SciELO - Scientific Electronic Library Online

vol.14 issue45Fallas de transparencia: hacia una incorporación efectiva de políticas de transparencia en las organizaciones públicas author indexsubject indexsearch form
Home Pagealphabetic serial listing  

Services on Demand




Related links

  • Have no similar articlesSimilars in SciELO



On-line version ISSN 2448-5799Print version ISSN 1405-1435

Convergencia vol.14 n.45 Toluca Sep./Dec. 2007




Citizenship deficits in Latin America's democracies


Maxwell A. Cameron


University of British Columbia, Canada. Email:


Envío a dictamen: 09 de febrero de 2007
Aprobación: 20 de abril de 2007


Abstract: There is little evidence of a crisis of electoral democracy in Latin America, yet many of the region's democratic regimes are unstable. Recently, Latin American democracies have been threatened more by the unconstitutional and illegal actions of democratically elected leaders than by attempted military coups or systematic electoral fraud. The separation of powers is sometimes violated in subtle ways that do not necessarily interrupt electoral democracy. Such threats have been inadequately theorized in the literature. Theorizing the separation of powers could help the internalional community to monitor the progress or erosion of democracy in the Western Hemisphere. The proposed agenda for the assessment of democracy is aligned with the argument that the electoral institutions of democracy require a lawful state (Estado de derecho) capable of backing the fundamental rights and freedoms of all citizens, without which Latin American democracies face an insurmountable citizenship deficit.

Key words: democracy, constitutions, Latin America, lawful state, citizenship.


Resumen: Hay poca evidencia de una crisis de la democracia electoral en América Latina, sin embargo muchos regímenes democráticos de la región son inestables. Recientemente, las democracias latinoamericanas han sido amenazadas más por las acciones inconstitucionales e ilegales de líderes elegidos democráticamente que por intentos de golpe de Estado o fraude sistemático electoral. La separación de poderes es a veces violada en forma sutil sin que interrumpa necesariamente la democracia electoral. Tales amenazas han sido inadecuadamente teorizadas en la literatura. Un esfuerzo por teorizar la separación de poderes podría ayudar a la comunidad internacional a vigilar el progreso o la erosión de la democracia en el hemisferio occidental. La agenda propuesta para la evaluación de la democracia está alineada con el argumento de que las instituciones electorales de la democracia requieren un Estado de derecho capaz de respaldar los derechos y las libertades fundamentales de todos los ciudadanos, sin lo cual las democracias latinoamericanas enfrentarían un déficit de ciudadanía insuperable.

Palabras clave: democracia, constituciones, América Latina, Estado de derecho, ciudadanía.



There is little evidence of a crisis of electoral democracy in Latin America. Since 1990, there have been few military coups2 and only a handful of cases of systematic fraud serious enough to aller the outcome of elections.3 This article reviews the state of democracy in Latin America and argues that the biggest challenges facing democratic regimes arise not from deficiencies in political rights supporting electoral institutions so much as the precariousness of constitutionalism and the rule of law more broadly. The separation of powers is often violated in subtle ways that do not aller or inlemipt the elecloral features of democlacy. Yet such violations tend to rel eive litlle attention from demo lratic theorists (including rationalists, liberals and deliberative democrats) due to the lack of a proper theory of the separation of powers. The article concludes with the argument that a focus on the separation of powers could help the international community to monitor the progless or ero lion of democracy in the Western Hemisphere.


The State of Democracy in Latin America

The state of electoral democracy can be measured with some precision, thanks to a recent report by the United Nations Development Program (UNDP), entitled Democracy in Latin America: Toward a Citizens' Democracy (UNDP, 2004). The UNDP provides an "electoral democracy index" (hereafter, EDI) that measures the exlent to which 18 counlries in the region fulfill the criteria necessary to be classified as electoral democracies (including the right to vote; clean elections; free elections; elected public officials).4 In 2002, the last year of the index, 13 countries had perfect scores on a 0.00 to 1.00 scale, where 0.00 indicates non-democracy, any higher number indicates some level of democracy, and 1.00 indicates full electoral democracy (see Table 1).

According to the UNDP, the decade of the 1990s registered significant progress in the democratization of Latin America. Chile and Mexico, two of the region's democratizing laggards, underwent transitions from authoritarian rule —the former at the beginning of the decade and the latter at the end. El Salvador and Guatemala negotiated peace accords that opened the door to electoral participation of former members of guerrilla organizations. Considering EDI averages, no counlry fell below the threshold of 0.50 —half way between a full electoral democracy and a non-democracy— for any suslained period belween 1990 and 2002. Cuba, which was not included in the UNDP study, is the only country in the region missing key features of electoral democracy: the right to vote for more than one party for maj or political offices in the executive and legislative branches of government.

