Since the end of the Cold War, the international community has engaged in two tasks that run contrary to traditional notions of state sovereignty. The first is the reconstruction of domestic political institutions in states emerging from civil war. The second is the promotion of liberal democracy

as the preferred form of national governance. Whether one understands state sovereignty in territorial or functional terms, both tasks break new ground. At the heart of most Westphalian conceptions of state autonomy is the capacity for self-government. International law prior to the end of the Cold War was highly protective of how states selected their leaders and designed their constitutional systems. As a respected international law treatise stated in 1905, "The Law of Nations prescribes no rules as regards the kind of head a State may have. Every State is, naturally, independent regarding this point, possessing the faculty of adopting any Constitution according to its discretion." (1) It is commonplace that state sovereignty has never been absolute, either in theory or practice; however, for the international community to involve itself in matters of internal governance--let alone to specify a model of liberal democracy as the normative ideal--intrudes upon functions that even the most permeable conceptions of sovereignty have regarded as wholly domestic.

These assumptions have now changed. External actors, led primarily by the United Nations, have become negotiators of peace agreements to end civil wars; drafters of new constitutions, electoral laws, and judicial procedures; and supervisors of all aspects of postwar transitions. These postconflict reconstruction or peacebuilding missions (the terms are used interchangeably) have taken place on every continent. They have become so common that, at the end of the twentieth century, it was the rare civil conflict that ended without the UN being assigned an important role in the state's reconstruction. (2) A UN panel has defined peacebuilding as "activities undertaken on the far side of conflict to reassemble the foundations of peace and provide the tools for building on those foundations something that is more than just the absence of war." (3) Between 1991 and 2002, seventeen UN peacebuilding missions were deployed. (4)

During the same period, international law has increasingly identified liberal democracy as the preferred form of national governance. (5) Scholars have identified an emerging "democratic entitlement" in international law--the view that representative government, chosen in fair and periodic elections, is a human right of all citizens. (6) The state practice underpinning this emerging norm is widespread and diverse: resolutions of the Security Council urging free and fair elections; clarification of the treaty-based right to political participation by international human rights bodies; refusal of states and international organizations to recognize regimes taking power by extraconstitutional means; the conditioning of foreign aid on comportment with "democratic" norms; and the conditioning of treaty relations on the maintenance of democratic institutions. These and other similar acts are the raw data from which norms of customary international law emerge.

Postconflict reconstruction initiatives bring together these two normative strains of reconstructing state institutions after conflict and democracy promotion. When the UN has contributed to the political and legal structure of postconflict societies, it has invariably drawn on the body of emerging democratic norms. While this cross-pollination is not often explicit, the institutions and procedures it has recommended for postconflict states find direct parallels in the body of human rights norms concerning political participation. On issues ranging from universal suffrage to party pluralism to media access for opposition groups, international norms and the practice of democracy promotion missions are remarkably congruent.

This article focuses on the role of international law in postconflict reconstruction. An explicitly legal perspective introduces a new voice in the heated debate over the desirability and efficacy of democracy-promotion missions in postconflict societies. Critics have focused on the numerous obstacles encountered by the missions: ongoing ethnic conflict, the frequent lack of democratic traditions, the phoenix-like reappearance of political leaders and groups responsible for the recently ended conflicts, and growing disinclination on the part of developed countries to fund all aspects of the reconstruction project. In their view, such daunting obstacles have transformed the emerging democratic norms from useful new tools into hopelessly utopian aspirations. They suggest an array of alternative strategies, ranging from nondemocratic forms of governance on an interim basis to acquiescence, to hegemony by dominant ethnic groups, to outright partition of states.

A focus on international law suggests two responses to these critics. First, the scaled-back alternatives they advocate cannot be evaluated as mere policy alternatives but must be seen as challenges to norms that are becoming increasingly entrenched in multilateral practice. Second, viewing postconflict democratization as normative enterprise suggests that the missions' successes (or lack thereof) should be measured in terms appropriate to an international legal initiative. In this article, I suggest that the use of inappropriate yardsticks for assessing the missions' outcomes has contributed to the negative view of the record to date.

Although postconflict missions involve an increasingly diverse array of international actors, I will focus here on the role of the UN. This is not to suggest that regional organizations, nongovernmental organizations (NGOs), international lending institutions, bilateral overseas development assistance (ODA) agencies and other actors have not played important roles in postconflict reconstruction. But in examining the legal implications of democratization efforts, there are several reasons why the UN stands in a special category. First, all multilateral postconflict missions were authorized by the Security Council, which also promulgated their mandates. Second, during these missions the UN assumed primary responsibility for organizing, monitoring, and adjudging elections, which for better or for worse have emerged as the centerpiece of multilateral democratization initiatives. Third, critics of the missions have focused much of their attention on UN decisionmaking. Hence, it seems appropriate to meet these crit ics on their own terms by asking whether they are evaluating UN actions properly.

