Africa: The Limits of Power-Sharing

Publisher Website
Google Scholar
Abstract
Journal of Democracy 13.3 (2002) 123-136 An examination of recent high-profile peace processes in Sierra Leone, Angola, and Rwanda suggests that power-sharing is a surprisingly unstable form of government that, even at the best of times, provides only a short-term reprieve from violent conflict. Other than as transitional remedies, power-sharing agreements are virtually unworkable. As I have argued elsewhere, in the aftermath of civil war, power-sharing agreements are difficult to arrive at, are even more difficult to put into practice, and when implemented rarely stand the test of time. 1 Indeed, the problem with power-sharing is even more fundamental: It does not resolve conflict but instead may only temporarily displace it or disguise disputants' more malevolent intentions. There is much that is intuitively appealing about power-sharing, and it is no surprise that it is repeatedly proposed as a form of postconflict governance. Since each group is given a slice of power and access to state resources, disputants should find, at least in theory, less to fight about. Moreover, since recent efforts at power-sharing have also included provisions that allow each group to contribute its own troops to an integrated military, a semblance of group security is sustained. Indeed, former South African president F.W. de Klerk noted in his memoirs that his attachment to power-sharing was not "out of step with constitutional theory relating to the maintenance of democracy in divided societies," and he cites Vernon Bogdanor's claim that "some kind of power-sharing has been a feature of government in all societies that have successfully overcome their internal divisions." 2 Scholars have also argued that power-sharing is the only means of managing deeply divided societies. In the aftermath of the 1994 Rwandan genocide, Crawford Young noted that the "haunting question remains: whether [End Page 123] ethnocide as a mode of politics can ever be supplanted by any political order which does not include power-sharing buttressed by a liberalized politics." 3 Because power-sharing would appear to satisfy the desire of civil war disputants for political and economic power, it also has an appeal to the international community insofar as it reduces the need for a substantive international commitment. In many African conflicts, neither disputant can impose an outright military defeat on the other, yet the parties are often capable of continuing violence. Few outsiders, however, are willing to provide the long-term security guarantees that some analysts argue are necessary to enforce peace when it is attained. By default, power-sharing becomes the only option for a commitment-averse international community because it offers a logically attractive approach to conflict management. Amid a growing literature emphasizing inclusive governance in multiethnic societies, two early works continue to stand out: Arthur Lewis's Politics in West Africa and Arend Lijphart's Democracy in Plural Societies. Lewis provides the most eloquent justification for the need to share power in ethnically plural societies. 4 He identifies two meanings of democracy: 1) Everyone affected by a decision should have the opportunity to participate in making that decision, either directly or through representatives; and 2) the preferences of the majority should prevail. Lewis also identifies two types of society—a class society and an ethnically plural society—and argues that the form of democracy must be appropriate to the type of society. For example, a competitive party system is appropriate to a class-based society in that it allows the larger poor and middle classes the opportunity to avoid domination by a smaller wealthy elite. Such a system, however, is entirely unsuitable for an ethnically plural society because, by enabling those with the largest number of votes to win, it necessarily excludes or wastes the votes of the losers. If a losing group will never attain a plurality, its votes will never influence political decision making in an environment where electors consistently vote along ethnic lines. For nations that are attempting to forge a broader national loyalty, Lewis argues, such...

This publication has 4 references indexed in Scilit: