The concept of coup d’état under stifling stress: Legal theory’s political-science–based response

Abstract

The resignation of the Bolivian President Morales in November 2019 has sparked an international polemic over whether a coup d’état took place. This article concludes that the assumption of a coup d’état would be misleading. The problems faced by the Bolivian case stem not only from the lax manner in which scholars have gathered facts, but also from the fact that the concept of coup may have reached its limits. Building on a legal approach, this article proposes a new conceptual strategy that overcomes both the under-theorization of the (partial) legal order variable used to define a coup and the underrating of this variable. Introducing this variable as the target during an anomalous seizure of executive power means moving from a threefold to a fourfold conceptual scheme of definition (victim, perpetrator, tactic, and target). A coup must not only be illegal but must also break the partial legal order. As a result, the notion of regime restoration is proposed for cases in which the partial legal order underlying the regime has already been broken by the overthrown ruler well prior to the day of their anomalous exit.

1. Introduction

On the afternoon of November 10, 2019, Bolivian President Morales resigned after more than a decade’s rule and sought asylum in Mexico the next day. Immediately following this event, a worldwide debate was sparked across many disciplines as to whether this interruption of mandate was a coup d’état. In an account of these events, Levitsky and Murillo summarized the implications of this debate, concluding that “if foreign governments choose sides in the region’s conflicts, tolerating coups that favor their ideological allies rather than consistently defending democracy, it will encourage a return to the violence and instability that Latin Americans struggled so hard to end.”¹ Despite these facile admonitions, however, we should be able to assess the situation in Bolivia without fearing being portrayed as indirect supporters of violence in Latin America if we were to conclude that no coup took place. Before we systematically address the implications of this event, we should pause a moment and briefly consider the Bolivian crisis itself.

Morales’s resignation marks a deep rupture. It should be recalled that after assuming presidency in 2006, Morales hailed the new Constitution of 2009, which he had sponsored, as the foundation of a new state replacing the old republic installed in 1825. A few years later, Morales announced that his political movement would last 500 years, as did the rule of the previous elites.² But after an unusually long term of office and landslide electoral victories, Morales’s aspirations came to an end. His resignation was preceded by twenty-one days of massive protests in which various social groups, from urban middle classes to popular sectors,³ demonstrated in anger over the sudden interruption of the transmission of the Preliminary Electoral Results on election day October 20, 2019. The suspected fraud only added to a long-standing series of crises in many domains that had eroded Morales’s legitimacy. While his strained relations with the indigenous population in the lowlands,⁴ the decline of commodity prices,⁵ his arbitrary rule,⁶ and the accusations of corruption made up a huge part of this legitimacy crisis,⁷ the pinnacle of voter frustration was Morales’s refusal to concede he had lost a referendum in 2016, which he himself had called in order to amend a constitutional clause that barred more than one continuous presidential term.⁸ Viewed from this medium-term perspective, the alleged fraud in 2019 that precipitated Morales’s exit was a secondary event. A citizen uprising ensued after alternative mechanisms to hold Morales accountable had failed or were thwarted by Morales himself.⁹ Unlike street protests that challenge policy and eventually lead to a presidential interruption,¹⁰ the interruption in this case was preceded by citizens demonstrating for compliance with rules. Eventually, Morales lost both the ability to control the mass movements that roiled the entire country,¹¹ and the command over the security forces. These forces also had to weigh up whether Morales’s fourth candidacy was legal,¹² not to mention that the suspicion of fraud had further shaken their loyalty to Morales.¹³ All characteristics of an “endgame scenario” were thus fulfilled, i.e. a presidential crisis that can no longer be resolved by a change of policy but only by a change of government; a massive public opposition that transcends class, regional, and sectoral boundaries; stalled political negotiations; and, above all, security forces unwilling or unable to stop social mobilizations against a stubborn president.¹⁴

This article sees Morales’s awkward exit in November 2019 as a unique opportunity to test the concept of coup d’état.¹⁵ The goal is not only to shed light on anomalous seizures of the executive power in other regions, but to discourage the misuse of terms such as coup, impeachment, or presidential interruptions for political and ideological purposes.¹⁶ Remarkably, coups d’état as a technique of power disruption have become less likely worldwide, but the practice of using the term to describe events that might not be a coup seems on the rise.¹⁷ Scholars have questioned the concept of coup d’état as the standard tool for addressing any “unscheduled change of government.”¹⁸ Marsteintredet and Malamud have exposed the cognitive and political costs of this practice. However, although not every removal of a ruler is a coup, the Bolivian event has once again confirmed that scholars still tend to equate presidential interruptions with coups. As Marsteintredet and Malamud show, the overuse of adjectives to qualify a coup has only worsened the prospects for clarity.

