The American diplomat Richard Holbrooke pondered a problem on the eve of the September 1996 elections in Bosnia, which were meant to restore civic life to that ravaged country. "Suppose the election was declared free and fair," he said, and those elected are "racists, fascists, separatists, who are publicly opposed to [peace and reintegration]. That is the dilemma." Indeed it is, not just in the former Yugoslavia, but increasingly around the world. Democratically elected regimes, often ones that have been reelected or reaffirmed through referenda, are routinely ignoring constitutional limits on their power and depriving their citizens of basic rights and freedoms. From Peru to the Palestinian Authority, from Sierra Leone to Slovakia, from Pakistan to the Philippines, we see the rise of a disturbing phenomenon in international life—illiberal democracy.

It has been difficult to recognize this problem because for almost a century in the West, democracy has meant liberal democracy—a political system marked not only by free and fair elections, but also by the rule of law, a separation of powers, and the protection of basic liberties of speech, assembly, religion, and property. In fact, this latter bundle of freedoms—what might be termed constitutional liberalism—is theoretically different and historically distinct from democracy. As the political scientist Philippe Schmitter has pointed out, "Liberalism, either as a conception of political liberty, or as a doctrine about economic policy, may have coincided with the rise of democracy. But it has never been immutably or unambiguously linked to its practice." Today the two strands of liberal democracy, interwoven in the Western political fabric, are coming apart in the rest of the world. Democracy is flourishing; constitutional liberalism is not.

Today, 118 of the world's 193 countries are democratic, encompassing a majority of its people (54.8 percent, to be exact), a vast increase from even a decade ago. In this season of victory, one might have expected Western statesmen and intellectuals to go one further than E. M. Forster and give a rousing three cheers for democracy. Instead there is a growing unease at the rapid spread of multiparty elections across south-central Europe, Asia, Africa, and Latin America, perhaps because of what happens after the elections. Popular leaders like Russia's Boris Yeltsin and Argentina's Carlos Menem bypass their parliaments and rule by presidential decree, eroding basic constitutional practices. The Iranian parliament—elected more freely than most in the Middle East—imposes harsh restrictions on speech, assembly, and even dress, diminishing that country's already meager supply of liberty. Ethiopia's elected government turns its security forces on journalists and political opponents, doing permanent damage to human rights (as well as human beings).

Naturally there is a spectrum of illiberal democracy, ranging from modest offenders like Argentina to near-tyrannies like Kazakstan and Belarus, with countries like Romania and Bangladesh in between. Along much of the spectrum, elections are rarely as free and fair as in the West today, but they do reflect the reality of popular participation in politics and support for those elected. And the examples are not isolated or atypical. Freedom House's 1996-97 survey, Freedom in the World, has separate rankings for political liberties and civil liberties, which correspond roughly with democracy and constitutional liberalism, respectively. Of the countries that lie between confirmed dictatorship and consolidated democracy, 50 percent do better on political liberties than on civil ones. In other words, half of the "democratizing" countries in the world today are illiberal democracies.

Illiberal democracy is a growth industry. Seven years ago only 22 percent of democratizing countries could have been so categorized; five years ago that figure had risen to 35 percent. And to date few illiberal democracies have matured into liberal democracies; if anything, they are moving toward heightened illiberalism. Far from being a temporary or transitional stage, it appears that many countries are settling into a form of government that mixes a substantial degree of democracy with a substantial degree of illiberalism. Just as nations across the world have become comfortable with many variations of capitalism, they could well adopt and sustain varied forms of democracy. Western liberal democracy might prove to be not the final destination on the democratic road, but just one of many possible exits.


From the time of Herodotus democracy has meant, first and foremost, the rule of the people. This view of democracy as a process of selecting governments, articulated by scholars ranging from Alexis de Tocqueville to Joseph Schumpeter to Robert Dahl, is now widely used by social scientists. In The Third Wave, Samuel P. Huntington explains why:

Elections, open, free and fair, are the essence of democracy, the inescapable sine qua non. Governments produced by elections may be inefficient, corrupt, shortsighted, irresponsible, dominated by special interests, and incapable of adopting policies demanded by the public good. These qualities make such governments undesirable but they do not make them undemocratic. Democracy is one public virtue, not the only one, and the relation of democracy to other public virtues and vices can only be understood if democracy is clearly distinguished from the other characteristics of political systems.

