention on the Law of the Sea in 1982, withdrew from UNESCO in 1985, rejected the International Court of Justice’s Nicaragua decision in 1986, and denounced the UNGA as hopelessly biased against US interests.

To be clear, I do not claim that the APs did not achieve important change, that they were not influential, or that they were broadly rejected or ignored by Western states. The opposite is probably true.³² Archival evidence suggests that after a few years of domestic (and NATO) legal and military analysis, British concerns with the APs were successfully met through a combination of understandings and declarations. France found APII acceptable enough to ratify it 1984, but did not warm up to API until 2001, once its decolonization and defence woes had relaxed.

My precise argument here is simply that the experience of negotiating the APs, and aspects of their contents, turned former gatekeeper Western states off the traditional path of change – multilateral treatymaking – and encouraged them to use alternative ways to shape the IHL’s future, particularly interpretation via customary law claims-making.

The US became particularly recalcitrant. Not content with (rather boisterously) refusing to ratify API,³³ in the mid-1980s American officials conducted private diplomacy to convince the UK, France, and West Germany³⁴ to follow suit, pressing them instead to endorse a bespoke declaration, to include select principles from API (redrafted in language acceptable to the Reagan administration) and exclude, among others, the provisions on national liberation, the protections of installations containing dangerous forces, and the prohibition of reprisals against civilians.³⁵

The purpose of the American-designed document seemed threefold: relieving external pressures to ratify API, generating a consensus among core NATO states on desirable rules for future military interoperability, and – more to the point here – shaping international debate about the elements of API that might eventually attain the status of customary law, blunting some of the Third World coalition’s key innovations, especially regarding national liberation. Conversely, the US actively promoted the ratification of APII, which it considered to be a step forward in the humanization of internal conflicts hampered by a restrictive scope of application.³⁶

American resort to customary claims-making about IHL thus had both ‘positive’ and ‘negative’ sides.³⁷ On the positive side, American officials wished to further the idea that most of APII could and eventually should attain the level of customary law.³⁸ Meanwhile, their ‘negative’ agenda involved preventing the rules on national liberation war and freedom fighters from attaining customary status, as well as shaping future customary law-ascertainment regarding the articles protecting civilian persons and objects.³⁹

A British telegram summarizing a 1986 meeting held with the US State Department to discuss the American declaration of principles underscored US intentions ‘undoubtedly to use the present opportunity to warn the world at large to avoid using their voting strength in multilateral negotiation to override the wishes of the major powers’.⁴⁰ British officials further reasoned that the American declaration was also linked to the International Court of Justice (ICJ) adverse judgment regarding US involvement in Nicaragua, presumably viewing the ICJ’s decision as evidence of the Court’s anti-US politicization, hence a potential tendency to foment undesirable customary law formation.⁴¹

Although it failed to persuade their European peers to endorse their bespoke declaration, the US government nevertheless made public its contents in 1987 at an academic workshop specifically organized to elucidate the parts of the APs that could be considered customary law.⁴² Ironically, although the participation of high-level American officials at this event was a deliberate effort to shape API’s process of customary law-attainment via state opinio juris, it was other actors – not obviously beholden to the military interests of the US or NATO – who would take up the baton and run far with it.⁴³

3.2 Non-state entrepreneurs

The APs’ negotiation also left certain non-state legal entrepreneurs dissatisfied. Two international legal scholars who in the 1970s had served as advisors to their home governments during the APs process, Antonio Cassese and Theodor Meron, afterwards became active proponents of informal tactics for IHL development, eventually coalescing around the use of interpretation to advance the law, including customary law claims-making.⁴⁴

Writing as early as 1979 in a volume assessing the APs, Cassese offered the following recommendation:

