Elections Without Democracy
Thinking About Hybrid Regimes
Is Russia a democracy? What about Ukraine, Nigeria, Indonesia, Turkey, or Venezuela? There was a time when these were simple questions of regime classification. But the empirical reality in these countries is a lot messier than it was two decades ago, and so, in a way, is the never-ending dialogue on how to think about and classify regimes.
Few conceptual issues in political science have been subjected to closer or more prolific scrutiny in recent decades than this problem of "what democracy is . . . and is not," 1 and which regimes are "democracies" and which not. We are replete with definitions and standards and tools of measurement. But the curious fact is that—a quarter-century into the "third wave" of democratization and the renaissance it brought in comparative democratic studies—we are still far from consensus on what constitutes "democracy." And we still struggle to classify ambiguous regimes.
Some insist on a fairly robust (though still procedural) definition of democracy, like Robert Dahl's "polarchy." By this conception, democracy requires not only free, fair, and competitive elections, but also the freedoms that make them truly meaningful (such as freedom of organization and freedom of expression), alternative sources of information, and institutions to ensure that government policies depend on the votes and preferences of citizens. Some measure democracy by a "minimalist" standard like Joseph Schumpeter's: a political system in which the principal positions of power are filled "through a competitive struggle for the people's vote." 2 Yet contemporary applications of this electoral conception heavily overlap with Dahl's polyarchy by also implying the civil [End Page 21] and political freedoms necessary for political debate and electoral campaigning.
Even if we agree to apply a minimalist, electoral standard for democracy, vexing questions remain. If, following Samuel Huntington, a system is democratic when "its most powerful collective decision makers are selected through fair, honest, and periodic elections in which candidates freely compete for votes," 3 what constitutes "fair, honest, and free" elections? How can we know that parties have had a fair chance to campaign and that voters around the country (especially countries as large and diverse as Russia, Nigeria, and Indonesia) have been able to exercise their will freely? How—especially where elections do not benefit from parallel vote tabulations 4 —can we know that the reported results accurately reflect the votes that were cast? And how do we know that the officials elected are really the "most powerful decision makers," that there are not significant "reserved domains" of military, bureaucratic, or oligarchical power? 5
These questions have taken on a heightened relevance in recent years for several reasons. First, more regimes than ever before are adopting the form of electoral democracy, with regular, competitive, multiparty elections. Second, many of these regimes—indeed, an unprecedented proportion of the world's countries—have the form of electoral democracy but fail to meet the substantive test, or do so only ambiguously. And third, with heightened international expectations and standards for electoral democracy, including the rise of international election observing, there is closer international scrutiny of individual countries' democratic practices than ever before.
Yet even with this closer scrutiny, independent observers do not agree on how to classify regimes. Freedom House classifies all six regimes mentioned at the beginning of this essay as democracies. Yet by the logic of the three articles that follow, they are all (or mostly) something less than electoral democracies: competitive authoritarian systems, hegemonic-party systems, or hybrid regimes of some kind. At best, Ukraine, Nigeria, and Venezuela are ambiguous cases. We may not have enough information now to know whether electoral administration will be sufficiently autonomous and professional, and whether contending parties and candidates will be sufficiently free to campaign, so as to give the political opposition a fair chance to defeat the government in the next elections. Regime classification must, in part, assess the previous election, but it must also assess the intentions and capacities of ambiguously democratic ruling elites, something that is very hard to do. Increasingly, independent observers view Russia as an electoral authoritarian regime. Many so view Nigeria as well, given the massive (and quite characteristic) fraud in the 1999 elections. Indonesia's constitutional assignment of some parliamentary seats to unelected military representatives contradicts a basic principle of democracy. But even if that provision were removed, [End Page 22] the military would remain a major veto player (like the Turkish military, which has repeatedly forced the disqualification of a popular, moderately Islamist party).
