The nonretroactivity principle and custom

The most important time constraint on IHL is a principle older than IHL: nonretroactivity. First mentioned in the 1789 Constitution of the United StatesFootnote ¹⁰ and the 1789 Déclaration des Droits de l'Homme et du Citoyen,Footnote ¹¹ it was conceptualized by Anselm von Feuerbach in 1801. Feuerbach coined the phrase under which the principle became famous: nullum crimen, nulla poena sine praevia lege (no crime, no penalty without previous law).Footnote ¹² Nonretroactivity is a principle of general law, but from Feuerbach's formula we can infer that its most important application is in criminal law. That criminal aspect is probably the reason why we do not see any prominent appearance of the principle in IHL until 1945, when the punishment of war criminals was tackled in earnest with the establishment of the International Military Tribunal at Nuremberg. Indeed, this Tribunal was quickly accused of administering retroactive justice in trying the Nazi war criminals on the basis not only of existing crimes (war crimes) but also of new crimes (crimes against peace and crimes against humanity).Footnote ¹³ The Tribunal's response to this accusation was double: in its charter it stated that it would punish crimes against humanity “whether or not [committed] in violation of the domestic law of the country where perpetrated”, and in its judgment it argued that its charter was “the expression of international law existing at the time of its creation” and that “individuals ha[d] international duties which transcend[ed] the national obligations of obedience imposed by the individual state”.Footnote ¹⁴ As to war crimes committed by the Nazi war criminals in particular, it further stipulated that “[t]he evidence relating to War Crimes has been overwhelming, in its volume and its detail. … The truth remains that War Crimes were committed on a vast scale, never before seen in the history of war.”Footnote ¹⁵ In other words, the Tribunal argued that by 1939 – the year in which its jurisdiction ratione temporis beganFootnote ¹⁶ – the 1907 Hague Conventions and the 1929 Geneva Conventions had been recognized by all civilized nations,Footnote ¹⁷ meaning that they had acquired the status of customary international law applicable to all States, whether parties to the Hague and Geneva Conventions or not.Footnote ¹⁸ The Tribunal thus argued that the Nazi crimes had breached already existing customary international law and that, therefore, it did not violate the nonretroactivity principle in dealing with the Nazi crimes.Footnote ¹⁹ In other words, it appealed to “international custom”, one of the sources of international law recognized by the Permanent Court of International Justice (the world court from 1921 to 1946) and its successor, the International Court of Justice (ICJ).Footnote ²⁰

The Tribunal's argument became known as the “Nuremberg clause”. It was reaffirmed by the United Nations (UN) General Assembly in 1946Footnote ²¹ and repeated in the Universal Declaration of Human Rights in 1948.Footnote ²² In 1950, the International Law Commission reformulated the argument as a principle: “The fact that internal law does not impose a penalty for an act which constitutes a crime under international law does not relieve the person who committed the act from responsibility under international law.”Footnote ²³ The nonretroactivity and Nuremberg principles would later be integrated together into the 1966 International Covenant on Civil and Political Rights (with a non-derogable status),Footnote ²⁴ the 1977 Additional Protocols,Footnote ²⁵ and the 1998 Rome Statute of the ICC (Rome Statute).Footnote ²⁶

By definition, the nonretroactivity principle restricts the scope of time.Footnote ²⁷ For the Nuremberg Tribunal, the jurisdiction ratione temporis stretched back to 1939 (six years before its establishment); for the ICC, the jurisdiction is prospective, not retrospective.Footnote ²⁸ However, the reference to customary international law endowed the time-constraining nonretroactivity principle with unexpected breadth. The applicability of the Hague and Geneva Conventions as customary international law in 1939 meant a de facto temporal scope of at least four decades (namely from 1899 to 1939).Footnote ²⁹ In this context, it is also noteworthy that after 1949, most provisions of the Geneva Conventions have gradually been considered as customary IHL themselves.Footnote ³⁰ What custom does is shift the critical starting date of temporal jurisdiction backwards.

The humanity principle and the Martens Clause

Another principle that helped expand the temporal scope of IHL towards the past was the principle of humanity. This principle was formulated most famously in the so-called Martens Clause, which in its original wording in the preamble of Hague Convention II on the Laws and Customs of War on Land of 1899 read:

This residual clause – originally not an expression of lofty ideals but the solution to a pressing problemFootnote ³² – regulated all the problems that Hague Convention II did not foresee. In metaphysical language reminiscent of natural law, it attempted to bridge morality and law by promoting the triad of custom (“usages”), humanity and conscience as a compass for all conduct in armed conflict not explicitly covered in the Convention. The clause was repeated with minor variations in scores of IHL instrumentsFootnote ³³ and recognized as part of customary international law by the ICJ.Footnote ³⁴ Vividly invoking custom, the Martens Clause became one itself.