This definition also accords with the commonsense view of the term. If a country holds competitive, multiparty elections, we call it democratic. When public participation in politics is increased, for example through the enfranchisement of women, it is seen as more democratic. Of course elections must be open and fair, and this requires some protections for freedom of speech and assembly. But to go beyond this minimalist definition and label a country democratic only if it guarantees a comprehensive catalog of social, political, economic, and religious rights turns the word democracy into a badge of honor rather than a descriptive category. After all, Sweden has an economic system that many argue curtails individual property rights, France until recently had a state monopoly on television, and England has an established religion. But they are all clearly and identifiably democracies. To have democracy mean, subjectively, "a good government" renders it analytically useless.

Constitutional liberalism, on the other hand, is not about the procedures for selecting government, but rather government's goals. It refers to the tradition, deep in Western history, that seeks to protect an individual's autonomy and dignity against coercion, whatever the source—state, church, or society. The term marries two closely connected ideas. It is liberal because it draws on the philosophical strain, beginning with the Greeks, that emphasizes individual liberty. It is constitutional because it rests on the tradition, beginning with the Romans, of the rule of law. Constitutional liberalism developed in Western Europe and the United States as a defense of the individual's right to life and property, and freedom of religion and speech. To secure these rights, it emphasized checks on the power of each branch of government, equality under the law, impartial courts and tribunals, and separation of church and state. Its canonical figures include the poet John Milton, the jurist William Blackstone, statesmen such as Thomas Jefferson and James Madison, and philosophers such as Thomas Hobbes, John Locke, Adam Smith, Baron de Montesquieu, John Stuart Mill, and Isaiah Berlin. In almost all of its variants, constitutional liberalism argues that human beings have certain natural (or "inalienable") rights and that governments must accept a basic law, limiting its own powers, that secures them. Thus in 1215 at Runnymede, England's barons forced the king to abide by the settled and customary law of the land. In the American colonies these laws were made explicit, and in 1638 the town of Hartford adopted the first written constitution in modern history. In the 1970s, Western nations codified standards of behavior for regimes across the globe. The Magna Carta, the Fundamental Orders of Connecticut, the American Constitution, and the Helsinki Final Act are all expressions of constitutional liberalism.


Since 1945 Western governments have, for the most part, embodied both democracy and constitutional liberalism. Thus it is difficult to imagine the two apart, in the form of either illiberal democracy or liberal autocracy. In fact both have existed in the past and persist in the present. Until the twentieth century, most countries in Western Europe were liberal autocracies or, at best, semi-democracies. The franchise was tightly restricted, and elected legislatures had little power. In 1830 Great Britain, in some ways the most democratic European nation, allowed barely 2 percent of its population to vote for one house of Parliament; that figure rose to 7 percent after 1867 and reached around 40 percent in the 1880s. Only in the late 1940s did most Western countries become full-fledged democracies, with universal adult suffrage. But one hundred years earlier, by the late 1840s, most of them had adopted important aspects of constitutional liberalism—the rule of law, private property rights, and increasingly, separated powers and free speech and assembly. For much of modern history, what characterized governments in Europe and North America, and differentiated them from those around the world, was not democracy but constitutional liberalism. The "Western model" is best symbolized not by the mass plebiscite but the impartial judge.

The recent history of East Asia follows the Western itinerary. After brief flirtations with democracy after World War II, most East Asian regimes turned authoritarian. Over time they moved from autocracy to liberalizing autocracy, and, in some cases, toward liberalizing semi-democracy. Most of the regimes in East Asia remain only semi-democratic, with patriarchs or one-party systems that make their elections ratifications of power rather than genuine contests. But these regimes have accorded their citizens a widening sphere of economic, civil, religious, and limited political rights. As in the West, liberalization in East Asia has included economic liberalization, which is crucial in promoting both growth and liberal democracy. Historically, the factors most closely associated with full-fledged liberal democracies are capitalism, a bourgeoisie, and a high per capita GNP. Today's East Asian governments are a mix of democracy, liberalism, capitalism, oligarchy, and corruption—much like Western governments circa 1900.

Constitutional liberalism has led to democracy, but democracy does not seem to bring constitutional liberalism. In contrast to the Western and East Asian paths, during the last two decades in Latin America, Africa, and parts of Asia, dictatorships with little background in constitutional liberalism have given way to democracy. The results are not encouraging. In the western hemisphere, with elections having been held in every country except Cuba, a 1993 study by the scholar Larry Diamond determined that 10 of the 22 principal Latin American countries "have levels of human rights abuse that are incompatible with the consolidation of [liberal] democracy." In Africa, democratization has been extraordinarily rapid. Within six months in 1990 much of Francophone Africa lifted its ban on multiparty politics. Yet although elections have been held in most of the 45 sub-Saharan states since 1991 (18 in 1996 alone), there have been setbacks for freedom in many countries. One of Africa's most careful observers, Michael Chege, surveyed the wave of democratization and drew the lesson that the continent had "overemphasized multiparty elections . . . and correspondingly neglected the basic tenets of liberal governance." In Central Asia, elections, even when reasonably free, as in Kyrgyzstan and Kazakstan, have resulted in strong executives, weak legislatures and judiciaries, and few civil and economic liberties. In the Islamic world, from the Palestinian Authority to Iran to Pakistan, democratization has led to an increasing role for theocratic politics, eroding long-standing traditions of secularism and tolerance. In many parts of that world, such as Tunisia, Morocco, Egypt, and some of the Gulf States, were elections to be held tomorrow, the resulting regimes would almost certainly be more illiberal than the ones now in place.

Many of the countries of Central Europe, on the other hand, have moved successfully from communism to liberal democracy, having gone through the same phase of liberalization without democracy as other European countries did during the nineteenth century. Indeed, the Austro-Hungarian empire, to which most belonged, was a classic liberal autocracy. Even outside Europe, the political scientist Myron Weiner detected a striking connection between a constitutional past and a liberal democratic present. He pointed out that, as of 1983, "every single country in the Third World that emerged from colonial rule since the Second World War with a population of at least one million (and almost all the smaller colonies as well) with a continuous democratic experience is a former British colony." British rule meant not democracy—colonialism is by definition undemocratic—but constitutional liberalism. Britain's legacy of law and administration has proved more beneficial than France's policy of enfranchising some of its colonial populations.

While liberal autocracies may have existed in the past, can one imagine them today? Until recently, a small but powerful example flourished off the Asian mainland—Hong Kong. For 156 years, until July 1, 1997, Hong Kong was ruled by the British Crown through an appointed governor general. Until 1991 it had never held a meaningful election, but its government epitomized constitutional liberalism, protecting its citizens' basic rights and administering a fair court system and bureaucracy. A September 8, 1997, editorial on the island's future in The Washington Post was titled ominously, "Undoing Hong Kong's Democracy." Actually, Hong Kong has precious little democracy to undo; what it has is a framework of rights and laws. Small islands may not hold much practical significance in today's world, but they do help one weigh the relative value of democracy and constitutional liberalism. Consider, for example, the question of where you would rather live, Haiti, an illiberal democracy, or Antigua, a liberal semi-democracy. Your choice would probably relate not to the weather, which is pleasant in both, but to the political climate, which is not.


John Stuart Mill opened his classic On Liberty by noting that as countries became democratic, people tended to believe that "too much importance had been attached to the limitation of power itself. That . . . was a response against rulers whose interests were opposed to those of the people." Once the people were themselves in charge, caution was unnecessary. "The nation did not need to be protected against its own will." As if confirming Mill's fears, consider the words of Alexandr Lukashenko after being elected president of Belarus with an overwhelming majority in a free election in 1994, when asked about limiting his powers: "There will be no dictatorship. I am of the people, and I am going to be for the people."