I submit that those who have the lot of humanitarian law at heart should not overemphasize the deficiencies and pitfalls of the Protocols. Stressing the (inevitable) demerits and loopholes of these international instruments can only lead to increased skepticism about international humanitarian law … I therefore believe that scholars … should do their utmost to strengthen the possible role of these momentous treaties … Legal scholars can serve a useful purpose in their expert capacity as well, by propounding interpretations of the Protocols that aim at emphasizing the humanitarian purpose of their rules. As neither international law in general nor the Protocols themselves entrust anybody with the task of giving authoritative interpretations of their provisions, there is much room in this area for forward-looking jurists. The Protocols offer much space for interpretation … many rules … are therefore open to divergent interpretations.⁴⁵

For his part, in 1995 Meron claimed:

International lawmaking and various diplomatic conferences, for example, the conference that adopted the Additional Protocols to the Geneva Conventions in 1977, have, on the whole, been unsympathetic toward extending the protective rules applicable to international wars to civil wars, an attitude that has dampened prospects for redress through orderly treaty-making. Because conferences often make decisions by consensus and try to fashion generally acceptable texts, even a few recalcitrant governments may prevent the adoption of more enlightened provisions.⁴⁶

This analysis was not new for Meron. Concerned with situations of internal violence falling below APII’s scope of application, in a 1983 article he had already proposed:

a short, simple, and modest instrument to state an irreducible and non-derogable core of human rights that must be applied at a minimum in situations of internal strife and violence (even of low intensity) that are akin to armed conflicts.⁴⁷

Meron envisioned, as a first step, a ‘solemn declaration, which would not require formal accession or ratification by states’.⁴⁸ This ‘solemn declaration’ idea echoed an initiative made by the ICRC in 1972 after states flatly rejected plans for binding rules applicable to internal troubles and disturbances.⁴⁹ Meron’s project produced a document endorsed by distinguished experts in Turku (Finland) in 1990, but which ultimately failed to secure formal state approval at the UN.⁵⁰

Meron had also participated in the 1987 Washington workshop alluded to earlier (where American officials publicized their government’s views regarding the APs and customary law). His remarks there revealed an understanding of the importance of the customary path for IHL’s development through the identification of state practice and opinio juris together.⁵¹ Finding fault with the ICJ’s bare assertion regarding the customary status of CA3 in their Nicaragua decision ‘for the virtual absence of discussion of the evidence and reasons supporting this conclusion’, as well as the strategy of deeming certain rules to be customary law ‘without judicious attention to counter-practices’, in the end he acknowledged that this ‘method cannot but influence future consideration of customary law in various fields, including the Geneva Conventions’. He added:

Despite perplexity over the reasoning and, at times, the conclusions of a tribunal, states and scholarly opinion in general will probably accept judicial decisions confirming the customary law character of some of the provisions of the Geneva Conventions as statements of the law. Eventually, the focus of attention will shift from the inquiry into whether certain provisions reflect customary law to the judicial decisions establishing that status.⁵²

Meron was prescient: the practice of customary law claims-making endured, sometimes by assertion, others by more careful demonstration. He himself embraced the customary law path in short order through a study of the complementarity between human rights and humanitarian law.⁵³

Cassese also developed his approach to IHL advancement via customary law claims-making through his scholarship since the early 1980s.⁵⁴ In 1984, he authored a law review article offering a comprehensive survey of the IHL rules that could be said to have reached customary status, including some from the APs.⁵⁵ Like Meron’s, Cassese’s approach was academic, so he was careful to draw fine distinctions between rules from the 1949 Geneva Conventions and those from the APs. It was clear, however, that his tactic aimed at highlighting the broad importance of customary IHL as a means to augment the law’s reach.

Until the late 1980s the ideas of Cassese and Meron remained scholarly opinions aired out in academic conferences, journals, and books. Similarly, the American bespoke declaration of API principles (and the ICJ’s opinion on CA3) appeared unique. It remained unclear whether and how the use of customary law claims-making would actually make a dent and change IHL. Ultimately, it was its adoption by the ICRC that wound up cementing its influence.

