<%BANNER%>

Popular Support for Democracy in Newly Established Democracies: The Case of Peru and the Development of an Alternative T...


PAGE 1

POPULAR SUPPORT FOR DEMOCRACY IN NEWLY ESTABLISHED DEMOCRACIES: THE CASE OF PERU AND THE DEVELOPMENT OF AN ALTERNATIVE THEORETICAL AND METHODOLOGICAL APPROACH By JORGE ARAGON A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2006

PAGE 2

Copyright 2006 by Jorge Aragon

PAGE 3

To Sebastian, Andres and Mariana

PAGE 4

iv ACKNOWLEDGMENTS This dissertation would not have been possible without the support of the Department of Political Science at the University of Florida for almost six years. I am particularly grateful to my supervisory committee of Leslie Anderson, Philip Williams, Charles Wood, Peggy Kohn, and Bryon Moraski. They have been invaluable to me in helping me to improve this dissertation and also in helping me to be aware of the main weaknesses of my argument. Although Michael Martinez was not part of my supervisory committee, I want to thank him as well, because he was always very supportive of my development as scholar. Also, I want to thank two amazing and very generous colleagues, David Armstrong and Lee Walker with whom I had the opportunity to discuss my data analysis during the 2006 Inter-university Consortium for Political and Social Research (ICPSR) Summer Program in Quantitative Methods of Social Research. My fieldwork in Lima was possible because Guillermo Loli was not only a great friend but the most supportive one. Without his help, I would not be able to collect my data. I want to acknowledge also the help of Beatriz Venero and Ricardo Sanchez during my fieldwork in Peru. Working with Beatriz and Ricardo was a pleasure because of their professionalism and understanding. Furthermore, I want also to thank Lula Garca for being one of the most supportive librarians that I have met. Finally, I would like to thank three institutions: Apoyo Opinin y Mercado, the Instituto de Estudios Peruanos, and the ICPSR at the University of Michigan, for

PAGE 5

v providing me several key resources and the best environment to work in an academic endeavor.

PAGE 6

vi TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv ABSTRACT.......................................................................................................................ix 1 INTRODUCTION........................................................................................................1 Popular Support for Democracy in the Context of New Democracies.........................1 Research Questions.....................................................................................................10 Methodology...............................................................................................................13 Relevance of the Peruvian Case for the Study of Popular Support for Democracy in New Democracies...............................................................................................15 Dissertation Outline....................................................................................................19 2 NEW DEMOCRACIES, DEMOCRATIC CONSOLIDATION, AND POPULAR SUPPORT FOR DEMOCRACY...............................................................................23 New Democracies: Beyond Democratic Transition...................................................23 Main Contributions of the Literatu re on Democratic Consolidation...................25 Main Limitations and Problems of the Literature on Democratic Consolidation...................................................................................................38 New Democracies and Popular Support for Democracy............................................43 3 ADDRESSING AND EXPLAINING POPULAR SUPPORT FOR DEMOCRACY...........................................................................................................54 Political Culture Approach.........................................................................................56 Critical Review of the Political Culture Approach.....................................................64 Political Culture and Popular Support for Democracy...............................................71 4 POPULAR SUPPORT FOR DEMOCRACY IN COMPARATIVE PERSPECTIVE..........................................................................................................80 Established Democracies: Popular Support for Democracy and the Critical Citizens Thesis.......................................................................................................81 Popular Support for Democracy in New Democracies...............................................89 Post-communist Countries...................................................................................89 Meanings of democracy among post-communist citizens...........................92

PAGE 7

vii Measuring and explaining popular support for democracy in postcommunist countries...............................................................................105 Latin American Countries.................................................................................128 5 THE 1992 DEMOCRATIC BREAKDOWN IN PERU AND THE ROLE OF POPULAR SUPPORT FOR DEMOCRACY..........................................................139 Peru 1980........................................................................................................142 Explaining the 1992 Democratic Breakdown...........................................................155 Role of Popular Support for Democracy in the 1992 Democratic Breakdown........167 6 POPULAR SUPPORT FOR DEMOCRACY IN PERU..........................................174 Measuring Popular Support for Democracy in Developing Countries: The UNDPs Report on Democracy in Latin America................................................175 Popular Support for Democracy in Pe ru: Constructing Outcome Variables............183 Explaining Popular Support for Democracy in Peru................................................192 Impact of Socio-Economic Level on Popular Support for Democracy in Peru.......220 Conclusions...............................................................................................................223 7 CONCLUSIONS......................................................................................................232 Comparative and Critical Assessment of the Literature on Popular Support for Democracy............................................................................................................233 Popular Support for Democracy and the Prospects of Further Democratization in New Democracies.................................................................................................237 Alternative Theoretical and Methodological Approach toward Popular Support for Democracy in New Democracies..........................................................................242 APPENDIX A THE DEFINITION AND C ODING OF THE EXPLANATORY VARIABLES....251 Socio-Demographic Variables..................................................................................251 Citizen Definitions of Democracy Variables............................................................251 Civic or Political Culture Variables..........................................................................252 Social and Economic Expectations Variables..........................................................252 Assessment of Government and Re gime Performance Variables............................254 Political and Electoral Re presentation Variables.....................................................255 Fujimoris Variables.................................................................................................256 B QUALITATIVE DATA ON POPULAR SUPPORT FOR DEMOCRACY IN PERU........................................................................................................................258 LIST OF REFERENCES.................................................................................................276 BIOGRAPHICAL SKETCH...........................................................................................294

PAGE 8

viii LIST OF TABLES Table page 1-1 Citizen perceptions of democracy in Latin America (2000)..........................17 5-1 Number of victims reported to the Commission of Truth and Reconciliation.......145 5-2 Main problems in the country (1990)...........................................................146 5-3 Approval of Fujimori and Autogolpe (April-August 1992)...................................149 5-4 Confidence in institutions in Peru (1989)....................................................152 5-5 Approval of the president, legisl ature, and judiciary in Lima (1987)..........152 6-1 Selected questions to assess popular support for democracy in Latin America, United Nations Development Programmes Report on Democracy in Latin America ..................................................................................................................177 6-2 Factorial loads for the eleven questions used to assess popular support for democracy, United Nations Development Programmes Report on Democracy in Latin America .........................................................................................................179 6-3 Sample and Metropolitan Lima population distributions: Gender, age, and socioeconomic level........................................................................................................184 6-4 Factorial loads for the eleven questions used to assess popular support for democracy, VARIMAX Rotation (Peru, 2004).....................................................186 6-5 Reliability coefficients for the set of variables associated with each different dimension of popular support for democracy (Peru, 2004)....................................189 6-6 Support for democracy as a form of government, ordinary least squares (OLS) regression estimates (Peru, 2004)...........................................................................205 6-7 Support for democratic procedures, ordinary least squares (OLS) regression estimates (Peru, 2004)............................................................................................207 6-8 Support for democratic institutions, ordinary least squares (OLS) regression estimates (Peru, 2004)............................................................................................209

PAGE 9

ix Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy POPULAR SUPPORT FOR DEMOCRACY IN NEWLY ESTABLISHED DEMOCRACIES: THE CASE OF PERU AND THE DEVELOPMENT OF AN ALTERNATIVE THEORETICAL AND METHODOLOGICAL APPROACH By Jorge Aragon December 2006 Chair: Leslie Anderson Major Department: Political Science My study concerns popular support for democracy, particularly in newer and less consolidated democracies. Political scientists who deal with contemporary processes of democratization and the prospects for democratic consolidation have devoted much attention to the role of political elites, and to the merits and liabilities of different institutional designs. They have devoted less attention to the role played by ordinary citizens. Without neglecting the importance of elites and institutions in the creation and maintenance of democracy, I argue that overlooking the role of ordinary citizens is a serious limitation when we try to understand the advances and setbacks common to newly established democratic regimes. As numerous historical and contemporary cases show, much of what elites attempt to do is conditioned by their judgments of what ordinary people think and how they behave. My focus on the political development of these newly and less consolidated democracies is organized in terms of two fundamental questions. First, what are the

PAGE 10

x political consequences of a prevalent and persistent lack of support for democracy among the majority of a countrys population? Second, what is the explanation for the current degree of support for democracy that we witness in countries that have only recently become democratic? Unfortunately, much of the recent literature on popular support for democracy in newly established democratic regimes is of limited utility when it comes to providing answers to the two main question posed here. In most cases, studies of popular support for democracy adopt theoretical approaches and methodological strategies that are illequipped to grasp the full complexity of the phenomenon. In the face of these deficiencies, I developed a new theory and method to account for the multi-dimensional nature of the relationships at hand, and the multiple factors that influence the degree to which ordinary citizens support democracy.

PAGE 11

1 CHAPTER 1 INTRODUCTION Popular Support for Democracy in the Context of New Democracies This dissertation addresses the topic of popular or mass support for democracy in the context of new democracies, and, more specifically, the case of the most fragile or less consolidated ones. In doing that, I have tried to provide answers to the two following questions: (1) what are the main the consequences of a lack of a considerable and stable level of popular support for democracy in new democratic regimes and (2) how to address and explain current levels of popular support for democracy in new democratic regimes. Regarding the first question, I contend here that, in the context of several new democracies, the lack of advances in democratization, and even the cases of democratic setbacks, are highly related to a lack of popular support for these new democratic regimes; in particular, a lack of popular support for democratic procedures and institutions. As Hagopian (2006:320) points out for the case of Latin America, unlike the 1960s and 1970s, when democracies were weakened by the clash of ordinary people demanding the rights that elites were reluctant to grantto become citizenstoday the crisis of democracy is associated with the widespread and deep disillusion among citizens. Accordingly, citizen dissatisfaction has led to low levels of public trust in

PAGE 12

2 political institutions and political parties, and to popular support for strong-arm, antiestablishment, or neopopulist1 leaders with very ambiguous attitudes toward democracy. I am aware that not everyone will share the conviction that popular support for democracy matters to a great extent vis--vis the current and future situation of new democracies. However, I think it is critical to recognize that much of what elites attempt to do in these new democratic regimes is heavily conditioned by their judgments of how ordinary people will behave (Rose, Mishler, and Haerpfer 1998; Bermeo 2003). By the same token, I am aware that someone can argue that new democracies are not the only ones in trouble regarding current levels of popular support (Pharr and Putnam 2002). However, I think that it is also important to acknowledge that the level of citizen support for democracy is far lower in new democracies that in the established ones, and, moreover, that the dissatisfaction and skepticism about government and authorities within established democracies has not produced yet the same degree of regime instability. Regarding the consequences of lack of popular support for democracy, I contend here that answering why support for democracy is declining in several democracies is not an easy task. Nevertheless, as it has been noted for several Latin American countries, the performance of democratic governments has been very poor where new democratic regimes have been less stable. Specifically, when new democratic regimes have failed to promote growth, reduce poverty, address the causes of social and economic inequality, provide security, and establish the rule of law; or, in other words, when they have failed 1 For a definition of neopopulism, particularly in the context of Latin American countries see Weyland (1996, 2001).

PAGE 13

3 to address citizen needs, popular satisfaction with democracy has declined (Hagopian and Mainwaring 2006). Therefore, in trying to understand what can make new democratic regimes more democratic, it is critical to address citizen expectations of government, perceptions of the policy course that governments set, and evaluations of the success of those policies (Hagopian 2006). In that regard, one of the main contributions of this dissertation is that it shows how the combination of structural variables such as poverty and inequality, and the poor performance of both democratic governments and regimes in addressing citizen needs and expectations has indeed a corrosive effect on popular support for democracy, and, as a consequence of this, on the stability and quality of the new democratic regimes. The recognition of both that popular support for democracy matters and that democratic performance is a key determinant of it is extremely relevant to the study of new democracies. Accordingly, what allows a democracy to emerge and survive does not guarantee that democracy will be effective or immune from anti-system challenges and citizen dissatisfaction. In other words, whereas elite attitudes toward democracy and a favorable international political environment have been key factors in understanding the maintenance of competitive elected regimes in Latin America, the consolidation of new democratic regimes, or, simply, further democratization (i.e., the democratization of semi-democratic regimes) is better explained by the interplay of structural factors, regime performance, and popular political attitudes (Mainwaring and Prez-Lin 2006) In addressing the topic of popular support for democracy in new democracies, I have confronted several and different challenges. First, despite the common use of the term popular or mass support for democracy it is only rarely acknowledged that this

PAGE 14

4 particular political attitude involves different components or dimensions.2 In that regard, it is critical to address the extent to which citizens living in new and fragile democracies lack support for democracy as a form of political regime based on public contestation and the right to participate in elections and office (Dahl 1971) or, on the contrary whether the lack of support is with respect to current democratic regimes and democratic authorities. As it will be addressed later, our perceptions and conclusions regarding the future of new and fragile democracies will change considerably depending on the dimensions of popular support for democracy we consider. Second, despite the growth of an important literature about the recent processes of democratization outside the realm of deve loped countries (Bunce 2000), very few works have specifically addressed the topic of popular support for democracy, and often, the topic has been frequently addressed in a tangential way.3 Most of the time, popular support for democracy has just been considered as one among several factors that can play a role in the future prospects of these new democracies, and very little has been said about how this political phenomenon is interrelated to these other critical factors.4 Therefore, in spite of the widespread r ecognition of the theoretical importance of popular support for democracy for the maintenance and consolidation of new democracies, relatively few studies have investigated, in a systematic and comprehensive way, its main determinants. As a consequence of this, we lack a comprehensive view regarding the challenges that most of these new democratic regimes are confronting to 2 Norris (1999a) and Dalton (1999) provide a theoretical and empirical treatment of the main dimensions of political support in a democratic setting. 3 Rose, Mishler, and Haerpfer (1998) is one of the few cases where, in dealing with new democracies, explaining support for democracy and for its alternatives was its main goal. 4 See for example Diamond, Linz, and Lipset (1989) and Diamond (1999).

PAGE 15

5 achieve higher levels of democratization. Part of the explanation for the lack of a systematic and comprehensive approach toward popular support for democracy in the context of new democracies is related to the difficulties of studying political support. According to Kornberg (1990), investigators dealing with political support in democratic societies have found it extremely difficult to unravel and separately measure the different types or components of political suppor t (e.g., diffuse and specific support).5 Also, and equally important, investigators have generally been unable to provide satisfactory answers to so what questions linking differences in support levels to variations in peoples political behavior. Therefore, in studying popular support for democracy in new democracies, scholars have faced not only difficulties in dealing empirically with this type of political support, but also difficu lties in connecting individuals attitudinal orientations with individuals political behavioral orientations.6 There is no doubt such difficulties may explain why scholars dealing with new democracies have been reluctant to address the topic of popular support for democracy in more specific and systematic terms. Third, there have been also important limitations in the theoretical and methodological ways in which popular support for democracy in new democracies has been approached. Regarding the theoretical limitations, a large majority of scholars who 5 According to Easton (1965), the distinguishing feature of specific support is that it is based on a kind of rational calculus, i.e., support is giving or withheld de pending upon whether a politi cal object generates affects that are considered beneficial or detrimental to ones self or group. On the contrary, the essential feature of diffuse support is its affective quality, i.e., support is granted or withheld on the basis of simple but powerful feelings of like or dislike. 6 It is true that there are several so what questions that we are far from having satisfactory answers for, e.g., to what extent is popular support for democracy connected with electoral behavior, and to conventional and unconventional forms of political participation; or, to what extent is popular support for democracy shaping citizen willingness to comply w ith the authoritative governments decisions and policies. However, it is also true that the erosion of popular support for democracy appears frequently as one of the most powerful causes of democratic reversals (see Chapter 5).

PAGE 16

6 have dealt directly or indirectly with this topic seem to ignore that individuals political orientations are strongly influenced by (a) deep-seated political values and (b) specific assessments about the ability of the regime and its authorities to perform its ascribed functions Unfortunately, most of the time, the level of popular support for democracy among citizens from new democracies has been addressed as a function of the level of citizens democratic values; more specifically, regarding the existence or not of a civic culture (Almond and Verba 1963).7 As a reaction to all these limitation, and trying to fill a critical void in the literature on popular support for democracy, this dissertation will test the extent to which nonpolitical values variablessuch as individuals social and economic expectations, and individuals assessment of democratic performanceplay a critical role in shaping it. I will try to demonstrate that neglecting the role of individuals expectations as well as the role of individuals assessments regarding the performance of the current democratic regime produces partial and misleading interpretations, not only about the level of popular support for democracy and current democratic regimes, but also about the prospects of new democracies. Consequently, one of the main contributions of this dissertation is to make clear that accounting for the main determinants of popular support for democracy requires that analysts consider a complex and diverse set of explanatory variables, and to pay attention to what some of these main determinants have in common.8 Accordingly, the empirical analysis conducted in this dissertation makes clear that some structural factors such as 7 See Chapter 4, and the section dealing with the study of popular support for democracy within Latin American countries, in particular. 8 For an example of how this theoretical approach is used for studying support for democracy in established democracies, see Dalton (2004).

PAGE 17

7 class or socio-economic level exert an important influence on popular support for democracy. In this regard, I have found that the most disadvantaged population in Metropolitan Lima is, at the same time, the most dissatisfied with the performance of their current democratic government and democratic regime, and the one that presents the lowest levels of support for democracy as political regime, democratic procedures, and democratic institutions. Therefore, I believe that this dissertation reveals the urgency of overcoming the tension between individual-agency and structural explanations that have characterized the study of the recent process of democratization.9 By the same token, it is worth noting here also that the existing literature on popular support for democracy among new democracies has focused mostly on between differences (i.e., differences among countries considered as a whole) and much less on within differences (i.e., differences within one specific country). Accordingly, most scholars have been much more concerned with how to explain the differences in popular support for democracy between established democracies and new democracies, or between new democracies, rather than addressing the critical differences in terms of popular support for democracy that can be found within specific new democracies. This explains why the effect of class or differe nt socio-economic levels on popular support for democracy within new democracies has been largely neglected. This is a critical limitation with respect to how political scientists have been approaching the study of popular support for democracy among new democracies.10 9 See Landman (2000) for how this has been the case fo r the literature dealing with democratic transitions in the late twentieth century. 10 This issue needs to be address w ith a greater sense of urgency becau se there is some evidence showing that within differences are, very frequently, as important as between differences (Moreno 2001, Aragn 2001).

PAGE 18

8 With respect to the most common methodological shortcomings regarding the study of popular support for democracy within new democracies, I will show here that it is absolutely critical to review and improve the way that most scholars have operationalized and measured this complex phenomenon. Perhaps too frequently, it is assumed that one or two survey questions about the preferred political regime (i.e., democratic or non-democratic) and the individual level of satisfaction regarding the way in which democracy is working in the country are sufficient to provide reliable and valid measures of popular support for democracy. This unsatisfactory methodological strategy can be found even in some of the most influential comparative works about democratization in the late twentieth century.11 As a reaction to this situation, this dissertation will present an innovative way to deal empirically with the topic of popular support for democracy. For several reasons, I will contend that this new methodological strategy constitutes a significant improvement for the study of popular support for democracy in new democracies. In that regard, I have been able to identify and empirically measure three different components or dimensions of popular support for democracy (i.e., support for democracy as form of political regime, support for democratic procedures,12 and support for democratic institutions). Additionally, and departing from previous scholarly works, I have identified a set of variables, and construct indexes, that seem re liable and valid tools to account for the main determinants of popular support for democracy. 11 See for example, Linz and Stepan (1996) and Diamond (1999). 12 See ODonnell (1994) for the relationship between the lack of institutionalization of democratic procedures and the development of a delegative version of democracy. In that regard, ODonnell makes absolutely clear why delegative democracies in Latin America are not good news for the development and institutionalization of democratic regimes in this region.

PAGE 19

9 Fourth, regarding the main challenges for the study of popular support for democracy, it was clear to me that the most important theoretical and methodological advancements regarding the study of popular support from democracy come from scholars who are dealing with this topic from the perspective of established democracies. This poses some problems because the phenomenon of popular support for democracy between established and new democracies differs in important ways. Established and several new democracies are alike in the sense that the consent of the people, not coercion, explain its persistence. However, it is only within established democracies that citizens clearly separate support for their political authorities from support for their political regimes and political communities. In a typical established democracy, we frequently find a high level of support for the political community, a low level of support for political authorities, and a level of support for the political regimeincluding the main democratic norms and procedureswhich fall between these two extremes. This is not the necessarily the case among several new democracies. In fact, in these cases, low popular political support is not constrained to political authorities, but may also include democratic rules and procedures, and, in some cases, the current democratic regime itself. In this sense, the main forces shaping popular support for democracy are entirely different between established and new democracies. Finally, it is my main expectation that a better understanding of the phenomenon of popular support for democracy in new democracies will contribute to improve our understanding regarding some key issues related to these recent processes of democratization. In particular, why is it difficult to develop popular support for current democratic regimes and key democratic rules and procedures among several new

PAGE 20

10 democracies; and, why and how ordinary citizens13 ideas, beliefs, and expectations matter for the prospects of a process of further democratization. The need for addressing this connection between popular support for democracyon the handand the possibility of democratic setbacks, stagnations or advancementson the other handled me to carefully evaluate the existing literature, identify a relevant empirical case, and elaborate an adequate research design. On the latter point, and after identifying important limitations within the literature dealing with popular support for democracy, I also had to develop particular and innovative theoretical and methodological approaches to address this political phenomenon. Accordingly, this dissertation is not only an argument in favor of the relevance of popular support for democracy as one of the most important factors shaping the political development of new democratic regimes, but it also makes an important contribution to the existing literature on this topic, in that it seeks to improve upon the theoretical and methodological approaches used to address popular support for democracy, particularly in the context of new and fragile democracies. Research Questions In addressing the topic of popular support for democracy, this dissertation attempts to make a contribution to the discussion on the survival and quality of new democratic regimes. Regarding these recent process of democratization, some of the most important issues are framed by the following questions: why is that so few of them are on route to becoming well-functioning and stable democracies? Or, why is that so few of them are 13 The term ordinary citizens comes mainly from Bermeo (2003), Linz and Stepan (1996) and Tarrow (1994). As Bermeo (2003:3) contends, the term ordina ry underscores the fact that we are considering people or citizens that do not have extraordinary powers vis--vis the state in whic h they live. Accordingly, they are not politicians or military officers, but people or citizens that spend most of their lives in personal endeavorsearning money, supporting families, and pursui ng whatever leisure activities their social status allows. Also, they are the people who compose the vast majority of the citizenry in virtually every country in the world.

PAGE 21

11 characterized by still having a positive dynamic in terms of further democratization? Alternatively, why is that an overwhelming majority of these new democratic regimes do not seem to be deepening or advancing whatever democratic progress they have made as part of their democratic transitions? (Carothers 2002) In summarizing this situation, ODonnell (2004b, 2004c) contends that the problem with new democracies is that a majority of them are characterized as being low quality democracies not only in terms of governance (i.e., the capacity to formulate and implement policies in an effective manner), but also in terms of providing the necessary conditions to make these new democratic governments more representative and accountable, particularly regarding the needs and desires of the most disadvantaged populations. I believe that this deficit in the quality of democracy is one of the main reasons why, as is often the case, there is a lack of stable and strong popular support for the new democratic regimes. This deficit of popular democratic legitimacy is not necessarily related to democracy as an ideal form of political regime, but instead related to current democratic regimes, democratic procedures, democratic institutions, and democratic authorities. Nevertheless, it is not the case that a low level of popular support for democracy will always lead to an undisputed democratic breakdown (e.g., Peru in 1992). More frequently, this lack of popular support for some democratic procedures and institutions will be a critical component for the development of hybrid regimes (e.g., current Venezuela14 and Russia15). Accordingly, it is evident to me that democratic breakdowns or the development of stable hybrid regimes (Hartlyn 2002) are among the reasons why 14 See McCoy and Myers (2004). 15 See Fish (2005) and McFaul, Petrov and Ryabov (2004).

PAGE 22

12 several new democracies are experiencing significant difficulties in terms of making sustained advances with regard to further democratic development. Therefore, this dissertation seeks to address two main questions. First, what are the main determinants of popular support for democracy in new democracies? Second, in the context of new democracies, what does the topic of popular support for democracy reveal about the nature of these political regimes and their challenges for a further process of democratization? Regarding the first question, I will consider the impact of (a) individuals understandings of democracy, (b) individuals social and economic expectations toward their own society, (c) individuals assessments of regime and governments performances, (d) the level of individuals civic culture orientations, and (e) some of the main individuals sociodemographic factorsincluding socio-economic levelon popular support for democracy. These findings should provide me with the opportunity to address a related question: to what extent does poverty and inequalitytwo of the most frequent characteristics found within new democracies in developing countriesexert a powerful influence on popular support for democracy either in a direct or indirect way? On the latter point, for example, I was expecting that living in poverty could be strongly related to some of the most important determinants of popular support for democracy such as particular ways of defining democracy, clear preferences vis--vis how society should be organized and the role of government in regulating the economy, and/or very poor evaluations of democratic regime and governments performance. With respect to the second question, and noting the connection between low quality democracies, lack of popular support for democracy, and the development of authoritarian and stable hybrid regimes, I contend that having a better understanding of

PAGE 23

13 the determinants of popular support for democracy should provide important insights regarding the main sources of citizen dissatisfaction and frustration with their current democratic experience and current democratic regimes. Such insights may reveal valuable information that may prove helpful for achieving finally gradual improvement of the democratic quality of these new democratic regimes. Methodology The theoretical part of this work, as well as the methodological discussion of how popular support for democracy has been addressed in the last yearsparticularly in the context of new democracieshave been conducted in the tradition of comparative studies. Looking for similarities and differences regarding what other scholars have found on this topic for different countries and regions provide, I tried to acquire a better understanding of this phenomenon. Additionally, I have been also able to develop an argument on how addressing the topic of popular support for democracy contributes, in a fundamental way, to an improvement of our understanding of the particular nature of these new democratic regimes. Unfortunately, this time, I was limited to work only with one empirical case. Accordingly, the empirical component of this dissertation is based on a public opinion survey that I designed and conducted in Peru during the second half of 2004. In that regard, I was able to design a tailored questionnaire that specifically addressed the topic of popular support for democracy. I was assisted by twenty experienced interviewers, and I was able to collect 420 face-to-face interviews with Peruvians older that 18 years old, who resided in the countrys capital, Lima. The quantitative orientation of this dissertation has, however, two main distinctive features. First, it was heavily influenced by my qualitative research on this topic. Between 1997 and 2004, I carried out several focus groups with Peruvian citizens from

PAGE 24

14 different socio-economic levels. My analysis of these sessions revealed to me that there are critical differences in the ideological and attitudinal orientations of Peruvians of different socio-economic levels toward politics and democracy. With specific reference to citizens from the lower socio-economic levels, I learned that more than an authoritarian orientation they hold either a profound indifference toward what can happen to the current democratic regime or a very critical view regarding how the current democratic regime is unable to deliver what a democratic regime is supposed to provide. All these insights were very useful for the development of my questionnaire about popular support for democracy. Second, I conducted an extensive review of some of the most important public opinion research projects that have been carried out in the last few years in Peru and other Latin American countries.16 These projects constituted an invaluable source of information and expertise regarding the use of public opinion surveys to generate valid and reliable measures on popular support for democracy in new democracies. For my analysis of these data, I used factor analyses to identify three different and critical dimensions of popular support for democracy and to operationalize each one of theses dependent or outcome variable. For the analysis of the main determinants of popular support for democracy, I used ordinary least squares (OLS) regression models. After analyzing different sets of expl anatory variables separately, e.g., sociodemographic factors, meanings and expectations toward democracy, economic and social expectations, civic culture variables, and assessments of democratic government and 16 Latinobarmetro, Latin American Public Opinion Project (LAPOP), a public opinion project founded by the Hewlett Foundation, and Apoyo Opinin y Mercados monthly public opinion report.

PAGE 25

15 regimes performance. I developed some general models to account for the main determinants of popular support for democracy. Relevance of the Peruvian Case for the Study of Popular Support for Democracy in New Democracies In this section I address the extent to which Peru can be considered a representative case in terms of popular support for democracy in new democracies. In that regard, and without neglecting the existence of some exceptional features in its recent political experience,17 I will contend that Peru is a very representative case of one particular kind of new democracies in Latin America, as well as in other regions of the world. Accordingly, some scholars (Hagopian and Mainwaring 2006, Mainwaring and Prez-Lin 2006) have suggested that curre nt new democracies in Latin Americaor, post-third-wave democratic regimescan be classified between countries where their democratic regimes have become more solid (e.g., Uruguay and Chile) and countries where their democratic or semi-democratic regimes have avoidedalmost all the time democratic regression, but where there is also clear that there has been a clear stagnation in their processes of democratization, particularly since the beginning of the 1990s (e.g., Bolivia, Ecuador, Colombia, Guatemala, and Venezuela).18 I do not have any doubts that Peru clearly belongs to this latter group, and th at its political and democratic experience it is not very different from what is happeni ng in other cases of democratic stagnation. One way to support my last contention is to look at some of the most standard public opinion survey questions that are used to gauge the citizen perception of 17 See Chapter 5 for a description of how Peru during the late 1980s and early 1990s confronted not only a severe economic crisis, but also an internal war conflict that definitivel y. This combination, definitively, increased to a large extent the level of desperation and the dissatisfaction with the democratic authorities among Peruvians. 18 Similar classifications can be found also within sc holars dealing with new democratic regimes in subSaharan Africa (Villaln a nd VonDoepp 2005, Hyden 2006).

PAGE 26

16 democracy. In that direction, Hagopian ( 2006:333) has provided a summary of these indicators using findings from the Lati nobarmetro between 2000 and 2004 for 17 Latin American countries (Table 1-1). This summary includes: levels of support for democracy (2000 average); numbers responding that democracy is the best governmental system (2000 average), levels of satisfaction with democracy (2000), levels of trust in government (2003), and numbers responding that voting can make a difference (2004). As Table 1-1 clearly shows, countries such as Uruguay, Costa Rica, Chile and Argentina are systematically at the top part of the table. On the contrary, Peru is almost always below the regional average, together with El Salvador, Colombia, Bolivia, Ecuador, Paraguay and Guatemala. As a consequence of this, it is not a coincidence how frequently non-constitutional depositions of democratically elected presidents,19 strong electoral support for anti-party or anti-systems leaders,20 and even democratic regressions21 can be found within this latter group of Latin American countries. By the same token, if we consider that there is some evidence that in the case of Latin America there is a relationship between the level of support for democracy and satisfaction, and the quality of democratic governance, particularly in terms of effectiveness, transparency, and fairness (Hagopian 2006); it is possible to contend that addressing the topic of popular support for democracy considering performance variables 19 Jamil Mahuad in Ecuador in 2000, Gonzalo Snchez de Lozada in Bolivia in 2003, and Lucio Gutirrez in Ecuador in 2005. 20 Hugo Chvez in Venezuela, Lucio Gutirrez in Ecuador, and Alberto Fujimori and Ollanta Humala in Peru. 21 Peru in 1992 and Venezuela since 1998.

PAGE 27

17 can contribute to a large extent to improve our understanding of these cases of democratic stagnation within new democracies. Table 1-1. Citizen perceptions of democracy in Latin America (2000) Country Support for democracya Satisfaction with democracyb Importance of votingc Trust in governmentd Democracy is the best systeme Uruguay 79 53 71 19 84 Costa Rica 75 56 51 23 80 Argentina 65 28 65 45 80 Chile 52 32 53 46 71 Panama 53 34 72 20 71 Venezuela 67 43 68 28 74 Honduras 56 41 49 25 72 Brazil 36 23 59 42 69 Nicaragua 52 30 50 12 66 Mexico 52 23 52 24 75 El Salvador 45 31 61 21 61 Colombia 43 20 57 33 64 Peru 56 15 54 14 59 Bolivia 53 21 37 14 62 Ecuador 47 19 45 10 55 Paraguay 41 10 50 16 55 Guatemala 38 26 54 7 46 Latin America 54 30 60 24 68 a Percentage agreeing with statement, "Democracy is preferable to any other kind of government," Latinobarmetro (2000 average). b Percentage responding they are "very satisfied" and "fairly satisfied" with the functioning of democracy, Latinobarmetro (2000 average). c Percentage responding "How one votes can make th ings different in future," (Latinobarmetro 2004). d Percentage expressing "a lot of confiden ce" and "some confidence" in government, Latinobarmetro (2003). e Percentage agreeing with the statement, "Democracy may have some problems, but it is the best system of government, Latinobarmetro (2003 average). Source: Hagopian (2006), pp. 334. Specifically about the study of popular support for democracy, addressing the case of Peru provides a new opportunity to advance our understanding of some common patterns that have been found in other regions of the world: the existence of a widespread but shallow or superficial popular support for democracy. As it will be evident, Peru

PAGE 28

18 shares with several new democracies in sub-Saharan Africa the coexistence of a widespread popular support for the idea of democracy or the rule for the people; with some pockets of authoritarianism, and very ambiguous popular orientations toward democratic authorities, institutions, and procedures (Bratton, Mattes and Gyimah-Boadi 2004:345). Finally, I want to mention something about to what extent the findings that I will be reporting here can be generalized to the entire country. My public opinion survey has not included Peruvian population outside Metropolita n Lima. Therefore, in a country with a long tradition of citizen resentment of Lima-dominated political institutions and deep desire for greater decentralization and regional autonomy, the fact that my survey does not get the urban-rural divide is an important limitation for the generalization of these findings. However, the fact that Metropolitan concentrates almost a third of the Peruvian population, and around 40% of the total urba n population makes these findings highly significant regarding what is happening in Peru in terms of popular support for democracy. Even more, I would expect that research on popular support for democracy in coastal cities in Peru (e.g., Piura, Chiclayo, Trujillo, Chimbote, Ica, and Tacna) will likely produce similar results. Finally, I think it is possible to contend that some of the trends found within the Metropolitan Lima population are even stronger within the rural and non-coastal urban Peruvian population. In that regard, a national public opinion survey conducted in Peru in 2005 reveals systematically higher levels of political and democratic dissatisfaction and alienation within the rural and non-coastal urban population.

PAGE 29

19 Dissertation Outline This dissertation is divided into two parts. Part I comprises Chapter 2 through Chapter 4; here, the relevant theories and research literature related to the topic of popular support for democracyparticularly in the context of recent processes of democratizationare presented and discusses. Part II comprises two chapters (Chapter 5 and 6) and focuses on a specific case, that of contemporary Peru, which provides the empirical evidence for this work. Additional details on each chapter are given below. Chapter 2 examines the connection between popular support for democracy and the consolidation of new democracies. I contend that the prospects of further and gradual democratization depend to a large extent on the level and resilience of popular support for democracy. It is not my intention here to claim that there is an automatic and mechanic correspondence between the level of popular support for democracy and the level of democratization; or, to claim that popular support for democracy is the most powerful factor shaping the political future of new democracies. On the contrary, as Rose, Mishler, and Haerpfer (1998) have suggested for the case of post-communist countries, I believe that the lack or presence of a considerable level of popular support for democracy will exert a powerful influence on political elite decisions. In that regard, and in the context of weakly institutionalized democratic regimes, it is not going to be particularly difficult to break democratic norms and procedures when ordinary citizens are either profoundly dissatisfied with the performances of the current democratic regime or profoundly indifferent regarding the possibility of a regime change. Chapter 3 reviews the main theoretical approaches that can be used to address the phenomenon of popular support for democracy in new democracies. I contend that the typical political culture approach (i.e., the one that is mainly interested in political

PAGE 30

20 socialization and the acquisition of deep-seated political values) is insufficient to account for this complex political phenomenon. Some non-typical political values, such as specific preferences regarding their own society, play a critical role in shaping popular political support; but so do evaluations regarding the different performances of both democratic governments and democratic regimes, and some structural factors such as class or socio-economic level. Chapter 4 presents a comparative and critical assessment of the literature on popular support for democracy in established democracies, post-communist countries and Latin American countries. The main goal was to acquire more insights on cases with which I am less familiar, that is, non-Latin American countries. At present, the highest level of theoretical and methodological advancement regarding the study of popular support arises from critical evaluation of cases of established democracies in advanced industrial countries. However, it is also the case that research on popular support for democracy within post-communist countries is more abundant, and, overall, better grounded in theoretical and methodological terms than comparable research on Latin American countries. Chapter 5 deals with the Peruvian political experience over the last few decades and in particular with the 1992 democratic breakdown and the role that popular support for democracy played with respect to this outcome. After presenting a very brief description of the main political events in Peru between 1980 and 2001, I address and discuss some of the main explanations of this democratic breakdown. I will contend that despite the relevance of some institutional explanations, there is no doubt that a severe decline among Peruvian citizen support for some of the main democratic institutions and

PAGE 31

21 procedures played a critical role in the rise of an authoritarian regime between 1992 and 2000.22 Chapter 6 presents and analyzes the data that were collected in Peru in November 2004. These data are representative of the Peruvian population that is older than 18 years, living in the capital of the country. In the first part, the aim is to operationalize and measure the concept of support for democracy. Several questions were analyzed using factor analyses to test the multidimensional nature of popular support for democracy. In the second part, different OLS regression models were developed to account for the each one of the dimensions of support for democracy previously identified. Overall, I am expecting that this work can be used in the development of more accurate methodology to measure and assess popular support for democracy among new democracies.23 Finally, Chapter 7 summarizes the main findings and conclusions of this work. It is clear for me that one of these main conclusions is related to the fact that despite the relevance of popular support for democracy for the future of new democracies, it is the case that existing research literature on this topic, particularly with respect to new democracies, has several theoretical and methodological shortcomings. However, as this dissertation will show, some of these theoretical and methodological challenges and shortcomings can be resolved in satisfactory ways. My contention is that improved research on popular support for democracy will benefit, to a considerable extent, our understanding of contemporary cases of democratization. Regarding the maintenance of 22 Carrin (2006) provide a very recent and comprehensive work about the na ture and legacy of Fujimoris regimes. The Fujimoris regime is classified as one of the cases of a new and contemporary breed of authoritarian regimes (i.e., electoral authoritarianism). 23 The UNDPs report of democracy in Latin America (2004) constitutes the best methodology that I have found on how to approach and measure popular support for democracy in the context of new democracies.

PAGE 32

22 these new democratic experiences and the possibility of a gradual process of further democratization, in particular, the research about popular support for democracy is providing several critical insights. Overall, it is evident that it may be very difficult to change the current lack of popular support for key democratic rules and procedures among several new democracies, if these citizens do not experiment and perceive a minimal improvement in the quality of their democratic regimes. However, the good news is that, in the context of several new democracies, there is evidence that an improvement in the quality of political representation, political accountability, and the rule of law can be as effective as an improvement of peoples socio-economic conditions.

PAGE 33

23 CHAPTER 2 NEW DEMOCRACIES, DEMOCRATIC CONSOLIDATION, AND POPULAR SUPPORT FOR DEMOCRACY This chapter addresses the experience of third-wave democracies once they have completed a transition to democracy. The first section of this chapter focuses on the literature most relevant to recent democratization processes, and democratic consolidation.1 The second section addresses the issue of popular support for democracy in these new democracies. In that regard, I contend that the literature on democratic consolidation is correct in its affirmation that the consolidation or institutionalization of these new democratic regimes it impossible without an important attitudinal and behavioral changeboth at the elite and ordinary citizens levelregarding democracy, current democratic regime, and main democratic procedures, norms, and institutions. However, I also discuss why it is problematic to assume that these pro-democratic attitudes and behaviors are the final outcome of a democratic consolidation process rather than addressing the main challenges for its gradual development. Finally, I argue for the need of a more systematic and comprehensive study of popular support for democracy within new democracies. New Democracies: Beyond Democratic Transition Beginning in the mid-1970s in Southern Europe (Greece, Portugal, and Spain), and then spreading to Latin America, East Asia and Africa, a large number of formerly authoritarian regimes began to shift to democracy. Also, during the period 1989, 1 It is worth noting here that some scholars are particular skeptical a bout the possibility of a democratic consolidation process in several new democracies.

PAGE 34

24 with the collapse of the Soviet Union and other communist regimes in Eastern Europe, a large number of formerly communist countries began transitions to democracy as well. This global expansion of democracyknown as the third-wave of democratization (Huntington 1991)made a strong impact on the study of democracy and democratization, particularly in comparative perspective. Accordingly, several scholars have been dealing with two main topics: (1) to explain why, when, and where these processes of democratic transition happened2 and (2) to assess the future prospects of these new democracies, particularly in the context of developing countries and postcommunist countries (Landman 2000). More specifically, within this second main issue, several scholars have been interested in addressing the prospects and challenges regarding the consolidation of these new democracies. Accordingly, this group of scholars has developed what is known as the literature on democratic consolidation. This section deals with this literature on democratic consolidation,3 assessing its main contributions and problems regarding the understanding of the democratization processes of the new members of the democratic world (Bunce 2000). The literature on democratic consolidation shows an interesting mix of critical contributions and important limitations regarding the study of new democracies. Regarding its main limitations, it is worth noting that the study of democratic consolidation processes is particularly 2 This is what constitutes the literature on democra tic transitions. As Landman (2000) points out, this literature has either focused on the role of elites and the nature of pacts that are formed between them, or has focused on the role of members of civil society and the ways they struggle for democracy. See for example, ODonnell and Schmitter (1986), Levine ( 1988), Diamond, Linz, and Li pset (1989), Di Palma (1990), Karl (1990), Huntington (1991), Przeworski (1991), Munck (1994), and Linz and Stepan (1996). 3 I am using here a broad defin ition of the literature on democratic consolidation because I am considering both scholars that have directly addre ssed the issue of democratic consolidation of new democraciese.g., ODonnell (1992), Valenzuela ( 1992), Mainwaring (1992), Linz and Stepan (1996), ODonnell (1996), Schedler (1998), Diamond (1999), Encarnacin (2000)and scholars that have been particularly skeptical about its current prospects of consolidatione.g., Carothers (2002), Schedler (1998), and ODonnell (1996).

PAGE 35

25 challenging not only because they are very complex, but also because we do not have yet clear empirical thresholds regarding the different stages of a process of democratic consolidation. Main Contributions of the Literature on Democratic Consolidation It is clear that the literature on democratic consolidation has been particularly important in establishing a set of relevant questions about new democracies that go beyond its mere processes of transition. Specifically, this literature has established that the consolidation of democracy is an analytical and empirical differentiated aspect of any process of democratization. Accordingly, the literature on democratic consolidation has been concerned about the main challenges regarding the development of democratic regimes once new democratic governments have been established. Furthermore, several scholars have proposed the idea that a process of democratization implies two transitions (ODonnell 1992, Valenzuela 1992, and Mainwaring 1992, Diamond 1999). The first transition is from the previous authoritarian regime to the installation of a democratic government. The second transition is from this democratic government to the consolidation of democracy or, to the specific development of a stable, functioning, and institutionalized democratic regime. The distinction between a democratic government and a democratic regime is critical. As Mainwaring (1992:296) points out there is huge difference between the installation of a democratic government, and the development of a democratic regime. The former basically denotes a government elected by the people. The later, however, is a broader concept that refers to the rules that control the interaction of the major actors in the system. Therefore, regime involves a process of institutionalization, i.e., the idea that such rules are widely understood and accepted, and that political actors pattern their

PAGE 36

26 behaviors accordingly. Any successful democratic transition from an authoritarian rule will consequently necessitate the installation of a democratic government. However, the installation of a democratic government does not guarantee the development of a democratic regime. Focusing on the second transition, the literature on democratic consolidation has sought to address why, in the context of many new democracies, the consolidation of democracy or the development of an institutionalized democratic regime has been particularly difficult. In other words, why do these new democracies often become diminished, incomplete, or illiberal democracies (ODonnell 1994, Collier and Levitsky1997; Diamond 1999; Zakaria 1997, 2003) or, hybrid or semi-democratic regimes (Hartlyn 2002, Carothers 2002, Schedler 2002, Levitsky and Way 2002, Diamond 2002). Several factors have been mentioned as the main obstacles for the development, institutionalization, or consolidation of thes e new democracies. Among the ones that have been more frequently considered, we have: (1) the persistence of authoritarian actors who control important resources of power, (2) the prevalence of many social spheres of profoundly authoritarian patterns of domination, (3) an attitude of neutrality and indifference among elites and ordinary citizens about the new political regimes in place, (4) the lack of strong and stable political institutions (e.g., political executives, congresses, and political parties), and (5) the impact of economic crises and rampant levels of social and economic inequality (ODonnell 1992, Diamond 1999, Encarnacin 2003/2004, Garretn 2003).

PAGE 37

27 It is in relation to the previous point that it is possible to identify one critical contribution of the literature on democratic consolidation. As Encarnacin (2000) correctly argues, the literature on democratic consolidation has incited a debate that has helped to overcome some of the limitations in the study of democratization. It is clear that the literature on democratic transition responded to the abundance of structuralist approaches accounting for the failure of democracy in developing countries4 by placing political elites and their choices at the center of the analysis. Democratization was thus seen as the product of strategic calculations made by political elites. Elite pacts and other forms of consensus-driven politics were perceived as the factors most conducive to the development of democracy in developing countries.5 To a considerable extent, the literature on democratic consolidation challenges the previous dominant view about the understanding of democratization processes (i.e., democratization as an elite-crafting and voluntaristic phenomenon). Scholars working within the democratic consolidation literature have been more open to consider the role of structural fact orsranging from socio-ec onomic structures to institutional settingsin constraining and shaping both elite and ordinary citizens political attitudes and behaviors. In doing so, some scholars have depicted the process of democratic consolidation as less dependent on elite crafting, and more determined by preexisting conditions and historical legacies often linked to the nature of these societies and the preceding authoritarian regime (Encarnacin 2000). Consequently, the recent literature on democratic consolidation has provided a unique opportunity to overcome the 4 See for example ODonnell (1973). 5 See for example ODonnell and Schmitter (1986) and Przeworski (1991).

PAGE 38

28 dichotomy between structural-oriented and process-oriented approaches that have characterized the study of democracy and democratization (Kitschelt 1992). Regarding this more complex approach toward democracy and democratization, and as part of the effort to define the meaning of consolidation, the literature on democratic consolidation has emphasized the centrality of support for democracy among elite and ordinary citizens. Specifically, several scholars (Gunther, Diamandouros and Puhle 1995; Linz and Stepan 1996; Diamond 1999) have affirmed that any process of democratic consolidation or institutionalization is impossible without both elites and ordinary citizens supporting it. Such support has been defined in terms of key attitudinal (i.e., both elites and ordinary citizens believing that a democratic regime is the most acceptable and convenient kind of political regime) and behavioral changes (i.e., all politically significant group respecting and adhering to democratic norms and procedures). I revisit this argument extensively in the second section of this chapter. Similarly, the literature on democratic consolidation has been significant because it has generated an important debate regarding different understandings of democracy and democratic regimes. This literature has been characterized by an intense debate about the advantages and disadvantages of assuming either a minimalist or procedural conception of democracy6 or a more maximalist or substantive one. For those scholars defending the use of a procedural or minimalist conception, democracy means Dahls (1971) definition of polyarchy: (1) political contestation or competitive elections (assuming that these 6 It is important to mention here that I am not distinguishing between a minimalist and a procedural definition of democracy because I believe that there is an important overlap between them. However, it is possible to contend that Dahls procedural definiti on of democracy departs from the minimalist definition provided by Schumpeter (1942) i.e., democracy is a political system in which the principal positions of power are filled through a competitiv e struggle for the peoples vote because he contends the need for civil and political freedoms to render political participation and competition meaningful.

PAGE 39

29 elections are free and fair), (2) citizens right to participative in public contestation, and (3) a set of civil rights (e.g., freedom of the press, freedom of speech, freedom to join organizations, etc.) to guarantee that competitive elections and political participation will politically effective and meaningful. This concept of democracy sharply contrasts with more substantive definitions of democracy that link democracy with some specific outcomes such as economic equality, the satisfaction of peoples needs, and the achievement of higher levels of direct democracy. The main argument favoring a minimalist conception of democracy contends that thinking about democracy in more substantive, maximalist or outcome-oriented terms is problematic in third-wave democracies. It assumes that democracy should provide outcomes like economic equality and redistribution. As a consequence of this, it could generate a situation in which ordinary citizens become impatient or ambivalent toward the key components of a procedural definition of democracy; i.e., a set of procedures and norms that guarantees a process of negotiation and dialogue to define what goals are chosen and pursued.7 Moreover, the argument in favor of a minimalist conception of democracy also claims that even the most established democracies in the world cannot be defined as substantive or maximalist forms of democracy because democracy is only a form of political regime thatin the context of some constitutional parameters and state protection of civil liberties and human rightsgives citizens more opportunities to 7 See Anderson (2005) for an interesting account of the negative consequences of rejecting a minimalist or procedural conception of democracy in dealing with the study of democratization in Latin America. According to her, dismissing the importance of a minimalist or procedural definition of democracy not only leads to pessimistic view about the current situation of democracy in Latin America, but also to the possibility of creating unrealistic popular expectations toward these demo cratic regimes that eventually could endanger their persistence.

PAGE 40

30 choose and influence governments decisions. Therefore, the specific content of governments decisions and policies is up to the citizens and their leaders. Finally, if established democracies represent the most advanced cases of democratization and democratic quality, this has been the outcome of a long-term experience with a procedural democracy (Fuchs 1999, Dalton 2004, Anderson 2005). However, it is important to acknowledge that there have been some important criticisms regarding the adoption of a strict minimalist or procedural definition of democracy in the context of new democracies. Accordingly, it has been affirmed that it is possible to find several cases where the presence of all the components of a procedural or minimalist democracyat least in the way defined by Dahlis not enough to nurture a democratic regime. For example, Karl (1990) has argued that in several Latin American countries, the presence of competitive elections, broad adult citizenship, and civil rights does not guarantee that the civilian authorities will have the power to control tutelary powers such as the military. As a consequence of this, popularly elected authorities confront severe policymaking constraints. Additionally, Diamond (1999) has pointed out that in many of the worlds new democracies competitive elections have not ensured a minimal level of political responsiveness. A similar argument has been made by ODonnell (1996) when he mentions that several political regimes in Latin America have adopted all procedures and norms contained in Dahls definition of polyarchy. However, the problem is that along with these formal democratic procedures and norms, these political regimes maintain several informal rules (e.g., particularism, clientelism, neopatrimonialism, etc.) that are antagonistic to the full institutional packag e of a polyarchy. Interestingly, ODonnell does

PAGE 41

31 not reject the convenience of developing pro cedural democracies within Latin American or other new democracies countries; on the contrary, he is trying to understand why these formal rules and procedures have a limited capacity in constraining and shaping the political behavior of relevant political actors. As a reaction to this type of criticism, it is common to find procedural or minimalist definitions of democracy that have added other components or requirements to Dahls definition of polyarchy. For example, it has been stated that in order to have the minimal conditions of democracy, elected authorities must be able to govern without being subject to military or clerical leaders (Diamond 1999, Levitsky 2002). Something similar is happening in regards to the rule of law because it is clear that the political significance of civilian and political rights will be very weak if they are not effectively protected and enforced by a rule of law (ODonnell 2004a, 2004b). It is worth noting here that this new trend does not imply the adoption of a substantive or maximalist definition of democracy. It seems more an effort to understand the limited capacity of democratic norms and procedures, in several new democracies, in making possible effective political representation and inclusion. Accordingly, some scholars have recently challenged the assumption that there is a fundamental tension between a procedural definition and a more substantive definition of democracy. ODonnell (2004c) contends that this tension is somewhat artificial because what really matters in the context of new democracies is improving the performance and democratic quality of these democratic regimes. Following Dahls contention (1971) that a democracy or a polyarchy needs a set of individual and political rights to effectively formulate and signify citizens preferences (e.g., the freedom to form and join

PAGE 42

32 associations, alternative sources of information, etc.), ODonnell discusses the impact of the political and social context on the functioning of the political regime. He goes on to discuss the extent to which the social and political context enable or disable the capacity of citizens to exert their political rights and citizenship. Furthermore, ODonnell insists on the contradiction of recognizing equal and universal political rights when citizens lack fundamental civil rights (e.g., fair and expeditious access to the legal system). He also discusses how severe economic deprivations hamper the capacity of an individual to be a political agent.8 Addressing the situation of several Latin American countries, he shows how the existence of a biased and intermittent rule of law makes impossible the generalization of the most basic civil rights for all citizens. Similarly, the combination of huge levels of poverty and inequality in these societies alongside a limited implementation of social rights definitively disables the political capacities of larger segments of the population. Therefore, several democratic regimes in Latin America are characterized by a low quality in terms of the two basic components of democracypolitical representation and political inclusion. Nevertheless, instead of claiming that the problem with these new democracies is its formal or minimalist nature, ODonnell suggests that the most effective strategy is to place more emphasis on the development and extension of civil rights because they not only protect, but also empower citizens givi ng them a better framework to define and defend their political interests, and exert all their political rights. This would be particularly relevant for most disadvantaged citizens and their struggle for the generation 8 According to ODonnell (2004c) the definition of an i ndividual as an agent implies the recognition of his or her capacity to make reasonable choices.

PAGE 43

33 of policies against poverty and inequality, and the recognition of some key social rights by their political regimes. Overall, ODonnells arguments are relevant for several reasons. He suggests a critical connection among political, civil, and social rights. He argues that the quality of the democratic regime will largely depend on the connections between political, civil, and social rights. By the same token, he is not suggesting that the political or procedural rights are irrelevant, but rather to understand why some social and political context features block their potential. In this regard, democracy can be considered as a valuable and necessary experience because it assumes that all individuals have the intrinsic capacity and the right to make reasonable choices, and because it provides a framework where a gradual and subsequent improvement in civil, political, and social right is possible. Finally, he also suggests that advancements in social rights will not be easily reversed if they can be exercised throughout the formal democratic procedures. All these contentions provide important insights for our understanding of democracy and democratization processes. First, democracy does not need to be defined either entirely in procedural or political terms, or entirely in substantive terms. Second, as some scholars have considered, the critical issue is to understand why in the context of several new democracies, democratic regimes have a very low quality in terms of political representation and political inclusion. The main challenge is then to improve the quality of these democratic regimes. Accordingly, it becomes important to recognize the relevance of democratic norms and procedures; but also to address how the political, social and economic contexts diminish the potential of these democratic norms and procedures.

PAGE 44

34 It is important to note here that in the last few years, more scholars have started to address specifically the quality of the new democratic regimes. So far, we know that some key factors have the capacity to make a negative impact on the quality of new democratic regimes. Among these factors we have the lack of horizontal accountability, the lack of a strong and consistent rule of law, the weaknesses of some key political institutions such as political parties, and extreme levels of economic and social inequality (ODonnell 2004a, 2004b, 2004c; Beetham 2004; Schmitter 2004; Rueschemeyer 2004; Powell 2004). Similarly, as Diamond (2004:21) points out, we know that all these different factors not only overlap, but also depend upon one another, forming a system in which improvement along one of these democratic dimensions (e.g., participation) can have beneficial effects along others (e.g., e quality and accountability). However, we also know that very frequently it is going to be very difficult to improve all these democratic dimensions at once. Finally, another important contribution of the literature on democratic consolidation has been the debate about what the experiences of these new democracies are revealing in terms of its possibilities of further democratization. The consolidation of the new democracies is doubtlessly a desirable out come, but it is clear that the majority of these new democratic regimes do not necessarily advance in that direction. Furthermore, it is becoming increasingly clear that the mere survival of formal democratic regimes does not necessarily indicate a transition toward democratic consolidation. Consequently, some scholars have called attention to a potential bias in the study of new democracies (Carothers 2002, Levitsky and Way 2002). Carothers (2002), for example, has pointed out the limitations of the democratic consolidation literature that

PAGE 45

35 was attached to what he defined as the transition paradigm. According to him, this paradigm was based on a set of assumptions that were called into doubt by the political trajectories of most third-wave countries. The key assumptions of this transition paradigm were the following: (1) that any country moving away from dictatorial rule can be considered a country in transition to democracy; (2) that democratization tends to unfold in a set of stages: opening, breakthrough, and consolidation; (3) high expectations for what the establishment of regular, genuine elections will do for democratization; (4) that a countrys chances for successful democratization primarily depend on the political intentions and actions of its political elite s without significant influence from underlying economic, social, and institutional conditions and legacies; and (5) that state-building is a secondary challenge to democracy building and largely compatible with it. The main problem, according to Carothers, is that all these assumptions have been proven wrong. For example, the assumptions that any country moving away from authoritarian rule was in transition to democracy and that there will follow a sequence of opening, breakthrough, and consolidation have often been inaccurate and misleading because some of these countries hardly democratized at all. Moreover, most common pattern has been the development of stable feckless pluralism regimes that are not in transition to be liberal democracies. There are also problems with the assumption that regular, genuine elections will not only give government democratic legitimacy but also foster a greater deepening of democratic participation and accountability. Regular and genuine elections are often held but they do not guarantee the development or renovation of political parties, and an important level political accountability. To add to the complication, elections do not change long traditions of personalistic leadership, and

PAGE 46

36 stagnant patronage-based politics. Finally, it has become clear that various structural conditions (e.g., the level of economic development, past experiences with political pluralism, the lack of a coherent state, etc.) heavily shaped the political outcomes of postauthoritarian regimes. Regarding the shortcomings of this transition paradigm, it is worth noting that they should not be seen as evidence that third-wave democracies are doomed to never become functioning liberal democracies. Rather, they should help us to assume that the middle ground between full-fledged democracy and outright dictatorship is actually todays most common political outcome of countries in the developing and the post-communist world (Carothers 2002:18). As expected some scholars have reacted negatively to Carothers criticisms. ODonnell (2002) affirms that Carothers lumps together a large and uneven body of work under the heading of transition paradigm, and then proceeds to concentrate his criticisms on some of the weakest arts of it. Moreover, ODonnell presents several examples showing that some of the Carothers main criticisms were already acknowledged in the literature about third-wave democracies. For example, he mentions that ODonnell and Schmitter (1986) in one of the first works about democratic transitions were explicit in affirming that these transitions do not necessarily lead to democracy. Rather, they may instead lead to authoritarian regressions, to revolutions, or to hybrid regimes. Regarding the role of structural factors in shaping the political outcomes of thirdwave democracies ODonnell cites the work by Przeworski et al. (2000) that contends there is a connection between a countrys level of socio-economic development and the democracys durability. Indeed, it is not hard to find evidence that this was not an entirely

PAGE 47

37 new topic within the literature on democratic consolidation. Besides Przeworski et al. (2000), other scholars have been disappointed that with few exceptions the literature on democratic consolidation has emphasizedas the literature on democratic transition didthe role of political processes in affecting regime change and asserted the autonomy of these processes from structural fact ors (Remmer 1996, Gasiorowski and Power 1998). According to Remmer (1996:630) one of the main problems with this literature is that comparativists have all but abandoned efforts to generalize about the macro-social prerequisites for political democracy in favor or more contingent understandings emphasizing the strategic choices of political actors. Without exploring how structural factors may affect the development of democracy within new democracies, this literature has neglected the possibility of working in a more comprehensive understanding of the determinants of democratic consolidation that must take into account political processes and structural factors. Therefore, it seems reasonable to contend that Carothers main contribution has not been the novelty of his propositions, but rather the effort of summarizing several ideas that were already part of the literature on democratic consolidation. In doing that, he has encouraged an interesting debate about what we have learned regarding the recent processes of democratization and the development of democratic regimes in developing countries. By the same token, this exchange between Carothers and some scholars dealing with the issue of democratic consolidation in new democracies has been particularly useful for the development of a new research agenda about these recent experiences of democratization and that is not constrained by the transition paradigm.

PAGE 48

38 Main Limitations and Problems of the Literature on Democratic Consolidation The literature on democratic consolidation has not been exempt from important limitations and problems. On a more general level, ODonnell (1996) has argued that democracy and consolidation are terms too polysemic to make a good pair. Schedler (1998) has also contended that one of the ma in problems with the literature on democratic consolidation was the multiple usage and meanings of the term democratic consolidation. According to him, the original meaning of democratic consolidation (i.e., the challenge of making new democracies secure, of extending their life expectancy beyond the short term, of building dams against eventual reverse waves) was expanded in so many directions that there is now a conceptual disorder. As a consequence of this trend, the list of the problems contained in the democratic consolidation literature includes such divergent items as popular legitimation, the diffusion of democratic values, the neutralization of anti-system actors, civilian supremacy over the military, the elimination of authoritarian enclaves, party building, the organization of functional interests, the stabiliza tion of electoral rules, the routinization of politics, the decentralization of state power, the introduction of mechanisms of direct democracy, judicial reform, the alleviation of poverty, and economic stabilization. Moreover, Schedler claims that there are at least five very different definitions of democratic consolidation. According to him, democratic consolidation can mean: (1) avoiding democratic breakdown; (2) avoiding democratic erosion; (3) completing democracy; (4); deepening democracy, or (5) organizing democracy. Interestingly, and very frequently, the particular meaning of democratic consolidation that is embraced will depend on where scholars stand in empirical terms (i.e., what particular cases among new democracies they are considering) and what they are thinking about what is possible or

PAGE 49

39 desirable to aim for these particular cases (i.e., the role of different definitions of democracy). For example, if an electoral de mocracy or a group of electoral democracies are facing the possibility of returning to an authoritarian regime, democratic consolidation in this context will tend to mean avoiding a democratic breakdown. On the contrary, if an electoral democracy has the possibility of moving forward toward becoming a liberal democracy, democratic consolidation in that context will tend to be defined as a process of completing democracy. To a large extent, the existence of such different definitions of democratic consolidation is consistent with the high level of diversity that we can observe within the group of new democracies. Additionally, there is no doubt that the existence of different levels of advancement in terms of the democratization of these political regimes is one of the most important sources for these empirical differences. In that regard, some of them seem to move forward, some of them seem stuck in a gray zone between democracy and authoritarianism, and some of themif they have not already suffered a democratic breakdownseem to be in risk of moving backwards. Therefore, when scholars are dealing with issues of democratic consolidation, they are not necessarily sharing the same concerns and the same questions. Another problem with the literature on democratic consolidation is that it is characterized by an uneven development. The literature has been extensive and systematic regarding the shortcomings of new democracies vis--vis established democracies, and regarding which should be considered the final outcome of democratic consolidation, but less so in accounting for the particular features of new democracies and the particularities of its processes of democratization. In addressing this shortcoming,

PAGE 50

40 Encarnacin (2000) contends that very frequently the literature on democratic consolidation has failed to explain the internal dynamics of democratization and consolidation. Specifically, regarding the shortcomings of new democracies, scholars working on third-wave democracies soon realized that they needed more than the distinction between authoritarian and democratic regimes because in so many cases these post-authoritarian and formally democratic regimes were lying somewhere between democracy and authoritarianism. The problem has been that the majority of cases have been solved by labeling new regimes as a diminished subtype of democracy (Collier and Levitsky 1997). A diminished type of democracy is a political regime that either lacks one or more attributes that we find in the most consolidated cases of democracy, or that has such attributes in a diminished way (e.g., popular elections that are not free and fair). A diminished type of democracy tends to be in gray zone area because it combines democratic and non-democratic features; and it is the case that the majority of new democracies can be classified as gray zone democracies (ODonnell 1996). The main problem with this strategy, as ODonnell (1996) points out, is that these negative definitions address these new democracies mainly according to some missing attributes, assuming that there is only one type of democracythe one found in most developed countries. According to ODonnell, it is clear that this strategy represent an ethnocentric and teleological approach unabl e to account for the different types of political regimes that exist among third-wave democracies.9 Other scholars have been 9 Franco (1998) presents a similar argument when he contends that the definiti on of political regimes in developing countries as "diminished" or "anomalous" subtypes of demo cracy must be avoided because it means that we are giving up regarding the understanding of the concrete and typical features of these political regimes.

PAGE 51

41 even more critical about the extensive use of diminished types of democracies. For example, Collier and Levitsky (1997) have asked whether these diminished democracies should in fact be treated as subtypes of democracies rather than subtypes of authoritarian regimes. However, it is fair to say that not all these formulations of a diminished type of democracy among new democracies are equally problematic. Some of them have meant an important contribution to understanding new democracies either because they were related to one specific region or because they were devoted to address one specific aspect of these new democratic regimes. For example, the concept of delegative democracies proposed by ODonnell (1994) focused on the level of horizontal accountability among new democracies in Latin America. In doing so, he provided not only a good characterization of some of the new democratic regimes that can find in this region (i.e., Argentina, Brazil, Peru, etc.), but also an interesting discussion regarding some challenges for its institutionalization. In dealing with this explosion of diminished types of democracies, some scholars have been particularly clear about the need of identifying particular types of political regimes between liberal democracies and closed authoritarian regimes (Diamond 2002, Levitsky and Way 2002, Schedler 2002 and van de Walle 2002). For example, all these aforementioned scholars think that it is possible to make a distinction between electoral democracies and competitive authoritarianism within new democracies. These different types of political regimes will present some democratic elements, but the extent to which these democratic elements pattern political competition varies considerably.

PAGE 52

42 Electoral democracies, as liberal democratic regimes, meet the minimum requirements of democracy, but fail to institutionalize vital dimensions of democracy such as the rule of law and political accountability (Schedler 2002). Indeed, several Latin American countries can be classified as electoral democracies. On the contrary, competitive authoritarianism is a political regime where formal democratic institutions are viewed as the principal means of obtaining and exercising political authority, but incumbents violate those rules so often and severely that the regime fails to meet the minimum standards for democracy. Specifically, incumbents routinely manipulate formal democratic rules but they are unable to eliminate them or reduce their potency. Rather than openly violating democratic rules (e.g., by banning or repressing the opposition and the media), incumbents are more likely to use bribery, co-optation, and more subtle forms of persecution, such as the use of tax authorities, complaint judiciaries, and other state agencies to legally harass or extort cooperative behavior from critics. However, even when the cards are stacked in favor of autocratic incumbents, the persistence of democratic institutions creates arenas through which opposition forces may pose significant challenges (Levitsky and Way 2002). There is no doubt that all these theoretical efforts are important because they improve our understanding and classification of new democracies, and make possible to have a better idea, depending of the circumstances, about the main challenges and prospects of subsequent processes of democra tization. Nevertheless, it is also clear that we are still far from reaching an important level of consensus regarding how to classify the diverse group of new democracies. Equally im portantly, we still need a great extent of theoretical and empirical work to improve our understanding of the particular nature of

PAGE 53

43 these political regimes, as well as our understanding of its democratic advancements and setbacks. New Democracies and Popular Support for Democracy As mentioned before, after the massive installations of democratic regimes in developing countries and former communist countries during the 1980s and 1990s, it is evident that only a relatively small number have made a significant progress in terms of further democratization, and are still enjoying a positive dynamic of incremental democratization. On the contrary, most of these countries experimenting with democracy are neither dictatorial nor clearly headed toward democracy (Diamond 2002, Levitsky and Way 2002, Schedler 2002, van de Walle 2002). They have some attributes of democratic political life, such as limited political space for opposition parties, an independent civil society, regular elections, and democratic constitutions. At the same time, they suffer from serious democratic deficits, including poor representation of citizens interests, low levels of political participation beyond voting, frequent abuse of the law by government officials, elections of uncertain legitimacy, very low levels of public confidence in state institutions, and persistently poor representation by the state (Carothers 2002:9). In general, it is no only the case that most of these new democracies lack one or some aspects of democratic rule, but that this overall democratic quality is very low. As a direct consequence of this democratic deficit trend, a large majority of these troubled new democracies suffer the lack of a significant and stable level of popular support for democracy (Diamond 1999, Moreno 2001). In some cases, this situation has

PAGE 54

44 led to a clear authoritarian reversal, e.g., Peru in 199210 (Carrin 2006) and to the development of stable political regimes that can be best described as electoral authoritarianism11 (Schedler 2006). Also, in other cases the lack of support for democratic procedures and institutions has led to a high level of political instability caused by very frequent non-constitutional depositions of democratically elected presidents, e.g., Ecuador in 2000 and 2005 and Bolivia in 2003 (Mainwaring and Hagopian 2006). In turn, it is clear that one of the most negative consequences of authoritarian reversals and processes of democratic erosion is that they make it difficult for these recent experiences with democracy to persist, which is absolutely necessary for the development of a democratic learning process and the possibility of reaching a dynamic characterized by incremental democratization.12 For example, it seems unreasonable to expect improvements regarding the current democratic deficits in several new democracies if some political actors, relying on the indifference or lack of support for democracy among ordinary citizens, can constantly manipulate and eventually suspend the main democratic procedures. Consequently, it is reasonable to contend that there is an important connection between the lack of support for democracy or democratic legitimacyamong elites and 10 Chapter 5 provides a detailed descri ption and interpretation of this r ecent Peruvian political experience. 11 According to Schedler (2006:3), elect oral authoritarian regimes play th e game of multiparty elections by holding regular elections for the chie f executive and a national legislative assembly. However, they violate the liberal-democratic principles of freedom and fa irness so profoundly and systematically as to render elections instruments of authoritarian rule rather than instruments of democratic rule. Contemporary political regimes in Russia, Algeria, Egypt, Ca meroon, Chad, Ethiopia, Cambodia, Malaysia and Singapore, among many other cases, can be considered as clear examples of electoral authoritarianism. 12 See Anderson and Dodd (2005) for how they defined a democratic learning process in the case of contemporary Nicaragua.

PAGE 55

45 ordinary citizensin many developing countries and the lack of democratic stability and advancement of its new democratic regimes. However, the problem is that for a long period of time, thinking carefully about the development of popular democratic legitimacy was not one of the most important issues in the literature on democratic consolidation. In this regard, several scholars (ODonnell 1992, Valenzuela 1992) dismissed the importance of having a majority of democrats among the population as a key condition for the transition toward democratic rule and the development of an institutionalized democratic regime. For example, ODonnell (1992:19) stated that there are no cases in the world where there has been a majority of democrats before the advent of political democracy. Moreover, according to him, the development of this majority of democrats will be only possible throughout a continuous practice of political democracy.13 Nevertheless, the main problem with this contention is that the experience of several new democracies is showing that a continuous practice of democracy is almost impossible to achieve without some considerable and stable level of democratic legitimacy among ordinary citizens.14 More specifically, as Rose, Mishler, and Haerpfer (1998) affirm, it is hard to think that political elitesthe supply side in a political systemwill have any incentives to play according the main democratic norms and procedures if there is not a minimal demand in that direction from ordinary citizensthe demand side of a political system. 13 A similar argument can be found in Rustow (1970) when he affirms that democratic institutions become stable and acquire legitimacy in pa rt through the passage of time, main ly because political actors become habituated to democratic practices. 14 In the context of Latin America, and besides Peru Venezuela represents another case where a limited and deficient representative democracy has been replaced by a hybrid regime that in several aspects does not represent a democratic advancement (McCoy and Myers 2004).