The tension between constitutional liberalism and democracy centers on the scope of governmental authority. Constitutional liberalism is about the limitation of power, democracy about its accumulation and use. For this reason, many eighteenth- and nineteenth-century liberals saw in democracy a force that could undermine liberty. James Madison explained in The Federalist that "the danger of oppression" in a democracy came from "the majority of the community." Tocqueville warned of the "tyranny of the majority," writing, "The very essence of democratic government consists in the absolute sovereignty of the majority."

The tendency for a democratic government to believe it has absolute sovereignty (that is, power) can result in the centralization of authority, often by extraconstitutional means and with grim results. Over the last decade, elected governments claiming to represent the people have steadily encroached on the powers and rights of other elements in society, a usurpation that is both horizontal (from other branches of the national government) and vertical (from regional and local authorities as well as private businesses and other nongovernmental groups). Lukashenko and Peru's Alberto Fujimori are only the worst examples of this practice. (While Fujimori's actions—disbanding the legislature and suspending the constitution, among others—make it difficult to call his regime democratic, it is worth noting that he won two elections and was extremely popular until recently.) Even a bona fide reformer like Carlos Menem has passed close to 300 presidential decrees in his eight years in office, about three times as many as all previous Argentinean presidents put together, going back to 1853. Kyrgyzstan's Askar Akayev, elected with 60 percent of the vote, proposed enhancing his powers in a referendum that passed easily in 1996. His new powers include appointing all top officials except the prime minister, although he can dissolve parliament if it turns down three of his nominees for the latter post.

Horizontal usurpation, usually by presidents, is more obvious, but vertical usurpation is more common. Over the last three decades, the Indian government has routinely disbanded state legislatures on flimsy grounds, placing regions under New Delhi's direct rule. In a less dramatic but typical move, the elected government of the Central African Republic recently ended the long-standing independence of its university system, making it part of the central state apparatus.

Usurpation is particularly widespread in Latin America and the states of the former Soviet Union, perhaps because both regions mostly have presidencies. These systems tend to produce strong leaders who believe that they speak for the people—even when they have been elected by no more than a plurality. (As Juan Linz points out, Salvador Allende was elected to the Chilean presidency in 1970 with only 36 percent of the vote. In similar circumstances, a prime minister would have had to share power in a coalition government.) Presidents appoint cabinets of cronies, rather than senior party figures, maintaining few internal checks on their power. And when their views conflict with those of the legislature, or even the courts, presidents tend to "go to the nation," bypassing the dreary tasks of bargaining and coalition-building. While scholars debate the merits of presidential versus parliamentary forms of government, usurpation can occur under either, absent well-developed alternate centers of power such as strong legislatures, courts, political parties, regional governments, and independent universities and media. Latin America actually combines presidential systems with proportional representation, producing populist leaders and multiple parties—an unstable combination.

Many Western governments and scholars have encouraged the creation of strong and centralized states in the Third World. Leaders in these countries have argued that they need the authority to break down feudalism, split entrenched coalitions, override vested interests, and bring order to chaotic societies. But this confuses the need for a legitimate government with that for a powerful one. Governments that are seen as legitimate can usually maintain order and pursue tough policies, albeit slowly, by building coalitions. After all, few claim that governments in developing countries should not have adequate police powers; the trouble comes from all the other political, social, and economic powers that they accumulate. In crises like civil wars, constitutional governments might not be able to rule effectively, but the alternative—states with vast security apparatuses that suspend constitutional rights—has usually produced neither order nor good government. More often, such states have become predatory, maintaining some order but also arresting opponents, muzzling dissent, nationalizing industries, and confiscating property. While anarchy has its dangers, the greatest threats to human liberty and happiness in this century have been caused not by disorder but by brutally strong, centralized states, like Nazi Germany, Soviet Russia, and Maoist China. The Third World is littered with the bloody handiwork of strong states.