Reasoning by analogy and the principle of continuing breaches

Other tools with the potential to expand the temporal scope backwards – the method of analogy and the concept of continuing breaches – were not taken up in IHL. The Rome Statute stipulates that definitions of crimes must be strictly construed and prohibits extending them by analogy.Footnote ³⁵ This prohibition, however, refers to analogy as a tool of law-making, not as a tool of interpretation. As tools of law-making – for example, by creatively widening the list of war crimes – analogies are detrimental to fair trial rights. As tools of interpretation, however, analogies between cases often endow investigations with a historical dimension that provides relevant context. Obviously, when analogies are invoked, the presumed precedents should be selected and used with methodological delicacy.Footnote ³⁶

The notion of continuing breaches – breaches of obligations enduring over time – was first introduced by Heinrich Triepel in 1899. A continuing breach is a breach which started before the critical moment that temporal jurisdiction comes into effect but continues after that critical moment. Phenomena such as enforced disappearances, confiscation of property, sexual slavery, conscription of children, forcible population transfer, unlawful occupation, and maintenance of colonial domination by force extend over time and are seen as continuing breaches.Footnote ³⁷ This frequently discussed notion is underexplored in IHL, and it was not taken up in the Rome Statute, although it has been used by the UN Human Rights CommitteeFootnote ³⁸ and the UN Working Group on Enforced or Involuntary Disappearances,Footnote ³⁹ among others, and proposed in the 1991 Draft Articles on Responsibility of States for Internationally Wrongful ActsFootnote ⁴⁰ and the 2005 UN Reparation Principles.Footnote ⁴¹ It could also become part of a future Convention on Responsibility of States for Internationally Wrongful Acts.Footnote ⁴²

Imprescriptibility versus legal forgetting

In some circumstances, the temporal scope can also be expanded towards the future. In law, the finality principle reigns: interest rei publicae ut finis litium sit (it is in the public interest that lawsuits should end).Footnote ⁴³ This principle imposes time limits (statutes of limitations) on the prosecution of crimes. After World War II it gradually dawned, however, that atrocity crimes (an umbrella term for genocide, crimes against humanity and war crimes) should be exempt from the principle – not only because these crimes complicated and thus extended the duration of investigations, but also because they constituted such an affront to humanity that taking responsibility for them was inescapable, however long after the fact.Footnote ⁴⁴ Nevertheless, the history of the idea of imprescriptibility – the waiving of time bars – is long and twisted. This is partly due to a curious legal reasoning: the logic is that where international criminal law does not mention any time bars for prosecution, imprescriptibility applies; only explicit mention of time bars is interpreted as a rejection of imprescriptibility.Footnote ⁴⁵ This oddity may partly explain why the Geneva Conventions of 1949 remained silent about the lifting of time bars for war crimes, although by then it had become quite clear that the search for and prosecution of World War II criminals would continue for many years:Footnote ⁴⁶ the silence about time bars in the Geneva Conventions was later interpreted as support for the imprescriptibility of war crimes. Nevertheless, the UN convention that explicitly blocked the pending prescription, after twenty-five years, of World War II crimes by declaring atrocity crimes imprescriptible in 1968 had less effect than expected.Footnote ⁴⁷ In the end, resistance to the idea slowly waned and the imprescriptibility provision in the Rome Statute was adopted in 1998 without major problems.Footnote ⁴⁸ The provision fits the ICC philosophy of working “for the sake of present and future generations” well.Footnote ⁴⁹ Most legal scholars have since accepted imprescriptibility as a rule of customary international law.Footnote ⁵⁰

In tandem with the imprescriptibility discussion came the troubling problem of forgetting the past under international law: under which conditions could past crimes be legally forgotten through amnesties? It would be decades before this tough question was tentatively regulated in Additional Protocol II (AP II) of 1977, which requires authorities “to grant the broadest possible amnesty to persons who have participated in the armed conflict”.Footnote ⁵¹ It was logical to think that if atrocity crimes violated a jus cogens norm (a non-derogable and peremptory norm) – namely the humanity principle – the prosecution of these crimes was a jus cogens norm as well.Footnote ⁵² The practice, however, was different: the urge to reconcile and forget was often more powerful than the urge to prosecute, especially in situations where massive violence had left the hands of many dirty.Footnote ⁵³ Behind this tension between principle and practice lay the deeper ambition to balance two basic values: peace and justice. After armed conflicts, the need for peace required broad amnesties but the need for justice made exemptions for atrocity crimes imperative. The Rome Statute, which is silent about amnesties,Footnote ⁵⁴ did not tackle the issue, and this debate is still not settled. Amid ever stronger campaigns to combat impunity, legal forgetting remains a rock. The equilibrium is fragile.

State succession

Another mechanism for enlarging the temporal scope is State succession. In the law of treaties, new States are not bound by treaties of their predecessors because they experience a fundamental change in circumstances (a doctrine known as rebus sic stantibus).Footnote ⁵⁵ There is, however, near-consensus that when new States are established, the rule of continuity with the predecessor State still applies with respect to one particular type of obligations: the humanitarian and human rights of citizens.Footnote ⁵⁶ The UN Human Rights Committee has explained the fundamental reason for this continuity:

Like imprescriptibility, State succession stretches the temporal scope of IHL provisions into the future.

Looking at the entire panorama, then, we can conclude that the applicability of IHL is extended towards the past by means of the customary strand of nonretroactivity and the humanity principle as embodied in the Martens Clause. It is also extended towards the future by means of the State succession principle in IHL matters and the imprescriptibility principle in criminal matters, although a countertrend of legal forgetting under the guise of amnesties curbs this expansive tendency. On balance, IHL's time regime is an ingenious mixture of immediacy and longue durée. Its broad horizons constitute the background for discussing the IHL views of memory and history.

