LJIL Symposium: A Comment on Ambrus by William Schabas

[William Schabas is professor of international law at Middlesex University in London]

This article is about the fine points of how we construe a legal text adopted through a complex process of negotiation more than sixty years ago. The post-Second World War codification of genocide is notoriously narrow in scope. The reasons are relatively easy to explain.

At Nuremberg, the four ‘great powers’ had been nervous about their potential liability for crimes against humanity because of the persecution of minorities for which they were themselves responsible within their colonies or inside their own borders. For purely selfish reasons, they insisted that crimes against humanity be linked to aggressive war, failing which they would not be deemed offences at international law. In that way, Nazi atrocities against Jews within Germany could be punished as crimes under international law, while segregation and lynching in the southern United States (and similar acts attributable to Britain, France and the Soviet Union) escaped the net of international criminal liability.

When these same powers concurred in the adoption of the Genocide Convention by the United Nations General Assembly two years later, they agreed to remove the perverse nexus with armed conflict for a category of atrocity crime that they understood to be much more limited in extent than crimes against humanity. This is the definition of genocide that scholars, lawyers and judges have struggled to deconstruct over the ensuing decades.

The text adopted in 1948 provides explicitly that ‘genocide, whether committed in time of peace or in time of war, is a crime under international law’ (my emphasis).  These words were necessary in order to clarify a fundamental distinction between genocide and crimes against humanity, as the concepts were conceived of at the time. 

In 1948 the exclusion of certain acts and categories of victims from the scope of the crime of genocide contributed to impunity. Terrible acts, many of them associated with racial hatred, did not comply with the definition of genocide. Yet they could not be labelled crimes against humanity because of the lack of a nexus with aggressive war.

As a result, there was much benefit in efforts to expand the definition of the crime of genocide through judicial interpretation. Alternatively, as many urged, formal amendment of the text of the definition offered another solution to the problem. Finally, though, the difficulty was solved in another way. Instead of altering the definition of genocide, the international legislator enlarged the notion of crimes against humanity. This was posited in the Tadic Jurisdictional Decision of 2 October 1995 and confirmed at Rome in 1998. During negotiations of the Statute, there were only a few isolated attempts to amend the definition of genocide applicable to the International Criminal Court. These were resoundingly dismissed. The expansion of crimes against humanity to cover peacetime atrocity meant there was no real need to repair the alleged blind spots in the definition of genocide. For similar reasons, international judges have generally resisted attempts at interpretative enlargement of the definition.

This article by Monika Ambrus argues for an approach to construction that may slightly expand the definition of genocide. It does not contend that ethnic cleansing or similar acts of persecution falling short of genocide should fall within the ambit of the crime. Nor does it argue that the genocide label be attached to attacks on groups other than the four classic categories: national, ethnic, racial and religious. It is really only addressed to individual victims on the fringes of acts that unquestionably fall within the scope of the crime.

But isn’t this a bit like arguing that when a rapist attacks the bystander who attempts to rescue the victim, he also rapes the bystander? In any functional justice system, both the rape itself and the assault on the good Samaritan will be punishable. There is no need to convert the assault on the bystander into a rape so that it can be dealt with by criminal justice.

The poet Philip Larkin famously wrote about a time ‘between the end of the Chatterley ban and the Beatles’ first LP’. So it is with crimes against humanity in peacetime, which began at some point between Nuremberg and Tadic. Unlike the rapist who also attacks the bystander and is convicted of assault as a result, in the past the génocidaire might avoid accountability at the international level for attacks on individuals who were not clearly members of the victimized national, ethnic, racial or religious group. These victims did not seem to meet the terms of article 2 of the Convention, which required ‘killing members of the group’. And they might not be liable for crimes against humanity because of the nexus with aggressive war. The impunity gap no longer exists.

Debating the precise boundaries of genocide with respect to individual victims now seems a rather sterile exercise. It is not really relevant to genuinely important controversies such as whether situations like Darfur, Cambodia or the war between Croatia and Serbia should be described as genocide. It will not impact upon the jurisdiction of the International Criminal Court. It only means that in cases like Nahimana, the offender will be convicted of crimes against humanity or perhaps war crimes rather than genocide for the killing of an individual victim. Is this really a problem that requires as solution? Is there any real need to broaden ever so slightly the notion of genocide as it has been interpreted by the courts? Some may find the theoretical issues to be intriguing but the practical consequences seem unimportant.