Using the alleged coup against Morales in 2019 as an illustration, this article advances the critique of the overstretching of the concept of coup and, more importantly, proposes a different concept to describe cases like the one in Bolivia. To achieve this, a two-stepped approach is adopted. The first step is reasoned as follows: a widely endorsed definition of coup has settled around a threefold structure consisting of target, perpetrators, and tactics.¹⁹ I propose that a fourth element should be added. This element is the partial legal order (PLO), understood to be a set of formal rules legally regulating the dimension of the state’s political regime that concerns access to, and exit from, the state’s highest executive post.²⁰ The addition of a variable to a previous conceptual structure is the result of using the strategy of precising, which is one of the three conceptual strategies developed by Collier and Levitsky (the remaining two being the use of diminished subtypes and shifting the overarching concept itself) that they proposed after evaluating Sartori’s classical conceptual strategy.²¹

Just as earlier developments in Latin America have justified narrowing the definition of coup d’état by now presupposing illegal tactics of the seizure of power,²² the recent Bolivian case forces us to go one step further, because a coup implies not only illegal tactics but also the breaking of a PLO. The PLO’s role in defining a coup has been both under-theorized and underrated. This article demonstrates that assessing the PLO’s fate in the wake of an interruption of a ruler’s mandate helps to explain why the events in Bolivia cannot be described as a coup.

In the second step, an alternative concept to the coup is introduced. This alternative concept seems justified because even an improved concept of coup seems to fall short in the Bolivian case. This alternative concept, which is conceivable only after the revelation of the PLO variable, is based on the following argument: if we assume that a coup must break a PLO, but this order has long since been broken by the time of the anomalous seizure of executive power, then the coup concept loses its grip. For those cases where an anomalous seizure of executive power is accompanied not by the breaking but by the reinstatement of a PLO that had long since collapsed, I propose the term restoration of the regime.²³ The term restoration is not new in this context. Linz uses it in distinction to the “instauration” of a new democracy.²⁴ My version of restoration, however, is about a regime that is not necessarily a democracy, nor a ruling elite but a set of rules, and finally restoration means bringing back a broken regime that has not been replaced and is therefore still influential and retrievable. While introducing a new term is risky, sometimes “new forms. . . of mandate interruptions” cannot be described using a well-established concept.²⁵ Although the concept of regime restoration is better suited for describing cases like the Bolivian one and presents a strong argument for including the PLO, since the PLO’s reinstatement is a precondition for a regime restoration, it remains only one possible technique among others such as the coup d’état to describe such anomalous political processes.

While this approach represents a departure from the contemporary literature on the anomalous seizure of power,²⁶ it connects to this research tradition in order to contribute to the accumulation of knowledge in this field. The following sections undergird both the precising of the coup concept (Section 2) and the rationale of the concept of regime restoration (Section 3). Section 4 details the ousting of Morales to test the new conceptual model.

2. One decisive conundrum around the concept of coup d’état

When considering the Bolivian case, it is reasonable to ask whether the concept of coup has not reached its cognitive boundaries.²⁷ After Morales’s resignation, renowned scholars, Morales’s endorsers, and Morales himself swiftly spoke of a coup. The advocates of the theory of a coup focused on the role of the military.²⁸ Although its role has been overrated and, as will be shown below, the subsequent interpretations were flawed, the role of the military has been persistently endorsed.²⁹ Scholars either skeptical about, or opposed to, the coup theory,³⁰ have viewed the citizen revolt as decisive to understanding these events.

Without revisiting the theoretical premises of what a coup is and examining how these fit the facts, the risk of hastily assuming a coup is high. To begin with, while a coup expresses itself through the illegal interruption of a ruler’s mandate, it also encompasses the breach of the PLO. Thus, Balkin and Levinson view the idea of a coup as both an illegal act and as a subversion of “constitutional structures.”³¹ The breach of a PLO comes close to Bermeo’s description of a “rapid and radical change across a broad range of institutions,” leading thus “to outright. . . breakdown and to regimes that are unambiguously” different.³² However, compared to Bermeo’s “broad range of institutions,” the term “partial legal order” is more precise: it entails only those legal rules that regulate the exit from, and access to, the executive office of the national government, which constitutes a core aspect of a country’s political regime. When scholars do consider an order dimension when defining a coup, they may go to the other extreme and view a coup as implying the revolutionary replacement of one “legal realm” with another, as happened in 1789 in France.³³ Contrary to this extreme, and to Kelsen’s assumption that a coup overthrows “the entire legal order,”³⁴ I believe that a coup breaks something close to what Kelsen referred to as “certain laws of paramount political relevance,” while “a great part of the old legal order remains valid.”³⁵ Of course, theoretically, it is possible for a coup to end up impacting the entire constitution, but there is no need to go that far to define a coup. While Mahmud’s assertion that “coups. . . do not aim at destruction of the entire legal order, but only at usurpation of political offices” is true in the sense that a coup does not target the “entire” legal order, it is less so in reducing coups to “usurpation.”³⁶ The point, then, is to envision the coup as something that means more than seeking the illegal usurpation of a political office, but less than the replacement of an entire legal order. Consequently, and deviating from Kelsen’s assumption that a coup overthrows “the constitution,”³⁷ Mahmud concludes that “a coup d’etat typically suppresses only that part of a constitution that deals with political organs of the state; the rest of the constitution and the larger legal system is expressly left in place.”³⁸ The PLO entails both constitutional and infra-constitutional rules regulating the political regime, whereby breaching them culminates in the breakdown of the regime itself.