Dead persons and gravesites as portals to memory

The Geneva Convention of 1864 did not pay attention to the war dead, but this would soon change. Deeply affected by the battlefield scenes of the 1866 Austro-Prussian war, the German physician Julius Naundorff wrote about “the hyenas of the battlefield” who pillaged dead bodies.Footnote ⁶⁰ In 1867–68, conferences in Paris and Geneva studying the gaps in the 1864 Convention expressed wishes (voeux) to protect the dead against desecration and pillage, to bury them in conformity with sanitary prescriptions, to identify them and to notify other countries by exchanging death lists.Footnote ⁶¹ This early sensitivity to the dead was dominated by fear: fear of burying the living and fear that unattended corpses would endanger public health.Footnote ⁶²

The idea of codifying treatment of the war dead was taken up again in the Hague Conventions, which made compulsory some regulations about death certificates, last wills and burial of war prisoners.Footnote ⁶³ The 1906 Geneva Convention translated the wishes of 1867–68, including the need to clarify the whereabouts of the dead, into binding provisions.Footnote ⁶⁴ The duties flowing from these provisions were seen as duties of conduct, not of result.Footnote ⁶⁵ The 1929 Conventions expanded upon them with a crucial new duty: the duty to search for the dead. At the same time, identification requirements and disposal practices were tightened and became duties to honourably bury the dead and to respect and mark their graves, including for exhumation purposes.Footnote ⁶⁶ Thus, the notion of respect for the dead, itself stretching back into the mists of time, received solid codification in IHL in 1929.Footnote ⁶⁷

After World War II, IHL prescriptions about the dead would multiply and become more systematic. The 2005 ICRC Customary Law Study summed up this post-war system elaborated in the 1949 Geneva Conventions and perfected in their 1977 Additional Protocols: search for and collection of the dead, protection and respectful treatment of the dead, return of the remains and personal effects of the dead upon request, respectful individual disposal of the dead, respect for and maintenance of graves, and accounting for the dead through their identification prior to and after disposal and through marking and accessing graves.Footnote ⁶⁸ These prescriptions now form an uncontested part of customary IHL.

Mutilating dead bodies was seen as an offence from relatively early on. It was treated as a war crime at the US General Military Government Court in Dachau in 1947.Footnote ⁶⁹ The idea that mutilating the war dead was a serious IHL breach which could be subsumed under the provision of “outrages upon personal dignity” prohibited under Article 3 common to the four 1949 Geneva Conventions matured slowly into treaty and custom. It was clearly expressed for the first time in the Elements of Crimes of the ICC in 2002.Footnote ⁷⁰

The influence of World War II, and particularly the Holocaust, can be noticed in two aspects. Whereas in the 1929 Conventions cremation was a disposal option equivalent to burial, this changed completely under the influence of the discovery of the Nazi death camps, with their crematoria. Although still permitted, cremation had to meet stricter conditions in the 1949 Conventions because it was irreversible, prevented identification and effaced traces of crimes: it could be carried out only for imperative reasons of hygiene, for a wish expressed in a will or for religious motives, and the reason had to be stated on the death certificate.Footnote ⁷¹ Similarly, mass disposal was rejected in 1949 unless absolutely unavoidable because it conflicted with the principle of respect for the dead and made identification, grave visits by families and exhumation for reasons of overriding public necessity (i.e., for public health, investigative, reburial or return purposes) exceedingly difficult.Footnote ⁷² We can conclude that sustained attention to the dead has been a constant hallmark of IHL since at least 1929.

The principle of intertemporal law

It is unclear, however, how long IHL duties to the dead last. Searching for, identifying and protecting the war dead are open-ended tasks that continue to be performed in peacetime. Similarly, the maintenance of, and access to, gravesites may last quasi-indefinitely, and the duty to exhume and criminally investigate deaths or return remains may come up decades after an armed conflict has ended.Footnote ⁷³

When the need arises to solve disputes about the exhumation of human remains, the removal of a war cemetery or the transfer of a war memorial, the question is which IHL norms must be followed: those in vigour at the time of the victim's interment, the cemetery's construction or the memorial's creation, or those at the time of the disputes. Anna Petrig has suggested invoking the principle of intertemporal law to solve this problem.Footnote ⁷⁴ This principle was developed in 1928 by Swiss arbitrator (and later ICRC president) Max Huber at the Permanent Court of Arbitration. He wrote:

Bearing in mind the nonretroactivity requirement, this means that the law contemporaneous with the facts must prevail. But what are the facts? Applied to our problem, the principle has to factor in that remains, gravesites or memorials are “facts” that have not ceased to exist but persist over time. Their “continuing manifestation” imposes the use of IHL as evolved at the time of the dispute.Footnote ⁷⁶ We notice that whereas the idea of continuity is not applied to IHL breaches (see above), it is quietly introduced here through the intertemporal principle as applied to “persistent facts”.

Cultural heritage as a portal to memory

Sites of cultural heritage – including gravesites and memorials for the war deadFootnote ⁷⁷ – are sites of memory. From this perspective, the Hague Conventions of 1899 and 1907 were the first IHL instruments to tackle the problem of how to reconcile, in times of war, military necessity with respect for historical monuments and places of worship and works of art. They used a legal fiction and categorized all monuments as private property even when State-owned. The Hague Regulations then prohibited the confiscation of private property. Seizure of, destruction of, and intentional damage to historical monuments would be made the subject of proceedings unless the monuments were used for military purposes.Footnote ⁷⁸ The 1949 Geneva Conventions did not refine these views, but the neglect was temporary as a specialized convention, the Hague Convention for the Protection of Cultural Property in the Event of Armed Conflict, was concluded in 1954. It aimed at shielding a subset of the objects protected by the Hague Conventions: cultural property. The new convention defined “cultural property” as “movable or immovable property of great importance to the cultural heritage of every people” and grouped monuments, archaeological sites, museums, archives and libraries under the concept. On top of this, the 1977 Additional Protocols aimed at protecting a subset of this subset – cultural property which “constitute[s] the cultural or spiritual heritage of peoples”. In this way, the provisions of 1907, 1954 and 1977 formed a pyramid of protective layers for heritage, with the latter two shielding unique and transcending categories of heritage of humanity by prohibiting acts of hostility against them as well as their use in support of military efforts or as objects of reprisal.Footnote ⁷⁹

