Major Episodes of Political Violence
1946-2006

The following table lists 324 episodes of armed conflict (including 28 ongoing cases) that comprise a comprehensive accounting of all forms of major armed conflicts in the world over the contemporary period: 1946-2006. Cases highlighted in red were ongoing at the end of 2006 (24 cases); ongoing cases highlighted in orange have diminished substantially in magnitude during the last six months of the most recent year (2006) and may be ending (4 cases). All episodes that are denoted as having ended within the past three years (i.e., since January 2004; cases highlighted in yellow are considered at high risk of return to warfare (10 cases).

Note that several revisions were made to the list in regard to episodes in Africa. A thorough re-examination of the case list for African countries was conducted under the auspices of a Summer 2005 contract with the UK Government's Department for International Development (DFID) and the African Conflict Prevention Pool for a study and report on "Conflict Trends in Africa." The final report was delivered in October 2005; a full copy of the report, including seven data annexes, is available on the CSP Web site. Click here to view or download the report in PDF format.

The variables listed in the "Major Episodes of Political Violence" table are as follows:

Inclusive years (Begin and End): The beginnings and endings of most political violence episodes are difficult to determine exactly; various researchers "pinpoint" and denote various dates. The "begin" and "end" years listed for each episode (below) are those considered by the author to be those most likely to capture the transformative "moments" (beginning and ending) of the episodes, according to a comparison of the varying claims of the sources noted. No "end" year is listed for episodes that began and ended in the same year.

Episode type (Type): Episode type is listed according to two character codes. The first character denotes either a (C)ivil-intrastate involving rival political groups; (E)thnic-intrastate involving the state agent and a distinct ethnic group; or (I)nternational event-interstate, usually two or more states, but may denote a distinct polity resisting foreign domination (colonialism). The second character connotes either an episode of (V)iolence-the use of instrumental violence without necessarily exclusive goals; (W)ar-violence between distinct, exclusive groups with the intent to impose a unilateral result to the contention; or i(N)dependence-an attempt to forcibly remove an existing foreign domination.

Magnitude of societal-systemic impact (Mag): The rationale and methodology for assessing the societal and systemic impact of warfare episodes is discussed and described in detail in the accompanying text. The number listed represents a scaled indicator of the destructive impact, or magnitude, of the violent episode on the directly-affected society or societies on a scale of 1 (smallest) to 10 (greatest). Magnitude scores reflect multiple factors including state capabilities, interactive intensity (means and goals), area and scope of death and destruction, population displacement, and episode duration. Scores are considered to be consistently assigned (i.e., comparable) across episode types and for all states directly involved. For a more detailed explanation of the coding scheme used, click here.

Episode location (States Directly Involved): Countries listed are only those upon whose territory the political violence episode actually takes place, that is, those state-societies directly affected by the warfare. Countries intervening in the episodes are not listed as the violence does not take place on their territory and, so, these intervening actors are considered to be indirectly, or remotely, affected by the violence.

Estimates of "directly-related" deaths (Deaths): Accountings of the number of deaths resulting directly from an episode of political violence are difficult to determine and estimates often vary widely. This difficulty is especially problematic as the distinction between combatants and non-combatants has grown increasingly obscure as "less formal" civil conflict interactions in less institutionalized societal systems predominate in the contemporary era. As argued in the text, such estimates of "battle-related deaths" should be regarded simply as estimates of the general magnitude of the violence. The numbers listed here reflect the median or mean of often widely disparate estimates listed in the various sources and are provided solely as a referent point. Casualties among non-combatants directly related to the violent conflict are inconsistently estimated (if at all) in the various source estimates. Far more problematic than "battle-related deaths" for societal systems are the much larger numbers of persons directly and indirectly, physically and psychologically, distorted and disturbed by violence during episodes of armed conflict (for this we have no estimation procedure).