PAGE 56

46 By the same token, in dealing with the relevance of popular support for the stability and eventually consolidation of new democracies, some scholars have mentioned that universal democratic legitimacy is not necessary in the beginning stages of democratization. However, if commitment to democracy does not emerge over time, and common citizens do not develop some considerable level of democratic commitment, democracy will likely fail because these new democratic regimes will be vulnerable to disloyal leaders and groups (Mainwaring 1992:307) Therefore, the argument in favor of the importance of democratic legitimacy is not claiming that a strong popular commitment toward democracy is a pre-requisite for democratic consolidation. On the contrary, the idea is that democratic consolidation will be almost impossible without a gradual development of popular support for democracy. Accordingly, democratic consolidation has been clearly defined as a process that implies fundamental attitudinal and behavioral changes among elites and ordinary citizens. Specifically, it has been argued that a successful process of democratic consolidation implies that the majority of public opinion, even in the midst of major economic problems and deep dissatisfaction with incumbents, should hold the belief that democratic procedures and institutions are the most appropriate way to govern collective life (Linz and Stepan 1996). In a very similar way, according to Gunther, Diamandouros, and Puhle (1995), a democratic regime can be considered as consolidated when all politically significant groups, in both attitudinal and behavioral dimensions, adhere to the democratic rules of the game and regard the main political institutions as the only legitimate framework for political contest.

PAGE 57

47 Unfortunately, the recognition of the importance of democratic legitimacy particularly at the level of ordinary or ordinary citizens and as a sign of democratic consolidationhas not been accompanied by a more comprehensive and systematic theoretical and empirical consideration of this topic. In that regard, very frequently in the last years, the topic of popular support for democracy has been considered as one among several factors that has the capacity either to facilitate or hinder the development of democracy. Moreover, little attention has been devoted to its interaction with the rest of these critical factors (Diamond, Linz, and Lipset 1989).15 This situation is problematic not only because in order to develop theories about democracy and democratization processes we have to move beyond general lists of factors that matter in almost all the cases toward more systematic statements about what factors the matter most in particular cases and situations (Geddes 1999); but also because it seems reasonable to expect that the possibilities of consolidation or institutionalization of new democratic regimesespecially within developing countrieswill depend, to a large extent, on the current and future levels of popular support for democracy, democratic regimes, and democratic procedures and institutions. Hence, it is the case that our understanding of popular support for democracy in new democracies is still limited. Among other things, we are still struggling with how to 15 Along with the need of a democratic political culture, these long lists te nd to include all types of factors that can influence the development of democracy in developing countries (e.g., the previous authoritarian regimes, the effect of the international context, the ex istence of an effective stat e, the existence or not the representative and strong political institutions, th e economic and political performance of the new democratic governments, the level of institutionali zation of the party system, the quality of political leadership, the impact of the level of socio-economic development, the existence of a pluralistic and autonomous civil society, etc.).

PAGE 58

48 operationalize and measure this complex political phenomenon,16 and we are not clear about what can be considered the main determinants of this type of political support.17 In trying to change this situation, it is critical to pay close attention to some recent works that have been dealing, in direct or indi rect way, with the topic of popular support for democracy. One of the main strengths of these recent efforts is that they go beyond what can be considered as the traditional political culture approach. This approach basically contends that peoples attitudes and orientations toward democracy are the product of depthpolitical values (e.g., interpersonal trust, political tolerance, etc.) rather than the product of short-term evaluations of government and regime performances, or the product of individuals particular social and economic conditions (e.g., class or socio-economic levels). Specifically, some scholars have pointed out that there are key differences in the way ordinary citizens in developing countries perceive and define democracy vis--vis ordinary citizens of the most developed countries. In the case of developing countries, and particularly for those who lack the means to satisfy their basic needs, democracy is not only associated with political components (i.e., rule of law, civil liberties, competitive elections, etc.), but also with other components, such as welfare and the dispersion of economic resources (Camp 2001). 16 In the context of new democracies, there has been some advancement in the case of post-communist countries, but still there are so many controversial i ssues. See Chapter 4 for a discussion of how the issue has been addressed in established democracies post-communist and Latin American countries. 17 Unfortunately, scholars frequently dealing with this topic tend to emphasize only one main theoretical perspective (e.g., a political culture pe rspective, structural explanations or the importance of the economic, social, and political performance of these new democratic regimes) rather than addressing the complexity of popular support for democracy in new democracies.

PAGE 59

49 The main idea behind this contention is that there are high expectations toward democracy within developing countries, and that these expectations include both political and economic aspects. Therefore, the level of democratic legitimacy in these countries will depend not only on the new democratic regimes ability to provide a set of rules and procedures that make possible authentic political competition and participation (i.e., a popular election of government), but also in its ability to provide effective inclusion and representation of the needs and aspirations of the most disadvantaged citizens (Roberts 1998, Diamond 1999, Panizza 2000). Therefore, it seems that the poor performa nce of democratic governments in several developing countries is one of the main reas ons for the lack of popular support for current democratic regimes and one of the main sources of popular alienation regarding democratic institutions and democratic procedures (Mainwaring and Hagopian 2006, Hagopian 2006). Furthermore, these very negative assessments are not only related to the economic performance of democratic governments and regimes but also to their performance in terms of the quality of the rule of law and the precarious situation of civil or individual rights in these countries (F oweraker and Krznaric 2002; ODonnell 2004a, 2004b, 2004c). It is critical to note here that in dealing with the lack of popular support for democracy within several new democracies, it is the same people who have high economic and social expectations regarding their current democratic regimes who live under an irregular and weak rule of law that severely limited their civil and political rights. This is what ODonnell (2004b, 2004c), for the case of Latin America, has defined as a strong connection between poverty and weak rule of law. Accordingly, there is not a

PAGE 60

50 coincidence that the rule of law is particular ly weak or non-existent for the sectors of the population most marginalized socially and economically. Individuals with higher education, more access to different resources and more capacity to exert their civil and political rights will inevitably have more power to shape the states political agenda. In other words, it is possible to contend that the lack of popular support for democracy in new democracies is related to the fact that a large part of the population is suffering simultaneous social, economic, and political exclusion. As a consequence of this, scholars who claim that political equality is almost impossible in a context of huge wealth and status inequalities (Rueschemeyer 2004; PNUD 2004; ODonnell 2004b, 2004c) seem to be pointing in the right direction.18 It is also possible to be more specific about how particular social, economic and political conditions (i.e., high levels of povert y and inequality) explain the lack of popular democratic support if we consider the development of main political attitudes among the most disadvantaged citizens of several developing countries. For example, in the case of some Latin American countries, some scholars have addressed the relationship between non-democratic attitudes (i.e., clientelism and plebiscitarianism) and social and economic structural conditions (Parodi 1993, Stokes 1995, Murakami 2000). According to these researchers, poor Peruvian citizens have the tendency to develop a clientelistic or plebliscitarian patterns of interaction with governments and authorities. These practices or patterns of political interactions are based on a highly instrumental attitude toward governments and political regimes rather than by a clear support to the main democratic 18 On contrary to ODonnell (2004) who contends that th e best strategy to overcome this situation is the expansion of civil rights, Rueschemey er (2004) believes that the best stra tegy is the mitigation of economic and social inequalities, declaring that certain goods su ch as health, education and a minimal income are social rights.

PAGE 61

51 principles and procedures. These clientelistic and patrimonialistic attitudes are particularly strong among the poorest and marginalized population.19 In conclusion, it is important to develop a more systematic approach toward the topic of popular democratic legitimacy or popular support for democracy in the study of new democracies and the debate about its further democratization prospects. This is especially urgent in those cases where the installation of formal democratic regimes, following Dahls (1971) definition of polyarchyi.e., extensive competition for power through regular, free and fair elections; highly inclusive citizenship conferring rights of participation on virtually all adults; and even civil and political libertieshas not been sufficient to generate a strong commitment toward democratic principles and institutions among ordinary citizens and elites (Shin 1994, Diamond and Plattner 1996). Specifically, there are some issues that demand a closer examination. First, although it is appropriate to define democratic consolidation as a process of gradual changes in ordinary citizens and elites attitudinal and behavioral dimensions, the main remaining challenge is to understand what factors shape the political attitudes, perceptions, and beliefs of ordinary citizens. It is critical to address and understand how ordinary or ordinary citizens in new democracies approach and define politics and democracy, and assess the performance of their democratic governments. Second, it is important to recognize that one of the main difficulties for the development of popular support for democracy is the combination of a large population that as a consequence of their acute economic and social marginalization have diverse 19 Bratton and van de Walle (1997) make a similar ar gument for several new African democratic regimes. According to these authors, neopatrimonialismdefined as an informal political institution associated with the prevalence of big men syndrome and their networks of personal loyalty in government and public lifemakes it difficult to develop popular legitimacy towa rd main democratic institutions and procedures.

PAGE 62

52 and high economic and social expectations toward democracyon the one hand, and the poor performance of current democratic governments in these areason the other hand. Nevertheless, it is also critical to acknowledge that several new democracies are having also problems in delivering the minimum conditions that can make individual and political rights effective instruments to increase the level of political representation, accountability, and inclusions (e.g., an effective and unbiased rule of law). This situation is even more critical if we consider that social and economic marginalization goes hand by hand with a very precarious situation in terms of the protection of individual, civil, and political rights (Tokman and ODonnell 1998). Accordingly, it is critical to consider that political regimes are embedded in particular social and economic contexts (ODonnell 2004b, 2004c). Finally, it is important to be aware that dealing with the topic of popular support for democracy, and recognizing how some social and economic structural factors play a critical role in the development of democracy, is a good opportunity to overcome the traditional dichotomy between structural-oriented and process-oriented approaches that has characterized the study of democratization (Kitschelt 1992). Indeed, there is nothing wrong in assuming that the democratic consolidation of new democracies is a democratic crafting process (Di Palma 1990). The problem lies in assuming that structural factors and conditions, such as economic and social inequality or the quality of the rule of law, are not particularly important. This is a good opportunity to recognize that one of the main failures of the literature on democratic consolidation has been its neglect of the insights from structural approaches (Gasiorowski and Power 1998).

PAGE 63

53 Accordingly, a systematic approach toward attitudes, perceptions, and beliefs of ordinary citizens in new democracies is one of the most important steps toward the understanding of the limits and possibilities of further democratic advancement, and, eventually, consolidation of new democratic regimes. By the same token, in order to understand the lack of popular legitimacy of the current democratic regimes in these countries, it is necessary to assume the challenge of thinking more systematically not only about the several and different explanat ory variables, but also about the best alternative to operationalize and measure our outcome or dependent variables: the different key dimensions of support for democracy.

PAGE 64

54 CHAPTER 3 ADDRESSING AND EXPLAINING POPULAR SUPPORT FOR DEMOCRACY This chapter addresses one of the most common theoretical approaches used in accounting for popular support for democracy: the political culture approach. I will make a critical assessment of this theoretical approach and stress its main contributions and limitations in dealing with popular support for democracy, particularly in the context of new democracies. There is no doubt that the political culture approach will be an important component of any systematic approach toward popular support for democracy, but this endeavor also requires additional efforts to overcome some of its limitations, and to consider alternative theoretical approaches.1 One of the most common ideas among scholars dealing with democracy and democratization is that elites and ordinary citizens support for democracy2 is vital for the maintenance and development of democracy. 3 Nevertheless, as Mishler and Rose (1996, 1999) point out it is impossible to identify an absolute level of support sufficient for democracy to flourish because, among other reasons, popular support for democracy is a 1 According to Norris (1999a) there are three main theoretical approaches to explain cross-national variation in popular support for democracy: (1) the role of cultural values, (2) the performance of democratic governments, and (3) the infl uence of different institutional designs. 2 It is worth noting here that one of the main problems with the concept of support for democracy is that scholars do not necessarily agree on its particular meani ng. As a consequence of this, their definitions range from support for democratic values to support for democratic institutions. 3 Among other critical factors for de mocracy and democratizationpartic ularly in the context of new democracieswe can consider the level of socio-economic development, the severity of social cleavages, the levels of economic and political stress, the design of political institu tions, the structure and strength of the state, an active and autonomously organized civil society, the performance of democratic governments and institutions, and even some international factor s (Diamond, Linz, and Lips et 1989). Unfortunately, one of the aspects less developed in this literature is th e interrelation among these criti cal factors; and how they shape popular support for democracy.

PAGE 65

55 necessary but not sufficient condition for democratic stability.4 Therefore, the most certain thing that can be said about the relationship between support for democracy and democratic stability is that a considerable or stable level of this support facilitates the maintenance of democracy, while a declining level can undermine democracy and, eventually, cause is its collapse. A stable or upward trajectory of public support is even more important for new democracies because these new regimes not only confront higher levels of political and economic stress, but also lack the long and continued democratic experience that enables established democracies to cope with political challenges. Additionally, it is also important to highlight that democratic legitimacy is not only vital for the stability of democratic regimes, but also for its governance (i.e., the capacity to formulate and implement policies in an effective manner). The citizenrys confidence in the regime, its institutions and authorities, enhances the governments abilities to make decisions and commit resources without resorting to coercion or the need to obtain approval from the citizens for every decision (Mishler and Rose 1999). As Levi (2006) points out, the lack of governance is a recurrent problem within new democracies, as well as a source of democratic regressions. Starting during the 1960s, an extensive literature has documented the phenomenon of popular support for democracy, mainly in the context of established democracies. More recently, scholars have devoted considerable attention to this topic within new democracies. From this trend, it has become clear that there are substantial cross-national variations in support for democracy within established democracies, within new democracies and, particularly, between these two groups. Next section, addresses in more 4 The same argument can be found in Lipset (1959), Almond and Verba (1963), Dahl (1971), Linz and Stepan (1978), ODonnell and Schmitter (1986), Diamond (1994), and Linz and Stepan (1996).

PAGE 66

56 detail what I define as a mainstream political culture approach in regards to the topic of popular support for democracy. Given that popular support for democracy is, to a large extent, a political attitudinal phenomenon, it seems relevant to pay attention to this political culture approach and how it has been used to account for ordinary citizens values, and belief orientations toward politics and democracy. It is worth noting here that I recognize that the orientations and decisions of political elites are critical to understand the fate of processes of democratic transition and democratic consolidation (ODonnell and Schmitter 1986, Villaln and VonDoepp 2005). Nevertheless, I have decided to focus on the attitudinal dimension of ordinary citizens because although political elites exert a critical influence on political developments, it is the case that their strategic calculations are shaped by the distribution of political and regime preferences at the popular level (Alexander 2002). Political Culture Approach The contemporary trajectory of the political culture research is particularly interesting. During the 1950s and 1960s there was an exceptional level of interest in this topic because the behavioralist revolution at the time considered individuals as the main unit of analysis, emphasized the use of empirical data, and looked to universal theories (Lane 1992). Following this period, interest in political culture declined especially because of two main challenges. First, during 1960s and 1970s, the legitimacy of political culture research was challenged by the idea that political culture basically reflected class or ethnic status. Second, during this time, rational choice also challenged the relevance of political culture, stating that it was unnecessary because self-interest and short-term rationality were enough to explain individual behavior. However, in the last years both Marxist interpretations of political culture and rational choice approaches have

PAGE 67

57 been weakened as the most comprehensive explanations of individual behavior. This trend has encouraged the return of political culture approaches. Moreover, as part of this trend, it has been widely acknowledged that political culture can make important contributions to our understanding of key processes like democratization and economic development (Almond 1994). In the context of a renaissance of political culture,5 it is particularly important to pay attention to the main theoretical and empirical contentions of this particular approach toward politics. According to the seminal work on political culture by Almond and Verba (1963:14), the political culture of a nation is the particular distribution or patterns of orientations toward political objects among the members of the nation. These different orientations include: a cognitive orientation (i.e., knowledge of and belief about the political system, its roles, and the roles of the incumbents); an affective orientation (i.e., feelings about the political system, its roles, personnel, and performance); and an evaluational orientation (i.e., judgments and opinions about political objects based on the combination of values, feelings, and informa tion). These different orientations are not entirely independent from each other, for exam ple, political information and feelings play an important role in evaluational orientations. Unfortunately, as I will address in the next section, this consideration is missing in many contemporary works that use a political culture approach. By the same token, several factors, such as socialization, education, media exposure, and adult experiences with governmental, social, and economic performance play an important role in shaping political culture. Additionally, political culture is 5 About the current renaissance of political cultu re see Inglehart (1988), Almond (1994), Diamond (1994), and Power and Clark (2001).

PAGE 68

58 expected to affect political and governmental structures because values, feelings, and beliefs are key determinants of political behavior. Finally, regarding the connection between political culture and democracy, one of the Almond and Verbas main conclusions is that democratic stability requires a very specific type of political culture and democratic citizenship (i.e., the civic culture). This civic culture is defined as a mixed-model where participant political attitudes (i.e., a participant political culture) are balanced by considerable levels of passivity, trust, and deference toward authority (i.e., subject and parochial political cultures). Interestingly, a very similar theoretical approach can be found in several other works that, during the 1960s, were also dealing with political culture and its impact on the political system. For example, Pye (1965) contended that the notion of political culture assumes that the attitudes, sentiments, and cognitions that inform and govern political behavior in any society are not just random congeries but coherent patterns that mutually reinforce each other. By the same token, Dahl (1966), when discussing the impact of different types of political orientations on patterns of political partisanship, claimed that the orientation toward the political system as a whole affected the distribution of loyalty in a national society. Further, he claimed that attitudes toward cooperation, individuality, and other people aff ected the formation of political groups and their interactions. Finally, in dealing with the connection between political culture and political development, Verba (1966) stated that the most important components of a political culture included a sense of national identity, attitudes toward oneself as participant, attitudes toward ones fellow citizens, attitudes and expectations regarding

PAGE 69

59 government output and performance, and knowledge about and attitudes toward the political processes of decision making. There is no doubt that this seminal political culture approach has been very influential on the work of several scholars that came later and were also interested in political culture issues. Accordingly, several of its main contentions can be found in most recent works dealing in one way or another with the topic of political culture (Diamond 1994; Turner 1995; Lagos 1997, 2001, 2003a, 2003b; Inglehart 1998, 2003; Power and Clark 2001; Moreno and Mendez 2002). As a consequence of this, one of the main contemporary developments within scholars dealing with political culture issues has been the identification and measurement of specific political culture components (e.g., values) considered as highly significant in regards to the maintenance and development of democracy. Therefore, some of the most intense debates about the connection between political culture (i.e., political values and attitudes) and democracy, among scholars dealing with the political culture approach, have not been related to the shortcomings of the political culture approach, but instead to which ordinary citizens political values and orientations can exert the strongest influence on the stability of a democratic regime. Inglehart (1988), for example, has sustained that citizens of different societies are characterized by durable cultural orientations and that cross-national differences reflect the distinctive historical experience of the respective nationalities. Equally important, Inglehart has stated that the evolution and persistence of a popular-based democracy requires the emergence of certain supportive habits and attitudes among the general public. His own empirical research has aimed at identifying key habits and attitudes: a high level of interpersonal trust, satisfacti on with ones life, and long-term commitment

PAGE 70

60 to democracy or the diffuse feeling that democracy is inherently a good thing. As he stated, this cultural syndrome of satisfaction and trust is somewhat different from the definition of civic culture provided by Alm ond and Verba (1963)the right balance of parochial, subject and participant orientationsnevertheless, it is clear that both works share the same approach. More recently, Inglehart (2003) has revised so me of his previous assertions, but he fits still closely to what I have defined as the mainstream political culture approach. In one of his most recent works, he affirms that the level of popular support for democracy, as measured frequently in public opinion surveys, is not the best predictor for stable democracy. Rather, democratic stability is better predicted by interpersonal trust, tolerance to outgroups, post-materialist values, economic success and subjective wellbeing. Indeed, these different individuals valu es and orientations conform to a syndrome (i.e., self-expression values); such that a ny society that ranks high on one of them tends to rank high on all them. Nevertheless, as Turner (1995) contends, it has been also the case that, in the last years, there has been a great deal of room for improving this mainstream political culture approach. Accordingly, Muller and Seligson (1994) have argued that in Latin American countries, and particularly in Central American countries, there is no association between Ingleharts civic culture syndrome (i.e., interpersonal trust, life satisfaction, and opposition to revolutionary change) and democracy (i.e., democratic values and behaviors). They contend that there are three possible explanations for these findings that contradict Ingleharts contentions about th e link between civic culture and democracy. First, that Ingleharts theory is wrong and that the maintenance of a democratic regime

PAGE 71

61 may have nothing to do with political culture. A democratic regime may emerge and be sustained by forces entirely removed from ordinary citizens values, such as elite pacts and elite consensus, class structure, and level of economic development. Second, political culture may be significant, but the variables selected by Inglehart (1988, 1990) and Putnam (1993) may be wrong. Interpersonal trust might have little or no relevance to attitudes toward democratic institutions and processes. Third, the variables may be correct, but their operationalization may be flawed. The operationalization of the interpersonal trust item, a highly North American notion, is especially suspicious, because cross-cultural validation of the measure has been virtually absent. Nevertheless, I believe that this seems much more an attempt to try to accommodate the mainstream political culture approach to the reality of developing countries, rather than an effort to establish its lack of relevance. Therefore, it is not hard to find several scholars trying to adapt this political culture approach to the reality of new democracies, particularly in developing countries. For example, contending that political attitudes and values can be considered not only as an independent variable but also as a dependent variable. Similarly, contending that the relationship among different political attitudes is not necessarily linear, but possibly Vshaped. Finally, contending that short-term evaluations can be a powerful influence of key political attitudes and beliefs. Regarding this last point, some scholars have found that, in the context of Latin American countries, deep dissatisfaction with presidents performance can explain an increase in the acceptance of military governments (Seligson and Carrin 2002).

PAGE 72

62 Nevertheless, in most of these cases, the idea is not to question the explanatory power of the mainstream political culture approach regarding how individuals political values and orientations affect political processes. On the contrary, it seems that idea is to make the case for the need of both further theoretical development within the political culture approach and the use of different theoretical perspectives to account satisfactorily for individuals political orientations. Specifically, for example, in trying to explain the decline of citizen trust on political institutions and political authorities among established democracies, Dalton (2004) claims the need for using four different theoretical perspectives: (1) performance of governments, political institutions, and authorities, (2) political culture and value change, (3) social capital, and (4) media effects. Summarizing, the mainstream political culture approach is composed by the following main contentions. First, political culture is concerned with some persistent and stable components, such as basic political value commitments and beliefs (e.g., interpersonal trust, political tolerance, leve l of personal life satisfaction, etc.). These attitudes and beliefs are a deep-seated aspect of given populations and are, consequently, independent of short-term economic, social or political conditions. Second, the expectation is that it is going to be possible to find consistent and substantial crossnational and cross-sectional differences regarding these deep-seated political values, beliefs, and attitudes. Therefore, it is not surprising to find that some societies and some segments of these societies have different levels of interpersonal trust and political tolerance. Third, some specific political culture syndromes are particularly linked with the stability of democracy. Therefore, one of the main contributions of the political culture approach is to provide a set of political values and beliefs that can be considered

PAGE 73

63 as critical for the maintenance and development of democratic governments. Finally, it is the case that political culture has only a relative autonomy. In that sense, it is impossible to rule out the effect that modernization and structural changes have on the content of political culture. However, social and economic transformations by themselves will not produce democracy. Unless specific changes occur in the political culture, there is no guarantee that socio-economic structural changes will nurture a democratic regime. It is worth noting here, however, that this mainstream political culture approach does not represent a specific theory of the relationship among individuals values and attitudes, political behavior, and democratic stability. Rather, it is a more general theoretical approach to formulate specific th eories about individuals orientations and political development. In more practical terms, a political culture approach meant the identification of some individuals psychol ogical and subjective variables that were critical in testing hypotheses and developing theories for example about the development and maintenance of democratic regimes. However, it is equally important to recognize that scholars working within this mainstream political approach always impute an explanatory power to the psychological and subjective dimensions of politics (Almond 1980:26). Finally, given that both the research on popular support for democracy and the political culture approach deal with individuals attitudinal and behavioral orientations. And, given that it is very difficult to expl ain individuals political behaviors without considering the role of individuals political values, feelings and beliefs (Almond 1994). It is important now to assess the contributions that the political culture approach provides

PAGE 74

64 to our understanding of popular support for democracy and its impact on democratization. Critical Review of the Political Culture Approach The development of this mainstream political culture approach has not been free of criticisms. As a consequence of this, the dialogue concerning these different positions has generated insightful debates, particularly regarding the relationship between political culture and democratization. One of the first criticisms advanced about the political culture approach is that it imputes a causal relationship from political culture toward other political and economic structures (e.g., the level of democratic or socio-economic development). Nevertheless, it is worth noting here that Almond (1980:29) believes that this criticism is incorrect because political culture in The Civic Culture was treated as both an independent and a dependent variable, as causing political structures and as being caused by them. In any case, this type of criticism has been part of a broader concern about the main determinants of the content of political values and attitudes. Accordingly, the relationship between political culture and socio-economic development has drawn considerable attention. Specifically, part of the debate has focused on whether it is the countrys level of socio-economic development which basically generates the attitudes and values that are more favorable for the development of democracy (e.g., tolerance, political efficacy, higher predisposition toward political participation, civic life, etc.); or, whether the experience with democracy and alternative regimes generates an independent effect on the nations political attitudes and values. However, it is important to mention here that in both cases the idea is that these political values and attitudes could in turn overpower the effect of the country's levels of socio-economic development, and the regime's political

PAGE 75

65 and economic performance (Diamond 1999:162). Additionally, it is worth noting here that this specific debate has been particularly relevant within the discussion of the existence of some structural prerequisites for the development and consolidation of democracy (Lipset 1959; Rustow 1970; Gasiorowski 1998; Diamond 1994, 1999). Regarding these different interpretations of the relationship between political culture, and political and economic processes and structures, I believe that it is most useful to assume that the formation of political culture is a complex phenomenon that is influenced by different social, economic, and po litical factors. Additionally, I also believe that these political values, beliefs, and attitudes, once established, have a momentum of their own, and may act as autonomous influences on political and economic behaviors and processes. Accordingly, some scholars have made the case that some political values, beliefs and attitudes can have a critical role in the development and maintenance of democracy (Lipset 1981; Diamond 1994, 1999) Highly related to this debate, there has been an important discussion about whether political culture should be treated as a depende nt, or independent variable. Some scholars have stressed that a political culture approach cannot take these beliefs, attitudes, values, ideals, sentiments, and evaluations as given. In other words, political culture must not be treated as an uncaused factor that explains why people behave as they do, yet incapable of being explained in itself (Thompson, Ellis and Wildavsky 1990:218). About the previous point, I consider that it is more appropriate to assume that political culture can be both a dependent and an independent variable. Therefore, a political culture approach should not only be aimed at addressing the impacts of political culture on ordinary citizens and elites political orientations, and regime and government

PAGE 76

66 performances; but also what factors shape the content of this particular political culture. Predictably, assuming this dual character of political culture (i.e., as dependent and independent variable) is a challenging enterprise. However, considering the challenge of understanding the most recent processes of democratization, there is no doubt that it will be extremely important to address both how a phenomenon like popular support shapes the current experience of several new democracies, and what is playing a critical role regarding the development and content of political culture within these societies. There have been also some criticisms regarding the way some scholars use the political culture approach. It has been affirmed that rather than assuming that political culture is about testing single classificatory schemes or measuring some political cultures a priori defined,6 political culture should be seen as a method for analyzing the basic beliefs of any given group of people. As a consequence of this, it has been contended that a political culture model needs to be defined as a complex structure of logically linked beliefs, norms, and goals; and that scholars dealing with political culture issues need to departure always from the viewpoints of the persons that they are studying (Lane 1992: 364).7 Furthermore, there have been some criticisms about how the dependent and independent political culture variables have been operationalized.8 Concerns have arisen about how well this political culture approach is travels from the reality of established 6 Indeed, this was one of the problems with Almond and Verbas work (1963) because they departed from considering the existence of three main or pure types of political cultures (i.e., parochial, subject, and participant) that could be identified in almost any political system. 7 Lane (1992) also questions that in the field of political culture, th ere is too much emphasis on betweennation differences and much less on within-nation differe nces. Fortunately, this is something that has been changed recently. 8 Chapter 6 addresses this issue specifically.

PAGE 77

67 democracies to the reality of new democracies. In this regard, it is revealing how Inglehart (1988) seems to suggest that a c onservative attitude (i.e., individuals thinking that our present society must be valiantly defended against all subversive forces) vis-vis both revolutionary (i.e., individuals thi nking that the entire way our society is organized must be radically changed by a revolutionary action) and reformist ones (i.e., individuals thinking that our society must be gradually improved by reforms) can be considered always as the most pro-democratic attitude. Regarding the capacity of the political culture approach to deal with political and support for democracy, it is the case that other theoretical approaches are also relevant. For example, it has been affirmed that democratic politics is a social contract whereby governments perform certain functions in exchange for popular support (Dalton 2004).9 Accordingly, not only the economic performance, but also a more general policy performance is seem as one of the most important determinants of citizens political attitudes and orientations, including support for democracy. Nevertheless, it is important to mention that the empirical evidence regarding the impact of government performance on individuals political attitudes and orientations reveals that this effect is weak at the aggregate national level, but significant at the individual level (Clarke, Dutt, and Kornberg 1993; McAllister 1999; Nye and Zelikow 1997; Listhaug 1995; Kornberg and Clarke 1992) More specifically, there is evidence that objective economic circumstances do not generate clear feelings toward the polity. On the contrary, assessments of these objective conditions are heavily influenced and 9Easton (1976:436) was one of the first scholars to recognize this situation. According to him, () leadership assumes the responsibilities fo r tending to the problems of the soci eties. In return the leadership gains power to enable it to make and implement binding decisions, a power that it loses to an alternative set of leaders if is unable to supply some average level of satisfaction to its supporters.

PAGE 78

68 mediated by psychological judgments (Dalton 2004:65). By the same token, this empirical evidence suggests that individuals ideological positions and expectations toward government play also a critical role in shaping support for the government, political institutions, and the political regime. This is particularly relevant because seminal works about political culture (e.g., Almond and Verba 1963) argued for an evaluational orientation as one of the key components of political culture. Nevertheless, the large majority of contemporary works within the mainstream political approach has neglected this political culture dimension, focusing instead on the cognitive and affective orientations. Similarly, very little attention has been given to the connection between deep-seated political values and attitudeson the one handand individuals evaluations of government performanceon the other hand. It is clear for me that overcoming these limitations will contribute enormously to our understanding of popular support for democracy. Interestingly, this is what some scholars are trying to accomplish when they sustain that individuals definitions and expectations toward democracy and current democratic regimes factors into the citizenrys satisfaction and support of current democratic regimes in new democracies. Accordingly, in those cases characterized by a low level of popular support for democracy among Latin American countries, the problem seems to be a huge gap between democratic expectations and the overall performance of these current democratic regimes (Camp 2001, 2003). There is no doubt for me that this way of approaching the phenomenon of popular support for democracy constitutes a very

PAGE 79

69 plausible explanation for the high levels of political and democratic alienation among the most disadvantaged citizens in developing countries.10 Finally, it is important to recognize that some scholars have been very skeptical about the relationship between the level of popular support for democracy and the level of societal democracy. The main argument is that overt support for the idea of democracy is widespread among publics throughout the world. However, at the same time, a large number of countriesparticularly new democraciesare struggling not only with the quality of their democratic regimes but also with their stability. By the same token, surveys are showing that among many of these countries high levels of popular support for democracy coexist with other less democratic attitudes (e.g., the popular preference of strong leaders who does not have to bother with elections and the parliament). One of the main conclusions of this way of thinking is that what really matters for the development and stability of democracy are some specific citizens attitudes, like interpersonal trust and political tolerance rather than a general or overt support for democratic rule (Inglehart 2003, Moreno and Mendez 2002). I will discuss in subsequent chapters how it is possible to challenge this contention about the existence of a considerable level of popular support for democracy in new democracies. For now, it is enough to say that this misleading impression has been the result of the way several scholars have operationalized and measured popular support for democracyparticularly throughout the use of public opinion surveys.11 This way of 10 See Aragn (2001) for the specific case of Peru. 11 The most extensive survey question used to measure popular support for democracy has been the following: which of the following statements do you agree most? (a) Democracy is preferable to any other kind of government; (b) in certain situations an authoritarian government can be preferable to a democratic one; or (c) it does not matter to people li ke me whether we have a democratic government or a non-democratic government . This question has been included in all the independent associated with the

PAGE 80

70 operationalizing popular support for democracy presents several problems, beginning with the expectation that one or two genera l survey question can be enough to empirically account for this complex phenomenon. Not surprisingly, several works that relied on this type of operationalization of popular support for democracy tend to provide too general and unclear conclusions.12 Just to provide one example about the limitations of this approach toward popular support, we can consider how in the case of Latin America, and according to the last findings from the Latinobarmetro between 1996 and 2003, some politically stable countries (e.g., Chile and Brazil) present levels of both support for democracy and democratic satisfaction that are lower in comparison to some other countries that have suffered democratic reversals (such as Peru and Venezuela). Hence, one way to respond to the scholars who contend that there is no such relationship between popular support for democracy and the actual level of democratization is to contend that the problem is not that popular support for democracy has a very limited impact on the process of democratization of several developing countries. On the contrary, that the problem has been the standard, but not less unsatisfactory, ways that have commonly been used to deal with this phenomenon empirically. It seems that a comprehensive understanding of popular support for democracy in dew democracies will require a considerable improvement in its operationalization and measurement. Global Barometer (i.e., the Eurobarometer, the Latinobarmetro, the Africanbarometer, and The New Democracies barometer). Also, an alternative has b een to use general questions about the level of satisfaction with the functioning of democracy in their own country. The World Values Survey phrased this question in the following manner: on the whole are you ve ry satisfied, rather satisfied, not very satisfied or not at all satisfied with the way democracy is developing in our country; while the Global Barometer phrased it in a slightly different way: on the whole are you very satisfied, fairly satisfied, not very satisfied or not satisfied at all with the way democracy works in (own country) . 12 See Lagos (2003).

PAGE 81

71 I want to finish this section contending that all these debates and criticisms regarding the mainstream political culture approach do not mean that individuals political cognitive, affective, and evaluational orientations are of little relevance for our understanding of both individuals political beha vior and broader political developments. Rather, it means that assuming both the most important insights contain in the political culture approach (e.g., popular support for democracy as both dependent and independent factor) and its mains limitations (e.g., avoiding a ethnocentric or normative approach regarding political culture in developing countries) can constitute a unique opportunity to improve our understanding of democratization, particularly regarding the experience of new democracies. Political Culture and Popular Support for Democracy As I mentioned before, a political culture approach has been extensively used to provide insights about the connection between the subjective dimension of politics and the functioning and maintenance of democratic regimes. To a large extent, this has been the case because the topic about political support has a long tradition among scholars dealing with this subjective dimension of politics. Easton (1975), for example, provides a critical distinction between diffuse and specific support. According to this author, the former is related to the regime itself and to political principles, while the latter is related to more concrete evaluations about how the regime is performing. Therefore, it is possible for some citizens to be dissatisfied with the way democracy works in their country, but still committed to the principles of democracy. Although, this seems to be what is happening in several developed countries, it does not seem as true in the context of developing countries.

PAGE 82

72 More recently, several scholars dealing with political support in established democracies have identified an important decline in peoples trust regarding political institutions and authorities in the last decades. Nevertheless, the same empirical evidence shows that this political dissatisfaction is not affecting popular support for democratic norms and principles (Craig 1993; Dalton 1999, 2004; and Pharr and Putnam 2000). On the contrary, the situation of several developing countries is much more complicated because ordinary citizens political perceptions and attitudes combineon the one handa high level of dissatisfaction regarding the performance of their current democratic government (i.e., specific support) withon the other handa somewhat contradictory orientation regarding the current democratic regime (i.e., diffuse support). Specifically, a considerable level of ordinary citizens preference for a democratic regime exists alongside an almost total lack of support regarding some key democratic procedures and the institutions that are supposed to implement them (e.g., congress and political parties).13 Therefore, regarding the distinction between diffuse and specific support, the experience of developing countries is making clear two things. First, that it is necessary to develop a more detailed approach regarding these two broad categories of popular political support. Accordingly, we need to identify different political objects within each one of these general categories. Second, at least in the context of developing countries, one or two decades of persistent low specific support for democratic governments has a negative impact on the level of diffuse support for the current democratic regimes. 13 In the case of Latin American countries, the Latinobarmetro shows consistently that in the last years while a clear majority of citizens in this region believe that democracy is preferable to any other form of government, a similar majority also believes that congress and political pa rties are not indispensable for the functioning of a democratic regime (1999, 2001, 2002, 2003, 2004 and 2005).

PAGE 83

73 However, as it will become evident later, the study of popular political support and popular support for democracy requires much more than Eastons initial distinction between diffuse and specific support. Fortunately in the last years, the study of support for democracy in theoretical and empirical terms has made important advancements. A majority of scholarsconfronting the difficulties in measuring and opertaionalizing popular support for democracy14have agreed on the need and convenience of distingui shing at least five different levels or objects of political support within democratic re gimes. One of the main justifications for this is that citizens do distinguish among these different levels or objects when they are assessing their democratic political system. In this direction, Dalton (1999, 2004) and Klingemann (1999) have shown that factor anal ysis strongly suggests that citizens indeed make these distinctions, and that it is possible to identify specific trends regarding these different dimensions of popular support for democracy. Therefore, the concept of support for democracy has been defined as a multidimensional concept that includes, at least, five levels: (1) support for the political community; (2) support for the core regime principles, norms, and procedures; (3) assessment of the regime performance; (4) support for the regime institutions; and (5) support for the political actors and authorities.15 These scholars have provided a more detailed elaboration of Eastons seminal ideal (1965, 1975) that political support can be 14 Dalton (1999) contends that operationalizign and measuring popular support for democracy is always difficult, especially in a compara tive way, because democracy is a complex phenomenon, and it can have several and different meanings. 15 See Seligson and Carrion (2002), Norris (1999a, 1999b) Klingemann (1999), Dalton (1999), Mishler and Rose (1999), Holmberg (1999), Fuchs (1999), Rose, Shin and Munro (1999), Newton (1999), Mcallister (1999), and Miller a nd Listhaug (1999).

PAGE 84

74 defined as a continuum from evaluations of the immediate actions of government officials to the identification with the nation-state. This distinction among different levels or objects of popular support for democracy is critical in improving our understanding of democracy and democratization in both established and new democracies. For example, a low level of popular support for democratic authorities will have a different impact in comparison to a low level of popular support for some democratic procedures on the stability of a democratic regime. Indeed, one of the most important challenges of popular support for democracy among new democracies is the combination of a high level of dissatisfaction with the performance of authorities and some political in stitutions with a lack of support for some key democratic procedures.16 Equally important have been the advancements regarding the differences in approaching popular support for democracy between established democracies and new democracies. Accordingly, it has been affirmed that the underlying dynamics of popular support for a new democracy are different from those of an established democracy. While support for the democratic principles, norms, and procedures is not an issue in established democracies, and tends to change slowly over time in these cases; it tends to be much more volatile in the case of new democracies. One of the key explanations for the volatility of popular support for the core democratic principles, norms, and procedures within new democracies lies in the fact that this kind of popular support for democracy depends basically on regime and governments performance. More specifically, in the context of new democracies, the new democratic regime may benefit from a degree of popular acceptance and approval resulting from the rejection of the former and non16 See Chapter 5 and 6 for a discussion about this topic in case of contemporary Peru.

PAGE 85

75 democratic regime. However, in the longer term, support for the new democratic regime will be mostly performance-based. In other words, in the context of new democracies, some level of individual and collective good is indispensable to avoid the rapid disappearance of the initial popular support for the new political regime (Mishler and Rose 1999). Furthermore, in trying to assess the dynamic of popular legitimacy or support for democracy in developing countries, Rose, Mish ler, and Haerpfer (1998) have made clear that the main problem of democratic legitimacy in these cases is the competition between democratic and non-democratic forms of government. The reason for that is that citizens in new democracies have experienced both types of regimes. Therefore, the fate of these new democratic regimes will depend on the competition between democratic and nondemocratic alternatives. Consequently, we need to pay attention not only to how democratic governments are perceived by ordinary citizens but also to how they are perceived vis--vis some lessor non-democratic alternatives. Specifically, in regards to the main determinants of popular support for democracy in the context of new democraciesparticularly within developing countries, it is critical to pay close attention to a particular set of possible explanations. First, on the role of popular definitions of and expectations toward democracy, some scholars have argued that there are key differences between deve loped and developing countries. According to Camp (2001, 2003), in developing countries, democracy is not only associated with a set of political rights that make possible the election of the authorities, but also associated with the provision of some specific outcomes such as employment, well-being, and a reduction of social and economic inequality. Accordingly, it has been affirmed that in

PAGE 86

76 these countries, democracy is not perceived as a way to regulate political competition and participation, but at the same time, people expect democracy to solve all social problems, especially income inequality (Przeworski et al. 1996). Second, regarding the current performance of new democratic regimes, some scholars have suggested that it has been a poor performance one of the main reasons for the lack of popular support for current democratic regimes and one the main sources of popular alienation regarding democratic institutions and democratic procedures (Mainwaring and Hagopian 2006, Hagopian 2006). Additionally, it is important to note that the problems regarding the performance or outcomes of these new democratic regimes in developing countries are not only related to the expectations regarding some very specific socio-economic outcomes (i.e., economic prosperity, economic stability, employment, etc.). A poor performance in other non-economic critical areas (such as an effective rule of law, equality before the law, effective mechanisms of political representation, etc.) can also be considered as problematic for the development of strong and stable support for democratic procedures and institutions (Foweraker and Krznaric 2002; ODonnell 2004a, 2004b, and 2004c). Moreover, it seems that the development of popular support for democracy in these countries will require an overall increase in the quality of political participation, inclusion, and representation; particularly in regards to the most disadvantages citizens. Interestingly, it seems possible to assert that, in many developing countries experimenting with democracy, there is a huge gap between popular ideas and expectations toward democracy, and the performance of the new democratic regimes. This gap, in turn, would explain why democratic procedures and institutions can be very

PAGE 87

77 frequently perceived as ineffective or unnecessary mechanisms of political participation and representation. Chapter 5 and Appendix B provide a very good example of this situation. Third, on the effect of acute levels of economic and social inequality on popular support for democracy, it has been argued that there is a clear tension between extreme inequality and democracy on the other hand (Lipset 1959; Beetham1994; Diamond, Linz, and Lipset 1989; ODonnell 2004b, 2004c). According to this explanation, living in povertyand in some cases in extreme povertycan have a strong impact on most disadvantaged citizens political perceptions, attitudes and behaviors. More specifically, it is highly likely that for people living in such conditions, the democratic regime itself, and its formal procedures and institutions, will not be perceived as a necessary or efficient mechanism to represent their particular interests and needs. Additionally to this, it is possible to expect that living in poverty does not provide the necessary resources, opportunities and time to develop political and democratic concerns. Therefore, a general attitude of indifference toward democracy and politics could be one of the most common features among poor populations. Accordingly, the existence of widespread poverty is what explains the development of clientelistic (Stokes 1995), plebliscitarian (Parodi 1993, Murakami 2000) and/or neopopulist (Weyland 2006) political orientations among poor citizens within new democracies in developing countries. In turn, all these patterns of political interaction between poor citizens and authorities explain why very frequently, particularly in contexts or crises, a considerable segment of the population of these new democracies are

PAGE 88

78 willing to accept the idea that there is a trade-off between the implementation of some specific policies and the respect for the democratic procedures. In conclusion, in dealing with popular support for democracy in new democracies, it is critical to consider the following issues. First, popular support for democracy or democratic legitimacy is a complex phenomenon with several and different dimensions; and each one of these dimensions can have a different impact on the overall level of support for democracy and current democratic regimes.17 Second, in the context of new democracies, popular support for democracy requires addressing the competition between democratic and non-democratic alternatives. Third, particularly in the context of new democracies in developing countries, it is impossible to identify a set of factors and variables that seem to be critical determinants of popular support for democracy (e.g., peoples ideas and expectations toward democracy; individuals assessments of democratic governments and current democratic regimes; the effect of social and economic inequality on these ideas, expectations and assessments, etc.). Fourth, thinking of the possibility of further democratization among new democracies, it is important to assume that popular support for democracy should not be seen as the final step of a democratic consolidation transition, but a process itself that can facilitate or hinder a gradual and constant process of democratization. In others words, what is at stake in several new democracies is the gradual development of ordinary citizens attitudes and orientations that can help to maintain and improve current democratic regimes. Chapter 6 in this dissertation uses survey data collected in Peru at the end of 2004 as an empirical assessment of these different explanatory variables regarding popular 17 Indeed, popular support for democracy is a complex phenomenon that needs to be address both as an outcome or dependent variable, as an explanatory or independent variable.

PAGE 89

79 support for democracy. The goal is not only to identify which ones of these explanatory variables exert the strongest and most consistent impact on popular support for democracy, but also to account for how they interact. Similarly important, Chapter 6 provides an innovative and reliable way of operationalizing some key different dimensions of popular support for democracy as dependent variables.

PAGE 90

80 CHAPTER 4 POPULAR SUPPORT FOR DEMOCRACY IN COMPARATIVE PERSPECTIVE Although, it is the case that some of the basic underlying dynamics of popular support for democracy in new democracies differ from the experience of the most established democracies. This chapter begins with a consideration of how this topic has been addressed in the last years in the context of established democracies. As it will become evident these differences in the dynamic of popular support for democracy do not make irrelevant a comparative analysis between established and new democracies because we can learn a great deal in both theoretical and empirical terms from this effort. It is also the case that among new democracies, the literature on popular support for democracy within post-communist countries shows a considerable level of sophistication and advancement. As a consequence of this, I will contend that scholars who are concerned with popular support for democracy in a region such as Latin America can learn a great deal from what has been done for Central and Eastern European postcommunist countries; particularly regarding how citizens can attach different meanings and expectations to the term democracy, and how it is possible to measure and explain popular support for democracy. Therefore, the second section in this chapter provides a separate assessment of the literature on popular support for democracy in post-communist and Latin American countries.

PAGE 91

81 Established Democracies: Popular Support for Democracy and the Critical Citizens Thesis In addressing the relationship between public opinion and support for democracy in established democracies there are two main contentions that are widely accepted. First of all, the recognition that in the last decadesstarting in the late 1960sthere has been a considerable decline in public confidence in the main institutions of representative democracy, in particular, the government, the congress, the judiciary, and political parties; and in the individuals that are in charge of these institutions (i.e., authorities and politicians). By the same token, citizens in many established democracies are giving poor marks to how they governments are functioning. Second, that these trends in public confidence and public assessments among established democracies are not implying an erosion of support for the democratic principles and procedures. In that direction, there is no evidence among established democracies of a crisis regarding the legitimacy of democracy as the best alternative for political regime.1 In summary, contemporary publics in established democracies are dissatisfied with the incumbents of office and even with the political institutions of representative democracy, but these feelings of dissatisfaction have apparently not yet affected basic support for their democratic regime (Craig 1993; Miller and Listhaug 1999; Mcallister 1999; Newton 1999; Holmberg 1999; Klingemann 1999; Fuchs 1999, 1995; Norris 1999a, 1999c; Dalton 1999, 1996; Torcal 2001; Pharr and Putnam 2000). The combination of these two trends, that seem more than just a transient phenomena merely linked to distrust of incumbents, have suggested the possibility that a 1 Klingeman (1999) and Dalton (1999) provide empirical evidence about the coexistence of strong and current commitment for democratic principles a nd processes, and citizen frustration with how contemporary democratic systems are or are not working within established democracies.

PAGE 92

82 large part of citizens in established democracies can be defined as critical citizens or dissatisfied democrats, individuals who value democracy as an ideal yet who remain dissatisfied with the performance of their political systems, in particular the core institutions of representative democracy (Dalton 1999; Klingemann 1999; Norris 1999a, 1999c; Torcal 2001). The important point here is that the rise of critical citizens can be hardly considered as a sign of a crisis of democracy in the most developed countries. On the contrary, several scholars maintain that the growth of more critical citizens who are dissatisfied with established authorities and traditional political institutions, who feel that existing channels for participation fall short of democratic ideals, and who want to improve and reform the institutional mechanisms of representative democracy can be a very positive influence for the future development of these democratic regimes. More specifically, it has been affirmed that the rise of critical citizens has increased the pressure for several constitutional reforms to make elected governments more accountable to the public (e.g., the introduction of direct democracy mechanisms as a way to allow more opportunities to participate in the decision-making process, and the reform of electoral systems). Therefore, according to this perspective, critical citizens are pushing for the reform of democratic institutions, the widening citizen involvement in government, and the establishment of new channels to link citizens and state (Norris 1999c, Shugart and Wattenberg 2001, Dalton 2004). It is important to note here that the critical citizens thesis is challenging not only any contention about the crisis of democracy in developed countries, but alsoat least for now and in the context of established democraciesthe idea that a successive dissatisfaction with the performance of democr atic government will lead inevitably to the

PAGE 93

83 erosion of legitimacy of democracy. On the contrary, it seems to confirm the contention that established democracies posses a reservoir of public support for democracy that can help democratic regimes to overcome hard times. Although some authors believe that the rise of critical citizens or dissatisfied democrats is to some extent a phenomenon that can be found in almost all democratic regimesincluding new democracies, it is clear that there is a big difference in terms of its extension and relevance between new and established democracies. Critical citizens are not a majority or even a considerable part of public opinion in new democracies. By the same toke, as I will show later, the key difference in support for democracy between new and established democracies is not only that newer democracies show a remarkably low level of confidence in democratic institutions vis--vis established democracies, but also that there is evidence of a very strong public dissatisfaction with governments performance and even some erosion of support for democratic principles and procedures among new democracies. Therefore, even when the notion of critical citizens or dissatisfied democrats can present some limitations in addressing the issue of support for democracy in new democracies, I believe that a revision of the explanations about the decline of popular support for democratic institutions and politicians among established democracies can provide very useful insights regarding the understanding of support for democracy in new democracies. In this direction, it is important to note that there is not a high level of consensus regarding the causes and consequences of the rise of critical citizens in established democracies. Accordingly, it seems thatin the context of established

PAGE 94

84 democraciessupport for democratic institutions, authorities, and politicians are shaped by different factors (Norris 1999b). Consistently with the empirical evidence showing that the problem with support for democracy in established democracies operates basically at the level of public support for institutions and main political actors (i.e., authorities and politicians) (Klingemann 1999, Dalton 1999), several authors have worked on the explanations of these trends, as well as in its main consequences for these democratic regimes (Inglehart 1990, 1997, 1999; Holmberg 1999; Fuchs 1999; Newton 1999; Mcallister 1999; Miller and Listhaug 1999; Norris 1999b). Before going any further, it is important to mention that there is no consensus regarding how to operationalize these outcome and explanatory variables. For example, in some cases the outcome variable is simply either the level of trust in politicians (Holmberg 1999) or the level of trust in government (Newton 1999). In other cases, the outcome variable is operationalize as a combined index of confidence in several institutions such as the parliament, the civil service, the legal system, the police and the army (Mcallister 1999, Norris 1999b). Basically the same happens regarding the operationalization of the explanatory variables. Therefore, it is the case that some caution is required in generalizing some of these findings for the entire universe of established democracies. Regarding the explanations of the decline in public trust in institutions and politicians, probably one of the most important conclusions is that current peoples evaluation of political and economic government performance appears as one of the main determinants. Nevertheless, it is also the case that there are other factors that can be

PAGE 95

85 considered as important determinants of the level of trust in institutions and politicians. Among these other important factors we have peoples understandings of democracy and citizens expectations toward government (Fuchs 1999, Miller and Listhaug 1999); the difference between being a winner (i.e., supporting the government) or a loser (i.e., supporting the opposition) in the political system (Newton 1999, Norris 1999b); the distance between citizens ideological positi onin the ideological left-right dimension and the ideological positions of their preferred political party (Holmberg 1999); the time the democratic regime has been in place and, as a consequence of that, the time it has played an important role in the political socialization of its citizens (Mcallister 1999); deep-rooted processes of value change that are gradually transforming citizens relationship toward government (Abramson and Inglehart 1995; Inglehart 1977, 1990, 1997, 1999); and even the type of political institutions that are in place (Anderson and Guillory 1997, Norris 1999b).2 Therefore, it is possible to contend that a comprehensive explanation of the lack of confidence and support for political institutions and politicians in established democraciesand eventually in new democracieswill demand the use of different theories and approaches (i.e., theories about citizens assessments of government performance, theories about political culture and political socialization, and institutional theories). This is even more important once we recognize, for example, that there are 2 It is important to note that in the process of identifying the determinants of public support for democratic institutions and politicians, some other factors, previ ously thought as important, have been ruled out, for example, objective measures of government economic performance measures as the rate of inflation and unemployment (Mcallister 1999), the countrys level of social trust and the participation in voluntary organizations and community involvement (Newton 1999, Holmberg 1999), and the impact of standard socio-economic variables (Newton 1999).

PAGE 96

86 important connections between the way citizens assess their government and regimes, and their political values. Equally important, the experience of established democracies reveals that citizens assessments of government performance are very complex phenomena. First of all, in assessing the performance of their government, citizens can use different criteria. They may use absolute levels of outputs or benefits received from government, or they may compare current benefits with past benefits, or they may compare the benefits that they are getting with what they think other citizens are getting, or, even, they may use some ethical principles. For example, Miller and Listhaug (1999) have found that ethical standards such as the expectation that government follows procedures that produce outcomes that neither advantage nor disadvantag e particular groups unfairly play also an important role in the overall assessment of government performance. In a similar vein, it is critical to recognize, as Norris (1999b) does, the importance of considering non-economic aspects of government performance such as the provision and respect of civil liberties and the quality of political representation. Regarding the relationship between political representation and citizens assessments of government performance, there is strong evidence that c itizens identified with the party in power are more likely to trust government and political institutions, while citizens identified with the parties in the opposition are more likely to express political distrust in government and political institutions (Anderson and Guillory 1997, Newton 1999, Norris 1999b). In a similar vein, Holmberg (1999) found that in the case of Scandinavian countries, and particularly in the case of Sweden, one of the most important explanations for the decline of public trust in politicians is the decline in the number of people perceiving that they

PAGE 97

87 have the same positionin a left-right ideological continuumas their preferred party; and a negative evaluation of their own partys policies. Second, citizens assessments of government performance are highly dependent on political culture issues. For example, Fuchs (1999) contends that in current Germany coexist two different models of democracy. While Western Germans hold a liberal model of democracy (i.e., a minimalist or procedural definition of democracy), Eastern Germans hold a socialists model of democracy (i.e., a model of democracy that adds to the minimal or procedural dimension supplemental elements such as the constitutional guarantee of social rightsa comprehensive list of welfare servicesand the political realization of these social rights). The problem is that the current democratic system is consistent with the liberal model of democracy. Consequently, support for this political regime is relatively high in West Germany and relatively low in East Germany; and these differences can be explained basically by different fundamental attitudes toward democracy. In other words, the political culture in East Germany is not congruent with the democratic system of the unified Germany. More specifically, the tension between these two different models of democracies is that a liberal model of democracy lives from permanent disputation about the goals that are to be attained through political processes, and that this permanent dissensus requires to be regulated through a set of rules and procedures legally defined by the constitution and embodied in the political institutions. These rules and procedures must be accepted by the citizens is they are to fulfill their regulatory function; or, in other words, the support of citizens must primarily relate to these rules and procedures that cannot guarantee any outcome in particular. And this exactly what is not working among citizens

PAGE 98

88 that were socialized, intentionally or unintentionally, in a socialist model of democracy (Fuchs 1999:142).3 Another example of how political culture plays a role in shaping citizens assessments of government performance is Ingleharts contentions about the postmaterialist revolution in established democracies (1990, 1997, 1999). According to him, a steady process of economic development (i.e., modernization)implying an important increase on the levels of individual prosperity and securityleads to the emergence of new and more demanding standards by which governmental performance is evaluated. The difficulties of current government to satisfy these demands is what is mainly causing the lower levels of trust in political institutions and politicians, and a general low level of satisfaction with the functioning of the political system. Briefly said, this process of values change is not undermining citizen support for democracy, but making democratic government more difficult (Dalton 2004). In conclusion, the scholarly literature about support for democracy in established democracies is basically addressing the cause and consequences of an important decline in citizen trust in political institution and the main political actors (i.e., authorities and politicians). In doing that, it has become evident that one of the most critical factors shaping support for democracy in these countries is the way citizens are assessing the performance of their governments. Moreover, another important contribution of all these works has been several insights about the dynamic of citizens assessments of 3 I believe that something similar is happening not only with other post-communist countries but also with other developing countries without a communist legacy. In this direction, there is evidence of a lack of congruence between this model of liberal democracy a nd the political culture of large segments of Latin American citizens (Camp 2001, 2003). Regarding this lack of congruency, an optimistic view would contend that a continued experience with a liberal democratic regime will make possible the development of people values and orientation congruent with this type of democratic regime.

PAGE 99

89 government performance. It is not only case that citizens can use different standards and that they do care for both the economic and non-economic performance, but also that several other factors play a critical role in the way citizens finally assess the performance of their governments. In this direction, two factors deserve a special mention. First, political representation exerts an important effect on how government and institutions are assessed. The identification with the political party that is currently in power or the perceptions that there is a political party that can represent individuals policy preference will have a positive impact in the level of trust in government and political institutions. Second political culture issues are another important influence for citizens evaluations of the performance of their governments and political regimes. Different expectations toward government and different ideas about the own political regime will have different effects on the level of political trust in institutions and politicians. Additionally, these political culture issues can provide a critical contribution to understand some persistence and long-term difference in popular support for democracy that are insufficiently explained by short-term changes in governments and regimes performances. Popular Support for Democracy in New Democracies Post-communist Countries The breakdown of state socialism in the Soviet Union and Eastern Europe between 1989 and 1991, and the subsequent rise of new regimes and states throughout this region allow us to broaden the discussion of recent processes of democratization. Several studies of these post-communist cases have addressed questions such as why democracies arise,

PAGE 100

90 and why some of the new democracies succeed while others fail.4 Therefore, important similarities exist between this literature on post-communist countries and the literature about the latest processes of democratization in Latin American countries. To a large extent, these similarities exist because several countries in both regions are trying to develop functioning and stable political regimes organized in a democratic way. Nevertheless it is worth nothing that there are also important differences regarding the processes of democratization in both regi ons. Regarding the type of regime that was abandoned in order to move toward democracy, there are important differences between post-communist and Latin American countries. Typically in the former, the transition toward democracy meant the collapse of a totalitarian communist regime. In the case of the latter, the transition toward democracy meant the withdrawal of an exhausted and divided military from power and a compromise between the military and conservative groups on the basic rules for a new civilian regime (Geddes 1999; Bunce 2000, 2003). Similarly, we can expect some important differences in the political experience and socialization processes between citizens from post-communist and Latin American countries. For example, citizens from post-communist countries not only lived most of their lives in a communist society and were socialized in a communist ideology, but also mostly lacked any previous experience with democracy. However, the majority of Latin American citizens had experienced non-democratic regimes in the form of military governments, and had experience with both military and civilian governments. Also, regarding the transition process itself, Latin American countries moved toward formal democratic regimes, whereas post-communist countries had a dual 4 See Bunce (2003) for a more detailed description of the main topics within the literature on postcommunist countries.

PAGE 101

91 transitioni.e., toward a formal democratic regime and a market economy.5 Consequently, this particular experience of post-communist countries provides the possibility of addressing the interaction between pro-democracy and pro-market attitudes, and to what extent the issue of government responsibilitiesa key component of the socialist democracy that existed in the communist regime (Fuchs 1999)played a role in shaping these two pro-reform attitudes. Finally, we can use other important differences to address the prospects of future democratization in both regions. Communist regimes effectively achieved strong political control over the military, something absent in several Latin American countries. By the same token, communist regimes also effectively reduced the levels of economic and social inequality among their citizens, another situation that contrasts with the inequality in several Latin American societies. Overall, the similarities and differences between Latin American and postcommunist countries provide a unique oppor tunity to improve our understanding of recent processes of democratization in both regions. Accordingly, in this section, I address mainly the research and the debate that scholars of post-communist countries have developed to discuss the role and the relevance of popular support for democracy within these recent democratization experiences. At the onset, I would like to mention that the literature about citizens attitudes toward democracy and their democratic regimes in post-communist countries is not only much more extensive in comparison to the literature about Latin American countries, but, also better grounded theoretically and better designed methodologically. Consequently, 5 It is important to note that some post-communist countries such as Belarus actually experienced a triple transition because before the development of a demo cratic regime and a market economy they faced the need of developing basic state institutions.

PAGE 102

92 this literature provides a very interesting debate about the nature and main determinants of ordinary citizens political perceptions and attitudes,6 as well as its contribution to the process of democratization. Nevertheless, it is also the case that the literature on popular support for democracy in post-communist countries presents the same problems and limitations that characterize this type of research. First, it is difficult to attempt generalizations because there is significant variation in the empirical base of the different works. Specifically, in the case of post-communist countries, few works include Central, Eastern and post-Soviet Union countries.7 Second, there are high degrees of variation in the way the outcome and explanatory variables have been operationalized; and this is particularly relevant in the way scholars have operationalize popular support for democracy. Finally, there is a lack of works trying to systematize and making sense of different researchers findings.8 In the remaining of this section, I will address two of the main topics contained in the research about popular support for democracy within post-communist countries: (1) the meanings of democracy among post-communist citizens, and (2) the measurement and explanation of popular support for democracy in post-communist countries. Meanings of democracy among post-communist citizens One of the most important questions regarding popular support for democracy in post-communist countries has been how post-communist citizens understand and define 6This debate contains different approaches towa rd popular support for democracy. Some scholars use a more typical political culture approach, while other scholars give more attention to citizens assessments regarding the economic and political performance of th e new democratic regimes. There are even some scholars that contend what really is important to understand popular support for democracy is to pay attention to the impact of some structural factors. 7 Mishler and Rose (2002) provide an exception to this trend. 8 Fleron and Ahl (1998) provide an interesting exception to this trend, although they focus exclusively on Russia.

PAGE 103

93 democracy. In general, citizens ideas about democracy are clear indicators of the expectations people have of their regime, government and society. In addition to this, this topic was particularly relevant in the context of post-communist countries because major discrepancies between the meanings of democracy expressed by, for example, the ordinary citizens and the elites would be revea ling a lack of a shared political outlook in a time where these societies were precisely redefining their social goals (Miller, Hesli and Reisinger 1997). More specifically, one of the main goals of this research was to know whether postcommunist citizens understandings and definitions of democracyand consequently citizens expectations from their democratic regime and governmentswere mainly about political rights and procedures that can be used to influence governments decisions; or about particular economic and so cial rights, or specific economic and social (Fleron and Ahl 1998). Moreover, in a context were these countries were developing both a democratic regime and a market economy, some scholars have devoted considerable attention to address to what extent the popular preference for a democratic regime included some economic and social considerations such as certain degree of economic and social equality. In that direction, particularly in the case of some former Soviet Union countries (i.e., Russia and Ukraine), empirical evidence shows that a large part of the ordinary citizens in these countries tend to associate democracy with specific outcomes such as economic prosperity and rule of law. As a consequence of this, much less important in the way these citizens approach democracy are political and procedural definitions of democracy such as multiparty competition, freedom of speech, and the protection of

PAGE 104

94 minority rights (Whitefield and Evans 1994; Miller, Hesli, and Reisinger 1995; Miller, Hesli, and Resinger 1997; Fleron and Ahl 1998).9 Not less relevant have been the findings regarding how different understandings and definitions of democracy are particularly related to some specific social groups within the post-communist societies. In that direction Miller, Hesli and Reisinger (1997) were aimed to determine if there was a meaning of democracy that was widely held in common by all Russians and Ukrainians citizens. Their analysis revealed major differences between the meaning of democracy as espoused by the ordinary citizens and elite, as well as systematic variation in the meaning of democracy across different subsets of both the ordinary citizens and elite respondents (170). Particularly regarding ordinary citizens in both countriesRussia and Ukrainethey found that respondents who lived in rural areas, who felt that their economic condition would improve in the coming year, who thought that individuals should be responsible for their own economic well-being, who were relatively more involved in politics, as well as the better-educated, all gave relatively greater emphasis to freedom when discussing the meaning of democracy. By comparison, those from urban areas, those who were pessimistic about their future economic situation, those looking to the government to provide a good standard of living, those less likely to participate and the less well-educated all gave relatively more emphasis to the rule of law. Additionally, the identification with a political party showed an important effect on citizens perceptions of democracy. Those citizens who felt that their interests were represented best by some particular party gave greater emphasis to freedom and majority rule, and significantly less emphasis to 9 For a more philosophical and less em pirical approach about the difficulties of introducing key democratic principles such as representation and separation of power in Russia see Biryukov and Sergeyev (1994).

PAGE 105

95 negative comments about democracy than did non-partisans in both Russia and Ukraine (173). In conclusion, according to these scholars, people from these countries do not have a widely shared prototype of democracy in mind when they are evaluating their emerging political regimes. Equally important is the fact that these differences appeared systematically associated to specific demographic, attitudinal and behavioral categories (175). More specifically, in the context of a dual transition in Russia and Ukraine, a sense of economic security (i.e., a positive assessment of the current economic situation), a sense of confidence in the own capacities (i.e., a higher level of education), and also the sense that it is possible to get political repres entation (i.e., being identified with a political party) are clearly related to an understanding of democracy that stresses freedom and to more positive comments about democracy. On the contrary, a sense of economic insecurity (i.e., a negative assessment of the current economic situation), a sense or a conviction that the own capacities are not enough to deliver minimum living standards (i.e., believing that the state should be responsible for peoples living standards and a lower level of education), and the lack of identification with any political party are clearly related to a understanding of democracy that stresses the rule of law and to more negative comments about democracy. At the beginning of this section, I contend that the relevance of identifying different understandings and definitions of democracy lies in the strong impact that they can have on citizens expectations toward democracy and, as a consequence of that, on the way citizens assess the performance of their current democratic regimes. In this direction, I certainly believe that the evidence presented by Miller, Hesli and Reisinger (1997)

PAGE 106

96 supports this claim. In the case of Russia and Ukraine, we have seen that citizens who thought about democracy in terms of various political and economic freedoms were more likely than the average citizen to see their democratic regimes as fulfilling their expectations of a democracy. On the contrary, citizens who thought of democracy in terms of the rule of law and economic prosperityand probably, as a consequence of that, a higher level of governments social and economic responsibilitywere largely negative in their evaluations of the extent to which the current regimes fitted their notion of democracy. One additional lesson would be that in the context of new democracies, if citizens approach democracy in terms of specific social and economic outcomes such as rule of law and economic prosperity, these new democratic regimes will face a situation where it is going to be more difficult to be perceived as fulfilling citizens expectations. Another way to deal with understandings and definitions of democracy in postcommunist countries has been to address the level of coherence, consistency or constraint10 that ordinary citizens within these countries exhibit in terms of their democratic ideas. Accordingly, Miller, Hesli and Reisinger (1995, 1997) have found repeatedly that ordinary citizens in Russia and Ukraine have a low level of consistency and coherence among their attitudes toward democratic principles.11 Again, the ordinary citizens more highly involved in politics and with some level of party identification 10 The idea of a level of popular or mass attitudinal c oherence or constraint came from the belief system framework provided by Converse (1964). This theoretical and empirical framework was originally used to measure the level of mass citizens attitudinal consistency in regards to some policy issues. A lack of mass citizens attitudinal consistency will be revealing basi cally the existence of a non-attitude (Miller, Hesli, and Reisinger 1995, 1997). 11 For example, they found that within the mass respondents, who in general tend to hold a pro-democratic orientation, a large majority believed that participati on of people is not necessary if you have a few trusted competent leaders. By the same token, there was not a clear consensus at the level of mass respondents regarding the need of party competition and the i ndividual right to organi ze opposition (Miller, Hesli, Reisinger 1997:176).

PAGE 107

97 showed the highest level of coherence and consistency in terms of their democratic ideas and beliefs. Regarding these findings, Miller, Hesli and Reisinger concluded that on the one hand it was true that a majority of ordinary citizens in Russian and Ukraine preferred a democratic rule to an authoritarian one; however, on the other hand, it was also true that this attitudinal commitment to democratic principles was not very consistent (1995:33). This was the case because, at the level of ordinary citizens, beliefs about democracy have not yet formed a deeply rooted democratic political culture. The main risk that these scholars saw in this situation was that given this lack of consistency, coherence, and constraint among democratic beliefs and ideas, the current ordinary citizens preference for democracy could shift dramatically in the face of continuing social and economic difficulties (1997:187). Another alternative, and not less interesting, way to address to what extent popular understandings and definitions of democracy entail an important degree of economic and social considerations and expectations (e.g., economic and social equality) is to review the research and debate regarding the relationship between support for the political transition (i.e., development of a democratic regime) and support for the economic transition (i.e., support for a market economy).12 12 In terms of the overall debate about post-communi st citizens attitudes and perceptions, it is worth nothing that the topic regardi ng the connection between political support and economic support has concentrated much more attention than the efforts directly related to understand how post-communist citizens approach and define de mocracy. I believe that this was the case because understanding the relationship between political and econom ic support was seen as a critical e ffort to assess the prospects of these two different sets of refo rms. As Gibson (1996) points out one possibility was to find out that economic attitudes were shaping po litical attitudes. In this scenar io, democracy would be supported basically for instrumental reasons (e.g., the need of economic reform). Other possibility was that political attitudes (e.g., desire for greater i ndividual autonomy, a reduction in the arbitrary power of state, etc.) would be shaping economic attitudes. In this scen ario, the preference for a market economy would be considered as an important component of new demo cratic regime. In any case, depending if the main

PAGE 108

98 To a large extent this debate started with Finifter and Mickiewicz (1992) contending that, among citizens of the Soviet Union, attitudes toward the locus of responsibility (i.e., individual responsibility versus state responsibility for the individual well-being), and attitudes toward political reform represented two different dimensions, thereby permitting political and economic attitudes to move in different directions. Being the case that those who were supporting political change and democratization were also supporting the idea that the staterather than individuals themselvesshould be responsible for the well-being of the citizenry. Interestingly, these authors concluded that despite all transformations in the Soviet Union, the socialist view of the role of the state remained a strong component of its citizens believes, even among those who were more strongly supporting the political reform. It is important to note that few years, and when Finifter (1996) was re-analyzing her empirical original data, she was still confident that among citizens from the former Soviet Union, it was possible to have a large majority of citizens holding both a widespread commitment to political democracy, and a strong support for social and economic guarantees provided by the state. Therefore, according to her, we should expect that the eventual shape of th e new political systems in these countries will likely resemble egalitarian social democracy, democratic socialism, or market socialism than the more individualistic capitalistic democracy found in the United States. As we could expect, several scholars challenged the possibility of a lack of correspondence between support for the political reform (i.e., democracy) and support for the economic reform (i.e., market economy). For example, Miller, Hesli and Reisinger concern was either the success of th e political reforms or the success of the economic reforms, it was clear that some particular forms of relationship between political and economic mass attitudes would be more beneficial than others.