The States that ratified the 1998 Rome Statute made it clear in its preamble that they were “conscious that all peoples are united by common bonds, their cultures pieced together in a shared heritage, and concerned that this delicate mosaic may be shattered at any time”.Footnote ⁸⁰ The Statute criminalized IHL violations against cultural heritage in two ways: indirectly, by defining the persecution of groups on cultural grounds as a crime against humanity,Footnote ⁸¹ and directly, by labelling intentional attacks against historical monuments and other heritage during armed conflicts as war crimes unless imperatively required by military necessity and regardless of whether the attack results in actual damage.Footnote ⁸² The 2005 ICRC Customary Law Study maintained that the protection of historical monuments and broader classes of cultural heritage in wartime had been part of customary law since 1899 (and even decades earlier).Footnote ⁸³ In short, we see that in IHL, codification awareness of cultural heritage has an even longer pedigree than awareness of respect for the dead. The IHL view of memory is strongly embedded in the IHL provisions about dead persons and cultural heritage.

The right to the truth, archives and families

From early on, the IHL treaty drafters tried to solve three problems that were preconditions for collecting facts about war victims in order to write their life stories: how to prevent war prisoners and war dead from going missing in the turmoil of war; how to repatriate last wills, money, articles of sentimental value and other possessions found on the dead; and how to address the need of families to obtain corroborated information about the whereabouts of the dead in their own as well as enemy countries. Dealing with these issues, which today are grouped together under a “right to the truth”, required the setting up of a documentary infrastructure in often chaotic circumstances. The ICRC quickly perceived this enormous need for collecting and exchanging data and objects while simultaneously protecting them against abuse.

As early as 1870, during the Franco-Prussian War, the ICRC established an information bureau for war prisoners (known since under various names, including the International Prisoners of War Agency from 1914 and the Central Tracing Agency from 1960). Over the years, and especially during the World Wars, data on many millions of war prisoners and war dead were collected and exchanged – an invaluable service, as war-torn countries always struggle with disorganized bureaucracies and scattered archives. Following the idea of establishing inquiry offices set out in the Hague Conventions of 1899 and 1907, the 1929 Geneva Conventions imposed the establishment of a Central Agency of Information regarding Prisoners of War and of official national information bureaus and grave registration services in conflict-ridden countries. From World War I, ICRC delegates also began tracing people in the field by visiting camps and prisons. The Central Agency of Information regarding Prisoners of War was also active in World War II and scores of post-war conflicts and disasters.Footnote ⁸⁴

The unusual documentary awareness of the ICRC is also demonstrated by the archives that it has kept since its foundation in 1863, and by its journal, the International Review of the Red Cross, which still exists after more than 150 years.Footnote ⁸⁵ All this time, the ICRC has had to carefully balance the transparency needed to research data with the privacy of the victims and the confidentiality of its missions.Footnote ⁸⁶ The archives of the International Prisoners of War Agency from 1914 to 1923 became part of UNESCO's Memory of the World Register in 2007. Data collection on prisoners in World War II was started in 1943 under the aegis of the Allied authorities and from 1948 found a home in the International Tracing Service established in Bad Arolsen under the auspices of successively the International Refugee Organization (1948–51), the Allied High Commission for Occupied Germany (1951–55), the ICRC (1955–2012) and an International Commission (2013–present). This work became part of UNESCO's Memory of the World Register in 2013 and was renamed the Arolsen Archives – International Center on Nazi Persecution in 2019. This data collection on Nazi concentration and death camps between 1933 and 1945 is unparalleled. Between 1983 and 2006, however, the archive was criticized for its isolation and poor accessibility. It took time to restore the balance of competing interests.Footnote ⁸⁷

From this thumbnail sketch we can see that, although the right to the truth as a concept was only fully developed during the 1990s, its constituent elements – identification of victims, repatriation of objects, and recognition of the need of families to obtain information about victims – were prime IHL concerns from the beginning. In moving passages, commentators Gustave Moynier (in 1870), Carl Lueder (in 1876) and Paul des Gouttes (in 1930) conjured up an image of families torn for years or even decades by the moral anguish of ignorance and uncertainty, begging for crumbs of information, struggling to receive due benefits and estates, and, in the end, longing to start a mourning process.Footnote ⁸⁸ Later commentators made similar remarks.Footnote ⁸⁹ The need of families in particular was recognized in the Hague Regulations from 1899 and in the Geneva Conventions from 1929.Footnote ⁹⁰

It is hard to explain why it took half a century – until the preparatory talks for Additional Protocol I (AP I) – before a discussion was started about whether this need of families was actually a right. This author's thesis, which might be called the conceptual salience thesis, is that the common traits of three distinct phenomena – the dead, the missing and the disappeared – only became apparent in the mid-1970s. Whereas IHL had paid sustained attention to the dead from virtually the beginning and had begun codifying its principles from 1906, it long treated missing persons – who could be dead or alive – as a loose category. In 1870, Moynier touched upon the phenomenon of missing persons, but by linking them to the dead: some of these disparus of the battlefield, he wrote, were deserters, but most were soldiers, hastily buried without identification.Footnote ⁹¹ And in 1930, while evoking the image of the unknown soldier, des Gouttes also recalled the fate of missing persons.Footnote ⁹² But while the need to clarify the whereabouts of the dead had explicitly emerged in the 1906 Geneva Convention, as we saw, Geneva Convention IV of 1949 only stipulated the facilitation of enquiries made by members of dispersed families seeking to renew contact with each other.Footnote ⁹³ A codification of missing persons similar to the 1906 one did not appear until 1977. Article 33 of AP I is devoted to “Missing Persons”. It was widely seen as filling a gap.Footnote ⁹⁴

