A short while ago, one of the world’s most brutal and entrenched dictatorships was swiftly toppled by the military force of the United States and the United Kingdom. The 2003 Iraq war was launched to disarm Saddam Hussein, but for many of its advocates and supporters, the more compelling aim was to bring about regime change. In fact, the goal is not simply “regime change” but a sweeping political transformation in that country — and, it is hoped, in states throughout its neighborhood — towards what has never existed there before: democracy.

This is the most ambitious effort to foster deliberate political change since European colonial rule drew to a close in the early post-World War ii era. Can it succeed? Since Iraq lacks virtually all of the classic favorable conditions, to ask whether it can soon become a democracy is to ask, really, whether any country can become a democracy. Which is to ask as well, can every country become a democracy?

My answer here is a cautiously optimistic one. The current moment is in many respects without historical precedent. Much is made of the unparalleled gap between the military and economic power of the United States and that of any conceivable combination of competitors or adversaries. But no less unique are these additional facts:

• This breathtaking preponderance of power is held by a liberal democracy.
• The next most powerful global actor is a loose union of countries that are also all liberal democracies.
• The majority of states in the world are already democracies of one sort or another.
• There is no model of governance with any broad normative appeal or legitimacy in the world other than democracy.
• There is growing international legal and moral momentum toward the recognition of democracy as a basic human right of all peoples.
• States and international organizations are intruding on sovereignty in ever more numerous and audacious ways in order to promote democracy and freedom.

In short, the international context has never mattered more to the future of democracy or been more favorable. We are on the cusp of a grand historical tipping point, when a visionary and resourceful strategy could — if it garnered the necessary cooperation and effort among the powerful democracies — essentially eliminate authoritarian rule over the next generation or two.1

A quarter-century of progress

As samuel p. huntington has documented in his seminal work, The Third Wave: Democratization in the Late Twentieth Century (University of Oklahoma Press, 1991), a powerful wave of democratic transitions began in April 1974, when the Portuguese dictatorship was overthrown in a military coup. It was far from clear then that Portugal would become a democracy. It had never been one before. It had just been through half a century of quasi-fascist rule. The Spanish dictator Francisco Franco held on to power over the border. Both countries were steeped in a Latin, Catholic culture that was dismissed by many political scientists and commentators as being unsuited to democracy. (That logic was also used to explain the virtual absence of democracy in Latin America at the time.) The Portuguese armed forces movement was split into ideological factions, and the country was plagued for 18 months by coups, counter-coups, and a succession of fragile provisional governments. Yet the triumph of democracy in Portugal was the beginning of a long wave of democratic expansion in the world that continues to this day.

When the third wave of democratization began in 1974, there were only about 40 democracies in the world, and these were mainly in the advanced industrial countries. There were a few other democracies scattered through Africa, Asia, and Latin America — such as India, Sri Lanka, Botswana, Costa Rica, and Venezuela. But military and one-party dictatorships held sway in most of Latin America, Asia, Africa, and the Middle East, while all of Eastern Europe and the Soviet Union were under communist rule.

Since 1974, democracy — a system of government in which the people choose their leaders at regular intervals through free, fair, and competitive elections — has expanded dramatically in the world. The number and percentage of democracies in the world expanded gradually after April 1974, spreading first to Greece and Spain in the mid-1970s. From 1979 to 1985, the military withdrew in favor of elected civilian governments in about nine Latin American countries. Where military rule was more economically successful, in Chile, the transition was delayed, but it came in 1989, after a heroic effort of peaceful political mobilization.

By then, the third wave of democratization had spread to Asia, first toppling the dictatorship of Ferdinand Marcos in the Philippines in February 1986, then forcing the complete withdrawal of the Korean military in 1987. That same year, martial law was lifted in Taiwan and a more gradual transition to democracy began there, not to be completed until the first direct elections for president in 1996. But by 1991, Pakistan, Bangladesh, and Nepal had all become democracies. Also that year, Thailand suffered what I believe will prove to have been its last military coup, followed by its shortest period of military rule.

By 1987, the third wave had spread to the point where about two of every five states in the world were democracies: all of Western Europe, much of Asia, and most of Latin America. But that still left gaping holes in Eastern Europe, Africa, and the Middle East. Democracy was still a regional phenomenon. This changed dramatically with the fall of the Berlin Wall in 1989 and then the collapse of the Soviet Union in 1991. By 1990, most of the states of Eastern Europe — and even poor and isolated Mongolia — held competitive elections and began to institutionalize democracy.

Freed from the prism of the two superpowers’ struggle for geopolitical dominance, and reeling from desperate fiscal crises, African countries began to liberate themselves. In February 1990 two seminal events launched a new wave of democratic transitions in Africa. In Benin, a coalition of forces in civil society, organized in a “sovereign national conference,” claimed governing authority and launched a transition to democracy. In South Africa, the apartheid regime released Nelson Mandela from prison and launched a process of political dialogue and normalization that gave birth to democracy in 1994. When these two events occurred, there were only three democracies in Africa — the Gambia, Botswana, and Mauritius. But starting in 1990, Africa experienced a rolling tide of democratic change. Under heavy pressure from international donors as well as their own peoples, most African states by 1997 had at least legalized opposition parties, opened space in civil society, and held multiparty elections. Many of these openings were largely a facade, marred by continued repression and blatant rigging of the vote. But well over a dozen met the minimum conditions of democracy, and in several cases, long-ruling incumbent parties were defeated.

To appreciate the depth and breadth of the third wave of democratization, consider this: In 1974, there were 41 democracies among the existing 150 states. Of the remaining 109 states, 56 (more than half) of them subsequently made a transition to democracy, and of those 56, only Pakistan, Sudan, and Russia are not democracies today. Moreover, 26 states since 1974 have become independent of colonial rule; 15 of these became democracies upon independence and have remained so, and another six have become democratic after some period of authoritarian rule. Of the 19 new postcommunist states, 11 (58 percent) are democracies. Overall, of the 45 new states created since the third wave began, almost three-quarters (71 percent) are democracies, though in the case of the former Soviet Union, some of them (such as Ukraine, Georgia, and Armenia) are only ambiguously democratic.