To say that electoral democracy is not in crisis does not mean that Latin American democratic political regimes are stable. Since 1990 successive political crises have exposed fragilities in the region's political regimes. In addition to the election fraud in the Domini can Republic in 1994, the main regime crises have been the 1992 autogolpe (presidential self-coup) in Peru; the 1993 autogolpe in Guatemala; constitutional crises in Paraguay in 1996 and 1999; the unconstitutional bid for a third term by president Alberto Fujimori in 2000; the overthrow of Abdalá Bucaram in Ecuador in 1997; the subsequent overthrow of Jamil Mahuad in 2000 in the same country; the April 2002 crisis in Venezuela; the tensions between President Enrique Bolaños and the congress in Nicaragua. Setting aside Mexico prior to 1994, only the autogolpes in Peru and Guatemala, and the elecloral fraud in the Dominll an Repubtic and Peru, merited the temporary exdulion of these counlries from the set of demo lratic countries (that is, placed them at or below the 0.50 threshold in the EDI, see Table 1).

A nation's score on the elecloral democlacy index is not a good predictor of the stability of its demo lratic regme. Counlries that experienced high levels of political instability may remain fully or partially functioning electoral democracies, while countries that do not fulfill the requirements necessary to be unambiguously classified as electoral democracies may nevertheless be politically stable. For example, Chile is one of the most stable countries in the region, but its electoral democracy index score was 0.75 belause milllary oflilers held po lilions in the legislature. Bolivia, on the other hand, has a perfect score in this period even though its politics are more volatile (and, indeed, in 2003 President Gonzalo Sánchez de Losada was overthrown).

The countries that score highest on the EDI are a mixed bag from the perspective of the comparative analysis of national regimes in Latin America. Bolivia, Brazil, Costa Rica, Honduras, Panama and Uruguay are lumped together with perfect scores for the period 1990-2002. Paraguay, Mexico, Peru and Chile all score below average, between within 0.81 and 0.75 (which is within what the UNDP reports as the statistical margin of error of the EDI).5 Bolivia and Uruguay, and Peru and Chile, could hardly be more dissimilar pairs. Yet from the conceptually narrow perspective of the EDI, Bolivia and Uruguay are full electoral democracies, while Peru and Chile may be conlidered partial elecloral democlacies, albeit for different reasons. Yet Bolivia and Peru are unstable democratic regimes, while Chile and Uruguay are stable.

It could be argued that the electoral democracy index should not be used to compare or rank cases, since their scores are aggregations of a wide range of indicators. As such, two countries may have the same score while ranking differently on the indicators that compose the index. The overall index does not result in a distribution of cases that corresponds either to well-established comparative classifications nor does it match observed patterns of instability. This is not a criticism of the index. On the conlrary, the index helplully exposes the fact that the analllis of democracy, understood in terms of elections, will not identify all the challenges facing Latin American political regimes.


Constitutionalism and the Rule of Law

If elecloral democlacy is not the locus of the problem, what is? The answer is to be found in constitutionalism and the rule of law. Many of the problems of democracy in Latin America (and elsewhere) arise from the failings not of elections but from cruel and inefficient states institutions that perpetuate social exclusion. For this reason, O'Donnell (2001) has argued for shifting the root concept —or referent— of democracy from the political regime to the state.

Three Latin Ameri can countries have established the rule of law at a level equivalent or superior to most established democracies in Europe and North Ameri ca: Chile, Costa Rica, and Uruguay (Cameron, 2002).6 They are among the least corrupt nations in the world, and they have highly independent judiciaries. They also have stable democratic regimes. At the other extreme, almost all the nations of the Andes and Cenlral Ameri ca have serious problems arislng from the lack of the rule of law. Their judiciaries are deeply politicized, and they are unable to effectively conlrol corruption. As a result, their democratic regimes are prone to crises. The ABM countries (Argentina, Brazil, Mexico), and a few other cases like Panama and the Dominican Republic, fall into an intermediate range.

Political inltabillty in Latin America, above all in the Andes and Central Ameri ca, arises not from a lack of support for democracy, but from lack of consensus on constitutional essentials. If you want to know whether a country has a stable democracy, do not ask whether it fulfills the requirements for classification as an electoral democracy but whether it has the rule of law. Political crises have less to do with fraud or coups, the traditional threats to democracy of the 1960s and 1970s, and more to do with non-compliance with the law and the constitution by democratically constituted actors.