In contrast to the dead and the missing, the disappeared refer to only one specific criminal subset of the missing: those who went missing after unacknowledged abduction. The practice of enforced disappearances was discussed at the Nuremberg Tribunal in 1945–46, under the counts of war crimes and crimes against humanity. This happened mainly in the context of the 1941 Nacht und Nebel Erlass (Night and Fog Decree), which had prescribed either the execution or the disappearance in complete secrecy of those who resisted Nazism, with the express intent to intimidate their families and others. The Nuremberg judges noted that this decree was a terror technique which violated the family rights protected in the Hague Regulations.Footnote ⁹⁵ From 1966, disappearances also emerged as a large-scale practice in Guatemala, as they did shortly after elsewhere in Latin America.Footnote ⁹⁶ However, we have to wait until a resolution of the UN General Assembly in 1978 for a conceptualization of the phenomenon. This resolution on disappearances – which, curiously, does not mention the broader category of the missing at all – refers to previous ICRC experience: the general provision on the missing in AP I may have spurred more specific thinking on disappearances eighteen months later.Footnote ⁹⁷

Whatever the validity of the conceptual salience thesis, AP I includes a section entitled “Missing and Dead Persons”, the opening article of which – Article 32 – formulates the “general principle” that activities for missing and dead persons should be “prompted mainly by the right of families to know the fate of their relatives”.Footnote ⁹⁸ Although discussions about this principle went back to resolutions adopted by the International Conference of the Red Cross in 1973 and the UN General Assembly in 1974,Footnote ⁹⁹ these early resolutions had not spoken about a “right” of families. Commentator Yves Sandoz recalled that the drafters of AP I adopted the term “right” “after careful reflection and … in full consciousness”, emphasizing that it was not a right of governments, but a right of families.Footnote ¹⁰⁰

Therefore, Article 32 of AP I, with its right of families, was revolutionary for two reasons: it constituted the real birth of the right to know, or as it is now called, the right to the truth; and, by using a rights vocabulary, it connected IHL to human rights.Footnote ¹⁰¹ Once seen as a right, and not merely a need, and as a right of families and not of governments, the principle opened up three new perspectives: families could make claims regarding investigation and reparation, governments had duties to respond to these claims, and violations of the right of families to know the truth were forms of inhuman treatment separate from, and additional to, violations of the rights of the missing themselves.Footnote ¹⁰² Following this trend, the Rome Statute two decades later paid attention to the safety and psychological well-being of families of crime victimsFootnote ¹⁰³ and emphasized that enforced disappearances were crimes against humanity.Footnote ¹⁰⁴

Over time, it also became clear that the right to know, although a strong procedural right, was not absolute in two respects. First, while providing strong impetus to governments to investigate specific cases, individual families could not force governments to take particular actions.Footnote ¹⁰⁵ Second, the right had to be balanced against all governmental duties, including duties to gather forensic evidence for criminal investigations into violent deaths, which could hinder rather than help communications with families.Footnote ¹⁰⁶

In sum, the ICRC has consistently advocated for State duties to the war dead from the beginning, organizing large-scale data collection and analysis in the process. This investigative work was a necessary, but not sufficient, condition for addressing the need of families to write the life stories of the war dead and for the broader right to search for the historical truth. Following the logic of its own thought, by converting this need to a right, the ICRC became a pioneer, from the mid-1970s, in developing the right to the truth. On a side note, the problem of the denial of war crimes – a pervasive abuse of history – and the powerful role of the right to the truth in countering it were never systematically raised within IHL.

IHL breaches, impunity and repression

The IHL treaty drafters did not reason in the abstract – they reacted to the problems of their time. In order to see how they mobilized the past in these reactions, we must therefore also examine how they responded to major historical challenges. Five of these challenges are singled out here. These are, in chronological order: IHL breaches, the Holocaust, human rights, domestic conflicts and colonialism. The intention of these five exercises is not to write a complete history of IHL but rather to evaluate whether, how and how quickly the IHL community responded to these challenges and whether those reactions have resulted in continuity or change.

The first historical challenge was the problem of the paper tiger: how to respond credibly to breaches of IHL. Within the present analysis, this question will only examined at the level of international accountability. An 1872 proposal by Red Cross co-founder Gustave Moynier to establish an international criminal court failed.Footnote ¹⁰⁷ The Hague Conventions of 1899 and 1907 attempted to peacefully settle disputes, including disputes about breaches, through good offices, mediation, inquiries and arbitration, but not through prosecution. The drafters of these conventions reasoned in terms of State responsibility, not individual responsibility.Footnote ¹⁰⁸ This changed after World War I, when the 1919 Versailles Treaty, in a section on penalties inspired by the Hague Regulations, stipulated that nationals of the defeated countries who had committed “acts in violation of the laws and customs of war” would be punished.Footnote ¹⁰⁹ Although the term “war crime” did not literally appear in this treaty – and would not appear in any conventional IHL instrument until sixty years laterFootnote ¹¹⁰ – the year 1919 can be considered the birth date of the notion of war crimes. However, the attempts to punish individual perpetrators largely failed at the time.Footnote ¹¹¹