PAGE 109

99 (1994) claimed that they found the opposite among citizens from the former Soviet Union, that ordinary citizens attitudes on political and economic reform were interconnected and consistent. More specifically they contend that their empirical evidence was showing that individuals whom preferred democratic political reform also favored a decreased role for the state as a guarantor of economic well-being. In other words, being a political reformist implied the abandonment of the socialist system in favor of an increased individual responsibility regarding the welfare of the citizens. By the same token, in trying to explain the relationship between popular support for democracy and support for a market economy among Russians and Ukrainians, Gibson (1996) contend that the dynamic of this relationship was heavily influenced by the transition period that that came after the collapse of communism, particularly between 1990 and 1992. Accordingly, he mentioned that political attitudes (i.e., support for democratic institutions and processes) were stable between 1990 and 1992, but this was not the case with economic attitudes (i.e., support for a market-based institutions and processes) in part because for the lack of experience with the operation of markets in 1990, and in part because commitment to market institutions were less strongly ideological and more dependent on perceptions of current and future economic conditions. However, he added that by 1992 there was a significant shift in the structure of ordinary citizens attitudes (i.e., a crystallization process of mass attitudes, probably as a consequence of people acquiring more experience with democracy and a market

PAGE 110

100 economy), with democrats coming to support a market economy and anti-democrats evolving into opposition to markets.13 One fundamental problem with Gibsons argument (1996) is the lack of consistency between affirming that by 1992 democrats, particularly in Russia, supported market economy and that political and economic reforms were becoming synonymous; and, concluding later that attitudes toward the market and toward democracy are only moderately interrelated; and that this interrelationship is not overwhelmingly strong because many democrats oppose market principles and many marketers oppose fullfledged democracy. Regarding the contradiction within Gibsons argument, I believe that the main explanation for this is that ordinary citizens attitudes toward the political economic reforms in former Soviet countries can hardly be described either as fundamentally disconnected (e.g., Finifter and Mickiewicz 1992, Finifter 1996) or as highly correlated (Miller, Hesli and Reisinger 1994; Gibson 1996). Accordingly, a large number of scholars and empirical research have shown that this relationship is particularly complex, and that these two sets of attitudes are only strongly related. Duch (1993), for example, found some correlation between support for both political and economic institutional reforms among former Soviet Union citizens. However, it was also the case that support for democratic reforms were significantly higher than support for free-market reforms. The main explanation for this trend was related to the fact that a large part of this population was still holding strong beliefs about social egalitarianism. In other others, thes e citizens embraced democracy but continued to 13 See Evans and Whitefield (1998) for an interesting explanation about how the definition of left-right orientations, as well as the connection between ec onomic and political preferen ces and specific social groups changed in post-Soviet Russia during 1993 and 1996.

PAGE 111

101 be reticent about market economy. Therefore, according to Duch, the social and economic model for the former Soviet Union will more likely resemble that of the social democracies of Western Europe, which tend to champion egalitarianism, rather than the more laissez-faire model found in the United States for example. In a similar way, McIntosh et al. (1994) addressed the distribution of public opinion in Bulgaria, Hungary, Poland and Rumania in terms of: (1) reform enthusiasts (those who support both market reforms and liberal democracy), (2) reform skeptics (those whose oppose both market reforms and liberal democracy), (3) democratic enthusiast / market skeptics, and (4) democratic skeptics / market enthusiasts. Interestingly, these scholars found that those who support both market reforms and liberal democracy constituted only a minority in all these countries. Similarly, those tending to support market economy but view liberal democracy skeptically also constituted also a minority. Those who supported liberal democracy but were skeptical about market reforms constituted the larger group in all these countries. Overall, there was some evidence that people who support either democracy or a market economy tend to support the other as well; nevertheless, it was also the case that a significant percentage of these publics hold mixed views. Finally, Mason (1995) recognized that within several post-communist countries14 there was an important consensus against socialism and in favor of a free-market economy. However, once we get away from these ideological and general terms (e.g., democracy or market economy), this consensus disappears. He contends that when respondents were asked more specific questions, they tended to support important policies and values associated with the socialist regime that they had left behind. These socialist principles include both support for a strong role of the government in the 14 Russia, Poland, Slovenia, Bulgaria, Hungary, Czechoslovakia, Estonia, and East Germany.

PAGE 112

102 economy, and a strong sense of social and economic egalitarianism (i.e., the government should provide a job for everyone who wants one, the government should place an upper limit on the amount of money a person can make, and rewards should be distributed on the basis of need as much as merit). As a consequence of this, Mason concluded that, at least at the initial years after the political and economic transition, most people in post-communist countries were committed to the democratic aspects of these reforms, but at the same time many of these citizens favored the development of an economic model with a strong role for the state (i.e., something closer to the Swedish or German model rather than the United States model). Finally, he pointed out we should expect some negative impact in terms of democratic legitimacy and stability among post-communist countries if governments were not perceived as responsive to these particular citizens preferences. This more balanced and complex view about the relationship between ordinary citizens political and economic attitudes within post-communist countries was to a considerable extent reinforced by some works that found an important and massive attitudinal change in these countries between 1992 and 1995. Miller, Reisinger and Hesli (1996) contend that during this period, Russian and Ukrainian preferences shifted sharply in a pro-socialist direction (i.e., state responsibility for well-being). According to these scholars, these changes regarding the locus of responsibility were related to three different processes: a rising nationalism, the political re-emergence of communists, and a growing disillusionment with democratic reformers. More specifically, nationalist sentiments were correlated with preferences for a greater state responsibility for citizens economic well-being, as well as less positive attitudes toward West and Western ideas.

PAGE 113

103 By the same token, those who were relatively more positive toward communists were also more favorable toward state responsibility for social well-being. Finally, it was also the case that the difficult economic circumstances associated with the transition to a market economy created a growing disillusionment with the policies of reformers, particularly regarding economic reforms, and to lesser extent regarding political reforms. Hence, facing now this new empirical evidence, these scholars concluded that political and economic views among post-Soviet citizen s were consistently, but not strongly related. Therefore, even when this debate has not been particularly conclusive about the relationship between economic and political support within post-communist countries, it has provide very important insights for the understanding of the type of democracy and democratic regimes that a large majority of post-communist citizens was expecting. In this direction, there is important evidence revealing that in the context of these societies an attitude of political support for democracy does not necessarily entail similar support for a market economy. As several scholars have acknowledged (Duch 1993, McIntosh et al. 1994, Mason 1995, Mason and Kluegel 2000), the majoritarian support for political reform and democracy among post-communist citizens coexist with a strong level of skepticism toward market principles, and with strong statist and egalitarian orientations. This is particularly relevant for mainly two reasons. First, because it means that the preference for democracy as a form of political regime does not preclude the existence of different citizens preferences and expectati ons regarding the most desired social and economic model. Second, because it could be the case that the type of social and economic model associate with a specific democratic regime and the inability of this

PAGE 114

104 political regime to incorporate alternative preferences regarding the social and economic model can affect in a negative the level and nature of popular support for this democratic regime. In conclusion, it seems safe to affirm that within some post-communist countries particularly former Soviet Union countriesa large majority of the population understand and define democracy in a way that sharply contrast with the more procedural definition of democracy that characteri zed citizens from developed countries.15 It is also the case that the way ordinary citizens from some post-Soviet countries define democracy, and the lack of consistency or cons traint in regards to their ideas and attitudes toward democracy will have critical implications in terms of popular support for democracy. On the one hand, the acceptance or rejection of democratic procedures and institutionsand eventually the acceptance or rejection of their current democratic regimeby citizens from some post-communist countries will depend to a large extent on the performance of these regimes in providing specific outcomes such as rule of law and economic prosperity. More specifically, future support for democracy will depend basically on the ability of these democratic regimes to satisfy these citizens expectations (Fleron and Ahl 1998:305). On the other hand, as Miller, Hesli and Reisinger (1995, 1997) suggest, the fact that a majority Russians and Ukrainians scores high in terms of a general or abstract preference for democracy, as well as in terms of some democratic values, is not necessarily guarantee of a strong and stable commitment with their democratic regimes and the main democratic procedures. And this is the case because 15 Given that something similar have been identified in the case of Latin American countries (Camp 2001, 2003), it seems reasonable to contend that non-procedur al definitions of democr acy can be found in both typical developing countries (i.e., Latin American c ountries), and countries with a previous communist experience and experimenting acute processes of economic, social, and political dislocation (i.e., former Soviet countries).

PAGE 115

105 these democratic ideas and attitudes do not show a strong level of consistency, coherence or constraint (i.e., if they are not an integr al part of an enduring, firmly held set of fundamental beliefs regarding the centrality of key democratic procedures). Therefore, it could be the case that they could easily shift depending of the social, economic and even political circumstances; or, that these citizens could eventually tolerate and even in some cases support some non-democratic alternatives. Measuring and explaining popular suppo rt for democracy in post-communist countries As part of the research on popular support for democracy among post-communist countries, trying to measure and explaining the current levels of this support has been one of the topics that have received a great deal of attentionparticularly at the beginning of the political and economic transition and throughout the 1990s.16 More specifically, this particular concern has addressed two main questions: (1) to what extent ordinary citizens from post-communist countries support democracy?, and (2) what are the most important determinants of popular support for democracy in the context of post-communist countries; or, in other words, what is the kind of distribution of this popular support for democracy that exist in these societies once we consider key demographic, social, economic, and political variables? In this direction, Rose, Mishler, and Haerpfer (1998) aimed to test the Churchills hypothesis17 in the context of new democracies within post-communist societies. 16 It is important to note that among scholars dealing with post-communist countries, it was a consensus that popular support for democracy was going to play a critical role in the democratization prospects of these countries. 17 In 1947, when Winston Churchill made the case fo r democracy, he affirmed that many forms of government have been tried and will be tried in the wo rld and no one pretends that democracy is perfect or all wise. Indeed, it is possible to consider that democracy is the worst for of government, except all those other forms that have been tried from time to time (Rose, Mishler, and Haerpfer 1998:11).

PAGE 116

106 Testing this hypothesis meant to address to what extent citizens from post-communist countries considered that the new democratic regime was better in comparison to other possible types of political regimes, including non-democratic alternatives. Accordingly, using data from the third New Democracies Barometer gathered between 1993 and 1994,18 they first measured the level of support for the current democratic regime as well as the level of support for some non-democratic alternatives (i.e., the return to communist rule, a government rule by the army, a strong leader that can get rid of the parliament so he can decide things quickly, the return of the monarchy, decisions about the economy made by experts and not by the government and parliament). Then, they tried to find the main determinants of these popular support attitudes. Overall, these scholars concluded that the main obstacles to complete democracy in Central and Eastern Europe come not from the demand side (i.e., popular support for democratic and non-democratic alternatives), but from the supply side (i.e., the shortcoming in what is supplied by the political elites). Accordingly, they found a considerable level of popular support for the current democratic regime, and also a considerable level of popular rejection of non-democratic alternatives. As important as these findings about the level of popular support for the current democratic regime and for the non-democratic alternatives, were the discussion and conclusions provided by Rose, Mishler, and Haerpfer (1998) regarding the main determinants of these ordinary citizens attitudes and orientations. According to these scholars, social structure variables (e.g., age, education, gender, urbanization, religion, and national identity) had very little influence on attitudes toward the new regime and 18 The third-wave of the NDB included the following post-communist countries: Bulgaria, Czech Republic, Hungary, Poland, Romania, Slovakia, Slovenia, Belarus and Ukraine.

PAGE 117

107 non-democratic alternatives. Therefore, they conclude that there was almost no evidence to support any theories about the structural determinism of political attitudes, which was good news for the prospects of democratization of these countries. Moreover, they found that political perceptions and assessments explained far more these levels of support for the current democr atic regime and non-democratic alternatives. Specifically, they found that the dislike of the communist regime was the strongest influence on the rejection of non-democratic alternatives; and that people who saw the new democratic regime as betterwhether in terms of freedom or openness to influence government decisionswere more likely to reject non-democratic alternatives and to support the current regime. These findings allo wed Rose, Mishler, and Haerpfer (1998) to argue for the importance of avoiding what they called economic reductionism or the idea that all political attitudes are simply derivative of economic conditions. Being the case that an assessment of political conditions under the previous communist and nondemocratic regime (i.e., the lack of a key political good such as freedom) was clearly a more powerful influence on the level support for the current democratic regime in comparison to standard sociotropic and pocketbook economic perceptions, and even some current economic indicators such as the level of unemployment, deprivation, and income. It is worth noting that regarding the role of economic variables, Rose, Mishler, and Haerpfer (1998) found that individuals percep tions of the economic situation had more influence on regime support than material conditions like income, deprivation, or employment. Furthermore, peoples perceptions of macroeconomic conditions were more important than the way in which they perceived their own finances. However, it is

PAGE 118

108 important to keep in mind that political perceptions and political evaluationseven after controlling by economic perceptionsremain as the most important influence on support for the current democratic regime and for the rejection of non-democratic alternatives. The interesting point here, as these authors highlighted, is that a group of postcommunist countries is revealing a situation where the lack of a positive assessment regarding the economic and, to some extent, political performance of the current democratic institutions was not a critical factor affecting the levels of popular support for the democratic regime. On the contrary, what we are seeing is that a very negative perception about the past communist regime and the perception that the current democratic regime provided much higher levels of political freedomin comparison to the communist regimewere enough to produce a high level of support for the current democratic regime, as well as an important level of rejection toward non-democratic alternatives. Nevertheless, according to Rose, Mishler, and Haerpfer (1998), we could expect that sooner or later current performance can become dominant either because popular exhaustion of patience with the new regime or because people start to take political freedom for granted and start to evaluate the new democratic regime in terms of its overall political and economic performance. Therefore, it is clear that Rose, Mishler, and Haerpfer (1998) were providing a very positive and somewhat optimistic view about popular support for the current democratic regimes within post-communist countries. However, there are three issues that deserve further attention. First, Rose, Mishler, and Haerpfer (1998) empirical data do not include Russia, which is a notable absence. As such, it is problematic to attempt any generalizations about the phenomenon of popular support for democracy in the context of

PAGE 119

109 post-communist countries if we are not considering the Russian case.19 Second, according to Rose, Mishler, and Haerpfer (1998) analysis, main socio and demographic variablesi.e., structural variableswere almost irrelevant as determinants of popular support for democracy in some post-communist societies. However, there are several works that challenge these findings and contend that in terms of the distribution of support for democracy, socio and demographic variables play an important role among post-communist countries (Gibson and Duch 1993; Reisinger, Miller, Hesli, and Maher 1994; McIntosh et al. 1994; Reisinger, Miller and Hesli 1995).20 Regarding popular support for democracy in Russia, Gibson, Duch and Tedin (1992) concluded that at least during the first years of the political and economic transition, Soviet political culture posed few obstacles to democratization in the Soviet Union. More specifically, upon analyzing a public opinion survey conducted in the Moscow Oblast between February and March 1990, they found that many ordinary Muscovites embraced a variety of democratic values: widespread support for the institution of competitive elections, opportunities for dissent, and general norms of democracy such as equal protection under the law. In addition to this, most respondents had a relatively high level of rights consciousness, and they seemed to value liberty strongly even in the face of possible disruptions. Nevertheless, they also found that 19 As several scholars have pointed out there is an important level of diversity in terms of the postcommunist experience, and this is particularly true for the case of Central and Eastern European former communist countrieson one hand, and the case for the countries that emerged from the former Soviet Unionon the other hand(Rutland 2004, Bunce 2003). 20 A similar argument could be made regarding the importance of considering more contemporary public opinion data about popular support for democracy within post-communist countries in Central and Eastern Europe. Unfortunately, even for the case of former Soviet Union countries, it seems that in the last years this type of research has lost centrality am ong scholars dealing with post-communist countries.

PAGE 120

110 support for independent media was more mixed, and that the levels of political tolerance were low. Using a public opinion survey conducted in 1990, and representative of the entire European portion of the Soviet Union, Gibs on and Duch (1993) extended their previous research. The main goal in this new work was to pay attention at the degree to which ordinary citizens from the European Soviet Union value the rights, liberties, and institutions of democratic government, and how these values were distributed across this larger population (72). Consistent with their previous wor k, Gibson and Duch (1993:87) discovered several elements in the political culture of the European portion of the Soviet Union that were, in their opinion, conducive to democracy. Mainly, fairly broad support for competitive electoral structures, and a willingness to claim a variety of rights associated with democratic citizenship (e.g., personal safety, freedom of conscience, equality before law, property, freedom of speech and privacy, etc.). Nevertheless, these scholars also discovered among these citizens a set of more mixed attitudes toward other democratic values such as the support for a fully independent media, the willingness to embrace liberty if its cost is social disruption, and political tolerance. Particularly, regarding political tolerance, it was clear that Soviets were not willing to allow their most hated enemies the right to compete for political power. Gibson and Duch (1993) interpreted these findings contending that there seemed to be widespread support for democracy when it was defined in terms of majoritarian institutions and processes, but less support when it was defined in terms of the rights and guarantees that could make possible the political participation and inclusion of political

PAGE 121

111 minorities. In other words, in a context where so many Soviet citizensfrom the European part of the Soviet Unionwere supporting at least some democratic values, it was the case that they were embracing the easier ones (i.e., majority rule sort of values), while the much more difficult ones (i.e., minority right values) were much less assimilated. In terms of the distribution of these democratic values, Gibson and Duch (1993) contend that two variables stand out as the strongest predictors of democratic attitudes: age and education. Those with higher education and younger people were significantly more likely to hold democratic values. Unfort unately, rather than confronting that support for democracy in the form of democratic values were concentrated in some specific segments of the population, these authors preferred to concluded that these findings particularly the fact that younger people tended to support democratic valuesmeant an advantage for a democratization process. It is important to note that Gibson and Duch (1993) were somewhat more cautious about the presence of a democratic political culture among citizens from the European part of the Soviet Union in comparison to th eir previous articles (Gibson, Duch and Tedin 1992). Nevertheless, even in the presence of some problematic trends regarding the distribution and consistency of a set of democratic values, they preferred to sustain a more optimistic assessment about the emergence of a democratic political culture among Soviet citizens. It is critical to recognize that the two works previously mentioned (Gibson, Duch and Tedin 1992; and Gibson and Duch 1993), were pa rt of a first-wave of survey research

PAGE 122

112 into political values and norms in Russia.21 As Fleron and Ahl (1998) contend, these pioneering studies mainly examined support for a variety of norms thought to be associated with a political culture supportive of democratic institutions. The list of these democratic norms or values was particularly extensive (i.e., political efficacy, political trust, political interest, support for a multiparty system, political knowledge, political tolerance, valuation of liberty, support for the norms of democracy, rights consciousness, support for dissent and opposition, support for an independent media, acceptability of non-conventional forms of political expression, priority of speech over public order, etc.). Not less relevant is the fact that all these scholars were particularly impressed by the generally high levels of popular acceptance of democratic norms expressed by Soviet, and particularly, Russian respondents. Moreover, what came after this first-wave of research about democratic values in Soviet and former Soviet countrieswith a particular emphasis on Russiahas been less consensual, less certain, and to some extent less optimistic. For example, some scholars like Fleron and Ahl (1998:301) have contended that levels of commitment to democratic norms in Russia were not only more modest than originally thought, but that they have declined in the late 1990s. By the same token, they have stated that the evidence on commitment to democracy is mainly characterized by both conflicting results across different surveys, and varying levels of support depending of the specific democratic norm or value. Similarly, McFaul (2000) believes that a normative commitment to the democratic process on part of both the elite and society is still not present in Russia. 21 Finifter and Mickiewicz (1992); Hahn (1993); and Reisinger, Miller, Hes li, and Maher (1994) can be also considered as part of this fi rst-wave of survey research.

PAGE 123

113 However, on the contrary, some other scholars have challenged the contention that support for democratic regime has been waning in Russia during the last years. Gibson (2001), for examplein analyzing data from a three-wave panel survey of ordinary Russians collected between 1996 and 2000have concluded that Russians, at the end of the 1990s, have a moderate support for democratic institutions, and only a weak support for market-based institutions and processes. Equality important is that both set of attitudes have not changed substantially between 1996 and 2000; therefore, there is little evidence of substantial erosion of support for the new political and economic system in Russia. According to him, these findings reveal that perceptions of economic malaisethat characterized Russians citizens during a great part of the 1990shave not undermined support for democracy.22 To a large extent, this is the case because support for democracy in Russia is not contingent upon the perception that the economy is performing adequately. It is also the case that those who support democratic institutions and processes tend also to support a market economy, although the overall association between these two pro-reform attitudes is not very strong. In light of these findings, Gibson (2001:123) ends his work saying that the nascent democratic culture in Russia has not eroded over the course of the last part of the 1990s, and that Russians may not be very supportive of market economy, but very few prefer an alternative to democratic governance. Also, he mentions that this mass public orientation 22 Gibson (2001) operationalizes popular support for democr acy using a extensive number of variables that includes support for independent and pluralistic medi a; a relative evaluation of order and liberty; the importance of respecting rights and liberties such as freedom of speech, associa tion, religion, equality before the law, protest, private privacy, etc.; support for dissent; support for the rule of law; support for a multi-party system with competitive elections; political tolerance; etc. However, although he mentions several times that he was measuring support for democr atic institutions and processes, it is important to note that a considerable part of these variables are indeed measuring typical de mocratic or civic values.

PAGE 124

114 does not guarantee the consolidation of democracy in Russia, but rather reduces its chances of failing. Overall, regarding the most current level of popular support for democracy in Russia, Fleron and Ahl (1998) seem to be right when they mention that the end of the 1990s scholars are, in general, less optimistic about popular commitment toward democracy in comparison to the beginning of the 1990s. However, their contention that, at the end of the 1990s there has been a considerable decline in popular support for democracy among Russian citizens seems problematic. Mainly because others scholars analyzing public opinion data from the mid and late 1990s have found some decline in popular support for a market economy, but no similar trend regarding popular support for democracy (Gibson 2001). The last part of this section aims to cover part of the debate about how to operationalize and measure popular support for democracy in post-communist countries, and how different ways of operationalizing and measuring popular support for democracy could lead to different and, in some cases, contradictory results. Nevertheless, for now, I believe it is important to note that when popular support for democracy in Russia is defined as the level of commitment toward some specific democratic values (e.g., Gibson 2001) the situation does not seem problematic for the development of democracy. On the contrary, when popular support for democracy is addressed as the level of commitment toward specific democratic procedures, and even in terms of the electoral support for non-democratic leaders, the situation seems less clear and more problematic for the development of democracy (McFaul 2000).

PAGE 125

115 In general, there is no doubt that the inclusion of Russia in the study of popular support for democracy within post-communist countries improves significantly our understanding of this political phenomenon. In that direction, and although there is a lack of comparative studies between post-communi st countries from Central and Eastern European and former Soviet Union countries, it seems that popular support for democracy is more widely distributed and more strong in the case of the former countries. Regarding this point, it is also th e case that socio-demographic and structural variables (e.g., education, age, and location) seem to be much important as determinants of popular support for democracy within former Soviet Union countries. Being the case that several scholars have found that popular support for democracy, for example in the case of Russia and Ukraine, is particularly concentrated in specific segments of the society. Not less important is the fact that research about popular support for democracy in Russia has demonstrated that different ways of operationalizing and measuring this kind of popular support can generate different results. For example, it is the case that the operationalization of support for democracy in terms of individuals attitudes toward general democratic values (e.g., support for independent and pluralistic media, support for dissent, political tolerance, etc) has tended to produce a more optimistic assessments about the level of democratic commitment among Russians citizens, as well as about prospects of future democratization of this countries. A more detailed consideration regarding the debate about the main determinants of popular support for democracy within post-communist countries, and about the different alternatives that can be used to

PAGE 126

116 define, operationalize, and measure this kind of political support is what follows in the remaining of this section. As I mentioned at the beginning of this section, the second main topic regarding popular support for democracy within post-communist countries has been related to the main demographic, social, political and economic determinants of this type of popular support; or, in other words, how this support for democracy is distributed among different social groups within post-communist societies. In that direction, and as I have mentioned before, Rose, Mishler, and Haerpfer (1998) have dismissed the relevance of social structure variables (e.g., age, education, gender, urbanization, religion, and national identity) in explaining popular support for current democratic regimes and nondemocratic alternatives. Nevertheless, it is the case that several other scholars have challenged this contention. For example, Reisinger et al. (1994)analyzing data from public opinions surveys conducted in Lithuania, Ukraine and the European Russia during 1990, 1991 and 1992 found that political and democratic values were concentrated in certain societal groups. Specificallyin all these three countriesurban residents, male, younger, better educated, less religious, wealthier, and more politically active people were the most supportive of a set of democratic values.23 They interpreted these findings as evidence that, in the case of three these post-Soviet so cieties, the main societal transformations associated with the process of modernization24 have indeed meant an important processes 23 This set of democratic values included: interpersonal trust; agreement with the statement that party competition strengthens the political system; agreement with the statement that anyone has the right to organize opposition or resistance to any governmental initiative; pref erence for materialist or postmaterialist values; and an index of support for individual rights such as speech, association, and conscience. 24 According to the modernization, and thinking in the context of post-communist countries, we should expect that variables such as age, education, income, location (rural vs urban), and even gender appeared

PAGE 127

117 of values change and values differentiation. Therefore, they suggest the importance of paying attention to how the social and demographic cleavages produced by modernization were shaping citizens belief systems, and, more specifically, the distribution of democratic values. Accordingly, in a second article Reisinger, Miller, and Hesli (1995) addressed the differences between Russian urban and rural population in terms of political outlooks. Unfortunately, this time the main outcome variable was the support for economic reform and not the support for democratic values. Nevertheless, their main findings seemed to reinforce the role of socio and demographic variables in explaining individuals economic and political attitudes. More specifically, they found that along with some rural-urban differences in terms of support for market reformsbeing the rural population less supportive of these reforms, there were also another key determinants of these individuals attitudes such as a sense of economic insecurity (i.e., level of perceived threat regarding unemployment, inflation, shortage of consumer goods, and income inequality), the perception about the family finances over the past year, occupation (i.e., entrepreneurial vs. non-entrepreneurial status), age, level of education, and degree of attention to the news. Interestingly, in controlling the effect of the ruralurban cleavage for all these other significant variables, they obtain that the effect of the rural-urban cleavage remain significant. Reisinger, Miller, and Hesli (1995) interpreted these findings contending that it is not the case that the rural-urban differences are unimportant. Nevertheless, what it seem to be more relevant is that these rural-urban differences are producedto a large as important determinants of opposition to socialis m, support for democracy, and a preference for an opportunities society or a market ec onomy (Miller, Reisinger, and Hesli 1996).

PAGE 128

118 extentby the fact that the distribution of some critical socio-demographic variables differs in important ways between the population living in cities and the population living in the countryside. Moreover, they proposed that a political economy perspective has an important explanatory power to account for individuals attitudes and beliefs differences. Specifically in this case, the individuals place in society (concrete or perceived) shapes the support for market reform. Therefore, this explains why the differences regarding the sense of economic insecurity, the situation of family finances, occupation, and level of education appeared as the strongest determinant of support for market reform. Overall, one of the main contributions of both articles (Reisinger et al. 1994 and Reisinger, Miller, and Hesli 1995) was to show how much explanatory power could be gained by attending to identifiable aspects of individuals place in society, either concrete or perceived. In this case, it was clear that citizens from the former Soviet Unionindeed like people in any societywere forming their opinions and choosing their behaviors, to a considerable extent, on the basis of what they were seeing as their main threats and opportunities, and what realistic potential they had for succeeding in different courses of action. Indeed, because different individuals start from different places in society, the way they change their outlook and behaviors in reaction to institutional and policy change will not be identical. Just as it was reasonable for some Russianseven some rural Russiansto support Western-style market reforms and democratic reforms, so it was reasonable for others to oppose them (Reisinger, Miller, and Hesli 1995:1040). In a similar direction, McIntosh et al. (1994) provided interesting insights regarding the main determinants of support for democracy and for a market economy in Bulgaria,

PAGE 129

119 Hungary, Poland and Rumania. Regarding the main explanation of these political and economic attitudes, they found that that those who support both market reforms and liberal democracy constituted only a minority in all these countries; however, they were basically younger, better educated, male and politically interested. By the same token, they found that those who oppose both political and economic reforms also constituted a minority; however, they tended to be older, less educated, and less-politically active. Additionally, these scholars found also that both symbolic politics25 variables (e.g., preferred society in terms of individual or government responsibility in providing what people need; choosing between freedom and equality; acceptance of change and new ideas; interpersonal trust; etc.) and self-interest variables (e.g., perceptions about the current economic situation, the retrospective economic situation, and the prospective economic situation) play an important role in shaping support for liberal democratic and market principles. In accounting for all these empirical findings, McIntosh et al. (1995) concluded that it seemed the case that those better prepared socially, psychologically and demographically to win in the new system or who are already winning economically or politically tend to support both liberal democratic principles and free markets.26 Nevertheless, there was also important evidence showing that symbolic attitudes (i.e., deep-seated values) were, at least, as important as self-interest variables in explaining support for democracy and for a market economy. 25 Although the authors do not use the term political culture I think that symbolic politics is really close to the definition of political culture as deep-seated values. 26 Duch (1993) provides a similar conclusion regarding support for ec onomic reform among citizens from the European part of the Soviet Union. For him, it was clear that younger people and well-educated people were more supportive of economic reform because they were in the position to get important benefits from these reforms.

PAGE 130

120 After revising the debate about the main determinants of popular support for democracy in post-communist countries, I believe that one of the main conclusion is that dismissing either the role of individuals political and economic assessments about the performance of the current and previous political regime or the role of more long-term socio-demographic variables renders it difficult to account for the dynamic of popular support for democracy in new democracies. In other words, in trying to explain the dynamic of some key political and economic attitudes (i.e., support for democracy or support for a market economy)particularly in the context of new democracieswe should be prepared to deal with multiple and different influences, ranging from normative and ideological issues to sociotropic and pocketbook considerations, and including also the effect of structural conditions such as level of education, location, age, and gender. Finally, as important as accounting for all these different determinants will be to account for the relationships among them (e.g., to what extent normative and ideological factors, and economic and political assessment of the performance of the current democratic regime are related to the particular concerns and needs of different segments of the population). The last issue that I want to address in this section is what the literature about popular support for democracy in post-communist countries tells us regarding the challenge of operationalizing and measuring this complex phenomenon. In this direction, it is important to mention that this literature also contains an important debate regarding this issue. For example, some scholars have been very critical about some of the standard ways that have been frequently used to measure this support for democracy (Rose and Mishler 1994; McIntosh et al. 1994; Mishler and Rose 1995; Rose, Mishler, and Haerpfer

PAGE 131

121 1998; Mishler and Rose 2001). An important part of this argument is that in new democracies, democracy is but one possible outcome. Therefore, in a context of competition between alternative regimesparticularly between democratic and nondemocratic alternatives, popular support will play a critical role because will be one of the most important forces shaping the behavior of the political elites. In that direction, and thinking in the persistence of a democratic regime, democracy only needs to be judged as the better of all the possible altern atives. Accordingly, citizen assessment of the nature and performance of the current democratic regime will be what matters for the maintenance and, eventually, the consolidation of new democracies (Rose, Mishler, and Haerpfer 1998). Given this particularity of new democraci es, Rose, Mishler, and Haerpfer (1998) have contended that some strategies frequently used to measure popular support for democracy are misleading. Specifically, asking ordinary citizens about their attitude toward democracy in general or in abstract terms presents the problem that the word democracy means different things for different people, and, equally important, that most of these meanings are positive. Therefore, when someone says that he or she supports democracy we cannot be sure what type of democracy is been supported. By the same token, we cannot be sure to what extent this positive general attitude toward democracy is related to the current democratic regime. Similar problems can be found anytime that some scholars use the political culture or civic culture approach (Almond and Verba 1963) to measure the proportion of individuals in a society endorsing the values and beliefs that are a priori deemed as components of a democratic political culture. As Rose, Mishler, and Haerpfer (1998:97

PAGE 132

122 98) pointed out: (1) we are still not clear if whether a democratic or civic culture is a cause or a consequence of democracy, (2) we are frequently facing empirical evidence revealing that people in new democracies have a mix of values (i.e., most of them positive toward democracy, but also some more negative and even in some cases a considerable level of indifference toward democracy), (3) it is the case that an exclusive focus on democratic or civic political values leaves out the issue about the support for competing undemocratic alternatives, and (4) it is also the case that there is no consensus about which questions best measure the commitment to the values and beliefs of a democratic or civic political culture. Moreover, Rose, Mishler, and Haerpfer (1998) have also pointed out that the recognition that the endorsement of abstract democratic ideals is not sufficient to establish a democracy has led some scholars to measure popular support for democracy in terms of the level of satisfaction or dissa tisfaction with democracy and/or the current democratic regime. The main problem hereas with the previous strategiesis that we cannot be sure about (a) what it is behind a high or low level of satisfaction or dissatisfaction, (b) what constitutes a satisfactory democracy, and (c) how exactly these levels of satisfaction or dissatisfaction are related to the overall support for the current democratic regime Given all these considerations, Rose, Mishler, and Haerpfer (1998:104) have advocated the use of a realist approach to measure popular support for democracy and its alternatives in new democracies. The realist approach meets four criteria: (1) it must refer to a real political regime rather than an abstract ideal; (2) it must invite a judgment on the real and current regime as a whole, with all its imperfections, rather than ask piecemeal

PAGE 133

123 questions; (3) it must compare between democratic and undemocratic alternatives; and (4) it should allow the expression of degrees of satisfaction and dissatisfaction because alternative regimes can be greater or lesser evils.27 This alternative seeks the evaluation of the current regime as a whole and the use of comparisons avoids the limbo of subjectivism in the satisfaction with democracy approach. Finally, offers the possibility to know to what extent the current democratic regime is seen as the lesser evil (i.e., a regime that is bad but that however is preferred because all other alternatives are worse). It is important to note that according to Rose, Mishler, and Haerpfer (1998) this realist approach should be complemented with a measure of popular support for some undemocratic alternatives (e.g., the return of the communist rule, a military government, strong leaders that do not need parliament and elections, a monarchy, etc.) as a way to know better what different regime alternatives are competing for popular support. Mishler and Rose (2001:303) have expanded and refined their previous arguments about the problems contain in different approaches used to measure popular support for democracy. In that direction, they affirm that the idealist satisfaction and cultural approaches used to measure citizen support for democracy do not capture the realities that citizens in new democracies or in transitional regimes moving from an undemocratic past to an uncertain future are facing. Therefore, as a consequence of this, these approaches mismeasure citizen political support for democracy in a fundamental way. More specifically, citizens from new de mocraciesgiven their lack of experience with democracymay not have a clear understanding of democracy and democratic 27 This realist approach has been used in the New Democracies Barometer and it is based in the following set of questions: Here is a scale for ranking how the government works. The top, +100, is best; the bottom, -100, is the worst. Where you would place: (a) the fo rmer communist regime, (b ) our current system of governing with free elections and many parties, and (c) our system of governing in five years.

PAGE 134

124 ideals, but they certainly have strong feelings about the performance of the new regime, especially in comparison to the past. Again, in the cases of new or incomplete democracies, what it is critical for the sustainability of a democratic course during the transition period is a sufficient level of popular support for the current and new regime.28 Specifically, regarding the idealist approaches (i.e., measuring citizen commitment to democracy as an abstract ideal), Mishler and Rose (2001:305) affirm that even if citizens in incomplete regimes have meaningful and positive attitudes about democracy, their commitment to democratic ideals may be of a little help for the survival of the new democratic regime. For example, a citizen who embraces democracy as an ideal but who opposes the current regime because it falls short of its standard would be wrongly classified as supporting the current democratic regime. Therefore, very easily the idealist approach can measure incorrectly the popular support for these transitional democratic regimes. Regarding the satisfaction approach (i.e., emphasizing popular assessments of democratic performance), Mishler and Rose (2001:305) contend that while this approach calls attention to the actual performance of a regime, individuals responses always remains ambiguous regarding the level support for the current democratic regime. In addition to this, this approach assumes that the current regime is democratic and that citizens view it this way. However, neither of these assumptions holds in new or incomplete democracies. Regarding cultural approaches (i.e., measuring support for democracy through an indirect way), Mishler and Rose (2001:306) points out that they begin with the 28 Assuming of course that we are considering a democratic regime that has possibilities to advance in its level of democratization and not a more stable semidemocraticor a semiauthoritarianregime.

PAGE 135

125 assumption that democracy requires a civic culture in which citizens manifest such basic values as tolerance and trust. Citizens are surveyed to determine how widely those attributes are distributed across society. The assumption is that societies with low levels of trust and tolerance are poorly suited to the establishment of democratic institutions. Increasingly, however, researchers disagree over the causal direction of the relationship between democratic culture and institutions. For established democracies, this poses few problems given the stability of institutions and culture, but in transitional societies the question of whether political attitudes are l eading or lagging indicators of democracy raises serious problems. I want to stress here that these criticisms aimed toward idealist and cultural approaches toward popular support for democracy is highly consistent with my previous observation regarding the fact that there is evidence within some post-communist countries as well as other new democracies showing that when this support for democracy is operationalize and measure as the level of peoples commitment toward some generaland in some cases very abstractdemocratic values, it tend to be less problematic because it seems that a considerable part of the population embrace these democratic values. However, once we consider more specific questions about democratic procedures or about the performance of the current democratic regime, it is less clear that there is strong popular consensus in favor of democracy. Finally, regarding the realist approach, Mishler and Rose (2001) emphasize that rather than evaluating incomplete democracies against abstract and ambiguous democratic ideals, this approach assumes that citizens of transitional regimes are better able to assess the current regime against the performance of the other regimes with which

PAGE 136

126 they have first-hand experience. While citizens may have little knowledge of democratic principles, they have a lifetime of experience with undemocratic regimes. At the start of a new regime the natural tendency is to evaluate the new regime by comparison with the regime it has replaced. Citizens may develop the capacity for more abstract and nuance evaluations as they acquire more experience with democracy, but at the beginning of the transition, they will tend basically to compare the performance of the new and the old political regime, and the will judge the new democratic regimes as better or worse in comparison to the previous non-democratic regimes. As important as the discussion about the possible shortcomings related to an idealist, satisfaction and cultural approach used to measure popular support for democracy in new democracies is the empirical evidence presented by Mishler and Rose (2001). Using data from the World Values Survey 1995, these scholars found very similar levels of idealist support for abstract democracyand even similar levels of trust in political institutionsamong stable democracies, new democracies, transitional regimes, and even stable non-democratic regimes.29 However, they found also that the level of a realist support for current regimes was lower in comparison to the idealist support for democracy across all different types of regime. More interestingly, they found also that the level of a realist support for democracy was not substantially and consistently higher for stable democracies than for stable non-democracies, but also higher in comparison to new democracies and transitional regimes. Therefore, they interpret all these findings as strong evidence that the realist support for democracy 29 Mishler and Rose (2001) define new democracies are those regimes which Freedom House codes as free and which have experienced a major regime cha nge (i.e., a shift toward democracy) over the past twenty years. On the contrary, th ey define transitional regimes are those regimes which also have experienced a major regime change, but are currently coded by Freedom House as only partly free or unfree.

PAGE 137

127 approach has greater validity in comparison to the idealist one because it distinguishes much more clearlyand in ways more consistent with theorydifferent levels of popular support for democracy among stable democracies, new democracies, transitional regimes and stable non-democratic regimes. In reacting to these all these contention about the strengths and weaknesses of the different approaches toward the measurement of popular support for democracy, I am not entirely convinced about the convenience of assuming that a realist approachas it has been defined by Rose and Mishleris the only valid way to assess popular support for democracy in the context of new democracies. Rather, I believe that its main contribution is to alert us about the problems and limits of the idealist, satisfaction, and cultural approaches. In a similar vein, this idealist approach is important because it claims for the need of paying close attention to the level of popular support for the current democratic regimes as well as toward non-democratic alternatives. In this direction, it is clear that the success or failure of the democratic experiences in post-communistbut also in developing countrieswill depend on the possibility of maintaining these democratic regimes. And for the maintenance of these democratic regimes, probably one of the most important factors will be a considerable of popular support for the current democratic institutions, rules, and procedures. In any case, it is clear that we need to develop more comprehensive and rigorous ways of operationalizing and measuring popular support for democracy in the context of new democracies. In that direction, I believe that a very promising alternative is to combine realist with idealist, satisfaction, a nd cultural approaches. This combination will allow us, for example, to understand how far or close the support for the current regime is

PAGE 138

128 from a conviction about the need to advance some democratic procedures and norms. Similarly, this combination can allow us to understand to what extent the assessment of the current democratic regime is influenced by particular ideas about how the society and the economy should be organized. Latin American Countries If we consider the number of works about political culture and popular support for democracy within Latin American countries that have been published in the last years, as well as the number of survey research projects that have provided empirical evidence for all these works (e.g., Latinobarmetro, the Latin American Public Opinion Project, the Hewlett Project) we can affirm that this topic has gained an unusual importance in the field of Latin American politics. All this is remarkably because until very recently social research on Latin American political culture has been quite fragmentary (Turner 1995, Power and Clark 2001). Equally important is the fact that this new research trend have meant an important break point in regards to some previous works about the political culture in Latin America that, without any systematic analysis of empirical evidence, stated that it was essentially traditional and authoritarian one, and inimical to the development of liberal democracy (Wiarda 1974, 1992; Dealy 1974, 1977, 1992; Harrison 1985). In this direction for example, Dealy (1974:73) ha d asserted that Latin Americans do not understand the very term of democracy in the conventional Western sense of political pluralism, representation, and competing interests but as the centralization and control of potentially competing interests as an attempt to eliminate competition among different interests. By the same token, Wiarda (1974:274) had argued that the US-style of democracy is probably ill suited to the nations of Iberia and Latin America because these

PAGE 139

129 societies are Catholic, corporate, stratified, authoritarian, hierarchical, patrimonialist and semi-feudal. It is clear that this view about political culture in Latin America relies in a very traditionalistic view of political culture. Trying to uncover the political culture or an entire societyor a region in this case, reducing it to one or two fundamental categories (i.e., either democratic or authoritarian), and assuming that the values that constitute a political culture hardly will change with the pass of the time. It is also clear, that this view has been challenged by the most recent experience of Latin American societies adopting and trying to develop democratic regimes, and, specifically, by the cases that are showing important signs of democratic consolidation (e.g., Costa Rica and Uruguay). Therefore, rather than assuming an almost total incompatibility between political culture and democratic development in Latin America, the most recent research literature on this topic has been aimed to understand the dimensions of Latin Americans political culture that can support democratic norms and institutions (Turner 1995). In that direction, for some scholars (Diamond, Linz, a nd Lipset 1989) there is no doubt that there has been, in the last years, a widespread acceptance of democratic values among elites and ordinary citizens in Latin America. Also, it is important to note, that all these more recent efforts have been characterized by the use of survey research which has made possible to test empirically different theories and hypotheses.30 30 Although it is true that the literature on political cu lture and democracy among La tin American countries has gained some centrality in the la st year, it is also the case that th e literature research on popular support for democracy among post-communist countries has not only a considerable advantage in the number of works that have been published, but also in terms of theoretical and methodological sophistication.

PAGE 140

130 Nevertheless, along with all these more positive features regarding the treatment of the relationship between political culture and popular support for democracy within Latin American countries, there have been some important limitations. In terms of research design, it is the case that only a very small sub-set of countries (i.e., Costa Rica, Nicaragua, Mexico, Ecuador, and Bolivia) has concentrated almost all the attention of scholars working within this field. By the same token, it has been also the case that the majority of these efforts have been a one-country study, being the exception two or morecountries studies (Seligson and Booth 1993, Booth and Seligson 1994). Similarly important, is the lack of any advanced statistical analysis of the survey research data collected in the last years (Lagos 1997, 2001, 2003a, 2003b). Moreover, there have been important limitations in the way the relationship between political culture and popular support for democracy has been conceptualized in the majority of these academic works. Very frequently, they have been mainly focus in trying to measure to what extent there is a civic or democratic political culturemostly in the way Almond and Verba (1963) defined itwithin Latin American countries that can generate popular support for democracy for both democracy and current democratic regimes. As a consequence of this, I believe that the relationship between ordinary citizens political attitudes and beliefs, and democracy has been reduced to the consideration of very specific topics such as the level of interpersonal trust (Lagos 1997, 2001, 2003; Power and Clark 2001), support for the right of political participation (Booth and Seligson 1994; Lagos 1997, 2001, 2003a, 2003b), tolerance (Booth and Seligson 1994, Seligson 2001), and respect for social norms and the rule of law (Lagos 1997,

PAGE 141

131 2001, 2003a, 2003b; Power and Clark 2001; Seligson 2001). All these topics can be considered as specific components of the civic or democratic political culture approach.31 In my opinion, the main problem with the civic culture framework is that it can become easily a very normative approach. Accordingly, it is possible to contend that democracymore specifically, democratic institutions, rules, and proceduresis having so many problems in Latin America because the majority of people of this region lack a democratic or civic political culture. Therefore, the main challenge for the development of democracy in this region is the need of transforming the non-democratic political cultures of its habitants following basically the path of the the civic culture suggested by Almond and Verba (i.e., interpersonal trust, a sense of internal and external political efficacy, etc.). As Burbano de Lara (2003) suggests, in order to understand the relationship between political culture and democratic governance in Latin America, the key should not be to measure how far or close peoples political attitudes are from a model of civic culture or any ideal democratic culture. Instead, the main goal should be to address the rationality, causes and consequences of these political attitudes. Accordingly, some Ecuadorian scholars have devoted a considerable amount of efforts in the last years working on the phenomenon of populism as a key component of their political system (de la Torre 2000a, 2000b; Ramrez 2003; Xavier Andrade 2003; Xavier Andrade 2003). Something similar with the topic of clientelism has occurred in Peru (Parodi 1993, Murakami 2000). In any case, what it is important to note here is that this way of approaching the issue of ordinary citizens political attitudes within Latin American countries does not preclude the possibility of considering that some of them 31 Moreno and Mendez (2002:353) have de fined this democratic political culture as a set of values and attitudes that are compatible with democratic principles and practices, such as tolerance, interpersonal trust, emphasis on civil liberties and rights, politi cal participation, support for democracy.

PAGE 142

132 (e.g., populism and clientelism) can constitute an obstacle for the stability and development of democratic regimes. To a considerable extent, the main findings and conclusions about the civic or democratic political culture in Latin America are no less problematic. First of all, some of the civic culture elements (e.g., social tolerance) that were supposed to have a strong and positive impact on popular support for democracy failed to be significant (Seligson 2001). In a similar vein, even when several scholars have found a modest positive impact of interpersonal trust on the preference for a democratic regime, this effect is very small. For example, the most significant determinant of the high level of preference for democracy found among Costa Ricans is not the level of interpersonal trust or any socioeconomic or demographic characteristic, but th e fact of being Costa Rican instead of being Chileans or Mexicans (Power and Clark 2001, Seligson 2001). Accordingly, country effects seem to be much more relevant than interpersonal trust. Second, even the contention that the level of civic or democratic political culture can be considered as the main factor to explain the level of democratization has questioned in the context of different Latin American countries. For example, Muller and Seligson (1994) founded that almost all civic culture variables suggested by Inglehart (1990) as relevant for democracy (i.e., interpersonal trust, life satisfaction, and support for a revolutionary change) failed to have a significant effect on the levels of political rights and civil liberties. On the contrary, interpersonal trust appeared to be a product of a democratic experience rather than a cause of it. Moreover, the single most important determinant of the level of democratization was not an attitude of the general public but rather a macroeconomic variable: income inequality (i.e., the size of the income share of

PAGE 143

133 the richest quintile of households). Higher levels of income inequality tend to create a feeling of threat among elites regarding the possibility of lower-class rebellion or revolution. Therefore, military coups to quell mass rebellion and preserve elite privileges are likely to occur in countries with a very inegalitarian distribution of income (647). I think that all these criticisms were not necessarily aim to reject the role that a democratic or civic political can play in terms of democratization of Latin American countries, but to show that this type of political culture is only one among other key factors that can explain the establishment and development of democracy in this region, and that a useful political culture approach required to go beyond the traditional civic or democratic culture approach. In this direction, most contemporary research about peoples democratic attitudes and perceptions in Latin America seems to be less attached to the traditional civic or democratic culture approach. For example, Camp (2001) addresses directly to what extent citizens definitions of democracy in some Latin American countries (i.e., Costa Rica, Chile, and Mexico) resembles the procedural definition of democracy that is widely used among scholars. According to him, it is clear that most Latin Americans do not conceptualize democracy in the same way as do theorists and North Americans citizens. Similarly, the vast majority of Latin Americans do not have the same expectations from democracy, as do their North American counter parts. Being possible that some type of relationship exists between how citizens conceptualize democracy and what they expect from democracy as a political model. Regarding both the definition and expectations towards democracy, what most distinguishes the Latin American version is an emphasis on social and economic equality

PAGE 144

134 and progress (9). Nevertheless, it is interesting to note also that there is not a clear consensus about what democracy means for Latin American citizens. In that direction, the definition of democracy in Costa Rica is closer to the procedural definition of democracy and to the way people in the United States define it. On the contrary, Mexicans and Chileans view democracy in more social and economic terms, not in political terms (15). Camp (2001:23) concluded his work aski ng for further efforts at understanding citizens conceptualizations of democracy in Latin America, and suggesting the possibility that social inequality, social injustice, and poverty can be molding Latin Americans views of democracy. Therefore, it could be the case that peoples perception of the degree of inequality in their societies can constitute a powerful determinant in explaining individuals political orientations. Similarly, Moreno (2001) provides another case of research about political culture in Latin America that is less attached to the civic culture framework. He was particularly interested in understandingi.e., finding the main determinantsof (a) popular support for democracy and (b) ordinary citizens meanings of democracy. Regarding the issue of support for democracy and using data from the World Value Survey, Moreno (2001:35) found that the higher the income level and the education level, the more pro-democratic the individual was. Moreover, the gap between the highest and the lowest income level was significantly greater that the gap between the highest and lowest education levels. In other words, income seemed more important than education in explaining the variance in support for democracy. The strongest differences in support for democracy based on class differences (income, education, and occupation) led him to suggest the need of revising

PAGE 145

135 several works on Latin American political culture thatperhaps with strong empirical limitationshave minimized the impact of class in shaping attitudinal support for democratic values and institutions (37). Regarding ordinary citizens meanings of democracy, Moreno (2001:29) stated that democracy might be associatedon the one handwith procedures such as free and fair election and abstract terms such as liberty or the extension of political rights to minorities (i.e., the minimalist or procedural definition of democracy); oron the other hand might be associated with concrete goals such as fighting crime or redistributing wealth. One of the problems with the latter definition, he suggested, is that it includes expectations toward democracy that are not exclusive of democratic rule. Specifically, he found, first, that individuals who are more educated, younger, and more informedin terms of news receptionare more likely to associate democracy with liberty. On the contrary, the lower the level of political knowledge, the more likely the individual will be to say that fighting crime is the main task of democracy.32 Finally, it is important to note that according to Moreno, Mexicans and Chileans place less importance on liberty as a defining feature of democracy than Costa Ricans do, and more emphasis in wealth distribution. Summarizing the contributions of this most contemporary research trend about the relationship between political culture and popular support for democracy, I certainly believe that we are now in a better position to address the relationship between popular support for democracy and democratization within Latin American countries. Peoples political perceptions and attitudes are not anymore approached following the civic culture 32 Unfortunately, Moreno does not explore more the relationship betw een political sophistication, measured as political knowledge, and class.

PAGE 146

136 framework (i.e., measuring what was considered as deep-seated beliefs and values such as the level of interpersonal trust or social and political tolerance), and, as a consequence of that, they are less treated as an external or independent phenomenon shaping the development of democracy in this region. More specifically, these most recent works on political attitudes and perceptions in Latin America are less oriented to trying to reveal the lack of a civic or democratic culture within Latin America societies, and much more interested in addressing the causes of a popular disenchantment with the current democratic regimes and institutions in Latin American. By working in this direction, we can start considering that political values, perceptions, and attitudes are both inde pendent and dependent variables. That, on one hand, they can constitute an obstacle for the maintenance and development of Latin American democratic regimes; but that, on the other hand, they are also an outcome of specific factors and processes such as class differences, different expectations toward government and democracy, and the accumulated frustration with the very poor economic and political performance of the current democratic governments. By the same token, this new research trend is showing a much higher level of methodological sophistication regarding the operationalization of the independent and dependent variables related to popular support for democracy. For example, some scholars have started to unpack the general notion of support for democracy making a distinction between support for democracy as an ideal type of political system and support for the current democratic regime (Seligson 2002). Similarly, some scholars have started to pay attention to the support for non-democratic alternativese.g., military governments(Seligson and Carrin 2002).

PAGE 147

137 I would like to end this section mentioning something about the current situation in terms of popular support for democracy in Latin America. First of all, it is worth nothing that there has been an important decrease in the level of optimism regarding the commitment and spread of support for democracy and democratic institutions and procedures among ordinary citizens in this region. Until mid-1990s was common to hear that despite the lack of a civic culture within Latin American countries, this region presented a strong level of democratic legitimacy or support (Lagos 1997). At the end of the 1990s and the beginning of the 2000s, we have started to hear about an important erosion of support for democracy (Lagos 2001, 2003). Very frequently, the main explanation for this erosion lies in the accumulated frustration with the performance of current democratic governments.33 A similar concern about the nature and level of popular support for democracy in Latin America can be found in the UNDPs report about democracy in this region (2004). Analyzing the data from the 2002 Latinobarmetro, this report finds that in the context of Latin American countries a large number of individuals that manifest a preference for democracy as the best way of government can have also strong non-democratic attitudes such as the tolerance for an authoritarian government capable of solving the economic problems of the country, and the perception that democracy can work without parliament and political parties. Therefore, this report alert us that the preference for democracy in Latin Americaat least measured in the standard waydoes not imply a strong support 33 It is interesting to note that even is the case of Costa Ricathe most stable a nd consolidated democratic regime in the regionSeligson (2002) has also found evidence of some erosion in the level of mass system support. According to him, some of the most important consequences of this trend have been an important increase in electoral abstentionism and the support for minor parties. Nevertheless, Costa Ricans seem following the trend that has been observed among citizens from established democracies. They are basically dissatisfied with the way the democratic regime is performing, but they are still holding a high level of support for their democratic regime (Seligson 2002, 2001).

PAGE 148

138 for democratic principles and institutions. Additionally, this report makes clear that these non-democratic attitudes are more likely to be found within: (a) people with lower levels of education, (b) people considering that government is not addressing their main problems, and (c) people thinking that they have not yet improved their living conditions in comparison to their parents, and very pessimistic about the future living conditions of their children. As a consequence of this, it not a surprise that support for democratic procedures and institutions is higher among Latin American countries with lower levels of social and economic inequality.

PAGE 149

139 CHAPTER 5 THE 1992 DEMOCRATIC BREAKDOWN IN PERU AND THE ROLE OF POPULAR SUPPORT FOR DEMOCRACY This chapter deals with the Peruvian political experience in the last decades, in particular with the 1992 democratic breakdown and the role that popular support for democracy or democratic legitimacy played in regards to this outcome. In a first section, I will present a very brief description of the main political events in Peru between 1980 and 2001. In the second section, I will present and discuss some of the main explanations of this democratic breakdown. Finally, the last section of this chapter is aimed to make a balance about the role that a decline in citizen support or legitimacy toward some of the main democratic institutions and procedures had on this democratic step. It is critical to start noting that when Fujimori took office on 28 July 1990, Peru was a democratic regime; although, a troubled one that showed signs of crisis and even deterioration. Therefore, the 5 April 1992 autogolpe meant the breakdown of a democratic regime. Specifically, what happened was a coup by the president, with the decisive support of the armed forces, against the other branches of government, particularly the legislature, the judicial system, and the party system. These successesthe coup and the democratic breakdownwere complex events that do not respond to a single cause. To a large extent what happened in Peru in 1992 was the unfortunate convergence of different factors and trends, mainly an acute economic crisis, an internal armed conflict between the state and some guerilla and terrorist movements, the deterioration of states capacities to regulate the society, a

PAGE 150

140 president who lacked a majority in both chambers, and a collapsed party system. There are, therefore, several different plausibl e explanations for the 1992 Peruvian breakdown. Nevertheless, the majority of these explanations can be classified in two main categories: (a) institutional oriented approaches,1 and (b) civil society oriented approaches.2 Without neglecting the relevance and contribution of the institutional interpretations, this chapter contends that the scrutiny of the role of the civil society in the 1992 democratic breakdown in Peru provides a more comprehensive explanation of these political events. I am aware that someone can question the relevance of analyzing these events because it is hardly likely to observe something similar in the Perus near future or in other Latin American countries. Indeed, within the Latin American region, only current Venezuela resembles some important similarities with the Peruvian experience (McCoy and Myers 2004); and other countries, very similar to Peru in political, social, economic and cultural terms (e.g., Bolivia and Ecuador), seem to be transiting a somewhat different path.3 There are in my opinion, however, several ways to challenge this argument. First, democratic breakdownsincluding ones with popular consenthave happened elsewhere and we cannot rule out the possibility that they might happen in the near future of other new democracies. Additionally, there is no doubt that democratic breakdowns 1 See, for example, Kenney (2004) and Aguayo (2004). 2 The concept of civil soci ety oriented approaches includes both th e role of political and economic elites and the role of ordinary citizens. See, for example, Burt (2004), Degregori (2000), Cotler and Grompone (2000), and Mauceri (1997). 3 Considering that there are many similarities among Peru, Bolivia and Ecuador, why a clear democratic breakdown has only happened in the case of Peru is a cr itical question for all scholars interested in these Andean countries. This dissertation w ill not provide an answer to thus que stion. Nevertheless, in addressing specifically the Peruvian case and the role of popular support for democracy in this chapter and in Chapter 6, I expect to make some contributions for a further discussion in that direction.

PAGE 151

141 constitute a very important component of the study of democratization in both current established and new democracies,4 and, therefore, every opportunity we have to understand democratic breakdown, especially with citizen consent, is an opportunity we need to take advantage of and learn from. Second, as I mentioned in Chapter 1, Peru clearly belongs to one very particular group within new democratic regimes. This group includes several new democratic regimes that are showing an important degree of democratic stagnation after the installation of formal democratic regimes. These new democratic regimes are also characterized by having serious deficits in terms of governance and the quality of democratic governance,5 and by having high levels of popular democratic dissatisfaction. If it is true that not all these new democratic regimes have suffered open democratic regressions, it is also true that the Peruvian case reveals, in a dramatic way, how the lack of popular support for democracy play an important role in these processes of democratic stagnation, and even regressions. Finally, I am convinced that addressing the 1992 Peruvian democratic breakdown is critical to understand Peruvian politics. Accordingly, I believe that the significance of these political events does not lie only in the combination of some unfortunate short-term circumstances, but in the way they interact with some more long-standing features of the Peruvian political system and the Peruvian society. Overall, these more long-standing factors are to a large extent responsible for a democratic regime of very low democratic quality and for the lack of an important level of support for democracy among ordinary 4 Among current established democracies we have the case of Germany between 1933 and 1994 (Childers 1983); and, among current new democracies in Latin America we have Brazil in 1964, Chile and Uruguay in 1973, and Argentina in 1976 (Bermeo 2003). 5 See Levine (2006) about the common governan ce deficits within developing countries.

PAGE 152

142 Peruvian citizens.6 Before presenting and discussing some of the main explanations of the Peruvian democratic breakdown in 1992, the next section provides a brief description of the main political events in this country between 1980 and the 2000. Peru 1980 Peru is one case of authoritarian reversal within the last wave of democratization across developing countries. Indeed, it can be considered an exceptional case for understanding the erosion of legitimacy and support for democratic institutions, rules and procedures. To have a better idea of this particular case, I will present below an overall view of its political experience during the last two decades. After a military government that lasted twelve years (1968), Peru held a democratic election in 1980. The winner of that election, Fernando Belande of Accin Popular, a center-right party, obtained 46% of the valid votes and defeated several more progressive parties of the center-left and left. In the next election in 1985 Alan Garca of the center-left APRA party won 53% of the popular vote and his party obtained a majority bloc in the legislature (51% of valid votes for the Senate; 50% of valid votes for the House of Deputies). Nevertheless, neither of these two governments ever proved successful in addressing either Perus acute economic crisis of the 1980s or the growing political violence unleashed mainly by the Shining Path during the same decade. In retrospect, it is now clear that these dual failures by administrations of both the right and 6 It is important to note that some of the conditions that were part of the election of Fujimori in 1990 are also present in the current electoral process in Peru that it is going to conclude in April 2006 with the election of a president and members of the Peruvian congress. As in 1990, public opinion polls are revealing that a considerable majority of most economic and social disadvantaged Peruvians are supporting a candidate (Ollanta Humala) that represent a radical criticism to the more traditional and institutional political groups and parties, to the effectiveness of the main democratic institu tions, and the mainstream neoliberal discourse that seems to be the one that dominated any discussion about public policy in Peru (Aragn 2006).

PAGE 153

143 left endangered Perus democracy, shook public faith in the democratic system, and ultimately contributed to the fall of democracy in Peru in 1992. Perus economic crisis, like similar crises elsewhere in Latin America in the 1980s and 1990s was characterized by a lack of econom ic development, deterioration in real salaries, rising unemployment and the loss of jobs and high levels of inflation to the point of hyperinflation. This crisis reached its peak at the end of the Garca administration in 1990. By that year Perus economic situation was critical. Perus GNP had shrunk by 25% in the last two years, inflation had reached an annual level of 7,600%, and poverty engulfed more than half the population.7 As in the case of other episodes of democratic breakdown, people in Peru also felt that the government was the main responsible of the economic crisis and for the peoples suffering (Childers 1983). Nevertheless, while Perus economic crisis was repeated in many other Latin American nations at the time, the crisis of political violence and guerrilla terrorism was more unusual. The Shining Path with another, lesser-known guerrilla movement, the Movimiento Revolucionario Tpac Amaru challenged and defied the democratic state and demanded a radical transformation of the democratic political order. These groups used terrorism and violence in rural and urban areas to strike fear into the population, engender economic chaos, shake the foundations of the polity, and create a generalized atmosphere of instability, panic, and a willingness to resort to extreme measures to regain safety and stability. According to the final report of the Commission of Truth and Reconciliation (2003) the internal armed conflict between 1980 and 2000 constituted the most intensive, 7 Several scholars have provided a detailed description of the Peruvian economic, social, and economic situation at the end of the 1990s. See for example, Palmer (1996), Dietz (1998), Cameron and Mauceri (1997), and Tanaka (1998).

PAGE 154

144 extensive, and prolonged episode of violence in the entire history of the Peruvian republic. In that direction, this report has estimated that the most probable figure of victims who died is 69,280 individuals. Moreover, this report has stated that there was a significant relationship between poverty and social exclusion and the probability of being a victim. More than 40% of the reported8 deaths and disappearances to this Commission were concentrated in the Andean department of Ayacucho, and these victims taken together with those documented in the departments of Junn, Hunuco, Huancavelica, Apurmac and San Martn add up to 85% of the victims registered by this commission. Similarly, almost 80% of the victims lived in rural areas and 56% were engaged in farming or livestock activities. By the same token, 75% of these victims spoke, as their mother tongue, Quechua or other native language (Table 5-1). Finally, the report concluded that this conflict demonstrated serious limitations of the State in its capacity to guarantee public order and security, as well as the fundamental rights of its citizens within a framework of democratic action. Table 5-2 provides additional evidence regarding the dual nature of the crisis that hit Peru at the end of the 1980s and early 1990s. When asked about the main problems of the country at that time, a large majority of citizens in Metropolitan Lima mentioned both problems related with the economic crisis (i.e., unemployment, inflation, cost of living, and the economic crisis itself) and problems related with the political violence crisis (i.e., terrorism and human rights violations). It is worth noting that at the end of the 1990s, two key problems (i.e., inflation and terrorism) were not perceived anymore as the most 8 The Commission of Truth and Rec onciliation in Peru received direct reports of approximately 24,000 people death or disappeared, and in 18,397 cases it was possible to identify the victims by his or her complete names and last names. This number is smaller in comparison to the estimation of the total victims (69,280 individuals) because not all victims of the armed conflict were directly reported to this Commission.

PAGE 155

145 salient problems in the country. As it will be addressed later, Fujimori obtained most of the credit for this change. Table 5-1. Number of victims reported to the Commission of Truth and Reconciliation Victims' socio-demographic characteristics Gender Total 22,468 Female 4,524 Male 17,944 Age Total 14,350 0 582 10 2,046 20 3,939 30 3,189 40 2,325 50 1,323 60 654 70more 292 Level of education Total 14,119 None 2,904 Kindergarten 50 Primary 6,756 Secondary 3,115 Superior 1,294 Mother tongue Total 15,177 Quechua 11,376 Spanish 3,593 Other native languages 208 Occupation Total 13,470 Peasants 6,897 Local authorities 2,359 Housewives 910 Salesman/saleswoman and tradesman/tradeswoman 899 Independent workers 625 Teachers 381 College students and technical superior students 324 Employees 322 Members of armed and police forces 217 Workers 193 Professionals and intellectuals 78 Other 265

PAGE 156

146 Table 5-1. Continued Victimss sociodemographic characteristics Place of Birth Total 14,881 Ayacucho 6,899 Hunuco 1,956 Junn 1,471 Huancavelica 1,323 Apurmac 832 San Martn 537 Lima-Callao 342 Puno 308 Cusco 182 Ucayali 96 Other 935 Total number of victims 22,507 Note: Totals for each variable differ from the general total as a consequence of missing information for some cases. Table 5-2. Main problems in the country (1990) 1990 (%) 1991 (%) 1992 (%) 1993 (%) 1994 (%) 1995 (%) 1996 (%) 1997 (%) 1998 (%) 1999 (%) Economic problems Unemployment 51 46 39 44 48 67 57 57 66 74 Inflation 81a 11 14 11 10 6 9 11 10 17 Cost of living 17 21 24 22 12 12 17 14 28 Economic crisis 58 Political violence problems Terrorism 55 40 49 33 15 12 4 17 4 4 Human rights violations 8 12 16 11 15 17 37 15 Social problems Poverty 21 21 45 54 53 54 51 36 38 Quality of education 14 15 21 14 22 13 12 7 12 13 Drugs consumption 15 14 10 25 14 15 24 18 10 24 Drug trafficking 15 9 5 10 11 19 15 7 Corruption 15 16 31 19 20 29 29 37 29 33 a Includes both mentions to inflation and cost of living as one of the main problems in the country. Note: All data were collected in April, with the exception of 1990 data that were collect in June. Apoyo Opinin y Mercado: Informe de Opinin 1990.

PAGE 157

147 As Garcas popularly elected government finished its term in 1990, Peru had reached a peak of economic crisis and political fear. Peruvians had become angry with government for these crises and the continued inability to deal with them effectively. Indeed, citizens had become disillusioned with democratic institutions and the entire political party system for its apparent failings and for its perceived contribution to the crises (Diamond and Plattner 1996, Tanaka 1998). It is in this context that a political outsider9 such as Alberto Fujimori was elected president in 1990 (Cotler 1994). He, himself, lacked a political party and instead ran on a personalist ticket based on a kind of social movement, Cambio 90. Rather than offering a party-based platform and future agenda, he used a personalistic, charismatic appeal and very vague general promises to solve the political crisis and stabilize the economy without using shock economic therapy. The electorate resonated with this kind of appeal. He appeared to represent new hope and a departure from the traditional parties and institutions that seemed connected with crisis and disaster. After a first round where no candidate got a majority of the votes, Fujimori won 62% of the votes in the second round. However, his Cambio 90 only obtained 22% in the Senate and 17% in the House of Deputies. Although he had promised that his government would be different, Fujimori moved to implement the same economic package that had been advocated by the rightist, neoliberal party, Frente Democrtico (FREDEMO). In spite of his lack of domestic support, Fujimoris program received strong backing from the international financial community and from key countries like the United States and Japan (Cotler 1994). These 9 The definition of Fujimori as a political outsider is twofold. First, he represented a political alternative coming from outside the traditional party system. S econd, his political identity was based on an open confrontation with the traditional Peruvian political elites.