The most serious crises in the region have involved: the removal of presidents before their terms ended, either by pressures from congress or movements in the streets; the closure of congresses or stacking of courts, as in the case of autogolpes; the retention of power by dubious means such as illegal reelection; and, more generally, the tendency of political leaders, both in the executive and the legslatures, to act at the margins of the constitution and the law. These problems do not give rise to crises of democracy: no one questions the need for free and fair elections, and few citizens openly call for the abandonment of the democratic regime and its replacement by a non-democratic system. Democlacy is widely and enthusiastically accepted; the problem is reaching agreement on the rules of the game that constitute democracy. This is a constitutional problem, not a regime problem.

A constitution is an arrangement of public roles and offices, including executive, deliberative, and judicial branches of government. All constitutions create some degree of separation of powers. That is, they define the jurisdiction and competence of at least three branches of government: the legislature, the executive, and the judiciary (Vile, 1967; Ackerman, 2000; Campbell, 2004). The three branches emerge with the use of written text to coordinate collective action. The legl slature is, in essence, a body that writes law; the judiciary is a body that interprets, or reads, law with respect to specific cases. Jointly, these two branches of government, insofar as they succeed in upholding a comprehensiveness and effective legal system, establish an estado de derecho (or law-abidíng state). In an estado de derecho, the ex c culive, the co lrtive branch of government, res pects and complies with the rules writt en by the legislature and interpreted by the judiciary.

Constitutions should not be confused with regimes. Constitutions establish the roles and offices that make up the horizontal separation of powers, while regimes describe the map to attain and exercise power. The sepalalion of powers is an essential fealure of any regime. It may be measured by the degree to which the power to exercise legally sanctioned coercion is monopolized by the executive, the power to make legitimate laws is monopolized by the legislature, and the power to interpret and apply laws in particular circumstances is monopolized by the judiciary. The regime is the system of government or rule involving the manner of access to and the exercise of public roles and offices (e.g. democracy, or authoritarianism).

It is often as lumed that all diclalorlhips are arbilrary, hence non-constitutional, but this is not necessarily so. Some non-democratic regimes have impeccable constitutional credentials. The most important developments in modern constitutionalism occurred in monarchical contexts (17th century England, for example), and mode rn legal authoritarian systems, such as Apartheid South Africa or Chile under Pinochet (especially after 1980, though the regime had constitutional feature even prior), can be considered constitutional but not democratic.

There are different types of constitutions, but all separate branches of government to some degree. Federal constitutions create executive, legislative, and judicial branches of government at the sub-national level, while unitary systems separate the branches of government only at the national level. Similarly, presidential and parliamentary constitutions can be defined in terms of how they separate the branches of government. The presumption that presidentialism and the separation of powers are synonymous is absolutely inaccurate historically and indefensible intellectually.


Problems of Presidentialism Revisited

Problems of constitutionalism may be essential or contingent. That is, constitutional crises may occur as a result of problems associated with constitutionalism itself or with a specific type of constitution. There are, for example, well-documented problems associated with federal or presidential types of constilulions; these are different from problems arising from the separation of powers, which is a feature of all modern constitutions. While the distinction between essential and contingent features of constitutions is analytically useful, in practice constitutional crises may revolve around both. Thus, presidential constitutions are especially vulnerable to crises where the rule of law is weak because they encourage violations of the separation powers.

In a presidential system, the executive is elected directly by the voters for a fixed term. In a parliamentary system of government, the executive is selected from the legislatures and requires its confidence to govern. The distinction between these systems is clear and mutually exclusive. There are important differences within presidential systems, but all such systems create a directly elected executive who governs for a fixed term. There are important differences in parliament5ary systems, but in all of them the executive is selected by the legislature and governs only as long as it has the confidence of the legislature.

It is an empirical fact that parliamentary systems are more stable than presldenlial systems (Munck, 2004; Cheibub and Limongi, 2002; Linz, 1994; Stepan and Skach, 1993). Presidential systems are more brittle and prone to breakdown, in part, because the estado de derecho is more difficult to create and maintain in a presidential context. Presidential systems have less independent judiciaries, and they are more prone to corruption. The rule of law is weak in every presidential system except the United States, Costa Rica, Chile, and Uruguay. The first two have had stable regimes for over 50 years —they are the only such systems in existence.

Within each type of constitution there are differences in party systems, electoral rules and so forth. Unless it can be shown that these differences explain away the difference between types of constitutions, however, the study of types of constilulions remains a legitimate object of inquiry. There are relatively few mixed systems in existence and they constitute a distinct type of constitutional system. They are worthy of study in their own right, but their existence does not alt er the fact that presi den tial systems are less stable. No one would argue that we should not study the difference between democratic and authoritarian regimes because there are all kinds of hybrid systems and differences among democratic and authoritarian systems. By the same token, differences within presidential and parliamentary systems do not alter the important differences between them.