As democracy spread to Eastern Europe, a few states in the former Soviet Union, and a number in Africa, while extending deeper into Asia and Latin America, it came during the 1990s to be a global phenomenon, the predominant form of government, and the only broadly legitimate form of government in the world. Today, about three-fifths of all the world’s states (by the count of Freedom House, 121 of 193) are democracies. There are no global rivals to democracy as a broad model of government. Communism is dead. Military rule everywhere lacks appeal and normative justification. One-party states have largely disappeared, for what single party — in this day and age — can credibly claim the wisdom and moral righteousness to rule indefinitely and without criticism or challenge? Only the vague model of an Islamic state has any moral and ideological appeal as an alternative form of government — and then only for a small portion of the world’s societies. Moreover, the only actual example of such an Islamic state is the increasingly corrupt, discredited, and illegitimate Islamic Republic in Iran, whose own people overwhelmingly desire to see it replaced by a more truly democratic form of government.

Clearly, most states can become democratic because most states already are. Moreover, the overwhelming bulk of the states that have become democratic during the third wave have remained so, even in countries lacking virtually all of the supposed “conditions” for democracy. Pre-1990 Africa aside, only four democracies have been overthrown by the military in a conventional coup. Two of those (Turkey and Thailand) returned fairly quickly to democracy, and the other two (Pakistan and the Gambia) have felt compelled at least to institute civilian multiparty elections. Several democracies have been suspended in “self-coups” by elected civilian leaders, while other elected rulers have more subtly strangled democracy. Overall, however, only 14 of the 125 democracies that have existed during the third wave have become authoritarian, and in nine of these, democracy has since been restored.

If democracy can emerge and persist (now so far for a decade) in an extremely poor, landlocked, overwhelmingly Muslim country like Mali — in which the majority of adults are illiterate and live in absolute poverty and the life expectancy is 44 years — then there is no reason in principle why democracy cannot develop in most other very poor countries. In fact, if we examine the 36 countries that the United Nations Development Programme (undp) classifies as having “Low Human Development,” 11 are democracies today. If we widen our scope to look at the bottom third of states classified by the undp, the percentage of democracies rises from nearly a third to 41 percent. About a dozen of these have been democracies for a decade or longer. That there should be so many democracies among the world’s least developed countries is a phenomenon at least as noteworthy as the overall predominance of democracy in the world, and one profoundly in defiance of established social science theories. It deserves more attention, and I attempt to give it some below.

Conceptualizing democracy

To comprehend the nature and limits of democratic progress in this third wave, it is useful to conceive of democracy in terms of two thresholds. Countries above the first threshold are, as I have suggested, electoral democracies in the minimal sense that their principal positions of political power are filled through regular, free, fair, and competitive (and therefore multiparty) elections. Electoral democracy can exist in countries with significant violations of human rights, massive corruption, and a weak rule of law. But in order for a country to be a democracy, these defects must be sufficiently contained so that, in elections at least, the will of the voters can be reflected in the outcome and, in particular, unpopular incumbents can be booted from office.2

Many people have criticized my emphasis on free, fair, meaningful, and competitive elections as the minimal litmus test of democracy. They say that this may not amount to much. What is the point of having such an “electoral” democracy if the rights of women, minorities, and the poor are extensively violated; if those who are elected take turns plundering the national treasury and abusing power, as happened in Pakistan before the October 1999 coup; if elections merely crown a temporary presidential monarch who can use and abuse power without constraint for his term of office (what University of Notre Dame professor Guillermo O’Donnell calls “delegative democracy”)? Indeed, in his new book, The Future of Freedom: Illiberal Democracy at Home and Abroad (Norton, 2003), Fareed Zakaria questions whether we might not do better with less democracy and more rule of law.

My answer to this is twofold. Normatively, I do not argue that we should rest content with such an illiberal and hollowed-out democracy as our goal. The goal for every country should be a political system that combines democracy on the one hand with freedom, the rule of law, and good government on the other — in other words, liberal democracy. Beyond the electoral arena, liberal democracy encompasses a vigorous rule of law with an independent and nondiscriminatory judiciary; extensive individual freedoms of belief, speech, publication, association, assembly, and so on; strong protections for the rights of ethnic, cultural, religious, and other minorities; a pluralistic civil society, which affords citizens multiple channels outside of the electoral arena through which to participate and express their interests and values; and civilian control of the military.3

Empirically, the implication that authoritarian and conflict-ridden states should emphasize the rule of law rather than democracy is viable only as a transitional strategy. In reality, democracy and freedom are closely related in the world. Even if we forget about the wealthy countries of the West — all liberal democracies — and examine only the developing and postcommunist countries, we find that the countries where civil liberties and the rule of law are best respected are democracies, and the human rights (and humanitarian) emergencies are invariably to be found in non-democracies.

Each year Freedom House rates each country from 1 to 7 along two scales, political rights (basically to participate and compete democratically) and civil liberties, with 1 being most free and 7 most repressive. There are only two countries in the world that are not democracies and yet have a civil-liberties score below the midpoint on the seven-point scale: Tonga, and Antigua and Barbuda. One can hardly advance a general theory of political development based on these two microstates. To be sure, there are some pretty illiberal democracies in the world, with serious problems of human rights and the rule of law, but the only countries that give their citizens extensive civic freedom and a thorough rule of law are democracies.

