On page one of the preface to their book on Guillermo O’Donnell’s legacy, Daniel Brinks, Marcelo Leiras, and Scott Mainwaring note that the relatively long duration of the current democratic and semi-democratic regimes in Latin America is unprecedented. Yet they are not able to register many important achievements by those regimes. They remark that “foremost among the achievements is the fact that they have survived” (vii). Juan Pablo Luna and Cristóbal Rovira Kaltwasser also observe in their edited book that, “contrary to the predictions of many who in the 1980s and 1990s discussed the prospects of Latin American democracies, the third wave of democratization has turned out to be durable” (15). And in Kenneth Roberts’s words in the same book: “Latin America’s new democratic regimes proved to be far more durable and resilient than anyone imagined at the onset of the wave of regime transition” (30).

The books selected by the journal editors for this review try both to explain such resilience and to address the low quality of most of the democratic experiences in the last few decades. The two questions seem to be strongly related. Democratic and semi-democratic regimes have lasted precisely because they have not delivered much change, that is, they have not threatened established interests or power positions with significant policy changes that could have triggered negative reactions against the new regimes. As a result, elected officials also face decreasing incentives to take risky innovative venues; they rather tend to try to compete, to gain, and stay in power as goals in themselves.

The Low Quality of Really Existing Democracy

Several contributions to Reflections on Uneven Democracies note that, given the durability of the new regimes, O’Donnell’s intellectual concern moved from the possibility of breakdowns to the perils of low democratic quality. O’Donnell had observed the weak and weakening “horizontal accountability” (or checks and balances) of the new regimes; his label “delegative democracy” is presented here not as a different regime type but as a set of practices implying both low citizen participation and low rulers’ responsibility.

Steven Levitsky and María Victoria Murillo, in particular, focus on the weakness of formal liberal-democratic institutions, either by low enforcement of the existing rules or by frequent change of them. In their chapter titled “Building Institutions on Weak Foundations,” the authors present numerous subnational territories as deprived of the rule of law and constituting “hybrid” or nondemocratic regimes. They remark that changes of formal legal rules have been rather inconsequential while, in contrast, informal rules and underlying norms of social control supported by old dominant social groups have proven to be resilient. Similarly, in another chapter titled “Economic Performance, Political Competition, and Regime Stability in Postwar Latin America,” Marcelo Leiras notes that the contrast between the policies that democratic and nondemocratic regimes implement is not as sharp as previously assumed.

Drawing on some of his own previous discussions, Philippe Schmitter, in his chapter “Reflections on ‘Transitology’: Before and After,” unveils “the dirty secret” of the new regimes’ survival: their scarce profound policy and substantive change, in contrast to many earlier periods. “The neodemocracies of the past four decades have been much less threatening to established interests than previously experienced or expected at the time” of democratization, Schmitter writes. Autocratic rulers have learned to make concessions in order to protect their interests within the new regimes. “Many of the beneficiaries of the ancient regime—civilian and military, private and public—have survived (even thrived) and come to play an important role in the subsequent one” (83).

Surely Schmitter is aware of the Latin European precedents of such strategies. In the French Second Republic (1848–1852), it was famously said that the more things change, the more they stay the same. In the Italian new unified kingdom about ten years later, according to the fictional Sicilian prince in The Leopard, “For things to remain the same, everything must change.” Nowadays in Spain, observers blame the return of old traditions of lawlessness and corruption after the frustration of greater expectations for change and modernization. Something of the sort can also be said of many Latin American countries in the most recent period.

Juan Pablo Luna and Cristóbal Rovira Kaltwasser also find this kind of explanation in their excellent edited monograph on The Resilience of the Latin American Right. The authors remind the reader that if the rightist political and social groups were excluded from the political system, they could promote—as in the past—political or regime instability. But they observe, again, that democratic competition has posed little threat to the political right and its core elite constituencies.