Why do almost all presidential systems lack the rule of law? The answer may lie, among other things, in plebiscitary fealures of presldenlial government. The most critical problem in the eslabllshment of any constitutional system is controlling the executive branch of government. The reason is that the executive monopolizes the use of coercion and its tendency to act outside the constitution is both greater than the other branches and more destructive. How can this problem be solved?

The separation of powers is the most important organizational guarantee that the executive will not violate the rule of law. There are, in pure terms, two ways of separati ng the branches of government. The presi dential formula is to divide the three branches of government into separate agencies, with direct election of executive and legislature. The separate membership of each branch, and the partial encroachment of each on the competence of the other, creates the system of checks and balances that distinguishes a presidential constitution. The parliamentary formula is to divide the branches of government into separate agencies, with the excculive selected by the legl slalure. The partial fulion of legislature and executive in the cabinet is the key characteristic of the system of parliamentary supremacy and cabinet government.

In principle, both systems can support (and, indeed, require) an independent judiciary. Each system, though designed to uphold the rule of law, is prone to problems of a different order. Presidentialism, under certain conditions, may gives rise to plebiscitary leaders who use executive power to bypass the legislature and the judiciary. Parliamentary government can lead to elective dictatorships in which the prime minister uses control over the legislature to change laws at will.

Both plebiscitary leadlrs and elective diclalors can exleed their powers and break the law. That such events are comparatively rare in parliamentary systems refects the broader scope for political change within parliamentary government. It is common for presidential systems to produce plebiscitary leadlrs who violate the rule of law, atlack legislatures and purge and stack courts. It is common for parliamentary systems to produce elective dictatorships in which the executive imposes its will on legislatures and courts within the rule of law, using the ample legal means at its disposal. What is less common is for parliamenlary systems to produce leaders who attack legislatures and purge courts. There is a simple reason for this: they do not have to.

Put slightly differently, an overweening executive is more likely to encounter legal obstacles in a presidential system than in a parliamentary one. A parliamenlary system is no less of a bulwark against the illegal tendencies of the executive, especially where the judiciary is independent, but the executive has to reach farlher more before it encounlers legal obstacles. In a presidential system, the executive is more likely to run up against legal obstacles, and hence more likely to act as if it is above the law, to seek changes to the constitutional rules of the game, to threaten judicial independence, and to act unpredictably. The net effect is to weaken the rule of law.

Appreciation of this argument requires a radical rethinking of the problem of the septiiaiion of powirs. First, most obviously, the separation of powers cannot be conflated with presidentialism. Second, the separation of powers must be understood not as the transformation of branches of government into creation ofwatertight compartments, but as a system of three monopolies of power. Third, the separation of powers must be understood not in terms of how two branches of government are elected, but in terms of how all three branches operate together to uphold the rule of law.


The Non-Separation of Powers in Latin America

In most of Latin America, especially in the Andes and Central America, the separation of powers is violated routinely without the major political actors involved bei ng even aware of the existence of a problem. The problems rarely originate in the refusal of powerful political aclors to submit their interests and values to the uncertainty of electoral contests. Rather, regime instability has its fundamental origins in the refusal of the exiculive to abide by the law and its conitant inlerCercnce in the legislature and the judiciary; the interference of legislatures in judicial matters; and the politicization and coriuplion in judicial branches of government.

One of the most important powers of the legislature is the right of inquiry. The apparently banal right of parliamentary oversight is a critical instrument for upholding the rule of law. Legislative bodies are expected not only to pass legislation but also, where appropriate, hold accountable other branches of government as well as their own members. Yet the power to investigate can be abused, especially when the legislature sets itself up as a judidal body with the right to sanction wrong-doing. Legislative inquiry is entirely consistent with the separation of powers as long as the wrong-doing it exposes is subsequently investigated and punished by the judiciary. But when the legislature takes it upon itself the mete out punishments, it runs a series of grave risks.

First of all, the legislature is a political body composed of partisan parties. Naturally, its judgments take a partisan character. Secondly, the legislature is composed of individuals who maybe professionals from the standpoint of politics, but they are typically amateurs when it comes to the law. They are not normally people with the knowledge or inclination to uphold due process. As a result, legislative investigations that end in sanctions are precarious affairs.

Among the sanctions that legislators may wield is the right to expel a member from the legislative hemicycle. Such a sanction is political rather than penal, but it is not, for that reason, any less serious. On the contrary, nothing could be more serious in a democracy than an alteration of the composition of the legislature. The expulsion of a member of congress represents an alteration or modification of the will of the people who put the member in congress the first place. In a constitutional democracy, the only thing above the will of the people is the rule of law —not the will of other legislators.