Historically, unchecked centralization has been the enemy of liberal democracy. As political participation increased in Europe over the nineteenth century, it was accommodated smoothly in countries such as England and Sweden, where medieval assemblies, local governments, and regional councils had remained strong. Countries like France and Prussia, on the other hand, where the monarchy had effectively centralized power (both horizontally and vertically), often ended up illiberal and undemocratic. It is not a coincidence that in twentieth-century Spain, the beachhead of liberalism lay in Catalonia, for centuries a doggedly independent and autonomous region. In America, the presence of a rich variety of institutions—state, local, and private—made it much easier to accommodate the rapid and large extensions in suffrage that took place in the early nineteenth century. Arthur Schlesinger Sr. has documented how, during America's first 50 years, virtually every state, interest group and faction tried to weaken and even break up the federal government. More recently, India's semi-liberal democracy has survived because of, not despite, its strong regions and varied languages, cultures, and even castes. The point is logical, even tautological: pluralism in the past helps ensure political pluralism in the present.

Fifty years ago, politicians in the developing world wanted extraordinary powers to implement then-fashionable economic doctrines, like nationalization of industries. Today their successors want similar powers to privatize those very industries. Menem's justification for his methods is that they are desperately needed to enact tough economic reforms. Similar arguments are made by Abdalá Bucarem of Ecuador and by Fujimori. Lending institutions, such as the International Monetary Fund and the World Bank, have been sympathetic to these pleas, and the bond market has been positively exuberant. But except in emergencies like war, illiberal means are in the long run incompatible with liberal ends. Constitutional government is in fact the key to a successful economic reform policy. The experience of East Asia and Central Europe suggests that when regimes—whether authoritarian, as in East Asia, or liberal democratic, as in Poland, Hungary, and the Czech Republic—protect individual rights, including those of property and contract, and create a framework of law and administration, capitalism and growth will follow. In a recent speech at the Woodrow Wilson International Center in Washington, explaining what it takes for capitalism to flourish, Federal Reserve chairman Alan Greenspan concluded that, "The guiding mechanism of a free market economy. . . is a bill of rights, enforced by an impartial judiciary."

Finally, and perhaps more important, power accumulated to do good can be used subsequently to do ill. When Fujimori disbanded parliament, his approval ratings shot up to their highest ever. But recent opinion polls suggest that most of those who once approved of his actions now wish he were more constrained. In 1993 Boris Yeltsin famously (and literally) attacked the Russian parliament, prompted by parliament's own unconstitutional acts. He then suspended the constitutional court, dismantled the system of local governments, and fired several provincial governors. From the war in Chechnya to his economic programs, Yeltsin has displayed a routine lack of concern for constitutional procedures and limits. He may well be a liberal democrat at heart, but Yeltsin's actions have created a Russian super-presidency. We can only hope his successor will not abuse it.

For centuries Western intellectuals have had a tendency to view constitutional liberalism as a quaint exercise in rule-making, mere formalism that should take a back seat to battling larger evils in society. The most eloquent counterpoint to this view remains an exchange in Robert Bolt's play A Man For All Seasons. The fiery young William Roper, who yearns to battle evil, is exasperated by Sir Thomas More's devotion to the law. More gently defends himself.

More: What would you do? Cut a great road through the law to get after the Devil?

Roper: I'd cut every law in England to do that!

More: And when the last law was down, and the Devil turned on you—where would you hide Roper, the laws all being flat?


On December 8, 1996, Jack Lang made a dramatic dash to Belgrade. The French celebrity politician, formerly minister of culture, had been inspired by the student demonstrations involving tens of thousands against Slobodan Milosevic, a man Lang and many Western intellectuals held responsible for the war in the Balkans. Lang wanted to lend his moral support to the Yugoslav opposition. The leaders of the movement received him in their offices—the philosophy department—only to boot him out, declare him "an enemy of the Serbs," and order him to leave the country. It turned out that the students opposed Milosevic not for starting the war, but for failing to win it.

Lang's embarrassment highlights two common, and often mistaken, assumptions—that the forces of democracy are the forces of ethnic harmony and of peace. Neither is necessarily true. Mature liberal democracies can usually accommodate ethnic divisions without violence or terror and live in peace with other liberal democracies. But without a background in constitutional liberalism, the introduction of democracy in divided societies has actually fomented nationalism, ethnic conflict, and even war. The spate of elections held immediately after the collapse of communism were won in the Soviet Union and Yugoslavia by nationalist separatists and resulted in the breakup of those countries. This was not in and of itself bad, since those countries had been bound together by force. But the rapid secessions, without guarantees, institutions, or political power for the many minorities living within the new countries, have caused spirals of rebellion, repression, and, in places like Bosnia, Azerbaijan, and Georgia, war.