These are hardly the only issues or anomalies in regime classification. In the 1970s and 1980s, scholars and observers debated whether Mexico, Senegal, and Singapore were really democracies (as their governments insisted). These debates fizzled once other countries in their respective regions began to experience true democratization and the democratic deficiencies of these one-party hegemonies became more blatantly apparent. More recently, a growing number of scholars are questioning the tendency to classify regimes as democratic simply because they have multiparty elections with some degree of competition and uncertainty. In an important conceptual contribution, focused on Eurasia and Latin America, Steven Levitsky and Lucan Way argue in the pages that follow that regimes may be both competitive and authoritarian.
This set of articles exemplifies a new wave of scholarly attention to the varieties of nondemocratic regimes and to the rather astonishing frequency with which contemporary authoritarian regimes manifest, at least superficially, a number of democratic features. This new intellectual upsurge partly reflects the exhaustion of the "third wave" of democratic transitions, which essentially crested in the mid-1990s. 6 For some years now, it has been apparent that a great many of the new regimes are not themselves democratic, or any longer "in transition" to democracy. Some of the countries that fall into the "political gray zone . . . between full-fledged democracy and outright dictatorship" are in fact electoral democracies, however "feckless" and poorly functioning, but many fall below the threshold of electoral democracy and are likely to remain there for a very long time. 7
A Historical Perspective
Hybrid regimes (combining democratic and authoritarian elements) are not new. Even in the 1960s and 1970s, there existed multiparty, electoral, but undemocratic regimes. Of these electoral autocracies—Mexico, Singapore, Malaysia, Senegal, South Africa, Rhodesia, and Taiwan (which allowed dangwai, or "outside the party," competitors)—only the Malaysian and Singaporean regimes survive today. Historically, there have also been numerous cases in Europe and Latin America of limited (elite) party competition with a limited franchise. In Latin America, these nineteenth-century and early-twentieth-century "oligarchical" democracies "contributed to the ultimate development of full democracy" by establishing some of its major political institutions, as well as the principles of limitation and rotation of power. 8 Thus these countries epitomized Dahl's optimal path to stable polyarchy, with the rise of political competition preceding the expansion of participation, so that the [End Page 23] culture of democracy first took root among a small elite and then diffused to the larger population as it was gradually incorporated into electoral politics. 9 In the contemporary world of mass participation, this gradualist path has been closed off, and anxious elites have thus sought out other ways to limit and control competition.
Until the past decade or two, most efforts at political control included a ban on opposition political parties (if not on electoral competition altogether) and severe limits on the organization of dissent and opposition in civil society as well. Thus Juan Linz's encyclopedic Totalitarian and Authoritarian Regimes (originally published in 1975) contains barely a mention of multiparty electoral competition within authoritarian regimes. Party politics figures within the framework of a single (typically mobilizational) party, and only brief mention is made of "pseudo-multiparty systems." 10 Certainly Linz does not identify, among his seven principal authoritarian regime types, anything like the "competitive authoritarian" regime type discussed by Levitsky and Way—and for good reason. This type of hybrid regime, which is now so common, is very much a product of the contemporary world.
One term for this phenomenon (favored more or less in the following three essays) is "electoral authoritarianism." However, the term "pseudodemocracy" resonates distinctively with the contemporary era, in which democracy is the only broadly legitimate regime form, and regimes have felt unprecedented pressure (international and domestic) to adopt—or at least to mimic—the democratic form. Virtually all hybrid regimes in the world today are quite deliberately pseudodemocratic, "in that the existence of formally democratic political institutions, such as multiparty electoral competition, masks (often, in part, to legitimate) the reality of authoritarian domination." 11 All such regimes lack an arena of contestation sufficiently open, free, and fair so that the ruling party can readily be turned out of power if it is no longer preferred by a plurality of the electorate. While an opposition victory is not impossible in a hybrid regime, it requires a level of opposition mobilization, unity, skill, and heroism far beyond what would normally be required for victory in a democracy. Often, too, it requires international observation and intervention to preempt and prevent (as in Nicaragua in 1990) or to expose and delegitimate (as in the Philippines in 1986) the electoral manipulations and fraud of the authoritarian regime.