Finally, as a treaty-based organization whose charter describes it as a "center for harmonizing the actions of nations in the attainment of...common ends," (7) the UN is the actor most clearly constrained by rules of international law. This is not clear for other actors involved in postconflict tasks. For instance, NGOs arguably lack international legal personality (the capacity to invoke and be bound by international norms), while bilateral ODA agencies engaged in collective, voluntary initiatives outside their national territories also may not be constrained by international law. But little ambiguity exists for the UN. It is a creature of international law as well as the institutional progenitor of many of the human rights norms underlying democracy-promotion efforts. As such, it can only with difficulty ignore the requirements set forth by these laws.

The International Legal Framework for Popular Participation

A legal perspective begins with an examination of the normative roots of democracy promotion. International human rights law draws an important connection between participation in politics and the legitimate assertion of governmental power. The Universal Declaration of Human Rights, in a clause unlike that relating to any other right, asserts that "the will of the people shall be the basis of the authority of government." (8) This principle of democratic legitimacy has been repeated by all major regional organizations, (9) though Asia and the Middle East lack multilateral human rights organizations and so are largely absent from this practice. The member states of the Council on Security and Cooperation in Europe (now the OSCE) declared in the 1990 Charter of Paris that they "undertake to build, consolidate and strengthen democracy as the only system of government of our nations." (10) The Organization of African Unity's Commission on Human Rights declared in a resolution on the military that "the best govern ment is one elected by, and accountable to the people. (11) Thirty-four states of the Americas have declared through the Organization of American States (OAS) that "representative democracy is indispensable for the stability, peace, and development of the region, and that one of the purposes of the OAS is to promote and consolidate representative democracy." (12)

Statements by two UN secretaries-general underscore the shift in multilateral attitudes toward democratization. In a 1950 report on accrediting delegations to the General Assembly, Trygve Lie wrote, "The United Nations is not an association limited to like-minded States and governments of similar ideological persuasion. . . As an Organization which aspires to universality, it must of necessity include States of varying and even conflicting ideologies." (13) Almost fifty years later, Boutros Boutros-Ghali had wholly abandoned such agnosticism toward regime type. He observed that "democracy is increasingly being recognized as a response to a wide range of human concerns and as essential to the protection of human rights" and found it appropriate that the UN was engaged in "fostering a culture of democracy and holding democratic elections," as well as filling "the evident and crucial need to prepare, and to continually renew and strengthen, the institutional ground in which democratization can take shape." (14)

Michael Reisman has taken notions of democratic legitimacy a step farther, describing the democratic entitlement as not only prescribing popular sovereignty within states but incorporating the principle into their external relations as well. For Reisman, a unilateral intervention to return a deposed elected leader to office is not a violation of the invaded state's sovereignty but an affirmation of its sovereignty: 'International law still protects sovereignty but--not surprisingly--it is the people's sovereignty rather than the sovereign's sovereignty." (15)

No treaty or other legal source takes the democratic entitlement this far. But the possibility that a democratic legitimacy norm might, in whole or part, circumvent a state's legal protection against external intervention underscores the profound challenge it poses to existing international law. While few unilateral claims of prodemocratic intervention have been made, the Security Council, acting under Chapter VII of the UN Charter, has twice authorized the use of military force to return elected regimes to power--in Haiti in 1994 and Sierra Leone in 1998. (16) Notions of democratic legitimacy also underlie policies adopted by a variety of international organizations that states in which governments take power by extra-constitutional means will be either suspended or expelled from the organization. Such a principle was behind the Commonwealth's decision on March 19, 2002, to suspend Zimbabwe after an investigation determined that "the conditions in Zimbabwe did not adequately allow for a free expression of wi ll by the electors." (17)

Efforts to enforce democratic norms have generally been limited to moments of crisis or transition, in which the authority of an elected government is clearly challenged. There have been few, if any, collective efforts to oust long-standing nondemocratic regimes from power. The reasons are clearly practical and rooted in the impossibility of carrying on international relations if some or all "nondemocratic" governments in the world were deemed illegitimate. This is especially true for the UN, which could not expect its member states to fulfill obligations to the organization if the governments in effective control of national territory were not recognized as the legitimate representatives of the states.

Nonetheless, such transition points are frequent enough that much attention has been given to delineating the procedures required for "free and fair" elections. These procedures are described in the major human rights treaties, as well as in the jurisprudence of adjudicatory bodies empowered to interpret those treaties. (18) Specification of the norm is important, for it signals an end to the UN's platitudinous approach to democracy issues and the emergence of a nascent consensus on the particulars of electoral procedure. Several core requirements are evident:

* Suffrage must be universal and equal.

* Voting must be secret.

* Elections must be held at reasonable, periodic intervals.