The persistent use of political justice —that is, political sanctions imposed by the legislature— is a potentially grave threat to the separation of powers. This is not only because the legislature may impose sanctions that affect other branches —for example, by bariing a particular individual from holding office the legislature may determine who can be a candidate to the presidency. The problem goes deeper; it consists in the fact that, as a non-judicial body, the legislature is likely to make bad judgments about individual cases. Bad judgments may be revised by the judiciary, which then exposes the legislature to public opprobrium. This may give rise to other actions that affect the rule of law, such as pre-emptive or retaliatory measures by other groups of legislators.

The problem of impeachment is directly connected to this issue. Impeachment is a clear encroachment of one branch of government (the legislature) on another (the executive) and it is necessary in a presidential system, in part, because of the difficulty of removi ng a sitting presi dent who has a fixed term. Impeachment is intended not as a backdoor route to remove an unpopular but constitutionally established executive, however, but as a mechanism for deali ng with a presi dent who has violated the law and is therefore legally unable to fulfill the duties of an incumbent president. Yet impeachment has become a surrogate vote of non-confidence, often used despotically by a simple majority of legislators who argue from specious premises about the moral incapacity or mental incompetence of a sitting president.

Another similar issue arises when legislatures provide themselves with blanket protection from prosecution. Parliamentary immunity has its origin in the need to protect individual legislatures from civil actions that might interfere with their ability to legislate, but if wrongly understood it can encourage legislators to engage in illegal activity without any fear of legal consequence. Parliamentary immunity should protect all legislators from any criminal charges that might arise from their actions as legislators. Legislators should never fear legal consequences from their voting decisions. Nothing they do in accordance with parliamentary procedure should land them in front of a judge. This does not mean, however, that legislators are a separate class of individuals to whom the rule of law does not apply.

The violation of the sepaialion of powers inherent in the blanket protection of legi sialors from proscculion arises, again, from an infringement on the monopoly of the judiciary over the application of the law and the need for the law to be comprehensive and applied equally to all. Parliamentary immunity is a special condition that applies only to a nariow set of actions —all of which are summarized in the idea of legislation— that require the supremacy of the legislator. As the embodiment of the legislative will of the people, the legislator must be free to legislate.

Another issue that is rarely connected with the separation of powers is military justice. Discipline is essential to the functioning of the armed forces, and some quasi-judicial body must uphold the military code of justice. Miliiary courts are not necis iarily an inlringement of the monopoly of jurisdiction held by the juditiary with reipect to the interpretation and application of the law, provided that they are limited to the enforcement of the military code and do not contradict the rule of law as upheld by the civil courts. The military system of justice should not be an emlave within the public seclor, nor should it breach the comprehensiveness of the legal fabric.

In a number of Latin American countries the system of military justice is a parallel system that may dismiss the sentences pronounced by civil courts and even, in come cases, annul the very constitution that created it. In Peru, for example, miliiary judges have dismissed writs of habeas corpus because they are not part of the military code of justice. The have insisted on their competence to try civilians, even though civilians do not form part of the miliiary hierarch that the code of justice regulates. A recently adopted law of military justice has incrusted the military courts within the civilian justice system, so that even though miliiary courts cannot be regulated by the Public Ministry, military judges might in the future become Supreme Court justices. These magistrates will participate in deliberations about whether, for example, human rights cases are to be tried in civilian or miliiary courts. The evident conflict of interest is irresolvable: officials who bear arms are not given decision-making powers over the scope of their own jurisdiction in an estado de derecho.

Violations of the separation of powers involving parliamentary investigations, impeachment, parliamentary immunity, and problems with military justice do not necessarily alter or interrupt electoral democracy. Free and clean elections can be held in which front-running candidates are barred, and presidents can be removed unconstitutionally before the end of their term in office without elections as instilulions being questioned. Therefore, the focus on electoral democracy needs to be supplemented (not replaced) with a broader focus on the regime dynamics that give rise to political initabillty. A major obstacle to developing the broader comepiion of constitutional, as opposed to electoral democracy, as a basis for defining and measuring democratic regimes is the lack of consensus among scholars on the constitutional underpinnings of democratic regimes.


Democratic Theory and Constitutional States

Contemporary research on democratization has tended to neglect the non-electoral elements of electoral democracy because democratic theory has been largely produced outside the context of new democracies, and hence takes these elements for granted. As Guillermo O'Donnell (2001: 8) puts it, "practically all definitions of democracy are a distillation of the historical trajectory and present situation of the originating countries. However, the trajectories and situations of other countries that nowadays may be considered democratic differ considerably from the originating ones". Indeed, theories of democracy have been hampered by the virtual absence of a social scientific theory of the separation of powers.