If scholarly treatment of hybrid or "electoral authoritarian" regimes is relatively new, it is not without some intellectual foundations in the transitions paradigm and in other earlier comparative work on democracy. Guillermo O'Donnell and Philippe Schmitter emphasized the inherent uncertainty and variation in the outcomes of regime transitions. A transition from authoritarian rule could produce a democracy, or it could terminate with a liberalized authoritarian regime (dictablanda) or a restrictive, illiberal democracy (democradura). 12 During the early wave [End Page 24] of enthusiasm over the spread of democracy in Latin America, Terry Karl criticized the tendency to equate democracy with competitive multiparty elections. She argued that military domination and human rights abuses rendered the Central American regimes of the 1980s and early 1990s "hybrid regimes," not democracies. 13 Robert Dahl classified (circa 1969) as "near polyarchies" six competitive electoral regimes. 14 Juan Linz, Seymour Martin Lipset, and I labeled "semidemocratic" those regimes "where the effective power of elected officials is so limited, or political party competition so restricted, or the freedom and fairness of elections so compromised that electoral outcomes, while competitive, still deviate significantly from popular preferences; and/or where civil and political liberties are so limited that some political orientations and interests are unable to organize and express themselves." 15 Among our 26 cases, Senegal, Zimbabwe, Malaysia, and Thailand (during 1980-88, when the government was led by an unelected prime minister) fell into the category that Levitsky and Way call "competitive authoritarian." Mexico fit the model of a hegemonic party system, in which a relatively institutionalized ruling party monopolizes the political arena, using coercion, patronage, media control, and other means to deny formally legal opposition parties any real chance of competing for power. 16 Singapore remains a classic example of such a system.
The Rise of Pseudodemocracy
One of the most striking features of the "late period" of the third wave has been the unprecedented growth in the number of regimes that are neither clearly democratic nor conventionally authoritarian. If we use a very demanding standard of democracy, encompassing not only democratic elections but solid protection of civil liberties under a strong rule of law, then the proportion of intermediate regimes truly swells because so many of the new "democracies" of the third wave are "illiberal." 17 However, I believe a more analytically fruitful approach is to measure separately both electoral democracy, in the minimalist terms that Schumpeter, Huntington, and others have used, and liberal democracy. 18 We can also divide nondemocratic regimes into those with multiparty electoral competition of some kind (variously termed "electoral authoritarian," "pseudodemocratic," or "hybrid") and those that are politically closed. We can further divide electoral authoritarian regimes into the competitive authoritarian (following Levitsky and Way's formulation) and the uncompetitive or (following Schedler, and before him Giovanni Sartori) hegemonic. Tables 1 and 2 on pages 26 and 30-31, respectively, sort the world's regimes into these five categories, plus the residual one of ambiguous regimes.
During the third wave, both the number and proportion of democracies in the world have more than doubled. We find 104 democracies in [End Page 25] the world at the end of 2001, seventeen fewer than Freedom House counts 19 but well over twice the 39 democracies at the start of the third wave; this accounts for 54 percent of the world's regimes, twice the proportion in 1974 (see Table 1 above). About seven in ten democracies may then be considered liberal (in that they have a fairly liberal Freedom House score of 2.0 or lower on the seven-point scale averaging political rights and civil liberties). Another 31 democracies are electoral but not liberal; some are clearly illiberal, with no more than a middling score on civil liberties. I consider 17 regimes "ambiguous" in the sense that they fall on the blurry boundary between electoral democracy and competitive authoritarianism, with independent observers disagreeing over how to classify them. 20 Virtually all 17 could be classified as "competitive authoritarian." Doing so would raise the number of such regimes from 21 to as many as 38, and the proportion from 11 to 20 percent—quite a significant phenomenon. Another 25 regimes are electoral authoritarian but in a more hegemonic way. They do not exhibit the degrees and forms of competitiveness elucidated by Levitsky and Way and illustrated as well by some of the African cases discussed by Nicolas van de Walle in his essay. Their elections and other "democratic" institutions are largely façades, yet they may provide some space for political opposition, independent media, and social organizations that do not seriously criticize or challenge the regime. Finally, 25 regimes do not have any of the architecture of political competition and pluralism. These remain politically closed regimes.