The regional distribution

In one respect, democracy is still not quite a global phenomenon. In every region of the world — except for one — at least a third of the states are democracies. Thirty of the 33 states in Latin America and the Caribbean are democracies, and about half of them are now fairly liberal in terms of their levels of freedom. Two-thirds of the former communist countries, half of the Asian states, and even about two-fifths of the African states are now democracies. Only in the Middle East is democracy virtually absent. In fact, among the 16 Arab countries, there is not a single democracy and, with the exception of Lebanon, there never has been.

The exceptionalism of the Middle East becomes even more striking when we examine trends in freedom. Every region of the world has seen a rather significant improvement in the level of freedom — except for one. Regions that had been strongholds of authoritarianism have seen their average freedom score on the combined seven-point scale improve by at least a point. There is only one region of the world where the average level of freedom has declined, by almost half a point — again, the Middle East.

Some skeptics believe that democracy is largely a Western, Judeo-Christian phenomenon that is not well suited to other regions, cultures, and religious traditions. They have a ready answer for this freedom gap: Islam. I believe this answer is wrong on substantive grounds that I will come to shortly, but it is also questionable empirically. There are 43 countries in the world that pretty clearly have a Muslim majority. The 27 of these outside the Arab world have an average freedom score (5.04) appreciably better than the Arab states (5.81). A quarter (seven) of these 27 non-Arab, Muslim-majority states are democracies. Moreover, as Columbia University’s Alfred Stepan has shown, non-Arab Muslim countries have some considerable cumulative experience over the past 30 years with political freedom.4

Democracy, then, exists in virtually all types of states. It is significantly present in almost every region of the world. It is present in countries evincing every major religious or philosophical tradition: Christian, Jewish, Hindu, Buddhist, Confucian, and Muslim. It is much more common in developed countries (all of the top 20 countries in human development are liberal democracies), but it is now significantly present among very poor countries as well. It is much more common — and much more liberal — in small states of under 1 million. But most of the biggest countries — specifically, eight of the 11 countries with populations over 100 million — are democracies.5 By any category that is meaningful in the world today, there is only one set of countries that is completely undemocratic: the Arab world. >

Democracy as a universal value?

There is a possible retort to this claim that democracy is present in virtually every major region of the world and thus is nearly a universal phenomenon. One could dismiss this as a fad, or a contemporary concession to international pressure: Democracy may exist today in far reaches of the globe, but only temporarily and superficially. It is not really valued by the people, and it will not last.

What about persistence? Forty years ago Seymour Martin Lipset argued that the richer the country the greater the chance that it would sustain democracy. In a seminal and methodologically sophisticated study, Adam Przeworski and his colleagues found that there was in fact a striking and monotonic relationship between development level and the probability of sustaining democracy. During the period 1950-90, the poorest democracies had a 12 percent chance of dying in any particular year, or an average life expectancy of eight years. Several third-wave democracies in their lowest income category have now outlived that expected life span, including Benin, Mali, Malawi, Mozambique, and Nepal.6 Even among the poorest countries, there have been few breakdowns of democracy.

In fact, a strong case has been made that democracy is not an extravagance for the poor, but very nearly a necessity. Amartya Sen won the Nobel Prize for economics in 1998 in part for showing that democracies do not have famines. “People in economic need,” he argues, “also need a political voice. Democracy is not a luxury that can await the arrival of general prosperity.” Moreover, “there is very little evidence that poor people, given the choice, prefer to reject democracy.”7 He notes the vigor with which Indians defended their freedom and democracy in the 1977 election, tossing from office the prime minister, Indira Gandhi, who had suspended political and civil rights. But there have been countless other instances — from Burma and Bangladesh to Senegal and South Africa — where poor people have mobilized passionately for (and in defense of) democratic change. The fact that they have sometimes, as in Burma, been crushed by sheer force while a timid world watched and protested ineffectually does not negate the overwhelming expression of their sentiment.

Fortunately, we also have more precise evidence from public opinion survey data as to what ordinary people really think. So far, the data show that the understanding and valuing of democracy is widely shared across cultures. Two-thirds of Africans surveyed (by the Afrobarometer poll in 12 mainly poor countries in 2001) associate democracy with civil liberties, popular sovereignty, or electoral choice. About two-thirds of Africans surveyed (69 percent) also say democracy is “always preferable” to authoritarian rule. The same proportion rejects one-party rule, and four in five reject military or one-man rule. Even many who are not satisfied with democracy believe it is the best form of government, and most Africans who live in democracies recognize there are serious institutional problems that must be addressed. Latin Americans — who have had more time than Africans to become disillusioned with how democracy actually performs in their countries — are more ambivalent; but overall, 57 percent still believe democracy is always preferable, and only about 15 percent might prefer an authoritarian regime. In East Asia, according to data from the East Asia Barometer collected in 2001, only a quarter in Taiwan and Korea, about a fifth in Hong Kong and the Philippines, but less than a tenth in Thailand believe that democracy is not really suitable for their country. In all five of these systems, consistently strong majorities (usually upwards of two-thirds) reject authoritarian alternatives to democracy. So do strong majorities (about seven in 10 overall) in the 10 postcommunist countries now negotiating membership in the European Union.

Although much has been made of the “clash of civilizations,” especially since September 11, 2001, the Afrobarometer survey evidence indicates that “Muslims are as supportive of democracy as non-Muslims.” Large majorities of African Muslims as well as non-Muslims support democracy, and any hesitancy in supporting democracy among African Muslims “is due more to deficits of formal education and other attributes of modernization than to religious attachments,” as the Afrobarometer’s analysts put it. Data from Central Asia and the Middle East point in a similar direction.8 The Middle Eastern data are somewhat dated (from the 1990s) and severely limited by what could be asked, but in two of the four country contexts (Egypt and Palestine), a majority attached at least some importance to the value of democracy. Weighing the evidence, political scientist Mark Tessler concludes, “Islam appears to have less influence on political attitudes than is frequently suggested.” Indeed, “support for democracy is not necessarily lower among those individuals with the strongest Islamic attachments.”