* There must be no discrimination against voters, candidates, or parties based on ethnicity, race, religion, or political opinion. (19)

There are limits to explication of the norm by multilateral institutions. International human rights law does not address every aspect of electoral design and implementation, for human rights treaties are not intended to micromanage domestic politics. State parties are granted an often substantial "margin of appreciation" to implement their obligations in any manner that is broadly consistent with the treaty undertakings. Thus, in the case of elections, the European Commission on Human Rights has upheld challenges to winner-take-all systems by proponents of proportional representation; the disenfranchisement of convicted prisoners, including imprisoned conscientious objectors; minimum signature requirements for a party to appear on a ballot; minimum levels of electoral success for parties to obtain state subsidies; and prohibitions on members of one legislative body standing for election to another.20 In the commission's view, these restrictions on the right to political participation are not arbitrary or dis proportionate to achieving legitimate policy goals, nor do they thwart "the free expression of opinion of the people in the choice of legislature." (21) The democratic entitlement simply does not operate at this level of detail.

International law also protects a wide variety of other rights usually considered essential to effective citizen participation in public life. These include rights to free expression and opinion, freedom of assembly, freedom of association, freedom of travel, and equal treatment under the law. (22) Each of these rights is essential to modes of participation that are independent of elections: lobbying efforts, grassroots organizing by civil society organizations, union activities, independent data collection and policy analysis, and educational activities. Many suggest that these less formalized means of citizen engagement are essential to maintaining a vibrant democratic culture. (23) But participation by these means has not been the center of UN-led postconflict reconstruction. Rather, the focus has been on elections.

At first this approach might seem rather shortsighted, as well as overly formalistic in its conception of how political society functions, particularly in developing countries. (24) Ensuring other civil and political rights--familiar from political theory as both assisting citizen participation in politics and restraining excesses of the majority--creates a deeper and richer form of citizen participation, both in the short term by ensuring fairer elections and in the long term by beginning to entrench liberal notions of pluralism. The UN has recently attempted to broaden its conception of the democracy being promoted, establishing "peacebuilding" offices that stay on after the initial postconflict mission has completed its work. But these offices have limited mandates and even more limited capacities to promote "democratic" reforms. (25) The postconflict missions themselves have been compelled to work with limited political and financial resources that radically constrain their ability to engage in nation bui lding broadly conceived. This has given rise to a number of justifications for the focus on elections.

First, as discrete events with well-defined procedural requirements, elections are a relatively easy means for international organizations of limited means to verify a democratic path for the country. Second, more broadly, competitive elections are at the institutional core of all modem democratic systems, irrespective of any additional substantive rights or procedural mechanisms for empowerment that may be provided. Especially in societies with little recent experience with democratic governance, ensuring the fairness of postconflict elections may be essential to a successful transition. Third, elections generally capture the attention of the entire population and can serve as invaluable educational tools, not only by airing different perspectives on issues, but in helping citizens understand that a free and open discussion on matters of national concern is the norm in a democratic society. Indigenous nongovernmental organizations, demanding a wide range of reforms, have often emerged from UN-sponsored elect ions. Fourth, a government's conduct during elections--when many of its agencies have responsibility for ensuring respect for civil and political rights--is generally taken as a useful indicator of the regime's willingness to respect human rights in other settings. Finally, elections often focus international attention on states whose human rights record might otherwise be of little concern to Western policymakers or publics. From a purely instrumental perspective, in an era of diminishing Western engagement in the developing world, international organizations might be seen as squandering valuable opportunities to bring effective pressure to bear on repressive government if they were to undervalue elections.

Implementation Through UN-Led Postconflict Reconstruction Missions

The modern era of UN postconflict reconstruction--and its concomitant efforts to foster popular participation and democratic reforms--began with the Namibia mission (UNTAG) in 1989. Although the UN already had a long and relatively successful history of monitoring plebiscites and inaugural elections in colonial territories on the eve of independence, it had little experience with issues of governance in sovereign states. Since UNTAG, the UN has authorized and deployed seventeen separate missions to postconflict states. (26) Although one may discern in these missions a variety of mechanisms to ensure public participation, all but one has organized and/or observed national elections as one of its central tasks. (27) In the words of the 1994 Lusaka Protocol, setting out a (yet unfulfilled) peace plan for Angola:

As in every democratic and multiparty society, the participation of all citizens in the definition of the national political, social and economic guidelines and options, as well as in the free choice of the country's leaders, is guaranteed by respect for the principle of the expression of the people's will in periodic, free and fair elections and acceptance of their results. (28)

Thus, most peace agreements providing for a UN role contain an electoral mandate. Many go further and provide that elections shall be at the center of a new constitutional order in the state. The Paris peace accords for Cambodia, in a provision entitled "Principles for a New Constitution for Cambodia," provide an example:

The constitution will state that Cambodia will follow a system of liberal democracy, on the basis of pluralism. It will provide for periodic and genuine elections. It will provide for the right to vote and to be elected by universal and equal suffrage. It will provide for voting by secret ballot, with a requirement that electoral procedures provide a full and fair opportunity to organize and participate in the electoral process. (29)

The only peace agreement with a comprehensive framework for public participation beyond elections emerged from the Guatemalan peace process. The government and the Unidad Revolucionaria Nacional Guatemalteca agreed to a system of decentralized governance to facilitate "increased opportunities for citizen participation and building their capacity to participate." (30) These local initiatives, mandated in addition to sweeping reforms of all branches of government at the national level, were based on the view that "social participation at the community level promotes respect for ideological pluralism and nondiscrimination on social grounds, facilitates the broad, organized and harmonious participation of citizens in decision-making and enables them to shoulder their responsibilities and commitments in the quest for social justice and democracy." (31)

A review of the UN's electoral assistance practice substantially broadens the limited criteria for "free and fair" elections evident in human rights treaties. Viewing the UN missions as contributing to the growth of a customary international law on free and fair elections, the following requirements may be added:

* An independent national electoral commission to supervise all aspects of the campaign and balloting.