The data in Table 1 and the underlying scheme of classification raise a number of interesting points and issues. The most stunning is the dwindling proportion of politically closed regimes in the world. This transformation is partly reflected in the steady overall rise of freedom in the world (with the average score on the combined seven-point Freedom House scale improving from 4.47 in 1974 to 3.47 in 2001). And it is partly seen in the shrinking number and proportion of states with the two most repressive average freedom scores of 6.5 and 7.0. These most repressive regimes declined from 29 in 1974 to 21 in 2001, and as a proportion of all states, they shrank from one-fifth to barely a tenth (11 percent). [End Page 26]
Thus the trend toward democracy has been accompanied by an even more dramatic trend toward pseudodemocracy. Only about half a dozen regimes in 1974 (less than 5 percent) would have met Schedler's criteria of electoral authoritarianism: undemocratic but with multiparty elections and some degree of political pluralism. The rest were all military, one-party, or personalist regimes. Today, at least 45 and perhaps as many as 60 are electoral authoritarian—roughly between a quarter and a third of all states. In proportional terms, authoritarian forms of multiparty electoral competition have increased during the third wave much more rapidly than democratic ones.
At the same time, military regimes have virtually disappeared as anything more than a transitional type of rule. Today, ambitious soldiers either legitimize their rule by running for president in contested, multiparty elections (however fraudulent, coerced, and manipulated), or they carve out large, autonomous spheres of political influence and economic domination behind the veil of civilian, multiparty rule. The first path has been taken by a number of African military strongmen, such as Jerry Rawlings in Ghana and most recently Yahya Jammeh in the Gambia. Nigerian dictator Sani Abacha was engaged in such a maneuver when he was struck dead by a "heart attack" in 1998. General—now President—Pervez Musharraf may yet pursue a similar conversion in Pakistan, albeit perhaps with considerably more genuine popular support. The second course has been taken by the military in Indonesia, and to a lesser degree still characterizes the military in Turkey, Thailand, Nigeria, and parts of Latin America.
There is also a striking correlation between country size and regime type. As I noted a few years ago, 21 countries with populations under one million are much more likely to be both democracies and liberal democracies. Two-thirds of these countries are liberal democracies, while only 30 percent of countries with populations over one million are. Among the larger 150 countries, only half are democracies, while 70 percent of the small countries are. The countries with populations over one million are about twice as likely as small states to have an electoral authoritarian regime and half again as likely to have a closed authoritarian regime.
Electoral Democracy vs. Electoral Authoritarianism
Interesting issues revolve around the boundaries between regime types, which all the authors in this issue recognize to be blurry and controversial. When fitting messy and elusive realities against ideal types, it cannot be otherwise. This is why I classify so many regimes as ambiguous—a judgment, however, that only addresses the border between democracy and electoral authoritarianism. The distinctions between liberal and electoral democracy, and between competitive and hegemonic electoral authoritarianism, can also require difficult and disputable judgements. [End Page 27] Thus the country classifications in Table 2 are offered more in an illustrative than a definitive spirit.
As Schedler elaborates, the distinction between electoral democracy and electoral authoritarianism turns crucially on the freedom, fairness, inclusiveness, and meaningfulness of elections. Often particularly difficult are judgments about whether elections have been free and fair, both in the ability of opposition parties and candidates to campaign and in the casting and counting of the votes. Hence the frequency with which the validations by international observer missions of elections in ambiguous or electoral authoritarian regimes are, often convincingly, criticized as superficial, premature, and politically driven.