These popular orientations among the world’s Muslims correspond with the thinking of increasingly outspoken moderate Muslim intellectuals, who are making the case either for a liberal interpretation of Islam or for a broader liberal view that deemphasizes the literal meaning of sacred Islamic texts while stressing the larger compatibility between the overall moral teachings of Islam and the nature of democracy as a system of government based on such principles as accountability, freedom of expression, and the rule of law. Islam is undergoing a kind of reformation now, and there is growing momentum among Muslim religious thinkers for a separation of mosque and state.

Significantly, Arab thinkers, scholars, and civil society activists are themselves challenging the democracy and freedom deficit that pervades the Arab world. The Arab authors of the Arab Human Development Report — an extraordinary document published by the undp last year — recognize that the global wave of democratization “has barely reached the Arab states. This freedom deficit undermines human development and is one of the most painful manifestations of lagging political development.” It was this same broad team of Arab specialists who wrote these words about the reform imperative:

There can be no real prospects for reforming the system of governance, or for truly liberating human capabilities, in the absence of comprehensive political representation in effective legislatures based on free, honest, efficient and regular elections. If the people’s preferences are to be properly expressed and their interests properly protected, governance must become truly representative and fully accountable.

Amartya Sen argues that the mark of a universal value is not that it has the consent of everyone, but that “people anywhere may have reason to see it as valuable.” By this measure, there is growing evidence of all kinds that democracy is becoming a truly universal value.

Democracy drivers

To assess whether vastly more countries — and, someday, potentially all countries — can become democratic, we must answer four more questions. First, what has been driving democratization in the third wave? Why have so many more countries become democratic during this period? Second, why have so few of these new democracies broken down in the last quarter-century? Third, why do the remaining nondemocracies hold out? Logically, the answers to these questions will then provide essential insights with which to answer the most important question: Can the countries that are not now democratic become so? And how would they do that?

In this essay, I can only sketch the answers to these four questions, beginning with the causes of democratization.

Economic development. As Huntington notes, economic development has been a major driver of democratization in the third wave. However, increases in national wealth bring about pressures for democratization only to the extent that they generate several other intervening effects: rising levels of education; the creation of a complex and diverse middle class that is independent of the state; the development of a more pluralistic, active, and resourceful civil society; and, as a result of all of these changes, the emergence of a more questioning, assertive, pro-democratic political culture.

These broad societal transformations have accompanied economic development in a number of countries in recent decades. South Korea and Taiwan stand as the classic examples of economic growth bringing about diffuse social, economic, and cultural change that then generates diffuse societal pressure for democracy. At a somewhat lower level of economic development, this has also been the story of Thailand, Brazil, Mexico, and South Africa. However, where states have managed successfully to control and co-opt civil society and to manipulate cultural symbols and belief systems in a way that legitimizes semi-authoritarian rule, the internal pressure for democratization has been preempted or deflected. This has been the case with Malaysia and especially Singapore, the richest authoritarian state in the history of the world. Alternatively, some states that look economically developed in terms of their per capita income are much less so when we examine education levels, status of women, civic life, and state-society relations. These are the oil-rich states, whose economic and class structures are grossly distorted by the fact of centralized state control of the oil sector.

Economic development that seeps broadly into the social structure and culture of a society will, in most cases, generate powerful pressures for democratization. The authoritarian rulers capable of managing this process of social and economic change as adeptly as Lee Kuan Yew and his successors in Singapore are few and far between. China’s communist leaders think they can duplicate Lee’s path, but they are wrong.

Economic performance. The second factor that has driven democratic change during the third wave has also been economic, but in the inverse direction of economic crisis, or poor governance performance in general. To the extent that they make an effort to justify their rule on moral and political grounds, conventional authoritarian regimes do so on the basis of performance achievements and imperatives. They claim that their rule is necessary to clean up corruption, fight subversion, unify the country, and/or generate economic growth. This puts authoritarian regimes in a dilemma. If they fail to deliver on these promises, authoritarian rulers forfeit their moral entitlement to rule. Unlike democracy, which people value intrinsically, these run-of-the-mill dictatorships have no other grounds on which to justify their rule except what they can tangibly deliver. Even if they succeed in overcoming the crises of political instability or insecurity that brought them to power, after some time people may feel they have served their purpose (perhaps at great cost to other values) and should go.

However one frames it, performance-based legitimacy is a delicate and perilous strategy for sustaining authoritarian rule indefinitely. Most authoritarian regimes are, in the long run, damned if they do deliver and damned if they don’t. Those that care only about their own survival focus on funneling corrupt payoffs to a narrow support circle of cronies, functionaries, soldiers, and thugs who will repress any opposition. In this way, the regime may survive for some time, as in Iraq, while the country slowly crumbles.

International actions and pressures. The most distinctive feature of the third wave has been the sweeping change in the policies, actions, and expectations of the established democracies, particularly the United States, as well as regional and international organizations. Beginning under Jimmy Carter, with his new emphasis on human rights, and then, after a false start, continuing with the new emphasis on democracy promotion under Ronald Reagan, U.S. presidential administrations became active in pressing for democratic change. New U.S. institutions, such as the National Endowment for Democracy, were created to provide practical assistance and encouragement to democratic movements, civic organizations, interest groups, parties, and institutions. By the late 1990s, the United States was spending over half a billion dollars a year to foster and support democratic development abroad. Direct and indirect diplomatic pressure was exerted. With the end of the Cold War, these pressures widened, and a number of African states that had been pawns on a superpower chessboard were suddenly viewed on their own terms. Many African governments that had been lavishly financed and repeatedly bailed out from their misrule suddenly found themselves in acute fiscal crisis, and thus were forced to reform politically.