Against emphasizing the breaking of a PLO for defining a coup, some authors might argue that coups, unlike revolutions, tend to be stabilizing rather than disruptive.³⁹ While a revolution is barely understandable without pointing to a “major structural change,”⁴⁰ one could argue that a coup, being just as illegal as a revolution, takes care not to disrupt too much. Moreover, some critics might argue that emphasizing the disruption of the legal order would represent a demoted expression of legal positivism underestimating the contributions a coup may yield for institutional continuity.⁴¹ The problem with this reasoning is that even if coup perpetrators sought a rapid stabilization of the country, this does not nullify their prior breach of the PLO. Moreover, a coup’s impact may still be disturbing in normative and practical terms, even if it were far from bringing about profound social changes as revolutions do. Not to mention that this approach fails to distinguish the perpetrators’ anxiety to reactivate the state’s bare coercive capacity from their attempt of reinstating a PLO.

2.1. Do conventional coup definitions overlook the partial legal order?

Although common sense often associates a coup with the breaking of an order, it is striking that this is not mentioned in conventional definitions.⁴² If it is mentioned, it is as a trait explicitly judged to be unnecessary,⁴³ or else it is mentioned in an unsystematic, contradictory manner,⁴⁴ or as a post hoc recognition;⁴⁵ or it is mentioned, but the meaning of this dimension remains conceptually murky;⁴⁶ or it appears to be mentioned, but in reality is not.⁴⁷ One of the most widespread definitions of coup omits this variable altogether.⁴⁸ Powell and Thyne define a coup (or its attempt) by reference to who is unseated, who the perpetrators are, and what tactics are used to unseat the key official of the executive branch (the famous threefold scheme of target-perpetrator-tactic). Their conclusion is that a coup is defined as an attempt at unseating the highest official of the executive branch, whereby the perpetrators must come from the inner state apparatus and the process itself must be overt, actual, and illegal, although not necessarily accompanied by bloodshed. If anything, the variable PLO is remotely hinted at by pointing towards the illegal way in which the unseating is carried out. This, however, is not enough. As Balkin indicated in a comment on the Bush v. Gore court decision, “it is perfectly possible for a government official to break the law. . . and not create a revolution or a shift in legal regime.”⁴⁹ The assassination of a president, while leading to an interruption, is illegal, but is not in itself a proof of regime breakdown. In fact, the illegality of tactics not only alludes to merely one aspect of the PLO,⁵⁰ but emphasizes the method of an unscheduled change in the executive rather than the circuit of legal rules constituting the PLO. Once illegality was established as a defining characteristic of a coup, after having long been underrated,⁵¹ it helped to overcome the pitfall of “conflating coup with legal tactics for government replacement,” such as impeachments.⁵² The next step now is to introduce the PLO to complete the conceptual improvement.

2.2. Towards a four-fold conceptual scheme of coup

The target/perpetrator/tactic scheme has taken root, but if the PLO is to be considered, it becomes clear that this would burst the threefold scheme. Yet, there are signs that scholars feel the need to account for the PLO. For example, Marinov and Goemans concur, in principle, with the threefold pattern, though slightly modify it by dropping the idea of attempt; preferring to speak of the threat or use of force instead of illegality; and concentrating on power seizure rather than on unseating a ruler.⁵³ After presenting their basic definition, however, they note that a coup implies the violation of “constitutional or customary procedures.”⁵⁴ Since Marinov and Goemans insist on not confounding threat or the use of force with the violation of such procedures, their work is an example of how breaching the legal order resonates in efforts to define a coup.⁵⁵ It is not clear, however, whether these “constitutional or customary provisions of how power is supposed to change hands” refer to the political regime in general or only to the specific illegality of the act of seizing power.⁵⁶ In view of the removal of presidents since 2009 in Honduras, Paraguay, and Brazil, many scholars have questioned whether the concept of coup is still useful “in the absence of an institutional rupture.”⁵⁷ But again, it is not clear what institutional means in this context. Some authors believe that the coup assaults the “rule of law,”⁵⁸ others that it changes “policies” and involves the “regime,”⁵⁹ or even affects “the state.”⁶⁰ When authors mention that coups affect the “regime,” they may not be referring to any set of rules,⁶¹ but to the replacement of “the group of elite atop the regime,”⁶² or to the mere scheme of rule organization.⁶³ When authors present a regime as a set of rules, it remains unclear what this set implies. For instance, Marsteintredet and Malamud assert that the illegality characterizing a coup “also normally involves a suspension of the constitutional order.”⁶⁴ While this assertion indirectly alludes to the PLO, the connection is neither developed to its fullest consequence, nor is it unambiguous. In this assertion, ambiguity persists, since a coup always—not “normally”—leads to the “suspension” of those constitutional rules that belong to the PLO and must not necessarily go so far as to “suspend the constitutional order,”⁶⁵ not to mention that Marsteintredet and Malamud blend here the illegality of tactics with the PLO’s stance. In other words, does the idea of suspending the constitutional order involve the entire constitutional order of the polity or only a portion of it? Does it encompass changes between regime subtypes or only those transforming a democracy into an autocracy? Is the illegality of the tactics in a coup synonymous to unconstitutionality?⁶⁶ And once this is clarified, should an order dimension be introduced alongside the conventional elements to define coup d’état? While Marsteintredet and Malamud helpfully elaborate on the difficulties of distinguishing illegal from legal in the context of a coup definition,⁶⁷ they are less concerned with differentiating between (il)legality and (un)constitutionality,⁶⁸ between constitutional order, regime, and PLO, and overlook how the breach of the latter brings down the regime.