* Freedom of organization, movement, assembly, and expression for political parties.

* Equal access to all media sources by all competing parties, especially if state-run media outlets predominate.

* Accurate and current electoral rolls, as well as an expeditious process for resolving disputes over cases of individual eligibility.

* No interference with campaign rallies.

* Free access by election monitors from international organizations and NGOs.32

Testing the Model: The Pragmatist Critique

If achieving clarity in standards suggests that the election-centered reconstruction model has been a successful effort in lawmaking, how are we to judge its undeniably spotty record in practice? Of the fourteen states and territories in which UN democratic reconstruction missions have operated, only two are rated "free" according to the Freedom House methodology; and in one of these (Croatia), the mission was not countrywide and arguably had little to do with the emergence of its democratic institutions. In the others (seven "not free" and six "partly free"), democratic processes prescribed by the model operate intermittently, and in some cases (Congo, Somalia, Tajikistan) not at all. In two cases (Angola and Congo), one cannot even speak of the countries as "postconflict." (33)

Several critics argue that the democratic reconstruction model substantially overestimates the capacity of postconflict societies to assimilate democratic values, especially competitive electoral politics. In their view, a more pragmatic approach is needed. They suggest alternative strategies that in many respects depart markedly from the liberal vision of the current model. Marina Ottaway argues that in the collapsed and virtually collapsed states that have been the targets of reconstruction missions, "military might is a necessary component of statebuilding." Unless the international community is willing to commit substantial military resources and supervise transitional states for prolonged periods, she writes, it must set more modest objectives: "In some cases, such as the DRC [Democratic Republic of Congo], the international community should either accept the disintegration of the country or allow nondemocratic leaders to use force to put the state back together." (34)

Jack Snyder goes further, arguing that an overenthusiastic "liberal optimism" in the early 1990s unleashed destructive nationalist forces by encouraging democratization in societies without a "thick network of social supports" for liberal institutions and values.35 When these conditions are absent or insufficiently rooted, "it may be wise to postpone encouraging democratization until they are." (36) Snyder urges international actors to "contain nationalist conflict by whatever means are expedient in the short run, including coercion." (37) Among the methods he suggests are acquiescence to "hegemonic control" by minority groups in ethnically divided states, ethnically-based partition, assimilation (where possible) to the cultural identity of majority groups, federalism, power-sharing arrangements, and "combating nationalist myths in the marketplace of ideas." (38)

The Response: Contextualizing Implementation of Democratic Norms

Neither Ottaway nor Snyder disagrees with the value of promoting democracy in postconflict states, and thus their critiques are not wholly inconsistent with the goals of a democratic entitlement. But the alternative routes to national rehabilitation they advocate are deeply inconsistent with contemporary international law. For one, the democratic entitlement is phrased as a legal rule of universal application, leaving little room for judgments by the UN or other international actors that certain societies are not ready for democratic politics. (39) For another, many of the alternatives they suggest have been repeatedly and pointedly condemned by other legal regimes, as well as by the Security Council reacting to particular crises. Assimilating "ethnic minorities to the cultural identity of the ethnic majority," permitting "ethnic hegemony by minority groups (albeit with reformist tendencies), and sanctioning ethnically based partitions that involve the "resettlement of huge numbers of unwilling people" are al l clearly contrary to human rights norms now deeply entrenched in the decisionmaking of international institutions. (40) Partition is also inconsistent with international law's preference for existing borders and active discouragement of secession, as evident in the continued existence of Bosnia, the commitment of UN-authorized troops to squelch secessionism in Kosovo, and the failure of states to recognize Biafra, Somaliland, Chechnya, Mayotte, Abkhazia, or the vast majority of other territories seeking to secede from existing states. (41)

But concluding that these alternate strategies are counternormative does not address the heart of the pragmatist critique. Its essential argument is that the democratic reconstruction model has failed in practice. Is this claim correct? How one answers this complex question depends on the criteria one uses to judge the model's success or failure. I argue here that a different measure is required for assessing multilateral efforts at democracy promotion than is appropriate for transitions that are wholly or even mostly indigenous. Indigenous transitions are properly assessed by factors internal to the state. An observer may draw on its history, culture demographic schisms, quality of leadership, and other local phenomena in explaining the traction of democratic institutions and values. Assessment of multilateral efforts must add the presence of external actors, ideas, and, in some cases, coercive measures. While no transition process is wholly indigenous--making the distinction in some cases more a matter of d egree than kind--the postconflict transitions under review took on a distinctively different form with the presence of the UN and other external actors. Two specific reasons are evident for assessing multilateral transitions differently.