The authors describe how, in recent times, the electoral agenda in most countries is mostly made of valence issues, such as economic growth and modernization, anticorruption, crime and security, and other non-redistributive concerns, such as clientelism or religion. On socioeconomic issues, diminished ideological conflict lowers the stakes of political conflict, as market reforms were adopted in a bait-and-switch fashion by traditional populist or center-left parties or independent populist figures. Other position issues on which different parties could offer more clearly differentiated policy alternatives are not highly salient in most electoral campaigns.

In short: following the transformation of the left after the fall of communism, “The right has fewer reasons than before to be scared about the rise of left-of-center administrations” (Luna and Rovira, 16). They hold that even in countries where right-wing parties are almost nonexistent and left-of-center governments are in power, the right has the capacity to shape the political agenda and policy outcomes. Therefore, the rightist groups are much less inclined than before to promote the intervention of the military. At the same time, they are also responsible for the new regimes’ low institutional quality.

Most of these analyses suggest that changes in the international arena can help explain the combination of durable democracy and scarce policy change. This is in sharp contrast with the most usual analyses of democratization in the region a few decades ago. In their highly influential tentative conclusions on what at the time were still uncertain democracies, Guillermo O’Donnell and Philippe Schmitter barely mentioned the international context. Only the third editor of that collection, Laurence Whitehead, addressed the issue and expanded on it, in retrospect, fifteen years later. Now, in contrast, international factors seem to be an explanatory key.1

Several international developments, which are significantly interrelated, can be identified. First, the failure of Communism in the Soviet Union, the end of the Cold War, and the collapse of the Cuban revolutionary experiment have eliminated the fears of revolution in most countries. Second, and partly as a consequence of the first, the United States administrations have withdrawn support to anticommunist dictatorships. In contrast to the US support to a number of anticommunist SOBs during the Cold War, now the United States does not back military rule, rather, it has encouraged the establishment and consolidation of civil governments.

Yet there is another major external change that contributes to meager or postponed substantive performance. They are the new advances in economic globalization. Foreign pressures from both the United States and the global financial institutions include economic and political conditionality of loans and investments which overtly interfere in domestic affairs. This limits the capacity of national governments to develop certain policies and aggravates their deficient performance. The result is a still more scarce substantive policy change.

A major gap in these books is so-called populism, which, in Venezuela, Bolivia, and Ecuador, has indeed tried to introduce substantive policy changes, openly against the intentions of powerful international actors. Most authors in the set explicitly or implicitly bet on a gradual classification of political regimes rather than on a neat dichotomy democracy-nondemocracy; this permits them to identify “semi-democracies” and diverse forms of “hybrid” regimes. This may open the comparative lens to the entire region, but more specific analyses of those cases located on the closer side to authoritarian rule, such as some of those inspired by the populist surge, is lacking.

Overlooked Military

It is somewhat strange that relatively little attention has been given to the political role of the military during the last few decades. After all, the main immediate agency factor for nondictatorial regimes to be established and survive in most Latin American countries was the general withdrawal of the former dictators, mostly organized around the army, from ruling and politics.

Only some essays in Problems Confronting Contemporary Democracies, the collection in honor of Alfred Stepan, openly address the subject of the military’s involvement in politics. Stepan’s seminal analysis of the role of the military in both the breakdown and the reestablishment of democracy, initially in Brazil several decades ago and later on in a comparative perspective, appears still helpful to identify categories, types, and strategies. J. Samuel Fitch follows this path and sketches a sociological discussion about the armed forces in a few countries in his chapter “The Armed Forces and Society in Latin America: Change and Continuity.” He suggests that, under the current nonmilitary regimes, some potential discomfort among members of the military, as well as a lowered social esteem of their role, can be observed. Especially in countries that have shifted to all-volunteer armies, the military faces difficulties in recruiting soldiers and retaining talent due to reductions in military budgets and salaries and the existence of better alternative employments. The decreasing numbers of young people entering the military academies seem to be motivated primarily by economic and social mobility expectations rather than by “a calling” or vocation for military life.