Elections are "free" when the legal barriers to entry into the political arena are low, when there is substantial freedom for candidates and supporters of different political parties to campaign and solicit votes, and when voters experience little or no coercion in exercising their electoral choices. Freedom to campaign requires some considerable freedom of speech, movement, assembly, and association in political life, if not fully in civil society. It is hard, however, to separate these two spheres, or to weigh the significance of particular violations. How many opposition candidates and supporters must be killed or arrested before one discerns a blatantly undemocratic pattern? Typically more than one murder is necessary, but fewer than the 21 deadly assaults committed during the two months prior to Cambodia's 1998 elections. 22 In India, election-related killings have a long history and have recently risen to alarming levels in some states. No major observer denies that India is a democracy, but particularly in states (like Bihar) where corruption, criminality, murder, and kidnapping heavily taint the electoral process, it is an illiberal and degraded one. A crucial consideration in assessing a regime is whether political violence is clearly and extensively organized by the state or ruling party as a means of punishing, terrorizing, and demoralizing opposition.
Assessments about whether elections are free or not thus require careful and nuanced judgments about the scale, pattern, and context of violations. The same is true for the dimension of electoral fairness. Levitsky and Way argue that political systems descend into electoral authoritarianism when violations of the "minimum criteria for democracy" are so serious that they create "an uneven playing field between government and opposition." Yet even in many liberal and established democracies, there is not a truly level playing field. Often, governing parties or executives enjoy advantages of incumbency—readier access to the media, an easier time raising money from business, and the ability (strictly legal or not) to use government transport and staff while campaigning. No system is a perfect democracy, all require constant vigilance, and scattered violations do not negate the overall democratic character of elections.
When evaluating elections, it is crucial to examine their systemic [End Page 28] character. We have by now elaborate criteria to judge the fairness of elections. Elections are fair when they are administered by a neutral authority; when the electoral administration is sufficiently competent and resourceful to take specific precautions against fraud in the voting and vote counting; when the police, military, and courts treat competing candidates and parties impartially throughout the process; when contenders all have access to the public media; when electoral districts and rules do not systematically disadvantage the opposition; when independent monitoring of the voting and vote-counting is allowed at all locations; when the secrecy of the ballot is protected; when virtually all adults can vote; when the procedures for organizing and counting the vote are transparent and known to all; and when there are clear and impartial procedures for resolving complaints and disputes. 23 This is a long list, but serious efforts to compromise the freedom and fairness of elections form a pattern (beginning well before election day) that is visible across institutional arenas. The institutional biases and misdeeds are there for international observers to see if those observers have the time, experience, courage, and country expertise to do so. 24
Degrees of Authoritarian Competitiveness
No less difficult is the challenge of distinguishing between competitive authoritarian regimes and hegemonic electoral authoritarian ones. Levitsky and Way posit four arenas in which "opposition forces may periodically challenge, weaken, and occasionally even defeat autocratic incumbents." While contestation in the judiciary and the mass media is hard to quantify, contestation in elections and legislatures does allow for more structured comparison.
Table 2 on the following pages classifies the world's regimes by the sixfold typology explained above. Regimes are considered democratic if they have free, fair, and open elections for all the principal positions of political power, as defined above and by Schedler in his contribution. In addition to the Freedom House scores, three types of data are drawn upon in my classification of nondemocratic regimes: the percentage of legislative seats held by the ruling party, the percentage of the vote won by the ruling party presidential candidate, and the years the incumbent ruler has continuously been in power. The latter, as van de Walle shows in his essay on Africa, can be a telling indicator of the degree to which a country has opened up, as well as a predictor of its future openness to democratic change. Although I do not use any mathematical formula to combine these three indicators and the Freedom House scores, a formal index of authoritarian competitiveness is worth developing.