PAGE 158

148 initial financial reforms took effect and Fujimori began to have some economic success. He ended the hyperinflation and stabilized the economy. He then moved swiftly to use economic success for his own political goals. Fujimoris economic program and his popularity with the international community made him look like a rightist president pursuing a neoliberal agenda. As such, he shared much common ground with Perus rightist party, FREDEMO. A political alliance with FREDEMO appeared possible and logical and would have greatly increased Fujimoris ability to govern effectively through the use of the nations political institutions. Yet Fujimori deliberately chose to eschew any su ch alliance or cooperative position. He opted instead for a confrontational style and continued to assume a dismissive and derogatory stance toward the nations political parties, including the right, and toward the democratic institutions. In April 1992, with considerable public opinion support and some international support, Fujimori dissolved the Congress that had been elected with him, and suspended the 1979 constitution. He called this closure an autogolpe or coup against himself. However, far from striking out against his ow n government or himself, he was isolating himself and his government from all opposition and challenge, especially from any that might emerge through the legitimate democratic institutions of the Congress and the parties. Fujimori justified his closure of the Congress by saying that the legislators were blocking passage of the urgent reformist measures the nation needed. In particular, speed was needed to combat terrorism. A large majority of Peruvians were convinced by this rhetoric. As Table 5-3 shows, right after April 5 events and subsequent months, a large

PAGE 159

149 majority of Peruvians supported Fujimori and his decision of closing the Congress, intervening the judiciary, and suspending the constitutional order. It is important to note that at this point there were not signifi cant differences across socio-economic levels. Indeed, Fujimori and Peruvian citizens seemed to be in close accord. Table 5-3. Approval of Fujimori and Autogolpe (April-August 1992) Date Agency Universe Approve of Fujimori (%) Approve closure of Congress (%) Approve intervention in the judiciary (%) Approve rupture of constitutional order (%) 6-7 April APOYO Lima 71 89 7 April DATUM Lima 79 84 95 9-11 April CPI Peru 88 94 11 April APOYO Peru 79 80 11 April APOYO Lima 82 78 SES A 85 SES B 79 SES C 84 SES D 80 23 April APOYO Lima 81 70 7 June APOYO Lima 76 66 9 August APOYO Lima 62 60 SES A 61 SES B 61 SES C 57 SES D 62 Apoyo Opinin y Mercado: Informe de Opin in, April 1992, August 1992; DATUM and CPI: Ferrero Costa 1992, pp. 141; SES data: Pe r Econmico, February 1992, pp. 16 Note: SES A: Socio-economic Sector A = wealthi est 1.8% of Limas population and 3.2% of Lima families; SES B: Socio-economic Sector B = next wealthiest 10.1% of Limas population and 20.2% of Lima families; SES C: Socio-econom ic Sector C = next poorest 29% of Limas population and 40.3% of Lima families; SES D: Socio-economic Sector D = poorest 59% of Limas population and 36.3% of Lima families. Source: Kenney (2004:228). Drawing upon an impressive array of sources on democratic breakdown in Eastern and Western Europe and Southern Cone c ountries in Latin America (i.e., Brazil, Uruguay, Chile and Argentina), Bermeo (2003) has contended that average citizens do not and have not supported democratic breakdowns because in time of crisis, particularly economic crisis, the majority of citizens in these countries have gravitated toward the

PAGE 160

150 political center. On the contrary, when democracy has broken down, it has been because political elites deliberatively polarized among themselves such that political negotiation and compromise were not longer possible. Bermeo adds also to her argument that if it is true that, in general, ordinary citizens have not been the main cause of democratic breakdown, it is also true that they frequently failed to demonstrate their support for democracy. Some scholars have reacted to this too optimistic view regarding the role of ordinary citizens in processes of democratic breakdown. For example, Anderson (2005) has claimed that in some cases (e.g., Germany in 1933, and Argentina 1976) citizens have become extreme and have supported openly or tacitly anti-democratic elite behavior, contributing decisively to a democratic breakdown. I believe that this recent Peruvian experience is showing another possibility regarding the connection between ordinary citizens and democratic breakdown. Some democratic regimes can be breakdown not because ordinary citizens are openly supporting non-democratic leaders or because ordinary citizenssupporting more moderate alternativesare not able to articulate a defense of the democratic regime vis--vis some non-democratic alternatives. They can be also easily breakdown by some non-democratic leaders when there is a lack of considerable level of popular support for democracy; in other words, when citizens do not see any good reasons to defend and preserve their democratic regime or when they do not feel that they are going to lose something important if the democratic regime is suspended. As Levi (2006:5) points out good governments are those that are (1) representative and accountable to the population they are meant to serve, and (2) effectivethat is,

PAGE 161

151 capable of protecting the population from viol ence, ensuring security of property rights, and supplying other public goods that the populace needs and desires. The most representative and accountable governments are in democracies, but not all democracies have effective governments. In this regard, it is possible that, under some circumstances, people could perceive a non-democratic government as effective and even responsive. To a large extent this is what happened in Peru during the late 1980s and the early 1990s. A huge economic crises and a serious political violence crisis created not only a very high level of desperation among Peruvians citizens, but also a strong feeling that the current democratic authorities and institutions were offering little or almost nothing in terms of economic stability and security, two of the main things that governments are supposed to deliver in exchange for peoples quasi-voluntary compliance (Levi 2006:7 8). As a consequence of this, it is very likely that Peruvians citizens accepted the 1992 autogolpe as a short-term solution to preserve their society, and that they were looking the other way when they were confronted with a tension between democracy and governance. In that direction, I believe that Fujimori and his allies realized very soon the level of desperation among Peruvians and their profound level of frustration regarding the democratic authorities and institutions. Similarly, they calculated very well that no significant popular protest would occur as a consequence of a democratic breakdown and even future non-democratic actions. See Table 5-4 and 5-5 for the levels of Peruvian citizens confidence and approval regarding the main democratic institutions in Peru before April 1992.

PAGE 162

152 Table 5-4. Confidence in institutions in Peru (1989) March 1989 (%) September 1990 (%) March 1991 (%) September 1991 (%) March 1992 (%) September 1992 (%) Church 81 81 83 85 80 81 Armed Forces 42 58 53 47 54 57 Presidency 23 57 29 26 42 54 Legislature 20 45 19 19 17 Judiciary 20 23 16 22 14 28 Political parties 17 21 13 13 12 13 Apoyo Opinin y Mercado: Informe de Opinin 1989, 1990, 1991, and 1992. Source: Kenney (2004:234). Table 5-5. Approval of the president, legi slature, and judiciary in Lima (1987) President Legislature Judiciary APOYO (%) IMASEN (%) APOYO (%) IMASEN (%) APOYO (%) IMASEN (%) January 1987 26a 22a May 1990 5 August 1990 46 11b October 1990 53 44 17 15 December 1990 61 44 18 February 1991 35 33 28 15b May 1991 45 27 September 1991 32 47 21 21 October 1991 54 50 26 26 November 1991 58 55 23 19 18 December 1991 60 19 January 1992 65 71 22 February 1992 64 55 19 14b March 1992 53 58 15 April 1992c 59 17 12 Apoyo Opinin y Mercado: Informe de Opini n and IMASEN: Confidencial; various years. a Percentage rating performance of institutions as good or very good. b Percentage having a favorable image of institution. c Survey carried out before 5 April 1992 coup. Source: Kenney (2004:235). Unfortunately for the Peruvian democracy, soon after the closure of Congress, Fujimori scored several key successes in the battle against terrorism. The most notable of these events was the September 1992 capture, in Lima, of Abimael Guzmn and several

PAGE 163

153 other key leaders of the Shining Path guerrilla group. Soon after being detained, Guzmn made an overture to the Peruvian government, offering to end the Shining Path insurrection. These victories produced a feeling of relief among Peruvians and an even greater level of popularity for Fujimori. In September 1992, after Guzmns capture, Fujimoris approval rating in Lima rose to nearly 75% (Apoyo Opinin y Mercado 1992). This popularity continued through the first half of the 1990s and produced Fujimoris second overwhelming presidential victory in 1995. In contrast to the failures and inefficacy of the two previous democratic governments, Peruvians saw Fujimori as having delivered on his promises to solve the economic crisis and decrease terrorism. Citizens were considerably less concerned with Fujimoris anti-democratic behavior and measures; particularly Peruvians from the lower socio-economic levels. In the 2000 election the situation was even more complicated; there was a stronger opposition and some concern regarding the consequences of a third period in office. This situation forced the government toward a fraudulent election. Despite the fraudulent nature of the 2000 election, where Fujimori was again re-elected, it was clear that Fujimori still retained considerable electoral support, especially from the lower socio-economic levels.10 Nevertheless, at the beginning of his term period in office, the release of a video that showed one of the closest Fujimoris allied (Vladimiro Montesinos) in charge of the national intelligence office (SIN) passing a bribe to one of the popularly-elected 10 As Carrin points out (2006:126) () a defining feature of Alberto Fujimori was it sustained level of approval in public opinion polls. Although this support declined after 1997, it remained significantly high (in comparison with that for previous administrations) despite Fujimoris frequent and egregious abuses of power. More surprisingly, even afte r the public became highly critical of Fujimoris economic policy, the annual average approval rating during his second term never dropped below 40%. As Fujimori inaugurated his controversial third term in the wake of an el ectoral process marred by accu sations of fraud and heavyhanded tactics, he still enjoyed a popularity rating above 45%. See Carrin ( 2006:129) for data about comparative evolution of presidential approval rati ngs in Peru between 1980 and 2000, and about support for Fujimori by social class between 1993 and 2000.

PAGE 164

154 legislators from an opposition party to buy that legislators vote in favor of Fujimori programs in Congress was the beginning of a huge scandal of political corruption. In the following days it became clear the high level of government control of the judges, military and police forces, politicians, journalists, and even businessman. As a result of this scandal and to avoid facing the justice system Fujimori fled the country and resigned to the presidency. Subsequently Peru began a process of democratic transition, led by the Congress that culminated in a new general election in 2001. During his more than 10 years in power, Fujimori was successful in economic and terrorist policy; and he used thes e successes to reinforce not only a delegative presidential style (ODonnell 1994); but also a systematic demolition of the democratic procedures and institutions. The phenomenon that is even more striking for me was that during all this time, a large majority of Peruvian citizens was willing to look the other way with respect to these non-democratic behaviors as long as Fujimori was perceived as providing solutions to the countrys main problems. Again, this was particularly the case for most economic and social disadvantaged Peruvians. Both Peruvian11 and foreign12 scholars have extensively addressed this latest Peruvian political experience. As we could e xpect, they provide different interpretations regarding the main causes for the 1992 democratic breakdown. And different accounts regarding the role that popular support for democracy played in that process. The next section addresses some of these interpretations; particularly the ones that can be classified 11 See for example, Cotler (1994, 2000), Tuesta (1997, 1999), Tanaka (1998), Lynch (2000), Grompone (2000), Degregori (2000), Neyra (2001), Aguayo (2004), and Carrin (2006). 12 See for example, Conaghan (2005), Cameron (1994, 1997), Mauceri (1997), Cameron and Mauceri (1997), Burt (2004), Kenney (2004), and Weyland (2006).

PAGE 165

155 as civil society oriented interpretations. Nevertheless, I pay also to attention to some institutional oriented approaches. Explaining the 1992 Democratic Breakdown In general, democratic breakdowns are complex events; and the Peruvian case was not an exception. Additionally, in this particular case, it is clear that the democratic breakdown in Peru in 1992 was closely associated with the profound socio-economic and political violence crises that hit this country in the late 1980s and early 1990s. However, the main challenge is to understand why the Peruvian democratic regime and its main political actors and institutions were incapable of dealing with this critical juncture and to resist the non-democratic, anti-system and even authoritarian forces that questioned its relevance and maintenance. Following Cotlers analysis of Peru at that time (2000:22), we need to understand how in the search for governability, democracy and authoritarianism became antagonistic forces. Several scholars have dealt with the 1992 democratic breakdown in Peru. The majority of these explanations are based on the identification of a common set of factors or phenomena that played a crucial role in this development. In that direction, and besides the impact of the economic, social and political violence crises, it has been suggested that this democratic breakdown was particularly related to the collapse of the Peruvian state; the collapse of the party system and the rise of political outsiders; the weakening of the civil society, Fujimoris authoritarian leadership and the formation of an authoritarian coalition composed of himself, the armed forces, and some business sectors; and, the conflict between the executive and the legislative between 1990 and 1992. To a large extent, the key difference among all these scholars is the way they combine these different factors and phenomena in order to identify the main cause or

PAGE 166

156 causes of this democratic breakdown, and the dissimilar relevance that they attributed to each one of these particular factors or phenomena. For example, for some scholars what was particularly critical was the loss of democratic legitimacy associated with an economic and political crisis that revealed the collapse of both the Peruvian state and the limitations of the democratic regime inaugurated in 1980 (Mauceri 1997, Burt 2004). For others, the main cause is related to the political leadership of Fujimori and the collapse of the Peruvian party system (Cameron 1997, Cameron and Mauceri 1997, Tanaka 1998, Levitsky and Cameron 2003). Finally, there ar e other scholars that, without neglecting the intervention of several other explanations, contend that the main cause was the institutional design of the Peruvian democratic regime, more specifically the fact that Fujimoris first government was extremely weak in terms of legislative support (Kenney 2004).13 Therefore, most of the time, these different explanations are not mutually exclusive. Indeed, this is not a debate about irreconcila ble points of view, but about which particular factors or phenomena can be considered as the most important ones in explaining the breakdown of democracy in Peru in the 1990s. Accordingly, it is possible to find important disagreements, regarding the role that the popular support for democracy played as one of the main causes of the 1992 democratic breakdown.14 13 It is important to note that this institutional argument (i.e., the failure of presidential democracy) was originally developed by Linz (1994) and that Kenney ba sically applies it for Peru and other Latin American countries. 14 There are as well other interesting debates about the main cause or causes of the 1992 democratic breakdown in Peru that I am not going to address here. For example, some explanations tend to give more weight to structural vari ables (e.g., the economic and political crisis or the Peruvian institutional design) while others contend that the primordial explanation is the authoritarian leadership of Fujimori as his allies. About this particular debate, Cameron (1997) pr ovides an initial and very insightful discussion.

PAGE 167

157 For some scholars it is impossible to understand what happened in Peru in the 1990s without considering the effect of the weakening and partial collapse of Peruvian state institutions in the late 1980s, coupled with the debilitation of political and civil society. Indeed, it was the delegitimization of the Peruvian state which made possible the formation of new alliances among state elites, the armed forces, and the domestic business elitewith the implicit support of transnational capitalto reconstitute the Peruvian state along authoritarian and neoliberal lines in the 1900s (Burt 2004). According to this approach, Peru has traditionally been a weak state in which private elites have tended to dominate policy-making and where racial and socioeconomic structures have left the vast majority of the population on the margins of politics (Cotler 1978). However, by the late 1980s, the Peruvian state ceased, in many respects, to be a functioning state. Governments inability to manage the countrys economic crisis and rising labor conflict, pervasive government corruption, government paralysis in the face of a dramatic expansion of political violence, and in general the growing disengagement of the state from the society undermined public confidence in the states authority and in the governments ability to govern the country. The conjunction of all these elements constituted a growing crisis of legitimacy that finally undermined the very belief in the state as an institution that set the rules of the game for political and economic interaction. It was also the case that the crisis of the Peruvian state translated directly to a crisis of the traditional parties because citizens became increasingly disillusioned with the inability of them at the national and local level to alleviate the worst effects of the economic and political crisis (Burt 2004:251).

PAGE 168

158 Therefore, it was in the context of the Peruvian state collapse during the late 1980s and early 1990s that distinct social groupsfrom the bourgeoisie to the middle class to poor rural and urban votersdemonstrated their willingness to cede broad powers to Fujimori and his allies in a desperate search for the return to normalcy after several years of economic and political upheaval under the previous two democratic governments. On the other hand, the weakness of political and social society undermined the possibility of forging a more democratic model of state reconstitution. In this context, Fujimori was in the position or re-building state-society relations with little regard for democratic norms and procedures (Burt 2004:255). The subsequent legitimacy of Fujimoris regime (1992) can be to a large extent derived from its policy success. The social disarticulation resulting from the process of state breakdown exacerbated the tendency of average citizens to value concrete policy successes as opposed to more abstracts principles of democratic rule and procedures. Each policy success strengthened the authoritarian coalition and the publics tendency to measure political performance according to concrete outcomes (Burt 2004:263). In a similar note, Mauceri (1997:35) provides a very interesting argument about the erosion in the efficacy and legitimacy of democratic institutions and the breakdown of the Peruvian democratic regime in 1992. He traces the origins of all these problems to the process of democratic transition carried out in 1980. According to him, this transition represented a compromise that left many key political actors unsatisfied: business and economic elites disliked the inability to revise the economic reforms of the Velascos regime in the 1979 constitution, while popular sectors felt that the new institutions of

PAGE 169

159 democracy that emerged from the Constituent Assembly did not fully represent their interests. Therefore, not coincidentally, both of these sectors overwhelmingly supported Fujimoris overthrow of the constitutional democratic regime inaugurated in 1980. More specifically, the 1977 transition to democracy in Peru was an elitecontrolled process that basically represented the withdrawal of an exhausted and divided military from power and the compromise between the military and conservative groups on basic rules for the new civilian regime. Therefore, the political order that emerged from this transition had little input from other political forces. Moreover, part of that exclusion was voluntary because most groups of the left did not prioritize the creation of liberal democratic procedures and institutions because they defined the upcoming political struggle in terms of the need to restore and expand a process of social reforms in the country. At the same time, the inability of most conservative sectors to alter the structural reforms of the Velasco period in the new constitution created an additional source of dissatisfaction (Mauceri 1997:29). Therefore, the transition was successful as a mechanism to extricate the military from power, but the new political regime left both conservatives and radicals dissatisfied, and it was unable to transform the long authoritarian features of the Peruvian political system such as the minimal congressional authority vis--vis the executive, clientelism and corporatism, personalist party structures, and the persistence of a salient political role for the military (Mauceri 1997:3). Accordingly, between 1980 and 1992, we observed in Peru a process that failed to create an institutional framework to which all actors felt committed; and the final outcome was an intense process of under valuation of its institutional democracy.

PAGE 170

160 All of this during a period of radicalization and mobilization of the civil society particularly an the end of the 1970s and the beginning of the 1980sthat probably made even more evident the tension between popular democratic expectations and a restrictive political democracy; and the fact that the post-transition period did not provide political institutions and norms that could address the demands and pressures of these social actors. It is also important to note that according to Mauceri (1997) additional events made even more complicate the possibility of a further and gradual democratization of the Peruvian political system after 1980, and a consequent increase in the level of legitimacy regarding the democratic institutions. One of these events was the most profound socioeconomic crisis of the 20th century Latin America that seriously diminished the efficacy and capacity of the Peruvian public sector. Anot her of these events was the decline of the lower-class protests movements during the 1980s as a consequence of the economic crisis and the demobilization measures adopted by the military regime since 1977. The sharp increase in the cost of participating in protest, made very difficult the work of the organizations that were engaged in organizing the protests among these groups. Additionally, these leftist leaders turned their attention toward forming political alliances to participate in the coming electoral processes. In another explanation of the breakdown of democracy in Peru in 1992, Cameron (1997) also makes a case for the centrality of citizens attitudes toward Peruvian political institutionsparticularly political partiesin this process. Specifically, he contends that Fujimoris success, with its collateral damage to the body politic, is highly related to the peoples perception that Perus decline was caused by political parties that cared only for

PAGE 171

161 partisan gain and placed narrow interests above the national interests. Therefore, traditional political parties were seen not as a channel for representation but as vehicles for the political ambition of corrupt politicians and self-serving interests. It is important to note that part of this responsibility regarding the collapse of the Peruvian party system was in the hands of party leaders, like for example their incapacity to overcome internal factions and to democratize the internal functioning of their political parties. However, it is also the case that not everything was a complete responsibility of party leaders. In that direction, parties also suffered the dissolution of class solidarities under the stress of economic decline of the 1980s and early 1990s; a nd, in the long term, they were unable to adapt to the changing electoral preferences. In any case, in a context where the traditional parties were no longer seen as valid representatives of different interests and values within agreed-upon rules, they were unable to mobilize opposition to authoritarian measures that attacked the political system within which parties operated (69). So far, I have reviewed some explanations that emphasized that the decline of peoples support or legitimacy toward either the complete democratic regime or some specific political institutions was one of the key explanationsif not the main onefor the 1992 collapse of democracy in Peru. But what about other explanations that disagree with this contention or, in other words, what about some possible shortcomings of these arguments about the centrality of the decline of democratic, political, and institutional legitimacy in Peru. In order o address these issues, I will now revise the explanation that Kenney (2004) offers for the 1992 democratic breakdown in Peru. Without excluding the intervention of several other factorse.g., the economic crisis, Fujimoris political

PAGE 172

162 leadership, and the intervention of the militarythis scholar affirms that the conflict between the executive and the legislature, particularly between November 1991 and the 5 April 1992, was the main cause for this democratic breakdown. He supports his argument showing that Latin American minority presidencies (i.e., when in a presidential system, the executive lacks a parliamentary majority) suffered democratic breakdowns at a much higher rate than majority presidencies between 1960 and 1997. Moreover, in Latin American countries experiencing economic decline, all of the democratic breakdowns took place under minority presidencies. This evidence is even stronger in the case of Peru because all democratic breakdowns in the 20th century occurred in the context of elected minority presidencies. As a consequence of this, the 1992 democratic breakdown in Peru would be one of those cases where in the context of an urgent need for action but having weak presidents partisan powers in the legislature, the president felt the temptation to reject the constitutional rules of the game and seek to circumvent what he perceived as an obstacle for governability. Regarding the executive-legislative conflict in Peru, an initial cooperation between Fujimori and the legislature that started with the inauguration of the new government ended in November 1991 when legislative majorities repealed many important pacification and economic presidential legislative decrees, and placed an unprecedented legislative control on Fujimoris constitutiona l decree powers. Additionally, Fujimori saw his power and some of his policies compromised when the Tribunal of Constitutional Guarantees began also to repeal the presidential legislative decrees not blocked by the Congress and the judiciary released some convicted terrorists on parole (Kenney 2004:219).

PAGE 173

163 According to Kenney (2004:251), as a consequence of the intensification of the conflict between the executive and the legislature, by January 1992 Fujimori found essential elements of his national security policies blocked and his economic program threatened by policies imposed by an opposition congressional majority. Moreover, he became aware that his effective powers had been weakened as a consequence of the Congress exercising veto power over the presidents constitutional decrees authority and reluctant to extent delegated decree authority to the president. From the perspective of Fujimoris opponents this was as it should be (i.e., the legitimate exercise of legislative checks and balances over the executive). However, from the perspective of Fujimori and his allies, the presidents ability to govern was being seriously undermined. Therefore, the decision to carry out the coup was based basically on the conviction that the relationship between the executive and the legislative had reached a zero-sum fight for survival and the expectation that it could end with the deposition of Fujimori by the Congress. To argue that a minority presidency has played a critical role in the breakdown of democracy in Peru is not to suggest that the underlying sources of conflict in the polity were unimportant but that the institutions designed to process and channel that conflict within democratic bounds failed to do so. In some casesmaybe Peru, such conflict may be so severe that no set of institutions would be capable of keeping it channeled within democratic bounds, but it is always the case that institutional arrangements play an important role in democracys success or failure (Kenney 2004:291). Interestingly, Kenney provides not only an explanation of the 1992 democratic breakdown in Peru that disagrees with some of the main conclusions provided by several

PAGE 174

164 other scholars (McClintock 1994a, 1994b; Cameron 1997; Mauceri 1997; Burt 2004); but also a discussion regarding what are the main weaknesses of the explanations that give to the decline of popular support for democracy in Peru a central role in this process. Moreover, he did that in both theoretical and empirical terms. In more theoretical terms, he questions the concept of democratic legitimacy itself; or at least, one very common way in which this concept is defined. Rejecting the idea that a requisite for democratic stability is a widespread belief among elites and ordinary citizens in the legitimacy of the democratic system (Diamond, Linz, and Lipset 1995) or that democratic legitimacy is based on the belief that for that particular country at that particular historical juncture no other t ype of regime should could assure a more successful pursuit of collective goals (Linz 1978:18). Kenney, on the contrary, agrees with Przeworskis contention (1986:50) that it is false that every regime needs legitimacy to survive and that without legitimacy every regime will collapse. The interesting point here is the existence of two different ways of approaching the relationship between democracy and legitimacy. For scholars such as Diamond, Linz, and Lipset (1995) and Linz (1978), elite and popular legitimacy toward democracy plays a key role in the maintenance of democracy. On the contrary, Przeworski (1986) is suggesting the possibility that people can accept a particular type of regimedemocratic or notbecause there is no acceptable alternative. According to this view, democracy can be stable even in a context without a strong democratic legitimacy; and, even more interestingly, a democratic regime can show some persistence not because elites and ordinary citizens believe democracy is the best type of regime but because there is not, at that time, another acceptable political regime.

PAGE 175

165 Regarding this debate, I believe that Przeworski is right in the sense that a decline in support for democracy or the decline in democratic legitimacy will not generate necessarily and automatically a democratic breakdown. However, I believe also that, in the context of new democracies, a lack of support for democracy increases in a considerable extent the possibilities of a de mocratic breakdown. Therefore, if the main concern is to minimize the chances of democratic breakdowns among nee democracies, it is critical to avoid situations where current democratic regimes lack popular support. In empirical terms, Kenney relies on public opinion surveys carried out in Peru to demonstrate that majority of Peruvians citizens never lost a sense of legitimacy toward democracy and a democratic regime. In that direction, he contends that it is possible to explain the simultaneous preference for democracy in Peru and the support for Fujimoris coup. On the one hand, this was the case because for some people Fujimoris coup was not undemocratic.15 On the other hand, for some other people, Fujimoris coup was seen as non-democratic; but they nevertheless support this authoritarian action in the short term as a way of achieving democracy in the long term16 (2004:230). More specifically, regarding the perception that Fujimoris coup was not undemocratic, the key puzzle is how to explain that the post-coup regime was considered as democratic. In that direction, Kenney (2004:232) affirms that this was the case not 15 According to some polls taken in Lima in 1992 April and May by Apoyo Opinin y Mercado after the 5 April autogolpe just over one-half of those polled considered the post-coup regime as democratic, and onethird considered it as dictatorial. These views were not particularly limited to any socio-economic class. 16 According to a poll carried out by DATUM on 7 April 1992, 86% thought that Fujimoris actions would create a more efficient legislature and 84% thought that these measures would decrease corruption in the judiciary. Similarly, 71% thought that the autogolpe would overcome the economic crisis and 56% though it would defeat terrorism.

PAGE 176

166 because Peruvians had very unusual ideas about democracy17 but because the autogolpe presented a mixed picture: a freely and popular ly elected president, closing the Congress and intervening in the judiciary, but giving the image that he was acting in the name of the general will and that there will be little change in terms of freedom of expression and respect for human rights. Therefore, according to peoples understanding of democracy there was not a democratic breakdown 5 April 1992. Interestingly, Kenney suggests that this particular understanding of democracy is close to ODonnells definition of delegative democracy (1994), and that for many Peruvians, Fujimoris actions were only a legitimate extension of the all-encompassing power to rule delegated to him by the citizens. Kenney ends this discussion affirming that because the publics understandings of what constitutes a democratic regime may at times be so broad and so complex, it is always problematic to make arguments based on democracys loss of legitimacy. Regarding the perception that the 1992 presidential coup was a necessary evil, Kenney (2004:233) states that one way to e xplain why some Peruvians characterized the post-coup regime as dictatorial but also approved Fujimoris decision is to consider that the problem was not democracy itself but the poor performance of democratic institutions in the context of Perus crisis; particularly, the legislature, the judiciary, and the political parties. The main idea behind this argument would have been that the autogolpe was a necessary non-democratic short-term action in order to reform and strength on these political institutions not to eliminate them. He reinforces his point mentioning that according to some a national public opinion survey carried out in 1992 by Apoyo Opinin y Mercado, while 79% approved Fujimoris government, 71% said 17 According to a public opinion survey carried out by Apoyo Opinin y Mercado in 1992, people in Lima associated democracy with freedom of expression, respect for human ri ghts, a freely elected president, respect for the Constitution, and respect for the general will.

PAGE 177

167 that they would disprove it if he did not return to constitutional rule. Therefore, rather than embracing a temporarily unlimited authoritarian regime, a large majority of Peruvians had apparently adopted a more nuance position, supporting unconstitutional measures they believed necessary to defend a democratic regime in crisis. As a consequence of all of this, Kenney (2004:237) closes his discussion about the role of the legitimacy of democracy in the 1992 democratic breakdown in Peru saying that it is not clear that democracy had lost legitimacy for most Peruvians at that time, in part because of ambiguities in the use of the terms legitimacy and democracy. On the contrary, what can be concluded is that (1) democracy remained the preferred regime type for the vast majority of Peruvians throughout the period under study, (2) most Peruvians were highly critical of the actual performance of some of the institutions of democracy, and (3) most Peruvians saw Fujimoris actions as either (a) consistent with their understanding of democracy or (b) undemocratic but necessary for the long-term survival of democracy. Role of Popular Support for Democracy in the 1992 Democratic Breakdown After revising some of the main explanations of the 1992 democratic breakdown in Peru, there is enough evidence to contend that a sharp decline in ordinary citizen support for some of the main democratic institutions (i.e., the legislature, the judiciary, and political parties) and democratic procedures played a key role in the presidential coup against the Peruvian democratic regime inaugurated in 1980. Nevertheless, in understanding how this lack of support for democracy was one of the main causes for this democratic breakdown, it will be also critical to account for how this attitudinal dimension interacted with the other critical factors or causes.

PAGE 178

168 As a consequence of this, I have mixed reactions to some explanation of Fujimoris self-coup, like the one provided by Kenney, where the role of popular support for democracy does not play a preeminent role. On the one hand, I agree with Kenney when he suggests that the use of the term democratic legitimacy can be problematic because, to a large extent, this term is broad and sometimes even vague. Also, I think his institutional approach provides a key contribution for a better understanding about the political development in Peru in the last decades, the nature of its presidential systems, and how it shapes the behavior of political leaders. Nevertheless, on the other hand, I do not think it is possible to contend that ther e was not a serious problem of popular support for democracy or democratic legitimacy involved in the 1992 breakdown of democracy in Peru. Accordingly, even if it is the case that a majority of Peruvians have always preferred democracy, it is also the case that the erosion of the legitimacy of the main democratic institutions during the late 1980s and early 1990sparticularly the legislature, the judiciary, and political partiespl ayed a fundamental role in the rise of an authoritarian regime between 1992 and 2000. In a very brief way, the maintenance of democracy in Peru was problematic not because of a lack of peoples identification with the democratic ideals (i.e., participation, contestation, and civil and political rights) but because the institutions and procedures that were supposed to make possible these ideals were perceived as inefficient and unnecessary. Moreover, in some cases even some specific institutions (e.g., legislatures and political parties) or some fundamental procedures (e.g., deliberation, negotiation, and separation of powers) were perceived by ordinary citizens and elites as obstacles for the functioning of a democratic regime.

PAGE 179

169 Therefore, the particular way or ways ordi nary citizens approach and define democracy and the way how they assess the performance of the current democratic regime will have a strong impact regarding what actions can be considered as democratic or nondemocratic. Finally, as the Peruvian case shows, the negative effect of an important decline of popular support for some democratic institutions and procedures can be particularly powerful in the context of a prolonged and acute socio-economic and political crisis. Therefore, rather than talking in general about a process of erosion of popular support for democracy in Peru, we need to distinguish among different components of the democratic regime (i.e., principles, procedures, and institutions). It is going to be also the case that it is necessary to make some distinctions within each one of these components. Nevertheless, it is important to be aware that a lack of popular support regarding some specific political institutions and specific democratic procedures can be enough to create serious risks for the stability of the entire democratic regime, indeed, to create strong incentives for some political leaders with ambiguous orientations toward democracy. Additionally, the Peruvian case shows also some interesting trends regarding the dynamic of popular support for democratic institutions and procedures, as well as the role it might have on the overall functioning of a developing countrys political system. First, it will not be necessarily the case that any experience of a sharp decline of popular support for democratic institutions and procedures, in new democracies, will lead to a democratic breakdown. As I mentioned before, democratic breakdowns are complex phenomena; and the Peruvian experience confirms that the decline in democratic legitimacy or popular support for democracy was one among other several factors that

PAGE 180

170 made possible this outcome. Nevertheless, as the Peruvian case reveals, non-democratic political actors can take advantage of a situation characterized by a lack of popular support for democracy. Second, the Peruvian case also shows that the general level of popular support for democratic institutions and procedures is a very complex issue that is affected by both shortand long-term causes. Regarding the short-term causes, it is clear that peoples perception of a very serious policy failure regarding the Belande and Garcas administrations, and Fujimoris success in attacking the problems of economic decline and political instability were a key component for both the decline in popular support for the democratic regime inaugurated in 1980 and the acceptance of the non-democratic regime between 1992 and 2000. However, it remains as a puzzle the way Peruvians accepted the trade-off between economic and political stabilityon the one handand a situation characterized by an increase in poverty, the persistence of huge levels of social and economic inequality, and constant human rights abuseson the other hand (Cameron and Mauceri 1997:2). Regarding some of the long-term factors affecting the overall popular support for democratic institutions and procedures in Per u, we might consider the conservative nature of the democratic transition of 1980 (Mauceri 1997), the traditional weakness of the Peruvian state to regulate and represent society (Burt 2004), and the Peruvian institutional design (i.e., presidential system) (Kenney 2004). According to Mauceri (1997:36) the democratic transition in Peru in 1980 put in place a weak institutional structure for the consolidation of democracy. Then the Peruvian democratic regime that existed between 1980 and 1992 was a limited affair. The

PAGE 181

171 realities of peoples daily life were corruption, clientelism and abuse of power. By the same token, regarding the connection between weak states and democracy, the Peruvian experience shows that weak states provide a weak institutional basis for the affirmation of democracy. Among other things, weak states cannot guarantee a rule of law and universal citizenship for the entire population, and it is possible that a process of state breakdown can lead to a state re-building process that can be antithetical to the affirmation of democratic principles and procedures. To overcome the weakness and nondemocratic nature of the Peruvian state it will be necessary to enact a series of state reforms to make possible effective mechanisms of check and balances between the branches of government (horizontal accountability), to ensure a fair justice system based on the rule of law, to develop adequate forms of political representation and mediation between state and society, to promote the development of the civil and political society, and to guarantee that the military remains subject to civilian control (Burt 2004:267 268). To a large extent both the conservative nature of the Peruvian democratic regime between 1980 and 1992, and the traditional weakness of the Peruvian state will be part of what McClintock and Lowenthal (1997:xiv) have defined as the traditional vicious political cycle of the Peruvian political system: weak democratic institutions leading to failures of democratic governance leading to potential for legitimate authoritarian rule leading to weak democratic institutions. For example, in the specific case of the Fujimori regime, it is clear that the 1993 Constitution, by centralizing decision making in the executive and limiting legislative and judicial powers, not only created an institutional structure that was less democratic than the previous one, but also institutionalized a semi-

PAGE 182

172 authoritarian political system that went beyond ODonnell concept of delegative democracy (Cameron and Mauceri 1997:241). Cameron and Mauceri (1997) present a very similar argument when they contend that the dilemmas of democratization in Peru, particularly during the 1980s and 1990s, were highly related to the disintegration and failure of the main democratic institutions and the growing gap between state and societ y. More specifically, the disintegration and failure of key institutions such as the judiciary, the legislative, and the political parties created a tremendous crisis of state power and societal representation, which in turn provided a very unfavorable climate for the development of democracy. In this direction, the crisis and collapse of the traditional party system in Peru deserves a special mention. The failure of the traditional political parties to provide a meaningful and effective political representation was a fundamental cause for the disengagement of Peruvians citizens from their democratic regime and institutions (Cameron 1997, Levitsky and Cameron 2003). Finally, it is also the case that the institutional design of Latin American countries (i.e., presidential systems) plays a critical role in producing either minority presidencies that reinforce citizens feelings of ineffective political institutions, and that can lead to democratic breakdownshow it has been the case in Peru during the 20th century; or majority presidencies characterized by an excess of executive power and weak horizontal accountability and checks and balances, that are no less problematic in terms of improving the quality of democratic governance and helping the development of legitimacy or support for democratic institutions and procedures (Kenney 2004).

PAGE 183

173 In this direction, it has become clear that one of the problems in Peru is the weakness of its political institutions. Accordingly, the last thing these political institutions need is to be interrupted by a coup. Extensive literature about the political institutional development shows that political institutions develop and get stronger with age and just through the simple process of continuing to exist, doing their job, and creating a process of institutional learning. Therefore, there are no doubt that Fujimori makes Perus institutional problems worse, no better. Overall, there is no doubt that in the1992 democratic breakdown in Peru several different factorssome of them very exceptional such as the combination of severe economic and political violence crisesplayed an important role. Nevertheless, it is also clear that a civil society approach is not only relevant for the understanding of these events, but also to understand how it was possible for Fujimori to develop an authoritarian regime that lasted almost 10 years. In this direction, neither the 1992 democratic breakdown nor the permanence of Fujimori in government between 1992 and 2000 can be explained without considering a very weak popular support for some key democratic procedures and principles. Accordingly, Fujimori and his allies political and economic allies were the main responsible of the latest democratic reversion of Peru and this new interruption of a process of democratic learning and gradual advancement; however, during most of this time, a large part of the Peruvian populationparticularly the poorestdid not were particularly trouble by the suspension of some democratic principles and procedures.

PAGE 184

174 CHAPTER 6 POPULAR SUPPORT FOR DEMOCRACY IN PERU This chapter contains five sections. In section one, I present a significant improvement regarding the operationalization of support for democracy in the context of new democracies. This empirical strategy was developed initially as part of the United Nations Development Programmes Report on Democracy in Latin America (2004), and one of its main contributions is the identification and measurement of three different and critical dimensions of popular support for democracy. In section two, using a public opinion survey collected in Peru in 2004, I replicate all the relevant statistical analysis related to the operationalization of citizen support for democracy contain in the UNDPs report. This replication revealed also the existence of these three different dimensions of popular support for democracy within the 2004 data: (a) support for democracy as a form of regime, (b) support for democratic procedures, and (c) support for democratic institutions. In section three, I use different statistical models to identify the main important determinants of each one of these three different dimensions of support for democracy. In section four, I present and discuss how individuals socio-economic level has a significant effect on the most significant determinants of popular support for democracy in Peru. Section five is aimed to draw the conclusions of this empirical analysis, and to discuss their significance for process of democratization in Peru and similar countries.

PAGE 185

175 Measuring Popular Support for Democracy in Developing Countries: The UNDPs Report on Democracy in Latin America The UNDPs report about democracy in Latin America (2004) constitutes a very important work about support for democracy in new democracies. Departing from the consideration that support for democracy is key to democratic sustainability, this work was aimed to understand and analyze the support that democracy can count on in Latin America. Accordingly, one section of this report was specifically dedicated to analyze a public opinion survey of citizens perceptions of democracy. This survey was carried out as a section of the Latinobarmetro 2002. The total sample for this survey was 19,508 individuals and it included 18 Latin American countries. As this reports states, one of main puzzles of popular support for democracy in Latin America is the fact that an initial look at peoples perceptions about democracy gives the impression of a considerable level of popular democratic commitment. Consistently, the Latinobarmetro has found that a majority of Latin American citizens prefers democracy to any other kind of regime (e.g., 61% in 1996, and 57% in 2002). However, these kinds of findings do not necessarily imply a strong democratic commitment because many people who affirm that they prefer democracy vis--vis other regimes have as well highly undemocratic attitudes toward several political and social issues. For example, in 2002, almost 45% who said that they preferred democracy were willing to support an authoritarian government if it could solve the countrys economic problems. By the same token, a fair number of people who expressed a preference for democracy oppose some of its basic procedures, rules, and institutions. For example, in 2002, approximately one in three of those who preferred democracy believed that it could function without a legislature and political parties (131).

PAGE 186

176 Therefore, in an attempt to scratch beneath the surface with regard to popular support for democracy in Latin America, and as a way to address its inconsistencies, this report uses different statistical methods to analyze the 2002 data. Accordingly, one of the most important contributions of this work has been the formulation of different indices to address and measure perceptions and attitudes of Latin American citizens toward democracy. Among the most relevant ones, there is the Democracy Support Index (DSI) that it is used as a summary-measure to gauge citizen support for democracy. It is important to note that the elaboration and use of this index was particularly complex because the goal was not only to identify the main orientations1 toward democracy but also to estimate their magnitudein terms of the number of citizens that could be classified within each one of these orientations, the relative distance among these orientations, and the level of political activism among citizens in each one of these orientations. It is worth noting here, that in this chapter I am only interested in how this report deals with the operationalization of popular support for democracy and with the identification of its different dimensions. Regarding the identification of these different dimensions of popular support for democracy, the authors of this report, first, applied a series of factor analyses to a wide set of questions that operationalized citizens attitudes and perceptions toward democracy. As a consequence of this, eleven questions were selected, which reflected not only a general preference for democracy, but also attitudes concerning specific dimensions of support for democracy. Table 6-1 shows this set of questions, as well as 1 This report contends that there are three main orientations toward democracy among Latin American citizens: democrat s or loyal citizens to democracy, ambivalents or semi-loyal citizens to democracy, and nondemocrats or disloyal citizens to democracy. According to the authors of this report, the magnitude of these citizens orientations and the distance among them will represent a useful tool with which to approach the issue of vulnerability of democracy (2004:205).

PAGE 187

177 the three specific dimensions of popular support for democracy that were identified: (a) support for democracy, or the preference for having a democratic regime; (b) support for delegative attitudes or a delegative version of a democratic regime (ODonnell 1994); and, (c) support for democratic institutions. Table 6-1. Selected questions to assess popular support for democracy in Latin America, United Nations Development Programmes Report on Democracy in Latin America Support for democracy as a form of government (Factor 1) (1) Which one of the following sentences are you most in agreement with: a) democracy is preferable to all other forms of government, b) under some circumstances, an authoritarian government can be preferable, or c) for people like me, it doesn't matter whether we have a democratic or a non-democratic regime (2) If you had to choose between democracy and economic development, which you would say is more important: a) economic developmen t is more important, b) democracy is more important, or c) both are equally important (3) In order for (country) to become a deve loped country, do you believed democracy is indispensable as a system of government? Or do you think it is possible to become a developed country with a system of government other than democracy?: a) Democracy is indispensable for a country to become develope d, or b) it is not indispensable, it is possible to become a developed country with another system of government (4) I would not mind a non-democratic government in power if it could solve the economic problem: a) very much in agreement, b) in ag reement, c) in disagreement, or d) very much in disagreement (5) Some people say that democracy permits us to find solutions to the problems that we have in (country). Others say that democracy does not help solving problems. Which statement best expresses your viewpoint?: a) democracy solves problems, or b) democracy does not solve problems Delegative attitudes (Factor 2) If the country has serious difficulties, are you ve ry much in agreement, in agreement, in disagreement, or very much in disagreement that the president should?: (6) Not be limited by what the laws say (7) Secure order by force (8) Control the media (9) Bypass congress and the parties Support for democratic institutions (Factor 3) (10) a) You can't have democracy without a national congress, or b) democracy can function without a national congress (11) a) You can't have democracy without a political parties, or b) democracy can function without political parties

PAGE 188

178 Regarding the three main factors that were identifiedindeed, three different latent variables2 regarding citizens orientations toward democracy, clearly factor 1 comprised the degree of support for democracy as a system of government or the personal preference for a democratic regimes, and corresponded to 16.5% of the total variance. Similarly, it is clear that factor 2 was made up of the group of questions about delegative attitudes or a delegative version of a democratic regimen, and corresponded to 23.5% of the total variance. And factor 3 comprised the degree of support for two key institutions of a democratic regime (i.e., congress and political parties), and corresponds to 13.8% of the total variance. The accumulated variance of these eleven questions explained by these three factors was 53.8%. Table 6-2 shows the factorial loads for these eleven questions used to assess citizens orientations toward democracy.3 It is important to note here that according to the authors of this report, after the identification of these three factors (i.e., support for democracy as a form of political regime, support for delegative attitudes, and support for democratic institutions), subsequent analyses showed that the generation of a combined index for each one of these dimensions turned out to be not very useful (2004:225).4 Therefore, they chose to 2 For an explanation of factor analysis as a data reduc tion technique used to represent a set of variables with a smaller set of factors see Bartholomew et al. ( 2002) and Kachigan (1991). A dditionally, Bartholomew et al. (2002:143) provide a very clear explanation why these factors can be better described as latent variables. 3 Unfortunately, the authors of the UNDPs report do not mention which type of rotation solution they use to calculate the factor loadings. Nevertheless, given the structure of these fact orial loadings, my guess is that they probably have used an orthogonal rotation. 4 Unfortunately, the authors of the UNDPs report do not provide more information about these analyses.

PAGE 189

179 use cluster analysis as a way to group interviewees according to the level of similarity of their answers regarding the 11 questions previously selected.5 Table 6-2. Factorial loads for the eleven questions used to assess popular support for democracy, United Nations Development Programmes Report on Democracy in Latin America Dimension Questions Factor 1 Factor 2 Factor 3 Support for democracy as a form of government Preference for democracy 0.67 Democracy or development 0.58 Democracy indispensable for development 0.69 Authoritarian government acceptable if it solves problems 0.48 Democracy solves problems 0.57 Delegative attitudes President is above the law 0.74 President imposes order by force 0.81 President should control the media 0.80 President ignores congress and political parties 0.77 Support for democratic institutions Democracy without congress 0.84 Democracy without political parties 0.85 Explained variance 16.5% 23.5% 13.8% The total variance explained for this factor analysis model is 53.8% Source: La Democracia en Amrica Latina: Hacia una Democracia de Ciudadanos y Ciudadanas. Programa de las Naciones Unidas para el Desarrollo (2004). As the authors of the report stated (2004:225), the main objective of cluster analysis is to assign cases to groups called clusters, so that the members of the same group are similar with regard to the particular characteristics selected, while the members of different groups are relatively dissimilar. As cluster analysis can be conducted in several different ways, it is important to note here that, in this particular case, the 5 Cluster analysis can be defined as an explanatory tool employed to resolve classification problems. Accordingly, it can be used to rev eal associations and structures pres ent in the data that had not been previously observed (Bartholomew et al. 2002, Grimm and Yarnold 2000, Kachigan 1991).

PAGE 190

180 Euclidian distance method was used to measure the similarity among the different answers, and that a method of K-means partition was applied. Given that the K-means procedure requires the investigator to establish a priori the number of clusters (k) that he or she need to obtain, these authors decided to use three clusters in order to account for the main citizens orientations toward democracy.6 Following this strategy, the UNDPs was able to distinguish among democratic citizens, ambivalent citizens, and nondemocratic citizens within Latin American countries. As I mentioned before, the identification of different dimensions of popular support for democracy and the classification of Latin American citizens according to three different democratic orientations were only two of several steps or components of the DSI provided by this UNDPs report. Given that my main interest here is precisely the topic of popular support for democracy, I will not address the methodological consideration about the others steps or components of this DSI (i.e., the measurement of the size of each democratic orientation, the measurement of the level of political activism of citizens according to their democratic orientation, and the distance among these three democratic orientations). Consequently, I will present now what I consider are the most relevant conclusions of this UNDPs report on democracy in Latin America in regards to the study of popular support for democracy in this region. In that regard, this report concludes that in 2002 citizens with a democratic orientation constitute the largest group in Latin America (43% of the population). The second largest group is the one composed by the ambivalents (30.5% of the population) 6 It is clear that specific cluster analysis results will depend to a large extent on several decisions made by the investigators. Nevertheless, as Bartholomew et al. (2002:22) state, cluster analysis is a descriptive method and its success or otherwise is to be judged by whether or not it produ ces meaningful clusters rather than the means used for their construction.

PAGE 191

181 and the non-democrats are made up for a 26.5% of the population. Therefore, according to this report, the stability of democracy in this region will depend to a large extent in gaining the support of the ambivalents (2004:135). Even more important for me were the sub-regional differences found in this report. Accordingly, the distribution among these three main orientations seems to be more favorable for the stability of democracy in Central America and Mexico because democrats constitute almost 50% of the population, ambivalents are 33.8% and nondemocrats are less than 20%. On the contrary, in the region composed by Andean countries ambivalents are as many as the democrats (34.4 and 37.3% respectively) and non-democrats are almost 30% of the population. Finally, the region composed by the countries that are members of MERCOSUR pl us Chile seems to confront a different challenge. This region presents the smallest number of ambivalents (21.9%) but a considerable number of non-democrats (34.5%) (2004:135). Regarding the social profile of people with different orientations toward democracy in Latin America, this reports concludes that the social origin of those with a similar attitude tend to be particularly heterogeneous; in other words, people who hold certain orientation do not, for the most part, bel ong to a specific group or social class. Nevertheless, it is also the case that it is possible to identify some common patterns. Among the most relevant of these patterns we have that people with higher education are most likely to be democrats. On the contrary, young people figure relatively more among non-democrats; and non-democrats are, on average, people who believed that their economic position has worsened relative to that of their parents; and, non-democrats are

PAGE 192

182 who least expect their children to experience an improvement in their economic circumstances. Similarly, non-democrats tend to think more than others that the main problem for them is not being solved and that the country is going backwards, tend to perceive more frequently that the political sector that they belong to does not have an equal chance of coming to power as others, tend to be less satisfied with democracy, tend to trust institutions and public figures less than other groups do, and tend to believe more frequent than the rest that politicians lied in order to be elected. Finally, it is also the case that the non-democratsbut also the ambivalentstend to be slightly less actively in the political life (2004:137). As an effort of summing up all these common patterns, the report concludes also that these findings can be seen as evidence that deficiencies in social citizenship and poor prospects for economic and social mobility are the two main factors related to the development of non-democratic and ambivalent orientations. In other words, it seems highly likely that the countries with the lower levels of social inequality in the region are at the same the ones with a population more clearly in favor of democracy (2004:146 147). Despite the important empirical and substantive contributions that this UNDPs report has made regarding the study of popular support for democracy in Latin America, I want to finish this section mentioning what in are, in my opinion, its main shortcomings. First, this work lacks a systematic and a more conclusive analysis about the main determinants of these three different and critical dimensions of popular support for democracy (i.e., a personal preference for a democratic regime, the support for a

PAGE 193

183 delegative democracy, and the support for key democratic institutions such as congress and political parties). Second, and as a direct consequence of this, we do not know yet to what extent these different dimensions of popular support for democracy are shaped by the same citizens socio-demographic features and by the same citizens social, political, and economic experiences. Likewise, we do not know yet if some particular sociodemographic features and/or some particular citizens experiences exert a powerful influence in one of these dimensions, but not in the others. As I will show later, advancing in that direction will be extremely useful for the our understanding of the future prospects of the democratic regimes in Latin America, and, in particular, for the future of the ones that can be considered as the most fragile ones. In trying to fill this gap, the next section will present the analysis of a public opinion data that I collected in Peru in N ovember of 2004. After replicating with the 2004 Peruvian data some of the analyses used in the UNDPs report on democracy in Latin America that I have presented in this section (i.e., factor analyses), I will develop several statistical models to account for the main determinants of these three main dimensions of popular support for democracy in Peru. Popular Support for Democracy in Peru: Constructing Outcome Variables In this section I will present, first, the results of a replication of the factor analyses carried out in the UNDPs Report on Democracy in Latin America. Using data from a public opinion survey that I conducted in Peru in 2004, my main goal was to confirm the existence of three main and different dimensions of citizen support for democracy that were previously found in the Latinobarmetro 2002 data. Given that the Peruvian data presented also strong evidence regarding the multidimensional nature of popular support, for democracy in the second part of this section, I will present what I consider a valid

PAGE 194

184 strategy to construct three different support for democracy dependent or outcome variables: (1) support for democracy as a form of political regime, (2) support for democratic procedures,7 and (3) support for democratic institutions. Table 6-3. Sample and Metropolitan Lima population distributions: Gender, age, and socio-economic level Sample Distribution (%) Population Distribution (%) Gender Male 50.4 51a Female 49.6 49a Total 100 100 Age 18 23.2 26a 25 40.7 39.8a 40 36.1 34.2a Total 100 100 Socio-economic level High 7.3 2.7a Middle 16.9 14.1a Low 29.8 28.7a Very low 32.7 35.1a Extremely low 13.3 19.4a Total 100 100 aApoyo Opinin y Mercado: Niveles Socioeconmicos Lima Metropolitana 2004. Regarding the 2004 data, I surveyed the population of Metropolitan Lima residents 18 years and older. Working with a team of twenty professional interviewers, I was able to gather a sample of 413 cases. All the interviews were conducted face to face in November 2004. The sampling method used was a random multiple-stage one with a 7 This particular dimension of popular support for de mocracy was previously defined as the support for delegative attitudes or support for a delegative version of democracy. Nevertheless, I think it is possible to re-label this dimension as support for democratic procedures. I believe that this can be justified because in the original ODonnells work (1994) what constitutes the nature of a delegative democracy is precisely the lack of institutionalization of the main democratic procedures (e.g., the need for horizontal accountability between the president and other democratic branches or institutions).

PAGE 195

185 random selection of zones of blocks, blocks and households. The final selection of the interviewees was made on the basis of age and gender quota sampling. Table 6-3 compares the distribution of gender, age, and socioeconomic level between this sample and the population of Metropolitan Lima. As I mentioned before, I am aware about the risks of trying to generalize my findings, based on a Metropolitan Lima sample, to the entire Peruvian population. However, I believe that the fact that Metropolitan Lima concentrates more than a third of Peruvian population and around of 40% of the total urban national population makes these findings highly relevant regarding what is happening in Peru in terms of the main forces behind the current and prospective levels of popular support for democracy in Peru. Moreover, both some recent empirical evidence and the last electoral national campaign are suggesting that popular support for democracy is even more problematic in urban areas outside Metropolitan Lima and rural areas (PNUD 2006, Aragn 2006). As part of the questionnaire that I elaborated to gather this 2004 public opinion data, I included all the eleven questions used by the UNDPs analysis of Latin American citizens attitudes toward democracy. By the same token, I was able to include different questions used in some of the most important projects aimed to gather Latin American public opinion data that have been carried in the last years (i.e., Latinobarmetro, Latin American Public Opinion Project, Hewle tt Foundation Projects Democracy Through Latin American Lenses: Views of the Citizenry, and Apoyo Opinin y Mercados Informe de Opinin). Moreover, I was able to conduct several in-depth interviews and some focus groups with Peruvians living in Metropolitan Lima that helped me test the final version of the questionnaire.

PAGE 196

186 Table 6-4. Factorial loads for the eleven questions used to assess popular support for democracy, VARIMAX Rotation (Peru, 2004) Dimension Questions Factor 1 Factor 2 Factor 3 Support for democracy as a form of government Preference for democracy 0.68 Democracy or development 0.60 Democracy indispensable for development 0.56 Authoritarian government acceptable if it solves problems 0.53 Democracy solves problems 0.60 Delegative attitudes President is above the law 0.67 President imposes order by force 0.76 President should control the media 0.70 President ignores congress and political parties 0.80 Support for representative institutions Democracy without congress 0.80 Democracy without political parties 0.84 Explained variance 16.9% 20.4% 14.6% The total variance explained for this factor analysis model is 51.9% As in the case of the UNDPs report, factor analyses of the 2004 data also show that there are three main factors or latent variables regarding popular support for democracy. Table 6-4 presents the results of these factor analyses using a principalcomponents factor method and an orthogonal VARIMAX rotation.8 Analyzing these results, it is clear that there is a strong correlation between the questions about preference for a democratic regime, the idea that democracy is as important as economic development, the idea that democracy is indispensable for economic development, the rejection of an authoritarian government even if it is able to solve countrys problems, 8 The VARIMAX procedure attempts to find an orthogonal rotation that is close to a simple structure by finding factors with few large loadings and as many near-zero loadings as possible. This rotation does not alter the fit of the model or the communalities; but it does change the factors loadings. Therefore, the main advantage of this procedure is that it can lead to a simpler structure, eas ier to interpret (Bartholomew et al. 2002:158).

PAGE 197

187 and the conviction that democracy can solve the countrys problemson the one hand, and factor 1on the other hand. Also, factor 1 explains 16.9% of the total variance. Therefore, I can conclude that factor 1 is measuring an overall support or preference for democracy as a form of political regime. Similarly, there is strong correlation between all the four variables related to the support for democratic procedures (i.e., the level of agreement or disagreements with the possibility of the president acting above the law, imposing order by force, controlling the media, and ignoring the congress and political parties in a context of serious difficulties for the country) and factor 2. Also, factor 2 explains 20.4% of the total variance. Therefore, I can conclude that factor 2 is measuring the overall support for democratic procedures. Finally, the correlations between the questions about the possibility of having democracy without congress and without politi cal parties and factor 3 are also very strong. Also, factor 3 explains almost 15% of the total variance. Therefore, I can conclude that factor 3 is measuring citizens support for two of the main democratic institutions. Also, factor 2 explains 20.4% of the total variance. It is important to note that the results of the two factor analyses (i.e., the one contained in the UNDPs report and the one carried out with the 2004 Peruvian data) are very similar. Both identified the same three factors, latent variables or dimensions within the eleven questions used to measure support for democracy. Also, both models explained a similar amount of variance (53.8% in the case of the UNDPs model and 51.9% in the case of the model for the 2004 data). These remarkable similarities are relevant for several reasons. First, we obtained very similar results analyzing public opinion data from 2002 and 2004. This means that at

PAGE 198

188 least in the last years, popular support for democracy trends in Latin America show an important level of consistency. Second, Peru as a particular case reinforces some of the main findings contained in the UNDPs report on democracy in Latin America. Finally, it is not less relevant that the Peru 2004 data, based on a much smaller sample size in comparison to the UNDPs sample, seems to be a reliable data. More specifically, the analysis of the 2004 Peruvian data reinforces previous findings regarding the inadequacy of addressing support for democracy as a onedimensional phenomenon (Dalton 1999, Norris 1999a, PNUD 2002). In that regard, it is the case that Peruvian citizens have develope d a general orientation toward the possibility or convenience of having a democratic regime in their country. Nevertheless, they have developed also more specific attitudes toward the development of a delegative democracy and democratic institutions. More importantly, this is suggesting that these three different support for democracy dimensions can have their own dynamic, and that reducing the complex phenomenon of popular support for democracy to one of these dimension can constitute a critical mistake. This will be even more problematic if, as it seems very likely, a general support for democracy as a form of government does not imply necessarily support for democratic procedures and institutions. Finally, regarding the factor analyses fo r the 2004 Peruvian data, it is worth nothing that factor loadings are particularly high in the case of factor 2 (i.e., support for a democratic procedures) and factor 3 (i.e., support for democratic institutions).9 Considering again that factor loadings are the correlation between single variables and factors or latent variables, it is clear that va riables associated with factor 2 and factor 3 perform much better measuring these two specific dimensions of citizen support for 9 The results of factor analyses for UNDPs data revealed the same pattern.

PAGE 199

189 democracy. On the contrary, it seems that variables associated with factor 1 (support for democracy as a form of government) perform less satisfactorily in measuring this particular dimension of citizen support for democracy. Table 6-5. Reliability coefficients for the set of variables associated with each different dimension of popular support for democracy (Peru, 2004) Cronbach's alpha Support of democracy as a form of government (5 items) 0.56 Preference for democracy Democracy or development Democracy indispensable for development Authoritarian government acceptable if it solves problems Democracy solves problems Support for delegative attitudes (4 items) 0.73 President is above the law President imposes order by force President should control the media President ignores congress and political parties Support for democratic institutions (2 items) 0.65 Democracy without congress Democracy without political parties Eleven items 0.64 The identification of these differences in how well these groups of variables measured each one of the three main dimensions of citizen support for democracy is consistent with the Cronbachs alpha scores calculated for each one of these sets of variables. Cronbachs alpha is not a statistical test but a coefficient of reliability or consistency. Therefore, it measures how well a set of items or variables measure a single one-dimensional latent variable. There is a not generally agreed cut-off point to establish a Cronbachs alpha level that can be considered as acceptable. However, in most social sciences applications, a value of 0.70 or higher is considered acceptable (Nunnally 1978). Table 6-5 shows these reliability coefficients and makes clear that the four items or

PAGE 200

190 variables associated with the support for democratic procedures, the two variables associated with the support for democratic institutions, andto some extentthe five variables associated with the support for democracy as form of political regime can be considered as reliable in measuring these three different dimensions of popular support for democracy. Cronbachs alpha scores can be use also to determine if it is possible to use a set of items or variables to create combined indexes. In this regard, it is possible to affirm that the creation of a combined index for support for democratic procedures and support for democratic institutions can be justified almost without any problems (Cronbachs alpha score equals 0.73 and o.65 respectively). The situation is slightly different in regards to the possibility of creating a combined index for support for democratic institutions (Cronbachs alpha score equals 0.56). In this case, on the one hand, it is true, that the Cronbachs alpha score is less than 0.70. However, on the other hand, it also true that this score does not fall too far from the conven tional 0.70 cut-off point. Moreover, previous factor analyses findings have shown that th ey these five questions about the support for democracy as a form of government have a significant correlation among them, and, hence, they are measuring a latent variable. Therefore, and after recognizing and balancing all of this, I decided that it is possi ble to justify the use of a combined index for each one these three dimensions of popular support for democracy. What about the possibility of creating an overall combined index for popular support for democracy? The Cronbachs alpha score for the 11 questions is 0.64, which could be used as evidence in favor of the creation of such index for support for democracy as a political regime, support for democratic procedures, and support for

PAGE 201

191 democratic institutions. However, two things must be consider before constructing and using such index. First, to a large extent the observed high value of the Cronbachs alpha score for the 11 question (0.64) is the result of the considerable correlations within each one of these three different sub-sets of questions and not the result of strong correlations among these 11 questions. Second, and equally important, the creation of such overall combined index would contradict the main fact or analyses findings, that there are indeed three different factors or dimensions within these 11 questions. Therefore, I decided not to use one overall combined index but instead three separate ones. In order to construct these three separate indexes for support for democracy, support for democratic procedures, and support for democratic institutions, I had the possibility of pursuing two different strategies. First, it seemed reasonable to construct those indexes just adding the interviewees answers to the questions related to each one of these popular support for democracy dimensions. This strategy would allow having cumulative indexes that could be treated as ordinal indexes (Bernard 2002). By the same token, the nature of these outcome variables would make possible the use of cumulative logit models for ordinal responses to analyze their main determinants (Agresti 1996). The second strategy that seemed also reasonable was to use the factor loadings generated by the factor analyses to calculate an individual or interviewees score on each one these support for democracy dimensions (Bartholomew et al. 2002). This strategy would create outcome variables that could be treated as interval variables. In analyzing the main determinants of these new outcome variables, using OLS models could be considered as an appropriate strategy.

PAGE 202

192 At the end, I ruled out the possibility of using cumulative indexes because some of the cumulative logit models that I used to analyze the main determinants of these different dimensions of popular support fo r democracy violated the proportional or cumulative odds assumption that is required to use this kind of statistical analysis.10 As a consequence of this, I decided finally to construct three different dependent variables using the individual or interviewees factors scores and to use OLS regression models for their analysis. It is worth noting here that probably one of the main limitations in using the OLS strategy will be related to the interpretation of the regression coefficients mainly because these coefficients will be expressing changes in outcome variables that have not been directly measured. However, the direction, the magnitude, and the significance of these linear regression coefficients will allow me to identify and assess the impact of the main determinants of these three different dimensions of popular support for democracy. Explaining Popular Support for Democracy in Peru In the previous sections, I have presented and discussed what I consider a very innovative and useful way to operationalize popular support for democracy in new democracies. More specifically, factor analyses were used to identify three different and key dimensions of popular support for democracy, and to calculate individual or interviewees scores for each one of these dimensions. It is clear for me that using this methodological approach to address the issue of popular support for democracy in 10 The proportional odds or parallel assumption holds when the coefficients for the explanatory variables for each sub-group or category within th e outcome variable are similar. I ndeed, this is exactly what makes meaningful the calculation of cumu lative probabilities for the different sub-groups or categories of the outcome variable (Long and Freese 2006)

PAGE 203

193 countries such as Peru has several advantages vis--vis some other methodological strategies that have been used before.11 First, this empirical strategy unpacks the complex phenomenon of popular support for democracy in at least three different dimensions (i.e., support for democracy as a political regime, support for democratic procedures, and support for democratic institutions). This is particularly relevant because it seems very likely that citizens from new democracies, and particularly from the mo st fragile ones, hold different levels of support regarding different components of a democratic regime. By the same token, it is likely that these different dimensions of popular support for democracy are operating according to different logics. Second, these three outcome or dependent variables are not based on one or very few questions or variables; on the contrary, they combine information from different variables. This is particularly useful to develop more valid and reliable measures of popular democratic, and to advance in our understanding of the relationship between mass public opinion and democratization in new democracies (Hagopian 2006). The main challenge now is to clarify to what extent it is possible to identify, within the 2004 Peruvian data, a set of explanatory variables that could explain the level and possible direction of future changes of each one of these popular support for democracy dimensions. As well as to assess to what extent these different dimensions are shaped by similar or different individuals characteristics, experiences, attitudes and perceptions. To 11 See for example Lagos (1997, 2001, 2003a, 2203b), Linz and Stepan (1996), and Diamond (1999). All these scholars have gauged levels of popular support for democracy using the distributions of one or very few survey questions. Even more problematic, all th ese scholars have most of the time neglected the existence of different dimensions of support for democracy, particularly the support for democratic procedures and institutions.

PAGE 204

194 my knowledge this is something that has been largely neglected in the large majority of works dealing with popular support for democracy in new democracies. In that direction, and according to what has been discussed previously, it will be necessary to test the impacts of different types of explanatory variables for these three dimensions of popular support for democracy. Accordingly, I will consider the following sets of explanatory variables: (1) interviewees main socio-demographic features; (2) different definitions of democracy; (3) political or civic culture variables; (4) interviewees preferences for their economy and society; (5) interviewees evaluations of the performance of both their current government and democratic regime; (6) variables related to interviewees attitudes and orientations toward political and electoral participation; and (7) interviewees retrospective evaluation of Fujimoris government, and electoral support for Fujimori. Regarding the effect that individuals socio-demographic features can have on citizen support for democracy, preliminary analyses showed that there was a very high level of correlation between interviewees level of education and interviewees socioeconomic level (Pearson correlation = 0.662).12 Preliminary analyses showed also that interviewees socio-economic level has a significant effect on all the support for democracy dimensions considered here. Nevertheless, the significance of socio-economic level, as an explanatory variable, disappeared once their effect is controlled by other explanatory variables. Therefore, in accoun ting for the effect of socio-demographic 12 Pearson coefficients take values between and +1 and these extreme values represent the strongest possible association between two interval variables (Agresti and Finlay 1997). I am totally aware that both education and socio-economic level can be hardly consider ed as interval variables. Nevertheless, given that I will be using OLS regression models in this section, and the fact that in this caseas well as in other cases herePearson coefficients were almost identical to Spearman rho coefficients, I decided that it was possible to use Pearson coefficients to assess the corre lations within my explanatory variables. See Agresti and Finlay (1997) for a detailed di scussion about the similarities and differences between Pearson and Spearman rho correlations.

PAGE 205

195 features on popular support for democracy, and in addition to age and gender, I decided to use the level of education rather than the interviewees socio-economic level. Nevertheless, the impact of socio-economic level on popular support for democracy will be analyzed separately in the next section. The justification for the inclusion of in terviewees definitions of democracy as possible determinants of popular support for democracy is based on what Camp (2001, 2003) systematically found regarding critical differences in the way citizens from the United States and citizens from Latin American countries define democracy. In analyzing the 2004 Peruvian data, I discovered very similar trends to the ones observed by Camp among Mexicans and Chileans citizens. Peruvians citizens associate also democracy mainly with rule of law, equality, freedom, and welfare or progressin that order. As a consequence of this, I was convinced of the importance of assessing to what extent these different definitions of democracy could have an impact on citizen support for democracy in Peru. In order to be able to do that, I included in my public opinion survey some questions originally found in Camps questionnaire on citizens views of democracy in Latin America (2001). Therefore the multivariate models used to analyze the main determinants of popular support for democracy will include the three following dummy variables: (a) defining democracy as equality, (2) defining democracy as rule of law, and (c) defining democracy as welfare or progress. As I mentioned previously, in studying democratization, several authors have stressed the role of political culture variables, and more specifically, the role of a set of perceptions, attitudes, and practices that can be defined as a civic culture on the maintenance and stability of democratic regimes (Almond and Verba 1967, Putnam

PAGE 206

196 1993). For different reasons, this has been also the most used theoretical and methodological approach to address popular support for democracy in Latin American countries. Hence, I decided to assess to what extent some of these political or civic culture variables have an impact on popular support for democracy in Peru. In that regard, I analyzed the impact of (a) the leve l of interpersonal trust, (b) an index of political tolerance based on the idea of respecting the political rights of interviewees most disliked social groups, and (c) participation in several civil society organizations (i.e., parent-teacher associations, women or ganizations, religious organizations, and communal organizations). Considering the lack of significance of some of these variables in bivariate analysis, I decided not to include in the final OLS model the variables related to the participation in parent-teachers organizations and in women organizations. Similarly, it was not possible to include the index of political tolerance in the OLS final models because such models violated the assumption of a constant variance of the residuals (Fox 1997).13 Then, at the end, and regarding these civic or political culture variables as explanatory variables of support for democracy, final OLS model include only the level of interpersonal trust, the participation in religious organizations, and the participation in communal organizations. In addressing the possibility that citizens preferences and expectations regarding their own society and economy could have a significant impact on the different dimensions of popular support for democracy, I included also in my questionnaire a set of 13 Regarding the impact of political tolerance on s upport for democracy, there is indeed some evidence about the association between a lack of political to lerance and a lower level of support for democratic procedures, however, this time was not possible to account for this effect using OLS regression models.