2.3. The case of change of elites instead of breach of rules; and the case of change between or within regimes

To clarify both the nature of the PLO and its role in defining a coup, two additional considerations are necessary. The first concerns cases in which the interruption of a mandate is focused on replacing a person or even the elites accompanying this person, and not on the modification of the PLO. This distinction seems important for Latin America, because “while earlier forms of instability resulted in the breakdown of the democratic regime, the ‘new’ pattern of political instability threatens only the president.”⁶⁹ Some authors insist on calling coups,⁷⁰ or “reshuffle coups,”⁷¹ those removals that affect only the individual dimension but not the rules concerning the change in executive power. I disagree. One should distinguish between the victim of a coup (the “government,” the leader) and its target (the PLO).⁷² For a better understanding of the anomalous seizure of the highest executive post in the state, a fourfold conceptual scheme encompasses victim/perpetrator/tactic and target. Where the “irregular exit” hits the victim (the ruling persons)⁷³ but not the target—understood as “the set of rules that regulate how the state is governed”⁷⁴—we are dealing with a “palace revolt” rather than a coup.⁷⁵ In such cases, there is no breaching of the “governing rules,” let alone of the regime,⁷⁶ provided regime is understood as a “set of rules” and not as the “incumbents who control the government.”⁷⁷ Confusing target and victim, Marsteintredet and Malamud conclude that “by narrowing the victim to the government a coup needs not by definition change the nature of the regime. . . but. . . a coup will hold consequences for the regime,”⁷⁸ following thus the conventional undervaluation of the PLO in their definition of coup. I suggest, on the contrary, that a coup must attack both the victim (the ruler) and the target (the PLO), whereby the latter becomes a constitutive element for defining a coup and not something a coup would have “consequences” on that fall short of changing the regime’s nature.

The second consideration is whether a coup may also be assumed if more than the government is hit, but less than the whole regime. Paraphrasing Kotowski, the PLO can have a “minor” scope,⁷⁹ such as when it concerns the rules regulating the subregime of a polity, or it may have a “larger” scope when it concerns the regime itself. Thus, we have a coup (provided all other traits hold) when the anomalous seizure of the executive power entails either the transition, for example, from “monarchic dictatorship” to a “theocratic dictatorship,”⁸⁰ or when it comes to the more demanding transition from an autocratic regime to a democratic one or vice versa.

3. Introducing the concept of restoration of regime

In light of all these considerations, it is now possible to frame the following decisive question. If a coup implies breaching a PLO, is coup still the right category when the interruption of a president’s mandate coincides with the fact that the PLO in question was already broken? Why not look for another concept capable of straightforwardly conveying that an anomalous seizure of executive power may be the starting point for a potential reinstatement of that broken order? In this case, we should not speak of a coup but of a restoration of regime. Naudé in 1693 argued that among the many occasions that make a coup d’état plausible, there was one that called for “rétablissement, et restauration des États.”⁸¹ This is a precedent for what I mean by restoration of regime, the difference being that I am not concerned with the state as a whole, but only with one of its PLOs, and that for Naudé coups are illegal by default,⁸² which does not hold for restoration of regime.

3.1. Plausibility parameters for assuming a restoration of regime

Coup and restoration share certain traits: both imply an anomalous seizure of the executive power; both should not be diagnosed by an artificially reconstructed case ex post facto;⁸³ and both share speediness. Yet, only coups lead to a generalized breaking up of a PLO, whereas restoration enters the scene when the anomalous seizure of executive power coincides with a previously broken PLO, whereby reinstating this broken order is not inevitable. To ensure a correct assessment, the observer has three indicators for assuming a possible restoration. First, it is important to know whether the unseated systematically violated the rules before his mandate was interrupted. If it did, the likelihood is high that his removal may be the initiation of reinstating the affected PLO.⁸⁴ The removal of a longstanding despotic ruler is thus seen as beneficial to democracy.⁸⁵ Second, if the steps taken to fill the political vacuum following the interruption adhere to minimal constitutional procedures, restoration becomes more likely. Third, the observer has to scrutinize what the new government’s initial legal measures entail. If these measures reveal clear intentions not only to recover the minimal conditions of state coercion but to reinstate the previously broken PLO, restoration attains maximum plausibility. In all these instances, a restoration remains a punctual technique implemented within a short time span, whose nature cannot be deduced from particular contents of policy. Any substantive motivations that perpetrators might present in justification of their deeds are irrelevant for defining a restoration. Needless to say, it is one thing for a restoration to fail, another for the restored regime to recollapse. Moreover, a restoration should not be rejected only because it does not perform perfectly, nor should one discard the fact that leaders may feign a restoration. Philosophically speaking, a restoration is nothing but a negation of a negation, meaning a move against an already damaged setting.