One defining feature of competitive authoritarian regimes is significant parliamentary opposition. In regimes where elections are largely an authoritarian façade, the ruling or dominant party wins almost all the seats: [End Page 29] [Begin Page 32] repeatedly over 95 percent in Singapore, about 80 percent in Egypt in 2000 and Mauritania in 2001, 89 percent in Tanzania in 2000, and repeatedly over 80 percent in Tunisia during the 1990s. 25 In Cambodia the hegemonic character of rule by Hun Sen's Cambodian People's Party (CPP) was not apparent in the bare majority of parliamentary seats it won in 1998, but it became more blatant in early 2002 when the CPP won control of about 99 percent of the 1,621 local communes with about 70 percent of the vote.
Where, as in Kazakhstan and Kyrgyzstan, parties are so poorly developed that it is difficult to interpret legislative election results, presidential election returns offer other evidence of hegemony. After winning a presidential referendum with a 95 percent "yes" vote in 1995, Kazakhstan's President Nursultan Nazarbayev was reelected with 80 percent of the vote in 1999. In 1995 and again in 2000, Kyrgyz president Askar Akayev, in whom the West placed early (and nave) hopes for democratic progress, was reelected with 75 percent of the vote. One clear sign of hegemony is when the president "wins" three-quarters or more of the popular vote. This also happened in Algeria in 1999, in Azerbaijan in 1998, in Burkina Faso in 1998, in Cameroon (with an opposition boycott) in 1997, in Djibouti in 1999, and in Tanzania in 2000.
At the extreme end of the continuum, the presidents of Egypt, Tunisia, and Yemen were all "reelected" in the 1990s with well over 90 percent of the vote. These men have been in power now for 21, 15, and 12 years, respectively, pointing to another sign of authoritarian hegemony: prolonged presidential tenure. Other examples include 23 years in Angola, 20 years in Cameroon, 35 years in Gabon, 18 years in Guinea, and 16 years in Uganda. Yet some long-ruling autocrats have had to fight for their political lives in the 1990s. Daniel arap Moi (who has finally pledged to step down this year after 24 years in power) was reelected twice during the 1990s with less than 42 percent of the vote. Zimbabwe's President Robert Mugabe, in power for 22 years, was resorting to massive violence and intimidation in his unpopular presidential reelection bid as this article went to press. His ruling party won only a bare majority of seats in a rough 2000 election that marked a breakthrough from numbing hegemony to competitive authoritarianism.
These data become more revealing when weighed with the annual Freedom House ratings of political rights and civil liberties. Generally, electoral authoritarian regimes range from 4.0 to 6.0 on the combined seven-point scale. Regimes closer to the less repressive score (4.0) allow more political pluralism and civic space, and hence are more likely to be competitive authoritarian. Some examples include Peru under Fujimori (4.5 in 1995), Senegal under the hegemonic Socialist Party (which averaged 4.0 or 4.5 during the 1990s), and Côte d'Ivoire (4.5 today, with competitive presidential and legislative elections in 2000). Many observers consider Tanzania a democracy, with its relatively benign regime (4.0), despite persistent electoral irregularities. Yet if one [End Page 32] traces its pedigree back to President Julius Nyerere's original TANU party, the Chama Cha Mapizindi (CCM) is the only ruling party Tanzanians have known in nearly 40 years of independence.
The reason we must examine several variables is that levels of freedom and levels of electoral competitiveness do not always neatly align. Indeed, when longtime authoritarian rulers face serious challenges (as in Malaysia and Zimbabwe recently), they may turn to their nastiest levels of repression, deploying levels of violence and intimidation that are unnecessary when political domination can be more subtly secured at the ballot box. Tracking the interplay between changes in political competition and changes in political repression may thus help us understand when and how moments of possible transition open and close in electoral authoritarian regimes.
Black and White or Shades of Gray?