It was not just the United States that was pressing for democracy. The European Union became increasingly active and outspoken towards the same ends, particularly in its financial and organizational efforts to promote democracy in postcommunist Europe. The driving wedge of Western Europe’s democratizing impact was a simple and unyielding condition that all states seeking entry into the European Union had to manifest, in the words of the European Community at the time, “truly democratic practices and respect for fundamental rights and freedoms.” Much European Union technical and political assistance over the past 12 years has gone into helping the candidate states for entry meet these political (and other economic) conditions. First in Southern Europe and now in Central and Eastern Europe as well as Turkey, a regression away from democracy has become unthinkable because of the enormous economic and political costs it would impose through isolation from the community of European states and free trade. When we think about the prospects for democratic expansion in the world, this means by which the political will for democracy is generated and entrenched must be borne in mind.

More recently, regional pressure for democracy has begun to take hold in the Americas. In June 1991, the Organization of American States (oas) adopted the “Santiago Commitment to Democracy,” which required immediate consultation if a democracy is overthrown. Concerted action by the oas and by the U.S. and other member states deterred a planned autogolpe in Guatemala in 1993 and a rumored military coup in Paraguay later in the decade. And the oas effectively monitored transitional or controversial elections in a number of its emerging or transitional democratic states.

In fact, international election observation has become one of the most common means by which international actors — the United Nations, regional organizations, other governments, and ngos — intrude, often by invitation, on the internal politics of sovereign countries. These kinds of political intrusions are reshaping the very idea of sovereignty, negating the longstanding presumption that states are free to do what they like within their own borders.

Changing international norms and conventions. Finally, then, what has changed during the third wave is the normative weight given to human rights — and to democracy as a human right — in international discourse, treaties, law, and collective actions. The world community is increasingly embracing a shared normative expectation that all states seeking international legitimacy should manifestly “govern with the consent of the governed” — in essence, a “right to democratic governance” is seen as a legal entitlement.9 Already effectively implied by the Universal Declaration of Human Rights and the International Covenant on Civil and Political Rights, this right to democratic governance has been articulated more and more explicitly in the documents of regional organizations and affirmed by the growing number of interventions by those organizations and by the United Nations. In June 2000, 106 states, gathered in the “Toward a Community of Democracies” conference, agreed to “respect and uphold” a detailed list of “core democratic principles and practices” — including individual liberties, the rule of law enforced by an “independent and impartial judiciary,” and “the right and civic duties of citizens to choose their representatives through regular, free and fair elections with universal and equal suffrage, open to multiple parties, conducted by secret ballot, monitored by independent electoral authorities, and free of fraud and intimidation.”

At a minimum, this evolution has done two things. First, it has lowered the political threshold for intervention, not only for the multilateral actors but for states and ngos as well. Second, it has emboldened domestic advocates of democracy and human rights. No factor has been more important in driving and sustaining the third wave of democratization than this cluster of international normative and legal trends.

Fewer breakdowns

Drawing from the causal factors above, we can identify three factors that have provided a strong degree of immunity to democratic breakdown during the third wave. First, some countries became democracies after they had become relatively rich — in fact, richer than any country that has ever suffered a breakdown of democracy. Przeworski and his colleagues found that from 1950 to 1990, no country with a per capita income higher than $6,055 (in 1985 Purchasing Power Parity dollars) had ever suffered a breakdown of democracy. (This was the 1975 per capita income of Argentina, the richest country ever to have suffered a coup against democracy.) The equivalent level of economic development in 2000 dollars is $8,773. Taiwan and Korea became democracies at levels of economic development richer than this and are now much richer than this. Several democracies in Central and Eastern Europe and Latin America are also beyond this level.

The second factor is public opinion and normative change within countries. In many of the democracies that have emerged over the past two decades, citizens are broadly dissatisfied with the performance of the political system and distrustful of many of its institutions (especially parties and politicians). Yet they do not see an alternative to democracy. Even in Brazil, where active support for democracy stood at only 37 percent last year, people do not prefer authoritarian rule (only 15 percent could imagine wanting it). The alternative, rather, is apathy and withdrawal. This is bad for democracy, but not as bad as people actively clamoring for an authoritarian alternative. In the 10 postcommunist candidate states for eu accession, 61 percent are dissatisfied with the way democracy works in their country, according to the Centre for the Study of Public Policy’s New Europe Barometer survey. Yet, overall, 72 percent would not approve of its suspension.

Belief in the legitimacy — the moral rightness — of a political system is always a relative judgment. In the past several decades, almost every form of nondemocratic government imaginable has been tried: absolute monarchy, personal dictatorship, military rule, colonial rule, fascism, communism, Ba’athism, the socialist one-party state, other forms of one-party rule, the Islamic Republic, pseudodemocracy, semidemocracy, and numerous other permutations. At an accelerating rate, people have opted for democracy. “As democracy has spread, its adherents have grown, not shrunk,” Amartya Sen observes. Whatever their naïve assumptions at the beginning, people are sticking with democracy without illusions. They remember in their lifetime one or more of these other forms of rule, and they do not want to go back.

Of course it is possible that some new form of nondemocratic rule will be conjured up and capture the passions and imagination of some peoples; but at this point, more than a decade after the collapse of communism, there is no sign on the horizon of an antidemocratic ideology that could even begin to generate universal claims. Most likely, where authoritarian rule reasserts itself in the coming years, it will do so apologetically, wrapping itself in the moral purpose of democratic restoration and insisting — as General Pervez Musharraf did when he seized power in Pakistan in 1999 — that the suspension of democracy would be temporary. Or elected rulers will gradually whittle down the quality and competitiveness of democratic institutions. Or violent insurgencies will grind down the scope of their actual authority to the point where it is just very difficult to determine whether the country meets the minimal test of democracy.