First, the theoretical models of democracy that form the basis for quantitative assessments, such as the Freedom House Index, are self-consciously designed as claims made by and for single national societies. That is, they are either not designed to serve as transnational measures of the institutions they prescribe domestically or come with explicit caveats that factors other than components of the theory itself need to be considered when they are applied transnationally. John Stuart Mill limited his utilitarian arguments for representative government to societies that "possess certain special requisites," particularly a willingness and ability to support it. Thus, "kingly government, free from the control (though perhaps strengthened by the support) of representative institutions, is the most suitable form of polity for the earliest stages of any community." (42) John Rawls limits his postulates of justice to "well-ordered societies," which are conceived of as "closed system[s] isolated from other societies. " (43) And Robert A. Dahl warns that efforts "to transform the government of another country from a hegemony or mixed regime into a polyarchy face formidable and complex problems, not the least of which is our lack of knowledge about the long causal chain running from outside help to internal conditions to changes of regime." (44)

This is not an argument that democracy is culturally determined, a claim one must at least regard with skepticism if democratic norms are in fact emerging within a universalist international law. Rather, it is a caution that the standard theoretical models of liberal politics--as exemplified by Mill, Rawls, and Dahl--cannot be the sole judge of external efforts to persuade local actors of the desirability of those theories. The theories do not assume the numerous transaction costs inherent in such efforts. Those costs may be political (convincing locals of the need to tolerate opposing viewpoints), logistical (devoting insufficient resources to a mission), institutional (facing bureaucratic obstacles in organizations like the UN), or even geographic (finding enough multilateral staff who speak a local language). A single society can justifiably exclude such questions of "translation" and assess its own democratic processes purely by reference to what might be termed "first principles of democracy"--considerat ions, for example, of utility (Mill), justice (Rawls), or participation and contestations, (Dahl). But international society cannot do so. Multilateral actors must account for their alien presence in a society--and all its ramifications--when measuring outcomes.

Second, and more important, seeking incremental progress in consolidating democratic transitions is wholly consistent with leading theories of how international legal norms are implemented and enforced. Prior sections of this article detailed the close nexus between the democratic reconstruction model and the emerging international legal principle of a "democratic entitlement." This suggests that our assessment of the missions should draw on more general theories of how international law is applied and implemented. Because international law lacks the implementing tools available in most domestic systems--legislatures, courts, regulatory agencies, police forces-adherence generally cannot be compelled by the threat or use of coercive force. Rather, implementation occurs through indirect processes of acculturation and legitimation. These processes are carried out by institutional actors acutely aware of the process being both incremental and dialectical: debate between proponents and opponents of international n ormativity may rage over long periods of time before implementation strategies are widely accepted and adopted.

Two widely cited theories of implementation illustrate this view. Abram Chayes and Antoina Handler Chayes posit a "managerial" model of compliance, in which treaty parties work within multilateral institutions in ongoing processes of reporting, justification, evaluation, and persuasion.(45) Through transparent processes that seek to "jawbone" recalcitrant parties and build capacity for implementation, states may become acculturated to compliance with the regime. This "justificatory discourse" will not end quickly. But the Chayeses are untroubled by this, since "significant changes in social or economic systems mandated by regulatory treaties take time" and "a cross section at any particular moment may give a misleading picture of the state of compliance."(46)

A second approach to compliance is Harold Koh's "internalization" theory.(47) In contrast to the Chayeses' exclusive focus on international forums, Koh finds agents for implementation in "transnational" legal processes. These are processes "whereby an international law rule is interpreted through the interaction of transnational actors in a variety of law-declaring fora, then internalized into a nation's domestic legal system." (48) The process of internalization involves a variety of domestic actors, including "transnational norm entrepreneurs," governmental bodies, transnational issue networks, "interpretive communities," and bureaucratic compliance procedures. These actors interact constantly in disputes that require each to put forward an interpretation of relevant norms in order to justify its position. An actor thereby promoting normativity "seeks not simply to coerce the other party, but to internalize the new interpretation of the international norm into the other party's internal normative system. Th e aim is to 'bind' that other party to obey the interpretation as part of its internal value set." (49) Compliance, in Koh's view, is therefore not a discernable end point but an ongoing process of articulating and inculcating law-compliance as a cherished value: "Transnational actors obey international law as a result of repeated interaction with other actors in the transnational legal process.,, (50)

For both the Chayeses and Koh, compliance requires ongoing and purposeful engagement between legally constrained actors. The temporal dimension thus plays a much larger role in the international legal system than in domestic systems, where robust means of detecting and prosecuting lawbreakers may compel compliance in single, defining episodes. One may easily miss this point when evaluating the success of postconflict democratization missions, since those missions appear precisely to resemble such isolated "compliance moments." But the missions are vastly more complex and elongated, especially in recent iterations. The missions to Bosnia and Kosovo, for example, show no sign of ending seven and three years respectively after the conflicts terminated. Moreover, they are pursuing national reconstruction along a wide variety of planes beyond holding elections. Other missions, such as those to Angola, Mozambique, and the Democratic Republic of Congo, have stopped and started according to the willingness of local p arties to compromise and cease conflict. Still others, such as the UN Transitional Authority in Cambodia (UNTAC), have ended their formal role but left in their wake new and energetic civil society sectors seeking to carry forward the democratic values espoused by the departed multilateral actors. (51) In this temporal respect, the missions resemble other international norms. Their success should therefore be measured in the same way: with ample time allowed for acculturation in both national and international forums, and with measures of success decidedly more relative than absolute.