3.3 The ICRC

The ICRC (and the Swiss government, as the host of the CCDH) also found the APs’ negotiation process disappointing. They originally envisioned the revisions episode to last at most two sessions; negotiations, however, extended to four meetings (six if you count two meetings on conventional weapons), already on the heels of years of research, consultations, resources, and painstaking diplomacy. Archival evidence shows that, dreading a possible fifth session, in 1976 the Swiss government and the ICRC actively liaised behind-the-scenes with key delegations across political groupings to help broker compromise and limit the risk of prolongation.⁵⁶

Although the ICRC’s private balance-sheet after the CDDH refrained from overdramatizing the AP’s flaws, instead framing them as the ‘maximum attainable’ at the time, its overall tone was indicative of some measure of regret. ‘An inventory of negative points … is very difficult to establish, because although it is rather clear that the ideal was not attained from a humanitarian viewpoint, it is very difficult to set the contours of such an ideal …’⁵⁷

Nevertheless, the ICRC did recognize concrete issues with the APs in form and content. On form, the texts appeared to it regrettably ‘heavy’ and difficult, precisely due to the complex compromises struck in conference, often resolved through complicated and sometimes ambiguous formulations. Regarding API’s specific contents, however, it lamented only the formulation of a handful of rules, among which figured the absence of a complete prohibition of attacks on installations containing dangerous forces (which had been restricted to the chagrin of NATO states but not fully banned as the ICRC wished), the requirement of state consent to allow for the provision of humanitarian relief, and the deletion of a provision foreclosing the filing of reservations on certain ‘fundamental’ provisions of the APs. Concerning APII, the ICRC regretted the last-minute deletion of multiple articles from the working draft (the result of a behind-the-scenes compromise between representatives of the Third World and the West), as well as the requirement of state consent for providing humanitarian relief in rebel-held territory.

Importantly, the ICRC did not lament the extension of IHL to national liberation conflicts via API because:

[it demonstrated the] concern of Third World countries to build a humanitarian law that takes into account their specific problems. For these countries, liberation from colonialism and racism is a fundamental humanitarian principle and the central importance they give to conflicts against “colonial” or “racist” regimes cannot be ignored. To have taken it into consideration in Protocol I is the best way to show them that humanitarian law is now theirs and no longer just a body of law drawn up by “Westerners” according to “Western” norms. On substance, we cannot pretend that [this extension] is regressive, since it introduces a broader application of the law. We can thus only be reticent regarding the manner and form in which this expansion was formulated.⁵⁸

Overall, a reading of the ICRC’s private post-conference bilan suggests that the organization found three broad aspects of the APs disappointing. First, the confusing formulation of many of the rules. Second, the introduction of conditions and loopholes in areas as important as the protection of civilians and the provision of humanitarian relief. Third – and in agreement both with scholar-jurists and the US – APII’s limited ‘humanization’ of NIAC, which was largely chalked up to the Third World’s attachment to their national sovereignty and their ‘tendency to ardently, if not blindly, defend it’. Indeed, the ICRC quipped that ‘generally … Protocol II allows us to gauge, in the contemporary international context, the extent to which states are willing to sacrifice to humanitarianism their absolute right to do as they please within their territory’.⁵⁹

These issues notwithstanding, once the CDDH wrapped in June 1977 the ICRC turned decisively to the diffusion of the APs and the promotion of their ratification, allocating this task to a specific, dedicated official (Hans-Peter Gasser). However, despite Gasser’s efforts, those of the Swiss depositary, or of the UN General Assembly via multiple resolutions, the process of state ratification of the APs moved slowly. By the end of 1982, only 27 states had ratified API, and 23 APII.⁶⁰ Major Western states, including the US, Britain, and France wavered further throughout the 1980s, and with the American decision against API’s ratification in 1987 and non-ratification by major non-Western states including Afghanistan, Iran, Iraq, India, Israel, Pakistan, and Turkey, the goal of reaching universal commitment with the revamped IHL seemed to falter.