Elections require that politicians compete for peoples' votes. In societies without strong traditions of multiethnic groups or assimilation, it is easiest to organize support along racial, ethnic, or religious lines. Once an ethnic group is in power, it tends to exclude other ethnic groups. Compromise seems impossible; one can bargain on material issues like housing, hospitals, and handouts, but how does one split the difference on a national religion? Political competition that is so divisive can rapidly degenerate into violence. Opposition movements, armed rebellions, and coups in Africa have often been directed against ethnically based regimes, many of which came to power through elections. Surveying the breakdown of African and Asian democracies in the 1960s, two scholars concluded that democracy "is simply not viable in an environment of intense ethnic preferences." Recent studies, particularly of Africa and Central Asia, have confirmed this pessimism. A distinguished expert on ethnic conflict, Donald Horowitz, concluded, "In the face of this rather dismal account . . . of the concrete failures of democracy in divided societies . . . one is tempted to throw up one's hands. What is the point of holding elections if all they do in the end is to substitute a Bemba-dominated regime for a Nyanja regime in Zambia, the two equally narrow, or a southern regime for a northern one in Benin, neither incorporating the other half of the state?"

Over the past decade, one of the most spirited debates among scholars of international relations concerns the "democratic peace"—the assertion that no two modern democracies have gone to war with each other. The debate raises interesting substantive questions (does the American Civil War count? do nuclear weapons better explain the peace?) and even the statistical findings have raised interesting dissents. (As the scholar David Spiro points out, given the small number of both democracies and wars over the last two hundred years, sheer chance might explain the absence of war between democracies. No member of his family has ever won the lottery, yet few offer explanations for this impressive correlation.) But even if the statistics are correct, what explains them? Kant, the original proponent of the democratic peace, contended that in democracies, those who pay for wars—that is, the public—make the decisions, so they are understandably cautious. But that claim suggests that democracies are more pacific than other states. Actually they are more warlike, going to war more often and with greater intensity than most states. It is only with other democracies that the peace holds.

When divining the cause behind this correlation, one thing becomes clear: the democratic peace is actually the liberal peace. Writing in the eighteenth century, Kant believed that democracies were tyrannical, and he specifically excluded them from his conception of "republican" governments, which lived in a zone of peace. Republicanism, for Kant, meant a separation of powers, checks and balances, the rule of law, protection of individual rights, and some level of representation in government (though nothing close to universal suffrage). Kant's other explanations for the "perpetual peace" between republics are all closely linked to their constitutional and liberal character: a mutual respect for the rights of each other's citizens, a system of checks and balances assuring that no single leader can drag his country into war, and classical liberal economic policies—most importantly, free trade—which create an interdependence that makes war costly and cooperation useful. Michael Doyle, the leading scholar on the subject, confirms in his 1997 book Ways of War and Peace that without constitutional liberalism, democracy itself has no peace-inducing qualities:

Kant distrusted unfettered, democratic majoritarianism, and his argument offers no support for a claim that all participatory polities—democracies—should be peaceful, either in general or between fellow democracies. Many participatory polities have been non-liberal. For two thousand years before the modern age, popular rule was widely associated with aggressiveness (by Thucydides) or imperial success (by Machiavelli) . . . The decisive preference of [the] median voter might well include "ethnic cleansing" against other democratic polities.

The distinction between liberal and illiberal democracies sheds light on another striking statistical correlation. Political scientists Jack Snyder and Edward Mansfield contend, using an impressive data set, that over the last 200 years democratizing states went to war significantly more often than either stable autocracies or liberal democracies. In countries not grounded in constitutional liberalism, the rise of democracy often brings with it hyper-nationalism and war-mongering. When the political system is opened up, diverse groups with incompatible interests gain access to power and press their demands. Political and military leaders, who are often embattled remnants of the old authoritarian order, realize that to succeed they must rally the masses behind a national cause. The result is invariably aggressive rhetoric and policies, which often drag countries into confrontation and war. Noteworthy examples range from Napoleon III's France, Wilhelmine Germany, and Taisho Japan to those in today's newspapers, like Armenia and Azerbaijan and Milosevic's Serbia. The democratic peace, it turns out, has little to do with democracy.