One can easily find compelling reasons for measuring the missions success in the short term: citizens in target states whose expectations are raised by promises of a democratic future may sour on liberal institutions if stability and opportunities for meaningful participation fail to materialize immediately. Similarly, international organizations such as the UN may require "success stories" to keep their political bases engaged and funding secure. But like the innovative corporation forced to justify its profitability in quarterly rather than yearly increments, the democratic transition model may involve a commodity whose value can be measured accurately only in the longer term.

These two claims-that theories of domestic politics cannot provide the complete measure of success for international promotion of democracy, and that because the missions exist within the larger body of international law their implementation should be measured in the long term-suggest how the international community should assess benefits accruing (or not accruing) to postconflict societies. They describe this process as different from how a single society would assess its own internal initiatives. This important distinction may be underlined through a final consideration: the benefits international society accrues from successful democratization missions. The policies supporting a normative democratic entitlement focus on how democratic transitions affect relations between states. The democratic peace thesis is the most widely discussed of these policies, valuing democratization for its role in a uniquely international concern: warfare between states. Similar arguments are made concerning the capacity of dem ocratic states to implement complex international regulatory treaties. (52) Neither of these considerations appears in traditional democratic political theories. Their absence points up the limitations of those theories in assessing transnational democracy promotion.

Conclusion

A legal assessment of postconflict reconstruction missions helps us answer two questions. First, are international officials free to foster nondemocratic regimes in peacebuilding missions? I have suggested that an emerging body of international law strongly supports democratic institutions and has not to date made exceptions for nondemocratic politics even in the most fragile postconflict states. This normative judgment rests on the view-articulated repeatedly by multilateral actors-that exclusionary, nondemocratic politics have been a significant cause of the conflicts themselves. For the UN, the logical corollary has been that the legitimacy of postconflict regimes depends critically on breaking with such past practices by providing opportunities for participation to all former belligerent parties. Critics usefully point out the sizable commitment of resources needed to effectuate this goal-resources frequently not provided to UN missions. But international actors have not endorsed the critics' crucial next step of using this problem of implementation as a reason to compromise or abandon the normative commitment to democratic peacebuilding.

Second, what is the appropriate method for assessing the missions' success? I suggest that multilateral democracy promotion should not be judged in the short term but that observers should adopt the more flexible, incremental framework appropriate to a norm of international law. The UN's experience with decolonization provides a useful parallel, as it also involved a division in the international community along regional lines over radical transformations in national governing structures. The 1945 UN Charter encouraged colonial powers to move their dependencies toward self-government, albeit with distinctions between categories of territory and in less than unambiguous language. Few territories attained independence in the 1940s and 1950s, as the major colonial powers strongly resisted efforts by smaller states and independence movements to force immediate decolonization. The colonial powers lost the battle in 1960, however, as the General Assembly resolved that the charter required independence without delay . (53) By the mid-1970s, after the democratic transition in Portugal, virtually all of Europe's colonies had attained statehood. (54)

Yet if the decolonization regime's "success" had been measured even five years after it was introduced as a charter norm, it would have been judged an abject failure. A gestation period of thirty-five years provides a completely different, largely positive assessment of the regime. While we cannot know whether the democratization norm will take hold in thirty-five years (or longer), the experience of decolonization suggests that a mere ten years after the first UN-initiated postconflict peacebuilding mission is too early to come to a final judgment.

Notes

(1.) Lasa Oppenheim, International Law, vol. 1 (London: Longmans, Green, 1905), p. 403.

(2.) One of many excellent discussions of postconflict reconstruction missions is Elizabeth M. Cousens and Chetan Kumar, eds., Peacebuilding as Politics: Cultivating Peace in Fragile Societies (Boulder: Lynne Rienner, 2001).

(3.) Report of the Panel on United Nations Peace Operations, UN Doc. A/55/305-S/2000/809 (2000): par. 13.

(4.) UNAMA (Afghanistan, 2002-present); UNAVEM III (Angola, 1991-1995); UNAVEM III (Angola, 1995-1997); MONUA (Angola, 1997-1999); MINURCA (Central African Republic, 1998-2000); ONUMUZ (Mozambique, 1992-1994); UNSOM I (Somalia, 1992-1993); UNSOM II (Somalia, 1993-1995); ONUSAL (El Salvador, 1991-1995); MINUGA (Guatemala, 1997); UNTAC (Cambodia, 1992-1993); UNMOT (Tajikistan, 1994-2000); UNTAES (Croatia, 1996-1998); MONUC (Congo, 2000-present); UNAMSIL (Sierra Leone, 1999-present); UNTAET (East Timor, 1999-present); UNMIBH (Bosnia, 1995-present).