3.2. Are the techniques of restoration (and coup) also applicable to autocracies?

To make the discussion complete, the final question to be addressed is whether coup or restoration are also applicable to autocratic regimes. To begin with, scholarship now tends to accept that autocracies have constitutional arrangements.⁸⁶ I do not intend to discuss whether regulations of the exit from, and access to, power in autocracies can be termed constitutional. For the moment, the point is that autocracies normally have some sort of institutionality that regulates transfer of power.⁸⁷ Some types of autocracy, such as monarchies or elected autocracies,⁸⁸ do have recognizable rules, whereas others, such as personalistic variants,⁸⁹ are less clear on that. Either way, in an autocracy—as in a democratic regime—a coup must have something to transgress. If an anomalous seizure of autocratic executive power coincides with the breaching of the autocratic rules regulating the changes in the executive power, then a coup takes place.⁹⁰ If this seizure happens at a time when those autocratic rules have already been broken, then a restoration is the more likely technique. Of course, in the latter case, restoration is only one possible path. Theoretically, it is possible that perpetrators have no intention in reinstating the broken order. Figure 1 synthesizes the key findings.

Figure 1.

Distinction between a coup d’état and regime restoration.

¹ The figure of extra-state actors applies to the Bolivian case.

² The “functioning” partial legal order (PLO) translates into four regime variants. Founded here means no longer transitional; for stable though not consolidated regime, see Esther Brimmer, Vigilance: Recognizing the Erosion of Democracy, inProtecting Democracy International Responses 217 (Morton Halperin & Mirna Galic eds., 2005); severely eroded is warranted because, while some degree of erosion may affect a consolidated democracy, at some point the erosion is so severe that it makes little sense to speak of a stable PLO, much less of a consolidated one.

³ “Broken” goes along the prevalence of a not-yet-replaced-broken regime setting, as the new incoming regime is unable to replace the broken PLO.

⁴ Incumbent takeover applies for the Bolivian case.

3.3. Is there no better way than to introduce the new concept of restoration?

One could consider whether, instead of introducing restoration, better alternatives could be found by comparing the concept of coup with similar phenomena such as revolution,⁹¹ civil war, or what has been called “failed presidents,”⁹² or by elaborating different types of coups.⁹³ Let me present some caveats for each of these alternatives.

The alternative that immediately comes to mind when comparing the concept of coup to similar processes is presidential interruption, which is often associated with mass protests.⁹⁴ Indeed, two qualities are common to restoration and interruption theories: extra-state⁹⁵ perpetrators can be decisive and the struggle for the head of the executive is at stake.⁹⁶ Despite these similarities, the PLO is the issue where interruption does not fare better than restoration, even when considering that the presidential-interruption approach addresses the distinction between “presidential breakdown” and “democratic breakdown,”⁹⁷ thereby indirectly pointing to the PLO. For starters, while presidential breakdown theory is usually intended for analyzing democracies,⁹⁸ restoration can likewise be applied to non-democratic settings. Furthermore, although it is true that formal impeachment may be too narrow to explain “failed presidencies,”⁹⁹ the alternative concept of “resignations in response to street protests”¹⁰⁰ as part of a “broader sample of failed presidents” is equally debatable. The last idea reflects victim, perpetrators, and tactics, but leaves out the PLO. Not surprisingly, scholars pursuing the interruption approach may find themselves in murky waters. They might, for example, insist that the events in Honduras in 2009 were a coup, even acknowledging that the presidential removal did not imply a breakdown of the regime,¹⁰¹ which I believe blurs the distinction between palace revolt and coup. Or take Barracca’s assessment that presidential interruptions in Ecuador in 2000 and Venezuela in 2002 were coups.¹⁰² The problem here is twofold: first, a regime breakdown cannot be proved in either country, as is shown in the Online Appendix; and second, while Barracca links coup, at least in a first step, to a breakdown of a “constitutional order,” he calls cases in which “the military plays a role in removing an unpopular chief executive for the purpose of preserving the current democratic regime,” a “palace coup,” and ends up stretching the concept of coup to cover mere palace revolts.¹⁰³

As for the coup subtypes, I defer to Marsteintredet and Malamud, especially their critiques of the family resemblance approach of concept building. For illustrative purposes, however, let me mention some limitations of coup with adjectives.¹⁰⁴ Some examples seem to share the curative virtues of regime restoration, as the so-called “guardian coup,” which is supposed to bring back stability.¹⁰⁵ However, this effect most probably refers to reactivating state capabilities of minimal coercion, which is different from reinstating a PLO. Or take the idea of “democratic coups.”¹⁰⁶ Although coups more often lead to autocracies than to democracies,¹⁰⁷ the idea of a coup leading to democracy conveys an optimistic prospect similar to the one associated with restoration. Nevertheless, the idea of a “democratic coup” implies substituting a functioning autocratic order, while a restoration presupposes an order no longer functioning.