An American scholar recently traveled to Kazakstan on a U.S. government-sponsored mission to help the new parliament draft its electoral laws. His counterpart, a senior member of the Kazak parliament, brushed aside the many options the American expert was outlining, saying emphatically, "We want our parliament to be just like your Congress." The American was horrified, recalling, "I tried to say something other than the three words that had immediately come screaming into my mind: 'No you don't!'" This view is not unusual. Americans in the democracy business tend to see their own system as an unwieldy contraption that no other country should put up with. In fact, the adoption of some aspects of the American constitutional framework could ameliorate many of the problems associated with illiberal democracy. The philosophy behind the U.S. Constitution, a fear of accumulated power, is as relevant today as it was in 1789. Kazakstan, as it happens, would be particularly well-served by a strong parliament—like the American Congress—to check the insatiable appetite of its president.

It is odd that the United States is so often the advocate of elections and plebiscitary democracy abroad. What is distinctive about the American system is not how democratic it is but rather how undemocratic it is, placing as it does multiple constraints on electoral majorities. Of its three branches of government, one—arguably paramount—is headed by nine unelected men and women with life tenure. Its Senate is the most unrepresentative upper house in the world, with the lone exception of the House of Lords, which is powerless. (Every state sends two senators to Washington regardless of its population—California's 30 million people have as many votes in the Senate as Arizona's 3.7 million—which means that senators representing about 16 percent of the country can block any proposed law.) Similarly, in legislatures all over the United States, what is striking is not the power of majorities but that of minorities. To further check national power, state and local governments are strong and fiercely battle every federal intrusion onto their turf. Private businesses and other nongovernmental groups, what Tocqueville called intermediate associations, make up another stratum within society.

The American system is based on an avowedly pessimistic conception of human nature, assuming that people cannot be trusted with power. "If men were angels," Madison famously wrote, "no government would be necessary." The other model for democratic governance in Western history is based on the French Revolution. The French model places its faith in the goodness of human beings. Once the people are the source of power, it should be unlimited so that they can create a just society. (The French Revolution, as Lord Acton observed, is not about the limitation of sovereign power but the abrogation of all intermediate powers that get in its way.) Most non-Western countries have embraced the French model—not least because political elites like the prospect of empowering the state, since that means empowering themselves—and most have descended into bouts of chaos, tyranny, or both. This should have come as no surprise. After all, since its revolution France itself has run through two monarchies, two empires, one proto-fascist dictatorship, and five republics.

Of course cultures vary, and different societies will require different frameworks of government. This is not a plea for the wholesale adoption of the American way but rather for a more variegated conception of liberal democracy, one that emphasizes both parts of that phrase. Before new policies can be adopted, there lies an intellectual task of recovering the constitutional liberal tradition, central to the Western experience and to the development of good government throughout the world. Political progress in Western history has been the result of a growing recognition over the centuries that, as the Declaration of Independence puts it, human beings have "certain inalienable rights" and that "it is to secure these rights that governments are instituted." If a democracy does not preserve liberty and law, that it is a democracy is a small consolation.


A proper appreciation of constitutional liberalism has a variety of implications for American foreign policy. First, it suggests a certain humility. While it is easy to impose elections on a country, it is more difficult to push constitutional liberalism on a society. The process of genuine liberalization and democratization is gradual and long-term, in which an election is only one step. Without appropriate preparation, it might even be a false step. Recognizing this, governments and nongovernmental organizations are increasingly promoting a wide array of measures designed to bolster constitutional liberalism in developing countries. The National Endowment for Democracy promotes free markets, independent labor movements, and political parties. The U.S. Agency for International Development funds independent judiciaries. In the end, however, elections trump everything. If a country holds elections, Washington and the world will tolerate a great deal from the resulting government, as they have with Yeltsin, Akayev, and Menem. In an age of images and symbols, elections are easy to capture on film. (How do you televise the rule of law?) But there is life after elections, especially for the people who live there.

