Elections are a symbol of democracy, renewing the right of individuals to choose their rulers. In the horoscope of humanist politics, this ceremony of consent remains a potent sign as a central act of political commitment. It references our foundational regard for personal liberty and autonomy, as well as the necessity of compromise and agreement. And yet there is the 20th-century warning of Elias Canetti in Masse und Macht. Writing in German, with a family history that roamed from Ladino Spain and Bulgaria to Vienna, London and Zurich, he observed how men in political groups may act rashly, without diffidence or thought, abandoning moral restraint on the grounds that others are indulging and the event will happen anyway. The voice of the demos may ring so loud that it is taken as the voice of the volk, the organism of a state that mistakenly claims a right to exemption from ordinary standards of conduct. The legitimating effect of democracy can, at times, be its very danger.

There is thus a hazard in saluting democratic elections as the only measure of political ethics. Collective choice must respect the rights of individuals and a fundamental human decency. In the 18th century, international law played a particular role in this limitation. The “law of nature and nations,” as it was then styled, went beyond diplomatic relations among states. It claimed status as a common law or jus commune to frame how a ruler should act at home.

This authority may have seemed less radical in the English-speaking colonies against the background of the Scottish Enlightenment and “common sense” moral philosophers such as Adam Smith, Thomas Reid and James Ferguson. Their belief in a natural moral faculty and human sympathy informed the Declaration of Independence of the American republic. Self-evident truths, inalienable rights and the common opinion of mankind would necessarily inform democratic choice.

But even then, there existed a deep suspicion of what pure democracy could bring. The American Constitution provided for indirect rather than direct selection of the president and the Senate, working through the mediation of the Electoral College and state legislatures. The federal right of election was only to choose an intermediate group of men who would, in turn, deliberate and reflect upon the qualifications of chosen candidates for the Senate and the presidency. The architecture was republican, rather than democratic. Tellingly, the promise of Article IV of the Constitution was that the federal government would guarantee a “republican” government in the states. Pure democracy was still regarded with suspicion, against the history of the Roman republic, as a form likely to decay into a mob and tyranny.

Even into the early 19th century, this view persisted in many quarters. Supreme Court Chief Justice John -Marshall and his confidant Joseph Story were deeply skeptical of what popular economic policy might bring in its wake, clinging to a constitutional theory of “vested” rights to limit temptation. The franchise was restricted to men of property for the same reason and, at least until the Civil War amendments, the federal government deferred to the judgment of the states in matters of electoral qualification.

Recalling this older fear of a “tyranny of the majority,” we can afford some understanding of the difficulties of erecting new democratic structures in the rest of the world. States are not made of homogeneous ethnic, religious or cultural groups. The idea of “concurrent majorities” has been offered as an attractive but difficult account of democracy—supposing that a well-founded government should warrant the respect of quite separate communities. But a stable balance of power and common ground can be hard to reach, especially when consensus is lacking on fundamental issues.

One of the most divisive questions is whether the state should embrace a particular ethnic or religious identity. In a classic essay entitled Liberal Nationalism, Yael Tamir worried that the deep individualism of American society and liberal theory might blind us to the practical force of religious and nationalist identifications in other societies. These memberships might not be negotiable, she warned, and could not be treated as affectations, but rather often were constitutive of who people were and would choose to be. Democracy must take account of that limit, she suggested.

The utility of elections as a method of peacemaking over the last decade has warranted the force of this observation. After the bitter conflict in Bosnia, holding early elections proved to be a source of friction, rather than reconciliation, as nationalist parties ran for office by invoking the malevolent intent and past deeds of rival communities. The Sarajevo government is maintained, even now, through the support of international peacekeepers and might not survive on its own. In Rwanda, the moral failure of the West to intervene against the genocide has since been used to mask tendentious methods of political rule.

The difficulties of reconciling communities that lack common language and culture has been seen in Kosovo, where the visible lack of progress under UNMIK (the U.N. Mission in Kosovo) has led to an impatient demand to have done with the matter, in favor of a “referendum” whose outcome is foreordained. Iraq is not the only country where the violent frictions of competing communities have been hard to quell. And in widely disparate settings, the international community has often used the event of an election as an excuse to withdraw before there is political stability and peace on the ground.

The belief that a referendum or election can be a cure-all in matters of governance or self-determination is confounded by the profound difficulty of determining who should qualify as a voter. When the 1996 elections were organized in Bosnia to implement the Dayton Accords, the Organization for Security and Co-operation in Europe agreed to a ballot rule that permitted displaced persons to vote in their place of original residence. This was exploited to bring in busloads of “voters” from across international borders who lacked any prior residency.

So, too, the wistful hope of resolving the status of Western Sahara through a referendum has foundered for decades over who gets to vote as a member of the indigenous Sahrawi Arab-Berber tribes. There is no specification of the human genome that would suffice to say who has a legitimate claim as a member of a scattered and nomadic group, and the regional rivals have not been brought to agree.

Who votes in a majority election, in a world of particularistic identities, will likely determine the outcome.

Thus, in celebrating the importance of democracy and elections as an ambition of American foreign policy, we must pay as much attention to the other elements of political concord: structures of power-sharing, federalism, regional decentralization and monitoring by mediate institutions, including constitutional courts. The construction of a res publica—the political cultivation of a common cause and commitment—is perhaps the hardest task after a conflict.

Ruth Wedgwood is the Edward B. Burling Professor of International Law and Diplomacy and director of the International Law and Organizations Program.