PAGE 207

197 questions about the following topics: (a) the preference for a society where the majority of companies are private versus the prefer ence for a society where the majority of companies are public; (b) who should be the main responsible regarding the provision of basic services such as electricity and drinkable water (i.e., the private sector or the state); (c) an index of governments price control on basic services and goods such as electricity, drinkable water, telephone, gas, etc.;14 and (d) support for a more egalitarian income or wealth distribution in the country (i.e., leve l of agreement with the following statement: the fairest way of distributing wealth and income is to give everyone equal shares).15 Although all these variables appeared as significant predictors of support for democratic procedures in bivariate analysis, the high correlation between the variable about the main responsible regarding the provision of basic services, and the index of governments price control on basic services and goods (Pearson correlation = 0.447), forced me to exclude the former variable from the final OLS models. As I discussed previously, empirical research on support for democracy on both developed and post-communist countries has shown that the way citizens assess the performance of both governments and democratic regimes has a strong influence on popular support for democracy (Rose, Mishler, and Haerpfer 1998; McAllister 1999; Miller and Listhaug 1999; Dalton 2004). Therefore, it seems reasonable to contend that popular support for democracy is not only about depth-seated values and orientations, both also about specific political, economic, and social assessments of the performance of 14 The reliability analysis for these six basic serv ices and goods measured by the Cronbachs alpha coefficient is 0.92. 15 This question was originally included in the International Social Justice Project. One of the main goals of this study, conducted within post-comm unist countries, was to assess if th e perception of the citizens about economic and social fairness had some impact on the development of political and democratic legitimacy (Mason and Kluegel 2000).

PAGE 208

198 both democratic governments and democratic regimes. Accordingly, I decided also to address to what extent citizens evaluations regarding the performance of their current government and their democratic regime are playing a role in shaping popular support for democracy in Peru. Regarding citizens assessments of the performance of their current democratic governments, I have considered the impact of: (a) an index about the performance of president Toledos government regarding eight key policy areas or outcomes (i.e., economic policy, generation of employment, security policy, respect for the rule of law, respect for the freedom of expression, public education, reduction of poverty, and the respect for democratic norms and procedures);16 and (b) an index regarding the assessment of the retrospective, current, and prospective national and family economic situation.17 As we could expect there was a strong correlation between these alternative ways to measure citizen assessment of government performance (Pearson correlation = 0.438); therefore, the final OLS models for popular support for democracy will include only the index of Toledos government to account for the effect of the evaluation of the performance of the current government. In order to address the impact of citizen assessment of the performance of their democratic regime, I decided to consider the following variables: (a) the standard question about the level of satisfaction with the way democracy is working in the country; (b) an index about the performance of the Peruvian judicial system based on the 16 The reliability analysis for these eight different policy areas, measured by the Cronbachs alpha coefficient, is 0.95. 17 The reliability analysis for the six variables regarding the assessment of the retrospective, current, and prospective national and family economic situation, m easured by the Cronbachs alpha coefficient, is 0.71

PAGE 209

199 answers to three different questions: to what extent the judiciary in Peru provides fair trials, to what extent the judiciary in Peru is capable of punishing people that have broken the law, and to what extent the judiciary in Peru, although with some delays, finally will deliver justice;18 and (c) an index about the performance of the Peruvian authorities in terms of enforcing the law and treating all citizens in the same way.19 The effect of the level of satisfaction w ith the current democratic regime is only significant in a bivariate model accounting for the support for democracy; however, once other explanatory variables are included in the model, this effect loses its significance.20 On the contrary, both the index about the perf ormance of the Peruvian judiciary and the index about the performance of the Peruvian authorities appear as a having a significant and strong effect on the level of support for democratic procedures even when each one of these impacts is controlled by other explanatory variables. However, given the high correlation between these two indexes (Pearson correlation = 0.481), I was forced to include only one of them in the final OLS models; accordingly, I decided to use the index about the performance of the Peruvian judiciar y to account for the effect of the evaluation of the democratic regime performance on popular support for democracy. Chapter 4 made clear that in the context of developing countries one of the topics less developed is the relationship between popular support for democracyon the one handand political participation or political behavioron the other hand. As a 18 The reliability analysis for these three questions related to the performance of the Peruvian judiciary in delivering justice measured by the Cronbachs alpha score is 0.67. 19 The reliability analysis for these two questions related to the performance of Peruvian authorities in enforcing the law measured by the Cronbachs alpha score is 0.77. 20 See Canache, Mondak and Seligson (2001) for a disc ussion regarding the limitations of this question (i.e., the level of satisfaction with democracy) to acc ount for the level of democratic legitimacy among the mass public.

PAGE 210

200 consequence of this, our understanding about the prospects of future democratization in several new democracies will remain fragmented unless we start understanding better this relationship between support for democracy and political behavior. In that regard, I believe it is important to start filling this gap addressing the relationship between variables related to attitudes, perceptions, and orientation about political and electoral participation, and support for democracy, democratic procedures, and democratic institutions.21 In addressing this relationship I have analyzed the impact of the following variables: (a) party identification (i.e., as a dummy variable that distinguishes between respondents whom felt close to some political party and respondents who said that they were not close to any political party), (b) voting intention in the next presidential election (i.e., a dummy variable that distinguishes between respondents who will support some particular candidate in the next presidential election, and respondents who said that they will not support any candidate in the next presidential elections), (c) general perception about voting as an effective mechanism to make things different in the future, (d) perception about the nature of electoral processe s in Peru (i.e., fair vs. fraudulent), and (e) the perception about the equality of opportunities for different political parties. All these variables related to the topic of political ad electoral participation were included in the final multivariate OLS models accounting for the determinants of popular support for democracy. Finally, as I mentioned before, Rose, Mish ler, and Haerpfer (1998) provide a very interesting argument about how to deal empirically with the issue of popular support for 21 Anderson and Dodd (2005) provide a very interesting ex ample of how electoral processes in a context of high level of institutionalization of party competition and strong party identification have a very positive influence on popular democratic commitment.

PAGE 211

201 democracy in new democracies. They have contended that what really matters in this contextin terms of the maintenance of these democratic experiencesis the way citizens evaluate their new democratic regime vis--vis previous and, eventually, current non-democratic alternatives. From the perspective of democratic stability and eventually democratic consolidation of new democracies, citizens do not need to be necessarily satisfied with the current democratic regime; they just need to think that the democratic alternative is better than the non-democratic ones. It is implied in this perspective that citizens are particularly attentive to the economic, social, and political performance of these different regime alternatives. As I mentioned in Chapter 5, since 1992 until 2000, Peru experienced the development of a semi-authoritarian regime (Carrin 2006). This recent political experience matches very well Rose, Mishler, and Haerpfers contention that citizens in new democracies are confronted with and react to different political alternatives. In that sense, it was important for me to address what Peruvian citizens think about these alternative political regimes and how it can be related to the issue of support for democracy. Having that in mind, I included in my survey several questions addressing the evaluation of the performance of Fujimoris regime (i.e., the economic situation, the employment situation, crime, respect of the law, freedom of speech, the situation of public education, concern for poor people, and the situation of democracy in the country). Similarly, I included some questions about voting intention for Fujimori in the coming general election in 2006.22 This should provide me insights about to what extent the 22 Although in 2006, Fujimori was still banned to particip ate as ca candidate in a presidential election, I decide to ask these questions to assess his popular and elector al political support.

PAGE 212

202 current lack of support for democracy in Peru could be related to a positive assessment regarding the performance of a very recent non-democratic regime. As I could expect there is a strong correlation between the evaluation of Fujimori as president and the intention of supporting him in the next presidential election (Pearson correlation = 0.502). Therefore, it only made sense to include one of these variables in the final and general models of popular support for democracy. I decided to maintain in the analysis the index about the evaluation of Fujimoris governments for different reasons. First, the reliability analysis for the eight questions regarding Fujimoris performance shows that they can be indeed treated as one index (Cronbachs alpha score = 0.949). Second, it is possible to question the decision of treating Fujimoris electoral support as a determinant of popular support for democracy. In that regard, it seems likely that a positive evaluation of the performance of Fumitorys regime is having an impact on both the support for democratic procedures and democratic institution, and the willingness to vote for Fujimori in the next presidential election. Summarizing, the set of explanatory variables that I finally used to model the determinants of the three different dimensions of popular support for democracy included the following variables:23 SOCIO-DEMOGRAPHIC CHARACTERISTICS. Gender, age, and level of education. DEFINITIONS OF DEMOCRACY. Equality, rule of law, and welfare. EXPECTATIONS FROM DEMOCRACY. Equality, rule of law, and welfare. POLITICAL OR CIVIC CULTURE VARIABLES. Interpersonal trust, participation in religious organizations, and participation in communal organizations. 23 See Appendix B for a complete description of each one of these variables as well as for an explanation of how all these indexes have been constructed.

PAGE 213

203 ECONOMIC AND SOCIAL PREFERENCES. Preference for a society where the majority of companies are public, index of support for governments price control policies regarding basic services, and support for a more egalitarian income or wealth distribution in the country. EVALUATIONS OF GOVERNMENT PERF ORMANCE AND DEMOCRATIC REGIME PERFORMANCE. Index of evaluation of Toledos government performance, and index of evaluation of the performance of the Peruvian judiciary system. POLITICAL AND ELECTORAL PARTICIPATION. Party identification, electoral support in the next presidential election, opinion regard ing the political efficacy of voting, opinion regarding the nature of electoral processes in the country, and opinion regarding the equality of political opportunities in the country. EVALUATIONS OF FUJIMORIS REGIME. Index of a retrospective evaluation of Fumitorys regime performance. With this final selection of the explanatory variables, I proceeded then to model, using multivariate OLS regressions, the main determinants of (1) support for democracy as a form of political regime (i.e., the preference for having a democratic regime in their own country), (2) support for democratic procedures, and (3) support for democratic institutions. In that regard, the only thing that will be different regarding the specification of these three OLS regression models, that I am presenting here, is the dependent variable. Table 6-6 presents the results of modeling the determinants of the support for democracy as form of political regime using a multivariate OLS regression. Overall, this model, explains 15% of the total variance, and the Cook-Weisberg test of

PAGE 214

204 heteroskedasticity does not provide evidence of the violation of the constant variance assumption. Regarding the interviewees socio-demographic characteristics, and controlling for all other explanatory variables included in the model, only the interviewees level of education has a positive and significant effect on support for democracy as a form of political. If we consider that the dependent variable in this model (i.e., support for democracy) ranges from -1.40 to 1.21, the effect of one unit increase in the level of education on the support for democracy variable (beta coefficient = 0.04),24 and the level of significance (p value less than 0.05), it is possible to contend that education has a somewhat strong effect on the level of support for democracy as a form of political regime among Peruvians living in Metropolitan Lima. Similarly, and controlling for the rest of explanatory variables, thinking that democracy means equality has also a positive and significant impact on support for democracy as a form of political regime. By the same token, considering the range of the dependent variable, the magnitude of this co efficient (0.18), and its level of significance (p value of 0.04), I feel confident in contending that defining democracy as equality has also a moderate strong effect on popular support for democracy as a form of political regime among Peruvians living in Metropolitan Lima. Regarding the effect of the political or civic culture variables and controlling for all other explanatory variables in the model, it is the level of interpersonal trust the only variable that has a positive and significant effect on the level of support for democracy. Moreover, considering the nature of the dependent variable, the fact that interpersonal trust was measured using a 11-point scale, the size of the coefficient (0.04), and its level 24 Note that the level of education is an ordinal variable that ranges from 1 = no education, to 9 = complete college or post-grade.

PAGE 215

205 of significance (0.005), it is possible to contend that the level of interpersonal trust is one main determinants of support for democracy as form of political regime among Peruvians living in the capital of the country. Table 6-6. Support for democracy as a form of government, ordinary least squares (OLS) regression estimates (Peru, 2004) Independent variables Coefficient Standard error Sig. Socio-demographic characteristics Gender -0.064 0.073 0.377 Age 0.002 0.003 0.524 Level of education 0.039 0.018 0.035 Meanings of democracy Equality 0.183 0.090 0.042 Rule of law 0.074 0.085 0.385 Civic Culture Interpersonal trust 0.042 0.015 0.005 Member of a religious organization -0.110 0.091 0.233 Member of a communal organization 0.049 0.116 0.672 Economic and social preferences Preference for public companies -0.017 0.089 0.844 Government's price control policy index -0.003 0.015 0.865 Wealth distribution index -0.040 0.053 0.457 Government and regime performance Evaluation Toledo's government performance index 0.066 0.024 0.007 Evaluation judiciary's performance index 0.004 0.011 0.699 Political and electoral participation Party identification 0.084 0.086 0.329 Voting intention next presidential election 0.145 0.083 0.080 Political efficacy of voting 0.215 0.076 0.005 Nature of electoral processes 0.033 0.098 0.735 Equality of political opportunities 0.032 0.074 0.670 Fujimori Evaluation Fujimoris governments performance index -0.016 0.021 0.459 Constant -0.710 Number of observations 402 Prob > F 0.000 R2 0.15 As in the case of interpersonal trust, a positive evaluation of the performance of Toledos government appeared also as one of the strongest positive effects on the level of support for democracy as form of political regime, even after controlling for all other explanatory variables. One unit increase in the index used to evaluate this performance

PAGE 216

206 will have an impact of 0.07 on the level of support for democracy,25 with a level of significance less than 0.01. Indeed, this can be considered as the strongest effect on support for democracy within Metropolitan Lima population. Finally, regarding the variables relate to political and electoral participation, and after controlling by all other explanatory va riables, only holding the opinion that voting can be considered as an effective way of political participation has a positive and significant impact on the level of support for democracy as a political regime. The magnitude of this beta coefficient (0.21), as well as the level of significance (p value less than 0.01) makes possible to contend that belie ving that voting has political efficacy has a moderate strong effect on support for democracy. Table 6-7 presents the results of a multivariate OLS regression model for support for democratic procedures. Overall, this model explains around 18% of the total variance, and as in the case of the previous model there is no evidence of non-constant variance for the residuals. Regarding the impact of the socio-demographic variables considered in this model, no one appears as a significant predictor of support for democratic procedures. On the contrary, and controlling for the rest of explanatory variables, defining democracy as rule of law has a negative impact on the support for democratic. If we consider the range of this dependent variable (from -1.37 to 1.91), as well as the size of this coefficient (0.20), and its level of significance (p values = 0.035), it is possible to contend that thinking that democracy means the existence of a rule of law has a somewhat strong negative effect on the level of support for democratic procedures among Peruvians living in Metropolitan Lima. 25 Note that the evaluation of Toledos government performance index has an 8-point scale.

PAGE 217

207 Table 6-7. Support for democratic procedures, ordinary least squares (OLS) regression estimates (Peru, 2004) Independent variables Coefficient Standard error Sig. Socio-demographic characteristics Gender -0.022 0.080 0.780 Age -0.004 0.003 0.164 Level of education 0.001 0.020 0.942 Meanings of democracy Equality 0.064 0.099 0.519 Rule of law -0.200 0.094 0.035 Civic Culture Interpersonal trust -0.011 0.016 0.499 Member of a religious organization 0.086 0.101 0.395 Member of a communal organization 0.121 0.127 0.341 Economic and social preferences Preference for public companies -0.207 0.098 0.035 Government's price control policy index -0.044 0.016 0.007 Wealth distribution index -0.275 0.060 0.000 Government and regime performance Evaluation Toledo's government performance index 0.001 0.027 0.975 Evaluation judiciary's performance index 0.031 0.012 0.010 Political and electoral participation Party identification 0.003 0.095 0.974 Voting intention next presidential election 0.146 0.091 0.112 Political efficacy of voting -0.180 0.084 0.033 Nature of electoral processes 0.152 0.108 0.16 Equality of political opportunities 0.194 0.081 0.017 Fujimori Evaluation Fujimoris governments performance index -0.053 0.023 0.024 Constant 1.047 Number of observations 402 Prob > F 0.000 R2 0.18 Regarding the variables about interviewees social and economic preferences, it is remarkable the fact that almost of all them have a strong negative effect on the level of support for democratic procedures, even after controlling for all other explanatory variables. Accordingly, supporting a more egalitarian wealth or income distribution in the country is the strongest single determinant of support for democratic procedures (beta

PAGE 218

208 coefficient = -0.27, and a p value = .000).26 Similarly, the preference for a society with a majority of public companies (beta coefficient = -0.20, and p-value = 0.035), and the preference for a governments control price policy regarding basic services and goods (beta coefficient = -0.04, and p-value = 0.007),27 are also strongly associated with a decrease in the level of support for democratic procedures. In a similar vein, a positive evaluation of the performance of the Peruvian judiciary in delivering justice has a very strong and positive effect on the level of support for democratic procedures even after controlling for the rest of explanatory variables. This index has a scale of 14-point, and am increase of one unit in this index will have an impact of 0.03 on this dimension of support for democracy (p value = 0.01). If it is worth noting here that if the index of the performance of the Peruvian judiciary is replaced by the index of the assessment of Peruvian authorities performance in enforcing the law, this latter index appears also as a significant predictor of support for democratic procedures. Hence, there is also evidence that a positive assessment regarding how Peruvian authorities are performing in enforcing the law and making no distinctions in the way they treat Peruvian citizens has also a positive effect on the level of support for democratic procedures. Finally, and according to my expectations, and controlling for the other explanatory variables, thinking that there are equal opportunities for all political actors in the country has a somewhat strong effect on the level of support for democratic procedures (beta coefficient = 0.20 and p value = 0.017). However, contrary to my expectations, and controlling for all other explanatory variables, thinking that voting is an effective way of 26 This variable is based on a 4-point scale question. 27 Note that this index has a 7-point scale.

PAGE 219

209 political participation has a negative effect on this dimension of support for democracy (beta coefficient = -0.18, and p value 0.013). Table 6-8. Support for democratic institutions, ordinary least squares (OLS) regression estimates (Peru, 2004) Independent variables Coefficient Standard error Sig. Socio-demographic characteristics Gender 0.090 0.070 0.192 Age 0.008 0.002 0.002 Level of education 0.022 0.017 0.210 Meanings of democracy Equality -0.075 0.085 0.378 Rule of law -0.227 0.081 0.005 Civic Culture Interpersonal trust 0.023 0.014 0.099 Member of a religious organization -0.028 0.087 0.743 Member of a communal organization 0.218 0.109 0.046 Economic and social preferences Preference for public companies -0.053 0.084 0.530 Government's price control policy index -0.024 0.014 0.086 Wealth distribution index 0.030 0.050 0.552 Government and regime performance Evaluation Toledo's government performance index 0.019 0.023 0.409 Evaluation judiciary's performance index 0.002 0.010 0.845 Political and electoral participation Party identification 0.209 0.081 0.011 Voting intention next presidential election 0.035 0.078 0.658 Political efficacy of voting -0.014 0.072 0.847 Nature of electoral processes 0.192 0.093 0.039 Equality of political opportunities 0.103 0.070 0.140 Fujimori Evaluation Fujimoris governments performance index 0.026 0.020 0.189 Constant -0.780 Number of observations 402 Prob > F 0.000 R2 0.14 Table 6-8 presents the result for a multivariate OLS regression model accounting for the main determinants of support for democratic institutions. This model explains around 15% of the total variance, and, as in the two previous cases, there is no evidence of a non-constant variance for the residuals. In this model, and controlling for all other explanatory variables, the increase of one unit in the age of the respondents has an effect

PAGE 220

210 of 0.008 on the level of support for democratic institutions, being this coefficient significant at a 0.002 level. Therefore, it is po ssible to contend that older Peruvians living in Metropolitan Lima, in contrast to younger ones, tend to have a slightly higher level of support for democratic institutions (i.e., congress and political parties). As in the case of support for democratic procedures, and after controlling for all other explanatory variables, defining democracy as rule of law has also a negative impact on the level of support for democratic institutions (beta coefficient = -0.23, p value = .005). As I will address later, this is one of the few cases where the same explanatory variable has a similar effect in more than one dimension of popular support for democracy. Therefore, thinking that democracy means the existence of a rule of law is associated with lower levels of support for democratic procedures and institutions among Peruvians living in Metropolitan Lima. Similarly, the participation in a communal organization has also a positive impact on the level of support for democratic institutions, after controlling for the rest of explanatory variables. Given the range of this dependent variable (from -0.88 to 0.73), the size of this coefficient (0.22), and its level of significance (p value = 0.46), I can contend that the experience of being part of a communal organization has a somewhat strong positive effect on the support for democratic institutions. It is worth noting that this is the only case where a variable related to the participation in voluntary organizations appears as a significant determinant of any of the three dimensions of support for democracy that I am considering here. The last two explanatory variables that appear as significant determinants of support for democratic institutions are related to the political and electoral participation

PAGE 221

211 ones. Accordingly, the identification with a political party has a considerable strong effect on this dimension of support for democracy (beta coefficient = .21 and a p value = 0.011), after controlling for all other explanator y variables. By the same token, perceiving electoral processes in the country as fair has also a somewhat strong positive effect on support for democratic institutions (beta coefficient = 0.19 and p value = 0.039), after controlling for the rest of independent variables. For several reasons, I am convinced that these findings provide critical insights for a better understanding of the phenomenon of popular support for democracy in new democracies, and particularly in the ones that are less consolidated or more open to democratic setbacks. First of all, it is clear that different variables appear as the most significant determinants of the three critical and different dimensions of support for democracy that have been identified here. Moreover, in very few cases the same explanatory variables will have a significant impact on more than one of these dimensions of popular support for democracy. Regarding the support for democracy as a political regime or a general preference for having a democratic regime in the country, the evaluation of the performance of the current government, the level of interpersonal trust, and the level of education appear as the strongest determinants of this dimensions of support for democracy. In addition to this, believing that voting is an effective mechanism of political participation and defining democracy mainly as equality appear also as having a considerable impact of the level of support for a democratic system or regime. The fact that the assessment of the pe rformance of the current democratic government appeared as one of the most important determinants of support for

PAGE 222

212 democracy among urban Peruvians living in Lima provides strong empirical support for the group of scholars that have been contending, in recent years, not only that there are critical differences between the factors that explains democratic transitions and the factors that explain the maintenance of democratic regimes within Latin American countries, but also that the quality of democratic government and regime performances play a fundamental role in explaining democratization advances and setbacks in this region (ODonnell 2004b, 2004c; Mainwaring and Hagopian 2006; Mainwaring and Prez-Lin 2006; Hagopian 2006). As it will address later, the impact of the performance of the democratic regime will be even stronger in the case of support for democratic procedures. Regarding the relevance of interpersonal trust as one of the key determinants of support for democracy as a political regime, it is important to note this is still an open debate, particularly within Latin American countries. On the one hand, some scholars such as Power and Clark (2001) have also found that interpersonal trust is a positive predictor of support for democracy in Chile, Costa Rica and Mexico.28 However, on the other hand, other scholars such as Hagopian (2006) contend that it is not clear that within Latin American countries interpersonal trust engenders greater trust in government and political institutions. Therefore, my findings regarding a strong association between interpersonal trust and support for democracy among Peruvians living in Metropolitan Lima suggest that there is still need for more research and discussion about this topic, at 28 In the context of developed countries, Newton and Norris (2000) found also an strong association between an aggregate level of social trust, and c onfidence in government and other political institutions.

PAGE 223

213 least in Latin America.29 Moreover, it could be the case to find this relationship in some but not in all Latin American countries. Regarding the impact of the level of education on support for democracy as political regime, several scholars have found, within Latin American countries, a strong association between higher education and higher democratic orientations and attitudes., Moreno (2001)analyzing data from the World Values Survey for Latin American countriesfound that not only education but al so class and occupation were significantly linked to supportive or opposing views toward democracy. In a similar vein, Camp (2003) suggests that in Mexico the level of educationwhich is closely related to incomehas a strong but indirect effect on political preferences. Specifically, lower levels of education are clearly associated with lower levels of political interest, political knowledge, and political efficacy. Finally, anal yzing also Mexican public opinion survey data in Mexico, Moreno and Mendez (2002) found also a strong correlation between education and different political values, including support for democratic governance. Hence, not only the impact of education on support for democracy as a political regime found in Metropolitan Lima is not atypical vis-vis other Latin American countries; but also there is considerable evidence to contend that education matters as a predictor of an overt support for democracy in Peru and similar countries. I want to conclude the analysis of support for democracy as form of political regime saying that after identifying some of its more important determinants, the critical challenge that remains is to address if there is something in common among them. I will 29 In this regard, I totally agree w ith Power and Clark (2001) when they contend that there is still a great deal of room for an improvement regarding the operationization and measurement of the concept of social trust. By the same toke, I also agree with Camp (2003) when he contends that we have not been able so far to distinguish between the shortand long-term conditions that affect interpersonal trust.

PAGE 224

214 elaborate more on this in the next section, but now I would like to mention that I found that all these critical determinants are correlated with the interviewees socio-economic level.30 More specifically, it is the case that people that belong to the higher socioeconomic levels in Metropolitan Lima have not only higher levels of education, but have also higher levels of interpersonal trust and a more positive assessment of the performance of the current government. By the same token, defining democracy as equality rather than defining it as rule of law, and believing that voting is an effective way of political participation are positively correlated with socio-economic level. Therefore, it is possible to contend that there are some significant differences in terms of the distribution of support for democracy among Peruvians from different socioeconomic levels, and that these differences will be mediated mainly by the interviewees level of education, the level of interpersonal trust, and the evaluation of the performance of the current democratic regime. One of the main theoretical and empirical arguments in this dissertation has been that in addressing the relationship between support for democracy and democratic development within new democracies, it is absolutely critical to consider that popular support for democracy is not a one-dimensional phenomenon. In that regard, it is remarkable that the support for democratic procedures, vis--vis the support for democracy as political regime, has almost a totally different set of significant determinants. Regarding the specific support for democratic procedures, it is the case that most of its main determinants are related to either interviewees preferences regarding the functioning of their own society and economy, or the evaluation of the performance of 30 The variable socio-economic level used here has the following values: 1 = extremely low, 2 = very low, 3 = low, 4 = middle, and 5 = high.

PAGE 225

215 the current democratic regime. Accordingly, interviewees that support a more egalitarian income or wealth distribution in their country, interviewees whom prefer an economy with a majority of public companies, and in terviewees that would support a governments price control policy of the main basic services tend to have lower levels of support for democratic procedures or higher levels of support for a delegative version of democracy. By the same token, interviewees that are more critical or dissatisfied with the performance of the Peruvian judiciary in delivering justice tend also to have a lower level of support for democratic procedures. By the same token, and controlling for all explanatory variables, a positive retrospective assessment regarding the performance of Fujimoris governments has also a very strong negative effect on the support for democratic procedures (beta coefficient = .05, and p value = 0.024).31 Finally, defining democracy mainly as rule of lawrather than defining it in terms of equality, and be lieving that different political actors do not have the same opportunities to advance their programs is also associated with a lower level of support for democratic procedures among Peruvian citizens living in Metropolitan Lima. Overall, it is clear for me that the lack of support for democratic procedures which in turn has played a critical role in the development of a delegative version of democracies or neopopulist32 democratic regimes in Latin America (Weyland 2006)is clearly related to: (1) a sharp distance between some particular social and economic 31 Note that this index has an 8-point scale. 32 Weyland (2006:13) defines neopopulism as a majoritarian conception of political rule where the will of the people, as interpreted by a predominant chief executive reigns supreme, and largely unconstrained by parliament and the courts. In this context, checks and balances are weak, a nd horizontal accountability is low; but a vertical and unmediated relationship betw een the personalistic leader and the masses provides the necessary political legitimacy for the development of this type of political rule. He himself believes that the terms delegative democracy and neopopulist leadership are almost interchangeable.

PAGE 226

216 preferences within Peruvians living in Metropolitan Lima and current features of the Peruvian society and economy (i.e., a high level of social and economic inequality, and an almost complete state and government retrenchment regarding the functioning of the economy); and (2) a very negative assessment of the performance of current democratic regimes, particularly in regards to quality of the rule of law in the country (i.e., the poor performance of the judicial system to deliver justice and the unwillingness or the incapacity of the authorities to enforce the law).33 Accordingly, it seems less difficult now to explain why defining democracy as rule of law, and believing that not all political parties have the same opportunities are also related to lower levels of support for democratic procedures among Peruvians living in Metropolitan Lima. In this regard, it is very likely that the effect of this particular way of defining democracy on the support for democratic procedures is related to the perception that the lack of a rule a law in the country is one of the main deficits of the current democratic regime. By the same token, it could be also the case that a perceived lack of equality in terms of political opportunities is related to the fact that very clear citizens social and economic preferences are most of the time absent in the policy agenda of the last Peruvians governments (e.g., the implementation of reforms to change the current extreme levels of economic and social inequality, and more government intervention in regulating the economic activity). If we consider that some of the most important democratic setbacks in Latin America in the last yearsincluding the recent case of Peruhave not been an outcome 33 It is worth noting here that the index about the level of commitment of Peruvian authoritie s to enforce the law was also a strong determinant of support for democratic procedures. Unfortunately, it was not possible to include it in the final model as a consequen ce of its high correlation with the index about the performance of the Peruvian judiciary,

PAGE 227

217 related to the rise of an open authoritarian government, but an outcome related to the rise of extreme cases of delegative or neopopopulist versions of democracy (Roberts 1995; Weyland 1996, 2001, 2006; Barr and Dietz 2006) these findings regarding the main determinants of support for democratic procedures are extremely useful in understanding the main challenges for the maintenance, development, and eventual consolidation of democratic regimes in this region. Hence we will not be able to understand these democratic setbacks if we do not pay close attention to (1) how citizens are assessing the performance of the current democratic regimes; and, (2) to the extent democratic regimes are being accountable to some clear citizens social and economic preferences and demands. More specifically, it seems that new democratic regimes incapable of enforcing the rule of law, in an effective and unbiased way, and incapable or unwilling to address issues that are particularly salient for a majority of citizens (e.g., economic and social inequality, and a governments responsibility in regulating markets) are prone to democratic regressions. In other words, these performance and accountability deficits would explain why large segments of the population in new the aspiration to have a democratic regime, are much less convinced about the importance of respecting and following democratic procedures. The popular support for democratic institutionscongress and political parties can be considered also as one the most critical factors for the political future of new democratic regimes. In that regard, as in the case of support for democratic procedures, there is no doubt that a low level of support for democratic institutions can be considered also as a critical factor for the development of delegative or neopopulist versions of

PAGE 228

218 democracy, and, eventually, for democratic regressions. Accordingly, the fact that more than 30% of the interviewees in my 2004 data mentioned that it is possible to have democracy without congress and political parties are not good news for the Peruvian democratic regime. Likewise worrying regarding this specific dimension of support for democracy is the fact that younger Peruvians living in Metropolitan Lima tend to have lower levels of support for democratic institutions. I am aware that this finding contradicts what can be considered the conventional wisdom within the literature on popular support for democracy within post-communist countries (see Chapter 4), but I am confident that this finding is a significant one because I obtained almost the same coefficients and the same levels of significance using either a bivariate (i.e., support for democratic institutions regressed by age) or multivariate (i.e., support for democratic institutions regressed by the complete set of explanatory variables) OLS model. In addition to this, the UNDPs report o democracy in Latin America (2004) found also that young citizens in Latin America are among the one with a less democratic orientation. There is no doubt that this particular relationship between age and support for democratic institutions found in Peru and other Latin American countries deserves further attention. Similarly, as in the case of support for democratic procedures, defining democracy as rule of law is negatively correlated with support for democratic institutions. With all the limitations of these survey questions to assess different citizens meanings of democracy,34 it seems that citizens in Metropolitan Lima, thinking that democracy is mainly about having a rule of law, tend to have lower levels of support for some key 34 One of the main limitations of using survey research to assess citizens definitions of democracy is that very frequently we cannot be totally sure about what the interviewees are telling us when they choose equality or rule of law as their meaning of democracy.

PAGE 229

219 democratic components of a functioning democracy: the respect for democratic procedures, the need for horizontal accountability, and a vital party system.35 Equally important, some of the variables related to political and electoral participation are also clear determinants of support for democratic institutions. The identification with a political party and thinking that electoral processes in the country are fair bolster the support for democratic institutions. By the same token, participating in a communal organization increases the support for democratic institutions. It is worth noting here that this is the only case where the participation in voluntary organizations has a significant effect on any of the three dimensions of support for democracy that I am considering here. These findings are highly consistent to a large extent with what Hagopian (2006:360) has suggested as a possible explanation for the resilience of democratic regimes in Latin America, that a robust civil society linked to political institutions and a chance to participate may serve as the best inoculation that democracy has against disappointing economic performance. More specifically, where political representation is weak, the survival of government is dependent on performance, and poor performance may erode democratic regimes, as evidenced in Haiti, Paraguay, Ecuador, and Peru. Even given a similar set of economic problems, democracy has remained more solid in those countries that can count on matureor maturingnetworks of political association, participation, and representation. Where layers of political representation effectively texture the political process, democracy today appears stronger than where these layers have frayed or never existed. 35 See Appendix B for a discussion of how these topics have appeared also in several focus groups sessions that I conducted in Peru during between 1998 and 2004.

PAGE 230

220 Nevertheless, it is important to note that in the case of Peru both the identification with a political party and the participation in communal organizations do not appear to be related to a higher level of support for democratic procedures. Again, different dimensions of popular support for democr acy respond to different individuals characteristics, experiences, and perceptions. Impact of Socio-Economic Level on Popular Support for Democracy in Peru As I mentioned before bivariate OLS regressions between interviewees socioeconomic level and each one the three dimensions of support for democracy considered here showed a significant association between these variables. In particular, socioeconomic level has a strong effect on the support for democratic procedures (beta coefficient = 0.13 and p value = 0.000) and the support for democracy as a form of political regime (beta coefficient = 0.09 and p value = 0.005). Accordingly, Peruvians living in Metropolitan Lima that belong to the higher socio-economic levels have a much higher level of support for democratic procedures, democracy as a form of political regime, and democratic institutionsin that order. Or, in other words, people from the lower socio-economic levels in Limawhom constitute a large majority of the total population36score lower in each one of these dimensions of popular support for democracy. Nevertheless, all these effects lost their significance once other explanatory variables are included in the OLS regression models. To a considerable extent, socio-economic level is not a significant determinant of the three dimensions of support for democracy in the multivariate OLS models because a large part of the variance in the dependent variables that is explained by the interviewees 36 According to Apoyo Opinin y Mercado (2004) the distribution of the popula tion in Metropolitan Lima in 2004 was the following: high socio-economic level 3.3%, middle socio-economic level 15.2%, low socio-economic level 30.4%, very low socio-economic level 33.1%, and extremely low 18%.

PAGE 231

221 socio-economic level is also explained by the other explanatory variables. The next question, then, is if this is enough evidence to contend that it is possible to ignore the effect of socio-economic level on popular support for democracy. My answer to this question is that this is not case. In analyzing the relationship between the main determinants of these dimensions of popular support for democracy and socio-economic level, a similar pattern emerges regarding the sign and the strength of all these correlations. Regarding the support for democracy as a form of political regime, socio-economic level has a positive strong correlation with the level of education (Pear son correlation = 0.66), and with the level of interpersonal trust (Pearson correlation = 0.23).37 Similarly, it has also a positive moderate correlation with the assessment of the performance of Toledos government (Pearson correlation = 0.14). This means that systematically the effect of the explanatory variables that increase the level of support for democracy as form of political regime will be stronger among Peruvians in Metropolitan Lima that belong to the higher socioeconomic levels. By the same token, and regarding the support for democratic procedures, socioeconomic level has a strong negative correlation with an egalitarian orientation in terms of income or wealth (Pearson correlation = -0.30), the retrospective assessment of Fujimoris performance (Pearson correlation = -0.26), and with the preference for a society where the majority of companies are public (Pearson correlation = -0.24). Similarly, it has a moderate negative correlation with defining democracy as rule of law. Again, this means that systematically the effect of the explanatory variables that diminish 37 Note again that socio-economic level is coded in the following way: 1 = extremely low, 2 = very low, 3 = low, 4 = middle, and 5 = high.

PAGE 232

222 the level of support for democratic procedures will be stronger among Peruvians in Metropolitan Lima that belong to the lower socio-economic level. Finally, the same pattern appears regarding the association between the main determinants of support for democratic institutions and socio-economic level. As I just mentioned, there is a moderate negative correlation between socio-economic level and defining democracy as rule of law, and a considerable positive correlation between socioeconomic level and the perception that the electoral processes in the country are fair (Pearson correlation = 0.20). Again, more affluent people in Metropolitan Lima score higher in the variables the increase this dimension of support for democracy, and less affluent people score lower in the variables that decrease it. All these findings allow me to contend that there is in Peru an important, although indirect, effect between socio-economic level and popular support for democracy, in particular in regards to support for democratic procedures and support for democracy as a form of political regime. In other words, the effect of interviewees socio-economic class on the levels of support for democracy happens throughout systematic variations within its main determinants. The justification for this contention is based not only in the clear associations between socio-economic levelon the one handand the main determinants of these different dimensions of support for democracyon the other hand, but also in the fact that it is possible to affirm that there is a casual relationship between socio-economic level and the main determinants of support for democracy in Peru. Being the case that the individuals socio-economic level can explain an important part of the variation in the preference for a more egalitarian society, the preference for a society where governments play an important regulatory role, the level of interpersonal

PAGE 233

223 trust, the assessment of both governments and regime performance, and even the way democracy is defined. Therefore, in dealing with popular support for democracy in Peru and similar countries, the effect of class or socio-economic level cannot be neglected. Conclusions There are two main general conclusions for the empirical analysis that I carried out in this chapter. First, this chapter provides enough evidence to contend that it is possible to address and explain the phenomenon of popular support for democracy in the context of a new and fragile democracy using public opinion surveys. In that regard, popular support for democracy, more specifically, its different dimensions are heavily influenced by the combination of some structural factors such as education, socio-economic level, and even age; citizens performance assessments of both their democratic regime and their past and current governments; citizens social and economic preferences; citizens perceptions and orientations toward political and electoral participation; and even some of the most standard political or civic culture variables such as the level of interpersonal trust and the participation in voluntary organizations. Depending on the specific dimension of support for democracy that is considered (i.e., support for democracy as a political regime, support for democratic procedures, and support for democratic institutions), a very particular sub-set of these explanatory variables will appear as its most important determinants. In that regard, the assessment of the performance of the current government, the level of interpersonal trust, and the level of education are the three variables with the strongest influence on the level of support for democracy as a form or political regime. By the same token, a strong social and economic egalitarian orientation, a positive assessment of the performance of Fujimoris regime, the assessment of the current democratic regime in terms of the quality of the rule

PAGE 234

224 of law, and a clear preference for a governme nts price control policy and for an economy whit a majority of public companies appear as the most important determinants of support for democratic procedures. Finally, defining democracy as rule of law, being member of communal organization, being identified with a political party, and thinking that electoral process in the country are fair are the most important determinants of support for democratic institutions (i.e., congress and political parties). The second main and general conclusion is that the phenomenon of popular support for democracy is clearly a multidimensional political phenomenon. In that regard, it has been possible to identify at least three different dimensions of this political support: (a) a general support or preference for a democratic regime, (b) support for democratic procedures, and (c) support for democratic institutions. Equally important, as I just mentioned, there is strong evidence that each one of these dimensions is shaped by different interviewees characteristics, experiences, preferences and perceptions. Even more, in that regard, it is worth noting here that only between support for democracy as a political regime and support for democratic institutions seems to be a considerable level of association (Pearson correlation = 0.30). On the other hand, there is a weak association between support for democracy and support for democratic procedures (Pearson correlation = 0.15), and almost no association between support for democratic procedures and support for democratic institutions (Pearson correlation = 0.05). Therefore, it is not only the case that these different dimensions of popular support for democracy differ in terms of its main determinants, but also have a limited impact in shaping each other. For example, it seems highly likely that many Peruvians in

PAGE 235

225 Metropolitan Lima that are supportive of democracy as a political regime are much less supportive of democratic procedures. I think that this extremely relevant for the discussion about the future of new and fragile democracies because changes in these levels of supports regarding these different dimensions will have different consequences in the functioning of a new democratic regime. For example, in Peru and similar Latin America countries (e.g., Venezuela, Ecuador and Bolivia), democratic setbacks have been basically related to very low levels of support for democratic procedures, or, as a direct consequence of this, highly related to an important popular support for delegative or neopopulist political regimes (Roberts 1995, Weyland 2006, Hagopian 2006). Accordingly, having a better understanding about the factors that can affect the popular support for democratic procedures is particularly relevant for this region. Regarding some more specific conclusions, it is clear that in the context of a new democracy, citizens assessments of both democratic regime and democratic governments play a critical role in shaping popular support for democracy, in particular support for democratic procedures and support for democracy as a form of political regime. Therefore, popular support for democracy, in this context, hardly can be considered as independent from how citizens are assessing these performances. Being likely that no stable and considerable support for the current democratic regime or democratic procedures will arise where citizens are profoundly dissatisfied with how their governments and democratic regime are performing. Equally important is the fact that regarding these regime and governments performances, citizens do not only care for the

PAGE 236

226 economic performance but also for other policies such as security and public education, and, how authorities are performing in delivering and effective and unbiased rule of law. Moreover, there is evidence that, despite the recent political experience of the country, a considerable part of the Peruvian electoral was willing to support Fujimori in a presidential election because his governments were associated with the capacity to formulate and implement the needed policies. As Levi (2006:5) suggest this is related to the recurrent situation within developing countries where current democratic regimes have serious deficits in terms of being effective, that is, capable of protecting the population from violence, ensuring security of property rights, and supplying other public goods that the populace needs and desires. In a similar vein, one of the main challenges for the development of popular support for democracy among Peruvian citizens in Metropolitan Lima is related to a huge gap between some clear social and economic expectations, and current features of Peruvian society and economy. In that regard, citizens who are clearly dissatisfied with how wealth and income is distributed in the country, and, therefore have a strong egalitarian orientation; and citizens who clearly in favor of a higher level of government intervention in regulating the economy are considerable more supportive of delegative or neopopulist versions of democracy. Additionally, different perceptions and attitudes related to the quality of political and electoral participation are also important determinants of popular support for democracy in Metropolitan Lima. Thinking voting is an effective mechanism of political participation bolsters support for democracy as a form of government. Thinking that political actors have the same political opportunities to advance their political programs

PAGE 237

227 increases the level of support for democratic procedures. Finally, both the identification with a political party and thinking that electoral processes are fair increase the level of support for democratic institutions. Not less intriguing is the fact that some particular citizens definitions of democracy appear also as important determinants of popular support for democracy. Particularly, defining democracy as rule of law has a negative effect on support for democratic procedures and institutions. As I mentioned before, I think might be reflecting also the profound dissatisfaction with the current democratic regime in delivering an effective and unbiased rule of law.38 Finally, and considering the preeminent role that civic or political culture variables have had in previous works addressing the topic of popular support for democracy in Latin America, as well as in other regions, it is highly relevant that these variables do not appear as the most important determinants of this political phenomenon. However, two of them contribute in a significant way to explain the variations in the levels of support for democracy. In the case of support for democracy as a form of political regime, the level of interpersonal trust is one of its strongest determinants. Similarly, the participation in communal organizations has a somewhat positive effect on the level of support for democratic institutions. This finding is to a large extent consistent with what Seligson (1999:357) found analyzing survey data from Central American countries, that only 38 This observation came from several focus groups and in -depth interviews that I conducted in Peru since 1999. Particularly, regarding citizens from lower socio-economic levels, it is clear that they have very specific expectations from democracy (e.g., welfare, employment, economic progress, end of social discrimination, rule of law, etc.). At the same time, it is also clear that perceiving that the current regime is constantly falling in fulfilling these expectations is one of the main reasons for the development of a strong feeling of political and democratic alienation (Appendix B).

PAGE 238

228 participation in community development organizations was consistently related to democratic participation. Therefore, and until more research is accumulated about the effect of these civic or political culture variables, it does not seem a good idea to ignore some of them as possible determinants of some dimensions of popular support for democracy. However, how it has been pointed out, we will have to be able to distinguish and to address the relationship between social capital and political capital because not any form of organizational involvement will have necessarily a positive impact on the development of democracy (Bayer and Booth 2000). To a large extent, all these findings are consistent with what was found among post-communist citizens from Central and East Europe by Rose, Mishler, and Haerpfer (1998:227), that the main obstacle to completing democracy come from the shortcomings in what is supplied by political elites, particularly in regards to political outcomes such as rule of law, political freedom, and accountability to the electorate. Nevertheless, my own findings are also at odds with Rose, Mishle r, and Haerpfer (1998:140) contention that structural factors are almost irrelevant fo r the development of popular attitudes toward the new democratic regimes and undemocratic alternatives. At least in the case of Metropolitan Lima, there is evidence that some of these structural factors such as education, age, and even socio-economic level have an important impact on the levels of popular support for democracy, democratic procedures, democratic institutions, and even non-democratic alternatives. In particular, regarding the indirect impact of socioeconomic level, I found that there are strong correlations between class and the main determinants of these different dimensions of popular support for democracy. More

PAGE 239

229 disadvantaged citizens, systematically, score higher in the main determinants that have a negative impact on these dimensions of support for democracy, and score lower in the main determinants that have a positive impact on these dimensions of support for democracy. More specifically, what a found among Peruvian living in Metropolitan Lima is that citizens from lower socio-economic levels in Peru are the ones who are less positive about the performance of both their current democratic government and the performance of the Peruvian democratic regime, particularly in terms of the quality of the rule of law; very dissatisfied with the current situation in terms of income or wealth distribution; much less satisfied with how the Peruvian government is regulating the economic activity; much more skeptical about the quality of political and electoral representation in the country; and the one who have the lower level of interpersonal trust, and a clear tendency to define democracy as rule of law. Therefore, and at least for the population of Metropolitan Limaalmost a third of national population, individuals differences in terms of socio-economic level has a very perverse effect on the levels of popular support for democracy. The effect of the variables associated with an increase in the support for democracy, and democratic procedures and institutions are much weaker among individuals that belong to the lower socio-economic levels. Similarly, the effect of the variables associated with a decrease in the support for democracy, and democratic procedures and institutions are much stronger among individuals that belong to the lower socio-economic levels Overall, this empirical research provides important insights about a set of factors that matter for an increase or decrease in the levels of support for democracy as a form of

PAGE 240

230 government, and, more specifically, for democratic procedures and institutions among Peruvians in Metropolitan Lima.39 For example, and regarding what seems more relevant for the future of the Peruvian political system and its democratization process, we have now a better idea about what factors make the current democratic regime more or less resilient to democratic setbacks throughout the rise of delegative or neopopulist political leaderships. Finally, regarding the use of public opinion survey data to address the phenomenon of popular support for democracy in the context of new and fragile democracies, this chapter provides interesting insights. First, the generation and analysis of survey data constitute a very useful strategy to address empirically the nature and dynamics of popular support for democracy. Second, there is already a considerable amount of experience regarding a set of survey questions that can be used to address this political phenomenon (e.g., the way the UNDPs report on democracy in Latin America has operationalized support for democracy as an outcome or dependent variable) that has been produced by the community of scholars dealing with this phenomenon in the context of countries trying to develop democratic regimes. Accordingly, I used extensively this expertise in the process of gathering my own data, and I think this is one of the reasons why the Peruvian 2004 data has been particularly useful for my theoretical and empirical goals. Additionally, in analyzing the Peruvian data, it has become very clear the advantage of using combined indexes rather than single variables. In that direction, very frequently, these indexescomposed by 39 I do not see any good reasons to rule out the possib ility that similar patterns, and maybe clearer ones, could be found within the total Peruvian urban populati on, particularly in the co astal main cities of the country. Moreover, new empirical evidence suggests that the level of popular dissatisfaction with the performance of the Peruvian democratic regime is even stronger outside Lima (PNUD 2006).

PAGE 241

231 different single variableshave appeared as significant determinants of citizen support for democracy. Similarly, all this process has shown me the benefits of combining different research tools to address a very complex political phenomenon such as popular support for democracy.

PAGE 242

232 CHAPTER 7 CONCLUSIONS In dealing with the study of recent processes of democratization, scholars have given a great deal of attention to the role of political elites, and the merits and liabilities of different institutional designs. In this regard, it is true that both elites and political institutional designs play key roles in the creation and maintenance of democracy. Nevertheless, it is also true, for different reasons, that the role of ordinary people deserves close attention. Just to consider one of these reasons, as several historical and contemporary cases show, much of what elites attempt to do is conditioned by their judgments of how ordinary people will behave (Bermeo 2003). As a consequence of this, the purpose of this dissertation was to propose a fundamental theoretical and methodological revi sion of mainstream approaches that have been used to account for citizen political support for democracy and their current democratic regimes, particularly in the context of new democracies. This decision was made following a comparative and critical assessment of the concept of popular support for democracy, and the consideration of the role that this popular support can play in the development, maintenance, and consolidation of a democratic rule. In carrying out this critical assessment, I was able to frame an argument to explain the important role that popular support for democracy plays in the current situation of several new democracies; particularly in the less consolidated or more fragile ones. Specifically, and throughout this discourse, I have contended that the lack of a considerable level of popular support for democracy, particularly regarding the

PAGE 243

233 institutions and political procedures of these new democratic regimes is not only a serious problem in terms of the maintenance and further democratization of these political regimes, but that it is also a clear indicati on of the underlying deficiencies that these new democracies possess with respect to the quality of political inclusion, representation and accountability that they offer to their citizens. Moreover, I developed an alternative theoretical and methodological approach that can be use to gauge the level and main determinants of popular support for democracy in new democracies. As I will show in this chapter, I believe that this new approach can be extremely useful in overcoming some of the more critical limitations that previous scholars have encountered in dealing with the topic of popular support for democracy within new democracies, as well as with its main political implications. Comparative and Critical Assessment of the Literature on Popular Support for Democracy In both theoretical and methodological terms, I have identified the main strengths and weaknesses of the literature on popular support for democracy, particularly when it is used to address the experience of new democracies. Such critical evaluation is crucial in order to fully understand and contextualize the difficulties and main challenges encountered by new democracies as they seek to advance in terms of their democratization processes. In my opinion, it is within this framework that the topic of popular support warrants careful attention. Lack of a considerable and stable popular support for democracy, and, more specifically, lack of support for the main procedures and institutions of current democratic regimes often results in an emergence of non-

PAGE 244

234 democratic and authoritarian forces within new democracies.1 Consequently, these new democracies may fail to benefit from an interrupted experience with democracy. This is indeed unfortunate, for a continued experience with democracy is critical to facilitate a process of democratic learning for both elites and ordinary citizens (Anderson and Dodd 2005). In addition to this, a continued experience with democracy is equally crucial for a gradual advancement of the democratic quality of these current new democratic regimes (ODonnell 2004b, 2004c). Nevertheless, it is the case that there are important limitations in the way in which popular support for democracy has been addressed in the last years, particularly among new democracies. These limitations have arisen not only with respect to problems with the operationalization and measurement of popular support for democracy as a dependent variable, but also in terms of the identification of its main determinants. Accordingly, it is time to reject the common but misguided assumption that popular support for democracy can be treated as a one-dimensional phenomenon, and, as a consequence of this, can be accurately measured via one or two survey questions about very general preferences for a democratic system. Regarding the multi-dimensional nature of popular support for democracy, my empirical analysis has shown not only that it has indeed different dimensions,2 but also that each one of these dimensions is heavily influenced by a very particular set of explanatory variables. 1 In the context of Latin American countries, there is no doubt that this was the case in Peru between 1992 and 2002 (Carrin 2006), and it seems also to be the case in contemporary Venezuela (McCoy and Myers 2004). According to McFaul, Petrov, and Ryabov (2004), this is also the case in current Russia. 2 The analysis of the 2004 data confirms the existence of at least three different dimensions of popular support for democracy found initially by the UNDPs report on democracy in Latin America (2004): (a) support for democracy as a form of political regime, (b) support for democratic institutions, and (c) support for democratic institutions.

PAGE 245

235 Likewise, it is also time to recognize that it is absolutely necessarily to integrate different theoretical approaches in order to account more satisfactorily for this complex political phenomenon. Regarding these main determinants, my empirical analysis has revealed, for example, the importance of individuals preferences and expectations toward their own society and economy, individuals assessments of both democratic governments and democratic regime performance, and individuals evaluations in regards to the quality of political representation. Moreover, the empirical analysis undertaken in this dissertation has also shown the relevance of structural variables as key determinants of popular support for democracy. Socio-economic level is one such variable; it exerts a powerful, although indirect, influence on the different dimensions of popular support for democracy in shaping its main determinants (e.g., individuals social and economic preferences and expectations, and individual s assessments of democratic performance). In regards to the comparative assessment of the literature on popular support for democracy, I surveyed the most important theoretical and methodological works that have been developed on established democracies and new democracies.3 It is worth noting that the study of popular support for democracy has a long tradition among scholars dealing with the development of democracy,4 and that it gained some attention over the last several years as a consequence of the third-wave of democratization (Huntington 1991) or the latest global resurgence of democracy (Diamond and Plattner 1996). 3 In dealing with the research literature on popular support for democracy within new democracies, I have addressed separately the cases of postcommunist and Latin American countries. 4 See, for example, Lipset (1959), Almond and Verba (1963), and Dahl (1971).

PAGE 246

236 To a large extent, it is the literature on popular support for democracy within established democracies that is leading the way in terms of how to account theoretically for this complex political phenomenon, as well as how to operationalize and measure it. As part of the main contributions of this literature, we can consider the contentions that popular support for democracy is essentially a multidimensional phenomenon5 (Norris 1999a), and the necessity of using different theoretical approachesnot only a political culture approachin order to develop a comprehensive understanding of its main causes and consequences (Norris 1999c, Dalton 2004). By the same token, there are important differences in the way the topic of popular support for democracy has been addressed in the context of new democracies. Accordingly, it is the literature on popular support for democracy in the context of postcommunist countries experiences that presents the highest levels of theoretical and methodological advancements. In this regard, I found the debate about how postcommunist citizensparticularly the most disadvantaged ones in economic and social termsdefine democracy, and combine a preference for democracy with high expectations in terms of governments social and economic responsibilities, particularly useful. Additionally, in the context of post-communist countries, and regarding the determinants of popular support for democracy in those countries, I have found the theoretical and empirical insights regarding the critical role that governments and regimes perceived performance have on the overall level of popular support for democracy equally useful. Finally, it has been also very valuable to review the debate 5 These different dimensions incl ude: (1) support for the political comm unity; (2) support for democratic principles, norms, and procedures; (3) assessment of the regime performance; (4) support for democratic institutions; and (5) support for political actors and authorities.

PAGE 247

237 about how to measure popular support for democracy and to discover that there are important limitations in assuming that peoples general or abstract ideas about democracy, or the level of commitment toward some equally abstract democratic values, can be considered as reliable measures of political support for current democratic regimes. Therefore, in dealing with the phenomenon of popular support for democracy, it is critical to recognize that there is much to learn from a comparative study on this topic considering both established and new democr acies, and also considering different new democratic regions. By the same token, it is clear that this applies to both theoretical and methodological approaches toward this complex political phenomenon. Popular Support for Democracy and the Prospects of Further Democratization in New Democracies In connecting the research on popular support for democracy with the study of recent experiences of democratization, there are two bodies of literature that are extremely relevant. First, regarding the role that popular support for democracy can play in the current and future political development of new democracies, the literature on democratic consolidation appears particularly relevant. Second, regarding the need of explaining current levels and determinants of popular support for democracy among new democracies, a political culture approach seems to be the right place to start. Regarding the literature on political culture, there is no doubt that scholars such as Almond and Verba (1963) and Easton (1975) pr ovided the seminal formulation of what now is know as the political culture approach : the relationship between individuals cognitive, affective, and evaluational orientations and the stability of political regimes. It is worthwhile to note that despite a fundamental concern for the role of political values

PAGE 248

238 (i.e., deep-seated political orientations), these scholars were careful not to neglect the impact of regime and governments performance on individuals political orientations. However, in the large majority of contemporary works using a political culture approach toward politics, it would seem that the concern for political socialization and the role of political values overshadows a systematic concern for how assessments of regimes and governments performances exert also a powerful influence on the development of individuals political and democratic orientations. This is the main reason why I have contended, on the one hand, that in dealing with the topic of popular support for democracy is not possible to neglect the contribution of a political culture approach. However, on the other hand, it is also the case that the current mainstream political approach is insufficient to grasp satisfactorily this political phenomenon. Regarding the literature on democratic consolidation, I indicated in Chapter 2 that scholars dealing with the main challenges and prospects about the consolidation of new democracies are correct when they contend that certain citizens attitudinal and behavioral changes are indispensable for the maintenance and, eventually, further democratization of these new democratic regimes. To a large extent, these attitudinal and behavioral changes are captured by the concept of popular support for democracy. However, it is absolutely necessary to make two additional comments to this argument. First, it does not make any sense to affirm that new democracies will be consolidated when ordinary citizens and political elites recognize and accept that democracy is the only game in town.6 New democracies are, by definition, very open and elastic political experiences, therefore, in order to avoid democratic reversal or the development of very stable hybrid regimes (Hartlyn 2002; Levitsky and Lucan 2002; Van de Walle 2002; 6 See for example Linz and Stepan (1997), and Diamond (1999).

PAGE 249

239 Schedler 2002, 2006), a gradual and constant development of popular support for democracy will play a critical role. In other words, it is necessary to approach support for democracy within ordinary citizens as a process and not as the final sign of democratic consolidation. Further, the concept of popular support for democracy does not imply that the majority of ordinary citizens need to be convinced democrats, willing to fight for democracy if it is necessary.7 As some scholars have pointed out, popular support for democracy or popular democratic legitimacy in new democracies can be seen in a more realistic and less demanding way (Rose, Mishler, and Haerpfer 1998; Anderson 2005). According to this view, the legitimacy of democracy only requires that ordinary citizens perceive that the current democratic regime is better in comparison to other political alternatives, including some non-democratic ones. In other words, what is really important in terms of the relationship between popular support for democracy and the possibility of further democratization of new democracies is to reach a situation where ordinary citizens, even the ones that are not satisfied with their current democratic experience, will not support actively or passively some political discourses and actions aimed to question and/or suspend all or some key democratic procedures and institutionswhich are very frequently justified as a necessary step toward the realization of a real democracy. Moreover, it is clear that this kind of support for democracy is indispensable for new democracies in order to, throughout a long and sustained experience with democratic norms and procedures, improve the quality of its current democratic regimes. 7 See Bermeo (2003) for a very intere sting view about what we can exp ect from ordinary citizens in terms of support for democratic and non-democratic alternatives.

PAGE 250

240 It is important to mention here that this argument in favor of popular support for democracy does not imply that ordinary citizens have to accommodate or lower their social, political and economic expectations in order to avoid political conflict, polarization, and eventually the breakdown of democracy. On the contrary, I have found that one of the main explanations for a low level of popular support for democracy within new democracies is the very common perception, particularly among most disadvantaged citizens, that these current democratic regimes are highly ineffective in terms of its economic, social and political performance.8 Accordingly, it seems almost impossible to expect an increase in terms of popular support for democracy if new democratic regimes are unable to change the current popular perception that they are, very frequently, unaccountable to their citizens expectations and demands. This is particularly more relevant for the most social and economically disadvantaged citizens, whom in several new democracies constitute the large majority of ordinary citizens. Regarding this deficit of accountability, it is true that some of these citizens expectations toward democracy and current democratic regime are very difficult to satisfy in the shortand even medium-term (e.g., generation of employment, reduction of poverty). However, it is not the case that all these citizens expectations have an economic or socio-economic nature. In this regard, there is also a strong concern for how the rule of law is applied in a very intermittent and biased way. Similarly, maybe ordinary citizens are not necessarily expecting that democratic governments can provide, all at once, solutions for their concerns and demands, but rather 8 On how poor democratic regime performance is one of the main explanations of democratic setbacks in Latin America, see Mainwaring and Hagopian (2006) Mainwaring and Prez-Lin (2006), and Hagopian (2006). On how this could also the case in several new democracies in developing countries, see Levi (2006).

PAGE 251

241 some clear signs that they are making the effort to incorporate them in its political agendas. Unfortunately, this possibility sharply contrasts with what has been happening over the last few decades, for example, in Latin America, where there has been a dramatic reduction in the level of governments responsibilities regarding the social and economic well-being of their citizens (ODonnell 2004b, 2004c). Finally, in thinking of the democratic quality deficits among several new democracies and some alternative to change this situation, maybe the best strategy is not to blame the limitations of a procedural or minimalist definition of democracy (e.g., that a procedural or minimalist democracy, which by definition cannot guarantee the achievement of specific social and economic goals). Rather, the best strategy could be to assume that indeed democratic regimes are open in terms of the specific policies that can finally be implemented. Particularly for the most disadvantages citizens, these democratic procedures and rules collectively provide a viable alternative to achieve that some social rights could finally be recognized and incorporated by their political regimes.9 Nevertheless, I believe that two conditions need to be met if this political strategyworking within the formality of democratic procedures and democratic institutionsis to have any chance to be successful in increasing the quality of the current new democratic regimes. First, as Donnell (2004b, 2004c) has pointed out, it is critical to address why the existence and recognition of different political rights within new democraciesparticularly in developing countriesseem to be almost irrelevant in 9 Here, and regarding the level of governments social and economic responsibilities, it is important to consider that there are indeed different models of democracy. For example, countries like the United States have developed democratic regimes with a limited sc ope in terms of social rights and welfare system. Meanwhile, other countries, such as Western European countries have developed democratic regimes with a broader scope in terms of social rights and welfare system (Fuchs 1999). To a large extent, societies within these new democracies have the possibility to choose which m odel of democracy they want to pursue.

PAGE 252

242 terms of improving the quality of these democratic regimes, especially in terms of political representation and political accountability. As he explains, several features of the social and economic context (e.g., a weak rule of law and extreme levels of poverty and inequality) are the main causes for the difficulties of the most disadvantaged citizens in using intensively their political rights. Second, all these changes, such as making political rights more relevant, and extending and applying in a unbiased way civil and social rightsthat are so important for an improvement of the democratic quality of new democracieswill have a greater chance to be institutionalized and to endure if they are achieved within the framework of a procedural democracy. This is why it is so important for new democracies to avoid situations where some political leaders can suspend the democratic experience and to seriously harm a process of democratic development and learning. In this regard, this is where a better understanding of the main challenges for a gradual development of popular support for current democratic regimes can play a very significant role. Alternative Theoretical and Methodological Approach toward Popular Support for Democracy in New Democracies The empirical analysis that I conducted as part of this dissertation allowed me not only to identify three different dimensions within the phenomenon of popular support for democracy in a new democratic regime (i.e., support for democracy as a form of political regime, support for democratic procedures, and support for democratic institutions), but also to identify a particular set of variables th at appear as key determinants of each one of these dimensions. As it will be addressed in this section, these findings are extremely useful in understanding why is so difficult the development of democratic popular support for new democratic regimes, particularly in the context of developing countries.

PAGE 253

243 Regarding the role played by individuals expectations and preferences about their own society, it was the case that those who were in favor of a somewhat radical wealth of income distribution reform, those who wanted a higher level of government intervention in regulating the private economic activity, and those who preferred a higher presence of the state in regards to the provision of basic services and goods, were clearly the citizens less convinced about the need or the importance of respecting and following democratic procedures. According to these results, it seems very likely that in several new democracies within developing countries, the most dissatisfied citizens with the current wealth or income distribution, and/or the most dissatisfied citizens with the common current reluctance of governments to assume economic and social responsibilities, can constitute an important social base for political leaders proposing a delegative or neopopulist version of democracy (ODonnell 1994; Weyland 1996, 2001, 2006). By the same token, citizens performance assessments of both democratic governments and the democratic regime have also a strong influence on some dimensions of popular support for democracy. Specifically, a positive retrospective evaluation of the overall performance of Fujimoris regime, and a negative assessment of the performance of the democratic regime in delivering an effective and unbiased rule of law had a strong negative influence on the level of support for democratic procedures. Similarly, a positive evaluation of the overall performance of Toledos government had a positive impact on the preference for democracy as particular form of political regime. Therefore, in the context of new democracies, citizen support for democracy particularly, citizen support for democratic procedurescan hardly be considered as independent from how citizens assess the performance of both the current democratic

PAGE 254

244 regime and its different governments. It is also very relevant the fact that in assessing the performance of democratic governments, citizens care not only for the current family and national economic situation, but also for specific policies such as crime and security, the quality of public education, generation of employment, etc. Equally important is the fact that the concern for a clear non-economic performance, such as the implementation of an effective and unbiased rule of law, is one of the main determinants of support for one of the key components of current democratic regimes (i.e., democratic procedures). Accordingly, the stability and further development of new democratic regimes will depend not only in its capacity to be representative and accountable to the population that they are meant to serve, but also in its capacity to offer effective governments capable of providing the necessary conditions for economic growth, security and stability for its citizens, and even the goods that its population needs or demand (Levi 2006). In other words, in several new democracies, building a stable democratic regime will imply also the development of the necessary governments capacities to formulate and implement policies in an effective manner. Otherwise, as the Peruvian political experience of the late 1980s and early 1990s clearly shows, the combination of dramatic economic and political violence crises can generate a situation where ordinary citizens could give priority to governance over consensual and deliberative democratic procedures. Moreover, as it has been suggested, the perception about the quality of political representation could be considered also as an important determinant of citizens perceptions about democracy and current democratic regimes (Hagopian 2006). To a large extent, my findings support the claim that individuals assessments of the quality of political representation have an important effect on popular support for democracy.

PAGE 255

245 Regarding the support for democratic institutions, it was the case that both the identification with a political party and believing that electoral process are fair increased this particular dimension of popular support for democracy. Additionally, there were significant associations between perceiving voting as an effective mechanism of political participation and support for democracy as a form of political regime, and between perceiving that all political parties have the same opportunities to advance their political programs and the support for democratic procedures. It is worth noting here that it has been also suggested that when political representation is weak, the survival of governments is dependent basically on performance (Hagopian 2006). In this regard, and as I addressed in Chapter 5, the recent Peruvian political experience can be considered as a good example of how the lack of effective networks of political association, participation, and representation amplifies the negative impact of a poor performance on support for democratic regimes and democratic stability. Additionally, as this case is also suggesting, poor democratic performance makes very difficult the development or the recovery of popular trust in democratic authorities, institutions, and procedures. Also, even some civic culture variables appeared as important determinants of some of these dimensions of popular support for democracy. Specifically, the level of interpersonal trust had a very strong positive impact on the level of support for democracy as a form of political regime. By the same token, the participation in a communal organization had a positive impact on the level of support for democratic institutions. As I mentioned before, several scholars have either questioned the validity of using civic culture variables to address the processes of democratization in third-wave

PAGE 256

246 democracies (Muller and Seligson 1994) or have made the case for the need of more research to fully grasp how civic culture variables are related to the phenomenon of popular support for democracy (Seligson 1999, Bayer and Booth 2000, Power and Clark 2001). In this regard, I will contend that it is not possible to neglect the impact that some of these civic culture variables (e.g., interpersonal trust and participation in some voluntary organizations) have on some dimensions of popular support for democracy (e.g., support for democracy as a form of government and support for democratic institutions). However, I am also convinced that we are still very far from having a comprehensive understanding of how these civic culture variables interact with the different dimensions of popular support for democracy in new democratic regimes. Equally important is the role played by citizens ideas about democracy. For example, it was the case that those individuals who defined democracy as rule of law were characterized by lower levels of support for democratic procedures and democratic institutions. On the contrary, those individuals who defined democracy as equality were characterized by a higher level of support for democracy a form of political regime. As I have noted before there are some limitations in making clear statements about what are the meanings and implications of defining democracy as equality or as rule of law; however, based on my own findings and similar research,10 I believe that it is very likely that different ways of approaching democracy are producing different orientations toward current new democratic regimes, and that thinking that democracy is about the existence of a rule of lawin a context where this is very frequently one of the main democratic deficitsis creating an additional challenge for the development of popular support for new democratic regimes. 10 See for example Camp (2001, 2003).

PAGE 257

247 Furthermore, and on the contrary to some contentions regarding the low impact of structural factors on popular support for democracy within new democratic regimes (Rose, Mishler, and Haerpfer 1998), my analysis has revealed that some sociodemographic variables are clearly related to some dimensions of popular support for democracy. Specifically, the level of education had a strong impact on the support for democracy as form of government, and age had a positive impact on the support for democratic institutions. Finally, as important as the identification of some of the main determinants of these different dimensions of popular support for de mocracy was to find out a very particular relationship between most of them and one stru ctural variable: individuals class or socioeconomic level. The 2004 Peruvian data on which my analysis is based showed systematically that citizens from lower socio-economic levels in Metropolitan Lima were, at the same time, the same individuals who, in a greater number, were more critical of the current situation in terms of income and wealth distribution in the country, and the limited role of government in regulating and participating in the economy. They were also the more dissatisfied with the economic and social performance of their current democratic government (i.e., Toledos government), with the performance of their democratic regime in guaranteeing a rule of law, and with the quality of political representation. By the same token, they were the ones with the lowest levels of interpersonal trust, and the ones who more frequently defined democracy as rule of law. Hence, what we have here is that class or socio-economic level reinforces, in a negative way, the individual effect of almost all significant determinants regarding the different dimensions of support for democr acy. Accordingly, the most disadvantaged

PAGE 258

248 citizens in Metropolitan Lima tended to be more skeptical about the convenience of having a democratic system in the country, and about the benefits of maintaining and respecting democratic procedures; and tended also to be less supportive of two key democratic institutions (i.e., congress and political parties). In this regard, I believe that this makes absolutely clear that the lack of popular support for democracy among several new democracies will continue to be an issue unless the most disadvantaged, who in these cases constitute a large majority, can develop the conviction that maintaining democratic procedures and living under a democratic regime can provide the best alternative to advance their particular social, political, and economic concerns and needs. Summarizing, as this research and similar works have shown, the phenomenon of popular support for democracy is a multi-dimensional one. Therefore, different theoretical perspectives are needed to account satisfactorily for these different dimensions of support for democracy. In this regard, and particularly in the context of new democracies, it is absolutely necessary to address how citizens assess the performance of both governments and regimes, the role of citizens preferences and expectations toward their own society and economy, and how citizens evaluate the quality of political representation. Similarly, it is crucial to pay attention to how class or socio-economic level, in turn, is shaping most of these key determinants of the different dimensions of popular support for democracy. By the same token, my research has revealed that these different dimensions of popular support for democracy are influenced by a very particular set of explanatory variables or factors. In this regard, for example, the main determinants of a general support for democracy as a form of political regime are not the main determinants of

PAGE 259

249 support for democratic procedures or democratic institutions. There is no doubt that some correlation among these different dimensions, but no one of these dimensions explains a large amount of the variation of the other dimensions of support for democracy. Therefore, it seems that within new democratic regimes, there are different ways of being supportive or skeptical regarding the current democratic experience and the current democratic regime. As I mentioned before in Chapter 6, this particular finding is extremely relevant for the very large group of troubled or fragile new democratic regimes where democratic setbacks are not necessarily related to the installation of an open authoritarian regime or to the almost complete lack of popular support for democracy, but to the installation of delegative neopopulist, or electoral authoritarian regimes (Schedler 2006). In this regard, having a better idea of what is shaping the support or the rejection for delegative neopopulist or semi-authoritarian versions of democracy is a step forward in our understanding of the most recent experiences of democratization, as well as its further prospects. I would like to finish this chapter mentioning what I consider are the most valuable methodological lessons regarding the study of popular support for democracy in new democratic regimes. Although, there is no a specific chapter about the main findings of several focus groups sessions that I conducte d in Peru between 1998 and 2004, there is no doubt they made an important contribution to my research (Appendix B). First of all, they revealed to me that key political perceptions, attitudes, and orientations in Peru differ to a considerable extent along the different socio-economic levels. Also, they were particularly useful in helping me to discover that the problem with popular support for

PAGE 260

250 democracy in Peru was not related necessarily to the rejection of democracy as a form of government or to the existence of a non-democratic political culture. Rather, it was related to a high level of dissatisfaction and frustration with the political, social, and economic performance of both different democratic governments and the whole democratic regime. Additionally, these qualitative data made an important contribution for the development of my questionnaire about popular support for democracy. Specifically, regarding the use of survey data to address popular support for democracy, it is worth noting that survey research that is directly connected to specific research questions can be quite informative and useful. To a large extent, the Peruvian data used in this dissertation were extremely useful because there were produced from survey research that was designed specifically to deal with the phenomenon of popular support for democracy. Also, in trying to account for such complex political phenomenon as dependent variable, and in trying to account for its main determinants, it is evident that creating and analyzing multi-item indexes provide, very frequently, valid and reliable measures of these key variables than using single questions. Finally, there is no doubt that the use of different statistical methods (e.g., factor analysis, cluster analyses, OLS regression) can contribute significantly in advancing our understanding of complex political phenomena such as popular support for democracy in new democracies.