The reduction of the size and the resources available to most armies in the region can help to explain how, in Fitch’s words, “over the past twenty-five years, democratic governments have in fact done an impressive job in reducing, if not eliminating, long-standing border conflicts and military rivalries … [so that] expectations of interstate wars in the region are virtually nil.” Also, “expectations of significant intrastate violence and insurgency are very low” (41). Several countries hint at eliminating the existing military forces a la Costa Rica. In those circumstances, Fitch notes, “it has been a difficult task for the South American militaries to rally public support for their conventional military role as deterrent to potential military threats” (41). In other words, people seem to accept the armed forces as long as they stay within democratic limits, but do not respect them or value the little they do.

Doubtless, the end of the Cold War, which produced both the subsequent vanishing of revolutionary alternatives and the new US policy in favor of civil rights and the rule of law, was in the background of regime change. But most of the previous dictatorships were military, not communist or revolutionary. Even the current leftist regime in Venezuela was established not by means of a revolution but by a military coup d’état. The main novelty to explain the unprecedented diffusion and duration of nondictatorial regimes in the region was not the defeat of revolutionary regimes but the choice by a vast majority of members of the military not to rule anymore. That’s why, by default of military dictators, civilian rule was inevitable and this rule had to be based on elections, some degree of freedom, and sufficient popular acceptance.

A possible way to address the subject would be to ask what changed in the armies during the recent period in comparison with the long previous history when the military often became rulers. Since independence from Spain, military groups led by caudillos and local chiefs were creating new countries by splitting the previous colonial administration into smaller units and taking over their governments. Very weak states, in terms of financial and organizational resources, were substituted with “strong” governments organized by the military with a command and control mind-set.

In subsequent periods, the weakness of the new states was also manifested in the absence of a sufficiently well-organized, professional, competent political class able to deliver to its subjects and to obtain sufficient social consent or acquiescence to their rule. Then, the military replaced missing or failed politicians on many occasions. But since the 1980s, the military were fed up of ruling and, deprived from external support and internal revolutionary threats, in most countries they withdrew from the first line of politics.

It could be inferred that new military levantamientos would not find the social support that some caudillos, coups d’état and dictatorships found in the past. Yet it would still be worthy to explore broader or deeper political motivations of the current military, including the retrospective evaluation they may make of the results of previous dictatorial experiences. In fact, the military is not weakened in Venezuela, Colombia, and other countries; even in Mexico, fighting narcotraffic and crime seems to have strengthened the army. All this should be taken into account to forecast whether the current professional discomfort that Fitch detects might eventually be expanded, once again, into more ambitious actions one day.

The Role of Actors’ Preferences

Scott Mainwaring and Aníbal Pérez-Liñán’s monograph, Democracies and Dictatorships in Latin America: Emergence, Survival, and Fall, develops a different point of view. Rather than discussing the vanishing of revolutionary threats, the withdrawal of the military, or the new international context, the authors focus on some political actors’ changing normative preferences for different types of political regime. The book is presented as a funeral of previous theories of regime change and stability. The memorial is sung for those who have tried to explain democratization as derived from the general modernization of the country, including previous economic development, education, and civility, or from a balanced social class structure; and for those seeking explanations for the failure of democracy in the drawbacks of natural resource dependence, negative reactions to the regime economic performance, or the perils of presidential regimes that foster high ambitions and concentration of power. The authors’ own alternative theory, however, does not look particularly strong.

Basically, Mainwaring and Pérez-Liñán postulate that democracy and dictatorship are the results of normative preferences of some relevant actors (presidents, powerful organizations, organized movements). In the authors’ words: “Actors’ purposeful action largely determines regime outcomes” (11, 30), and “regime change usually results from incumbents being replaced by new rulers who prefer a different regime” (34).