I draw a direct connection here between three broad factors (disappointment with aspects of the negotiated APs, a recognition of the complications of treatymaking amidst intense international politicization, and concern with the instruments’ limping ratification process) and the ICRC’s eventual endorsement of customary law to strengthen IHL. Sharing similar goals, the context of the early 1990s provided both the ICRC and scholar-jurists entrepreneurs with a political opening to achieve them.

4. Enacting the shift towards customary IHL

4.1. Cassese and Meron: Scholars-cum-judges

The atrocity context of the early 1990s resurfaced enduring challenges regarding respect for and application of IHL, this time amid multiple conflicts occurring in the Balkans.⁶¹ Debates about the ‘crisis’ of the law rose anew, alongside a revitalized role for – and disappointment with – UN peacekeeping in Somalia, and later in Rwanda. Legal entrepreneurs, international organizations, and even the ‘P-5’ states within the UN Security Council now underscored a need to pursue criminal accountability as a means of enforcement, paving the way for institutional innovation via the creation of ad hoc tribunals, and soon after, the permanent ICC. Cassese and Meron were at this stage no longer just influential international legal scholars in their own right, but occupants of high-ranking positions in core international criminal tribunals, which they began to harness as mechanisms to develop IHL.⁶²

The UN Security Council (UNSC) took the first steps, however. In 1993 the Council unanimously adopted Resolution 827 establishing an International Criminal Tribunal for the former Yugoslavia (ICTY), the first such institution since the Second World War. The UNSC tasked the ICTY with prosecuting four types of offenses: (i) Grave breaches of the 1949 Geneva Conventions; (ii) violations of the laws and customs of war; (iii) genocide; and (iv) crimes against humanity. In the Spring of 1994 the Council agreed to create a second tribunal to deal with the atrocities committed in Rwanda. Unlike with the ICTY, the 1994 resolution establishing the International Criminal Tribunal for Rwanda (ICTR) expressly determined that the court had jurisdiction over ‘serious violations’ of the IHL rules governing internal armed conflict, Common Article 3 and APII.

The founding ICTR Resolution became the first international instrument to criminalize atrocities committed in internal conflict, a move whose importance cannot be overstated: Only twenty years prior, while negotiating the APs, states deemed the notion of war crimes in internal conflicts incongruous.⁶³ Still, it remained uncertain whether this UNSC Resolution could be deemed applicable to cases beyond Rwanda.

That uncertainty was soon mitigated. In October 1995 the ICTY, through the Appeals Chamber presided by Judge Antonio Cassese, delivered the landmark Tadić opinion. An appeal by Duško Tadić, a presumed Bosnian war criminal, had questioned the tribunal’s jurisdiction over acts committed in internal conflicts, arguing that its founding charter only authorized it to prosecute abuses perpetrated in international conflicts. Since violations of Common Article 3 were not technically ‘grave breaches’ of the Geneva Conventions, Tadić’s defence argued that the ICTY could not proceed.⁶⁴ The Appeals Chamber responded that while internal atrocities could not be deemed grave breaches, they constituted ‘violations of the law and customs of war’, a distinct category of acts considered under a separate article of the Tribunal’s charter. In this way, by resorting to an argument about the customary nature of Common Article 3, and ‘the core’ of APII, the tribunal set a clear legal precedent that resonated in later decisions.⁶⁵

Cassese’s touch was evident here. Years later he told an interviewer that during discussions about the ICC:

I was told there was also this fear of the ‘Cassese approach’, namely judges overdoing it, becoming dangerous by, say, producing judgments that can be innovative. For example, at the ICTY, we said for the first time that war crimes could also be committed in internal armed conflicts. This was breaking new ground. You go beyond the black letter of the law because you look at the spirit of law.⁶⁶

A decade and a half after formulating his plans to develop IHL through interpretation, Cassese deployed his legal entrepreneurship from within a new authoritative institution, an international tribunal. This became his trademark, if controversial, style.⁶⁷

The Tadić decision brought with it another crucial innovation. As some noted at the time, in its response to Tadić’s claims, instead of asserting its jurisdiction in internal conflicts with recourse to customary law, the ICTY Appeals Chamber could have simply determined that the conflict in the former Yugoslavia was international. This would have resolved the controversy in that particular case and allowed the tribunal to continue pursuing its work without complicated debates about conflict classification.