Conversely, the absence of free and fair elections should be viewed as one flaw, not the definition of tyranny. Elections are an important virtue of governance, but they are not the only virtue. Governments should be judged by yardsticks related to constitutional liberalism as well. Economic, civil, and religious liberties are at the core of human autonomy and dignity. If a government with limited democracy steadily expands these freedoms, it should not be branded a dictatorship. Despite the limited political choice they offer, countries like Singapore, Malaysia, and Thailand provide a better environment for the life, liberty, and happiness of their citizens than do either dictatorships like Iraq and Libya or illiberal democracies like Slovakia or Ghana. And the pressures of global capitalism can push the process of liberalization forward. Markets and morals can work together. Even China, which remains a deeply repressive regime, has given its citizens more autonomy and economic liberty than they have had in generations. Much more needs to change before China can even be called a liberalizing autocracy, but that should not mask the fact that much has changed.

Finally, we need to revive constitutionalism. One effect of the overemphasis on pure democracy is that little effort is given to creating imaginative constitutions for transitional countries. Constitutionalism, as it was understood by its greatest eighteenth century exponents, such as Montesquieu and Madison, is a complicated system of checks and balances designed to prevent the accumulation of power and the abuse of office. This is done not by simply writing up a list of rights but by constructing a system in which government will not violate those rights. Various groups must be included and empowered because, as Madison explained, "ambition must be made to counteract ambition." Constitutions were also meant to tame the passions of the public, creating not simply democratic but also deliberative government. Unfortunately, the rich variety of unelected bodies, indirect voting, federal arrangements, and checks and balances that characterized so many of the formal and informal constitutions of Europe are now regarded with suspicion. What could be called the Weimar syndrome—named after interwar Germany's beautifully constructed constitution, which failed to avert fascism—has made people regard constitutions as simply paperwork that cannot make much difference. (As if any political system in Germany would have easily weathered military defeat, social revolution, the Great Depression, and hyperinflation.) Procedures that inhibit direct democracy are seen as inauthentic, muzzling the voice of the people. Today around the world we see variations on the same majoritarian theme. But the trouble with these winner-take-all systems is that, in most democratizing countries, the winner really does take all.


We live in a democratic age. Through much of human history the danger to an individual's life, liberty and happiness came from the absolutism of monarchies, the dogma of churches, the terror of dictatorships, and the iron grip of totalitarianism. Dictators and a few straggling totalitarian regimes still persist, but increasingly they are anachronisms in a world of global markets, information, and media. There are no longer respectable alternatives to democracy; it is part of the fashionable attire of modernity. Thus the problems of governance in the 21st century will likely be problems within democracy. This makes them more difficult to handle, wrapped as they are in the mantle of legitimacy.

Illiberal democracies gain legitimacy, and thus strength, from the fact that they are reasonably democratic. Conversely, the greatest danger that illiberal democracy poses—other than to its own people—is that it will discredit liberal democracy itself, casting a shadow on democratic governance. This would not be unprecedented. Every wave of democracy has been followed by setbacks in which the system was seen as inadequate and new alternatives were sought by ambitious leaders and restless masses. The last such period of disenchantment, in Europe during the interwar years, was seized upon by demagogues, many of whom were initially popular and even elected. Today, in the face of a spreading virus of illiberalism, the most useful role that the international community, and most importantly the United States, can play is—instead of searching for new lands to democratize and new places to hold elections—to consolidate democracy where it has taken root and to encourage the gradual development of constitutional liberalism across the globe. Democracy without constitutional liberalism is not simply inadequate, but dangerous, bringing with it the erosion of liberty, the abuse of power, ethnic divisions, and even war. Eighty years ago, Woodrow Wilson took America into the twentieth century with a challenge, to make the world safe for democracy. As we approach the next century, our task is to make democracy safe for the world.

You are reading a free article.

Subscribe to Foreign Affairs to get unlimited access.

  • Paywall-free reading of new articles and a century of archives
  • Unlock access to iOS/Android apps to save editions for offline reading
  • Six issues a year in print, online, and audio editions
Subscribe Now