The authors have coded actors’ preferences and developed a trichotomous classification of political regimes per countries and years since 1900, for which task they have been able to use “sixteen research assistants over the course of a decade” (8). But they do not discuss why their “more stringent” classification is better than others, such as those of Polity IV (http://www.systemicpeace.org) or Freedom House (https://freedomhouse.org), for example, which are also trichotomous but based on a high number of measured variables, have a worldwide scope, and, in the case of Polity, covers a much longer period. They announce that, on the basis of their data set, they have proceeded to “reconstruct the probability of transitions and breakdowns in particular countries and years” (27). But although they claim to start in 1900, “because of data limitations, [their] quantitative explanation focuses exclusively on the period since 1945” (9).

At the end, the statistical model fails because it finds more variance for the independent variable than for the dependent one, that is, more variance for the set of values for preferences than for regime outcomes. In other words, the measures of actors’ normative preferences appear to be relatively dispersed in contrast to the simple, dichotomous or trichotomous, classification of political regimes and regime changes. This means that the supposedly explanatory variable, as it is widely dispersed, cannot explain a more compact set of political regime outcomes. The authors try to respond with the trick of “averaging actors’ preferences at the country-level” (296, also 78, 82), but then the explanatory power of specific actors’ preferences for every specific event of regime change disappears.

Mainwaring and Pérez-Liñán candidly collect a number of objections that they found at presenting their work to colleagues. They start with those that feel “patently obvious and hence trivial that if actors have a normative preference for democracy, transitions to competitive [democratic or semi-democratic] regimes will be more likely and breakdowns will be less likely.” Other criticisms are possible reverse causality (actual regimes induce normative preferences), or endogenous formation of preferences in political processes more generally. The classical question of how people’s preferences are formed remains unsolved (52–56).

To discard endogeneity or the possible interpretation that the most relevant actors experienced a kind of sudden religious conversion to the normative values of democracy, Mainwaring and Pérez-Liñán respond by emphasizing the role of previous experiences of democracy or dictatorship during the (unexplained) period 1900–1944 in the formation of preferences after 1945. But this does not completely discard endogeneity, as the previous preferences can also be endogenous. It seems to this reviewer that, if a researcher wanted to explain the formation of regime preferences on the basis of previous nonendogenous experiences, he should go at least to the origins of the movements for independence in the nineteenth century (which the Polity data set could help to approach) or even to the colonial legacy.

Absent such explorations, more traditional, well-grounded postulates remain unchallenged: democracy tends to be established without democrats (as dictatorships do not usually favor the diffusion of liberal and democratic values), by default when dictatorships collapse, and, in the best of the cases, it is the democratizing experience that makes and diffuses democratic preferences, rather than the other way around.

In the last half page of the book, the authors recognize that, after their exercise, still “several questions emerge.” As they list them, they look about the same as those initially presented: Why do regime preferences originate? How do they change? Under what conditions do certain preferences spread in society? The authors acknowledge that they have not definitively responded to them.

Methodological Issues

Some of these comments may be extended to certain methodological and style characteristics of other pieces of scholarship here reviewed. The numerous contributors to the collective books in homage to Stepan and to O’Donnell display different approaches and skills. But in the introduction and especially in the conclusion of the latter, titled “Studying Big Political Issues,” Mainwaring, Brinks, and Leiras try to summarize Whitehead’s previous summary of O’Donnell’s supposed hallmarks.

Guillermo O’Donnell was broadly appreciated for his insightfulness at identifying crucial questions in new situations and for his conceptual creativity (including “bureaucratic authoritarianism,” “transitions to democracy,” “delegative democracy,” etc.). His keen remarks and the coinage of new notions opened the path to the evaluation of the quality of democracy, among other developments. The editors of the book discuss these seminal contributions, but they do not succeed in converting them into a general methodology of political studies. As they present such a methodology, it basically includes readiness “to trade off big questions for methodological rigor,” inclusion of “strong normative convictions” as part of the academic work, “skepticism about universalistic theories,” deep knowledge of just a few cases, and openness to change the author’s mind and contradict oneself (16–17, 353–365).