Yet a further response by Tadić’s defence, claiming that in fact no armed conflict was taking place at the time in the former Yugoslavia, enabled the judges in the Appeals Chamber to provide a positive definition of armed conflict as occurring ‘whenever there is a resort to armed force between States or protracted armed violence between governmental authorities and organized armed groups or between such groups within a State’.

This phrasing, credited once again to Cassese, immediately became noteworthy.⁶⁸ The most prominent international legal circles immediately seized upon it and considered it, alongside the ICTR statute, to herald the new era of the criminalization of internal atrocities. Meron himself agreed that this decision demonstrated ‘the renewed vitality of customary law in the development of the law of war’. In his view, ‘the clarification of customary law on this subject is the most important normative contribution of the decision’.⁶⁹

The relationship between IHL and international criminal law (ICL) is complex: each body of law features different protagonists, follows particular logics and purposes, and responds to different though partially overlapping sources and communities of practice.⁷⁰ This being said, it appears that as regards normative development the ICL-IHL interplay has largely moved in one direction: international criminal tribunals producing jurisprudence that extends IHL through interpretation.⁷¹ This jurisprudence seems to generally have been welcome; with few exceptions, even states have failed to show opposition.⁷² As I note below, this is also true of the ICRC’s Customary Law Study.

4.2 The ICRC’s turn to customary law⁷³

The ICRC’s Customary Law Study is doubtless the most prominent source of recent legal change in IHL. Analyses of the Study’s contents and methodology abound,⁷⁴ so my exploration of them here only addresses aspects relevant to my argument. My focus lays chiefly on understanding the process behind the ICRC mandate to produce the report as a lens through which to understand the politics of the customary law turn.

The Customary Law Study idea can be traced back at least to August 1993.⁷⁵ There, after a three-day International Conference for the Protection of War Victims convened by Switzerland and attended by 160 states, the UN Secretary General, and major international human rights and humanitarian NGOs, participants issued a strongly-worded declaration calling for action to improve respect for IHL. Interestingly for my purposes here, the very first question then-ICRC Director for Principles, Law, and Relations with the Movement, Yves Sandoz asked state representatives was: ‘Why have [you] not acceded to all the instruments of humanitarian law?’ Sandoz had himself participated as delegate in the negotiation of the APs, and as he rose within the ICRC’s ranks, he had surely also witnessed (and become concerned with) their normative limitations and slow ratification.

Although this 1993 Conference lacked treatymaking powers, one of its central committees (the Drafting Committee) appears to have essentially functioned as a negotiation forum of the event’s Final Declaration. Absent archival evidence, details about most states’ positions in 1993 are hard to come by, yet interviews with participants indicate that important powers such as the United States, India, Cuba, China, and Nigeria expressed opposition to the idea of extending IHL via formal treaty-making. The ICRC seemed to agree. In a 1996 article, Sandoz characterized a potential new episode of wholesale treaty revision as ‘long, costly, and hazardous’, likening it to opening a ‘Pandora’s box’ which might even allow states to roll back existing treaty IHL.⁷⁶ Sandoz’ views may plausibly be read not merely as an assessment of the global politics of the early 1990s but as harkening back to the APs’ negotiation in the 1970s.