The third factor suppressing potential reversions to authoritarian rule has been the unfavorable climate for such reversals at the regional and international levels. Most of all in Europe, but in Latin America as well, political and military leaders know that they will pay a high price in terms of economic and political standing within their regions if they reverse democracy. On specific occasions, some such leaders who have been tempted to reverse democracy — in Guatemala, in Paraguay, perhaps in Venezuela, and probably in Turkey — have been deterred from doing so by explicit interventions from neighboring countries and from the United States

But the international environment is a discordant one. There are conflicting signals and incentives. If we can create a more coherent and vigorous international environment supporting democracy and democratization, we can more effectively bolster the existing democracies against reversion while inducing more transitions to democracy, both gradual and rapid. That is the overriding challenge of the moment.


Several factors explain the tenacious resistance to the democratic trend on the part of roughly 70 countries. The least common explanation is authoritarian success. This can account for Singapore and Malaysia, and perhaps to some extent China with its recent rapid economic growth. But China remains a lower-middle-income country, and the sars crisis is only the latest demonstration of the contradictions it confronts trying to sustain its phenomenal growth rates while the political system remains closed, corrupt, and unaccountable. Then there are the oil-rich states — the ones with staggering revenue and relatively small populations — which have been able to maintain authoritarian rule because they have had the wealth to buy off their peoples while lavishly financing structures of internal security and control. Even so, their peoples — in Kuwait, in Qatar, in Bahrain, and now in Saudi Arabia — are restive and want more self-determination. With the exceptions of Singapore, Malaysia, possibly for a while China, and these oil-rich states, there are no dictatorships in the world that survive today because they have brought prosperity to their people.

There are a few other holdout communist states in Asia (Vietnam, Laos, North Korea) and Cuba as well. Here, the insular, repressive logic of communist control persists. However, Vietnam is learning from China’s model of economic opening, and what is true for China is true for these states as well. The more they open to the outside world in terms of trade, investment, foreign study, foreign travel, and all the other aspects of globalization, the more their people become exposed to education and global culture, the more the insular, repressive logic will weaken. At some unpredictable point, a regime crisis, an economic downturn, a split within the elite could ignite a transition to democracy. There is of course an alternative strategy to bring about regime change: Isolate them from the world. Make it hard for the regime. Then wait for it to collapse. We have tried this strategy for 40 years in Cuba, and all it has done is impoverish the people and entrench their repressive rulers. Precisely in order to generate the social and economic changes that will finally undermine communist rule in Cuba, we should lift the embargo and promote as much exchange and interaction with that country as possible.

Most dictatorships in the world survive for a simple reason. Their leaders enjoy having unchallenged power as well as having the ability that power confers to accumulate great personal wealth. It is just not possible to look at the evidence from the ground (and from the public opinion surveys) in Africa, Asia, and the Middle East and argue that their peoples don’t mind living under dictatorship. Of course, no one could maintain that the majority of people in every country want a fully democratic system or that all peoples understand all the institutions of liberal democracy. But most people do want freedom. Given the choice, they would like to be able to constrain the arbitrary power of government, to replace bad and corrupt leaders, to have a predictable and secure life under some kind of just rule of law. When one assembles these basic political preferences, it begins to look an awful lot like democracy, even if the word may have different (or unsure) meanings in many places.

There is a lot of work to be done around the world to build the culture of democracy — the understanding of its rules, possibilities, obligations, and limits, the norms of tolerance, civility, participation, and mutual respect. Some of this cultural change happens with economic development, increasing education, and exposure to the global environment. Much of it can and should happen through deliberate programs of civic education and civil society construction. External democracy promotion programs and domestic civil society efforts have made some progress toward these goals. Much more remains to be done.

But the principal obstacle to the expansion of democracy in the world is not the people of the remaining authoritarian states. The problem is the ruling elites who have hijacked the structures of state power and barricaded themselves inside. As long as these rulers can corner a sufficient flow of resources to feed their apparatus of political predation and domination, they can survive.

That is where the international environment enters in. Predatory authoritarian regimes do not generate resources organically from within their own societies very well. Rather, they inhibit domestic investment, innovation, entrepreneurship, and hence economic growth by violating property rights and other individual freedoms. Such arbitrary rule also discourages foreign investment — except in the enclave economy of oil or other natural resource extraction. If predatory regimes do not have natural resources, they fall heavily in need of foreign loans and aid. This makes them vulnerable if the sources of those loans start insisting on responsible government.

For the most part, Arab states are rentier states, deriving their revenues mainly from oil or international aid flows rather than from taxes paid by their own people. But there has been another, entirely unique factor in their authoritarian survival. All of these dictatorships have been able to summon up a grand excuse for the failures and disappointments of their systems. First, it was the allegedly “colonial” existence of the state of Israel. For some it is still that, but now in particular it is the plight of the Palestinians and the Israeli-Palestinian conflict. Over the past several decades, this conflict has generated a heavy fog over Arab politics, diminishing political visibility and transparency. Arab governments have used it relentlessly to legitimate their rule — by stressing the authenticity of their commitment to something larger than themselves — and have relied on it more and more as the older forms of nationalism and pan-Arab solidarity have lost their luster. Much of the energy and emotion of Arab intellectuals and political activists has been drawn away from national political failings into protest over this larger political conflict. The debate about the true failings of Arab development — so eloquently expressed in last year’s Arab Human Development Report — has been distorted and deflected by this powerful symbolic struggle over Arab identity and dignity.

Until the fog of this struggle is lifted — so that the peoples of the Arab world can see and debate more clearly the real nature of the obstacles to national progress, and so that radical Islamists will be deprived of one of their most emotive instruments for mobilizing political support — genuine and lasting democratization will be unlikely in the region.

What is to be done?

Lenin had a revolutionary agenda for global dictatorship. Some American leaders, such as Woodrow Wilson and Ronald Reagan, have also had a revolutionary global vision, but a vision of democracy. If the whole world is ever to become democratic, the most powerful democracy cannot be passive or timid (yet neither can it transform the world alone). We must craft a global strategy, asking Lenin’s classic question: What is to be done?