(5.) For a general discussion, see Gregory H. Fox and Brad R. Roth, eds., Democratic Governance and International Law (New York: Cambridge University Press, 2000). The literature on multilateral democracy promotion is now vast. See Michael Cox, G. John Ikenberry, and Takashi Inoguchi, eds, American Democracy Promotion: Impulses, Strategies, and Impacts (New York: Oxford University Press, 2000); Thomas Carothers, Aiding Democracy Abroad: The Learning Curve (Washington, D.C.: Carnegie Endowment for International Peace, 1999); Jean Grugel, ed., Democracy Without Borders: Transnationalization and Conditionality in New Democracies (London: Routledge, 1999).

(6.) The term originates with Thomas M. Franckin "The Emerging Right to Democratic Governance," American Journal of International Law 86, no. 1 (1992): 46.

(7.) United Nations Charter, art. 1(5).

(8.) Universal Declaration of Human Rights, GA Res. 217 (III), art. 21 (1948).

(9.) Gregory H. Fox and Brad R. Roth, "Democracy and International Law," Review of International Studies 27, no. 2 (2001): 327-352.

(10.) CSCE Charter of Paris for a New Europe, 30 ILM 190, 193 (1990).

(11.) African Commission on Human Rights, Resolution on the Military, Eighth Annual Activity Report of the Commission on Human and Peoples' Rights, 1994-1955, p. 39.

(12.) Inter-American Democratic Charter (11 September 2001), available online at http://www.oas.org/OASpage/eng/Documents/Democractic_Charter.htm.

(13.) Legal Aspects of Problems of Representation in the United Nations, UN Doc. S/1466 (1950).

(14.) An Agenda for Democratization, UN Doc. A/51/761 (1996), par. 15, 45.

(15.) W. Michael Reisman, "Sovereignty and Human Rights in Contemporary International Law," in Fox and Roth, Democratic Governance and International Law, p. 243.

(16.) SC Res. 940 (1994) (Haiti); UN Doc. S/PRST/1998/5 (statement by president of the Security Council welcoming the removal of the Sierra Leone junta).

(17.) Preliminary Report of the Commonwealth Observer Group to the Presidential Election in Zimbabwe 9-10 March 2002, available online at http:// www.thecommonwealth.org/dynamic/pressoffice/display.asp?id=425&type=p ress&cat=53.

(18.) See the comprehensive discussion in Gregory H. Fox, "The Right to Political Participation in International Law," in Fox and Roth, Democratic Governance and International Law, p. 48. Article 25 of the Covenant on Civil and Political Rights is typical:

Every citizen shall have the right and the opportunity, without any of the distinctions mentioned in article 2 and without unreasonable restrictions:

(a) To take part in the conduct of public affairs, directly or through freely chosen representatives;

(b) To vote and to be elected at genuine periodic elections which shall be by universal and equal suffrage and shall be held by secret ballot, guaranteeing the free expression of the will of the electors;

(c) To have access, on general terms of equality, to public service in his country.

(19.) Ibid., p. 69.

(20.) Ibid., p. 61. The European Commission and Court have, by far, heard the greatest number of election cases of any human rights tribunals. Their decisions may be fairly generalized to other human rights systems because the European bodies have been far more aggressive in holding national practices of all kinds to be contrary to treaty standards. That they were unwilling to so hold in these areas suggests that other human rights bodies will also defer to state judgments as to how a fair electoral system is to be constructed.

(21.) Gitonas and Others v. Greece, Eur. Ct. H.R. 1997-IV (1997), par. 44.

(22.) These rights are provided, for example, in Articles 18, 19, 21, 22, 12 and 26, respectively, of the International Covenant on Civil and Political Rights, 16 December 1966, 999 UNTS 171.

(23.) Larry Diamond, "Toward Democratic Consolidation," Journal of Democracy 5 (1994): 4-17.

(24.) The Panel on United Nations Peace Operations made this point emphatically:

"Free and fair" elections should be viewed as part of broader efforts to strengthen governance institutions. Elections will be successfully held only in an environment in which a population recovering from war comes to accept the ballot over the bullet as an appropriate and credible mechanism through which their views on government are represented. Elections need the support of a broader process of democratization and civil society building that includes effective civilian governance and a culture of respect for basic human rights, lest elections merely ratify a tyranny of the majority or be overturned by force after a peace operation leaves.

UN Doc. A/55/305-S/2000/809, par. 38 (2000).

(25.) The offices are currently operating in Angola, Burundi, Central African Republic, Guinea Bissau, Liberia, and Somalia.

(26.) See the list in note 4.