PAGE 261

251 APPENDIX A THE DEFINITION AND CODING OF THE EXPLANATORY VARIABLES Socio-Demographic Variables The measurement of interviewees socio-demographic variables was straightforward and requires only brief explanation. Gender was coded using a dummy variable, where male = 0 and female = 1. Age was an interval variable reporting the age in years for each one of the interviewees. For the level of education I used a ordinal variable with eight categories: no education and primary education incomplete =1, primary education complete = 2, secondary education incomplete = 3, secondary education complete =4, technical education incomplete = 5, technical education complete = 6, college education incomplete = 7, and college education complete and post-grade education = 8. Socio-Economic Level was measured following the methodology developed by Apoyo Opinin y Mercado (2004), and the outcome was an ordinal variable with five categories: extremely low = 1, very low = 2, low = 3, middle = 4, and high = 5. Citizen Definitions of Democracy Variables The two dummy variables used to address the different citizens definitions of democracy were derived from asking interviewees: In one word, could you please tell me what is democracy for you? The alternatives for this question were the following: freedom, equality, voting, a form of government, welfare or progress, and rule of law. Defining democracy as freedom then, was a dummy variable where answering democracy was coded as 1, and any other answer was coded as 0. Similarly, defining

PAGE 262

252 democracy as rule of law was a dummy variable where 1 = mentioning rule of law, and 0 = any other answer. Finally defining democracy as welfare or progress was a dummy variable where 1 = mentioning welfare or progress, and 0 = any other answer. Civic or Political Culture Variables The level of interpersonal trust was obtained by asking: Using a scale from 0 to 10, where 0 means that it is not possible to trust other people and 10 means that it is possible to trust other people, what do you think about people in general? Therefore, this was an ordinal 11-point variable. Political tolerance was a multi-item index based on a four questions about the realization of public demonstrations by most disliked social groups (e.g., migrants, homosexuals, drug addi cts, people from a different religion, people with different political ideas, etc.), where 0 = the most political tolerant attitude, and 4 = the most intolerant political attitude. Finally, the participation in voluntary organizations was measured asking if the interviewees were members of a parent teacher organization, a women organization, a religious organization, or a communal organization. Therefore, in all these four cases, I used dummy variables where 1 = member and 0 = not a member. Social and Economic Expectations Variables Regarding interviewees social and economic expectations, the variable about the preferred type of society was obtained using the following question: What do you prefer, a society where the majority of companies are public, a society where the majority of companies are private, or a society where similar numbers of public and private companies? This variable was coded as a dummy variable where 1 = preference for a society with a majority of public companies and 0 = preference for either a society where

PAGE 263

253 the majority of companies are private or a society with equal number of public and private companies. The variable about the main responsible in delivering public services and goods was obtained by asking: Which one of the fo llowing statements best reflect your way of thinking: (a) the state should be the responsible for the provision of basic services such as electricity and drinkable water, or (b) basic services such as electricity and drinkable water should be provided by private companies. This variable was coded in the following way: 1 = the main responsible for the provision of basic services should be the state, and 0 = the private sector should be the main responsible for the provision of basic services. The index of governments price control policy was obtained using the answers to a set of questions about the prices of six basic services or goods that in the opinion of the interviewees should or should not regulated by the government (i.e., drinkable water, electricity, gas, telephone service, local transportation, and interest rates). Thin index ranges from 0 to 6, where 0 = the least favorable attitude toward a governments price control policy, and 6 = the most favorable attitude toward a governments price control policy. Finally the support for an egalitarian orientation regarding the distribution of income or wealth in the country was obtained by asking: Do you strongly agree, somewhat agree, somewhat disagree, or strongly disagree with the following statement: the fairest way of distributing wealth and income is to give everyone equal shares. The answers to this question were coded in the following way: 4 = strongly agree, 3 = somewhat agree, 2 = somewhat disagree, and 1 = strongly disagree.

PAGE 264

254 Assessment of Government and Regime Performance Variables The index of the performance of Toledos government was obtained using interviewees evaluations regarding eight different areas and/or policies: economic situation, generation of employment, internal security, respect for the rule of law, respect for the freedom of expression, public education, reduction of poverty, and respect for democratic norms and procedures. As it was finally included in the OLS models, this index has an 8-point scale, where 1 = lowest overall evaluation, and 8 = highest overall evaluation. Regarding the assessment of the retrospective, current, and prospective national and family economic situation I constructed an index adding the answers from these six different economic assessments (i.e., retrospective assessment of both national and family economic situation, current assessment of both national and family economic situation, and prospective assessment of both national and family economic situation). This index ranges from 0 = the most negative overall assessment of the economic situation, to 26 = the most positive overall assessment of the economic situation. The level of satisfaction with democracy was obtained using the following standard question: How satisfied are you with the way democracy is working in Peru. Are you very satisfied, somewhat satisfied, somewhat dissatisfied, or very dissatisfied? The answers to this question were coded in the following way: 4 = very satisfied, 3 = somewhat satisfied, 2 = somewhat dissatisfied and 1 = very dissatisfied. The index of the performance of the Peruvian judicial system was obtained adding the answers to three different: to what extent the judiciary in Peru provides fair trials, to what extent the judiciary in Peru is capable of punishing people that have broken the law, and to what extent the judiciary in Peru, although with some delays, finally will deliver

PAGE 265

255 justice. This index, as it was finally included in the OLS models, has a 14-point scale, where 1 = the most negative overall evaluation of the Peruvian judiciary, and 14 = the most positive overall evaluation of the Peruvian judiciary. The index of the performance of Peruvian authorities in enforcing the law was obtained adding the answer of two related questions: to what extent Peruvian authorities are committed to enforce the law and to what extent Peruvian authorities treat all citizens in the same way. This index, as it was finally included in the OLS models, ranges from 1 to 8, where 1 = the most negative overall evaluation of the performance of Peruvian authorities, and 8 = the most positive overall evaluation of the performance of Peruvian authorities. Political and Electoral Representation Variables The variable party identification was based on a simple question which asked: Do you feel close to one political party or not? The answers to this question were coded in the following way: 1 = yes, and 0 = no. The variable voting intention in the next presidential election was based also on a simple question: If tomorrow was the next presidential election, which political party you would vote for? Given the answer to this question, I coded this variable as a dummy variable where 1= interviewees that mentioned the party that they would support in the next presidential election, and 0 = interviewees that mentioned that said no one or did not mention a party. The variable political efficacy of voting was measured using the following question: Which of the following statements is closer to your way of thinking: the way one votes can make things different in the future; or it does not matter the way one votes, things will not be different in the future. The coding for this variable was the following: 1 =

PAGE 266

256 thinking that voting matters regarding the future of the country, and 0 = thinking voting has not impact on the future of the country. The variable about the perception about the nature of elections in own country was based on the question: In general, do you think that elections in your country are fair or do you think they are fraudulent? The answers were coded in the following way: 1 = thinking that elections are fair, and 0 = thinki ng that elections are fraudulent. Finally, the perception about equality of political opportunities in the country was measured using the following question: Do you think that the political group you support has the same opportunity as others to get into powers or do you think it does not have the same opportunity? The answers for this question were coded in the following way: 1 = same political opportunities, and 0 = there are not the same political opportunities. Fujimoris Variables The index of the performance of Fujimoris governments was obtained using interviewees evaluations regarding eight different areas and/or policies: economic situation, generation of employment, internal security, respect for the rule of law, respect for the freedom of expression, public education, reduction of poverty, and respect for democratic norms and procedures. As it was finally included in the OLS models, this index has an 8-point scale, where 1 = lowest overall evaluation, and 8 = highest overall evaluation. Electoral support for Fujimori in the next presidential election was based on a simple question which asked: If Alberto Fujimori was able to run in the 2006 presidential election, what do you think will be your situation: I will definitively vote for him, I will probably vote for him, I will probably not vote for him, and I will definitively not vote for him? The answers for this question were finally coded in the following

PAGE 267

257 way: 1 = definitively or probably vote for hi m, and 0 = definitively or probably do not vote for him.

PAGE 268

258 APPENDIX B QUALITATIVE DATA ON POPULAR SUPPORT FOR DEMOCRACY IN PERU The main empirical component of this dissertation has been the analysis of a public opinion survey conducted in Lima, Peru, in November 2004. However, it is important to mention here that the collection and analysis of qualitative data on popular support for democracy between 1999 and 2004 played an important role in the development of the research design and research hypotheses used in this dissertation. As Powers (2001:4) suggests in her research on grassroots expectations of democracy and economy in Argentina, and as I have realized myself, using a qualitative approach helps researchers not merely in collecting data but also in discovering what the right questions are. During the second half of 1999, I conducted six focus groups on attitudes and perceptions toward politics and democracy among poor1 Peruvians living in Metropolitan Lima of both genders and different age-groups.2 Subsequently, in 2000, 2001 and 2004, I was part of a research team in charge on conducting focus groups on citizens political attitudes and expectations.3 The large majority of participants in the 2000, 2001 and 2004 focus groups were also poor Peruvians that lived in Metropolitan Lima. Therefore, this 1 By poor people, I am considering citizens that belong to the extremel y low, very low, and low socioeconomic levels (Apoyo Opinin y Mercado 2004). 2 See Aragn (2000) for a detailed analysis of these data. 3 All these research projects were conducted by Apoyo Opinin y Mercado (2000, 2001, 2004). The nine 2000 focus group sessions were conducted between March 5th and March 9th. The eight 2001 focus groups sessions were conducted between March 7th and March 15th. Finally, the eight 2004 focus group sessions were conducted between August 23rd and August 27th. Overall, in this appendix I report and analysis findings from 29 focus group sessions.

PAGE 269

259 qualitative data represent mostly what poor Peruvians in the capital city of Peru think about democracy and their current democratic experience.4 One of the most important issues that I wanted to address in all these focus group sessions was the logic behind phenomenon of popular support for democracy5 in a developing country trying to develop a democr atic regime. One possible explanation for the lack of popular support for the democratic regime in Peru, that we saw during the 1992 democratic breakdown and in the following yearsincluding the re-election of Fujimori in 1995 and 2000, could be the existence of an authoritarian political culture in this country, particularly among poor Per uvians. Another possible explanation could be that this lack of popular support for the democratic regime is more the result of a complex process that includes not only political values, but also peoples expectations of democracy and government, and peoples assessments of the performance of democratic governments and their democratic regime in dealing with these expectations, and in providing solution so the most critical problems of the country. It is clear for me that these different alternative explanations regarding the logic behind popular support for democracy and democratic regimes are crucial for the debate about the prospects of democracy in Peru, and similar countries.6 If democratic 4 See Chapter 1 for a discussion regarding to what extent this Lima data can be generalize to other parts of the country. 5 As I mentioned before, I define popular support for de mocracy in this dissertation in a very particular way. By popular support for democracy I do not expect common citizens to be actively engage in the promotion or defense of their current democratic regime, but rather in the way they react in a situation where some political leaders challengein the discourse and/or practicethe need of respecting the basic democratic rules, procedures or in stitutions. According to this defin ition, the lack of popular support for democracy implies basically that ordinary citizens will be either indifferent and, eventually, supportive of these non-democratic discourses or actions. 6 As it has been addressed before in Chapter 1, th ere is enough evidence to contend that countries like Peru, at least in the context of South America, are Venezuela, Bolivia, Ecuador, and Paraguay (Hagopian and Mainwaring 2006).

PAGE 270

260 breakdowns are not necessarily the expression of an authoritarian predisposition among the majority of citizens, but rather they cannot be explained without taking into consideration how citizens were assessing the performance of their democratic regime or a particular democratic government; there are more opportunities to avoid democratic breakdowns. After analyzing these qualitative data on popular support for democracy in Peru, I was convinced that this type of political support was indeed a complex phenomenon, and that several and different factors appeared as its most important determinants. More specifically, I was convinced that the strength and consistency of popular support for the democratic regime was much less based on democratic or authoritarian political values, and much more based on the way democracy is defined, the expectations of democracy and democratic governments, the assessment of the performance of the current democratic government, and, particularly, how the current democratic government is dealing with the more salient issues at that time. Therefore, I believe that the maintenance and consolidation of a democratic regime is possible in new democraciesparticularly the ones within developing countries; although, this requires addressing the most important causes of popular democratic dissatisfaction and alienation. However, as the Peruvian case also shows, it is worth noting that the causes of popular democratic dissatisfaction and alienation can be profound and diverse.7 As I consequence of this, the possibility of being able to make some generalizations about these trends observed during the focus groups sessions, was one of the main motivations to conduct a public opinion survey in Metropolitan Lima. In that regard, I 7 Chapter 5 discusses with deta il the combination of political, ec onomic, social, and institutional circumstances that explain the lack of popular opposition toward Fujimoris actions against the democratic regime in April 1992 and his subs equent non-democratic actions.

PAGE 271

261 was convinced that collecting data on popular support for democracy that could be generalized to the population of Peruvians living in Limawhere more than third of Peruvian population resideswill constitute an important contribution in understanding the limits and possibilities of democratic maintenance and democratic consolidation in Peru, and similar countries. Specifically, the analysis of these focus groups sessions provided me insightful information about (1) how poor Peruvians in Lima approach and define democracy, (2) what they expect from democracy and from a democratic government, and (3) how they assess the performance of a democratic government, particularly vis--vis what they considered the main countrys problems at that time. In regards to this last topic, I found particularly illuminating the way poor Peruvians in Lima reacted to Fujimoris political style and the way they assessed the performance of his two governments. Almost without exception, a large majority of poor Peruvians participants in all these focus groups sessions considered that Fujimori represented the best government in the last years. Interestingly, his actions were not without criticismincluding for example his arbitrary use of power, but the general idea was that as a whole, and in comparison to previous democratic governments, Fujimori delivered a good government for the country. As I will show later this positive evaluation was mainly related to the high level of effectiveness perceived during Fujimoris governments, particularly regarding to what poor Peruvian in Metropolitan Lima considered the most salient countrys problems at that time (i.e., economic crisis and political violence),8 and regarding to what they considered their main particular problems (i.e., poverty, lack of opportunities, etc.). 8 See Weyland (2006).

PAGE 272

262 Regarding the first one of these topics, it is particularly interesting the existence of different ways of defining democracy among poor Peruvians Metropolitan in Lima. Some of definitions were related to somewhat abstract notions such as democracy means that citizens have rights ;9 democracy means equal rights for all citizens ;10 democracy as the respect of the law ;11 and democracy as a government that represents the people .12 Other definitions were much more specific, for example, democracy as freedom of speech or democracy as the right to elect government officials Therefore, it is clear that within this population, there is not consensus regarding the meaning of democracy; however, it is also clear that democracy is a concept that most of the time suggests positive and desirable attributes in a political systems (e.g., citizens having rights, equality for all citizens, the existence of a rule of law, freedom of speech, and a popular election of the authorities). Before addressing the expectations that poor Peruvians in Metropolitan Lima have of democracy and democratic governments, it is critical to note that the fact that democracy as concept is basically associated with positive or desirable attributes, should lead us to review the use survey questions about a general or abstract preference for democracy as a valid form of measuring popular support for democracy in new democracies. It is the case that even for the most disadvantaged population in these 9 Democracy means the respect for th e rights stipulated by the law (participant in one the 1999 focus group sessions). 10 Democracy means equal rights for all and the possibility of claiming for their enforcement (participant in one of the 1999 focus groups sessions). 11 Democracy means respecting and keeping the law (participant in one of the 1999 focus groups sessions). 12 Democracy is when the ruler takes into consideration what are the needs of the people (participant in one of the 1999 focus groups sessions). A democratic government is the one that it is able to listen to the people (participant in one of the 2001 focus group sessions).

PAGE 273

263 countries democracy represents very desirable attributes in a political regime. Therefore, it is not a surprise that a large majority of citizens in these countries have constantly mentioned that they prefer democracy over an authoritarian regime. However, as it has also found in other regions of the world (e.g., sub-Saharan Africa),13 it does not mean that these citizens can be considered as committed democrats.14 Accordingly, the situation is totally different if we consider how individuals in these countries assess the performance of their current democratic governments, institutions, and authorities; and, as a consequence of this, how they develop particular attitudes toward the possibility of using non-democratic means to overthrown democratic governments and reform current democratic regimes. Interestingly, it was the case that some of the participants, whom expressed a clear preference for a democratic regime, showed later on a strong support for the 1992 autogolpe in Peru, and a clear intention to support Fujimori in the next presidential election. Highly related to these different definitions of democracy was the topic about the expectations of democracy. Very frequently, the way democracy was defined by the participants turned out also to be their main expectation. This was particularly the case with democracy as rights equality respect of law and a representative government However, in all these years, some partic ipantsparticularly the poorest onesmentioned also that they expected also living quietly and having a good job15 from democracy. 13 See Bratton, Mattes and Gyimah-Boadi (2005). 14 I am not suggesting here that there is something intrinsically problematic in using public opinion surveys to address popular support for democracy. On the contrary, I am making the case about the urgent need for a more valid operationalization of th is political phenomenon. Something that I think I was able to provide in Chapter 6. 15 Democracy is supposed to provide good and decent jobs for the people (participant in one the 1999 focus group sessions).

PAGE 274

264 Therefore, it is not only the case that poor Peruvians in Metropolitan Lima have diverse expectations of democracy, but also that some of them can be considered particularly problematic for the development of popular support for their democratic regime (e.g., expecting good and decent jobs from democracy). Among other things, because it is clear that democratic governments in developing countries will not be able to deliver, in the near future, a considerable reduction of poverty and/or an important increase in living standards. Nevertheless, it is equally important to note that not all these popular expectations imply material or economic outcomes such as employment and economic progress. Very frequently poor Peruvians in Metropolitan Lima were also expecting things that one would not considered external to a procedural or even minimalist definition of democracy such as the respect for individual rights, and being treated equal before the law. In other words, they were expecting from democracy some things that have been considered as critical conditions or requirements for the development of a competitive and participant political regime (Dahl 1973). These focus groups sessions also provided another way to assess popular expectations of a democratic government in Peru as they addressed also what could be considered, according to the participants, the main responsibilities of a government. As in the case of meanings of democracy, and expectations of democratic governments, there were different opinions about the main governments responsibilities among poor Peruvians living in Lima. Some participants, mainly from the lowest socio-economic level in Peru, mentioned that governments should be responsible for the generation of employment, the

PAGE 275

265 reduction in the cost in living, and a decrease in the current level of poverty. However, other participants, equally worried about the current economic situation in the country, mentioned that the main responsibility of governments was not the generation of employment itself but the provision of a good public education that could prepare young Peruvians to be more competitive in the labor market, and the lack of labor stability. By the same token, it was also mentioned the high level of crime in the country. With regard to the expectations of democracy and democratic governments, the majority of participants from the lowest socio-economic also hold the most demanding attitude in regards to governments responsibilities. In trying to achieve a better understanding of what poor Peruvians think regarding the main responsibilities of their governments, the 2001, 2002, and 2004 focus group sessions are particularly useful. In these years, several questions about which could be considered the better economic system for the country were included. A large majority of these participants expressed a clear preference for something that can be defined as mixed economic system. Additionally, in terms of role of government, there was a clear preference for a regulatory state.16 This mixed economic system would be based on the main components of a free market economy such as competition, and the freedom of consumers to choose their providers of goods and services, but it would include also an active participation of the government in protecting people from monopolistic practices from the recent privatized enterprises, protecting workers rights, supporting popular and micro enterprises, and, if it 16 Privatization have brought good and bad things, but the government needs to find a way to control if the privatized enterprises are charging fair prices (participant in one of the 2001 focus group sessions). Government needs to ch eck what the privatized enterprises are doing (participant in one of the 2002 focus groups sessions).

PAGE 276

266 is necessary, to protect the national industry. Hence, more than asking for a total government control of the economic activity, or a return to the pre-privatization era where governments were in charge of providing basic services and goods, it seems that poor Peruvians, at least in the capital of the country, have a strong preference for a government that accepts a responsibility for economic regulation. As a way of summarizing these findings regarding popular expectations of democracy and democratic governments, I would like to say that it seems that poor people in developing countries tend to have not only in general high expectations but also clear economic expectations (e.g., employment, an improvement in living conditions). However, at the same time, it was the case that not all these expectations have necessarily an economic or material nature. Very frequently, poor Peruvians in Metropolitan Lima were also expecting equality before the law and a government capable of enforcing the existing laws from democracy. In this regard, I believe that these are good news for the maintenance of democracy in poor countries because democratic governments can compensate the lack of a substantive improvement in peoples socio-economic conditions with, for example, the development of an effective and unbiased rule of law. Therefore, according to these findings, it is possible to contend that poor Peruvians in Lima have a very positive attitude toward democracy, and some particular expectations of a democratic regime that there are not unreasonable or unachievable for current democratic governments in developing countries. Unfortunately, the situation was totally different regarding the assessment of their current democratic experience. There is, indeed, a generalized feeling that all democratic governments have fallen very short from

PAGE 277

267 what can be considered an authentic democracy.17 Moreover, as a constant in all these focus groups sessions, several participants expressed either a strong conviction that there is a very long way ahead before Peru could finally reach an authentic democratic government or a strong skepticism regarding the possibility that Peru could reach democracy. But what were the main reasons for this profound dissatisfaction with the democratic governments in Peru among its poorest population? First, one of the main reasons for this dissatisfaction was a generalized feeling regarding the huge gap between what a democratic government was supposed to deliver and what democratic governments were delivering in the last years. Second, and more specifically, there was a strong consensus regarding the lack of an authority in the country. Previous democratic governments were basically perceived as almost incapable of providing effective solutions to the main national problems.18 Again, this feeling regarding the ineffectiveness of democratic regimes was even stronger among poorest Peruvians that participated in these focus group sessions. Some of them, particularly during the 1999 focus groups sessions, mentioned clearly that the existence of a democratic government do not make any difference to their daily life. One of the main sources of this almost total dissatisfaction with democratic governments in Peru was its clear failure in providing solutions for the most important and critical problems during the 1980s and early 1990s (i.e., the economic crisis and the 17 In our country, democracy simply does not fulfill its promises (participant in one of the 1999 focus group sessions). 18 The main problem with democracy in Peru is that decisions are not made with the necessary strength and then things are not got done (participant in one of the 1999 focus group sessions). There is no democracy in Peru because there is no political leadership (participant in one of the 2004 focus groups sessions).

PAGE 278

268 political violence) was one of its main sources. By the same token, at the end of the 1990s and the beginning of the 2000s, there was also a high level of skepticism regarding the possibility that a democratic government could provide some solutions to the emerging and critical countrys problems at this time (i.e., lack of employment and poverty). Moreover, very frequently, this disappointment with the democratic governments was also related to the lack of equality before the law19 and the fact that the rights stipulated by the law are totally ineffective in Peru. In this direction, for some participants in one of the 2004 focus groups sessions, the main problem in the country was not the lack of employments but the lack of an authority that could enforce effectively the respect for the rights of the working people. Therefore, besides some recurrent complaints about the economic situation and the responsibility of the government in this matter, there was also the conviction that one of the main problems with democracy and democratic regimes in Peru is the lack of respect for the law and the capacity to enforce rights.20 Accordingly, some participants in the 1999 focus groups sessions mentioned that the best form of government for their country should be something in the middle a government that combines a democratic and an authoritarian side. The democratic side would guarantee that government would listen to the voice of the people and would rule taking into consideration their needs. The authoritarian side would guarantee a principl e of authority, making possible, among other things, the enforcement of the law and government decisions. 19 One of the main problems in our c ountry is that there is no equality before the law, the socio-economic level of a person will determine how the law will be applied (participant in one the 1999 focus group sessions). 20 The main problem with democratic governments is that they lack authority and leadership (participant in one of the 2004 focus group sessions).

PAGE 279

269 Similarly, several participants in different focus groups sessions addressed their image of what could be considered a good government for the country. According to them, the main feature of a good government would be the capacity to provide solutions to the main current problems in the country. In the same direction, it was also mentioned that a good government would have to be able to enforce the law, especially when the rights of the working or poor people were at stake. Although some participants also considered that allowing people to participate in the elaboration of governments plans was also another feature of a good governments, it was clear that this was not the first thing that poor Peruvians in Metropolitan Lima have in mind when they think in a good government. In the context of several developing countries the aspiration and the need for governments capable of exerting authority over the society and the economy, and being effective in providing solutions to the main countrys problems could lead to a tension between democracy and governance (Levi 2006). Regarding the 1992 democratic breakdown in Peru, there is no doubt that this tension between the capacity to formulate and implement policies in an effective manneron the one handand the deliberate and consensual nature associated with democratic procedureson the other handplayed a critical role (Burt 2004, Carrin 2006). More specifically, in a situation of extreme crisis such as that experienced by Peru in the late 1980s and early 1990s, the urgent need for a resolute leadership to implement economic reforms and to pacify the country seemed to exacerbated the tension between governance and democracy, and to convince to a large majority of Peruvians that under these circumstances the country could not afford neither

PAGE 280

270 a strict adherence to the democratic procedures nor the overhead of a government making decisions based on consensus, consultation, and compromise. Consistently with all of this, almost all participants in these focus groups sessions between 1999 and 2004 considered that the 1992 autogolpe or democratic breakdown was necessary.21 By the same token, they believed that Fujimori represented a good government if not the best government in th e contemporary history of Peru. However, this was not an unqualified contention. Some of these Fujimoris supporters were also very critical of him and his governmental style. They mentioned, for example, his arbitrary use of power; his lack of respect for some laws; the persistence of some economic problems, such as the lack of employment; his close connection with the armed forces, the owners of the main mass communication means, and Vladimiro Montesinos; and having turned his back to the needs of the people at the end of his second term because he was mainly concerned with his 2000 re-election. Therefore, not all components of Fujimoris governments were perceived as positive for poor Peruvian in Lima. Nevertheless, what Fujimori offered in terms of governance or governability,22 and concern for the pueblos problems and needs23 compensated largely any of his shortcomings.24 This suggests to me that poor Peruvians are capable of evaluating different aspect s of any government, including his government 21 At that time the congress was totally ineffective, and even more it was an obstacle for the plans of the president Fujimori (participant in one of the 1999 focus groups sessions). 22 Fujimori fulfilled what he offered; he defeated terrorism, provided security, and built many schools (participant in one of the 2000 focus groups sessions). Fujimori kept order in the country, and that was important (participant in one of the 2004 focus groups sessions). 23 Fujimori is close to the people (participant in one of the 2000 focus groups sessions). He [Fujimori] understood the problems of the country (participant in one the 2004 focus group sessions). 24 This explains why in both the 1999 and 2000 wave of focus group sessions, a majority of poor Peruvians in Lima said that they would vote for Fujimori in the next election.

PAGE 281

271 capacity and democratic nature. Nevertheless, they are also capable of reaching an overall and final evaluation about the performance of their governments. Accordingly, the popular perception regarding how Fujimori was capable of dealing with the most critical issues or problems of the moment exerted the biggest influence on the overall evaluation of their governments.25 In other words, there was an almost absolute consensus that Fujimori was able to defeat terrorism and delivering peace and order, stabilizing the economy, and supporting poorest people thorough the construction of infrastructure (e.g., roads and schools). Therefore, it was not the case that, during the 1990s, democracy was not valued under Fujimori, but it constituted a secondor third-order priority. Then, what it is possible to contend about the way poor Peruvians in Metropolitan Limaand eventually poor Peruvians in Peru and similar countriesthink about democracy and their democratic experiences? And, what it is possible to contend about how these perceptions can shape ordinary citizens political behavior in a context where an authoritarian leader is challenging the democratic regime? First, there is evidence that poor Peruvians in Metropolitan Lima perceive several advantages in having a democratic regime. For them, democracy basically means the recognition of rights by the governments and authorities, and the possibility of claiming for them; a political system where the law is respected, and, consequently, the possibility that the law will be enforce in the same way for all citizens; a government that is supposed to take into consideration the needs and concerns of its citizens; and very specific rights such as freedom of speech, and electing the authorities. 25 Fujimori is a mixed of good and bad things, but the balance is more positive and negative; and I think also that he would be able to find solutions for the pr oblems that our country still have (participant in one of the 1999 focus group sessions).

PAGE 282

272 Second, it seems that not many poor Peruvians in Metropolitan Lima think that democracy exists in their country. Even more, there are serious doubts regarding the possibility of a having a real democratic system in Peru. By the same token, it seems also that there is a generalized lack of satisfac tion with the performance of their democratic governments, and particularly with the performance of their democratic authorities and institutions. The main complaint is that, since the return of democracy in 1980, all democratic governments have shown a very limited capacity to find solutions to the main problems of the country (e.g., economic deterioration, political violence, poverty, inequality, etc.), as well as a very low capacity to effectively enforce the laws that already exist in Peru. As I mentioned before, this popular demand for an effective government and for a government with enough authority to enforce laws can be problematic for the maintenance of democracy in a developing country. Being the case that in some contextsparticularly the ones characterized by deep economic and political crisis ordinary citizens can perceive a clear trade-off between democracy and governance. Regarding the recent Peruvian political history, there is no doubt that the perception of this trade-off facilitated enormously the development of a successful neopopulist leader and the development of an electoral authoritarian regime (Weyland 2006, Mauceri 2006, Carrin 2006). Moreover, and considering the main challenges for the maintenance and development of democracy in developing countries, it is critical to acknowledge that there is even a stronger connection between the most disadvantaged citizens and the expectation and demand for an effective government and an authority. It seems that poor

PAGE 283

273 citizens in developing countries such as Peru are in greatest need of a government and authorities capable of providing solutions to the main problems of the country, and enforcing effectively the laws and rights that already exist. Accordingly, it is not a big surprise that for poor Peruvians in Metropolitan Lima Fujimoris governments stood out in comparison to previous governments; mainly because they deviated from this pattern of government ineffectiveness and lack of authority to enforce the law. This was even more significant in the critical juncture that Peru faced at the end of the 1980s and early 1990s. Therefore, it is not a big surprise that despite his authoritarian style, Fujimori was re-elected in 1995, supported by a large number of Peruvians in 2000, and that, even at the beginning of 2006, he appeared in public opinions surveys having a considerable electoral support. In a similar vein, it seems that what poor Peruvians mainly expect from a governmentwhat they could consider as a good governmentis some improvements regarding two of the main problems of the country, the lack of employment and an extensive poverty. But they expect also from a government things that are less problematic to deliver such as an improvement in the quality of public education, the use of more resources to fight crime, and some economic regulation, especially in regards to the private enterprisesformerly public onesthat provide basic services such as electricity and telephone. Equally important, they expect a government capable of enforcement individual rights, in particular labor rights, and to address the current situation characterized by the lack of equality before law. Those are, no doubt, huge expectations toward government and democratic governments. But they provide also

PAGE 284

274 different and several opportunities for democratic governments to prove to their citizens that they can be also able to deliver an effective government. Additionally, there is also evidence regarding a critical difference regarding how poor Peruvians in Lima could react to some non-democratic or authoritarian political leaders. For some of them basically the dramatic situation of the late 1980s and early 1990s justified the 1992 democratic breakdown and the installation of a decisive and authoritarian government. This makes me think that once the severe economic and political crisis disappeared; their support for this non-democratic style of government could also disappear. However, for other participants, not only the 1992 democratic breakdown was necessary, but a mano dura government will be always necessary in their country. According to this point of view, democracy will be never capable of providing a principle of authority. By the same token, this principle of authority was perceived as absolutely necessary in order to able to address the main problems of the country and the main problems of the most disadvantaged population. Finally, and although it is always problematic to make generalizations based on focus group data about the distribution of th ese two different attitudes toward democracy, it was the case that much more frequently were the participants from the lowest socioeconomic levels in Metropolitan Lima (i.e., the poorest population) the ones that subscribed the most skeptical about democracy. Even more, some of these participants clearly stated a fundamental contradiction between being poor and to be able to access to the benefits that democracy may offer to the citizens.26 26 Democracy only works for wealthy people, not for us (participant in one of the 1999 focus groups sessions). Only rich people is complaining by the lack of democracy with Fujimori (participant in one of the 2000 focus groups).

PAGE 285

275 Therefore, it is not the case that democratic breakdowns are unavoidable in Peru and similar countries. However, it is also clear that in a context where democratic governments are not be capable of being minimally effective in addressing main countrys problems, and where democratic government seem not able or not willing to be minimally accountable to some reasonable popular expectations in terms of its social and economic responsibilities, it will be nave to expect poor citizens to be supportive of their democratic political regimes. This is even more critical in a country where poor citizens do not necessarily have a clear distinction between the democratic regime and the democratic governments, and where these disadvantaged citizens seem not convinced that changing the government following the democratic procedures could be a solution for their concerns and problems.

PAGE 286

276 LIST OF REFERENCES Abramson, Paul R., and Ronald Inglehart. 1995. Value Change in Global Perspective Ann Arbor: University of Michigan Press. Agresti, Alan. 1996. An Introduction to Categorical Data Analysis. New York: Wiley. Agresti, Alan, and Barbara Finlay. 1997. Statistical Methods for the Social Sciences 3rd ed. Upper Saddle River: Prentice Hall. Aguayo, Julio Javier. 2004. The Legislature Strikes Back in Peru the Role of Congress in the Demise of Fujimori in 2000. University of Florida. Retrieved January 20, 2006 from http://purl.fcla.edu/fcla/etd/UFE0007840. Alexander, Gerard. 2002. The Sources of Democratic Consolidation Ithaca: Cornell University Press. Almond, Gabriel A. 1980. The Intellectual History of the Civic Culture Concept. In Gabriel A. Almond and Sidney Verba, eds., The Civic Culture Revisited: An Analytic Study Boston, MA: Little, Brown. Almond, Gabriel A. 1990. A Discipline Divided: Schools and Sects in Political Science Newbury Park: Sage Publications. Almond, Gabriel A. 1994. Foreword: The Return of Political Culture. In Larry J. Diamond, ed., Political Culture and Democracy in Developing Countries. Boulder, CO: Lynne Rienner. Almond, Gabriel A., and Sidney Verba. 1963. The Civic Culture: Political Attitudes and Democracy in Five Nations Princeton: Princeton University Press. Almond, Gabriel A., and G. Bingham Powell. 1978. Comparative Politics: System, Process, and Policy 2d ed. Boston: Little, Brown. Almond, Gabriel A., and Sidney Verba. 1980. The Civic Culture Revisited: An Analytic Study Boston: Little, Brown. Anderson, C. J., and C. A. Guillory. 1997. Political Institutions and Satisfaction with Democracy: A Cross-national Analysis of Consensus and Majoritarian Systems. American Political Science Review 91 (1): 66.

PAGE 287

277 Anderson, Leslie E. 2005. Idealism, Impatience, and Pessimism: Recent Studies of Democratization in Latin America. Latin American Research Review 40 (3): 390 402. Anderson, Leslie E., and Jorge Aragon. 2003. What Difference Could a Revolution Make: Political Values in Nicaragua and Peru. Paper Presented at the Twentyfourth Latin American Studies Association Congress. Anderson, Leslie E., and Lawrence C. Dodd. 2005. Learning democracy: citizen engagement and electoral choice in Nicaragua, 1990 Chicago: University of Chicago Press. Andrade, Xavier. 2003. Adis Cultura y Ha sta la Vista Cultura Poltica: Sobre el Tratamiento Sociolgico del Regionalismo y Populismo en Ecuador. In Felipe Burbano de Lara, ed., Democracia, Gobernabilidad y Cultura Poltica Quito, Ecuador: FLACSO Sede Ecuador. Apoyo Opinin y Mercado. 2000. Actitudes hacia la Poltica y el Proceso Electoral Lima: Apoyo Opinin y Mercado. Apoyo Opinin y Mercado. 2001. Actitudes hacia las Elecciones Presidenciales Lima: Apoyo Opinin y Mercado. Apoyo Opinin y Mercado. 2004. La Opinin Pblica ante la Situacin del Pas y las Prximas Elecciones Presidenciales Lima: Apoyo Opinin y Mercado. Apoyo Opinin y Mercado. 2004. Niveles Socioeconmicos en Lima Metropolitana Lima: Apoyo Opinin y Mercado. Aragn, Jorge. 2001. Attitudes and Perceptions toward Democracy and Politics in Peru Thesis (M.A.). University of Florida. Aragn, Jorge. 2006. Elecciones 2006: Democracia vs. Autoritarismo. Argumentos: Coyuntura Electoral 1 (5): 6. Lima: Instituto de Estudios Peruanos. Barr Robert R., and Henry Dietz. 2006. Fujimori and the Mayors of Lima, 1990: The Impact and Legacy of Neopopulist Rule. In Julio Carrin, ed., The Fujimori Legacy: the Rise of Electoral Authoritarianism in Peru University Park, PA: Pennsylvania State University Press. Bartholomew, David J., Fiona Steele, Irini Moustaki, and Jane I. Galbraith. 2002. The Analysis and Interpretation of Multivariate Data for Social Scientists Boca Raton: Chapman & Hall/CRC. Bayer, Patricia, and John A. Booth. 2000. Civil Society and Democratic Transition. In Thomas W. Walker, and Ariel C. Armony, eds., Repression, resistance, and democratic transition in Central America Wilmington, DE: Scholarly Resources.

PAGE 288

278 Beetham, David. 2004. Freedom as the Foundation. Journal of Democracy 15 (4): 61. Bermeo, Nancy G. 2003. Ordinary People in Extraordinary Times: the Citizenry and the Breakdown of Democracy Princeton: Princeton University Press. Bernard, Russell H. 2002. Research Methods in Anthropology: Qualitative and Quantitative Approaches Walnut Creek: Altamira Press. Biryukov, Nikolai, and Victor Sergeyev. 1994. T he Idea of Democracy in the West and in the East. In David Beetham, ed., Defining and Measuring democracy Sage modern politics series; v. 36 Thousand Oaks, CA: Sage Publications. Booth, John A., and Mitchell A. Seligson. 1994. Paths to Democracy and the Political Culture of Costa Rica. In Larry J. Diamond, ed., Political Culture and Democracy in Developing Countries. Boulder, CO: Lynne Rienner. Bratton, Michael, and Nicolas Van de Walle. 1997. Democratic Experiments in Africa: Regime Transitions in Comparative Perspective New York: Cambridge University Press. Bratton, Michael, Robert Mattes, and E. Gyimah-Boadi. 2005. Public Opinion, Democracy, and Market Reform in Africa Cambridge: Cambridge University Press. Bunce, Valerie. 2000. Comparative Democratization: Big and Bounded Generalizations. Comparative Political Studies 33 (6): 703. Bunce, Valerie. 2003. Rethinking Recent Democratization: Lessons from the Postcommunist Experience. World Politics 55 (2): 167. Burbano de Lara, Felipe. 2003. Democracia, Cultura Poltica y Gobernabilidad: Los Estudio Polticos de los Aos Noventa. In Felipe Burbano de Lara, ed., Democracia, Gobernabilidad y Cultura Poltica Quito, Ecuador: FLACSO Sede Ecuador. Burt, Jo-Marie. 2004. State Making against Democracy. In Jo-Marie Burt and Philip Mauceri, eds., Politics in the Andes: Identity, Conflict, Reform Pittsburgh, PA: University of Pittsburgh Press. Bustamante, Fernando. 2003. Cultura Poltica y Ciudadana en Ecuador. In Felipe Burbano de Lara, ed., Democracia, Gobernabilidad y Cultura Poltica Quito, Ecuador: FLACSO Sede Ecuador. Cameron, Maxwell A. 1994. Democracy and Authoritarianism in Peru: Political Coalitions and Social Change New York: St. Martin's Press.

PAGE 289

279 Cameron, Maxwell A. 1997. Political and Economic Origins of Regime Change in Peru: The Eighteenth Brumaire of Alberto Fujimori. In Maxwell A. Cameron and Philip Mauceri, eds., The Peruvian Labyrinth: Polity, Society, Economy University Park, PA: Pennsylvania State University Press. Cameron, Maxwell A., and Philip Mauceri, eds., The Peruvian Labyrinth: Polity, Society, Economy University Park, PA: Pennsylvania State University Press. Camp, Roderic A. 2001. Democracy Through Latin American Lenses: An Appraisal. In Roderic A. Camp, ed., Citizen Views of Democracy in Latin America Pittsburgh, PA: University of Pittsburgh Press. Camp, Roderic A. 2001. Citizen Views of Democracy in Latin America Pittsburgh: University of Pittsburgh Press. Camp, Roderic A. 2003. Politics in Mexico: the Democratic Transformation 4th ed. New York: Oxford University Press. Canache, Damarys, Jeffery J. Mondak, and Mitchell Seligson. 2001. Meaning and Measurement in Cross-National Research on Satisfaction with Democracy. Public Opinion Quarterly 65 (4): 506. Carothers, Thomas. 2002a. The End of the Transition Paradigm. Journal of Democracy 13 (1): 5. Carothers, Thomas. 2002b. A Reply to my Critics. Journal of Democracy 13 (3): 33 38. Carrin, Julio. 2006. Introduction. In Julio Carrin, ed., The Fujimori Legacy: the Rise of Electoral Authoritarianism in Peru University Park, PA: Pennsylvania State University Press. Carrin, Julio. 2006. Public Opinion, Market Reforms, and Democracy is Fujimoris Peru. In Julio Carrin, ed., The Fujimori Legacy: the Rise of Electoral Authoritarianism in Peru University Park: Pennsylvania State University Press. Carrin, Julio. 2006. The Fujimori Legacy: the Rise of Electoral Authoritarianism in Peru University Park, PA: Pennsylvania State University Press. Childers, Thomas. 1983. The Nazi Voter: The Social Foundations of Fascism in Germany, 1919 Chapel Hill: The University of North Carolina Press. Clarke, H. D., N. Dutt, and A. Kornberg. 1993. The Political Economy of Attitudes toward Polity and Society in Western-European Democracies. Journal of Politics 55 (4): 998. Collier, David, and Steven Levitsky. 1997. Democracy with Adjectives: Conceptual Innovation in Comparative Research. World Politics 49 (3): 430.

PAGE 290

280 Conaghan, Catherine M. 2005. Fujimori's Peru: Deception in the Public Sphere Pittsburgh: University of Pittsburgh Press. Converse, Philip E. 1964. The Nature of Belief Systems in Mass Public. In David E. Apter, ed., Ideology and Discontent New York, NY: Free Press. Cotler, Julio. 1978. Clases, Estado y Nacin en el Per 1st ed. Lima, Peru: Instituto de Estudios Peruanos. Cotler, Julio. 1994. Poltica y Sociedad en el Per: Cambios y Continuidades 1st ed. Lima, Peru: Instituto de Estudios Peruanos. Cotler, Julio. 2000. La Gobernabilidad en el Per: Entre el Autoritarismo y la Democracia. In Julio Cotler, and Romeo Grompone, eds., El Fujimorismo: Ascenso y Cada de un Rgimen Autoritario Lima, Peru: Instituto de Estudios Peruanos. Craig, Stephen C. 1993. The Malevolent Leaders: Popular Discontent in America Boulder: Westview Press. Dahl, Robert A. 1966. Political Oppositions in Western Democracies New Haven: Yale University Press. Dahl, Robert A. 1971. Polyarchy: Participation and Opposition. New Haven: Yale University Press Dalton, Russell J. 1999. Political Support in Advanced Industrial Democracies. In Pippa Norris, ed., Critical Citizens: Global Support for Democratic Government New York, NY: Oxford University Press. Dalton, Russell J. 2004. Democratic Challenges, Democratic Choices: The Erosion of Political Support in Advanced Industrial Democracies Oxford: Oxford University Press. Dealy, Glen C. 1977. The Public Man: An Interpretation of Latin American and Other Catholic Countries Amherst: University of Massachusetts Press. Dealy, Glen C. 1992. The Latin Americans: Spirit and Ethos Boulder: Westview Press. Degregori, Carlos Ivn. 2000. La Dcada de la Antipoltica: Auge y Huida de Alberto Fujimori y Vladimiro Montesinos Lima: Instituto de Estudios Peruanos. De la Torre, Carlos. 2000a. Populist Seduction in Latin America. The Ecuadorian Experience Athens, Ohio University Press. De la Torre, Carlos. 2000b. Veinte Aos de Populismo y Democracia. Revista ICONOS 9: 22.

PAGE 291

281 Di Palma, Giuseppe. 1990. To Craft Democracies: An Essay on Democratic Transitions Berkeley: University of California Press. Diamond, Larry J. 1994. Introduction to Political Culture and Democracy. In Larry J. Diamond, ed., Political Culture and Democracy in Developing Countries. Boulder, CO: Lynne Rienner. Diamond, Larry J. 1999. Developing Democracy: Toward Consolidation Baltimore: Johns Hopkins University Press. Diamond, Larry J. 2002. Thinking about Hybrid Regimes. Journal of Democracy 13 (2): 21. Diamond, Larry J., Juan J. Linz, and Seymour Martin Lipset. 1989. Democracy in Developing Countries: Latin America Boulder: L. Rienner Publishers. Diamond, Larry J., Juan J. Linz, and Seymour Martin Lipset. 1995. Politics in Developing Countries: Comparing Experiences with Democracy 2nd ed. Boulder: L. Rienner Publishers. Diamond, Larry J., and Marc F. Plattner. 1996. The Global Resurgence of Democracy 2nd ed. Baltimore: Johns Hopkins University Press. Diamond, Larry J. and Leonardo Morlino. 2004. The Quality of Democracy: An Overview, Journal of Democracy 15 (4): 20. Dietz, Henry A. 1998. Peru since 1990. Latin American Research Review 33 (2): 197 206. Duch, Raymond M. 1993. Tolerating Economic Reform: Popular Support for Transitions to a Free Market in the Former Soviet Union. American Political Science Review 87 (3): 590. Easton, David. 1965. A systems analysis of political life New York: Wiley. Easton, David. 1975. A Reassessment of the Concept of Political Support. British Journal of Political Science 5: 435. Easton, David. 1976. Theoretical A pproaches to Political Support. Canadian Journal of Political Science 9 (3):431. Encarnacion, Omar. G. 2000. Beyond Transitions: The Politics of Democratic Consolidation. Comparative Politics 32 (4):479. Encarnacion, Omar G. 2003. The Strange Persistence of Latin American Democracy. World Policy Journal 20 (4): 30.

PAGE 292

282 Evans, Geoffrey, and Stephen Whitefield. 1998. The Evolution of Left and Right in Post-Soviet Russia. Europe-Asia Studies 50 (6): 1023. Finifter, Ada W. 1996. Attitudes Toward Individual Responsibility and Political Reform in the Former Soviet Union. American Political Science Review 90 (1):138. Finifter, Ada W., and Ellen Mickiewicz. 1992. Redefining the Political System of the USSR: Mass Support for Political Change. American Political Science Review 86 (4): 857. Fish, Steven M. 2005. Democracy Derailed in Russia: The Failure of Open Politics New York: Cambridge University Press. Fleron, Frederic J. Jr., and Richard Ahl. 1998. Does the Public Matter for Democratization in Russia? What We Have Learned from Third Wave Transitions and Public Opinion Surveys. In Harry Eckstein, Frederic J. Fleron Jr., Erik P. Hoffman, and William Reisinger, eds., Can Democracy Take Root in PostSoviet Russia? Explorations in State-Society Relations Lanham, MD: Rowman & Littlefiled Publishers, Inc. Foweraker, Joe, and Roman Krznaric. 2002. The Uneven Performance of Third Wave Democracies: Electoral Politics and the Imperfect Rule of Law in Latin America. Latin American Politics And Society 44 (3): 29. Fox, John. 1997. Applied Regression Analysis, Linear Models, and Related Methods Thousands Oaks: Sage Publications, Inc. Franco, Carlos. 1998. Acerca del Modo de Pensar la Democracia en Amrica Latina Lima: Fundacin Friedrich Ebert Stiftung Fuchs, Dieter. 1999. The Democratic Culture of Unified Germany. In Pippa Norris, ed., Critical Citizens: Global Support for Democratic Government New York, NY: Oxford University Press. Garretn Merino, Manuel A. 2003. Incomplete Democracy: Political Democratization in Chile and Latin America Chapel Hill: University of North Carolina Press. Gasiorowski, Mark J., and Timothy J. Power. 1998. The Structural Determinants of Democratic Consolidation: Evidence from the Third World. Comparative Political Studies 31 (6): 740. Geddes, Barbara. 1999. What do We Know about Democratization after Twenty Years? Annual Review of Political Science 2: 115. Gibson, James L. 1996. Political and Economic Markets: Changes in the Connections Between Attitudes Toward Political Democracy and a Market Economy Within the Mass Culture of Russia and Ukraine. Journal of Politics 58 (4): 954.

PAGE 293

283 Gibson, James L. 2001. The Russian Dance with Democracy. Post-Soviet Affairs 17 (2): 101. Gibson, J. L., Raymond M. Duch, and Kent L. Tedin. 1992. Democratic Values and the Transformation of the Soviet-Union. Journal of Politics 54 (2): 329. Gibson, James L., and Raymond Duch. 1993. Political Intolerance in the USSR: The Distribution and Etiology of Mass Opinion. Comparative Political Studies 26 (3): 286. Grimm, Laurence G., and Paul R. Yarnold. 2000. Reading and Understanding More Multivariate Statistics Washington: American Psychological Association. Grompone, Romeo. 2000. Al Da Siguiente: El Fujimorismo como Proyecto Inconcluso de Transformacin Poltica y Social. In Julio Cotler, and Romeo Grompone, eds., El Fujimorismo: Ascenso y Cada de un Rgimen Autoritario Lima, Peru: Instituto de Estudios Peruanos. Gunther, Richard, Nikiforos P. Diam andouros, and Hans-Jrgen Puhle. 1995. The Politics of Democratic Consolidation: Southern Europe in Comparative Perspective Baltimore: The Johns Hopkins University Press. Hagopian, Frances. 2006. Conclusions: Government Performance, Political Representation, and Public Perceptions of Contemporary Democracy in Latin America. In Frances Hagopian and Scott P. Mainwaring, eds., The Third Wave of Democratization in Latin America: Advances and Setbacks New York, NY: Cambridge University Press. Hahn, Jeffrey W. 1993. Continuity and Change in Russian Political Culture. In Frederic J. Fleron Jr., and Erik P. Hoffmann, eds., Post-Communist Studies and Political Science: Methodology and Empirical Theory in Sovietology Boulder, CO: Westview. Huntington, Samuel P. 1991. The Third Wave: Democratization in the Late Twentieth Century. Norman: University of Oklahoma Press. Hyden, Goran. 2006. African Politics in Comparative Perspective New York: Cambridge University Press. Inglehart, Ronald. 1977. The Silent Revolution: Changing Values and Political Styles among Western Publics Princeton: Princeton University Press. Inglehart, Ronald. 1988. The Renaissance of Political Culture. American Political Science Review 82 (4): 1203. Inglehart, Ronald. 1990. Culture Shift in Advanced Industrial Society Princeton: Princeton University Press.

PAGE 294

284 Inglehart, Ronald. 1997. Modernization and Postmodernization: Cultural, Economic, and Political Change in 43 societies Princeton: Princeton University Press. Inglehart, Ronald. 1999. Postmodernization Er odes Respect for Authority, but Increases Support for Democracy. In Pippa Norris, ed., Critical Citizens: Global Support for Democratic Government New York, NY: Oxford University Press. Inglehart, Ronald. 2003. How Solid is Mass Support for Democracy and How Can We Measure It? PS: Political Science & Politics 36 (1): 51. Kachigan, Sam Kash. 1991. Multivariate Statistical Analysis: A Conceptual Introduction 2nd ed. New York: Radius Press. Karl, Terry L. 1990. Dilemmas Of Democratization In Latin America. Comparative Politics 23 (1): 1. Kenney, Charles. 2004. Fujimori's Coup and the Breakdown of Democracy in Latin America Notre Dame: University of Notre Dame Press. Klingemann, Hans-Dieter. 1999. Mapping Political Support in the 1990s: A Global Analysis. In Pippa Norris, ed., Critical Citizens: Global Support for Democratic Government New York, NY: Oxford University Press. Kornberg, Allan, and Harold D. Clarke. 1992. Citizens and Community: Political Support in a Representative Democracy. New York: Cambridge University Press. Lagos, Marta. 1997. Latin America's Smiling Mask. Journal of Democracy 8 (3): 125 138. Lagos, Marta. 2001. Between Stability and Crisis in Latin America. Journal of Democracy 12 (1): 137. Lagos, Marta. 2003a. A Road with No Return? Journal of Democracy 14 (2): 163. Lagos, Marta. 2003b. Support for and Satisfaction with Democracy. International Journal of Public Opinion Research 15 (4): 471. Landman, Todd. 2000. Issues and Methods in Comparative Politics: An Introduction New York: Routledge. Lane, Ruth. 1992. Political Culture: Residual Category or General Theory. Comparative Political Studies 25 (3): 362. Levi, Margaret. 2006. Why We Need a New Theory of Government. Perspectives on Politics 4 (1): 5. Levine, Daniel. H. 1988. Paradigm Lost: Dependence to Democracy. World Politics 40 (3): 377.

PAGE 295

285 Levitsky, Steven, and Lucan A. Way. 2002. The Rise of Competitive Authoritarianism. Journal of Democracy 13 (2): 51. Levitsky, Steven, and Maxwell A. Cameron. 2003. Democracy Without Parties? Political Parties and Regime Change in Fujimori's Peru. Latin American Politics and Society 45 (3): 1. Linz, Juan J. 1994. Presidential or Parliamentary Democracy: Does It Make a Difference. In Juan J. Linz and Arturo Valenzuela, eds., The Failure of Presidential Democracy Baltimore, MD: Johns Hopkins University Press. Linz, Juan J., and Alfred C. Stepan. 1978. The Breakdown of Democratic Regimes Baltimore: Johns Hopkins University Press. Linz, Juan J., and Alfred C. Stepan. 1996. Problems of Democratic Transition and Consolidation: Southern Europe, South America, and Post-communist Europe Baltimore: Johns Hopkins University Press. Lipset, Seymour M. 1959. Some Social Requisites of Democracy: EconomicDevelopment and Political Legitimacy. American Political Science Review 53 (1): 69. Lipset, Seymour M. 1981. Political Man: The Social Bases of Politics Baltimore: Johns Hopkins University Press. Listhaug, Olga. 1995. The Dynamics of Trust in Politicians. In Hans-Dieter Klingemann, and Dieter Fuchs, eds., Citizens and the State : Beliefs in Government; v. 1 New York: Oxford University Press. Long, Scott J. and Jeremy Freese. 2006. Regression Models for Categorical Dependent Variables Using Stata Texas: Stata Press. Mainwaring, Scott P. 1992. Transitions to Democracy and Democratic Consolidation: Theoretical and Comparative Issues. In Scott Mainwaring, Guillermo ODonnell, and J. Samuel Valenzuela, eds., Issues in Democratic Consolidation: The New South American Democracies in Comparative Perspective Notre Dame, IN: University of Notre Dame Press. Mainwaring, Scott P. and Frances Hagopia n. 2006. Introduction: The Third Wave of Democratization in Latin America. In Frances Hagopian and Scott P. Mainwaring, eds., The Third Wave of Democratization in Latin America: Advances and Setbacks New York, NY: Cambridge University Press. Mainwaring, Scott and Anbal Prez-Lin. 2006. Latin American Democratization since 1978: Democratic Transitions, Breakdowns, and Erosions. In Frances Hagopian and Scott P. Mainwaring, eds., The Third Wave of Democratization in Latin America: Advances and Setbacks New York, NY: Cambridge University Press.

PAGE 296

286 Mason, David S. 1995. Attitudes toward Market and Political Participation in the Postcommunist States. Slavic Review 54 (2): 385. Mason, David S., James R. Kluegel, and Ludmila A. Khakhulina. 2000. Marketing Democracy: Changing Opinion About Inequality and Politics in East Central Europe Lanham: Rowman & Littlefield Publishers. Mauceri, Philip. 1997. The Transition to Democracy and the Failures of Institution Building. In Maxwell A. Cameron and and Philip Mauceri, eds., The Peruvian Labyrinth: Polity, Society, Economy University Park, PA: Pennsylvania State University Press. Mauceri, Philip. 2006. An Authoritarian Presidency: How and Why Did Presidential Power Run Amok in Fujimoris Peru? In Julio Carrin, ed., The Fujimori Legacy: the Rise of Electoral Authoritarianism in Peru University Park, PA: Pennsylvania State University Press. McAllister, Ian. 1999. The Economic Performance of Governments. In Pippa Norris, ed., Critical Citizens: Global Support for Democratic Government New York, NY: Oxford University Press. McClintock, Cynthia. 1994a. Presidents, Messiahs, and Constitutional Breakdowns in Peru. In In Juan J. Linz and Arturo Valenzuela, eds., The Failure of Presidential Democracy Baltimore, MD: Johns Hopkins University Press. McClintock, Cynthia. 1994b. The Breakdown of Constitutional Democracy in Peru Atlanta: Latin American Studies Association. McClintock, Cynthia, and Abraham Lowenthal. 1997. Foreword. In Maxwell A. Cameron and and Philip Mauceri, eds., The Peruvian Labyrinth: Polity, Society, Economy University Park, PA: Pennsylvania State University Press. McCoy, Jennifer, and David J. Myers. 2004. The Unraveling of Representative Democracy in Venezuela Baltimore: Johns Hopkins University Press. McFaul, Michael. 2000. One Step Forward, Two Steps Back. Journal of Democracy 11 (3): 19. McFaul Michael, Nikolai Petrov and Andrei Ryabov. 2004. Between Dictatorship and Democracy: Russian Post-Communist Political Reform Washington: Carnegie Endowment for International Peace. McInstosh Mary E., Martha Abele Mac Iver Daniel G. Abele, and Dina Smeltz. 1994. Publics Meet Market Democracy in Central and Eastern Europe, 1991. Slavic Review 53 (2): 483

PAGE 297

287 Miller, Arthur H., Vicky L. Hesli, and William M. Reisinger. 1994. Reassessing Mass Support for Political and Economic Change in the Former USSR. American Political Science Review 88 (2): 399. Miller, Arthur H., Vicky L. Hesli, and William M. Reisinger. 1995. Comparing Citizen and Elite Belief Systems in Post-Soviet Russia and Ukraine. Public Opinion Quarterly 59 (1): 1. Miller, Arthur H., Vicky L. Hesli, and William M. Reisinger. 1997. Conceptions of Democracy Among Mass and Elite in Post-Soviet Societies. British Journal of Political Science 27: 157. Miller, Arthur H., William M. Reisinger, and Vicky L. Hesli. 1996. Understanding Political Change in Post-Soviet Societie s: A Further Commentary on Finifter and Mickiewicz. American Political Science Review 90 (1): 153. Miller, Arthur, and Ola Listhaug. 1999. Political Performance and Institutional Trust. In Pippa Norris, ed., Critical Citizens: Global Support for Democratic Government New York, NY: Oxford University Press. Mishler, William, and Richard Rose. 1996. Trajectories of Fear and Hope: Support for Democracy in Post-Communist Europe. Comparative Political Studies 28 (4): 553. Mishler, William, and Richard Rose. 1999. Five Years After the Fall: Trajectories of Support for Democracy in Post-Communist Europe. In Pippa Norris, ed., Critical Citizens: Global Support for Democratic Government New York, NY: Oxford University Press. Mishler, William, and Richard Rose. 2001. Political Support for Incomplete Democracies: Realist vs. Idealist Theories and Measures. International Political Science Review 22 (4): 303. Mishler, William, and Richard Rose. 2002. Learning and Re-learning Regime Support: The Dynamics of Post-Communist Regimes. European Journal of Political Research 41 (1): 5. Moreno, Alejandro. 2001. Democracy and Mass Belief Systems in Latin America. In Roderic A. Camp, ed., Citizen Views of Democracy in Latin America Pittsburgh, PA: University of Pittsburgh Press. Moreno, Alejandro, and Patricia Mendez. 2002. Attitudes Toward Democracy: Mexico in Comparative Perspective. International Journal of Comparative Sociology 43 (3 5): 350. Muller, Edward N., and Mitchell A. Seligson. 1994. Civic Culture and Democracy: The Question of Causal Relationships. American Political Science Review 88 (3): 635 652.

PAGE 298

288 Murakami, Yusuke. 2000. La Democracia Segn C y D: Un Estudios de la Conciencia y el Comportamiento Poltico de los Sectores Populares de Lima Lima: Instituto de Estudios Peruanos. Newton, Kenneth. 1999. Social and Political Trust in Established Democracies. In Pippa Norris, ed., Critical Citizens: Global Support for Democratic Government New York, NY: Oxford University Press. Newton, Kenneth, and Pippa Norris. 2000. Confidence in Public Institutions: Faith, Culture, or Performance. In Susan J. Pharr and Robert D. Putnam, eds., Disaffected Democracies: Whats Troubling the Trilateral Democracies? Princeton, NJ: Princeton University Press. Neyra, Hugo. 2001. El Mal Peruano: 1990 2001 Lima: SIDEA. Norris, Pippa. 1999a. Introduction: The Growth of Critical Citizens? In Pippa Norris, ed., Critical Citizens: Global Support for Democratic Government New York, NY: Oxford University Press. Norris, Pippa. 1999b. Institutional Explanations for Political Support. In Pippa Norris, ed., Critical Citizens: Global Support for Democratic Government New York, NY: Oxford University Press. Norris, Pippa. 1999c. Conclusions: The Growth of Critical Citizens and Its Consequences. In Pippa Norris, ed., Critical Citizens: Global Support for Democratic Government New York, NY: Oxford University Press. Nunnally, Jum C. 1978. Psychometric Theory 2d ed. New York: McGraw-Hill. Nye, Joseph S., and Zelikow, Philip. 1997. Conc lusion: Reflections, Conjectures, and Puzzles. In Joseph S. Nye, Philip Zelikow, and David C. King, eds., Why People Don't Trust Government Cambridge, MA: Harvard University Press. O'Donnell, Guillermo. 1973. Modernization and Bureaucratic Authoritarianism Berkeley: Institute of International Studies University of California. ODonnell, Guillermo. 1992. Transitions, Continuities, and Paradoxes. In Scott Mainwaring, Guillermo ODonnell, and J. Samuel Valenzuela, eds., Issues in Democratic Consolidation: The New South American Democracies in Comparative Perspective Notre Dame, IN: University of Notre Dame Press. ODonnell, Guillermo. 1994. Delegative Democracy. Journal of Democracy 5 (1): 55 69 ODonnell, Guillermo. 1996. Illusions about Consolidation. Journal of Democracy 7(2): 34.

PAGE 299

289 O'Donnell, Guillermo. 1998. Horizontal Accountability in New Democracies. Journal of Democracy 9 (3): 112. O'Donnell, Guillermo. 2000. The Judiciary and the Rule of Law. Journal of Democracy 11 (1): 25. O'Donnell, Guillermo. 2002. In Partial Defense of an Evanescent Paradigm. Journal of Democracy 13 (3): 6. O'Donnell, Guillermo. 2004a. Why the Rule of Law Matters. Journal of Democracy 15 (4): 32. ODonnell, Guillermo. 2004b. Presentacin: El Debate Conceptual Sobre la Democracia. In Programa de las Naciones Unidas para el Desarrollo, La Democracia en Amrica Latina. Hacia una Democracia de Ciudadanos y Ciudadanas: El Debate Conceptual sobre la Democracia Retrieved September 5, 2004 from http://democracia.undp.org/Default.asp. ODonnell, Guillermo. 2004c. Human Development, Human Rights, and Democracy. In Guillermo ODonnell, Jorge Vargas Cullel, and Osvaldo Miguel Iazzetta. 2004. The Quality of Democracy: Theory and Applications Notre Dame, IN: University of Notre Dame Press. O'Donnell, Guillermo, and Philippe C. Schmitter. 1986. Transitions From Authoritarian Rule: Tentative Conclusions About Uncertain Democracies Baltimore: Johns Hopkins University Press. O'Donnell, Guillermo, Jorge Vargas Cullel, and Osvaldo Miguel Iazzetta. 2004. The Quality of Democracy: Theory and Applications Notre Dame: University of Notre Dame Press. Palmer, David Scott. 1996. Peru: Collectivel y Defending Democracy, In Tom J. Farer, ed., Beyond Sovereignty: Collectively Defending Democracy in the Americas Baltimore, MD: Johns Hopkins University Press. Panizza, Francisco. 2000. Beyond delegative democracy, Old politics and New Economics. Journal of Latin American Studies 32: 737. Parodi Solari, Jorge. 1993. Los Pobres, la Ciudad y la Poltica Lima: CEDYS. Pharr, Susan J., and Robert D. Putnam. 2000. Disaffected Democracies: What's Troubling the Trilateral Countries? Princeton: Princeton University Press. Powell, Bingham G. 2004. The Chain of Responsiveness. Journal of Democracy 15 (4): 91.

PAGE 300

290 Power, Timothy, and Mary A. Clark. 2001. Does Trust Matter? Interpersonal Trust and Democratic Values in Chile, Costa Rica, and Mexico. In Roderic A. Camp, ed., Citizen Views of Democracy in Latin America Pittsburgh, PA: University of Pittsburgh Press. Powers, Nancy R. 2001. Grassroots Expectations of Democracy and Economy: Argentina in Comparative Perspective Pittsburgh: University of Pittsburgh Press. Progama de las Naciones Unidas para el Desarrollo. 2004. La Democracia en Amrica Latina. Hacia una Democracia de Ciudadanos y Ciudadanas Buenos Aires: Alfaguara. Progama de las Naciones Unidas para el Desarrollo. 2006. La Democracia en el Per 1: El Mensaje de las Cifras Lima: Progama de las Naciones Unidas para el Desarrollo. Przeworski, Adam. 1986. Some Problems in the Study of the Transition to Democracy. In Guillermo ODonnell, Philippe C. Schmitter, and Lawrence Whitehead, eds., Transitions from Authoritarian Rule: Prospects for Democracy. Baltimore, MD: Johns Hopkins University Press. Przeworski, Adam. 1991. Democracy and the Market: Political and Economic Reforms in Eastern Europe and Latin America New York: Cambridge University Press. Przeworski, Adam. 2000. Democracy and Development: Political Institutions and Wellbeing in the World, 1950 Cambridge: Cambridge University Press. Przeworski, Adam, Michael Alvarez, Jo se Antonio Cheibub, and Fernando Limongi. 1996. What Makes Democracies Endure? Journal of Democracy 7 (1): 39. Putnam, Robert D., Robert Leonardi, and Raffaella Nanetti. 1993. Making Democracy Work: Civic Traditions in Modern Italy Princeton: Princeton University Press. Pye, Lucian W. 1965. Political Culture and Political Development. In Lucian W. Pye and Sidney Verba ed., Political Culture and Political Development. Princeton, NJ: Princeton University Press. Reisinger, William M., Arthur H. Miller, Vicky L. Hesli, and Kristen H. Maher. 1994. Political Values in Russia, Ukraine and Lithuania: Sources and Implications for Democracy. British Journal of Political Science 24: 183. Reisinger, William M., Arthur H. Miller, and Vicky L. Hesli. 1995. Political Norms in Rural Russia: Evidence from Public-Attitudes. Europe-Asia Studies 47 (6): 1025 1042. Remmer, Karen L. 1996. The Sustainability of Political Democracy: Lessons from South America. Comparative Political Studies 29 (6): 611.