Nuremberg, discussed above, changed this. Following Nuremberg, a system of individual criminal responsibility, with penal sanctions for abuses and breaches, was incorporated – albeit reluctantly – in the 1949 Geneva Conventions and their 1977 Additional Protocols.Footnote ¹¹² War crimes for which individuals could be held liable regardless of their official capacity were codified in the statutes of a chain of international criminal tribunals, from Yugoslavia and Rwanda to the ICC in 1998.Footnote ¹¹³ Thus, while impunity prevailed for most of history, after World War II a system of war crimes trials was set up and then gradually perfected. As Horace said: “Raro antecedentem scelestum deseruit pede poena claudo” (Rarely does punishment, even at a slow pace, fail to overtake the criminal in his flight). But the pace was slow indeed, taking from 1872 to 1998 – more than a century.

IHL and the Holocaust

The Holocaust posed an unprecedented challenge to the IHL treaty drafters. This can be illustrated by the conduct of the ICRC between roughly 1942 and 1995. Notwithstanding – or perhaps because of – the fact that during World War II the German Red Cross operated under Nazi control and that its leadership even participated actively in crimes against the Jews,Footnote ¹¹⁴ the ICRC had become aware of the large-scale persecution inflicted on Jews by the summer of 1942. It considered launching a general public appeal to denounce these gross violations, among others, and it prepared a draft for such an appeal, with four central points of concern: aerial bombing raids; blockades; deportation, hostage-taking and massacres of civilians; and the fate of war prisoners not protected by the 1929 Geneva Convention.Footnote ¹¹⁵ On 14 October 1942, however, the ICRC decided against the launch of the appeal, believing that it would be unable to stop the atrocities. Its subsequent confidential diplomatic approaches to Reich authorities and further activities in individual countries were still impressive – it was awarded the 1944 Nobel Peace Prize for such efforts – but on the whole the conduct of the ICRC was a moral failure. This general feeling led to a lingering crisis of conscience within the organization for the decades to come.

The first post-war reaction manifested itself in a determination to adapt the instruments of IHL. The large-scale suffering of civilians under foreign occupation in World War II – apart from atrocity crimes, this also included the other points of the 1942 appeal – spurred the drafters of IHL treaties to redouble their efforts. The missed opportunities of the past and the atrocities of World War II forced a revision of the two 1929 Geneva Conventions and haunted the drafters of the new Geneva Conventions in 1949.Footnote ¹¹⁶

In order to understand this, we need to go back to 1919. If anything, World War I had shown that civilians lacked protection in wartime; provisions in the Hague Regulations to that effect had proved insufficient.Footnote ¹¹⁷ Therefore, as early as 1921, the ICRC proposed to prepare a convention for civilians. The initiative was rejected, however, because some government representatives feared that it could weaken the hard-won peace and general optimism following the Great War. The ICRC nevertheless continued studying the issues and prepared a draft, which was ready in 1934. In the end, a diplomatic conference, convened in 1939 to adopt this so-called “Tokyo draft” in early 1940, could not take place, as World War II had broken out. The belligerent States refused to bring the Tokyo draft into force but nevertheless agreed to apply the 1929 provisions for war prisoners to civilians at risk of internment for being in enemy territory when hostilities opened. This helped some 160,000 civilians in internment camps but left millions of others without protection against deportation and internment during the war. The frustration that efforts initiated in 1921 had had so few tangible results and the inability to stop the wartime atrocities, which prominently included the genocidal frenzy of the Holocaust, led to a vigorous drive “to bridge this tragic gap”Footnote ¹¹⁸ and extend IHL protection to civilians in a separate convention. This became Geneva Convention IV of 1949.Footnote ¹¹⁹

This resounding success did not dissipate the awareness of failure or the silence about the Holocaust in the International Review of the Red Cross,Footnote ¹²⁰ nor the criticism of third parties. In 1975, the ICRC reissued a 1946 report about its work for civilian detainees in German concentration camps, “by way of reply to the many questions from governments, National Red Cross Societies, associations and individual inquirers”.Footnote ¹²¹A few years later, in 1979,Footnote ¹²² it eventually decided that more decisive action was needed: it opened its archives and commissioned an independent external study about its response to the Holocaust to historian Jean-Claude Favez, who published his book in 1988.Footnote ¹²³ Although the initial reaction to this book by then ICRC president Cornelio Sommaruga was very defensive,Footnote ¹²⁴ the silence was broken and a debate opened. It took another seven years until Sommaruga, on the fiftieth anniversary of the liberation of Auschwitz in 1995, declared in Krakow that “Auschwitz … represents the greatest failure in the history of the ICRC, aggravated by its lack of decisiveness in taking steps to aid the victims of persecution”.Footnote ¹²⁵ Since 1995, the ICRC has reiterated this view, including in statements by the incumbent ICRC president, historian Peter Maurer.Footnote ¹²⁶ The history of the ICRC's reaction to the Holocaust tragedy reveals that processes of dealing with a difficult war past affect not only States but also international organizations, that these processes can be postponed, and that when confronted at last, they take much time to address.

