Civil wars are a feature of the modern political landscape and significant attention has been given to the increase in this type of conflict in recent years. This discussion centres on those civil wars that are primarily political in nature - conflicts that concern the supreme executive power in a given state. Although civil wars can have obvious winners and losers militarily, the judgment of international actors can often be decisive. The purpose of this paper is to examine the way in which international actors decide who holds the sovereign authority of a state during civil wars and what is needed to 'win' such wars in the eyes of the international community. Existing theories stress effective control over territory or the political interests of the recognising states. However, my hypothesis is that states observe a rule of recognition that equates control of the capital city with possession of a state's sovereign authority. I examine four civil wars (Chad, Somalia, Zaire/Democratic Republic of Congo, and Liberia) in order to test these three theories and in all four cases control of the capital city is shown to be necessary for recognition by other governments. Effective control and political expediency demonstrate less explanatory power. The second part of the paper investigates the possible reasons why capital cities should be so significant in civil wars and considers the arguments for the special circumstances of the African state and for the economic significance of capital cities. However, it is the symbolic value of capital cities and, more importantly, the long-standing perspective in military history that views capitals as political and territorial proxies for states, that explain the recognition pattern in civil wars. The conclusion argues that the merits of this practice of recognition should be debated, as should the 'de-certification' of capital cities as a way of changing the incentive structure of some civil wars.
Occasional Paper No. 6, London, UK; Crisis States Research Centre, 26 pp.