The resulting 1993 Conference declaration omitted plans to proceed towards treatymaking. Crucially, however, it reaffirmed participants’ determination ‘to apply, and to clarify and, where it is deemed necessary, to consider further developing the existing law governing armed conflicts, in particular non-international ones’.⁷⁷

The Conference created an Intergovernmental Experts Group (IEG) to follow up on this work. The IEG then met twice, in September 1994 and January 1995, featuring widespread governmental participation (60 states in 1994 and 107 in 1995) alongside scores of other governmental and non-governmental organizations.⁷⁸

Official accounts of the IEG meetings in 1994 and 1995 refer to ‘intensive discussions and negotiations’ occurring over several days, culminating in a set of recommendations. Among these, recommendation V invited the ICRC:

to prepare, with the assistance of experts in IHL representing various geographical regions and different legal systems, and in consultation with experts from governments and international organizations, a report on customary rules of IHL applicable in international and non-international armed conflicts, and to circulate the report to States and to competent international bodies.⁷⁹

This mandate was given final approval at the 1995 International Red Cross Conference, another event attended and sanctioned by governments as well as National Societies of the Red Cross and Red Crescent. There, only one (unnamed) state is recorded as having voiced opposition to the idea of turning to customary law.⁸⁰

Two points are crucial here. First, while in the past government expert meetings of this size constituted the prelude to formal treaty revisions, now the ICRC, and the Red Cross and Red Crescent movement seemed ready to use them differently. Second, despite not constituting multilateral treatymaking by states, this preparatory phase did rely on a measure of state endorsement, in that the adoption of these different resolutions always required widespread approval by participants, including governments. States could have hardly guessed the expansive results of a study authorized in the context of an International Red Cross and Red Crescent conference, and for that reason it seems a stretch to attribute them an entrepreneurial role here. It is more likely that many states acquiesced to what might have seemed a relatively unassuming proposal, or one whose outcomes they may have been able to steer in their favour. Whether or not all states were aware of the stakes involved and supportive of a transformative endeavour, these resolutions granted governmental authority upon the ICRC to conduct the Customary Law Study.

Together then, the ICRC, National Societies of Red Cross and Red Crescent, and states all contributed to steering IHL’s pathway change towards customary law. Treatymaking was not openly discarded, yet the words and actions of these stakeholders confirm how it seemed to have acquired a reputation of risky and limited tool, potentially desirable for specific purposes (e.g., weapons regulation) but not for the broader consideration of IHL corpus of principles and rules.⁸¹ To be sure, the shift to customary law was not inevitable; had states, National Societies, or the ICRC pushed hard to reignite a codification process, they may well have succeeded. Yet a confluence of views regarding the undesirability of treatymaking in the mid-1990s foreclosed the traditional pathway of IHL change while bolstering the ICRC’s authority as legal entrepreneur, this time via the customary route.

Requiring over a decade to complete and a multitude of international and in-country experts from all regions of the world, the Customary Law Study itself constituted a mammoth effort.⁸² Legal expert teams carried out research in the ICRC’s own archives, considered national sources of nearly 50 states (nine in Africa, 11 in the Americas, 15 in Asia, one in Australasia, and 11 in Europe) international organization resolutions and reports, and jurisprudence of international courts and judicial bodies.⁸³ Scholars and governmental experts also participated in the process, including Meron and Georges Abi-Saab.

The Study was finally published in 2005 as a 4000-page three-volume report.⁸⁴ As is now well-known, its findings were sweeping: of over 161 rules of conduct in armed conflict, nearly all (146) were found to be applicable as customary to both international and internal conflicts, while a few merited the ‘arguably applicable in non-international armed conflict’ qualifier, and some others were deemed applicable only to either IAC or NIAC.⁸⁵ The most important rules extended to NIAC were those relating to the conduct of hostilities, the use of means and methods of warfare, and the treatment of persons in the hands of a party to the conflict.⁸⁶