Presented this way, this type of scholarship sounds almost as informal and weak as the democracies that are analyzed in the book. Foremost among its achievements may also be the fact that it has survived. Let me point out a major example. One of the most celebrated of O’Donnell’s insights was a nonlinear relationship between economic development and the probability of a durable democracy, as he held that at some middle levels of development, democracy was more vulnerable to an authoritarian breakdown than at both lower and higher levels of development. This was certainly not a “universalistic” model but a rather local one strongly based on deep knowledge of the special case of Argentina, O’Donnell’s home country, whose dictatorship in the 1970s Adam Przeworski identified as an outlier and almost an exception for the general relationships between democracy and development.2 The correlation hypothesized by O’Donnell has not generally been observed again during the last forty years. Undeniably, O’Donnell had “strong normative convictions” against the military regime in Argentina. The editors suggest that he also “changed his mind and contradict himself” about the importance or some implications of his analysis of that case and on other topics. But that may not fit the standard methodology able to produce cumulative knowledge based on scientifically inspired scholarship.

The book by Ignacio Walker, Democracy in Latin America: Between Hope and Despair, which is also included in the set to review, is more amateurish. His exercise is introduced as “a dialogue between academia and politics,” as the author seems to pretend to be on both sides of the dialogue. Actually, he is an active Chilean politician with an academic degree. Walker’s diagnosis is similar to those above reviewed: “Democracy has been more robust than the literature predicted for a region like Latin America (underdeveloped, Catholic, and presidentialist)”; at the same time, “there remains a great deficit in the quality of institutions and democratic governability” (7).

But the book is neither a high-quality academic monograph, as it does not make any substantial new contribution, nor a clearly targeted political speech. The author dares to say that he has “serious reservations about the current trends in political science” (xv), which, as he characterizes them, are nothing but a few high-school-level logical and mathematical techniques that he seems to have been unable to digest. He basically reproduces and repeats a few mantras about despair, hope, siren songs, the need to recognize that reality is much more complex and diverse than others presume, and the virtues of greater pragmatism, peppered with references between brackets to academic publications.

A Revisionist Proposal

Let’s finish with a more positive note. The most interesting piece of all in the five books reviewed may be the one signed by Juan J. Linz in collaboration with Thomas J. Miley in the book in honor to Stepan, which is titled “Cautionary and Unorthodox Thoughts about Democracy Today.” This is one of Linz’s very last published works and, in spite of, or perhaps thanks to his free style (mostly reflecting oral conversations), summarizes some of his most pungent legacies.

During most of his life, Juan Linz was obsessed with the drawbacks of democracy that led to its failure in the 1920s and 1930s. He celebrated the third wave of democratization, especially after the dismissal of the Soviet Union. But, thanks to his previous critical analyses of past experiences, he was probably better prepared than most to recognize the weaknesses of the new regimes and practices during the last fifteen or twenty years and to set the alarm. Juan Linz and Thomas J. Miley present their reflections “from a disillusioned or at least realistic view of democracy” (227). Although they do not focus strongly on Latin America, their concerns fit many worries about the region awfully well and can be widely shared.

The authors use some normative foundations of democracy to present a critical assessment of the existing democratic regimes and to identify a few crucial points for its possible improvement. The basic premise is that a real democracy should—“must”, they say—be based on a mix of three properties: “responsiveness, responsibility, and accountability” (228).

For the authors, it is unfortunate that contemporary democratic theory has focused on responsiveness to the neglect of responsibility. Responsiveness refers to the demos. It means that democratic politicians should respond to people’s demands. But in reality, many politicians running for office limit themselves to rehearsing banalities during electoral campaigns.

Responsibility, in contrast, does not refer to the demos, but it should imply a focus on the kratos. Linz reconsiders representative democracy and, like a number of authors with different backgrounds and specializations in the last few years, he states that it should be based on the Burkean ideal. This basically implies the absence of imperative mandates from the voters and the autonomy of the representatives to make decisions on behalf of the voters, but by using their own judgment. Linz remarks that politicians should have their own opinions and should make their decisions on the basis of (presumably) greater knowledge, information, and analysis than most voters. To do so, they should form the opinion of the citizens, educate them, tell them about the complexity of the issues, and disabuse them of their simplest predispositions, ideas, and sentiments. For Linz, real democracy works best when it is based, rather than on responsiveness to citizens’ alleged demands, on the responsible action of the elected leaders.