Other variants, such as the term autogolpe, blur the rationale behind a coup. Still, it has been used to describe Morales’s exit.¹⁰⁸ This position makes sense colloquially, but is technically ambiguous, since Morales’s infractions, however deep, did not lead to an irregular ousting of any ruler. In contrast, both coup and restoration require such an irregular ousting of the ruler as a definitional trait. The idea of autogolpe is misleading, since the perpetrator is far from harming himself, and hence it seems more appropriate to speak of an “incumbent takeover,”¹⁰⁹ or an “endogenous termination.”¹¹⁰ While Svolik’s category of incumbent takeover describes well how Morales fueled the “lingering demise” of democracy many years before his fall,¹¹¹ it does not account for what occurred to the broken order the day of Morales’s exit. Furthermore, Svolik and Maeda assume a collapse of a (democratic) order,¹¹² whereas restoration does not equate the anomalous seizure of the executive power to the breaking of an order. Additionally, the idea of autogolpe may lead to conflating the concepts of government and regime¹¹³ by extending the victim of a coup to the congress¹¹⁴ or any state organ,¹¹⁵ rather than having the victim embodied by the “state’s primary leader,” meaning the “chief executive.”¹¹⁶

A final example is the so-called soft coup or neo-coup.¹¹⁷ The reasoning behind this type of coup is based on a lax, if not biased, handling of concepts.¹¹⁸ In their view, a soft coup could have occurred even if the removal was legal. Moreover, they criticize why civilians are not considered to be perpetrators of a coup—as if all coup definitions presuppose the military as the only perpetrator.¹¹⁹ Needless to say, considering the PLO is not in their interest.¹²⁰ In fact, they seem to have found the new element to improve the definition of a coup elsewhere, by making its definition dependent on whether the ousted ruler was part of a radical leftist policies wave or not. Thus, what technically was not a coup would become one if the victim were a leader, such as Chavez, Morales, or Rousseff, regardless of whether the unseating was legal or not,¹²¹ and no matter how these leaders treated democratic rules in the past. Patently, the definitional task moves into the domain of emotional attachment.¹²²

In this soft-coup approach, the elected nature of the ousted presidents is perceived as the paramount virtue to neutralize criticism of their possible undermining of democracy;¹²³ citizen protests against elected and leftist rulers are assumed as intrinsically golpist (coup-plotter);¹²⁴ judicial or parliamentary measures that do not conform to the leftist rulers’ visions are declared intrinsically illegal; and the risk of an incumbent takeover¹²⁵ is viewed at best sarcastically.¹²⁶ Moreover, soft-coup analysts reduce the figure of impeachment to a ploy of oligarchic elites.¹²⁷ Not surprisingly, it is stated that “the wave of soft coups d’état in Latin America started in 2009 in Honduras. . . followed by Paraguay in 2012, Brazil in 2016, and Bolivia in 2019.”¹²⁸

Besides, this approach is opportunistic. If its proponents were to apply their own parameters, they ought to evaluate Sánchez de Lozada’s abnormal exit in Bolivia in 2003 as a soft coup. In fact, they reject this as they consider him a neoliberal.¹²⁹ What occurred to Sánchez de Lozada must have been a virtuous removal triggered by popular mobilizations, reserving the term coup to when public pressure removes leftist rulers. That the right has used impeachment to interrupt leftist presidents is beyond doubt, yet this must not lead to overstretching the coup concept, or to presuming that anything politically inconsistent with one’s preferences is necessarily illegal, or that all such interruptions follow this pattern.¹³⁰ This opportunistic, ideologically driven theory, banalizing intra-state institutional checks, is hardly suitable for scientific assessment, however important its insights might be for understanding the politics of presidential interruptions.

3.4. Restoration works for describing paradox regime settings

Restoration must face one further challenge. What if there is always a regime to break in any polity? The Bolivian case is unique as it also serves to clarify this puzzle. Let us create a hypothetical scenario A in which Morales’s resignation could have followed a “slow death”¹³¹ of Bolivia’s democratic regime, until it became replaced by an autocratic one that Morales himself helped install during his previous terms. In other words, on the day of his resignation, an (autocratic) regime was functioning whose PLO could be breached, giving the coup its necessary target. In this scenario, Morales’s violations could have led Bolivia to competitive authoritarianism well before his anomalous exit.¹³² Interrupting Morales’s mandate in this scenario could possibly be a kind of “good coup”—a coup insofar as, in addition to the other three requirements, there existed an (autocratic) order to be broken; and “good coup,” given the probable contribution to what Linz would call reinstaurating democracy. However, this is only true if before this reinstauration the old democratic regime had actually been fully replaced by an autocratic regime; otherwise, restoration is the better concept, because without a replacement of the broken regime, there is still the possibility of restoring the broken regime rather than building a new regime from scratch. Let us now consider the reverse hypothetical scenario B and assume that Bolivia was still a democracy, despite all severe damages Morales inflicted on the regime. Bolivia could have been a case of a “borderline regime” still anchored in democracy.¹³³ If so, the abnormal unseating of Morales would not have coincided with a broken regime, but with a severely eroded democracy. Under such circumstances, Morales’s exit could have either (i) been accompanied by the continuation of the “borderline” situation; (ii) coincided with moving the eroded regime away from the border, reverting democracy’s decay;¹³⁴ or (iii) brought the “blow of mercy” to the eroded democracy, thus paving the way to autocracy as the legacy of the unseated leader. Since breaking the PLO happens neither in (i), as the borderline situation is deadlocked, nor in (ii), as democracy at least restabilizes, a coup could be the correct concept only in (iii), unless, although the PLO breaks, the exit of the president comes along with a legal tactic led by intra- or extra-state actors (recall: a coup presupposes illegality), or it is not legal but led by extra-state actors (recall: a coup presupposes intra-state actors). Restoration requires a scenario C, in which a “broken-not-yet-replaced” regime must be the prevailing (transitional) constellation. Admittedly, stopping the erosion of a regime as in setting (ii) of scenario B shares a similar positive sensation with the idea of regime restoration; nevertheless, both alternatives differ conceptually.