PAGE 301

291 Roberts, Kenneth M. 1995. Neoliberalism and the Transformation of Populism in Latin America: The Peruvian Case. World Politics 48: 82 Roberts, Kenneth M. 1998. Deepening Democracy? The Modern Left and Social Movements in Chile and Peru . Stanford: Stanford University Press. Rose, Richard, and William Mishler. 1994. Mass Reaction to Regime Change in Eastern Europe: Polarization or Leaders and Laggards. British Journal of Political Science 24: 159. Rose, Richard, William Mishler, and Christian W. Haerpfer. 1998. Democracy and Its Alternatives: Understanding Post-Communist Societies Baltimore: Johns Hopkins University Press. Rose, Richard, Doh C. Shin, and Neil Munro. 1999. Tensions Between the Democratic Ideal and Reality: South Korea In Pippa Norris, ed., Critical Citizens: Global Support for Democratic Government New York, NY: Oxford University Press. Rueschemeyer, Dietrich. 2004. Addressing Inequality. Journal of Democracy 15 (4): 76. Rustow, Dankwart A. 1970. Transitions To Democracy: Toward A Dynamic Model. Comparative Politics 2 (3): 337. Schedler, Andreas. 1998. What Is Democratic Consolidation? Journal of Democracy 9 (2): 91. Schedler, Andreas. 2002. The Menu of Manipulation. Journal of Democracy 13 (2): 36. Schedler, Andreas. 2006. The Logic of El ectoral Authoritarianism. In Andreas Schedler, ed., Electoral Authoritarianism: The Dynamics of Unfree Competition Boulder, CO: Lynne Rienner Publishers. Schmitter, Philippe C. 2004. The Ambiguous Virtues of Accountability. Journal of Democracy 15 (4):47. Schumpeter, Joseph A. 1942. Capitalism, Socialism, and Democracy New York: Harper & brothers. Seligson, Amber L. 1999. Civic Association and Democratic Participation in Central America: A Test of the Putnam Thesis. Comparative Political Studies 32 (3):342 362. Seligson, Mitchell A. 2000. Trouble in Paradise? The Erosion of System Support in Costa Rica, 1978. Latin American Research Review 37 (1): 160.

PAGE 302

292 Seligson Mitchell A. 2001. Costa Rican Exceptionalism: Why Ticos are Different. In Roderic A. Camp, ed., Citizen Views of Democracy in Latin America Pittsburgh, PA: University of Pittsburgh Press. Seligson, Mitchell A., and John A. Booth. 1993. Political Culture and Regime Type: Evidence from Nicaragua and Costa Rica. Journal of Politics 55 (3): 777. Seligson, Mitchell A., and Julio F. Carrion. 2002. Political Support, Political Skepticism, and Political Stability in New Democracies: An Empirical Examination of Mass Support for Coups d'etat in Peru. Comparative Political Studies 35 (1): 58 82. Shin, Doh C. 1994. On The 3rd Wave Of Demo cratization: A Synthesis And Evaluation of Recent Theory And Research. World Politics 47 (1): 135. Shugart, Matthew S., and Martin P. Wattenberg. 2001. Mixed-Member Electoral Systems: The Best of Both Worlds? Oxford: Oxford University Press. Stokes, Susan C. 1995. Cultures in Conflict: Social Movements and the State in Peru Berkeley: University of California Press. Tanaka, Martin. 1999. Los Espejismos de la Democracia. El Colapso del Sistema de Partidos en el Per Lima: Instituto de Estudios Peruanos. Tarrow, Sidney. 1994. Power in Movement New York: Cambridge University Press. Thompson, Michael, Richard Ellis, and Aaron B. Wildavsky. 1990. Cultural Theory Boulder: Westview Press. Tokman, Vctor E., and Guillermo A. O'Donnell. 1998. Poverty and Inequality in Latin America: Issues and New Challenges Notre Dame: University of Notre Dame Press. Torcal, Mariano. 2001. La Desafeccin en las Nuevas Democracias del Sur de Europa y Latinoamrica. Revista Instituciones y Desarrollo 8 (9): 229. Tuesta Soldevilla, Fernando ed. 1997. Los Enigmas del Poder: Fujimori 1990 1996 Lima: Fundacin Friedrich Ebert Stiftung. Tuesta Soldevilla, Fernando ed. 1999. El Juego Poltico: Fujimori, la Oposicin y las Reglas 1. ed. Lima: Fundacin Friedrich Ebert Stiftung.. Tulchin, Joseph S., and Amelia Brown. 2002. Democratic Governance and Social Inequality Boulder: L. Rienner Publishers. Turner, Frederick C. Reassessing Political Culture. In Peter H. Smith, ed., Latin America in Comparative Perspective: New Approaches to Methods and Analysis Boulder, CO: Westview Press.

PAGE 303

293 Valenzuela, Samuel J. 1992. Democratic Consolidation in Post-Transitional Settings: Notion, Process, and Facilitating Conditions. In Scott Mainwaring, Guillermo ODonnell, and J. Samuel Valenzuela, eds., Issues in Democratic Consolidation: The New South American Democracies in Comparative Perspective Notre Dame, IN: University of Notre Dame Press. van de Walle, Nicolas. 2002. Africa's Range of Regimes. Journal of Democracy 13 (2): 66. Verba, Sidney. 1965. Conclusions: Comparative Political Culture. In Lucian W. Pye and Sidney Verba ed., Political Culture and Political Development. Princeton, NJ: Princeton University Press. Villaln, Leonardo A., and Peter VonDoepp. 2005. The Fate of Africa's Democratic Experiments: Elites and Institutions Bloomington: Indiana University Press. Weyland, Kurt. 1996. Neopopulism and Neoliberalism in Latin America: Unexpected Affinities. Studies in International Comparative Development 31 (3): 3 Weyland, Kurt. 2001. Clarifying a Contested Concept: Populism in the Study of Latin American Politics. Comparative Politics 34 (1): 1 Weyland, Kurt. 2006. The Rise and Decline of Fujimoris Neopopulist Leadership In Julio Carrin ed., The Fujimori Legacy: The Rise of Electoral Authoritarianism in Peru University Park, PA: Pennsylvania State University Press. Wiarda, Howard J. 1974. Politics and Social Change in Latin America: The Distinct Tradition Amherst: University of Massachusetts Press. Wiarda, Howard J. 1992. Politics and Social Change in Latin America: Still a Distinct Tradition? 3rd ed. Boulder: Westview Press. Whitefield, Stephen and Geoffrey Evans. 1994. The Russian Election of 1993: Public Opinion and The Transition Experience. Post-Soviet Affairs 10 (1): 38. Zakaria, Fareed. 1997. The Rise of Illiberal Democracy. Foreign Affairs 76 (6): 22. Zakaria, Fareed. 2003. The Future of Freedom: Illiberal Democracy at Home and Abroad New York: W.W. Norton.

PAGE 304

294 BIOGRAPHICAL SKETCH Jorge Aragon graduated in 1992, from the Pontificia Univesidad Catlica del Per, with a B.A. in sociology. Between 1992 and 1998, he was teaching at university and high school levels, and working as a research analyst in a public opinion and market research company in Peru. In 1998, he enrolled in the Center for Latin American Studies at the University of Florida, where he received an M.A. in Latin American Studies with a concentration in political science in 2000. He began his doctoral studies in the Department of Political Science at the University of Florida in 2000. Aragon specializes in comparative politics, political methodology, and political behavior.


Permanent Link: http://ufdc.ufl.edu/UFE0017001/00001

Material Information

Title: Popular Support for Democracy in Newly Established Democracies: The Case of Peru and the Development of an Alternative Theoretical and Methodological Approach
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0017001:00001

Permanent Link: http://ufdc.ufl.edu/UFE0017001/00001

Material Information

Title: Popular Support for Democracy in Newly Established Democracies: The Case of Peru and the Development of an Alternative Theoretical and Methodological Approach
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0017001:00001


This item has the following downloads:


Full Text












POPULAR SUPPORT FOR DEMOCRACY IN NEWLY ESTABLISHED
DEMOCRACIES: THE CASE OF PERU AND THE DEVELOPMENT OF AN
ALTERNATIVE THEORETICAL AND METHODOLOGICAL APPROACH














By

JORGE ARAGON


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA


2006





























Copyright 2006

by

Jorge Aragon
































To Sebastian, Andres and Mariana















ACKNOWLEDGMENTS

This dissertation would not have been possible without the support of the

Department of Political Science at the University of Florida for almost six years. I am

particularly grateful to my supervisory committee of Leslie Anderson, Philip Williams,

Charles Wood, Peggy Kohn, and Bryon Moraski. They have been invaluable to me in

helping me to improve this dissertation and also in helping me to be aware of the main

weaknesses of my argument.

Although Michael Martinez was not part of my supervisory committee, I want to

thank him as well, because he was always very supportive of my development as scholar.

Also, I want to thank two amazing and very generous colleagues, David Armstrong and

Lee Walker with whom I had the opportunity to discuss my data analysis during the 2006

Inter-university Consortium for Political and Social Research (ICPSR) Summer Program

in Quantitative Methods of Social Research.

My fieldwork in Lima was possible because Guillermo Loli was not only a great

friend but the most supportive one. Without his help, I would not be able to collect my

data. I want to acknowledge also the help of Beatriz Venero and Ricardo Sanchez during

my fieldwork in Peru. Working with Beatriz and Ricardo was a pleasure because of their

professionalism and understanding. Furthermore, I want also to thank Lula Garcia for

being one of the most supportive librarians that I have met.

Finally, I would like to thank three institutions: Apoyo Opinion y Mercado, the

Institute de Estudios Peruanos, and the ICPSR at the University of Michigan, for










providing me several key resources and the best environment to work in an academic

endeavor.
















TABLE OF CONTENTS



A C K N O W L E D G M E N T S ................................................................................................. iv

A B ST R A C T .................................................................................................. ix

1 IN T R O D U C T IO N .................................................................. .. ... .... ............... 1

Popular Support for Democracy in the Context of New Democracies.........................
Research Questions ................... ........... ............................... 10
M eth o d ology ...................................................... ..... ... .. ............... ...............13
Relevance of the Peruvian Case for the Study of Popular Support for Democracy
in N ew D em ocracies .............. .................. ................................................ 15
Dissertation Outline ...................... ........... ............................. 19

2 NEW DEMOCRACIES, DEMOCRATIC CONSOLIDATION, AND POPULAR
SU PPORT FOR D EM O CRA CY .......................................................... ................ 23

New Democracies: Beyond Democratic Transition ..............................................23
Main Contributions of the Literature on Democratic Consolidation................25
Main Limitations and Problems of the Literature on Democratic
Consolidation ................. .. ...... ............ ............... 38
New Democracies and Popular Support for Democracy .......................................43

3 ADDRESSING AND EXPLAINING POPULAR SUPPORT FOR
D E M O C R A C Y ........................................................................................................... 54

P political C culture A approach .................................................................. .................. 56
Critical Review of the Political Culture Approach................................ ................ 64
Political Culture and Popular Support for Democracy .........................................71

4 POPULAR SUPPORT FOR DEMOCRACY IN COMPARATIVE
P E R S P E C T IV E ............................................................. ............................................. 8 0

Established Democracies: Popular Support for Democracy and the Critical
C itizen 's T h esis ............................................................ ............... ..... ... ............ 8 1
Popular Support for Democracy in New Democracies..........................................89
P ost-com m unist C ountries.............................................. .... ......... ............... 89
Meanings of democracy among post-communist citizens ........................ 92









Measuring and explaining popular support for democracy in post-
com m unist countries...... .......... .......... ..................... 105
Latin A m erican Countries ......................................................... 128

5 THE 1992 DEMOCRATIC BREAKDOWN IN PERU AND THE ROLE OF
POPULAR SUPPORT FOR DEMOCRACY ...............................................139

Peru 1980-2001 .................................................. . ... ............ ............... 142
Explaining the 1992 Democratic Breakdown................ ..... ............................... 155
Role of Popular Support for Democracy in the 1992 Democratic Breakdown ........167

6 POPULAR SUPPORT FOR DEMOCRACY IN PERU ..................................... 174

Measuring Popular Support for Democracy in Developing Countries: The
UNDP's Report on Democracy in Latin America ........................... ................ 175
Popular Support for Democracy in Peru: Constructing Outcome Variables......... 183
Explaining Popular Support for Democracy in Peru..................... ................... 192
Impact of Socio-Economic Level on Popular Support for Democracy in Peru .......220
C o n c lu sio n s.............................................................................................................. 2 2 3

7 CONCLUSIONS ................................................................ 232

Comparative and Critical Assessment of the Literature on Popular Support for
Democracy ........ .. ................ .. .... ..... .. .. ............... 233
Popular Support for Democracy and the Prospects of Further Democratization in
N ew D em ocracies ......................................................... .. .. ....... ... .............. 237
Alternative Theoretical and Methodological Approach toward Popular Support for
Democracy in New Democracies...... .... ...... ...................... 242

APPENDIX

A THE DEFINITION AND CODING OF THE EXPLANATORY VARIABLES .... 251

Socio-D em graphic V ariables ................................ ....................... .....................251
Citizen Definitions of Democracy Variables...... .... ..................................... 251
Civic or Political Culture Variables...... .... ...... ...................... 252
Social and Economic Expectations Variables ..................................... ................ 252
Assessment of Government and Regime Performance Variables ..........................254
Political and Electoral Representation Variables .......................... ................... 255
Fujim ori's V ariables .............................. ............................................ 256

B QUALITATIVE DATA ON POPULAR SUPPORT FOR DEMOCRACY IN
P E R U ...................................................................................................... ........ .. 2 5 8

LIST O F R EFEREN CE S .. .................................................................... ................ 276

BIOGRAPH ICAL SKETCH .................. .............................................................. 294














LIST OF TABLES


Table page

1-1 Citizen perceptions of democracy in Latin America (2000-2004). ......................17

5-1 Number of victims reported to the Commission of Truth and Reconciliation....... 145

5-2 M ain problems in the country (1990-2000)...... .... ..................................... 146

5-3 Approval of Fujimori and Autogolpe (April-August 1992). ..............................149

5-4 Confidence in institutions in Peru (1989-1992) ......................... ................... 152

5-5 Approval of the president, legislature, and judiciary in Lima (1987-1992)..........152

6-1 Selected questions to assess popular support for democracy in Latin America,
United Nations Development Programme's Report on Democracy in Latin
A m e ric a .............................................................................................................. .. 1 7 7

6-2 Factorial loads for the eleven questions used to assess popular support for
democracy, United Nations Development Programme's Report on Democracy in
L a tin A m erica ......................................................................................................... 17 9

6-3 Sample and Metropolitan Lima population distributions: Gender, age, and socio-
eco n o m ic lev el ........................................................................................................ 184

6-4 Factorial loads for the eleven questions used to assess popular support for
democracy, VARIMAX Rotation (Peru, 2004) .......................... .................. 186

6-5 Reliability coefficients for the set of variables associated with each different
dimension of popular support for democracy (Peru, 2004)...............................189

6-6 Support for democracy as a form of government, ordinary least squares (OLS)
regression estim ates (Peru, 2004)...... .......... ........ ...................... 205

6-7 Support for democratic procedures, ordinary least squares (OLS) regression
estim ates (Peru, 2004) .............. .. ............ ................................................ 207

6-8 Support for democratic institutions, ordinary least squares (OLS) regression
estim ates (Peru, 2004) .............. .. ............ ................................................ 209















Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

POPULAR SUPPORT FOR DEMOCRACY IN NEWLY ESTABLISHED
DEMOCRACIES: THE CASE OF PERU AND THE DEVELOPMENT OF AN
ALTERNATIVE THEORETICAL AND METHODOLOGICAL APPROACH

By

Jorge Aragon

December 2006

Chair: Leslie Anderson
Major Department: Political Science

My study concerns popular support for democracy, particularly in newer and less

consolidated democracies. Political scientists who deal with contemporary processes of

democratization and the prospects for democratic consolidation have devoted much

attention to the role of political elites, and to the merits and liabilities of different

institutional designs. They have devoted less attention to the role played by ordinary

citizens. Without neglecting the importance of elites and institutions in the creation and

maintenance of democracy, I argue that overlooking the role of ordinary citizens is a

serious limitation when we try to understand the advances and setbacks common to

newly established democratic regimes. As numerous historical and contemporary cases

show, much of what elites attempt to do is conditioned by their judgments of what

ordinary people think and how they behave.

My focus on the political development of these newly and less consolidated

democracies is organized in terms of two fundamental questions. First, what are the









political consequences of a prevalent and persistent lack of support for democracy among

the majority of a country's population? Second, what is the explanation for the current

degree of support for democracy that we witness in countries that have only recently

become democratic?

Unfortunately, much of the recent literature on popular support for democracy in

newly established democratic regimes is of limited utility when it comes to providing

answers to the two main question posed here. In most cases, studies of popular support

for democracy adopt theoretical approaches and methodological strategies that are ill-

equipped to grasp the full complexity of the phenomenon. In the face of these

deficiencies, I developed a new theory and method to account for the multi-dimensional

nature of the relationships at hand, and the multiple factors that influence the degree to

which ordinary citizens support democracy.














CHAPTER 1
INTRODUCTION

Popular Support for Democracy in the Context of New Democracies

This dissertation addresses the topic of popular or mass support for democracy in

the context of new democracies, and, more specifically, the case of the most fragile or

less consolidated ones. In doing that, I have tried to provide answers to the two following

questions: (1) what are the main the consequences of a lack of a considerable and stable

level of popular support for democracy in new democratic regimes and (2) how to

address and explain current levels of popular support for democracy in new democratic

regimes.

Regarding the first question, I contend here that, in the context of several new

democracies, the lack of advances in democratization, and even the cases of democratic

setbacks, are highly related to a lack of popular support for these new democratic

regimes; in particular, a lack of popular support for democratic procedures and

institutions. As Hagopian (2006:320) points out for the case of Latin America, unlike the

1960s and 1970s, when democracies were weakened by the clash of ordinary people

demanding the rights that elites were reluctant to grant-to become citizens-today the

crisis of democracy is associated with the widespread and deep disillusion among

citizens. Accordingly, citizen dissatisfaction has led to low levels of public trust in









political institutions and political parties, and to popular support for strong-arm, anti-

establishment, or neopopulist1 leaders with very ambiguous attitudes toward democracy.

I am aware that not everyone will share the conviction that popular support for

democracy matters to a great extent vis-a-vis the current and future situation of new

democracies. However, I think it is critical to recognize that much of what elites attempt

to do in these new democratic regimes is heavily conditioned by their judgments of how

ordinary people will behave (Rose, Mishler, and Haerpfer 1998; Bermeo 2003). By the

same token, I am aware that someone can argue that new democracies are not the only

ones in trouble regarding current levels of popular support (Pharr and Putnam 2002).

However, I think that it is also important to acknowledge that the level of citizen support

for democracy is far lower in new democracies that in the established ones, and,

moreover, that the dissatisfaction and skepticism about government and authorities

within established democracies has not produced yet the same degree of regime

instability.

Regarding the consequences of lack of popular support for democracy, I contend

here that answering why support for democracy is declining in several democracies is not

an easy task. Nevertheless, as it has been noted for several Latin American countries, the

performance of democratic governments has been very poor where new democratic

regimes have been less stable. Specifically, when new democratic regimes have failed to

promote growth, reduce poverty, address the causes of social and economic inequality,

provide security, and establish the rule of law; or, in other words, when they have failed




1 For a definition of neopopulism, particularly in the context of Latin American countries see Weyland
(1996, 2001).









to address citizen needs, popular satisfaction with democracy has declined (Hagopian and

Mainwaring 2006).

Therefore, in trying to understand what can make new democratic regimes more

democratic, it is critical to address citizen expectations of government, perceptions of the

policy course that governments set, and evaluations of the success of those policies

(Hagopian 2006). In that regard, one of the main contributions of this dissertation is that

it shows how the combination of structural variables such as poverty and inequality, and

the poor performance of both democratic governments and regimes in addressing citizen

needs and expectations has indeed a corrosive effect on popular support for democracy,

and, as a consequence of this, on the stability and quality of the new democratic regimes.

The recognition of both that popular support for democracy matters and that

democratic performance is a key determinant of it is extremely relevant to the study of

new democracies. Accordingly, what allows a democracy to emerge and survive does not

guarantee that democracy will be effective or immune from anti-system challenges and

citizen dissatisfaction. In other words, whereas elite attitudes toward democracy and a

favorable international political environment have been key factors in understanding the

maintenance of competitive elected regimes in Latin America, the consolidation of new

democratic regimes, or, simply, further democratization (i.e., the democratization of

semi-democratic regimes) is better explained by the interplay of structural factors, regime

performance, and popular political attitudes (Mainwaring and Perez-Lifian 2006)

In addressing the topic of popular support for democracy in new democracies, I

have confronted several and different challenges. First, despite the common use of the

term popular or mass support for democracy, it is only rarely acknowledged that this









particular political attitude involves different components or dimensions.2 In that regard,

it is critical to address the extent to which citizens living in new and fragile democracies

lack support for democracy as a form of political regime based on public contestation and

the right to participate in elections and office (Dahl 1971) or, on the contrary whether the

lack of support is with respect to current democratic regimes and democratic authorities.

As it will be addressed later, our perceptions and conclusions regarding the future of new

and fragile democracies will change considerably depending on the dimensions of

popular support for democracy we consider.

Second, despite the growth of an important literature about the recent processes of

democratization outside the realm of developed countries (Bunce 2000), very few works

have specifically addressed the topic of popular support for democracy, and often, the

topic has been frequently addressed in a tangential way.3 Most of the time, popular

support for democracy has just been considered as one among several factors that can

play a role in the future prospects of these new democracies, and very little has been said

about how this political phenomenon is interrelated to these other critical factors.4

Therefore, in spite of the widespread recognition of the theoretical importance of

popular support for democracy for the maintenance and consolidation of new

democracies, relatively few studies have investigated, in a systematic and comprehensive

way, its main determinants. As a consequence of this, we lack a comprehensive view

regarding the challenges that most of these new democratic regimes are confronting to


2 Norris (1999a) and Dalton (1999) provide a theoretical and empirical treatment of the main dimensions of
political support in a democratic setting.

3 Rose, Mishler, and Haerpfer (1998) is one of the few cases where, in dealing with new democracies,
explaining support for democracy and for its alternatives was its main goal.

4 See for example Diamond, Linz, and Lipset (1989) and Diamond (1999).









achieve higher levels of democratization. Part of the explanation for the lack of a

systematic and comprehensive approach toward popular support for democracy in the

context of new democracies is related to the difficulties of studying political support.

According to Kornberg (1990), investigators dealing with political support in democratic

societies have found it extremely difficult to unravel and separately measure the different

types or components of political support (e.g., diffuse and specific support).5 Also, and

equally important, investigators have generally been unable to provide satisfactory

answers to "so what" questions linking differences in support levels to variations in

people's political behavior. Therefore, in studying popular support for democracy in new

democracies, scholars have faced not only difficulties in dealing empirically with this

type of political support, but also difficulties in connecting individual's attitudinal

orientations with individual's political behavioral orientations.6 There is no doubt such

difficulties may explain why scholars dealing with new democracies have been reluctant

to address the topic of popular support for democracy in more specific and systematic

terms.

Third, there have been also important limitations in the theoretical and

methodological ways in which popular support for democracy in new democracies has

been approached. Regarding the theoretical limitations, a large majority of scholars who


5 According to Easton (1965), the distinguishing feature of specific support is that it is based on a kind of
rational calculus, i.e., support is giving or withheld depending upon whether a political object generates
affects that are considered beneficial or detrimental to one's self or group. On the contrary, the essential
feature of diffuse support is its affective quality, i.e., support is granted or withheld on the basis of simple
but powerful feelings of like or dislike.
6 It is true that there are several "so what" questions that we are far from having satisfactory answers for,
e.g., to what extent is popular support for democracy connected with electoral behavior, and to
conventional and unconventional forms of political participation; or, to what extent is popular support for
democracy shaping citizen willingness to comply with the authoritative government's decisions and
policies. However, it is also true that the erosion of popular support for democracy appears frequently as
one of the most powerful causes of democratic reversals (see Chapter 5).









have dealt directly or indirectly with this topic seem to ignore that individual's political

orientations are strongly influenced by (a) deep-seated political values and (b) specific

assessments about the ability of the regime and its authorities to perform its ascribed

functions Unfortunately, most of the time, the level of popular support for democracy

among citizens from new democracies has been addressed as a function of the level of

citizens' democratic values; more specifically, regarding the existence or not of a civic

culture (Almond and Verba 1963).7

As a reaction to all these limitation, and trying to fill a critical void in the literature

on popular support for democracy, this dissertation will test the extent to which non-

political values variables-such as individual's social and economic expectations, and

individual's assessment of democratic performance-play a critical role in shaping it. I

will try to demonstrate that neglecting the role of individual's expectations as well as the

role of individual's assessments regarding the performance of the current democratic

regime produces partial and misleading interpretations, not only about the level of

popular support for democracy and current democratic regimes, but also about the

prospects of new democracies.

Consequently, one of the main contributions of this dissertation is to make clear

that accounting for the main determinants of popular support for democracy requires that

analysts consider a complex and diverse set of explanatory variables, and to pay attention

to what some of these main determinants have in common.8 Accordingly, the empirical

analysis conducted in this dissertation makes clear that some structural factors such as


7 See Chapter 4, and the section dealing with the study of popular support for democracy within Latin
American countries, in particular.

8 For an example of how this theoretical approach is used for studying support for democracy in established
democracies, see Dalton 2.l'" 4).









class or socio-economic level exert an important influence on popular support for

democracy. In this regard, I have found that the most disadvantaged population in

Metropolitan Lima is, at the same time, the most dissatisfied with the performance of

their current democratic government and democratic regime, and the one that presents the

lowest levels of support for democracy as political regime, democratic procedures, and

democratic institutions. Therefore, I believe that this dissertation reveals the urgency of

overcoming the tension between individual-agency and structural explanations that have

characterized the study of the recent process of democratization.9

By the same token, it is worth noting here also that the existing literature on

popular support for democracy among new democracies has focused mostly on between

differences (i.e., differences among countries considered as a whole) and much less on

ii i/hin differences (i.e., differences within one specific country). Accordingly, most

scholars have been much more concerned with how to explain the differences in popular

support for democracy between established democracies and new democracies, or

between new democracies, rather than addressing the critical differences in terms of

popular support for democracy that can be found within specific new democracies. This

explains why the effect of class or different socio-economic levels on popular support for

democracy within new democracies has been largely neglected. This is a critical

limitation with respect to how political scientists have been approaching the study of

popular support for democracy among new democracies. 10



9 See Landman (2000) for how this has been the case for the literature dealing with democratic transitions
in the late twentieth century.

10 This issue needs to be address with a greater sense of urgency because there is some evidence showing
that within di',...... a are, very frequently, as important as between hdi. "...... (Moreno 2001, Arag6n
2001).









With respect to the most common methodological shortcomings regarding the

study of popular support for democracy within new democracies, I will show here that it

is absolutely critical to review and improve the way that most scholars have

operationalized and measured this complex phenomenon. Perhaps too frequently, it is

assumed that one or two survey questions about the preferred political regime (i.e.,

democratic or non-democratic) and the individual level of satisfaction regarding the way

in which democracy is working in the country are sufficient to provide reliable and valid

measures of popular support for democracy. This unsatisfactory methodological strategy

can be found even in some of the most influential comparative works about

democratization in the late twentieth century.11

As a reaction to this situation, this dissertation will present an innovative way to

deal empirically with the topic of popular support for democracy. For several reasons, I

will contend that this new methodological strategy constitutes a significant improvement

for the study of popular support for democracy in new democracies. In that regard, I have

been able to identify and empirically measure three different components or dimensions

of popular support for democracy (i.e., support for democracy as form of political regime,

support for democratic procedures,12 and support for democratic institutions).

Additionally, and departing from previous scholarly works, I have identified a set of

variables, and construct indexes, that seem reliable and valid tools to account for the main

determinants of popular support for democracy.



1 See for example, Linz and Stepan (1996) and Diamond (1999).
12 See O'Donnell (1994) for the relationship between the lack of institutionalization of democratic
procedures and the development of a (, ./,.i i,, ,.a version of democracy. In that regard, O'Donnell makes
absolutely clear why (,4. ki..,,h 1C.' democracies in Latin America are not good news for the development and
institutionalization of democratic regimes in this region.









Fourth, regarding the main challenges for the study of popular support for

democracy, it was clear to me that the most important theoretical and methodological

advancements regarding the study of popular support from democracy come from

scholars who are dealing with this topic from the perspective of established democracies.

This poses some problems because the phenomenon of popular support for democracy

between established and new democracies differs in important ways. Established and

several new democracies are alike in the sense that the consent of the people, not

coercion, explain its persistence. However, it is only within established democracies that

citizens clearly separate support for their political authorities from support for their

political regimes and political communities. In a typical established democracy, we

frequently find a high level of support for the political community, a low level of support

for political authorities, and a level of support for the political regime-including the

main democratic norms and procedures-which fall between these two extremes. This is

not the necessarily the case among several new democracies. In fact, in these cases, low

popular political support is not constrained to political authorities, but may also include

democratic rules and procedures, and, in some cases, the current democratic regime itself.

In this sense, the main forces shaping popular support for democracy are entirely

different between established and new democracies.

Finally, it is my main expectation that a better understanding of the phenomenon of

popular support for democracy in new democracies will contribute to improve our

understanding regarding some key issues related to these recent processes of

democratization. In particular, why is it difficult to develop popular support for current

democratic regimes and key democratic rules and procedures among several new









democracies; and, why and how ordinary citizens' 13 ideas, beliefs, and expectations

matter for the prospects of a process of further democratization. The need for addressing

this connection between popular support for democracy-on the hand-and the

possibility of democratic setbacks, stagnations or advancements-on the other hand-led

me to carefully evaluate the existing literature, identify a relevant empirical case, and

elaborate an adequate research design. On the latter point, and after identifying important

limitations within the literature dealing with popular support for democracy, I also had to

develop particular and innovative theoretical and methodological approaches to address

this political phenomenon. Accordingly, this dissertation is not only an argument in favor

of the relevance of popular support for democracy as one of the most important factors

shaping the political development of new democratic regimes, but it also makes an

important contribution to the existing literature on this topic, in that it seeks to improve

upon the theoretical and methodological approaches used to address popular support for

democracy, particularly in the context of new and fragile democracies.

Research Questions

In addressing the topic of popular support for democracy, this dissertation attempts

to make a contribution to the discussion on the survival and quality of new democratic

regimes. Regarding these recent process of democratization, some of the most important

issues are framed by the following questions: why is that so few of them are on route to

becoming well-functioning and stable democracies? Or, why is that so few of them are


13 The term ordinary citizens comes mainly from Bermeo (2003), Linz and Stepan (1996) and Tarrow
(1994). As Bermeo (2003:3) contends, the term ordinary underscores the fact that we are considering
people or citizens that do not have extraordinary powers vis-a-vis the state in which they live. Accordingly,
they are not politicians or military officers, but people or citizens that spend most of their lives in personal
endeavors-earning money, supporting families, and pursuing whatever leisure activities their social status
allows. Also, they are the people who compose the vast majority of the citizenry in virtually every country
in the world.









characterized by still having a positive dynamic in terms of further democratization?

Alternatively, why is that an overwhelming majority of these new democratic regimes do

not seem to be deepening or advancing whatever democratic progress they have made as

part of their democratic transitions? (Carothers 2002)

In summarizing this situation, O'Donnell (2004b, 2004c) contends that the problem

with new democracies is that a majority of them are characterized as being low quality

democracies, not only in terms of governance (i.e., the capacity to formulate and

implement policies in an effective manner), but also in terms of providing the necessary

conditions to make these new democratic governments more representative and

accountable, particularly regarding the needs and desires of the most disadvantaged

populations. I believe that this deficit in the quality of democracy is one of the main

reasons why, as is often the case, there is a lack of stable and strong popular support for

the new democratic regimes.

This deficit of popular democratic legitimacy is not necessarily related to

democracy as an ideal form of political regime, but instead related to current democratic

regimes, democratic procedures, democratic institutions, and democratic authorities.

Nevertheless, it is not the case that a low level of popular support for democracy will

always lead to an undisputed democratic breakdown (e.g., Peru in 1992). More

frequently, this lack of popular support for some democratic procedures and institutions

will be a critical component for the development of hybrid regimes (e.g., current

Venezuela14 and Russia15). Accordingly, it is evident to me that democratic breakdowns

or the development of stable hybrid regimes (Hartlyn 2002) are among the reasons why

14 See McCoy and Myers 2" 1114).

15 See Fish (2005) and McFaul, Petrov and Ryabov (2' 114).









several new democracies are experiencing significant difficulties in terms of making

sustained advances with regard to further democratic development.

Therefore, this dissertation seeks to address two main questions. First, what are the

main determinants of popular support for democracy in new democracies? Second, in the

context of new democracies, what does the topic of popular support for democracy reveal

about the nature of these political regimes and their challenges for a further process of

democratization? Regarding the first question, I will consider the impact of (a)

individual's understandings of democracy, (b) individual's social and economic

expectations toward their own society, (c) individual's assessments of regime and

government's performances, (d) the level of individual's civic culture orientations, and

(e) some of the main individual's socio-demographic factors-including socio-economic

level-on popular support for democracy. These findings should provide me with the

opportunity to address a related question: to what extent does poverty and inequality-two

of the most frequent characteristics found within new democracies in developing

countries-exert a powerful influence on popular support for democracy either in a direct

or indirect way? On the latter point, for example, I was expecting that living in poverty

could be strongly related to some of the most important determinants of popular support

for democracy such as particular ways of defining democracy, clear preferences vis-a-vis

how society should be organized and the role of government in regulating the economy,

and/or very poor evaluations of democratic regime and government's performance.

With respect to the second question, and noting the connection between low quality

democracies, lack of popular support for democracy, and the development of

authoritarian and stable hybrid regimes, I contend that having a better understanding of









the determinants of popular support for democracy should provide important insights

regarding the main sources of citizen dissatisfaction and frustration with their current

democratic experience and current democratic regimes. Such insights may reveal

valuable information that may prove helpful for achieving finally gradual improvement

of the democratic quality of these new democratic regimes.

Methodology

The theoretical part of this work, as well as the methodological discussion of how

popular support for democracy has been addressed in the last years-particularly in the

context of new democracies-have been conducted in the tradition of comparative

studies. Looking for similarities and differences regarding what other scholars have found

on this topic for different countries and regions provide, I tried to acquire a better

understanding of this phenomenon. Additionally, I have been also able to develop an

argument on how addressing the topic of popular support for democracy contributes, in a

fundamental way, to an improvement of our understanding of the particular nature of

these new democratic regimes. Unfortunately, this time, I was limited to work only with

one empirical case. Accordingly, the empirical component of this dissertation is based on

a public opinion survey that I designed and conducted in Peru during the second half of

2004. In that regard, I was able to design a tailored questionnaire that specifically

addressed the topic of popular support for democracy. I was assisted by twenty

experienced interviewers, and I was able to collect 420 face-to-face interviews with

Peruvians older that 18 years old, who resided in the country's capital, Lima.

The quantitative orientation of this dissertation has, however, two main distinctive

features. First, it was heavily influenced by my qualitative research on this topic.

Between 1997 and 2004, I carried out several focus groups with Peruvian citizens from









different socio-economic levels. My analysis of these sessions revealed to me that there

are critical differences in the ideological and attitudinal orientations of Peruvians of

different socio-economic levels toward politics and democracy. With specific reference

to citizens from the lower socio-economic levels, I learned that more than an

authoritarian orientation they hold either a profound indifference toward what can happen

to the current democratic regime or a very critical view regarding how the current

democratic regime is unable to deliver what a democratic regime is supposed to provide.

All these insights were very useful for the development of my questionnaire about

popular support for democracy.

Second, I conducted an extensive review of some of the most important public

opinion research projects that have been carried out in the last few years in Peru and other

Latin American countries.16 These projects constituted an invaluable source of

information and expertise regarding the use of public opinion surveys to generate valid

and reliable measures on popular support for democracy in new democracies.

For my analysis of these data, I used factor analyses to identify three different and

critical dimensions of popular support for democracy and to operationalize each one of

theses dependent or outcome variable. For the analysis of the main determinants of

popular support for democracy, I used ordinary least squares (OLS) regression models.

After analyzing different sets of explanatory variables separately, e.g., socio-

demographic factors, meanings and expectations toward democracy, economic and social

expectations, civic culture variables, and assessments of democratic government and




16 Latinobar6metro, Latin American Public Opinion Project (LAPOP), a public opinion project founded by
the Hewlett Foundation, and Apoyo Opini6n y Mercado's monthly public opinion report.









regime's performance. I developed some general models to account for the main

determinants of popular support for democracy.

Relevance of the Peruvian Case for the Study of Popular Support for Democracy in
New Democracies

In this section I address the extent to which Peru can be considered a representative

case in terms of popular support for democracy in new democracies. In that regard, and

without neglecting the existence of some exceptional features in its recent political

experience,17 1 will contend that Peru is a very representative case of one particular kind

of new democracies in Latin America, as well as in other regions of the world.

Accordingly, some scholars (Hagopian and Mainwaring 2006, Mainwaring and

Perez-Lifian 2006) have suggested that current new democracies in Latin America-or,

post-third-wave democratic regimes-can be classified between countries where their

democratic regimes have become more solid (e.g., Uruguay and Chile) and countries

where their democratic or semi-democratic regimes have avoided-almost all the time-

democratic regression, but where there is also clear that there has been a clear stagnation

in their processes of democratization, particularly since the beginning of the 1990s (e.g.,

Bolivia, Ecuador, Colombia, Guatemala, and Venezuela). 18 1 do not have any doubts that

Peru clearly belongs to this latter group, and that its political and democratic experience it

is not very different from what is happening in other cases of democratic stagnation.

One way to support my last contention is to look at some of the most standard

public opinion survey questions that are used to gauge the citizen perception of

17 See Chapter 5 for a description of how Peru during the late 1980s and early 1990s confronted not only a
severe economic crisis, but also an internal war conflict that definitively. This combination, definitively,
increased to a large extent the level of desperation and the dissatisfaction with the democratic authorities
among Peruvians.

18 Similar classifications can be found also within scholars dealing with new democratic regimes in sub-
Saharan Africa (Villal6n and VonDoepp 2005, Hyden 2006).









democracy. In that direction, Hagopian (2006:333-334) has provided a summary of these

indicators using findings from the Latinobar6metro between 2000 and 2004 for 17 Latin

American countries (Table 1-1). This summary includes: levels of support for democracy

(2000-2004 average); numbers responding that democracy is the best governmental

system (2000-2004 average), levels of satisfaction with democracy (2000-2004), levels

of trust in government (2003), and numbers responding that voting can make a difference

(2004).

As Table 1-1 clearly shows, countries such as Uruguay, Costa Rica, Chile and

Argentina are systematically at the top part of the table. On the contrary, Peru is almost

always below the regional average, together with El Salvador, Colombia, Bolivia,

Ecuador, Paraguay and Guatemala. As a consequence of this, it is not a coincidence how

frequently non-constitutional depositions of democratically elected presidents, 19 strong

electoral support for anti-party or anti-systems leaders,20 and even democratic

regressions21 can be found within this latter group of Latin American countries.

By the same token, if we consider that there is some evidence that in the case of

Latin America there is a relationship between the level of support for democracy and

satisfaction, and the quality of democratic governance, particularly in terms of

effectiveness, transparency, and fairness (Hagopian 2006); it is possible to contend that

addressing the topic of popular support for democracy considering performance variables




19 Jamil Mahuad in Ecuador in 2000, Gonzalo Sanchez de Lozada in Bolivia in 2003, and Lucio Gutidrrez
in Ecuador in 2005.
20 Hugo Chavez in Venezuela, Lucio Gutidrrez in Ecuador, and Alberto Fujimori and Ollanta Humala in
Peru.
21 Peru in 1992 and Venezuela since 1998.










can contribute to a large extent to improve our understanding of these cases of democratic

stagnation within new democracies.

Table 1-1. Citizen perceptions of democracy in Latin America (2000-2004)
Country Support for Satisfaction Importance Trust in Democracy
democracy with of voting government is the best
democracy system
Uruguay 79 53 71 19 84
Costa Rica 75 56 51 23 80
Argentina 65 28 65 45 80
Chile 52 32 53 46 71
Panama 53 34 72 20 71
Venezuela 67 43 68 28 74
Honduras 56 41 49 25 72
Brazil 36 23 59 42 69
Nicaragua 52 30 50 12 66
Mexico 52 23 52 24 75
El Salvador 45 31 61 21 61
Colombia 43 20 57 33 64
Peru 56 15 54 14 59
Bolivia 53 21 37 14 62
Ecuador 47 19 45 10 55
Paraguay 41 10 50 16 55
Guatemala 38 26 54 7 46

Latin America 54 30 60 24 68
a Percentage agreeing with statement, "Democracy is preferable to any other kind of
government," Latinobar6metro (2000-2004 average).
b Percentage responding they are "very satisfied" and "fairly satisfied" with the functioning of
democracy, Latinobarometro (2000-2004 average).
c Percentage responding "How one votes can make things different in future," (Latinobar6metro
2004).
d Percentage expressing "a lot of confidence" and "some confidence" in government,
Latinobar6metro (2003).
e Percentage agreeing with the statement, "Democracy may have some problems, but it is the
best system of government, Latinobar6metro (2003-2004 average).
Source: Hagopian (2006), pp. 334.

Specifically about the study of popular support for democracy, addressing the case

of Peru provides a new opportunity to advance our understanding of some common

patterns that have been found in other regions of the world: the existence of a widespread

but shallow or superficial popular support for democracy. As it will be evident, Peru









shares with several new democracies in sub-Saharan Africa the coexistence of a

widespread popular support for the idea of democracy or the "rule for the people;" with

some pockets of authoritarianism, and very ambiguous popular orientations toward

democratic authorities, institutions, and procedures (Bratton, Mattes and Gyimah-Boadi

2004:345).

Finally, I want to mention something about to what extent the findings that I will be

reporting here can be generalized to the entire country. My public opinion survey has not

included Peruvian population outside Metropolitan Lima. Therefore, in a country with a

long tradition of citizen resentment of Lima-dominated political institutions and deep

desire for greater decentralization and regional autonomy, the fact that my survey does

not get the urban-rural divide is an important limitation for the generalization of these

findings.

However, the fact that Metropolitan concentrates almost a third of the Peruvian

population, and around 40% of the total urban population makes these findings highly

significant regarding what is happening in Peru in terms of popular support for

democracy. Even more, I would expect that research on popular support for democracy in

coastal cities in Peru (e.g., Piura, Chiclayo, Trujillo, Chimbote, Ica, and Tacna) will

likely produce similar results. Finally, I think it is possible to contend that some of the

trends found within the Metropolitan Lima population are even stronger within the rural

and non-coastal urban Peruvian population. In that regard, a national public opinion

survey conducted in Peru in 2005 reveals systematically higher levels of political and

democratic dissatisfaction and alienation within the rural and non-coastal urban

population.









Dissertation Outline

This dissertation is divided into two parts. Part I comprises Chapter 2 through

Chapter 4; here, the relevant theories and research literature related to the topic of

popular support for democracy-particularly in the context of recent processes of

democratization-are presented and discusses. Part II comprises two chapters (Chapter 5

and 6) and focuses on a specific case, that of contemporary Peru, which provides the

empirical evidence for this work. Additional details on each chapter are given below.

Chapter 2 examines the connection between popular support for democracy and the

consolidation of new democracies. I contend that the prospects of further and gradual

democratization depend to a large extent on the level and resilience of popular support for

democracy. It is not my intention here to claim that there is an automatic and mechanic

correspondence between the level of popular support for democracy and the level of

democratization; or, to claim that popular support for democracy is the most powerful

factor shaping the political future of new democracies. On the contrary, as Rose, Mishler,

and Haerpfer (1998) have suggested for the case of post-communist countries, I believe

that the lack or presence of a considerable level of popular support for democracy will

exert a powerful influence on political elite decisions. In that regard, and in the context of

weakly institutionalized democratic regimes, it is not going to be particularly difficult to

break democratic norms and procedures when ordinary citizens are either profoundly

dissatisfied with the performances of the current democratic regime or profoundly

indifferent regarding the possibility of a regime change.

Chapter 3 reviews the main theoretical approaches that can be used to address the

phenomenon of popular support for democracy in new democracies. I contend that the

typical political culture approach (i.e., the one that is mainly interested in political









socialization and the acquisition of deep-seated political values) is insufficient to account

for this complex political phenomenon. Some non-typical political values, such as

specific preferences regarding their own society, play a critical role in shaping popular

political support; but so do evaluations regarding the different performances of both

democratic governments and democratic regimes, and some structural factors such as

class or socio-economic level.

Chapter 4 presents a comparative and critical assessment of the literature on

popular support for democracy in established democracies, post-communist countries and

Latin American countries. The main goal was to acquire more insights on cases with

which I am less familiar, that is, non-Latin American countries. At present, the highest

level of theoretical and methodological advancement regarding the study of popular

support arises from critical evaluation of cases of established democracies in advanced

industrial countries. However, it is also the case that research on popular support for

democracy within post-communist countries is more abundant, and, overall, better

grounded in theoretical and methodological terms than comparable research on Latin

American countries.

Chapter 5 deals with the Peruvian political experience over the last few decades

and in particular with the 1992 democratic breakdown and the role that popular support

for democracy played with respect to this outcome. After presenting a very brief

description of the main political events in Peru between 1980 and 2001, I address and

discuss some of the main explanations of this democratic breakdown. I will contend that

despite the relevance of some institutional explanations, there is no doubt that a severe

decline among Peruvian citizen support for some of the main democratic institutions and









procedures played a critical role in the rise of an authoritarian regime between 1992 and

2000.22

Chapter 6 presents and analyzes the data that were collected in Peru in November

2004. These data are representative of the Peruvian population that is older than 18 years,

living in the capital of the country. In the first part, the aim is to operationalize and

measure the concept of support for democracy. Several questions were analyzed using

factor analyses to test the multidimensional nature of popular support for democracy. In

the second part, different OLS regression models were developed to account for the each

one of the dimensions of support for democracy previously identified. Overall, I am

expecting that this work can be used in the development of more accurate methodology to

measure and assess popular support for democracy among new democracies.23

Finally, Chapter 7 summarizes the main findings and conclusions of this work. It is

clear for me that one of these main conclusions is related to the fact that despite the

relevance of popular support for democracy for the future of new democracies, it is the

case that existing research literature on this topic, particularly with respect to new

democracies, has several theoretical and methodological shortcomings. However, as this

dissertation will show, some of these theoretical and methodological challenges and

shortcomings can be resolved in satisfactory ways. My contention is that improved

research on popular support for democracy will benefit, to a considerable extent, our

understanding of contemporary cases of democratization. Regarding the maintenance of



22 Carri6n (2006) provide a very recent and comprehensive work about the nature and legacy of Fujimori's
regimes. The Fujimori's regime is classified as one of the cases of a new and contemporary breed of
authoritarian regimes (i.e., electoral authoritarianism).
23 The UNDP's report of democracy in Latin America (2i "14) constitutes the best methodology that I have
found on how to approach and measure popular support for democracy in the context of new democracies.









these new democratic experiences and the possibility of a gradual process of further

democratization, in particular, the research about popular support for democracy is

providing several critical insights. Overall, it is evident that it may be very difficult to

change the current lack of popular support for key democratic rules and procedures

among several new democracies, if these citizens do not experiment and perceive a

minimal improvement in the quality of their democratic regimes. However, the good

news is that, in the context of several new democracies, there is evidence that an

improvement in the quality of political representation, political accountability, and the

rule of law can be as effective as an improvement of people's socio-economic conditions.














CHAPTER 2
NEW DEMOCRACIES, DEMOCRATIC CONSOLIDATION, AND POPULAR
SUPPORT FOR DEMOCRACY

This chapter addresses the experience of third-wave democracies once they have

completed a transition to democracy. The first section of this chapter focuses on the

literature most relevant to recent democratization processes, and democratic

consolidation.1 The second section addresses the issue of popular support for democracy

in these new democracies. In that regard, I contend that the literature on democratic

consolidation is correct in its affirmation that the consolidation or institutionalization of

these new democratic regimes it impossible without an important attitudinal and

behavioral change-both at the elite and ordinary citizens level-regarding democracy,

current democratic regime, and main democratic procedures, norms, and institutions.

However, I also discuss why it is problematic to assume that these pro-democratic

attitudes and behaviors are the final outcome of a democratic consolidation process rather

than addressing the main challenges for its gradual development. Finally, I argue for the

need of a more systematic and comprehensive study of popular support for democracy

within new democracies.

New Democracies: Beyond Democratic Transition

Beginning in the mid-1970s in Southern Europe (Greece, Portugal, and Spain), and

then spreading to Latin America, East Asia and Africa, a large number of formerly

authoritarian regimes began to shift to democracy. Also, during the period 1989-1991,


1 It is worth noting here that some scholars are particular skeptical about the possibility of a democratic
consolidation process in several new democracies.









with the collapse of the Soviet Union and other communist regimes in Eastern Europe, a

large number of formerly communist countries began transitions to democracy as well.

This global expansion of democracy-known as the third-wave of democratization

(Huntington 1991)-made a strong impact on the study of democracy and

democratization, particularly in comparative perspective. Accordingly, several scholars

have been dealing with two main topics: (1) to explain why, when, and where these

processes of democratic transition happened2 and (2) to assess the future prospects of

these new democracies, particularly in the context of developing countries and post-

communist countries (Landman 2000). More specifically, within this second main issue,

several scholars have been interested in addressing the prospects and challenges

regarding the consolidation of these new democracies. Accordingly, this group of

scholars has developed what is known as the literature on democratic consolidation.

This section deals with this literature on democratic consolidation,3 assessing its

main contributions and problems regarding the understanding of the democratization

processes of the new members of the democratic world (Bunce 2000). The literature on

democratic consolidation shows an interesting mix of critical contributions and important

limitations regarding the study of new democracies. Regarding its main limitations, it is

worth noting that the study of democratic consolidation processes is particularly


2 This is what constitutes the literature on democratic transitions. As Landman (2000) points out, this
literature has either focused on the role of elites and the nature of pacts that are formed between them, or
has focused on the role of members of civil society and the ways they struggle for democracy. See for
example, O'Donnell and Schmitter (1986), Levine (1988), Diamond, Linz, and Lipset (1989), Di Palma
(1990), Karl (1990), Huntington (1991), Przeworski (1991), Munck (1994), and Linz and Stepan (1996).

31I am using here a broad definition of the "literature on democratic consolidation" because I am
considering both scholars that have directly addressed the issue of democratic consolidation of new
democracies-e.g., O'Donnell (1992), Valenzuela (1992), Mainwaring (1992), Linz and Stepan (1996),
O'Donnell (1996), Schedler (1998), Diamond (1999), Encarnaci6n (2000)-and scholars that have been
particularly skeptical about its current prospects of consolidation-e.g., Carothers (2002), Schedler (1998),
and O'Donnell (1996).









challenging not only because they are very complex, but also because we do not have yet

clear empirical thresholds regarding the different stages of a process of democratic

consolidation.

Main Contributions of the Literature on Democratic Consolidation

It is clear that the literature on democratic consolidation has been particularly

important in establishing a set of relevant questions about new democracies that go

beyond its mere processes of transition. Specifically, this literature has established that

the consolidation of democracy is an analytical and empirical differentiated aspect of any

process of democratization. Accordingly, the literature on democratic consolidation has

been concerned about the main challenges regarding the development of democratic

regimes once new democratic governments have been established. Furthermore, several

scholars have proposed the idea that a process of democratization implies two transitions

(O'Donnell 1992, Valenzuela 1992, and Mainwaring 1992, Diamond 1999). The first

transition is from the previous authoritarian regime to the installation of a democratic

government. The second transition is from this democratic government to the

consolidation of democracy or, to the specific development of a stable, functioning, and

institutionalized democratic regime.

The distinction between a democratic government and a democratic regime is

critical. As Mainwaring (1992:296) points out, there is huge difference between the

installation of a democratic government, and the development of a democratic regime.

The former basically denotes a government elected by the people. The later, however, is a

broader concept that refers to the rules that control the interaction of the major actors in

the system. Therefore, regime involves a process of institutionalization, i.e., the idea that

such rules are widely understood and accepted, and that political actors pattern their









behaviors accordingly. Any successful democratic transition from an authoritarian rule

will consequently necessitate the installation of a democratic government. However, the

installation of a democratic government does not guarantee the development of a

democratic regime.

Focusing on the second transition, the literature on democratic consolidation has

sought to address why, in the context of many new democracies, the consolidation of

democracy or the development of an institutionalized democratic regime has been

particularly difficult. In other words, why do these new democracies often become

diminished, incomplete, or illiberal democracies (O'Donnell 1994, Collier and

Levitskyl997; Diamond 1999; Zakaria 1997, 2003) or, hybrid or semi-democratic

regimes (Hartlyn 2002, Carothers 2002, Schedler 2002, Levitsky and Way 2002,

Diamond 2002).

Several factors have been mentioned as the main obstacles for the development,

institutionalization, or consolidation of these new democracies. Among the ones that have

been more frequently considered, we have: (1) the persistence of authoritarian actors who

control important resources of power, (2) the prevalence of many social spheres of

profoundly authoritarian patterns of domination, (3) an attitude of neutrality and

indifference among elites and ordinary citizens about the new political regimes in place,

(4) the lack of strong and stable political institutions (e.g., political executives,

congresses, and political parties), and (5) the impact of economic crises and rampant

levels of social and economic inequality (O'Donnell 1992, Diamond 1999, Encarnacion

2003/2004, Garret6n 2003).









It is in relation to the previous point that it is possible to identify one critical

contribution of the literature on democratic consolidation. As Encarnacion (2000)

correctly argues, the literature on democratic consolidation has incited a debate that has

helped to overcome some of the limitations in the study of democratization. It is clear that

the literature on democratic transition responded to the abundance of structuralist

approaches accounting for the failure of democracy in developing countries4 by placing

political elites and their choices at the center of the analysis. Democratization was thus

seen as the product of strategic calculations made by political elites. Elite pacts and other

forms of consensus-driven politics were perceived as the factors most conducive to the

development of democracy in developing countries.5 To a considerable extent, the

literature on democratic consolidation challenges the previous dominant view about the

understanding of democratization processes (i.e., democratization as an elite-crafting and

voluntaristic phenomenon).

Scholars working within the democratic consolidation literature have been more

open to consider the role of structural factors-ranging from socio-economic structures to

institutional settings-in constraining and shaping both elite and ordinary citizens

political attitudes and behaviors. In doing so, some scholars have depicted the process of

democratic consolidation as less dependent on elite crafting, and more determined by

preexisting conditions and historical legacies often linked to the nature of these societies

and the preceding authoritarian regime (Encarnacion 2000). Consequently, the recent

literature on democratic consolidation has provided a unique opportunity to overcome the



4 See for example O'Donnell (1973).

5 See for example O'Donnell and Schmitter (1986) and Przeworski (1991).









dichotomy between structural-oriented and process-oriented approaches that have

characterized the study of democracy and democratization (Kitschelt 1992).

Regarding this more complex approach toward democracy and democratization,

and as part of the effort to define the meaning of consolidation, the literature on

democratic consolidation has emphasized the centrality of support for democracy among

elite and ordinary citizens. Specifically, several scholars (Gunther, Diamandouros and

Puhle 1995; Linz and Stepan 1996; Diamond 1999) have affirmed that any process of

democratic consolidation or institutionalization is impossible without both elites and

ordinary citizens supporting it. Such support has been defined in terms of key attitudinal

(i.e., both elites and ordinary citizens believing that a democratic regime is the most

acceptable and convenient kind of political regime) and behavioral changes (i.e., all

politically significant group respecting and adhering to democratic norms and

procedures). I revisit this argument extensively in the second section of this chapter.

Similarly, the literature on democratic consolidation has been significant because it

has generated an important debate regarding different understandings of democracy and

democratic regimes. This literature has been characterized by an intense debate about the

advantages and disadvantages of assuming either a minimalist or procedural conception

of democracy6 or a more maximalist or substantive one. For those scholars defending the

use of a procedural or minimalist conception, democracy means Dahl's (1971) definition

of polyarchy: (1) political contestation or competitive elections (assuming that these



6 It is important to mention here that I am not distinguishing between a minimalist and a procedural
definition of democracy because I believe that there is an important overlap between them. However, it is
possible to contend that Dahl's procedural definition of democracy departs from the minimalist definition
provided by Schumpeter (1942)-i.e., democracy is a political system in which the principal positions of
power are filled through a competitive struggle for the people's vote-because he contends the need for
civil and political freedoms to render political participation and competition meaningful.









elections are free and fair), (2) citizens' right to participative in public contestation, and

(3) a set of civil rights (e.g., freedom of the press, freedom of speech, freedom to join

organizations, etc.) to guarantee that competitive elections and political participation will

politically effective and meaningful. This concept of democracy sharply contrasts with

more substantive definitions of democracy that link democracy with some specific

outcomes such as economic equality, the satisfaction of people's needs, and the

achievement of higher levels of direct democracy.

The main argument favoring a minimalist conception of democracy contends that

thinking about democracy in more substantive, maximalist or outcome-oriented terms is

problematic in third-wave democracies. It assumes that democracy should provide

outcomes like economic equality and redistribution. As a consequence of this, it could

generate a situation in which ordinary citizens become impatient or ambivalent toward

the key components of a procedural definition of democracy; i.e., a set of procedures and

norms that guarantees a process of negotiation and dialogue to define what goals are

chosen and pursued.7

Moreover, the argument in favor of a minimalist conception of democracy also

claims that even the most established democracies in the world cannot be defined as

substantive or maximalist forms of democracy because democracy is only a form of

political regime that-in the context of some constitutional parameters and state

protection of civil liberties and human rights-gives citizens more opportunities to



7 See Anderson (2005) for an interesting account of the negative consequences of rejecting a minimalist or
procedural conception of democracy in dealing with the study of democratization in Latin America.
According to her, dismissing the importance of a minimalist or procedural definition of democracy not only
leads to pessimistic view about the current situation of democracy in Latin America, but also to the
possibility of creating unrealistic popular expectations toward these democratic regimes that eventually
could endanger their persistence.









choose and influence government's decisions. Therefore, the specific content of

government's decisions and policies is up to the citizens and their leaders. Finally, if

established democracies represent the most advanced cases of democratization and

democratic quality, this has been the outcome of a long-term experience with a

procedural democracy (Fuchs 1999, Dalton 2004, Anderson 2005).

However, it is important to acknowledge that there have been some important

criticisms regarding the adoption of a strict minimalist or procedural definition of

democracy in the context of new democracies. Accordingly, it has been affirmed that it is

possible to find several cases where the presence of all the components of a procedural or

minimalist democracy-at least in the way defined by Dahl-is not enough to nurture a

democratic regime. For example, Karl (1990) has argued that in several Latin American

countries, the presence of competitive elections, broad adult citizenship, and civil rights

does not guarantee that the civilian authorities will have the power to control tutelary

powers such as the military. As a consequence of this, popularly elected authorities

confront severe policymaking constraints. Additionally, Diamond (1999) has pointed out

that in many of the world's new democracies competitive elections have not ensured a

minimal level of political responsiveness.

A similar argument has been made by O'Donnell (1996) when he mentions that

several political regimes in Latin America have adopted all procedures and norms

contained in Dahl's definition of polyarchy. However, the problem is that along with

these formal democratic procedures and norms, these political regimes maintain several

informal rules (e.g., particularism, clientelism, neopatrimonialism, etc.) that are

antagonistic to the full institutional package of a polyarchy. Interestingly, O'Donnell does









not reject the convenience of developing procedural democracies within Latin American

or other new democracies countries; on the contrary, he is trying to understand why these

formal rules and procedures have a limited capacity in constraining and shaping the

political behavior of relevant political actors.

As a reaction to this type of criticism, it is common to find procedural or minimalist

definitions of democracy that have added other components or requirements to Dahl's

definition of polyarchy. For example, it has been stated that in order to have the minimal

conditions of democracy, elected authorities must be able to govern without being subject

to military or clerical leaders (Diamond 1999, Levitsky 2002). Something similar is

happening in regards to the rule of law because it is clear that the political significance of

civilian and political rights will be very weak if they are not effectively protected and

enforced by a rule of law (O'Donnell 2004a, 2004b). It is worth noting here that this new

trend does not imply the adoption of a substantive or maximalist definition of democracy.

It seems more an effort to understand the limited capacity of democratic norms and

procedures, in several new democracies, in making possible effective political

representation and inclusion.

Accordingly, some scholars have recently challenged the assumption that there is a

fundamental tension between a procedural definition and a more substantive definition of

democracy. O'Donnell (2004c) contends that this tension is somewhat artificial because

what really matters in the context of new democracies is improving the performance and

democratic quality of these democratic regimes. Following Dahl's contention (1971) that

a democracy or a polyarchy needs a set of individual and political rights to effectively

formulate and signify citizens' preferences (e.g., the freedom to form and join









associations, alternative sources of information, etc.), O'Donnell discusses the impact of

the political and social context on the functioning of the political regime. He goes on to

discuss the extent to which the social and political context enable or disable the capacity

of citizens to exert their political rights and citizenship.

Furthermore, O'Donnell insists on the contradiction of recognizing equal and

universal political rights when citizens lack fundamental civil rights (e.g., fair and

expeditious access to the legal system). He also discusses how severe economic

deprivations hamper the capacity of an individual to be a political agent.8 Addressing the

situation of several Latin American countries, he shows how the existence of a biased and

intermittent rule of law makes impossible the generalization of the most basic civil rights

for all citizens. Similarly, the combination of huge levels of poverty and inequality in

these societies alongside a limited implementation of social rights definitively disables

the political capacities of larger segments of the population. Therefore, several

democratic regimes in Latin America are characterized by a low quality in terms of the

two basic components of democracy-political representation and political inclusion.

Nevertheless, instead of claiming that the problem with these new democracies is

its formal or minimalist nature, O'Donnell suggests that the most effective strategy is to

place more emphasis on the development and extension of civil rights because they not

only protect, but also empower citizens giving them a better framework to define and

defend their political interests, and exert all their political rights. This would be

particularly relevant for most disadvantaged citizens and their struggle for the generation




8 According to O'Donnell (2i 1'4c) the definition of an individual as an agent implies the recognition of his
or her capacity to make reasonable choices.









of policies against poverty and inequality, and the recognition of some key social rights

by their political regimes.

Overall, O'Donnell's arguments are relevant for several reasons. He suggests a

critical connection among political, civil, and social rights. He argues that the quality of

the democratic regime will largely depend on the connections between political, civil, and

social rights. By the same token, he is not suggesting that the political or procedural

rights are irrelevant, but rather to understand why some social and political context

features block their potential. In this regard, democracy can be considered as a valuable

and necessary experience because it assumes that all individuals have the intrinsic

capacity and the right to make reasonable choices, and because it provides a framework

where a gradual and subsequent improvement in civil, political, and social right is

possible. Finally, he also suggests that advancements in social rights will not be easily

reversed if they can be exercised throughout the formal democratic procedures.

All these contentions provide important insights for our understanding of

democracy and democratization processes. First, democracy does not need to be defined

either entirely in procedural or political terms, or entirely in substantive terms. Second, as

some scholars have considered, the critical issue is to understand why in the context of

several new democracies, democratic regimes have a very low quality in terms of

political representation and political inclusion. The main challenge is then to improve the

quality of these democratic regimes. Accordingly, it becomes important to recognize the

relevance of democratic norms and procedures; but also to address how the political,

social and economic contexts diminish the potential of these democratic norms and

procedures.









It is important to note here that in the last few years, more scholars have started to

address specifically the quality of the new democratic regimes. So far, we know that

some key factors have the capacity to make a negative impact on the quality of new

democratic regimes. Among these factors we have the lack of horizontal accountability,

the lack of a strong and consistent rule of law, the weaknesses of some key political

institutions such as political parties, and extreme levels of economic and social inequality

(O'Donnell 2004a, 2004b, 2004c; Beetham 2004; Schmitter 2004; Rueschemeyer 2004;

Powell 2004). Similarly, as Diamond (2004:21) points out, we know that all these

different factors not only overlap, but also depend upon one another, forming a system in

which improvement along one of these democratic dimensions (e.g., participation) can

have beneficial effects along others (e.g., equality and accountability). However, we also

know that very frequently it is going to be very difficult to improve all these democratic

dimensions at once.

Finally, another important contribution of the literature on democratic

consolidation has been the debate about what the experiences of these new democracies

are revealing in terms of its possibilities of further democratization. The consolidation of

the new democracies is doubtlessly a desirable outcome, but it is clear that the majority of

these new democratic regimes do not necessarily advance in that direction. Furthermore,

it is becoming increasingly clear that the mere survival of formal democratic regimes

does not necessarily indicate a transition toward democratic consolidation.

Consequently, some scholars have called attention to a potential bias in the study of

new democracies (Carothers 2002, Levitsky and Way 2002). Carothers (2002), for

example, has pointed out the limitations of the democratic consolidation literature that









was attached to what he defined as the "transition paradigm". According to him, this

paradigm was based on a set of assumptions that were called into doubt by the political

trajectories of most third-wave countries. The key assumptions of this "transition

paradigm" were the following: (1) that any country moving away from dictatorial rule

can be considered a country in transition to democracy; (2) that democratization tends to

unfold in a set of stages: opening, breakthrough, and consolidation; (3) high expectations

for what the establishment of regular, genuine elections will do for democratization; (4)

that a country's chances for successful democratization primarily depend on the political

intentions and actions of its political elites without significant influence from underlying

economic, social, and institutional conditions and legacies; and (5) that state-building is a

secondary challenge to democracy building and largely compatible with it.

The main problem, according to Carothers, is that all these assumptions have been

proven wrong. For example, the assumptions that any country moving away from

authoritarian rule was in transition to democracy and that there will follow a sequence of

opening, breakthrough, and consolidation have often been inaccurate and misleading

because some of these countries hardly democratized at all. Moreover, most common

pattern has been the development of stable feckless pluralism regimes that are not in

transition to be liberal democracies. There are also problems with the assumption that

regular, genuine elections will not only give government democratic legitimacy but also

foster a greater deepening of democratic participation and accountability. Regular and

genuine elections are often held but they do not guarantee the development or renovation

of political parties, and an important level political accountability. To add to the

complication, elections do not change long traditions of personalistic leadership, and









stagnant patronage-based politics. Finally, it has become clear that various structural

conditions (e.g., the level of economic development, past experiences with political

pluralism, the lack of a coherent state, etc.) heavily shaped the political outcomes of post-

authoritarian regimes.

Regarding the shortcomings of this transition paradigm, it is worth noting that they

should not be seen as evidence that third-wave democracies are doomed to never become

functioning liberal democracies. Rather, they should help us to assume that the middle

ground between full-fledged democracy and outright dictatorship is actually today's most

common political outcome of countries in the developing and the post-communist world

(Carothers 2002:18).

As expected some scholars have reacted negatively to Carother's criticisms.

O'Donnell (2002) affirms that Carothers lumps together a large and uneven body of work

under the heading of transition paradigm, and then proceeds to concentrate his criticisms

on some of the weakest arts of it. Moreover, O'Donnell presents several examples

showing that some of the Carother's main criticisms were already acknowledged in the

literature about third-wave democracies. For example, he mentions that O'Donnell and

Schmitter (1986) in one of the first works about democratic transitions were explicit in

affirming that these transitions do not necessarily lead to democracy. Rather, they may

instead lead to authoritarian regressions, to revolutions, or to hybrid regimes.

Regarding the role of structural factors in shaping the political outcomes of third-

wave democracies O'Donnell cites the work by Przeworski et al. (2000) that contends

there is a connection between a country's level of socio-economic development and the

democracy's durability. Indeed, it is not hard to find evidence that this was not an entirely









new topic within the literature on democratic consolidation. Besides Przeworski et al.

(2000), other scholars have been disappointed that with few exceptions the literature on

democratic consolidation has emphasized-as the literature on democratic transition

did-the role of political processes in affecting regime change and asserted the autonomy

of these processes from structural factors (Remmer 1996, Gasiorowski and Power 1998).

According to Remmer (1996:630) one of the main problems with this literature is that

"comparativists have all but abandoned efforts to generalize about the macro-social

prerequisites for political democracy in favor or more contingent understandings

emphasizing the strategic choices of political actors." Without exploring how structural

factors may affect the development of democracy within new democracies, this literature

has neglected the possibility of working in a more comprehensive understanding of the

determinants of democratic consolidation that must take into account political processes

and structural factors.

Therefore, it seems reasonable to contend that Carother's main contribution has not

been the novelty of his propositions, but rather the effort of summarizing several ideas

that were already part of the literature on democratic consolidation. In doing that, he has

encouraged an interesting debate about what we have learned regarding the recent

processes of democratization and the development of democratic regimes in developing

countries. By the same token, this exchange between Carothers and some scholars

dealing with the issue of democratic consolidation in new democracies has been

particularly useful for the development of a new research agenda about these recent

experiences of democratization and that is not constrained by the transition paradigm.









Main Limitations and Problems of the Literature on Democratic Consolidation

The literature on democratic consolidation has not been exempt from important

limitations and problems. On a more general level, O'Donnell (1996) has argued that

democracy and consolidation are terms too polysemic to make a good pair. Schedler

(1998) has also contended that one of the main problems with the literature on democratic

consolidation was the multiple usage and meanings of the term "democratic

consolidation." According to him, the original meaning of democratic consolidation (i.e.,

the challenge of making new democracies secure, of extending their life expectancy

beyond the short term, of building dams against eventual "reverse waves") was expanded

in so many directions that there is now a conceptual disorder.

As a consequence of this trend, the list of the problems contained in the democratic

consolidation literature includes such divergent items as popular legitimation, the

diffusion of democratic values, the neutralization of anti-system actors, civilian

supremacy over the military, the elimination of authoritarian enclaves, party building, the

organization of functional interests, the stabilization of electoral rules, the routinization of

politics, the decentralization of state power, the introduction of mechanisms of direct

democracy, judicial reform, the alleviation of poverty, and economic stabilization.

Moreover, Schedler claims that there are at least five very different definitions of

democratic consolidation. According to him, democratic consolidation can mean: (1)

avoiding democratic breakdown; (2) avoiding democratic erosion; (3) completing

democracy; (4); deepening democracy, or (5) organizing democracy. Interestingly, and

very frequently, the particular meaning of democratic consolidation that is embraced will

depend on where scholars stand in empirical terms (i.e., what particular cases among new

democracies they are considering) and what they are thinking about what is possible or









desirable to aim for these particular cases (i.e., the role of different definitions of

democracy). For example, if an electoral democracy or a group of electoral democracies

are facing the possibility of returning to an authoritarian regime, democratic

consolidation in this context will tend to mean avoiding a democratic breakdown. On the

contrary, if an electoral democracy has the possibility of moving forward toward

becoming a liberal democracy, democratic consolidation in that context will tend to be

defined as a process of completing democracy.