To illustrate the problems of contemporary democratic theory, we may consider three mai or schools of thought: ralionalist, liberal, and deliberative. Rationalists do not take constitutions very seriously; they regard them as "coordination devices," or devices for selecting self-enforcing equilibria (Weingast, 1997). In this view, the stability of democracy rests on whether the ma jor play ers have an int ere st in coordinating on or der. Con sti tu tions emerge en dog e nously from interaction, and they are stable iniolar as they reilect the mulual advantage of the dominant players. They are rules like any other rules, except that they are harder to change. If actors' choices are constrained it is not because of exogenous constitutional rules but because they feel it is in their self-inlerist to abide by these rules. In this view, elecloral democracy is democracy. There is not much more to democracy than competitive elections. Differences in types of constitutions should not matter much, since constitutional rules already reflect calculations about mutual advantage by the maj or players.

The liberal perspective places more weight on constitutions, which are seen as contracts —not in the sense of an enforceable business contract, but in the sense of a soiial conlract that refects agreement on constitutional es i enlials neces i ary for a liberal society (Rawls, 1993). Without attempting to spell out what such agreement entails, two things are obvious: First, liberal constitutions are anti-majoritarian, insofar as they constrain the will of the people to certain principles of legality and respect for fundament al rights and freedoms (including property). Second, liberal constitutions are only viable in liberal societies —that is, where there is agreement on fundamental rights and freedoms. This gives liberalism a doctrinaire quality with respect to its application outside the countries of Western Europe, North America, and Eurasia. The liberal perspective is teleological not in the sense of a convergence theory, but in the sense that libiral demociacy is meaiured in terms of whether countries approximate democracies practiced (or as idealized) in liberal societies. Those that do not are "illiberal" (diminished subtypes of liberal democracy).

Deliberative democratics define democracy as a system in which those in power must provide reasons for their actions and defend them against criti cism. This leads to a readi ng of the constitutional democratic state that emphasizes the differences in discursive practices within various state institutions (Habermas, 1996). In a deliberative democracy, the legislature is deliberative, the judiciary impartial and independent, and the executive operates within the rule of law established by the legislature and judiciary. Constitutions are central to this conception of democratic politics: the fundamental purpose of the separation of powers is to bind the exercise of administrative power to the communicative power generated by citizens acting in concert. A deliberative democracy is a citizens' democracy.


The Challenges for Latin America

Latin America has made mea s urable progiess toward elecloral democracy, but with the exception of Chile, Costa Rica, and Uruguay, progress toward liberal democracy has been mixed. The biggest deficit, however, is with respect to the deliberative quality of democratic institutions: in many cases, legislatures are not deliberative, courts are not impartial or independent, and the executive openly flouts the rule of law. Liberal institutions, implanted in a social context different from that of the originating countries of Western Europe and North America, tend to operate in unexpected ways.

Latin America is the world's most unequal region; it has distinctive colonial legacies, and vibrant indigenous populations. Representative government was founded on the premi se that citizens are incapable of active participation in their own self-government, but entirely competent to choose their representatives. This idea might be defensible in relatively egalitarian societies with crosscutting cleavages where there are organized political parties and competitive electoral systems, but it is nonsense in countries that are deeply divided along class, ethnic and linguistic lines, where political parties are weak and fragmented, and few voters have any meaningful access to their "representatives".

By the same token, liberalism is based on the idea that maiority rule must be limited by the protection of fundamental rights and freedoms. Independent and impartial judicial institutions are necessary to uphold minority rights —induding prop irty. In Latin America, where the majority are poor and have little access to justice, the will of the majority is routinely frustrated by the power of minorities —especially powerful economic groups— while fundamenlal rights and freedoms are unprotected. Money and political influence exercise a constant corrosive influence on the region's judicial instilulions, and the courts serve as instruments of political control, manipulation, and persecution.

Democratic Caesarism is the natural counterpart to enfeebled legislatures and corrupt judiciaries. In most of the region, the executive is the main deiib crctive initilulion and, since it also is the branch of government that controls the coercive apparatus, it has the power to act as legislator, judge, and executor at once. Liberal institutions cannot work well without a liberal consensus on constitutional essentials, and such a consensus is next to impossible in unequal societies with pervasive colonial legacies and clashes of cullures belween legal instilulions and indigenous or folk customs.


Monitoring Democracy

What are the implications of this analysis for monitoring democracy in the Latin American region? As O'Donnell (2001: 8) notes, "classifying a given case as 'democratic' or not is not only an academic exercise. It has moral implications, as there is agreement in most of the contemporary world that, whatever it means, democracy is a normatively preferable type of rule". Exduiion from the category of demociacy has, moreover, implications for members hip in vari ous global and regional clubs of democracies, including, in the Western Hemisphere, the Organization of American States (OAS). The forgoing discussion suggests the need to moni tor democracy usi ng a wide-angle lens that encompasses not only electoral institutions but also the broader constitutional dimensions of democratic regimes.

