Title: Constitutional Design: Many Architects, No Buildings
Name: Donald L. Horowitz
Duke University
Year: 1997, 1999
Type: Research Grant

This study concerns constitutional design for severely divided societies--the political institutions apt for peaceful coexistence or interethnic accommodation of groups with a history of antipathy that must nevertheless live together in the same state. There are two main constitutional prescriptions for such societies. One, consociational democracy, relies on an elaborate set of elite agreements and guarantees of proportional quotas and group vetoes, to prevent majority domination of the minority. The other, the incentives view, aims to institutionalize reasons for politicians to behave moderately and accommodate the interests of groups other than their own. Both prescriptions have their adherents, but neither can point to many cases of adoption of coherent packages of their recommended prescriptions, with accompanying benign effects.

The study funded by the Foundation centers on the issue of adoptability. In the face of various obstacles, can either prescription be adopted in a sufficiently strong dose to do what its proponents claim for it? The research design selects cases in which the usual obstacles to the adoption of coherent packages are missing because the choice of institutions has been given over to external actors not subject to the standard constraints: Fiji (a constitutional commission chaired by a foreigner), Northern Ireland (the U.K. and Irish governments), Indonesia (a small group of foreign-trained experts), Bosnia (representatives of the international community), and Cyprus (the U.N.)

The findings are that coherent adoptions are possible but highly unusual, even when conditions should be favorable for them. Northern Ireland's Good Friday Agreement is a good example, but it was not configured by the two outside powers but by insiders in conditions so exceptional as to be sui generis. And its results thus far have been disappointing. In Fiji, Bosnia, and Indonesia, external dominance of the process did not last very long: insiders eventually took it over and negotiated compromises that had strong elements of incoherence to them. (In Cyprus, the U.N.'s consociational plan was rejected.)

In spite of the incoherent character of some of the constitutional packages adopted, benign results followed in certain cases. In Fiji, a multiethnic coalition won an election conducted under a conciliatory electoral system, despite the fact that its conciliatory features had been watered down. In Indonesia, a rather haphazard, incremental, multiyear constitutional process has produced what may turn out to be a durable democracy. But in Bosnia, international actors managed to select an electoral system that hurt the prospects of the most multiethnic Bosnian party. So incoherence is not incompatible with ethnic conciliation, but serious attention to the coherent package problem is still by no means amiss.

Bibliography: Horowitz, D. L. Constitutional Design: Proposals Versus Processes, in Andrew Reynolds, editor, The Architecture of Democracy: Constitutional Design, Conflict Management, and Democracy (Oxford University Press, 2002), pp. 15-36.
Horowitz, D. L. Domesticating Foreign Ideas in the Adoption of New Institutions: Evidence from Fiji and Indonesia, in John D. Montgomery and Nathan Glazer, editors, Sovereignty Under Challenge (New Brunswick: Transaction, 2002), pp. 197-220.
Horowitz, D. L. Explaining the Northern Ireland Agreement: The Sources of an Unlikely Constitutional Consensus, British Journal of Political Science, Vol. 32, no. 2 (April 2002), pp. 193-220.