Substantively, the Study addresses three of the core ‘flaws’ of the APs as identified by the ICRC in 1977 and by scholar-jurists in the following two decades. First, it adapted the often elaborate and sometimes awkward language of the Protocols – for instance with regard to the articles protecting the civilian population, civilian objects, and installations containing dangerous forces – reformulating them in simpler (if not altogether unambiguous) terms.⁸⁷ Second, it simplified and expanded the API rules regarding the protection of the natural environment,⁸⁸ or of humanitarian relief personnel.⁸⁹ Third, as noted, the Study ‘equalized’ the regulation of internal and international conflicts almost completely; only 12 rules in four main areas remained the ‘reserved domain’ of IACs: the definition of combatants and armed forces, conditions for POW status, the regulation of occupied territory, and the regulation of belligerent reprisals.⁹⁰

At the same time, the Study ostensibly steered clear of pronouncing on the national liberation-related provisions through which the Third World partially decolonized IHL in the 1970s. First, the Study made no effort to define armed conflicts as such, only the rules that apply to them, thereby eschewing debate about the customary status of API’s extension to conflicts against colonial, racist or occupying regimes. Second, without recreating API’s complex terminology regarding the protection of national liberation fighters, the Study’s short reformulation of the entire rule (Rule 106) may be read as incorporating such protections.⁹¹ Third, in its rephrasing of the rules dealing with the protection of civilian persons, objects, dangerous forces, and especially the natural environment, the Study furthered one of the goals most fervently endorsed by the Third World in the 1970s: a stricter limitation of the conduct of hostilities. At the same time, the Study also steered clear of advancing as customary other rules which major Western powers found (and still find) particularly objectionable, such as the general prohibition of reprisals against civilians.

The Study garnered much attention upon publication, particularly from scholars.⁹² From the side of states, a robust response came from the US in a 2006 memorandum. In it, American legal advisors John B. Bellinger III (State) and William Haynes (Defence) both recognized the Study’s general value and politely but forcefully critiqued its method, sources, and (some of) its findings.⁹³ They used four rules from the study to illustrate their methodological concerns with the Study, two of which had appeared in API in modified form: rule 51, stating that ‘humanitarian relief personnel must be respected and protected’, and rule 45, declaring that ‘the use of methods or means or warfare that are intended, or may be expected to, cause widespread, long-term and severe damage to the natural environment is prohibited’. Interestingly, as Sivakumaran and Milanovic observe, American analysis of these rules was ‘often reasonable but hardly devastating’, and none of the four critiqued rules were ‘genuinely pivotal to the structure of customary IHL as set out in the study’. For this reason, Sivakumaran and Milanovic conclude that the American memorandum did not ‘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’.⁹⁴

Besides the US, only Israel has expressed criticism of the Study.⁹⁵ Egypt and Finland have referred approvingly to it,⁹⁶ as have a handful of other states, including Malaysia, Sweden, Australia, and Azerbaijan. Beyond them, official state reaction has generally been muted. At the same time, the Study has garnered an important degree of authority through its use by an expansive range of actors, particularly domestic and international courts and tribunals but also international organizations and even national militaries.⁹⁷ The conclusion reflects further on the Study’s reception, and more generally on the politics and broader implications of the customary law turn in IHL.

5. Conclusion

This article has traced a notable shift in IHL’s path of change over the last four decades, from treatymaking to interpretation via customary law claims-making,⁹⁸ connecting this change to the experience and outcomes of the APs’ making in 1974–1977.

This conclusion offers tentative reflections on the question of how we may assess the turn to customary law politically and normatively. Given IHL’s historical development from a predominantly Western-led dominated project to a now (partially) decolonized body of law, what may we say about the turn to customary law? Has it worked to redress the ‘power-outcome’ decoupling of postwar universal lawmaking in the favour of Western states? Or has it instead furthered the move away from Western designs?