I maintain that, in Bolivia, the PLO had already been broken by 2019, but its substitute was still too immature to replace the entire democratic regime. Similar perplexing settings concerning the state regime are not new. Przeworski,¹³⁵ for instance, mentions settings in which no regime exists but turmoil. Another example of such a configuration is given by O’Donnell when he states that one thing is to have a democratic “government,” another is to have a democratic “regime.”¹³⁶ Likewise, scholars have recalled Linz’s description of a setting between democracy and autocracy, the fluidity of which compels one to speak of a “situation” rather than of a regime.¹³⁷ If a new regime is on its way but falls short of being the new functioning regime of a polity, then restoration becomes a possible technique to describe the regime effects accompanying an anomalous seizure of executive power, because here a broken, albeit not yet reinstated, PLO would underlie the prevailing regime constellation. It becomes clear that a successful regime replacement implies an irreparable disturbance of the formerly operative mechanisms, which until then had regulated the exit from, and the access to, state power, and presupposes the profound unsettling of the subjective perceptions of political actors and citizens about the right order of their country. Both conditions do not hold for Bolivia. Here, before October 2019, all these disturbances had been minor in terms of the autocratic regime promoted by Morales, to the extent that they did not condense into a PLO to be breached by a coup. On the contrary, the broken democratic regime was still radiating both operationally and subjectively onto the dynamics of political processes, propelled by the fact that it had not yet been replaced completely. As long as the new order is not “efficacious” because the “individuals” do not “behave by and large in conformity with the new order,”¹³⁸ it is not valid.

Until Morales resigned, the political actors in Bolivia had perceived the democratic order, however damaged, as still valid. Voters participated in elections in a disciplined way, including the elections of October 20, 2019, notwithstanding the skewed electoral playing field. Due to the normative, and functional, immaturity of the developing autocratic regime, in addition to the people’s subjective appraisal concerning the valid and legitimate order, what was at issue on November 10 was not the “sudden death”¹³⁹ of a regime but the opportunity of reinstating an already broken PLO.

Table 1 summarizes these findings by plotting the restoration of a regime alongside other “legal removals such as impeachment and popular recalls,” in addition to illegal ones,¹⁴⁰ the coup being the classical subtype. To underscore the usefulness of restoration as a concept, other techniques are added, which I refer to as quasi-impeachment, façade impeachment, and pseudo-impeachment. It suffices for now to say that quasi-impeachment refers to events in which the anomalous removal follows largely legal ways without applying the formally foreseen impeachment mechanism (as in Bolivia in 2003),¹⁴¹ or when no such mechanism is provided at all (Honduras in 2009),¹⁴² or has failed, whereby in all these variants no regime breakdown takes place.¹⁴³ If a quasi-impeachment is based primarily on legally backed citizen revolts representing the main driving force for the removal,¹⁴⁴ it is a non-state-led¹⁴⁵ quasi-impeachment and comes close to the idea of a “popular impeachment.”¹⁴⁶ This was the case in Bolivia in 2003,¹⁴⁷ where the main force was “insurrectional politics” of “social movements”¹⁴⁸ leading to a “resignation under massive protest.”¹⁴⁹ While Bolivian citizens acted as perpetrators, both in the 2003 non-state-led quasi-impeachment and the restoration in 2019, and the legislative branch played a “reactive role,”¹⁵⁰ these perpetrators did not seize power. Here, intra-state actors are usually involved, but do not play the primary role as in classical impeachments.¹⁵¹ I refer to state-led quasi-impeachment when the organs of a state are the perpetrators, even acknowledging that here street protests may play an important role, as was the case in the classical impeachment in Brazil in 2016.¹⁵²

Table 1.

Classifying restoration of regime in reference to other mechanisms of anomalous seizure of executive power^a

Anomalous seizure?	Legal?	Intra-state actors as perpetrators?	Breaking the PLO?	Technique	Examplesb
Yes	Yes	Yes	No	Impeachment	Paraguay 1999, 2012 Brazil 2016
Yes	Yes	No/(Yes)c	No, already broken	Regime restoration	Bolivia 2019
Yes	Yes	Yes/(No)d	No	Recall	Bolivia 2008
Yes	Yes	No	No	Non-state-led quasi-impeachment	Bolivia 2003, 2005
Yes	Yes	Yes	No	State-led quasi-impeachment	Honduras 2009
Yes	No	Yes	Yes	Coup d’état	Bolivia 1971 Chile 1973
Yes	No	Yes	No	State-led pseudo-impeachment	Honduras 2009
Yes	No	No	No	Non-state-led pseudo-impeachment	Venezuela 2002 Ecuador 1997, 2000, 2005
Yes	No	Yes	No	Façade impeachment	Paraguay 1999, 2012

^aFor details on the analytical justification of the classification of the cases and of the ambivalent attributions of Honduras 2009 and Paraguay 1999/2012, see Online Appendix. The Table omits more theoretical combinations such as “popular” pseudo-coup (Yes/Yes/No/Yes); state-led pseudo-coup (Yes/Yes/Yes/Yes); and semi-coup (Yes/No/No/Yes).