To a large extent, the existence of such different definitions of democratic

consolidation is consistent with the high level of diversity that we can observe within the

group of new democracies. Additionally, there is no doubt that the existence of different

levels of advancement in terms of the democratization of these political regimes is one of

the most important sources for these empirical differences. In that regard, some of them

seem to move forward, some of them seem stuck in a gray zone between democracy and

authoritarianism, and some of them-if they have not already suffered a democratic

breakdown-seem to be in risk of moving backwards. Therefore, when scholars are

dealing with issues of democratic consolidation, they are not necessarily sharing the same

concerns and the same questions.

Another problem with the literature on democratic consolidation is that it is

characterized by an uneven development. The literature has been extensive and

systematic regarding the shortcomings of new democracies vis-a-vis established

democracies, and regarding which should be considered the final outcome of democratic

consolidation, but less so in accounting for the particular features of new democracies

and the particularities of its processes of democratization. In addressing this shortcoming,









Encarnacion (2000) contends that very frequently the literature on democratic

consolidation has failed to explain the internal dynamics of democratization and

consolidation.

Specifically, regarding the shortcomings of new democracies, scholars working on

third-wave democracies soon realized that they needed more than the distinction between

authoritarian and democratic regimes because in so many cases these post-authoritarian

and formally democratic regimes were lying somewhere between democracy and

authoritarianism. The problem has been that the majority of cases have been solved by

labeling new regimes as a "diminished subtype" of democracy (Collier and Levitsky

1997). A "diminished" type of democracy is a political regime that either lacks one or

more attributes that we find in the most consolidated cases of democracy, or that has such

attributes in a diminished way (e.g., popular elections that are not free and fair). A

"diminished" type of democracy tends to be in gray zone area because it combines

democratic and non-democratic features; and it is the case that the majority of new

democracies can be classified as gray zone democracies (O'Donnell 1996).

The main problem with this strategy, as O'Donnell (1996) points out, is that these

"negative" definitions address these new democracies mainly according to some missing

attributes, assuming that there is only one type of democracy-the one found in most

developed countries. According to O'Donnell, it is clear that this strategy represent an

ethnocentric and teleological approach unable to account for the different types of

political regimes that exist among third-wave democracies.9 Other scholars have been



9 Franco (1998) presents a similar argument when he contends that the definition of political regimes in
developing countries as "diminished" or "anomalous" subtypes of democracy must be avoided because it
means that we are giving up regarding the understanding of the concrete and typical features of these
political regimes.









even more critical about the extensive use of "diminished" types of democracies. For

example, Collier and Levitsky (1997) have asked whether these "diminished"

democracies should in fact be treated as subtypes of democracies rather than subtypes of

authoritarian regimes.

However, it is fair to say that not all these formulations of a "diminished" type of

democracy among new democracies are equally problematic. Some of them have meant

an important contribution to understanding new democracies either because they were

related to one specific region or because they were devoted to address one specific aspect

of these new democratic regimes. For example, the concept of delegative democracies

proposed by O'Donnell (1994) focused on the level of horizontal accountability among

new democracies in Latin America. In doing so, he provided not only a good

characterization of some of the new democratic regimes that can find in this region (i.e.,

Argentina, Brazil, Peru, etc.), but also an interesting discussion regarding some

challenges for its institutionalization.

In dealing with this explosion of "diminished" types of democracies, some scholars

have been particularly clear about the need of identifying particular types of political

regimes between liberal democracies and closed authoritarian regimes (Diamond 2002,

Levitsky and Way 2002, Schedler 2002 and van de Walle 2002). For example, all these

aforementioned scholars think that it is possible to make a distinction between "electoral

democracies" and "competitive authoritarianism" within new democracies. These

different types of political regimes will present some democratic elements, but the extent

to which these democratic elements pattern political competition varies considerably.









Electoral democracies, as liberal democratic regimes, meet the minimum

requirements of democracy, but fail to institutionalize vital dimensions of democracy

such as the rule of law and political accountability (Schedler 2002). Indeed, several Latin

American countries can be classified as electoral democracies. On the contrary,

competitive authoritarianism is a political regime where formal democratic institutions

are viewed as the principal means of obtaining and exercising political authority, but

incumbents violate those rules so often and severely that the regime fails to meet the

minimum standards for democracy. Specifically, incumbents routinely manipulate formal

democratic rules but they are unable to eliminate them or reduce their potency. Rather

than openly violating democratic rules (e.g., by banning or repressing the opposition and

the media), incumbents are more likely to use bribery, co-optation, and more subtle forms

of persecution, such as the use of tax authorities, complaint judiciaries, and other state

agencies to "legally" harass or extort cooperative behavior from critics. However, even

when the cards are stacked in favor of autocratic incumbents, the persistence of

democratic institutions creates arenas through which opposition forces may pose

significant challenges (Levitsky and Way 2002).

There is no doubt that all these theoretical efforts are important because they

improve our understanding and classification of new democracies, and make possible to

have a better idea, depending of the circumstances, about the main challenges and

prospects of subsequent processes of democratization. Nevertheless, it is also clear that

we are still far from reaching an important level of consensus regarding how to classify

the diverse group of new democracies. Equally importantly, we still need a great extent of

theoretical and empirical work to improve our understanding of the particular nature of









these political regimes, as well as our understanding of its democratic advancements and

setbacks.

New Democracies and Popular Support for Democracy

As mentioned before, after the massive installations of democratic regimes in

developing countries and former communist countries during the 1980s and 1990s, it is

evident that only a relatively small number have made a significant progress in terms of

further democratization, and are still enjoying a positive dynamic of incremental

democratization. On the contrary, most of these countries experimenting with democracy

are neither dictatorial nor clearly headed toward democracy (Diamond 2002, Levitsky

and Way 2002, Schedler 2002, van de Walle 2002). They have some attributes of

democratic political life, such as limited political space for opposition parties, an

independent civil society, regular elections, and democratic constitutions. At the same

time, they suffer from serious democratic deficits, including poor representation of

citizens' interests, low levels of political participation beyond voting, frequent abuse of

the law by government officials, elections of uncertain legitimacy, very low levels of

public confidence in state institutions, and persistently poor representation by the state

(Carothers 2002:9-10). In general, it is no only the case that most of these new

democracies lack one or some aspects of democratic rule, but that this overall democratic

quality is very low.

As a direct consequence of this democratic deficit trend, a large majority of these

troubled new democracies suffer the lack of a significant and stable level of popular

support for democracy (Diamond 1999, Moreno 2001). In some cases, this situation has









led to a clear authoritarian reversal, e.g., Peru in 199210 (Carri6n 2006) and to the

development of stable political regimes that can be best described as electoral

authoritarianism11 (Schedler 2006). Also, in other cases the lack of support for

democratic procedures and institutions has led to a high level of political instability

caused by very frequent non-constitutional depositions of democratically elected

presidents, e.g., Ecuador in 2000 and 2005 and Bolivia in 2003 (Mainwaring and

Hagopian 2006).

In turn, it is clear that one of the most negative consequences of authoritarian

reversals and processes of democratic erosion is that they make it difficult for these

recent experiences with democracy to persist, which is absolutely necessary for the

development of a democratic learning process and the possibility of reaching a dynamic

characterized by incremental democratization.12 For example, it seems unreasonable to

expect improvements regarding the current democratic deficits in several new

democracies if some political actors, relying on the indifference or lack of support for

democracy among ordinary citizens, can constantly manipulate and eventually suspend

the main democratic procedures.

Consequently, it is reasonable to contend that there is an important connection

between the lack of support for democracy or democratic legitimacy-among elites and



10 Chapter 5 provides a detailed description and interpretation of this recent Peruvian political experience.

11 According to Schedler (2006:3), electoral authoritarian regimes play the game of multiparty elections by
holding regular elections for the chief executive and a national legislative assembly. However, they violate
the liberal-democratic principles of freedom and fairness so profoundly and systematically as to render
elections instruments of authoritarian rule rather than instruments of democratic rule. Contemporary
political regimes in Russia, Algeria, Egypt, Cameroon, Chad, Ethiopia, Cambodia, Malaysia and
Singapore, among many other cases, can be considered as clear examples of electoral authoritarianism.
12 See Anderson and Dodd (2005) for how they defined a democratic learning process in the case of
contemporary Nicaragua.









ordinary citizens-in many developing countries and the lack of democratic stability and

advancement of its new democratic regimes. However, the problem is that for a long

period of time, thinking carefully about the development of popular democratic

legitimacy was not one of the most important issues in the literature on democratic

consolidation. In this regard, several scholars (O'Donnell 1992, Valenzuela 1992)

dismissed the importance of having a majority of democrats among the population as a

key condition for the transition toward democratic rule and the development of an

institutionalized democratic regime.

For example, O'Donnell (1992:19-20) stated that there are no cases in the world

where there has been a majority of democrats before the advent of political democracy.

Moreover, according to him, the development of this majority of democrats will be only

possible throughout a continuous practice of political democracy.13 Nevertheless, the

main problem with this contention is that the experience of several new democracies is

showing that a continuous practice of democracy is almost impossible to achieve without

some considerable and stable level of democratic legitimacy among ordinary citizens.14

More specifically, as Rose, Mishler, and Haerpfer (1998) affirm, it is hard to think that

political elites-the supply side in a political system-will have any incentives to play

according the main democratic norms and procedures if there is not a minimal demand in

that direction from ordinary citizens-the demand side of a political system.




13 A similar argument can be found in Rustow (1970) when he affirms that democratic institutions become
stable and acquire legitimacy in part through the passage of time, mainly because political actors become
habituated to democratic practices.
14 In the context of Latin America, and besides Peru, Venezuela represents another case where a limited and
deficient representative democracy has been replaced by a hybrid regime that in several aspects does not
represent a democratic advancement (McCoy and Myers 2004).









By the same token, in dealing with the relevance of popular support for the stability

and eventually consolidation of new democracies, some scholars have mentioned that

universal democratic legitimacy is not necessary in the beginning stages of

democratization. However, if commitment to democracy does not emerge over time, and

common citizens do not develop some considerable level of democratic commitment,

democracy will likely fail because these new democratic regimes will be vulnerable to

disloyal leaders and groups (Mainwaring 1992:307). Therefore, the argument in favor of

the importance of democratic legitimacy is not claiming that a strong popular

commitment toward democracy is a pre-requisite for democratic consolidation. On the

contrary, the idea is that democratic consolidation will be almost impossible without a

gradual development of popular support for democracy.

Accordingly, democratic consolidation has been clearly defined as a process that

implies fundamental attitudinal and behavioral changes among elites and ordinary

citizens. Specifically, it has been argued that a successful process of democratic

consolidation implies that the majority of public opinion, even in the midst of major

economic problems and deep dissatisfaction with incumbents, should hold the belief that

democratic procedures and institutions are the most appropriate way to govern collective

life (Linz and Stepan 1996). In a very similar way, according to Gunther, Diamandouros,

and Puhle (1995), a democratic regime can be considered as consolidated when all

politically significant groups, in both attitudinal and behavioral dimensions, adhere to the

democratic rules of the game and regard the main political institutions as the only

legitimate framework for political contest.









Unfortunately, the recognition of the importance of democratic legitimacy-

particularly at the level of ordinary or ordinary citizens and as a sign of democratic

consolidation-has not been accompanied by a more comprehensive and systematic

theoretical and empirical consideration of this topic. In that regard, very frequently in the

last years, the topic of popular support for democracy has been considered as one among

several factors that has the capacity either to facilitate or hinder the development of

democracy. Moreover, little attention has been devoted to its interaction with the rest of

these critical factors (Diamond, Linz, and Lipset 1989).15

This situation is problematic not only because in order to develop theories about

democracy and democratization processes we have to move beyond general lists of

factors that matter in almost all the cases toward more systematic statements about what

factors the matter most in particular cases and situations (Geddes 1999); but also because

it seems reasonable to expect that the possibilities of consolidation or institutionalization

of new democratic regimes-especially within developing countries-will depend, to a

large extent, on the current and future levels of popular support for democracy,

democratic regimes, and democratic procedures and institutions.

Hence, it is the case that our understanding of popular support for democracy in

new democracies is still limited. Among other things, we are still struggling with how to






15 Along with the need of a democratic political culture, these long lists tend to include all types of factors
that can influence the development of democracy in developing countries (e.g., the previous authoritarian
regimes, the effect of the international context, the existence of an effective state, the existence or not the
representative and strong political institutions, the economic and political performance of the new
democratic governments, the level of institutionalization of the party system, the quality of political
leadership, the impact of the level of socio-economic development, the existence of a pluralistic and
autonomous civil society, etc.).









operationalize and measure this complex political phenomenon,16 and we are not clear

about what can be considered the main determinants of this type of political support.17 In

trying to change this situation, it is critical to pay close attention to some recent works

that have been dealing, in direct or indirect way, with the topic of popular support for

democracy.

One of the main strengths of these recent efforts is that they go beyond what can be

considered as the traditional political culture approach. This approach basically contends

that people's attitudes and orientations toward democracy are the product of depth-

political values (e.g., interpersonal trust, political tolerance, etc.) rather than the product

of short-term evaluations of government and regime performances, or the product of

individual's particular social and economic conditions (e.g., class or socio-economic

levels).

Specifically, some scholars have pointed out that there are key differences in the

way ordinary citizens in developing countries perceive and define democracy vis-a-vis

ordinary citizens of the most developed countries. In the case of developing countries,

and particularly for those who lack the means to satisfy their basic needs, democracy is

not only associated with political components (i.e., rule of law, civil liberties, competitive

elections, etc.), but also with other components, such as welfare and the dispersion of

economic resources (Camp 2001).



16 In the context of new democracies, there has been some advancement in the case of post-communist
countries, but still there are so many controversial issues. See Chapter 4 for a discussion of how the issue
has been addressed in established democracies, post-communist and Latin American countries.

1 Unfortunately, scholars frequently dealing with this topic tend to emphasize only one main theoretical
perspective (e.g., a political culture perspective, structural explanations, or the importance of the economic,
social, and political performance of these new democratic regimes) rather than addressing the complexity
of popular support for democracy in new democracies.









The main idea behind this contention is that there are high expectations toward

democracy within developing countries, and that these expectations include both political

and economic aspects. Therefore, the level of democratic legitimacy in these countries

will depend not only on the new democratic regime's ability to provide a set of rules and

procedures that make possible authentic political competition and participation (i.e., a

popular election of government), but also in its ability to provide effective inclusion and

representation of the needs and aspirations of the most disadvantaged citizens (Roberts

1998, Diamond 1999, Panizza 2000).

Therefore, it seems that the poor performance of democratic governments in several

developing countries is one of the main reasons for the lack of popular support for current

democratic regimes and one of the main sources of popular alienation regarding

democratic institutions and democratic procedures (Mainwaring and Hagopian 2006,

Hagopian 2006). Furthermore, these very negative assessments are not only related to the

economic performance of democratic governments and regimes but also to their

performance in terms of the quality of the rule of law and the precarious situation of civil

or individual rights in these countries (Foweraker and Krznaric 2002; O'Donnell 2004a,

2004b, 2004c).

It is critical to note here that in dealing with the lack of popular support for

democracy within several new democracies, it is the same people who have high

economic and social expectations regarding their current democratic regimes who live

under an irregular and weak rule of law that severely limited their civil and political

rights. This is what O'Donnell (2004b, 2004c), for the case of Latin America, has defined

as a strong connection between poverty and weak rule of law. Accordingly, there is not a









coincidence that the rule of law is particularly weak or non-existent for the sectors of the

population most marginalized socially and economically. Individuals with higher

education, more access to different resources and more capacity to exert their civil and

political rights will inevitably have more power to shape the state's political agenda.

In other words, it is possible to contend that the lack of popular support for

democracy in new democracies is related to the fact that a large part of the population is

suffering simultaneous social, economic, and political exclusion. As a consequence of

this, scholars who claim that political equality is almost impossible in a context of huge

wealth and status inequalities (Rueschemeyer 2004; PNUD 2004; O'Donnell 2004b,

2004c) seem to be pointing in the right direction. 18

It is also possible to be more specific about how particular social, economic and

political conditions (i.e., high levels of poverty and inequality) explain the lack of popular

democratic support if we consider the development of main political attitudes among the

most disadvantaged citizens of several developing countries. For example, in the case of

some Latin American countries, some scholars have addressed the relationship between

non-democratic attitudes (i.e., clientelism and plebiscitarianism) and social and economic

structural conditions (Parodi 1993, Stokes 1995, Murakami 2000). According to these

researchers, poor Peruvian citizens have the tendency to develop a clientelistic or

plebliscitarian patterns of interaction with governments and authorities. These practices

or patterns of political interactions are based on a highly instrumental attitude toward

governments and political regimes rather than by a clear support to the main democratic


18 On contrary to O'Donnell (21" 4) who contends that the best strategy to overcome this situation is the
expansion of civil rights, Rueschemeyer (21" 4) believes that the best strategy is the mitigation of economic
and social inequalities, declaring that certain goods such as health, education and a minimal income are
social rights.









principles and procedures. These clientelistic and patrimonialistic attitudes are

particularly strong among the poorest and marginalized population. 19

In conclusion, it is important to develop a more systematic approach toward the

topic of popular democratic legitimacy or popular support for democracy in the study of

new democracies and the debate about its further democratization prospects. This is

especially urgent in those cases where the installation of formal democratic regimes,

following Dahl's (1971) definition of polyarchy-i.e., extensive competition for power

through regular, free and fair elections; highly inclusive citizenship conferring rights of

participation on virtually all adults; and even civil and political liberties-has not been

sufficient to generate a strong commitment toward democratic principles and institutions

among ordinary citizens and elites (Shin 1994, Diamond and Plattner 1996).

Specifically, there are some issues that demand a closer examination. First,

although it is appropriate to define democratic consolidation as a process of gradual

changes in ordinary citizens and elites attitudinal and behavioral dimensions, the main

remaining challenge is to understand what factors shape the political attitudes,

perceptions, and beliefs of ordinary citizens. It is critical to address and understand how

ordinary or ordinary citizens in new democracies approach and define politics and

democracy, and assess the performance of their democratic governments.

Second, it is important to recognize that one of the main difficulties for the

development of popular support for democracy is the combination of a large population

that as a consequence of their acute economic and social marginalization have diverse



19 Bratton and van de Walle (1997) make a similar argument for several new African democratic regimes.
According to these authors, neopatrimonialism-defined as an informal political institution associated with
the prevalence of "big men" syndrome and their networks of personal loyalty in government and public
life-makes it difficult to develop popular legitimacy toward main democratic institutions and procedures.









and high economic and social expectations toward democracy-on the one hand-, and

the poor performance of current democratic governments in these areas-on the other

hand. Nevertheless, it is also critical to acknowledge that several new democracies are

having also problems in delivering the minimum conditions that can make individual and

political rights effective instruments to increase the level of political representation,

accountability, and inclusions (e.g., an effective and unbiased rule of law). This situation

is even more critical if we consider that social and economic marginalization goes hand

by hand with a very precarious situation in terms of the protection of individual, civil,

and political rights (Tokman and O'Donnell 1998). Accordingly, it is critical to consider

that political regimes are embedded in particular social and economic contexts

(O'Donnell 2004b, 2004c).

Finally, it is important to be aware that dealing with the topic of popular support for

democracy, and recognizing how some social and economic structural factors play a

critical role in the development of democracy, is a good opportunity to overcome the

traditional dichotomy between structural-oriented and process-oriented approaches that

has characterized the study of democratization (Kitschelt 1992). Indeed, there is nothing

wrong in assuming that the democratic consolidation of new democracies is a

"democratic crafting" process (Di Palma 1990). The problem lies in assuming that

structural factors and conditions, such as economic and social inequality or the quality of

the rule of law, are not particularly important. This is a good opportunity to recognize

that one of the main failures of the literature on democratic consolidation has been its

neglect of the insights from structural approaches (Gasiorowski and Power 1998).









Accordingly, a systematic approach toward attitudes, perceptions, and beliefs of

ordinary citizens in new democracies is one of the most important steps toward the

understanding of the limits and possibilities of further democratic advancement, and,

eventually, consolidation of new democratic regimes. By the same token, in order to

understand the lack of popular legitimacy of the current democratic regimes in these

countries, it is necessary to assume the challenge of thinking more systematically not

only about the several and different explanatory variables, but also about the best

alternative to operationalize and measure our outcome or dependent variables: the

different key dimensions of support for democracy.















CHAPTER 3
ADDRESSING AND EXPLAINING POPULAR SUPPORT FOR DEMOCRACY

This chapter addresses one of the most common theoretical approaches used in

accounting for popular support for democracy: the political culture approach. I will make

a critical assessment of this theoretical approach and stress its main contributions and

limitations in dealing with popular support for democracy, particularly in the context of

new democracies. There is no doubt that the political culture approach will be an

important component of any systematic approach toward popular support for democracy,

but this endeavor also requires additional efforts to overcome some of its limitations, and

to consider alternative theoretical approaches.1

One of the most common ideas among scholars dealing with democracy and

democratization is that elites and ordinary citizens' support for democracy2 is vital for the

maintenance and development of democracy. Nevertheless, as Mishler and Rose (1996,

1999) point out it is impossible to identify an absolute level of support sufficient for

democracy to flourish because, among other reasons, popular support for democracy is a


1 According to Norris (1999a) there are three main theoretical approaches to explain cross-national
variation in popular support for democracy: (1) the role of cultural values, (2) the performance of
democratic governments, and (3) the influence of different institutional designs.

2 It is worth noting here that one of the main problems with the concept of support for democracy is that
scholars do not necessarily agree on its particular meaning. As a consequence of this, their definitions range
from support for democratic values to support for democratic institutions.

3 Among other critical factors for democracy and democratization-particularly in the context of new
democracies-we can consider the level of socio-economic development, the severity of social cleavages,
the levels of economic and political stress, the design of political institutions, the structure and strength of
the state, an active and autonomously organized civil society, the performance of democratic governments
and institutions, and even some international factors (Diamond, Linz, and Lipset 1989). Unfortunately, one
of the aspects less developed in this literature is the interrelation among these critical factors; and how they
shape popular support for democracy.









necessary but not sufficient condition for democratic stability.4 Therefore, the most

certain thing that can be said about the relationship between support for democracy and

democratic stability is that a considerable or stable level of this support facilitates the

maintenance of democracy, while a declining level can undermine democracy and,

eventually, cause is its collapse. A stable or upward trajectory of public support is even

more important for new democracies because these new regimes not only confront higher

levels of political and economic stress, but also lack the long and continued democratic

experience that enables established democracies to cope with political challenges.

Additionally, it is also important to highlight that democratic legitimacy is not only

vital for the stability of democratic regimes, but also for its governance (i.e., the capacity

to formulate and implement policies in an effective manner). The citizenry's confidence

in the regime, its institutions and authorities, enhances the government's abilities to make

decisions and commit resources without resorting to coercion or the need to obtain

approval from the citizens for every decision (Mishler and Rose 1999). As Levi (2006)

points out, the lack of governance is a recurrent problem within new democracies, as well

as a source of democratic regressions.

Starting during the 1960s, an extensive literature has documented the phenomenon

of popular support for democracy, mainly in the context of established democracies.

More recently, scholars have devoted considerable attention to this topic within new

democracies. From this trend, it has become clear that there are substantial cross-national

variations in support for democracy within established democracies, within new

democracies and, particularly, between these two groups. Next section, addresses in more


4 The same argument can be found in Lipset (1959), Almond and Verba (1963), Dahl (1971), Linz and
Stepan (1978), O'Donnell and Schmitter (1986), Diamond (1994), and Linz and Stepan (1996).









detail what I define as a mainstream political culture approach in regards to the topic of

popular support for democracy. Given that popular support for democracy is, to a large

extent, a political attitudinal phenomenon, it seems relevant to pay attention to this

political culture approach and how it has been used to account for ordinary citizens'

values, and belief orientations toward politics and democracy.

It is worth noting here that I recognize that the orientations and decisions of

political elites are critical to understand the fate of processes of democratic transition and

democratic consolidation (O'Donnell and Schmitter 1986, Villal6n and VonDoepp 2005).

Nevertheless, I have decided to focus on the attitudinal dimension of ordinary citizens

because although political elites exert a critical influence on political developments, it is

the case that their strategic calculations are shaped by the distribution of political and

regime preferences at the popular level (Alexander 2002).

Political Culture Approach

The contemporary trajectory of the political culture research is particularly

interesting. During the 1950s and 1960s there was an exceptional level of interest in this

topic because the behavioralist revolution at the time considered individuals as the main

unit of analysis, emphasized the use of empirical data, and looked to universal theories

(Lane 1992). Following this period, interest in political culture declined especially

because of two main challenges. First, during 1960s and 1970s, the legitimacy of political

culture research was challenged by the idea that "political culture" basically reflected

class or ethnic status. Second, during this time, rational choice also challenged the

relevance of political culture, stating that it was unnecessary because self-interest and

short-term rationality were enough to explain individual behavior. However, in the last

years both Marxist interpretations of political culture and rational choice approaches have









been weakened as the most comprehensive explanations of individual behavior. This

trend has encouraged the return of political culture approaches. Moreover, as part of this

trend, it has been widely acknowledged that political culture can make important

contributions to our understanding of key processes like democratization and economic

development (Almond 1994).

In the context of a renaissance of political culture,5 it is particularly important to

pay attention to the main theoretical and empirical contentions of this particular approach

toward politics. According to the seminal work on political culture by Almond and Verba

(1963:14), the political culture of a nation is the particular distribution or patterns of

orientations toward political objects among the members of the nation. These different

orientations include: a cognitive orientation (i.e., knowledge of and belief about the

political system, its roles, and the roles of the incumbents); an affective orientation (i.e.,

feelings about the political system, its roles, personnel, and performance); and an

evaluational orientation (i.e., judgments and opinions about political objects based on the

combination of values, feelings, and information). These different orientations are not

entirely independent from each other, for example, political information and feelings play

an important role in evaluational orientations. Unfortunately, as I will address in the next

section, this consideration is missing in many contemporary works that use a political

culture approach.

By the same token, several factors, such as socialization, education, media

exposure, and adult experiences with governmental, social, and economic performance

play an important role in shaping political culture. Additionally, political culture is


5 About the current renaissance of political culture see Inglehart (1988), Almond (1994), Diamond (1994),
and Power and Clark (2001).









expected to affect political and governmental structures because values, feelings, and

beliefs are key determinants of political behavior. Finally, regarding the connection

between political culture and democracy, one of the Almond and Verba's main

conclusions is that democratic stability requires a very specific type of political culture

and democratic citizenship (i.e., the "civic culture"). This civic culture is defined as a

mixed-model where participant political attitudes (i.e., a participant political culture) are

balanced by considerable levels of passivity, trust, and deference toward authority (i.e.,

subject and parochial political cultures).

Interestingly, a very similar theoretical approach can be found in several other

works that, during the 1960s, were also dealing with political culture and its impact on

the political system. For example, Pye (1965) contended that the notion of political

culture assumes that the attitudes, sentiments, and cognitions that inform and govern

political behavior in any society are not just random congeries but coherent patterns that

mutually reinforce each other. By the same token, Dahl (1966), when discussing the

impact of different types of political orientations on patterns of political partisanship,

claimed that the orientation toward the political system as a whole affected the

distribution of loyalty in a national society. Further, he claimed that attitudes toward

cooperation, individuality, and other people affected the formation of political groups and

their interactions. Finally, in dealing with the connection between political culture and

political development, Verba (1966) stated that the most important components of a

political culture included a sense of national identity, attitudes toward oneself as

participant, attitudes toward one's fellow citizens, attitudes and expectations regarding









government output and performance, and knowledge about and attitudes toward the

political processes of decision making.

There is no doubt that this seminal political culture approach has been very

influential on the work of several scholars that came later and were also interested in

political culture issues. Accordingly, several of its main contentions can be found in most

recent works dealing in one way or another with the topic of political culture (Diamond

1994; Turner 1995; Lagos 1997, 2001, 2003a, 2003b; Inglehart 1998, 2003; Power and

Clark 2001; Moreno and Mendez 2002). As a consequence of this, one of the main

contemporary developments within scholars dealing with political culture issues has been

the identification and measurement of specific political culture components (e.g., values)

considered as highly significant in regards to the maintenance and development of

democracy. Therefore, some of the most intense debates about the connection between

political culture (i.e., political values and attitudes) and democracy, among scholars

dealing with the political culture approach, have not been related to the shortcomings of

the political culture approach, but instead to which ordinary citizens' political values and

orientations can exert the strongest influence on the stability of a democratic regime.

Inglehart (1988), for example, has sustained that citizens of different societies are

characterized by durable cultural orientations and that cross-national differences reflect

the distinctive historical experience of the respective nationalities. Equally important,

Inglehart has stated that the evolution and persistence of a popular-based democracy

requires the emergence of certain supportive habits and attitudes among the general

public. His own empirical research has aimed at identifying key habits and attitudes: a

high level of interpersonal trust, satisfaction with one's life, and long-term commitment









to democracy or the diffuse feeling that democracy is inherently a good thing. As he

stated, this cultural syndrome of satisfaction and trust is somewhat different from the

definition of civic culture provided by Almond and Verba (1963)-the right balance of

parochial, subject and participant orientations-nevertheless, it is clear that both works

share the same approach.

More recently, Inglehart (2003) has revised some of his previous assertions, but he

fits still closely to what I have defined as the mainstream political culture approach. In

one of his most recent works, he affirms that the level of popular support for democracy,

as measured frequently in public opinion surveys, is not the best predictor for stable

democracy. Rather, democratic stability is better predicted by interpersonal trust,

tolerance to outgroups, post-materialist values, economic success and subjective well-

being. Indeed, these different individual's values and orientations conform to a syndrome

(i.e., "self-expression" values); such that any society that ranks high on one of them tends

to rank high on all them.

Nevertheless, as Turner (1995) contends, it has been also the case that, in the last

years, there has been a great deal of room for improving this mainstream political culture

approach. Accordingly, Muller and Seligson (1994) have argued that in Latin American

countries, and particularly in Central American countries, there is no association between

Inglehart's civic culture syndrome (i.e., interpersonal trust, life satisfaction, and

opposition to revolutionary change) and democracy (i.e., democratic values and

behaviors). They contend that there are three possible explanations for these findings that

contradict Inglehart's contentions about the link between civic culture and democracy.

First, that Inglehart's theory is wrong and that the maintenance of a democratic regime









may have nothing to do with political culture. A democratic regime may emerge and be

sustained by forces entirely removed from ordinary citizens' values, such as elite pacts

and elite consensus, class structure, and level of economic development. Second, political

culture may be significant, but the variables selected by Inglehart (1988, 1990) and

Putnam (1993) may be wrong. Interpersonal trust might have little or no relevance to

attitudes toward democratic institutions and processes. Third, the variables may be

correct, but their operationalization may be flawed. The operationalization of the

interpersonal trust item, a highly North American notion, is especially suspicious,

because cross-cultural validation of the measure has been virtually absent. Nevertheless, I

believe that this seems much more an attempt to try to accommodate the mainstream

political culture approach to the reality of developing countries, rather than an effort to

establish its lack of relevance.

Therefore, it is not hard to find several scholars trying to adapt this political culture

approach to the reality of new democracies, particularly in developing countries. For

example, contending that political attitudes and values can be considered not only as an

independent variable but also as a dependent variable. Similarly, contending that the

relationship among different political attitudes is not necessarily linear, but possibly V-

shaped. Finally, contending that short-term evaluations can be a powerful influence of

key political attitudes and beliefs. Regarding this last point, some scholars have found

that, in the context of Latin American countries, deep dissatisfaction with president's

performance can explain an increase in the acceptance of military governments (Seligson

and Carri6n 2002).









Nevertheless, in most of these cases, the idea is not to question the explanatory

power of the mainstream political culture approach regarding how individuals' political

values and orientations affect political processes. On the contrary, it seems that idea is to

make the case for the need of both further theoretical development within the political

culture approach and the use of different theoretical perspectives to account satisfactorily

for individual's political orientations. Specifically, for example, in trying to explain the

decline of citizen trust on political institutions and political authorities among established

democracies, Dalton (2004) claims the need for using four different theoretical

perspectives: (1) performance of governments, political institutions, and authorities, (2)

political culture and value change, (3) social capital, and (4) media effects.

Summarizing, the mainstream political culture approach is composed by the

following main contentions. First, political culture is concerned with some persistent and

stable components, such as basic political value commitments and beliefs (e.g.,

interpersonal trust, political tolerance, level of personal life satisfaction, etc.). These

attitudes and beliefs are a deep-seated aspect of given populations and are, consequently,

independent of short-term economic, social or political conditions. Second, the

expectation is that it is going to be possible to find consistent and substantial cross-

national and cross-sectional differences regarding these deep-seated political values,

beliefs, and attitudes. Therefore, it is not surprising to find that some societies and some

segments of these societies have different levels of interpersonal trust and political

tolerance. Third, some specific political culture syndromes are particularly linked with

the stability of democracy. Therefore, one of the main contributions of the political

culture approach is to provide a set of political values and beliefs that can be considered









as critical for the maintenance and development of democratic governments. Finally, it is

the case that political culture has only a relative autonomy. In that sense, it is impossible

to rule out the effect that modernization and structural changes have on the content of

political culture. However, social and economic transformations by themselves will not

produce democracy. Unless specific changes occur in the political culture, there is no

guarantee that socio-economic structural changes will nurture a democratic regime.

It is worth noting here, however, that this mainstream political culture approach

does not represent a specific theory of the relationship among individual's values and

attitudes, political behavior, and democratic stability. Rather, it is a more general

theoretical approach to formulate specific theories about individuals' orientations and

political development. In more practical terms, a political culture approach meant the

identification of some individual's psychological and subjective variables that were

critical in testing hypotheses and developing theories for example about the development

and maintenance of democratic regimes. However, it is equally important to recognize

that scholars working within this mainstream political approach always impute an

explanatory power to the psychological and subjective dimensions of politics (Almond

1980:26).

Finally, given that both the research on popular support for democracy and the

political culture approach deal with individuals' attitudinal and behavioral orientations.

And, given that it is very difficult to explain individual's political behaviors without

considering the role of individual's political values, feelings and beliefs (Almond 1994).

It is important now to assess the contributions that the political culture approach provides









to our understanding of popular support for democracy and its impact on

democratization.

Critical Review of the Political Culture Approach

The development of this mainstream political culture approach has not been free of

criticisms. As a consequence of this, the dialogue concerning these different positions has

generated insightful debates, particularly regarding the relationship between political

culture and democratization. One of the first criticisms advanced about the political

culture approach is that it imputes a causal relationship from political culture toward

other political and economic structures (e.g., the level of democratic or socio-economic

development). Nevertheless, it is worth noting here that Almond (1980:29) believes that

this criticism is incorrect because political culture in The Civic Culture was treated as

both an independent and a dependent variable, as causing political structures and as being

caused by them.

In any case, this type of criticism has been part of a broader concern about the main

determinants of the content of political values and attitudes. Accordingly, the relationship

between political culture and socio-economic development has drawn considerable

attention. Specifically, part of the debate has focused on whether it is the country's level

of socio-economic development which basically generates the attitudes and values that

are more favorable for the development of democracy (e.g., tolerance, political efficacy,

higher predisposition toward political participation, civic life, etc.); or, whether the

experience with democracy and alternative regimes generates an independent effect on

the nation's political attitudes and values. However, it is important to mention here that in

both cases the idea is that these political values and attitudes could in turn overpower the

effect of the country's levels of socio-economic development, and the regime's political









and economic performance (Diamond 1999:162). Additionally, it is worth noting here

that this specific debate has been particularly relevant within the discussion of the

existence of some structural prerequisites for the development and consolidation of

democracy (Lipset 1959; Rustow 1970; Gasiorowski 1998; Diamond 1994, 1999).

Regarding these different interpretations of the relationship between political

culture, and political and economic processes and structures, I believe that it is most

useful to assume that the formation of political culture is a complex phenomenon that is

influenced by different social, economic, and political factors. Additionally, I also believe

that these political values, beliefs, and attitudes, once established, have a momentum of

their own, and may act as autonomous influences on political and economic behaviors

and processes. Accordingly, some scholars have made the case that some political values,

beliefs and attitudes can have a critical role in the development and maintenance of

democracy (Lipset 1981; Diamond 1994, 1999)

Highly related to this debate, there has been an important discussion about whether

political culture should be treated as a dependent, or independent variable. Some scholars

have stressed that a political culture approach cannot take these beliefs, attitudes, values,

ideals, sentiments, and evaluations as given. In other words, political culture must not be

treated as an uncaused factor that explains why people behave as they do, yet incapable

of being explained in itself (Thompson, Ellis and Wildavsky 1990:218).

About the previous point, I consider that it is more appropriate to assume that

political culture can be both a dependent and an independent variable. Therefore, a

political culture approach should not only be aimed at addressing the impacts of political

culture on ordinary citizens and elite's political orientations, and regime and government









performances; but also what factors shape the content of this particular political culture.

Predictably, assuming this dual character of political culture (i.e., as dependent and

independent variable) is a challenging enterprise. However, considering the challenge of

understanding the most recent processes of democratization, there is no doubt that it will

be extremely important to address both how a phenomenon like popular support shapes

the current experience of several new democracies, and what is playing a critical role

regarding the development and content of political culture within these societies.

There have been also some criticisms regarding the way some scholars use the

political culture approach. It has been affirmed that rather than assuming that political

culture is about testing single classificatory schemes or measuring some political cultures

a priori defined,6 political culture should be seen as a method for analyzing the basic

beliefs of any given group of people. As a consequence of this, it has been contended that

a political culture model needs to be defined as a complex structure of logically linked

beliefs, norms, and goals; and that scholars dealing with political culture issues need to

departure always from the viewpoints of the persons that they are studying (Lane 1992:

364-65).7

Furthermore, there have been some criticisms about how the dependent and

independent political culture variables have been operationalized.8 Concerns have arisen

about how well this political culture approach is travels from the reality of established


6 Indeed, this was one of the problems with Almond and Verba's work (1963) because they departed from
considering the existence of three main or pure types of political cultures (i.e., parochial, subject, and
participant) that could be identified in almost any political system.

7 Lane (1992) also questions that in the field of political culture, there is too much emphasis on between-
nation differences and much less on within-nation differences. Fortunately, this is something that has been
changed recently.

8 Chapter 6 addresses this issue specifically.









democracies to the reality of new democracies. In this regard, it is revealing how

Inglehart (1988) seems to suggest that a conservative attitude (i.e., individuals thinking

that "our present society must be valiantly defended against all subversive forces") vis-a-

vis both revolutionary (i.e., individuals thinking that "the entire way our society is

organized must be radically changed by a revolutionary action") and reformist ones (i.e.,

individuals thinking that "our society must be gradually improved by reforms") can be

considered always as the most pro-democratic attitude.

Regarding the capacity of the political culture approach to deal with political and

support for democracy, it is the case that other theoretical approaches are also relevant.

For example, it has been affirmed that democratic politics is a social contract whereby

governments perform certain functions in exchange for popular support (Dalton 2004).9

Accordingly, not only the economic performance, but also a more general policy

performance is seem as one of the most important determinants of citizens' political

attitudes and orientations, including support for democracy.

Nevertheless, it is important to mention that the empirical evidence regarding the

impact of government performance on individuals' political attitudes and orientations

reveals that this effect is weak at the aggregate national level, but significant at the

individual level (Clarke, Dutt, and Kornberg 1993; McAllister 1999; Nye and Zelikow

1997; Listhaug 1995; Kornberg and Clarke 1992). More specifically, there is evidence

that objective economic circumstances do not generate clear feelings toward the polity.

On the contrary, assessments of these objective conditions are heavily influenced and



9Easton (1976:436) was one of the first scholars to recognize this situation. According to him, "(...)
leadership assumes the responsibilities for tending to the problems of the societies. In return the leadership
gains power to enable it to make and implement binding decisions, a power that it loses to an alternative set
of leaders if is unable to supply some average level of satisfaction to its supporters."









mediated by psychological judgments (Dalton 2004:65). By the same token, this

empirical evidence suggests that individual's ideological positions and expectations

toward government play also a critical role in shaping support for the government,

political institutions, and the political regime.

This is particularly relevant because seminal works about political culture (e.g.,

Almond and Verba 1963) argued for an evaluational orientation as one of the key

components of political culture. Nevertheless, the large majority of contemporary works

within the mainstream political approach has neglected this political culture dimension,

focusing instead on the cognitive and affective orientations. Similarly, very little attention

has been given to the connection between deep-seated political values and attitudes-on

the one hand-and individual's evaluations of government performance-on the other

hand. It is clear for me that overcoming these limitations will contribute enormously to

our understanding of popular support for democracy.

Interestingly, this is what some scholars are trying to accomplish when they sustain

that individual's definitions and expectations toward democracy and current democratic

regimes factors into the citizenry's satisfaction and support of current democratic regimes

in new democracies. Accordingly, in those cases characterized by a low level of popular

support for democracy among Latin American countries, the problem seems to be a huge

gap between democratic expectations and the overall performance of these current

democratic regimes (Camp 2001, 2003). There is no doubt for me that this way of

approaching the phenomenon of popular support for democracy constitutes a very









plausible explanation for the high levels of political and democratic alienation among the

most disadvantaged citizens in developing countries. 10

Finally, it is important to recognize that some scholars have been very skeptical

about the relationship between the level of popular support for democracy and the level

of societal democracy. The main argument is that overt support for the idea of democracy

is widespread among publics throughout the world. However, at the same time, a large

number of countries-particularly new democracies-are struggling not only with the

quality of their democratic regimes but also with their stability. By the same token,

surveys are showing that among many of these countries "high" levels of popular support

for democracy coexist with other less democratic attitudes (e.g., the popular preference of

strong leaders who does not have to bother with elections and the parliament). One of the

main conclusions of this way of thinking is that what really matters for the development

and stability of democracy are some specific citizens' attitudes, like interpersonal trust

and political tolerance rather than a general or overt support for democratic rule

(Inglehart 2003, Moreno and Mendez 2002).

I will discuss in subsequent chapters how it is possible to challenge this contention

about the existence of a considerable level of popular support for democracy in new

democracies. For now, it is enough to say that this misleading impression has been the

result of the way several scholars have operationalized and measured popular support for

democracy-particularly throughout the use of public opinion surveys.11 This way of


10 See Arag6n (2001) for the specific case of Peru.

1 The most extensive survey question used to measure popular support for democracy has been the
following: "which of the following statements do you agree most? (a) Democracy is preferable to any other
kind of government; (b) in certain situations an authoritarian government can be preferable to a
democratic one; or (c) it does not matter to people like me whether we have a democratic government or a
non-democratic government". This question has been included in all the independent associated with the









operationalizing popular support for democracy presents several problems, beginning

with the expectation that one or two general survey question can be enough to empirically

account for this complex phenomenon. Not surprisingly, several works that relied on this

type of operationalization of popular support for democracy tend to provide too general

and unclear conclusions.12 Just to provide one example about the limitations of this

approach toward popular support, we can consider how in the case of Latin America, and

according to the last findings from the Latinobar6metro between 1996 and 2003, some

politically stable countries (e.g., Chile and Brazil) present levels of both "support for

democracy" and "democratic satisfaction" that are lower in comparison to some other

countries that have suffered democratic reversals (such as Peru and Venezuela).

Hence, one way to respond to the scholars who contend that there is no such

relationship between popular support for democracy and the actual level of

democratization is to contend that the problem is not that popular support for democracy

has a very limited impact on the process of democratization of several developing

countries. On the contrary, that the problem has been the standard, but not less

unsatisfactory, ways that have commonly been used to deal with this phenomenon

empirically. It seems that a comprehensive understanding of popular support for

democracy in dew democracies will require a considerable improvement in its

operationalization and measurement.



Global Barometer (i.e., the Eurobarometer, the Latinobar6metro, the Africanbarometer, and The New
Democracies barometer). Also, an alternative has been to use general questions about the level of
satisfaction with the functioning of democracy in their own country. The World Values Survey phrased this
question in the following manner: "on the whole are you very satisfied, rather satisfied, not very satisfied or
not at all satisfied with the way democracy is developing in our country"; while the Global Barometer
phrased it in a slightly different way: "on the whole are you very satisfied, fairly satisfied, not very satisfied
or not satisfied at all with the way democracy works in (own country)".
12 See Lagos (2003).









I want to finish this section contending that all these debates and criticisms

regarding the mainstream political culture approach do not mean that individual's

political cognitive, affective, and evaluational orientations are of little relevance for our

understanding of both individual's political behavior and broader political developments.

Rather, it means that assuming both the most important insights contain in the political

culture approach (e.g., popular support for democracy as both dependent and independent

factor) and its mains limitations (e.g., avoiding a ethnocentric or normative approach

regarding political culture in developing countries) can constitute a unique opportunity to

improve our understanding of democratization, particularly regarding the experience of

new democracies.

Political Culture and Popular Support for Democracy

As I mentioned before, a political culture approach has been extensively used to

provide insights about the connection between the subjective dimension of politics and

the functioning and maintenance of democratic regimes. To a large extent, this has been

the case because the topic about political support has a long tradition among scholars

dealing with this subjective dimension of politics. Easton (1975), for example, provides a

critical distinction between diffuse and specific support. According to this author, the

former is related to the regime itself and to political principles, while the latter is related

to more concrete evaluations about how the regime is performing. Therefore, it is

possible for some citizens to be dissatisfied with the way democracy works in their

country, but still committed to the principles of democracy. Although, this seems to be

what is happening in several developed countries, it does not seem as true in the context

of developing countries.









More recently, several scholars dealing with political support in established

democracies have identified an important decline in people's trust regarding political

institutions and authorities in the last decades. Nevertheless, the same empirical evidence

shows that this political dissatisfaction is not affecting popular support for democratic

norms and principles (Craig 1993; Dalton 1999, 2004; and Pharr and Putnam 2000). On

the contrary, the situation of several developing countries is much more complicated

because ordinary citizens' political perceptions and attitudes combine-on the one

hand-a high level of dissatisfaction regarding the performance of their current

democratic government (i.e., specific support) with-on the other hand-a somewhat

contradictory orientation regarding the current democratic regime (i.e., diffuse support).

Specifically, a considerable level of ordinary citizens' preference for a democratic regime

exists alongside an almost total lack of support regarding some key democratic

procedures and the institutions that are supposed to implement them (e.g., congress and

political parties). 13

Therefore, regarding the distinction between diffuse and specific support, the

experience of developing countries is making clear two things. First, that it is necessary

to develop a more detailed approach regarding these two broad categories of popular

political support. Accordingly, we need to identify different political objects within each

one of these general categories. Second, at least in the context of developing countries,

one or two decades of persistent low specific support for democratic governments has a

negative impact on the level of diffuse support for the current democratic regimes.


13 In the case of Latin American countries, the Latinobar6metro shows consistently that in the last years
while a clear majority of citizens in this region believe that democracy is preferable to any other form of
government, a similar majority also believes that congress and political parties are not indispensable for the
functioning of a democratic regime (1999-2000, 2001, 2002, 2003, 2004 and 2005).









However, as it will become evident later, the study of popular political support and

popular support for democracy requires much more than Easton's initial distinction

between diffuse and specific support.

Fortunately in the last years, the study of support for democracy in theoretical and

empirical terms has made important advancements. A majority of scholars-confronting

the difficulties in measuring and opertaionalizing popular support for democracy14-have

agreed on the need and convenience of distinguishing at least five different levels or

objects of political support within democratic regimes. One of the main justifications for

this is that citizens do distinguish among these different levels or objects when they are

assessing their democratic political system. In this direction, Dalton (1999, 2004) and

Klingemann (1999) have shown that factor analysis strongly suggests that citizens indeed

make these distinctions, and that it is possible to identify specific trends regarding these

different dimensions of popular support for democracy.

Therefore, the concept of support for democracy has been defined as a multi-

dimensional concept that includes, at least, five levels: (1) support for the political

community; (2) support for the core regime principles, norms, and procedures; (3)

assessment of the regime performance; (4) support for the regime institutions; and (5)

support for the political actors and authorities. 15 These scholars have provided a more

detailed elaboration of Easton's seminal ideal (1965, 1975) that political support can be




14 Dalton (1999) contends that operationalizign and measuring popular support for democracy is always
difficult, especially in a comparative way, because democracy is a complex phenomenon, and it can have
several and different meanings.
15 See Seligson and Carrion (2002), Norris (1999a, 1999b), Klingemann (1999), Dalton (1999), Mishler and
Rose (1999), Holmberg (1999), Fuchs (1999), Rose, Shin and Munro (1999), Newton (1999), Mcallister
(1999), and Miller and Listhaug (1999).









defined as a continuum from evaluations of the immediate actions of government

officials to the identification with the nation-state.

This distinction among different levels or objects of popular support for democracy

is critical in improving our understanding of democracy and democratization in both

established and new democracies. For example, a low level of popular support for

democratic authorities will have a different impact in comparison to a low level of

popular support for some democratic procedures on the stability of a democratic regime.

Indeed, one of the most important challenges of popular support for democracy among

new democracies is the combination of a high level of dissatisfaction with the

performance of authorities and some political institutions with a lack of support for some

key democratic procedures. 16

Equally important have been the advancements regarding the differences in

approaching popular support for democracy between established democracies and new

democracies. Accordingly, it has been affirmed that the underlying dynamics of popular

support for a new democracy are different from those of an established democracy. While

support for the democratic principles, norms, and procedures is not an issue in established

democracies, and tends to change slowly over time in these cases; it tends to be much

more volatile in the case of new democracies. One of the key explanations for the

volatility of popular support for the core democratic principles, norms, and procedures

within new democracies lies in the fact that this kind of popular support for democracy

depends basically on regime and government's performance. More specifically, in the

context of new democracies, the new democratic regime may benefit from a degree of

popular acceptance and approval resulting from the rejection of the former and non-

16 See Chapter 5 and 6 for a discussion about this topic in case of contemporary Peru.









democratic regime. However, in the longer term, support for the new democratic regime

will be mostly performance-based. In other words, in the context of new democracies,

some level of individual and collective good is indispensable to avoid the rapid

disappearance of the initial popular support for the new political regime (Mishler and

Rose 1999).

Furthermore, in trying to assess the dynamic of popular legitimacy or support for

democracy in developing countries, Rose, Mishler, and Haerpfer (1998) have made clear

that the main problem of democratic legitimacy in these cases is the competition between

democratic and non-democratic forms of government. The reason for that is that citizens

in new democracies have experienced both types of regimes. Therefore, the fate of these

new democratic regimes will depend on the competition between democratic and non-

democratic alternatives. Consequently, we need to pay attention not only to how

democratic governments are perceived by ordinary citizens but also to how they are

perceived vis-a-vis some less- or non-democratic alternatives.

Specifically, in regards to the main determinants of popular support for democracy

in the context of new democracies-particularly within developing countries-, it is

critical to pay close attention to a particular set of possible explanations. First, on the role

of popular definitions of and expectations toward democracy, some scholars have argued

that there are key differences between developed and developing countries. According to

Camp (2001, 2003), in developing countries, democracy is not only associated with a set

of political rights that make possible the election of the authorities, but also associated

with the provision of some specific outcomes such as employment, well-being, and a

reduction of social and economic inequality. Accordingly, it has been affirmed that in









these countries, democracy is not perceived as a way to regulate political competition and

participation, but at the same time, people expect democracy to solve all social problems,

especially income inequality (Przeworski et al. 1996).

Second, regarding the current performance of new democratic regimes, some

scholars have suggested that it has been a poor performance one of the main reasons for

the lack of popular support for current democratic regimes and one the main sources of

popular alienation regarding democratic institutions and democratic procedures

(Mainwaring and Hagopian 2006, Hagopian 2006). Additionally, it is important to note

that the problems regarding the performance or outcomes of these new democratic

regimes in developing countries are not only related to the expectations regarding some

very specific socio-economic outcomes (i.e., economic prosperity, economic stability,

employment, etc.). A poor performance in other non-economic critical areas (such as an

effective rule of law, equality before the law, effective mechanisms of political

representation, etc.) can also be considered as problematic for the development of strong

and stable support for democratic procedures and institutions (Foweraker and Krznaric

2002; O'Donnell 2004a, 2004b, and 2004c). Moreover, it seems that the development of

popular support for democracy in these countries will require an overall increase in the

quality of political participation, inclusion, and representation; particularly in regards to

the most disadvantages citizens.

Interestingly, it seems possible to assert that, in many developing countries

experimenting with democracy, there is a huge gap between popular ideas and

expectations toward democracy, and the performance of the new democratic regimes.

This gap, in turn, would explain why democratic procedures and institutions can be very









frequently perceived as ineffective or unnecessary mechanisms of political participation

and representation. Chapter 5 and Appendix B provide a very good example of this

situation.

Third, on the effect of acute levels of economic and social inequality on popular

support for democracy, it has been argued that there is a clear tension between extreme

inequality and democracy on the other hand (Lipset 1959; Beetham1994; Diamond, Linz,

and Lipset 1989; O'Donnell 2004b, 2004c). According to this explanation, living in

poverty-and in some cases in extreme poverty-can have a strong impact on most

disadvantaged citizens' political perceptions, attitudes and behaviors. More specifically,

it is highly likely that for people living in such conditions, the democratic regime itself,

and its formal procedures and institutions, will not be perceived as a necessary or

efficient mechanism to represent their particular interests and needs. Additionally to this,

it is possible to expect that living in poverty does not provide the necessary resources,

opportunities and time to develop political and democratic concerns. Therefore, a general

attitude of indifference toward democracy and politics could be one of the most common

features among poor populations.

Accordingly, the existence of widespread poverty is what explains the development

of clientelistic (Stokes 1995), plebliscitarian (Parodi 1993, Murakami 2000) and/or

neopopulist (Weyland 2006) political orientations among poor citizens within new

democracies in developing countries. In turn, all these patterns of political interaction

between poor citizens and authorities explain why very frequently, particularly in

contexts or crises, a considerable segment of the population of these new democracies are









willing to accept the idea that there is a trade-off between the implementation of some

specific policies and the respect for the democratic procedures.

In conclusion, in dealing with popular support for democracy in new democracies,

it is critical to consider the following issues. First, popular support for democracy or

democratic legitimacy is a complex phenomenon with several and different dimensions;

and each one of these dimensions can have a different impact on the overall level of

support for democracy and current democratic regimes.17 Second, in the context of new

democracies, popular support for democracy requires addressing the competition between

democratic and non-democratic alternatives. Third, particularly in the context of new

democracies in developing countries, it is impossible to identify a set of factors and

variables that seem to be critical determinants of popular support for democracy (e.g.,

people's ideas and expectations toward democracy; individual's assessments of

democratic governments and current democratic regimes; the effect of social and

economic inequality on these ideas, expectations and assessments, etc.). Fourth, thinking

of the possibility of further democratization among new democracies, it is important to

assume that popular support for democracy should not be seen as the final step of a

democratic consolidation transition, but a process itself that can facilitate or hinder a

gradual and constant process of democratization. In others words, what is at stake in

several new democracies is the gradual development of ordinary citizens' attitudes and

orientations that can help to maintain and improve current democratic regimes.

Chapter 6 in this dissertation uses survey data collected in Peru at the end of 2004

as an empirical assessment of these different explanatory variables regarding popular


1 Indeed, popular support for democracy is a complex phenomenon that needs to be address both as an
outcome or dependent variable, as an explanatory or independent variable.






79


support for democracy. The goal is not only to identify which ones of these explanatory

variables exert the strongest and most consistent impact on popular support for

democracy, but also to account for how they interact. Similarly important, Chapter 6

provides an innovative and reliable way of operationalizing some key different

dimensions of popular support for democracy as dependent variables.














CHAPTER 4
POPULAR SUPPORT FOR DEMOCRACY IN COMPARATIVE PERSPECTIVE

Although, it is the case that some of the basic underlying dynamics of popular

support for democracy in new democracies differ from the experience of the most

established democracies. This chapter begins with a consideration of how this topic has

been addressed in the last years in the context of established democracies. As it will

become evident these differences in the dynamic of popular support for democracy do not

make irrelevant a comparative analysis between established and new democracies

because we can learn a great deal in both theoretical and empirical terms from this effort.

It is also the case that among new democracies, the literature on popular support for

democracy within post-communist countries shows a considerable level of sophistication

and advancement. As a consequence of this, I will contend that scholars who are

concerned with popular support for democracy in a region such as Latin America can

learn a great deal from what has been done for Central and Eastern European post-

communist countries; particularly regarding how citizens can attach different meanings

and expectations to the term democracy, and how it is possible to measure and explain

popular support for democracy. Therefore, the second section in this chapter provides a

separate assessment of the literature on popular support for democracy in post-communist

and Latin American countries.









Established Democracies: Popular Support for Democracy and the Critical
Citizen's Thesis

In addressing the relationship between public opinion and support for democracy in

established democracies there are two main contentions that are widely accepted. First of

all, the recognition that in the last decades-starting in the late 1960s-there has been a

considerable decline in public confidence in the main institutions of representative

democracy, in particular, the government, the congress, the judiciary, and political

parties; and in the individuals that are in charge of these institutions (i.e., authorities and

politicians). By the same token, citizens in many established democracies are giving poor

marks to how they governments are functioning. Second, that these trends in public

confidence and public assessments among established democracies are not implying an

erosion of support for the democratic principles and procedures. In that direction, there is

no evidence among established democracies of a crisis regarding the legitimacy of

democracy as the best alternative for political regime.1 In summary, contemporary

publics in established democracies are dissatisfied with the incumbents of office and even

with the political institutions of representative democracy, but these feelings of

dissatisfaction have apparently not yet affected basic support for their democratic regime

(Craig 1993; Miller and Listhaug 1999; Mcallister 1999; Newton 1999; Holmberg 1999;

Klingemann 1999; Fuchs 1999, 1995; Norris 1999a, 1999c; Dalton 1999, 1996; Torcal

2001; Pharr and Putnam 2000).

The combination of these two trends, that seem more than just a transient

phenomena merely linked to distrust of incumbents, have suggested the possibility that a



1 Klingeman (1999) and Dalton (1999) provide empirical evidence about the coexistence of strong and
current commitment for democratic principles and processes, and citizen frustration with how
contemporary democratic systems are or are not working within established democracies.









large part of citizens in established democracies can be defined as "critical citizens" or

"dissatisfied democrats", individuals who value democracy as an ideal yet who remain

dissatisfied with the performance of their political systems, in particular the core

institutions of representative democracy (Dalton 1999; Klingemann 1999; Norris 1999a,

1999c; Torcal 2001). The important point here is that the rise of critical citizens can be

hardly considered as a sign of a crisis of democracy in the most developed countries. On

the contrary, several scholars maintain that the growth of more critical citizens who are

dissatisfied with established authorities and traditional political institutions, who feel that

existing channels for participation fall short of democratic ideals, and who want to

improve and reform the institutional mechanisms of representative democracy can be a

very positive influence for the future development of these democratic regimes.

More specifically, it has been affirmed that the rise of critical citizens has increased

the pressure for several constitutional reforms to make elected governments more

accountable to the public (e.g., the introduction of direct democracy mechanisms as a

way to allow more opportunities to participate in the decision-making process, and the

reform of electoral systems). Therefore, according to this perspective, critical citizens are

pushing for the reform of democratic institutions, the widening citizen involvement in

government, and the establishment of new channels to link citizens and state (Norris

1999c, Shugart and Wattenberg 2001, Dalton 2004).

It is important to note here that the critical citizen's thesis is challenging not only

any contention about the crisis of democracy in developed countries, but also-at least

for now and in the context of established democracies-the idea that a successive

dissatisfaction with the performance of democratic government will lead inevitably to the









erosion of legitimacy of democracy. On the contrary, it seems to confirm the contention

that established democracies posses a reservoir of public support for democracy that can

help democratic regimes to overcome hard times.

Although some authors believe that the rise of critical citizens or dissatisfied

democrats is to some extent a phenomenon that can be found in almost all democratic

regimes-including new democracies-, it is clear that there is a big difference in terms

of its extension and relevance between new and established democracies. Critical citizens

are not a majority or even a considerable part of public opinion in new democracies. By

the same toke, as I will show later, the key difference in support for democracy between

new and established democracies is not only that newer democracies show a remarkably

low level of confidence in democratic institutions vis-a-vis established democracies, but

also that there is evidence of a very strong public dissatisfaction with government's

performance and even some erosion of support for democratic principles and procedures

among new democracies.

Therefore, even when the notion of critical citizens or dissatisfied democrats can

present some limitations in addressing the issue of support for democracy in new

democracies, I believe that a revision of the explanations about the decline of popular

support for democratic institutions and politicians among established democracies can

provide very useful insights regarding the understanding of support for democracy in new

democracies. In this direction, it is important to note that there is not a high level of

consensus regarding the causes and consequences of the rise of critical citizens in

established democracies. Accordingly, it seems that-in the context of established









democracies-support for democratic institutions, authorities, and politicians are shaped

by different factors (Norris 1999b).

Consistently with the empirical evidence showing that the problem with support for

democracy in established democracies operates basically at the level of public support for

institutions and main political actors (i.e., authorities and politicians) (Klingemann 1999,

Dalton 1999), several authors have worked on the explanations of these trends, as well as

in its main consequences for these democratic regimes (Inglehart 1990, 1997, 1999;

Holmberg 1999; Fuchs 1999; Newton 1999; Mcallister 1999; Miller and Listhaug 1999;

Norris 1999b).

Before going any further, it is important to mention that there is no consensus

regarding how to operationalize these outcome and explanatory variables. For example,

in some cases the outcome variable is simply either the level of trust in politicians

(Holmberg 1999) or the level of trust in government (Newton 1999). In other cases, the

outcome variable is operationalize as a combined index of confidence in several

institutions such as the parliament, the civil service, the legal system, the police and the

army (Mcallister 1999, Norris 1999b). Basically the same happens regarding the

operationalization of the explanatory variables. Therefore, it is the case that some caution

is required in generalizing some of these findings for the entire universe of established

democracies.

Regarding the explanations of the decline in public trust in institutions and

politicians, probably one of the most important conclusions is that current people's

evaluation of political and economic government performance appears as one of the main

determinants. Nevertheless, it is also the case that there are other factors that can be









considered as important determinants of the level of trust in institutions and politicians.

Among these other important factors we have people's understandings of democracy and

citizens' expectations toward government (Fuchs 1999, Miller and Listhaug 1999); the

difference between being a "winner" (i.e., supporting the government) or a "loser" (i.e.,

supporting the opposition) in the political system (Newton 1999, Norris 1999b); the

distance between citizens' ideological position-in the ideological left-right dimension-

and the ideological positions of their preferred political party (Holmberg 1999); the time

the democratic regime has been in place and, as a consequence of that, the time it has

played an important role in the political socialization of its citizens (Mcallister 1999);

deep-rooted processes of value change that are gradually transforming citizen's

relationship toward government (Abramson and Inglehart 1995; Inglehart 1977, 1990,

1997, 1999); and even the type of political institutions that are in place (Anderson and

Guillory 1997, Norris 1999b).2

Therefore, it is possible to contend that a comprehensive explanation of the lack of

confidence and support for political institutions and politicians in established

democracies-and eventually in new democracies-will demand the use of different

theories and approaches (i.e., theories about citizens' assessments of government

performance, theories about political culture and political socialization, and institutional

theories). This is even more important once we recognize, for example, that there are





2 It is important to note that in the process of identifying the determinants of public support for democratic
institutions and politicians, some other factors, previously thought as important, have been ruled out, for
example, objective measures of government economic performance measures as the rate of inflation and
unemployment (Mcallister 1999), the country's level of social trust and the participation in voluntary
organizations and community involvement (Newton 1999, Holmberg 1999), and the impact of standard
socio-economic variables (Newton 1999).









important connections between the way citizens' assess their government and regimes,

and their political values.

Equally important, the experience of established democracies reveals that citizens'

assessments of government performance are very complex phenomena. First of all, in

assessing the performance of their government, citizens can use different criteria. They

may use absolute levels of outputs or benefits received from government, or they may

compare current benefits with past benefits, or they may compare the benefits that they

are getting with what they think other citizens are getting, or, even, they may use some

ethical principles. For example, Miller and Listhaug (1999) have found that ethical

standards such as the expectation that government follows procedures that produce

outcomes that neither advantage nor disadvantage particular groups unfairly play also an

important role in the overall assessment of government performance.

In a similar vein, it is critical to recognize, as Norris (1999b) does, the importance

of considering non-economic aspects of government performance such as the provision

and respect of civil liberties and the quality of political representation. Regarding the

relationship between political representation and citizens' assessments of government

performance, there is strong evidence that citizens identified with the party in power are

more likely to trust government and political institutions, while citizens identified with

the parties in the opposition are more likely to express political distrust in government

and political institutions (Anderson and Guillory 1997, Newton 1999, Norris 1999b). In a

similar vein, Holmberg (1999) found that in the case of Scandinavian countries, and

particularly in the case of Sweden, one of the most important explanations for the decline

of public trust in politicians is the decline in the number of people perceiving that they









have the same position-in a left-right ideological continuum-as their preferred party;

and a negative evaluation of their own party's policies.

Second, citizens' assessments of government performance are highly dependent on

political culture issues. For example, Fuchs (1999) contends that in current Germany

coexist two different models of democracy. While Western Germans hold a liberal model

of democracy (i.e., a minimalist or procedural definition of democracy), Eastern Germans

hold a socialists model of democracy (i.e., a model of democracy that adds to the minimal

or procedural dimension supplemental elements such as the constitutional guarantee of

social rights-a comprehensive list of welfare services-and the political realization of

these social rights). The problem is that the current democratic system is consistent with

the liberal model of democracy. Consequently, support for this political regime is

relatively high in West Germany and relatively low in East Germany; and these

differences can be explained basically by different fundamental attitudes toward

democracy. In other words, the political culture in East Germany is not congruent with

the democratic system of the unified Germany.

More specifically, the tension between these two different models of democracies is

that a liberal model of democracy lives from permanent disputation about the goals that

are to be attained through political processes, and that this permanent dissensus requires

to be regulated through a set of rules and procedures legally defined by the constitution

and embodied in the political institutions. These rules and procedures must be accepted

by the citizens is they are to fulfill their regulatory function; or, in other words, the

support of citizens must primarily relate to these rules and procedures that cannot

guarantee any outcome in particular. And this exactly what is not working among citizens









that were socialized, intentionally or unintentionally, in a socialist model of democracy

(Fuchs 1999:142).3

Another example of how political culture plays a role in shaping citizens'

assessments of government performance is Inglehart's contentions about the post-

materialist revolution in established democracies (1990, 1997, 1999). According to him, a

steady process of economic development (i.e., modernization)-implying an important

increase on the levels of individual prosperity and security-leads to the emergence of

new and more demanding standards by which governmental performance is evaluated.

The difficulties of current government to satisfy these demands is what is mainly causing

the lower levels of trust in political institutions and politicians, and a general low level of

satisfaction with the functioning of the political system. Briefly said, this process of

values change is not undermining citizen support for democracy, but making democratic

government more difficult (Dalton 2004).

In conclusion, the scholarly literature about support for democracy in established

democracies is basically addressing the cause and consequences of an important decline

in citizen trust in political institution and the main political actors (i.e., authorities and

politicians). In doing that, it has become evident that one of the most critical factors

shaping support for democracy in these countries is the way citizens are assessing the

performance of their governments. Moreover, another important contribution of all these

works has been several insights about the dynamic of citizens' assessments of



3 I believe that something similar is happening not only with other post-communist countries but also with
other developing countries without a communist legacy. In this direction, there is evidence of a lack of
congruence between this model of liberal democracy and the political culture of large segments of Latin
American citizens (Camp 2001, 2003). Regarding this lack of congruency, an optimistic view would
contend that a continued experience with a liberal democratic regime will make possible the development
of people values and orientation congruent with this type of democratic regime.









government performance. It is not only case that citizens can use different standards and

that they do care for both the economic and non-economic performance, but also that

several other factors play a critical role in the way citizens finally assess the performance

of their governments. In this direction, two factors deserve a special mention.

First, political representation exerts an important effect on how government and

institutions are assessed. The identification with the political party that is currently in

power or the perceptions that there is a political party that can represent individual's

policy preference will have a positive impact in the level of trust in government and

political institutions. Second political culture issues are another important influence for

citizens' evaluations of the performance of their governments and political regimes.

Different expectations toward government and different ideas about the own political

regime will have different effects on the level of political trust in institutions and

politicians. Additionally, these political culture issues can provide a critical contribution

to understand some persistence and long-term difference in popular support for

democracy that are insufficiently explained by short-term changes in governments and

regimes performances.

Popular Support for Democracy in New Democracies

Post-communist Countries

The breakdown of state socialism in the Soviet Union and Eastern Europe between

1989 and 1991, and the subsequent rise of new regimes and states throughout this region

allow us to broaden the discussion of recent processes of democratization. Several studies

of these post-communist cases have addressed questions such as why democracies arise,









and why some of the new democracies succeed while others fail.4 Therefore, important

similarities exist between this literature on post-communist countries and the literature

about the latest processes of democratization in Latin American countries. To a large

extent, these similarities exist because several countries in both regions are trying to

develop functioning and stable political regimes organized in a democratic way.

Nevertheless it is worth nothing that there are also important differences regarding

the processes of democratization in both regions. Regarding the type of regime that was

abandoned in order to move toward democracy, there are important differences between

post-communist and Latin American countries. Typically in the former, the transition

toward democracy meant the collapse of a totalitarian communist regime. In the case of

the latter, the transition toward democracy meant the withdrawal of an exhausted and

divided military from power and a compromise between the military and conservative

groups on the basic rules for a new civilian regime (Geddes 1999; Bunce 2000, 2003).

Similarly, we can expect some important differences in the political experience and

socialization processes between citizens from post-communist and Latin American

countries. For example, citizens from post-communist countries not only lived most of

their lives in a communist society and were socialized in a communist ideology, but also

mostly lacked any previous experience with democracy. However, the majority of Latin

American citizens had experienced non-democratic regimes in the form of military

governments, and had experience with both military and civilian governments.

Also, regarding the transition process itself, Latin American countries moved

toward formal democratic regimes, whereas post-communist countries had a dual


4 See Bunce (2003) for a more detailed description of the main topics within the literature on post-
communist countries.