The problem is that, for politicians to be responsible, they need to fulfill the third property of a good democracy: accountability of rulers for their performance, which is extremely difficult to achieve. The only mechanism for such an endeavor are elections, whether based on a party system or on individual representatives. In reality, electoral competition does not tend to advance the quality of representatives. In many countries, the different groups in the elites are very similar, the authors observe. Even if they may look engaged in feisty competition, they share paramount common self-interests. Linz wonders: “How can we find leaders … willing and able to serve some kind of collective interest?” (237).

The question of unaccountable responsibility does not affect only the competence and probity of the elected representatives, but also those of the executive officers in the administration. In comparison with previous regimes, democracies are worse off in terms of performance and results unless they have an expert, nonpoliticized, professional bureaucracy. Yet many high bureaucrats tend to be captured by partisan biases and crooked appointments. The diagnosis is cruel but realistic enough.

In Conclusion: Still a Minimalist Democracy?

All the previous discussion could be summarized like this: The most consequential thing that has changed in most Latin American countries during the last thirty years is the disappearance of sufficiently powerful and credible actors and alternative projects against democracy, both domestic and foreign. Naturally, this does not mean that the current democratic systems work substantially better than some of the past historical experiences in terms of delivering public goods and services to the citizens. The currently existing, politically underdeveloped democracy is “sustainable” due to the absence or weakness of both revolutionary and coup plotters. But in most countries, the states are still relatively weak, while old traditions of politicians’ petty clientelism, incompetence, corruption, and ineffectiveness persist. In addition, while Latin America is newly challenged by the economic, cultural, and social consequences of globalization, the rulers of the region are barely taken into account in the world forums where these issues need to be addressed.

This disappointing assessment may suggest that in the joyful moments of change, perhaps many scholars (including the author of this review) did embrace the “minimalist” conception of democracy with a little too much eagerness. The main, still valid reason for such an intellectual choice is the analytical advantage of having a parsimoniously defined concept focusing on formal procedures, which permits to analyze its relations with other political, economic, and cultural variables and with its own outcomes without prejudgments. But once the analysis of the actual performance of existing political regimes is made—as many contributions in the reviewed books help to do—the “minimalist” notion does not provide solid elements for an evaluation. Especially when some “minimal” things change but many others just remain the same.

When efficient policy making fails and the quality and honesty of the politicians is below expectations, the classical promise of representative democracy to provide rule by the best is unfulfilled. Scaling back from foundational democratic promises and limiting the requirements to just broad suffrage elections with uncertain results reflects failures and disillusionments. Then, the most essential evaluative criterion left is merely that, in contrast to civil wars and dictatorships, the rulers can be dismissed by the ruled without bloodshed, to be put in Karl Popper’s classical words. Winston Churchill anticipated the idea with his famous quip. It’s not usually remembered that he completed it with the observation that “the multitudes remain plunged in ignorance of the simplest economic facts, and their leaders, seeking their votes, do not dare to undeceive them.” Even more sarcastic, George Bernard Shaw summarized that “democracy substitutes election by the incompetent many for appointment by the corrupt few.” Sadly, many recent developments tend to confirm and even expand such verdicts. Really existing democracies, not only but particularly in Latin America, were accepted with minimal requirements and have, indeed, remained “minimalist” in their performance and results.

Some crucial questions for the foreseeable future, thus, remain open: Will the current democratic politicians in Latin America be more reliable in building efficient states and democratic regimes? If they were not capable of obtaining broader and sustained popular support, what the alternatives would be? Should Bolivarian revolutions, military discomfort, or people’s disappointments be dismissed as only unfruitful reactions to the really existing and poorly delivering democracies, or are they a real threat to subvert them? Will the current regimes continue to survive just by default?