First, with respect to the Arab world, we must relentlessly pursue a settlement of the Israeli-Palestinian conflict on the basis of the only broadly viable solution: the permanent coexistence and mutual recognition of two separate states, one an Israel that withdraws from most of its settlements in the West Bank and Gaza, the other an essentially demilitarized Palestine. Only the settlement of this conflict can strip Arab dictatorships of political cover for their abuses and free Arab societies to focus on the real sources of their misery and frustration.

Second, we need to open up the closed societies of the world. I do not propose to shower them with aid — far from it — but we should promote trade, travel, and exchanges of all kinds with countries like Cuba, Vietnam, Burma, and, yes, North Korea. The North Korean dictatorship is a house of cards resting on a tissue of lies. Its people, the most physically and intellectually isolated and totally brutalized of any in the world today, do not have a clue as to how the rest of the world lives. Once they find out, the regime will crumble, or else change very rapidly.

Third, we need a new deal in foreign aid and debt relief. Even after the end of the Cold War, even with the new standards and pressures on dictatorships, the resources to sustain them have largely continued to flow. Part of this has simply been inertia. Part of this has been the utterly perverse structural logic of aid agencies and especially the World Bank, whose officials are given portfolios of money to lend and projects to initiate with the understanding that their careers will suffer if they do not push the money out the door. Part of the reason has been fear that if we lean too heavily on weak, oppressive, rotten states, they will collapse altogether into new humanitarian emergencies. Instead, we dawdle and fund them while they disintegrate more slowly and millions of their people live shorter, nastier, more brutish lives because of abusive governance. Finally, part of the problem has been the conflicting priorities of bilateral donors (including the United States) that still want to maintain friendly client states around the world. Some thought this dualism — a polite word for hypocrisy — would come to an end with the demise of the Cold War. And indeed, it did subside for a time. But with the inception of the new war on terrorism since September 11, the problem of selling short our principles in order to nurture authoritarian clients has been reborn with a vengeance.

A new deal on aid would radically accelerate and institutionalize the tentative trends toward encouraging and expecting good governance in exchange for foreign aid. The Bush administration took an important step forward last year when it announced the creation of a Millennium Challenge Account, which will award a new $5 billion increment in development assistance (about a 50 percent increase over the current U.S. foreign aid budget) to a select number of low- and lower-middle-income countries that compete for it on the basis of three criteria: ruling justly (including democratically and accountably), investing in people, and promoting economic freedom. Hopefully, countries that qualify will start receiving substantial new sums of aid in the next year — if Congress allocates the funds.

This is an important departure — indeed, a conceptual revolution — in foreign aid. But it does not go nearly far enough. First, we still need to question what we are doing with the rest of our foreign aid budget. Much of it goes to countries ruled by corrupt, authoritarian regimes. If that aid is delivered to and through civil society rather than the corrupt state itself, it may do some good. But too much is wasted, and there is too little effort to generate leverage for real political change. The big problem is the other donors: the World Bank, the regional development banks, the Japanese, and many of the European aid agencies.

Democratic change is possible in the world’s remaining corrupt dictatorships, but it will require a radical manipulation of the incentives their leaders confront. They must know that the party is over, that they cannot any longer play one powerful donor off against another, or one country promoting its own oil industry over another.

This past January, the U.S. Agency for International Development released a report, Foreign Aid in the National Interest (available at www.usaid.gov), recommending a new set of strategies for our development assistance based on rewarding demonstrated performance and getting aid dollars into the hands of people in a given country who can do the most good with the money. These new directions, if adopted, could transform the international context in which dictatorships now maneuver to survive. In addition, a greater proportion of total U.S. foreign assistance should be devoted to political assistance to build democracy and improve the quality of governance. In truly intractable cases, helping to generate the demand for democracy and better governance by strengthening the capacity and reform understanding of independent organizations, interest groups, social movements, mass media, universities, and think tanks in civil society may be the main thing the United States can do to aid development. Whatever progress is made on governance will almost certainly have a positive impact on other sectors. Probably no other dimension of foreign assistance yields so many synergies.

The new strategy moves from the current exhausted approach of conditionality to a selectivity that rewards political freedom and accountability. As much as possible, rewards should be structured to lock into place the institutions and practices of democracy and good governance. As we seek to expand nafta into a Free Trade Agreement of the Americas, we should adopt a requirement similar to the European Union, that all members uphold democracy and human rights. In the case of debt relief for highly indebted poor countries, future relief should be granted only to countries that have demonstrated a basic commitment to good governance by allowing a free press and civil society, an independent judiciary, and a serious counter-corruption commission. Even in these cases, the debt should not be relieved in one fell swoop, but should be suspended and retired incrementally (for example, at 10 percent per year), generating ongoing incentives for adhering to good governance.

If the United States and the other major bilateral and multilateral donors were to move together toward such a comprehensive strategy affirming democracy and good governance as the basis of development (and hence development assistance), they would generate very powerful new pressures for democratic reform. Not all authoritarian states would be affected immediately because not all of them depend on these flows of assistance. But the overall global climate would shift emphatically in favor of democratic change, generating potent demonstration effects even on the stand-pat regimes.

Of course, this would still leave open the question of how democratic change could be accomplished in countries that have never been democracies before. There is no one formula for getting to democracy or for structuring it institutionally so that it works reasonably well. Different countries need different sequences, strategies, and structures. In some cases, the transition to democracy could and should proceed fairly rapidly, since governance is such a mess and viable democratic forces wait in the wings. In other cases — including many of the Arab states — the transition to democracy will need to proceed more cautiously and incrementally.

