Zenawi’s Ethiopia ranks 8th in the Minority Rights Group (MRG) International’s Peoples Under Threat 2009 list. The peoples most under threat in Ethiopia include: Oromo, Afars, Anuak, Somalis and other minorities. Ethiopia’s ranking rose by 1 compared to that of 2008’s. Minority Rights Group International (MRG) focuses its work on non-dominant ethnic, religious and linguistic communities, who may not necessarily be numerical minorities. The MRG’s Peoples Under Threat ranking is specifically designed to identify the risk of genocide, mass killing or other systematic violent repression, unlike most other early warning tools, which focus on violent conflict as such. Its primary application is civilian protection.
Oromo and Ogaden nations are currently waging armed resistance struggles against the Zenawi government for self-determination rights. The Anuak nation was targeted for genocide by the ruling party (Woyane) in Ethiopia in 2003. The Oromo Support Group (OSG) has reported 3,981 extra-judicial killings and 943 disappearances of Oromo civilians suspected of supporting groups opposing the government. Scores of thousands of Oromo civilians have been imprisoned. Torture and rape of prisoners is commonplace, especially, in unofficial detention centres, often in military camps.
How is Peoples Under Threat calculated?
Since the genocide in Rwanda in 1994, our ability to identify those situations most likely to lead to genocide or mass killing has improved. A number of comparative studies of the factors preceding historic episodes of political mass killing had been undertaken since the 1970s, including by Helen Fein and Ted Robert Gurr, but it was not until the 1990s that researchers such as Rudolf Rummel and Matthew Krain pioneered quantitative longtitudinal analysis of a wide range of such factors, enabling the testing of different causal hypotheses. Rummel, for example, showed the very strong relationship between concentration of government power and state mass murder; Krain demonstrated the correlation between existing armed conflict or political instability and the onset and severity of mass killing.
Following the early work of the Clinton administration’s policy initiative on genocide early warning and prevention, Professor Barbara Harff, a senior consultant with the US State Failure Task Force, constructed and tested models of the antecedents of genocide and political mass murder and her results were published in 2003 (‘Assessing Risks of Genocide and Political Mass Murder since 1955’, American Political Science Review 97, February 2003). Her optimal model identifies six preconditions that make it possible to distinguish, with 74 per cent accuracy, between internal wars and regime collapses in the period 1955 – 1997 that did, and those that did not, lead to genocide and political mass murder (politicide). The six preconditions are: political upheaval; previous genocides or politicides; exclusionary ideology of the ruling elite; autocratic nature of the regime; minority character of the ruling elite; and low trade openness.
Minority Rights Group International has drawn on these research findings to construct the Peoples Under Threat table, although responsibility for the final table is exclusively our own. Peoples Under Threat is specifically designed to identify the risk of genocide, mass killing or other systematic violent repression, unlike most other early warning tools, which focus on violent conflict as such. Its primary application is civilian protection.
Indicators of conflict are included in the table’s construction, however, as most, although not all, episodes of mass ethnic or religious killing occur during armed conflicts. War provides the state of emergency, domestic mobilization and justification, international cover, and in some cases the military and logistic capacity, that enable massacres to be carried out. Some massacres, however, occur in peacetime, or may accompany armed conflict from its inception, presenting a problem to risk models that focus exclusively on current conflicts. In addition, severe and even violent repression of minorities may occur for years before the onset of armed conflict provides the catalyst for larger scale killing.
The statistical indicators used all relate to the state. The state is the basic unit of enquiry, rather than particular ethnic or religious groups at risk, as governments or militias connected to the government are responsible for most cases of genocidal violence. Formally, the state will reserve to itself the monopoly over the means of violence, so that where non-state actors are responsible for widespread or continued killing, it usually occurs with either the complicity of the state or in a ‘failed state’ situation where the rule of law has disintegrated. Certain characteristics at the level of the state will greatly increase the likelihood of atrocity, including habituation to illegal violence among the armed forces or police, prevailing impunity for human rights violations, official tolerance or encouragement of hate speech against particular groups, and in extreme cases, prior experience of mass killing. Egregious episodes of mass killing targeted principally at one group have also seen other groups deliberately decimated or destroyed.
However, some groups may experience higher levels of discrimination and be at greater risk than others in any given state. Minority Rights Group International has identified those groups in each state which we believe to be under most threat. (This does not mean that other groups or indeed the general population may not also be at some risk.) It should be noted that although these groups are most often minorities, in some cases ethnic or religious majorities will also be at risk and in relevant cases are therefore also listed in the table. In some cases, for example in Iraq, all the groups in the country are at risk of ethnic or sectarian killing.
One indicator that has been tested and discarded by a number of studies is the general level of ethnic or cultural diversity in a society. Krain did not find any correlation between ‘ethnic fractionalization’ and the onset of genocide or political mass killing. Similarly, neither of the patterns of ethnic diversity tested by Harff had any effect on the likelihood of mass killing (although she did find the minority character of the ruling elite to be significant). These findings are supported by research on the relationship between diversity and conflict.
The overall measure is based on a basket of ten indicators. These include indicators of democracy or good governance from the World Bank, conflict indicators from the Center for Systemic Peace and other leading global conflict research institutes, indicators of group division or elite factionalization from the Fund for Peace and the Carnegie Endowment for International Peace, the State Failure Task Force data on prior genocides and politicides, and the country credit risk classification published by the Organization for Economic Cooperation and Development (as a proxy for trade openness). For citations and further information, see the notes to the table. For a fuller discussion of the methodology, see State of the World’s Minorities 2006.
Based on current indicators from authoritative sources, Peoples Under Threat seeks to identify those groups or peoples most under threat in 2009.
Sources of the indicators are as follows:
Conflict indicators: The base data used was Monty G. Marshall, ‘Major episodes of political violence 1946–2008’ (Center for Systemic Peace, 2009) and, for self-determination conflicts, Monty G. Marshall and Ted R. Gurr, ‘Peace and conflict 2005’ (CIDCM, University of Maryland, 2005), updated for 2008 using figures from Center for Systemic Peace, MRG and the Heidelberg Institute for International Conflict Research.
Self-determination conflicts in 2008 were ranked on a scale of 0–5 as follows: 5 = ongoing armed conflict; 4 = contained armed conflict; 3 = settled armed conflict; 2 = militant politics; 1 = conventional politics. Major armed conflicts were classified as 2 = ongoing in late 2008; 1 = emerging from conflict since 2005 or ongoing conflict with deaths under 1,000.
Prior genocide or politicide: Barbara Harff, US Political Instability Task Force (formerly State
Failure Task Force). 1 = one or more episodes since 1945.
Indicators of group division: Failed States Index, Fund for Peace and the Carnegie Endowment for International Peace, 2008.
Democracy/governance indicators: Annual Governance Indicators, World Bank, 2008.
OECD country risk classification: Organisation for Economic Co-operation and Development, Country risk classifications of the participants to the arrangement on officially supported export credits’, April 2009. Where no classification is given, a value of 8 was accorded.
Indicators were rebased as necessary to give an equal weighting to the five categories above, with the exception of the prior geno-/politicide indicator. As a dichotomous variable this received a lesser weighting to avoid too great a distortion to the final ranking. Resulting values were then summed.