The Cambodian ‘auto-genocide’ – musings on a concept that needn’t exist

Timothy Williams

Early on in my research stay I was at a conference organised by the Heinrich Böll Foundation, and I noticed there was one theme that was brought up repeatedly by Cambodian participants: how particularly terrible the Cambodian genocide was compared with others, because it was Khmer on Khmer violence, people killing their own. This is an argument, or maybe better said a statement, which I have heard many times again since, both at conferences and in interviews for my fieldwork. 

My first thought was how absurd it is to suggest that it should be seen as a lesser evil to mass kill people of a different ethnicity than to mass kill one’s own. But in the comments of these people one could hear the pain of the intimacy of these killings, the suffering elicited by the fact that these were their brethren, the people they would have expected to protect them from external aggressors, and how this violated the most fundamental assumptions of protection and safety. This tension between ideas of moral equality and subjective terribleness regarding the mass killing of one’s own opens up for debate what is particular or different about the nature of the Khmer Rouge killings compared with others. A term frequently used in the media to describe this phenomenon is auto-genocide, genocide of the self, and in this blog post I would like to explore and discuss this concept a little. 

While the Khmer Rouge did kill some ethnic minorities, notably Muslim Cham and ethnic Vietnamese, most of their victims were of Khmer ethnicity. More precisely, most victims actually starved to death or died of diseases, which spread like wildfire due to the horrific living conditions and lack of medical infrastructure. But of the people actually actively killed most were ethnic Khmer, the same ethnicity of the Khmer Rouge as their name suggests. So at first view, the term auto-genocide doesn’t appear too bad – the Khmer killing other Khmer. But let’s take a closer look – first at the term genocide and then at whether it is ‘auto’ or not.

Genocide is legally defined by the UN Convention of 1948 as “any of the following acts committed with intent to destroy, in whole or in part, a national, ethnical, racial or religious group, as such: (a) Killing members of the group; (b) Causing serious bodily or mental harm to members of the group; (c) Deliberately inflicting on the group conditions of life calculated to bring about its physical destruction in whole or in part; (d) Imposing measures intended to prevent births within the group; (e) Forcibly transferring children of the group to another group.”

For a number of reasons many social science academics find the definition problematic, including the quite exclusive selection of groups which are included in the definition, for instance excluding political, economic, social, or gender-based groups. Conversely, the five acts appear as quite expansive, broadening the crime of genocide from an act of mass killing, according to some scholars. Further discussion surrounding the UN definition focuses on what the clause ‘in whole or in part’ should actually mean in practice, often leading to a numbers game (that is how many people need to  have died before we call it a genocide), which is little helpful. 

There are many more critiques, too plentiful to go into here, but with these shortcomings, a plethora of definitions have sprouted all over the garden of the ivory tower. The definition I myself use is genocide as a form of one-sided and indiscriminate mass killing in which an authority intends to destroy a group in whole or in part, as that group and membership in it are defined by the perpetrator. I will spare you listing the merits of this definition, and get back with haste to the matter of interest here: auto-genocide. According to this definition then, auto-genocide would be this mass killing targeting the group which the perpetrators see as their own and trying to eliminate their entire group. So how does Cambodia fare under this definition?

Using this definition, it seems clear that the killing of Khmer during Democratic Kampuchea (as the Khmer Rouge regime was called) constitutes genocide. However, there must also be a debate on what constitutes this group as defined by the perpetrator and this is also important for evaluating the ‘auto’ part of the contested term. A key part of this definition of genocide is that people are killed because of the group in which they are a member, irrespective of their own personal attributes. 

If we rewind for a second and look at the legal definition, we would have to say that with this ethnic conception of genocide that genocide did not occur, much less auto-genocide, as that would mean that the Khmer Rouge indiscriminately killed Khmer because of them being Khmer (we are not talking about the ethnic minorities here, just the Khmer majority). Instead, they killed specific people who were against the Khmer Revolution, people they felt to be enemies of the people and people who were traitors to any of their ideals. This could be people who worked for the previous regime, intellectuals, people suspected of cooperating with the CIA or KGB. This group was defined politically by the Khmer Rouge. Thus the Khmer Rouge certainly didn’t see them as Khmer when they killed them, in fact they were explicitly robbed of their Khmer national identity and were said to be Khmer bodies, but Vietnamese heads. These people were placed outside the group of Khmer and thus were seen as a threat and a poison to society, thus necessitating that they be killed. 

But here, although the legal definition does not allow a classification as genocide, there is more to these killings than individual targeting and the non-legal, academic definition of genocide does fit more readily. The killing was not picking out specific individuals (which would mean it were not genocide), but occurred indiscriminately and in nigh totality within this group of enemies of the revolution as defined by the Khmer Rouge. Yes, the people who were arrested and killed as enemies of the people was from an outside perspective often arbitrary and dictated by local personal agendas, but from the perspective of the perpetrators (which is the key perspective for genocide as we have defined it here) it was a clearly demarcated group of individuals who could be identified and then subsequently killed. 

Altogether then it makes little sense to class this killing an auto-genocide. First, by legal standards, this killing of Khmer was not genocide. Second, adhering to the alternative definition, it is not an auto-genocide because from the perspective of the perpetrators they were not killing their own (the auto), but quite decidedly another group. For genocide to be present we need indiscriminate selection of victims within the victim group, otherwise it’s not genocide. And we have this according to my definition of genocide as any perceived traitor could be killed, but not according to the idea of auto-genocide as only specific Khmer were targeted.

What does this mean for the Cambodian victims sitting in conferences such as that of the Heinrich Böll Foundation that I participated in? In all honesty, probably not much. But I hope that this article has shed some light on the dynamics of the mass violence of the late 70s in Cambodia, and that it has become clear that the idea of an auto-genocide is not particularly helpful in describing it. From today’s perspective, the survivors feel strongly that it was their own killing their own, but back during Democratic Kampuchea this was not how the perpetrators saw it. This does not make the killing any less tragic, any less horrific.

Timothy Williams is a research fellow at the Centre for Conflict Studies at Marburg University and is pursuing a PhD at the Free University Berlin on the question of why people participate in genocide, funded by a scholarship of the Heinrich Böll Foundation; his doctoral thesis conceptually synthesizes motivations of perpetrators and tests the model empirically through field research in Cambodia and Armenia. He was recently awarded the Raphael Lemkin Fellowship 2015 by the Armenian Genocide Museum & Institute. He studied at Mannheim University (BA Political Science) and at the London School of Economics (MSc Comparative Politics) and is the Executive Director of the web platform