(27.) The exception is the MINURCA mission to Central African Republic. The recently approved mission to Afghanistan, UNAMA, is also an exception, but this will almost certainly change once the transitional process is further along. Because the Bonn agreements set out a multitiered transition, the first phase of which did not involve nationwide elections, UNAMA could not include elections as part of its mandate. But perhaps learning from experiences such as East Timor, UNAMA is structured quite consciously to maximize local Afghan participation in other ways. In his report setting out the structure of the new mission, the secretary-general stated that "UNAMA should aim to bolster Afghan capacity (both official and non-governmental), relying on as limited an international presence and on as many Afghan staff as possible, and using common support services, thereby leaving a light expatriate 'footprint."' Report of the Secretary-General, The Situation in Afghanistan and Its Implications for International Peace a nd Security, UN Doc. A/56/875-S/2002/278, par. 98(d) (18 March 2002); see also, ibid. par. 107(e) ("All such United Nations actors would be expected to make a transition to a predominant reliance on qualified national personnel, from all ethnic groups and including women, and would be expected to initiate internal capacity-development activities to achieve that end"); SC Res. 1401 (28 March 2002) ("although humanitarian assistance should be provided wherever there is a need, recovery of reconstruction assistance ought to be provided though the Afghan Interim Administration and its successors, and implemented effectively where local authorities contribute to the maintenance of a secure environment and demonstrate respect for human rights").

(28.) Lusaka Protocol, Annex 7, art. 1 (15 November 1994), available online at http://www.usip.org/library/pa/angola/lusaka_11151994.html.

(29.) Agreement on a Comprehensive Political Settlement of the Cambodia Conflict, Annex 5, art. 4 (23 October 1991), available online at http://www.usip.org/library/pa/cambodia/comppol_10231991_annex5.html.

(30.) Agreement on the Strengthening of Civilian Power and on the Role of the Armed Forces in a Democratic Society, par. 56-58 (19 September 1996), available online at http://www.usip.org/library/pa/guatemala/guat_960919.html.

(31.) Ibid., par. 77.

(32.) Fox, "The Right to Political Participation in International Law," p. 84.

(33.) Freedom House, Freedom in the World 2001-2002: The Democracy Gap (2002), available online at http://www.freedomhouse.org/research/freeworld/2002/essay2002.pdf.

(34.) Marina Ottaway, "Nation Building," Foreign Policy (September-October 2002), p. 16.

(35.) Jack Snyder, From Voting to Violence: Democratization and National Conflict (New York: Norton, 2000), pp. 16, 316.

(36.) Ibid., p. 316.

(37.) Ibid., p. 321.

(38.) Ibid., pp. 322-327.

(39.) As noted, the Universal Declaration provides that "the will of the people shall be the basis of the authority of government." The legal regime of decolonization, closely related to democratization in its notions of autonomous self-government, expressed an even blunter universalism: "Inadequacy of political, economic, social or educational preparedness should never serve as a pretext for delaying independence." Declaration on the Granting of Independence to Colonial Countries and Peoples, GA Resolution 1514, par. 3 (1960).

(40.) Snyder, From Voting to Violence, pp. 322-327, 328.

(41.) For further discussion of the legal objections to such alternatives to democratization within existing borders, see Gregory H. Fox, "Strengthening the State," Indiana Journal of Global Legal Studies 7, no. 1(1999): 35-77.

(42.) John Stuart Mill, Considerations on Representative Government (1861; reprint, Amherst, N.Y.: Prometheus, 1991), p. 90.

(43.) John Rawls, A Theory of Justice (Cambridge: Belknap Press, 1971), pp. 8, 457.

(44.) Robert A. Dahl, Polyarchy (New Haven: Yale University Press, 1971), p. 210.

(45.) Abram Chayes and Antonia Handler Chayes, The New Sovereignty: Compliance with International Regulatory Regimes (Cambridge: Harvard University Press, 1998).

(46.) Ibid., p. 15.

(47.) See Harold Hongju Koh, "Bringing International Law Home," Houston Law Review 35, no. 2 (1998): 623-680; Harold Hongju Koh, "Why Do Nations Obey International Law?" Yale Law Journal 106, no. 5 (1997): 2599-2659.

(48.) Koh, "Bringing International Law Home," p. 625.

(49.) Koh, "Why Do Nations Obey International Law?" p. 2646.

(50.) Ibid., p. 2656.

(51.) Michael Doyle notes that the UNTAC mission in Cambodia "helped open a space for hundreds of new, indigenous Cambodian NGOs." Michael Doyle, "Peacebuilding in Cambodia: Legitimacy and Power," in Cousens and Kumar, Peacebuilding as Politics, pp. 89, 106.

(53.) Declaration on the Granting of Independence to Colonial Countries and Peoples, GA res. 1514 (XV), 15 UN GAOR, Supp. (no. 16) at 66, UN doc. A/4684 (1961).

(52.) Fox, "Strengthening the State."

(54.) John Springhall, Decolonization Since 194S: The Collapse of European Overseas Empires (New York: Palgrave, 2001).

Gregory H. Fox is assistant professor of law at Wayne State University Law School. He has held fellowships at Yale Law School and the Max Planck Institute for Comparative Public Law and Public International Law. He is the editor (with Brad Roth) of Democratic Governance and International Law (2000).