Comparative politics is returning with new concepts and data to a very old issue: the forms and dynamics of authoritarian rule. If nothing else, the three articles that follow show that these divergent forms do matter. As democracies differ among themselves in significant ways and degrees, so do contemporary authoritarian regimes, and if we are to understand the contemporary dynamics, causes, limits, and possibilities of regime change (including possible future democratization), we must understand the different, and in some respects new, types of authoritarian rule.
At the same time, we must appreciate that classificatory schemes like the ones in these articles impose an uneasy order on an untidy empirical world. We should not ignore the critics of "whole system" thinking, who eschew efforts at regime classification altogether and seek to identify the ways in which each political system combines democratic and undemocratic features. 26 These approaches remind us that most regimes are "mixed" to one degree or another. 27 Even many politically closed regimes have quasi-constitutional mechanisms to limit power and consult broader opinion. For example, although China lacks competitive elections at any significant level, it has taken some steps to rotate power and to check certain abuses of corrupt local and provincial officials. Every step toward political liberalization matters, both for the prospect of a transition to democracy and for the quality of political life as it is daily experienced by abused and aggrieved citizens. As Levitsky and Way imply, significant steps toward a more open, competitive, pluralistic, and restrained authoritarian system can emerge in arenas other than electoral ones.
Democratic regimes are also "mixed" forms of government, not only in the
ways they empower institutions intentionally placed beyond the reach of
elected officials (such as constitutional courts or central banks), but in
less desirable respects as well. In their constant struggles to restrain
corruption, and in their ongoing frustration in trying to contain the role
[End Page 33]
of money in politics, even the world's most liberal democracies exhibit
the pervasive imperfections of responsiveness that led Robert Dahl
to adopt the term "polyarchy" instead of "democracy" for his seminal
study. As we add the forms and dynamics of electoral authoritarianism
to our long list of issues in comparative democratic studies, we should
not neglect these imperfections in our own systems. The transformations
of Taiwan, Mexico, and Senegal in the 1990s show that competitive
authoritarian regimes can become democracies. But democracies, new and
old, liberal and illiberal, can also become more democratic.
Larry Diamond is coeditor of the Journal of Democracy, codirector of the National Endowment for Democracy's International Forum for Democratic Studies, and a senior fellow at the Hoover Institution. His most recent books (published in 2001) are Political Parties and Democracy (edited with Richard Gunther) and The Global Divergence of Democracies (edited with Marc F. Plattner).
I am grateful to Terrence Blackburne for his research assistance and to Andreas Schedler, Steven Levitsky, and Nicolas van de Walle for their constructive comments.
1. Philippe C. Schmitter and Terry Lynn Karl, "What Democracy Is . . . and Is Not," Journal of Democracy 2 (Summer 1991): 75-88.
2. Joseph Schumpeter, Capitalism, Socialism, and Democracy, 2nd ed. (New York: Harper, 1947), 269.
3. Samuel P. Huntington, The Third Wave: Democratization in the Late Twentieth Century (Norman: University of Oklahoma Press, 1991), 7.
4. Larry Garber and Glenn Cowan, "The Virtues of Parallel Vote Tabulations," Journal of Democracy 4 (April 1993): 95-107.
5. J. Samuel Valenzuela, "Democratic Consolidation in Post-Transitional Settings: Notion, Process, and Facilitating Conditions," in Scott Mainwaring, Guillermo O'Donnell, and J. Samuel Valenzuela, eds., Issues in Democratic Consolidation: The New South American Democracies in Comparative Perspective (Notre Dame: University of Notre Dame Press, 1992), 64-66.
6. Larry Diamond, Developing Democracy: Toward Consolidation (Baltimore: Johns Hopkins University Press, 1999), ch. 2.
7. Thomas Carothers, "The End of the Transition Paradigm," Journal of Democracy 13 (January 2002): 5-21, quoted from pp. 9 and 18.
8. Larry Diamond and Juan J. Linz, "Introduction: Politics, Society, and Democracy in Latin America," in Larry Diamond, Juan J. Linz, and Seymour Martin Lipset, eds., Democracy in Developing Countries: Latin America (Boulder, Colo.: Lynne Rienner, 1989), 8.