These questions have no straightforward answers. Future research, especially in government archives, should examine state perceptions and tactics across Global North and South states regarding the Customary Law Study and the work of international courts, tribunals, and scholars. This is important but challenging work, not least given issues of access and document availability, among several others.⁹⁹

This article’s findings permit two general, provisional claims. On the one hand, the customary law turn seems to have done little to undermine the decolonization of IHL as regards API’s features on national liberation; as noted earlier, these were largely eschewed in the ICRC’s Study.¹⁰⁰ On the other hand, given its expansive approach to the humanization of internal conflicts far beyond the confines of APII, the Study clearly runs against the past wishes of a great majority of Third World states (though not just them), which in the 1970s refused to compromise their sovereignty to permit stronger rules for NIAC. Those states justified their position then as upholding the principles of non-intervention against potential neo-colonialism. In that sense, one’s assessment of the ‘revolutionary’ change in the area of NIAC regulation depends on whether one views IHL rules and principles for internal conflict as either authorizing neo-colonialism or, conversely, as tending to induce restraint and reduce atrocity.¹⁰¹

Two other possible ways to tentatively gauge the politics of the customary law turn involve interrogating its process-level legitimacy, and wider reactions to its outcomes. On process, the authorization of the Study’s mandate in various International Red Cross and Red Crescent forums, as well as the Study’s actual making via consultative global research, together suggest their reliance on widespread approval and participation. Indeed, according to Sivakumaran and Milanovic, the Study is underpinned by epistemic authority, not just because it was steered by the ICRC ‘with its 150-plus years of expertise in IHL’ but because it was based on the participation of many independent experts, academics, and governmental experts.¹⁰² In addition,

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.¹⁰³

Here I make no attempt to ascertain whether any aspect of the Study is correct, but simply underscore that its findings about customary IHL have been presented (and largely received) not as the predilection of the ICRC but as the product of a robust collective process.

In terms of outcomes, as noted earlier, official state reactions to the Study have been scarce and generally brief, with only the US issuing a dedicated (if incomplete) evaluation of it. There has been no concerted state pushback against the Study; meanwhile, ‘so many international legal institutions have treated it as authoritative’ that its ‘process of accretion is highly likely to continue’.¹⁰⁴

To be sure, states’ muted reaction to the Study remains a puzzle. Should we take it to mean acquiescence, obliviousness, or rather as a strategic choice by them neither to confirm nor deny an IHL rule’s purported customary status, maintaining freedom of action without ‘outing’ themselves as ‘anti-humanitarian’? Future research should attempt to investigate this crucial question. Yet it bears underscoring here that, when criticism or scepticism has emerged against the Study, it has come from the US and some of its allies, not from Global South states. In the end, as Sivakumaran and Milanovic shrewdly note, the Study is ‘exactly as authoritative as states have allowed it to be’,¹⁰⁵ and so far, states have opted for leniency.

Is the era of IHL treatymaking over? Such a claim seems premature and in the long run, probably erroneous. Yet, given the resurgence of geopolitical discord among great powers, and of major interstate war, it is much more likely that contemporary and future change in IHL will keep coming from customary law-ascertainment and soft law documents, among others, than from binding codification.¹⁰⁶ Processes taking those alternative forms have been in motion of years now regarding new areas of concern (e.g., cyber operations) as well as more traditional ones.¹⁰⁷ Comparative research should thus continue regarding the politics of contestation surrounding the many recent and ongoing efforts non-codification IHL-making.¹⁰⁸ In terms of ICRC-led initiatives, another critical case exists in the Interpretive Guidance on the Notion of Direct Participation in Hostilities,¹⁰⁹ and in the revised commentaries on the four Geneva Conventions and the APs. We may also further witness the ‘unilateralization’ of IHL-making by states,¹¹⁰ including through the publication of national military manuals that rephrase IHL in ways that suggest a muddying of existing law.¹¹¹ And all IHL stakeholders should watch with close attention developments in state practice and opinio juris emerging from contemporary conflicts, including that between Russia and Ukraine.

Whatever path of change IHL may take in the future, legal entrepreneurs, states, scholars, and civil society must continue interrogating its politics, norms, and procedures. The legitimacy of IHL – its authority, ownership, and likelihood of influence – lies in the balance.