IHL and human rights

Initially, IHL circles greeted the Universal Declaration of Human Rights of December 1948 with curiosity but without noticeable enthusiasm.Footnote ¹²⁷ There were several reasons for this.Footnote ¹²⁸ First, the genealogies of IHL and human rights differed: sensu largo, IHL went back to antiquity, human rights “only” to the Enlightenment; sensu stricto, IHL was born in 1864 and human rights in 1948. Second, their traditions diverged: IHL proceeded by discretion, human rights by revendication, and in addition, IHL regulators saw human rights as a political idea at odds with IHL's professed neutrality. Third, conventions (such as the Geneva Conventions) had more force than declarations (such as the Universal Declaration). Although many predicted the historic character of the Universal Declaration, nobody foresaw its future as an exceptional instrument with the status of customary international law.Footnote ¹²⁹ Fourth, IHL and human rights spoke different languages. The Universal Declaration contained only one oblique reference to armed conflict when in its preamble it mentioned the “barbarous acts which have outraged the conscience of mankind” and “tyranny and oppression”.Footnote ¹³⁰ Conversely, the 1949 Geneva Conventions – adopted eight months after the Universal Declaration – did not contain a single reference to human rights: proposals for a preamble which included such a reference did not reach unanimity and were dropped.Footnote ¹³¹ Finally, there was a basic antagonism between international human rights law, which prohibits the use of force, and IHL, which establishes rules for situations which according to human rights should not exist.Footnote ¹³²

Even the 1977 Additional Protocols still reflected the laborious relationship between IHL and human rights. AP I's preamble mentioned the UN, but the notion of human rights was conspicuously absent (although it was acknowledged at the beginning of the section on the treatment of persons in the power of belligerent States).Footnote ¹³³ In contrast, AP II's preamble ceded a place of honour to the concept. The official appearance of human rights in the Additional Protocols was the first significant result of the slow rapprochement between IHL and international human rights law that had started in earnest in 1968, when the International Conference on Human Rights in Tehran instructed the UN Secretary-General to prepare reports about respect for human rights in armed conflicts. This initiative, though, came from the side of human rights, not from IHL.Footnote ¹³⁴

The relatively minor role of human rights in AP I, which is dedicated to international armed conflicts, and its prominent role in AP II, which is dedicated to domestic conflicts, was no coincidence. The Uppsala Conflict Data Program shows why. Its database of armed conflicts between 1946 and 2020 reveals that inter-State and intra-State conflicts were on a par until about 1955, but after that date intra-State conflicts (including internationalized intra-State conflicts) massively outnumbered inter-State conflicts.Footnote ¹³⁵ Despite this predominance, the only IHL norms governing intra-State conflicts were those found in common Article 3 and additionally, from 1977, those in the modest second Additional Protocol. To close any gaps left by the IHL drafters in AP II, therefore, an appeal to apply human rights guarantees in domestic conflicts was indispensable. Conversely, international human rights bodies increasingly admonished States to apply IHL in armed conflict in order to reduce human rights violations.Footnote ¹³⁶ The rapprochement between the two types of law was crowned with success when both received pride of place in the 1998 Rome Statute. Together, they occupied the entire field.

An initiative launched in 1990 to draft Fundamental Standards of Humanity intended to fill any remaining gaps in IHL and human rights foundered in 2008.Footnote ¹³⁷ The view of the ICRC, expressed during this debate, was against such new standards.Footnote ¹³⁸ It is increasingly recognized that international human rights law applies in all circumstances, meaning that it operates simultaneously with IHL in armed conflicts and beyond these also provides protection during riots, rebellions and times of peace.Footnote ¹³⁹ In sum, the relationship between IHL and human rights was a rather chilly one in the first twenty years – until 1968 – but from then onwards they were increasingly seen as complementary.

IHL and civil war

In IHL's first decades, its norms regulated international wars, not internal armed conflicts. Before 1949, proposals to also monitor domestic conflicts failed because State-centred thinking dominated. The recognition of internal conflicts as being worthy of humanitarian codification was seen as a major attack on the Westphalian system that since 1648 had sanctioned State sovereignty as its prime principle. Proponents of the view that civil wars, despite the multiple difficulties in regulating them, also needed at least some IHL guarantees recalled in vain that one of the most influential early IHL instruments, the 1863 Lieber Code, had been drafted precisely to discipline the conduct of soldiers of the Union during the American Civil War (1860–65).Footnote ¹⁴⁰ In a similar vein, early modest proposals from the United States and Cuba to explore the role of the Red Cross during civil wars or insurrections, when submitted to a Red Cross Conference in 1912, were discussed only in a procedural sense but not put to the vote.Footnote ¹⁴¹ Resolutions adopted by Red Cross Conferences in 1921 and 1938, however, had some effect: they induced the parties in ongoing civil wars in Upper Silesia and Spain to respect IHL rules.Footnote ¹⁴² Over the decades, literally every term in the heated discussions addressing domestic conflicts proved explosive: the conflict's name (civil war, rebellion or riot), the conflict's motives (liberation or subversion), the conflict's combatants (belligerents, rebels or bandits) and the name of the breaches during the conflict (atrocity crimes or disturbances of public order).

The diplomats preparing the 1949 Geneva Conventions rejected an ICRC proposal (written with the experiences of the Spanish and Greek civil wars in mind) to make the Conventions applicable to all types of armed conflict, whether international or internal.Footnote ¹⁴³ After the dramatic events of World War II, however, they were open, although hesitantly, to incorporate a single provision to regulate the humane treatment of those involved in what they termed “conflicts not of an international character”. They refused, nevertheless, to accept a proposal to add to the term three examples between parentheses (“cases of civil war, colonial conflicts or wars of religion”) on the grounds that examples would weaken the provision.Footnote ¹⁴⁴ Even stripped of examples, the approved text – which would become common Article 3 – was revolutionary. For the first time, domestic conflicts came within the purview of IHL norms.