9. Robert Dahl, Polyarchy: Participation and Opposition (New Haven: Yale University Press, 1971), 33-36.
10. Juan J. Linz, Totalitarian and Authoritarian Regimes (Boulder, Colo.: Lynne Rienner, 2000), 60.
11. Larry Diamond, Juan J. Linz, and Seymour Martin Lipset, Democracy in Developing Countries, xviii.
12. Guillermo O'Donnell and Philippe C. Schmitter, Transitions from Authoritarian Rule: Tentative Conclusions about Uncertain Democracies (Baltimore: Johns Hopkins University Press, 1986), 9.
13. Terry Lynn Karl, "The Hybrid Regimes of Central America," Journal of Democracy 6 (July 1995): 72-86. See also Terry Lynn Karl, "Dilemmas of Democratization in Latin America," Comparative Politics 23 (October 1990): 14-15.
14. Robert Dahl, Polyarchy, 248.
15. Larry Diamond, Juan J. Linz, and Seymour Martin Lipset, Democracy in Developing Countries, xvii.
16. Giovanni Sartori, Parties and Party Systems: A Framework for Analysis (Cambridge: Cambridge University Press, 1976): 230-38.
17. Guillermo O'Donnell, "Delegative Democracy," Journal of Democracy 5 (January 1994): 55-69; Larry Diamond, "Democracy in Latin America: Degrees, Illusions, and Directions for Consolidation," in Tom Farer, ed., Beyond Sovereignty: Collectively Defending Democracy in the Americas (Baltimore: Johns Hopkins University Press, 1996), 52-104; Larry Diamond, Developing Democracy, 42-50; Fareed Zakaria, "The Rise of Illiberal Democracy," Foreign Affairs 76 (November-December 1997): 22-43.
18. Liberal democracy extends freedom, fairness, transparency, accountability, and the rule of law from the electoral process into all other major aspects of governance and interest articulation, competition, and representation. See Larry Diamond, Developing Democracy, 10-13.
19. See Adrian Karatnycky, "The 2001 Freedom House Survey," Journal of Democracy 13 (January 2002): 99.
20. The only exception in this ambiguous group is Tonga, the lone "liberal autocracy"—a nondemocracy with a Freedom House score on civil liberties better than the mid-point of 4—and thus difficult to classify in this framework.
21. Larry Diamond, Developing Democracy, 117-19.
22. Freedom House, Freeedom in the World: The Annual Survey of Political Rights and Civil Liberties, 2000-2001 (New York: Freedom House, 2001), 121.
23. This draws from Jørgen Elklit and Palle Svensson, "What Makes Elections Free and Fair?" Journal of Democracy 8 (July 1997): 32-46. See also the essays on electoral administration in Andreas Schedler, Larry Diamond, and Marc F. Plattner, eds., The Self-Restraining State: Power and Accountability in New Democracies (Boulder, Colo.: Lynne Rienner, 1999): 75-142.
24. For a thoughtful critique of international election observation, see Thomas Carothers, "The Rise of Election Monitoring: The Observers Observed," Journal of Democracy 8 (July 1997): 16-31.
25. Space does not permit presentation of the detailed election data in this article. Two tables with these detailed results for selected ambiguous and electoral authoritarian regimes may be found on the Journal of Democracy website at www.journalofdemocracy.org.
26. For a classic critical treatment in this vein, see Richard L. Sklar, "Developmental Democracy," Comparative Studies in Society and History 29 (October 1987): 686-724.
27. For an Africanist perspective, see Richard L. Sklar, "The Significance of Mixed Government in Southern African Studies: A Preliminary Assessment," in Toyin Falola, ed., African Politics in Postimperial Times: The Essays of Richard L. Sklar (Trenton, N.J.: Africa World Press, 2002), 479-87.