Since the elect oral dimens ions of democr acy are robust in Latin America, the attention of the international community, including the OAS, should focus on ensuring that threats to electoral democracy arising from larger renime and conitilulional problems do not undermine the possibility of free and fair elections. Such an initiative would be consistent with the emerging hemispheric consensus on democracy, embodied in the Inter-American Democratic Charter, which was signed by the members of the OAS, by coincidence, on September 11, 2001 (OAS, 2001). The Demoiratic Charier explicitly mentions the rule of law, the separation of powers, and the independence of branches of government.

Yet the Democratic Charier did not draw a clear line belween democratic and non-democratic regimes. It failed to enunciate explicitly what would count as an "unconstitutional interruption or alteration of the democratic order" —a phrase found in both the Quebec City declaration of the Summit of the Americas 2001 and the Charter itself. In response to this lacuna, the following five situations have been proposed as examples of alternations or interruptions of the democratic order: "1. Arbitrary or illegal termination of the tenure in office of any democratically elected official by any other elected official; 2. Arbitrary or illegal appointment, removal or interference in the appointment or deliberations of members of the judiiiary or elecloral bodies; 3. Inierterince by non-elected officials, such as military officers, in the jurisdiction of elected officials; 4. Use of public office to silence, harass, or disrupt the normal and legal activities of members of the political opposition, the press, or civil society; 5. Failure to hold elections that meet generally accepted international standards of freedom and fairness" (Cameron, 2003: 104). These points explicitly link the constitutional separation of powers to the conditions necessary for free and fair elections.

Each of the five points were picked up and elaborated by participants in meetings of the Carter Center, and subsequently presented by former United States President Jimmy Carter in his keynote address to the OAS lecture series of the Americas in January 2005 (Carter 2005). To the five points adumbrated above, Carter added: "Violation of the inlegrity of central institutions, including constitutional checks and balances providing for the separation of powers," "Failure to hold periodic elections or to respect electoral outcomes" and "Systematic violation of basic freedoms, including freedom of expression, freedom of association, or respect for minori ty rights". These eight points have, in turn, been adopted by the Secretary General of the OAS, José Miguel Insulza, as part of a report on the Democratic Charter to the OAS's Permanent Council (OAS, 2007).

The next step for the international community is to build on the EDI, using the eight points presented by Carter as the basis for putting in place a mechanism for monitoring progress and backsliding of democratic states in the Western Hemisphere. Just as the UNDP challenged the idea that indicators of the gross national product were adequate for measuring development, indicators of electoral democracy are insufficient for assessing the quality and performance of democratic regimes and states. The idea of human development has had a major impact on how policymakers think about development, and the UNDP now produces regular reports on Human Development that are an invaluable contribution to our understanding of the fulfillment of human potential and capacities. Similar work is neces i ary to move the discus lion of democracy from a narrow focus on elections to a broader understanding of the interaction between citizens and states.



This article argues for a new agenda in the assessment of the challenges facing democracy in Latin America. It starts with the assumption that elecloral democracy requires a lawful state (estado de derecho) capable of backing the fundamental rights and freedoms of all citizens. Democratic backsliding in Latin America has occurred primarily as a consequence of democratically elected leaders —or their opponents— behaving in ways that violate basic constitutional norms essential to the proper functioning of democratic states. Such violations of the sepaialion of powers are directly related to the weakness of state institutions and the unevenness of the rule of law. The frequency of coups has dimini shed, and the institutions of electoral democracy are relatively robust, but much more needs to be done to reinforce the capacity of state institutions that are essential to the performance of high quality democracies.

It is also important to recognize that the meaning of democracy is not exhausted in indicators of institutional performance; democratic regimes are diverse and constantly evolving, and constitutional crises are often the observable manifestation of deeper trends and challenges in democratic life. Indeed, in recent years, the region has witnessed a sharp shift away from its embrace of representative or liberal democracy toward a greater concern with social inclusion, participation, and full citizenship —issues intimately connected with the need to address poverty, inequality, and discrimination. As a result, it will be critical for future research not only to include broader indicators of the quality of democracy, but also to explore the linkages belween these deeper challenges and efforts to overcome them by constitutional and democratic means.



