Constitution writing in post-conflict settings: an overview.

• Author: Widner, Jennifer

During the past forty years, over 200 new constitutions have emerged in countries at risk of internal violence. Internationally brokered peace accords have entailed the development of constitutions not only in the Balkans but also in Cambodia, Lebanon, East Timor, Rwanda, Chad, Mozambique, Bougainville-Papua New Guinea, Nepal, the Comoros, and other places. (1) New constitutions have heralded the adoption of multiparty systems from Albania to Zambia. (2)

Policymakers have started to ask what we have learned and specifically whether some constitutional reform processes are more likely than others to deliver a reduction in violence or more rights-respecting fundamental documents. For example, over the past decade, the Commonwealth, the U.S. Institute of Peace, and the non-governmental organization International Institute for Democracy and Electoral Assistance (IDEA) have worked to develop good practice guidelines for the conduct of constitution writing. (3) Does the type of deliberative forum make a difference? Do better results emanate from elected constituent assemblies than from unelected bodies? Does the choice of decision rules shape the regard for a broader range of interests?

This challenge is difficult. Our instincts tell us that process makes a difference. Constitution writing has sometimes inflamed passions and sparked violence, as it did in the Solomon Islands, Iraq, Chad, and the Republic of the Congo, for example. (4) It has produced better than expected results in some other countries, including South Africa. (5) It is devilishly difficult to show, empirically, that procedures made the difference in these cases, however. A number of very serious analytical problems hamper the ability to give a social science answer to the question policymakers have asked. Mark Tushnet is right to wave warning flags. (6) Nonetheless, there may be some paths forward.

My primary intention here is to offer a description of the range of procedures currently in use and the "results," very narrowly defined, associated with these procedures. This overview draws on an original dataset, (7) as well as on conversations that took place among constitution drafters and scholars under the auspices of Princeton University, Interpeace, and International IDEA in May 2007. (8) It serves as a preface to some of the other contributions in this issue. Part I probes some of the expectations one might have about the effects of process on outcomes. Part II defines what drafters mean by "process" and offers a quick, general description of recent trends in the choice of procedure. Part III explores some of the patterns in the data. Part IV offers an agenda for research and discussion of constitution writing and conflict resolution.

1. EXPECTATIONS

   High hopes often attend efforts to write new constitutions. "Success" has many dimensions. A common aspiration includes the achievement of a durable agreement, an arrangement that will not be disregarded or suspended lightly and within a short period. More immediately and perhaps more importantly, people often hope for a reduction in violence and an increase in civility. The degree to which a constitution or a constitution-writing process displaces conflict from the streets and into institutions is an important measure of success. Said one participant in a conference at Princeton University in May 2007, "a successful process is transformational; it converts the spoilers." (9) The people most able to cause violence accept the basic terms and are willing to process disagreements in constitutionally acceptable ways. Their orientation toward political institutions and toward law changes in the course of negotiations.

   Success may have other dimensions as well. It may pertain to the choice of terms in the document itself. Order is not all that matters in today's world. Historically, constitutions often developed as agreements about how to design government so that the sovereign could not abuse citizens, especially those who had to foot the bill, in money or lives, of foreign misadventures and lavishness at home. (10) A process that produces a document that aggregates all power in the executive would be hard to countenance today. The future ability of citizens to participate in the affairs of their own countries--to act as members of a political community--matters too. And of course, where people feel excluded or where they can be thrown in jail for no good reason, grievances may also fester and breed violence. By both logics, a constitution that protects individual and civil liberties might be deemed more "successful" than one that does not.

   A third set of ambitions focuses on the degree to which a constitution can become self-enforcing. Public knowledge or awareness is likely to contribute to this aim, because people who are better armed with information about principles and institutions are more likely to police their governors than those who know little. Constitution-building processes that yield greater public awareness of government institutions and of the basic principles that lie behind those institutions are arguably more successful than others. Even where drafts fail at referendum, many consider that conditions have improved if public awareness is a product. (11) Whether a country has implemented the terms of a new constitution five years out from ratification may also capture an important dimension of the success of a process.

   Policymakers associated with the initiatives launched by the U.S. Institute of Peace and others expect that quite apart from the influence of the actual terms selected, the steps taken to draft a constitution matter for these three types of outcomes. (12) Those "steps" refer to the complex bundle of procedural rules that governs deliberation. They include incentives created to convince delegates to take the long view and to eschew short-term personal advantage; rules governing representativeness of deliberative bodies; decision rules used to regulate inclusion of passages (not just the voting rules, but also the form in which amendments may be introduced, the points at which votes are taken, etc.); ratification rules; and opportunities for public participation.

   Specifically, how might procedural choice matter? First, several aspects of process matter for the degree to which delegates may be willing to put aside short-term personal or partisan advantage and consider the long-term welfare of the broader political community. Veil rules increase any given delegate's uncertainty about his future position or the future position of his constituents. (13) They introduce prospectivity, generality, and durability into decision making. (14) That is, they structure deliberation in a way that increases the likelihood that a broader range of interests will be considered. One might include in this category rules that reduce the dominance of current interests by setting limits on eligibility, by forcing recusal of some kinds of incumbents, such as the military, or by restricting conflicts of interest in other ways, especially through mediation. Another example of a veil rule is the incorporation of a delay between the ratification of a new constitution and first elections under the new rules. Some countries hold elections immediately, while others stipulate that there will be a delay of eighteen months or longer. (15) A few countries have simply banned negotiators from running in the first set of post-ratification elections, spawning another set of difficulties. Informal practices may help promote a "long view" too. Informal bilateral contacts may build trust and obligation, making resort to strongly partisan positions more difficult. These arguments about procedure focus on incentives and their psychological impact.

   Representativeness is a function of the choice of forum or "reform model," as well as delegate eligibility and selection rules. (16) We generally think that elected constituent assemblies or legislatures will be more representative than other types of forums and ought to produce terms that are more "other-regarding" as well as constitutions that enjoy more public support and endure. Much depends, however, on the electoral rules put in place, the willingness of people to exercise their right to vote, social biases, and the degree to which people cast ballots to suit patrons. A number of countries have argued that reserving some seats for people who represent regions, women, youth, important occupational groupings or sectors, and other types of social groups might prove more representative than a purely elected assembly. (17) Further, where violence prevails or where there is no infrastructure for elections, a method of appointment that is not controlled by the incumbents may yield an assembly that the public is more likely to trust. Uganda is an example of a country that has employed a hybrid system. (18) These arguments about procedure focus on the isomorphism between popular conceptions of "representativeness" and the rules of delegate selection chosen.

   Procedural choices may also affect behavior on the floor of an assembly. For example, rules that lock delegates into positions or encourage public campaigning for subsequent political office are generally counterproductive. An example of this kind of claim is Jon Elster's proposition that highly public processes, in which negotiators deliberate in front of the media or audiences, promote grandstanding and are less friendly to compromise. (19) David Stasavage and others have picked up on this theme, though not with respect to constitution drafting. (20) Many constitution-writing processes have closed committee deliberations to the media for this reason, allowing coverage only of the occasional plenary session, although they have often reported extensively on the outcome of committee deliberations.

   The public's ability to monitor respect for the constitution and to punish those who infringe its terms is partly...