Another distinctive feature of the Arab world is that the formal political arena has been closed to all but a relatively narrow circle of establishment parties and interest groups, while the chief alternatives (both above ground and below) to these co-opted forces have been Islamist parties and groups mobilizing a considerable network of affection and support, but with an anti-democratic agenda. The more moderate and pro-democratic groups — both Islamist and secular — have been squeezed between the iron fist of the state and the superior popular organization of illiberal Islam. If they are to be given a fair chance to compete in electoral politics, these moderates will need time to surface, organize, advocate, and campaign. While electoral competition for genuine national power is phased in over a period of years, an interim period of political liberalization must be used to build the independent structures of horizontal accountability — the judiciary, audit agency, counter-corruption commission, human rights commission, ombudsman, and electoral administration — that can ensure free and fair electoral competition and constrain whoever wins election in the new system.

The challenge in postwar Iraq will be unique. There, the state and political system must be thoroughly reconstructed at the same time as we rebuild an economy devastated by three major wars and decades of colossal misrule. The one thing we absolutely must not do is impose a leader or solution of our choosing. Iraqis — both those within the country and those returning from exile — must be given the time and space to meet in local groups, select representatives, and assemble a broad-based transitional government. Then they must begin to take responsibility for administering and rebuilding the country while organizing a smaller and more professional new army, debating and drafting a new constitution, then submitting it to a referendum, forming political parties and civic associations, constructing new independent media, electing new municipal and provincial governments, operating these new governments, and — only some years later — holding national elections for a new constitutional government. During this period of political reconstruction and nation-building, which will require a huge international peacekeeping force if it is to have any chance of success, some international authority will need to provide initial supervision of the process, gradually withdrawing as the Iraqi authority is able to assume greater and greater responsibility. The sooner that authority is internationalized in some way (in military terms, for example, through nato), the longer it can stay in Iraq with some internal and international legitimacy.

Can it be done?

Even to think of democratizing the entire world is a bold endeavor. Perhaps it is too bold. Certainly the above prescriptions are not without flaws. In the near term, we will probably fall short of the courage, imagination, and nerve truly to transform the global political climate. Part of this challenge is deeply political, even in terms of our own national debate. It is important to realize how the current war has further altered the international climate. Increasingly, the United States is seen as an imperial power, imposing its will largely unilaterally on the rest of the world. That may serve our short-term interests in any one conflict or dispute, but it will not facilitate our deeper, long-term interest in building a world of democracies and good governance. We cannot invade and conquer every dictatorship in the world. In fact, our national appetite for forcible regime change will probably be quickly exhausted in postwar Iraq. Unless we learn to work with and through international partners and institutions while seeking to energize, transform, and democratize global structures, our scope to effect further democratic change in the world will shrink.

Whatever may happen in Iraq, we will face sobering challenges of international terrorism, rotten governance, and big-power division. Our moral and political prestige in the world has suffered in the past few months, even as respect for our sheer military power has increased. If the whole world is to keep moving toward democracy, we must recover the former while preserving the latter. Ironically, we will find that we have much greater leverage to advance the cause of freedom in the world if we build and maintain effective partnerships with the other democracies of the world. We cannot always lead from the battleship or the bully pulpit. Sometimes, we must do so more softly and subtly, as part of a team.

We should also not take the existing democracies for granted. There is a new sobriety even among the democracies that are not performing well. People know the alternatives and do not like them. They still embrace democracy. But will they continue to do so a decade or two hence if a new generation — with no direct experience of the costs and illusions of authoritarian rule — finds itself without education, without jobs, without justice, and pretty much without hope? We can travel only so far on democracy as the least bad system. If it turns out to govern badly for a long period of time, some new alternative will eventually come along.

The fully global triumph of democracy is far from inevitable, yet it has never been more attainable. If we manage to sustain the process of global economic integration and growth while making freedom at least an important priority in our diplomacy, aid, and other international engagements, democracy will continue to expand in the world. History has proven that it is the best form of government. Gradually, more countries will become democratic while fewer revert to dictatorship. If we retain our power, reshape our strategy, and sustain our commitment, eventually — not in the next decade, but certainly by mid-century — every country in the world can be democratic.

1 An earlier version of this essay was presented as the Harry Eckstein Lecture at the University of California, Irvine, April 10, 2003. The full original version of this paper, with statistical tables, may be viewed at http://repositories.cdlib.org/csd/03-05/. I am grateful for the research assistance of Terrence Blackburne and Benn Eifert with the data in this paper.

2 Larry Diamond, “Elections Without Democracy: Thinking About Hybrid Regimes,” Journal of Democracy (April 2002).

3 See Larry Diamond, Developing Democracy: Toward Consolidation (Johns Hopkins University Press, 1999): 10–12.

4 See the essay by Alfred Stepan forthcoming in the Journal of Democracy (July 2003).

5 The democracies in this set are Bangladesh, Brazil, India, Indonesia, Japan, Mexico, Nigeria, and the United States. The nondemocracies are China, Pakistan, and Russia.

6 Adam Przeworski, Michael E. Alvarez, Jose Antonio Cheibub, and Fernando Limongi, Democracy and Development: Political Institutions and Well Being in the World, 1950–1990 (Cambridge University Press, 2000): 92–103.

7 Amartya Sen, “Democracy as a Universal Value,” in Larry Diamond and Marc F. Plattner, eds., The Global Divergence of Democracies (Johns Hopkins University Press, 2001).

8 “Islam, Democracy, and Public Opinion in Africa,” Afrobarometer Briefing Paper No. 3 (September 2002); Richard Rose, “How Muslims View Democracy: Evidence from Central Asia,” Journal of Democracy (October 2002); Mark Tessler, “Islam and Democracy in the Middle East: The Impact of Religious Orientations on Attitudes Toward Democracy in Four Arab Countries,” Comparative Politics (April 2002).

9 See Thomas Franck, “The Emerging Right to Democratic Governance,” American Journal of International Law (January 1992).

overlay image