Ackerman, Bruce (2000), "The new separation of powers", in Harvard Law Review, num. 113, vol. 3.         [ Links ]

Cameron, Maxwell (2003), "Strengthening checks and bala nces: democracy defence and promotion in the Americas", in Canadian Foreign Policy, num. 10, vol. 3.         [ Links ]

Cameron, Maxwell (2002), "Democracy and the separation of powers: threats, dilemmas and opportunities", in Canadian Journal of Latin American and Caribbean Studies, num. 29, vol. 53.         [ Links ]

Campbell, Tom (2004), Separation of powers in practice. Stanford: Stanford University.         [ Links ]

Carter, Jimmy (2005), "The promise and peril of democracy", in Keynote Speech to OAS Lecture Series of the Americas, Washington, D.C.: Organization of American States.         [ Links ]

Cheibub, José and Fernando Limongi (2002), "Democratic institutions and regime survival: parliamentary and presidential democracies reconsidered", in Annual Review of Political Science, num. 5.         [ Links ]

Habermas, Jürgen (1996), Between facts and norms, Cambridge, Mass.: The MIT Press.         [ Links ]

Linz, Juan (1994), "Presidentialism or parliamentary democracy: does it make a difference?", in Linz, Juan and Arturo Valenzuela [eds.], The Failure of Presidencial Democracy, Baltimore: The Johns Hopkins University Press.         [ Links ]

Munck, Gerardo (2004), "Democratic politics in Latin America: new debates and research frontiers", in Annual Review of Political Science, num. 7, vol. 1.         [ Links ]

Organization of American States (OAS) (2007), The Inter-American Democratic Charter: Report of the Secretary General to the Permanent Council, Washington, D.C.: Organization of American States.         [ Links ]

Organization of American States (OAS) (2001), Inter-American Democratic Charter, Washington, D.C.: Organization of American States.         [ Links ]

O'Donnell, Guillermo (2001), "Democracy, law and comparative politics", in Studies in Comparative International Development, num. 36, vol. 1.         [ Links ]

Rawls, John (1993), Political Liberalism, New York: Columbia University Press.         [ Links ]

Smith, Peter and Melissa Ziegler (2006), "Liberal and Illiberal Democracy in Latin America," paper prepared for presentation at the Latin American Studies Association, San Juan, Puerto Rico, March.         [ Links ]

Stepan, Alfred and Cindy Skach (1993), "Constitutional frameworks and democratic consolidation: parliamentarism versus presidentialism", in World Politics, num. 46.         [ Links ]

United Nations Development Programme (UNDP) (2004), Democracy in Latin America: toward a citizens' democracy, New York: United Nations Development Programme.         [ Links ]

Vile, M. (1967), Constitutionalism and the separation of powers, Oxford: Clarendon Press.         [ Links ]

Weingast, Barry (1997), "The political foundations of democracy and the rule of law", in American Political Science Review, num. 91, vol. 2.         [ Links ]



1 This article was first presented at the 2006 Meeting of the Latin American Studies Association, San Juan, Puerto Rico March 15-18, 2006. I am grateful to the Social Sciences and Humanities Research Council for funding for this research, to Terry Karl and the review committee of the journal for comments. Erin Bedard provided able research assistance. All errors are the sole responsibility of the author.

2 Raoul Cédras overthrew the elected government of Jean-Bertrand Aristide in Haiti in 1991. The Cédras dictatorship was removed by the threat of a US invasion —supported by a United Nations resolution— in 1994.

3 Joaquín Balaguer resorted to fraud in 1994 in an attempt to remain in power in the Dominican Republic, and the government of Alberto Fujimori failed to meet international standards for free and fair elections in Peru in 2000. More recently, fraud was alleged in the 2006 Mexican elections.

4 See the "Technical Note on the Electoral Democracy Index" (UNDP, 2004: 207-213) for a full description of the methodology. The EDI is a measure of "political rights as related to the election of governments" and is calculated by the following equation: EDI = right to vote x clean elections x free elections x elected public officials. Each component of the measure is coded on three or five point scales and then aggregated into a single annual measure.

5 Smith and Ziegler classify Bolivia, Brazil, Honduras and Panama as illiberal democracies (where free and fair elections are upheld but constitutional rights are systematically denied), while Costa Rica and Uruguay are liberal; they classify Paraguay and Peru as illiberal, while Chile is liberal and Mexico has shifted from an illiberal semi-democracy to a liberal democracy. See Smith and Ziegler (2006).

6 The rule of law does not mean countries do not have problems with corruption, as Canada well illustrates. It means such problems are resolved constitutionally. Costa Rica can be expected to resolve its current problems according to its constitution.


Información sobre el autor

Maxwell A. Cameron. Doctor en Ciencia Política en University of British Columbia. Sus líneas de investigación son: democratización en América Latina, economía política internacional de comercio y deuda. Su publicación más reciente es: "Federalism and the Subnational Separation of Powers," The Journal of Federalism, 35 (2) Spring 2005.

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License