In its turn, common Article 3 would become a launch pad for an IHL instrument on non-international armed conflicts, AP II, in 1977. Although the ambitions of AP II were dramatically downsized in the last stages of the preparatory conference on the initiative of Pakistan,Footnote ¹⁴⁵ it signified a victory of sorts after sixty-five years of attempts. But as internal conflicts had become the dominant form of warfare since 1955, common Article 3 came just in time,Footnote ¹⁴⁶ whereas AP II arrived late. In retrospect, the IHL approach to civil wars was slow, and its codification showed many lacunae. Even so, its breakthrough in 1949 and its modest coming of age in 1977 were more than worth the midwifery.

IHL and colonialism

The belated recognition of colonial conflicts as IHL objects constitutes one of the tragic moments in the history of IHL (and of humanity at large). State sovereignty threw a veil of silence over these conflicts, despite the fact that the principle of self-determination had been proclaimed in the American and French Revolutions in the late eighteenth century, that it had almost been incorporated into the Covenant of the League of Nations at the instigation of American president (and historian) Woodrow Wilson,Footnote ¹⁴⁷ and that the UN Charter eventually recognized the principle of (but not the right to) self-determination in 1945.Footnote ¹⁴⁸ In this context, it is noteworthy that the Allies had agreed in 1945 to prosecute Nazi crimes committed against Germans within the borders of Germany as “crimes against humanity” on condition only that a nexus existed between these crimes and the other crimes of the Nuremberg statute – crimes against peace or war crimes. In so doing, they cleverly avoided any application of the notion of crimes against humanity to their own conduct against their minorities or in their colonies.Footnote ¹⁴⁹ And when the diplomats drafted common Article 3 on non-international conflicts in 1949, they had successfully countered, as we saw, a proposal to insert “cases of civil war, colonial conflicts or wars of religion” into the provision on the pretext that examples would weaken it.

Portugal's attitude was typical. This country entered a reservation to common Article 3 in 1949 (and stuck to it until 1961). In this reservation, it argued that there was no accepted definition of “conflicts not of an international character”; that when the term meant “civil wars”, there was no criterion to define the moment when an armed rebellion transformed into a civil war; and that it “reserve[d] the right not to apply the provisions of Article 3, in so far as they may be contrary to … Portuguese law, in all territories subject to her sovereignty in any part of the world”.Footnote ¹⁵⁰ Portugal's arguments singled out “civil wars” in its reservation regarding non-international conflicts, but there was little risk of civil war in Salazar's dictatorship. What it really wanted to exclude from the protection of common Article 3 were conflicts in its huge colonial empire. While other colonial powers did not make similar reservations, they reasoned in much the same way.Footnote ¹⁵¹

The discussions of the 1949 Geneva Conventions among the 63 participating States, only three of them from Africa (Ethiopia, Liberia and Egypt),Footnote ¹⁵² revealed that colonial powers saw the inclusion of colonial wars into the class of non-international wars, let alone the class of international wars, as outright blasphemy. In retrospect – but only in retrospect – this view was outdated. Indeed, the self-determination principle gained strength as it was incorporated into a spate of resolutions adopted by the UN General Assembly between 1952 and 1970, two of which were declarations – the 1960 Declaration on the Granting of Independence of Colonial Countries and Peoples and the 1970 Declaration on Principles of International Law Concerning Friendly Relations and Co-operation among States. The right to self-determination was also incorporated into the two UN Human Rights Covenants of 1966 as their common Article 1. All this demonstrates that self-determination crystallized as a right and a norm of customary international law during the 1960s.Footnote ¹⁵³

The 1970 Declaration was decisive: it stated that “[t]he territory of a colony or other Non-Self-Governing Territory has, under the Charter, a status separate and distinct from the territory of the State administering it”.Footnote ¹⁵⁴ It meant that colonies and non-self-governing territories were in fact proto-States with a status distinct from the metropolis and that, therefore, the conflict between the principles of territorial integrity and self-determination could be solved. In the preparatory meetings for the 1977 Additional Protocols, then, a majority saw the struggles of peoples against colonial domination, alien occupation and racist regimes in the exercise of their right of self-determination as armed conflicts of an international, not an internal, character. Their views prevailed and became part of Article 1 of AP I.Footnote ¹⁵⁵ This provision was largely the work of the new States from the South, which had increasingly cooperated since the 1955 Bandung Conference. They were supported by the socialist States.

The tragedy is that the 1977 provision on anti-colonial struggles would have made a substantial difference in 1949, when decolonization conflicts were in full swing, but meant little in 1977, when the decolonization process was largely over.Footnote ¹⁵⁶ The opinion of legal scholar Frits Kalshoven that the 1977 provision was applicable only to the peoples of Southern Africa and Palestine was not far from the truth.Footnote ¹⁵⁷ Two tragic paradoxes were at work here. The first is that even in 1977, the characterization of anti-colonial struggles as international armed conflicts would never have been adopted if the majority of colonies had not meanwhile become independent, largely between 1945 and 1965, and recognized as States in the first place. It took the voice of former colonies to force a change in the perception of colonial wars. The second paradox is that once they had won this cherished prize, many States of the South lost their interest in transforming common Article 3 into a strong second Additional Protocol, fearing in fact that it would undermine their stability.Footnote ¹⁵⁸ This opened the way to the dramatic downsizing of AP II.