^bPresidents affected were: Cubas (Paraguay 1999); Rousseff (Brazil 2010); Lugo (Paraguay 2012); Morales (Bolivia 2008 and 2019); Banzer, Sánchez de Lozada, and Mesa (Bolivia 1971, 2003, and 2005, respectively); Zelaya (Honduras 2009); Allende (Chile 1973); Chávez (Venezuela 2002); Bucaram, Mahuad, and Gutierrez (Ecuador 1997, 2000, and 2005, respectively).

^cRegime restoration can go along either intra- or extra-state perpetrators; in the case of Bolivia in 2019, it meant extra-state perpetrators.

^dRecall can go along either intra- or extra-state perpetrators; in Bolivia 2008, it meant intra-state perpetrators (see further Online Appendix, Section 1).

In a pseudo-impeachment, the removal not only bypasses the impeachment mechanism (if one exists) or its application’s conditions were uncertain (Ecuador in 2000),¹⁵³ but it correlates with illegality (e.g. Honduras, if Zelaya’s removal were deemed illegal),¹⁵⁴ and there is no breaking of a PLO, which makes the coup concept not applicable, as Table 1 indicates. Finally, façade impeachment refers to an impeachment process carried out according to the constitution,¹⁵⁵ but marred by manipulations in either the “legal justification” or the “legal procedure,” or both.¹⁵⁶Table 1 shows that scholars have rightly contrasted the coup as the Inbegriff [epitome] of an illegal seizure of power,¹⁵⁷ against a gamut of “legal procedures” (such as impeachment proceedings), all leading to a comparable outcome. Yet, the pseudo-impeachment and façade impeachments introduced here reveal additional non-legal paths of mandate interruption in addition to the coup, and allow to avoid labeling as a coup an impeachment carried out with “congressional abuses.”¹⁵⁸ Applying the rationale of Table 1, the 2009 case of Honduras loses the flavor of a “borderline case,”¹⁵⁹ and ceases to be “unequivocally” a coup.¹⁶⁰ It is now possible to account for non-legality in both the cases of Honduras 2009 and Paraguay 2012 (should illegality be confirmed),¹⁶¹ avoiding automatically declaring them as coups after verifying that no PLO was broken. There is one thing that quasi- and pseudo-impeachments, façade impeachments, and—logically—normal impeachments imply that justifies the restoration of regime as an alternative concept. In all the former cases, there is no breakdown of a regime (as in a coup), whereas in restoration the regime has long since been broken.

4. Restoration of regime applied to the Bolivian case

Morales’s systematic violations of the PLO well before he resigned are one of the decisive reasons to doubt that the coup concept is the best fit to describe these events.¹⁶² In 2008, aiming to unblock the holding of a referendum to approve of the new Constitution, Morales publicly promised to recognize his first term in office when the day comes that the term limits provided for in the draft constitution come into effect. After the new constitution was adopted in 2009, Morales, however, addressed the Constitutional Court in 2013 to bypass the constitutional two-term limit. The judges overtly sided with Morales in their decision, arguing that his first mandate did not count as it had been exercised during the old republic,¹⁶³ whose existence was dissolved through the new constitution. The boldest violation occurred in 2016, when Morales dismissed his defeat in the referendum he himself had organized with the aim of extending the presidential term limits, and, backed by the venal electoral court’s refusal to enforce the referendum results,¹⁶⁴ immediately sought new ways of flouting the constitutional limitation to two presidential terms. In effect, Morales disrespected the people both as a constituent power (bypassing the constituent power through a court decision in 2013) and as a constituted power (denying the results of the 2016 referendum).

However, Morales’s violations did not end there. To force yet another reelection after the lost referendum, he had his congressmen address the Constitutional Court once again in 2017. This time, the judges ruled that the two-term limit in Article 168 of the new Constitution was inapplicable.¹⁶⁵ They argued that under article 23 of the Inter-American Convention on Human Rights,¹⁶⁶ it was Morales’s human right to indefinitely compete for reelection. As in 2013, an obsequious Constitutional Court surpassed the constituent power by lifting the constitutional ban on limited terms of office. Finally, in the 2019 general elections, Morales, being an illegal candidate per se, once again disregarded the people by distorting the electoral playing field in many forms. The bottom line is that Morales’s violations were so systematic that he damaged not only demo-cracy (referring to the people as a regular electoral body periodically selecting state authorities), but also demo-archy (referring to the people as the ultimate source of the constitution). His (admittedly “unscheduled”) resignation, therefore, not only implied a change at the head of office, but simultaneously offered a chance of reinstating the order the ousted president had himself broken during his previous mandates.

4.1. The role of the military

There are further reasons that reinforce the theory of restoration. The rejection of a coup in Bolivia is justified not only because of the absence of an order to be broken, but is supported after a calm assessment of the military’s role. Levitsky’s conviction of a coup in the Bolivian case, in contrast, relies on General Kaliman’s advice for Morales to step down, approximately one hour before Morales’s public resignation.¹⁶⁷ Zeroing in on General Kaliman’s statement is a parsimonious approach in the sense of empirical and testable, and yet ill-advised. In fact is that, prior to General Kaliman’s declaration, the Bolivian Workers Union and the Ombudsman Office had already urged Morales to resign at around noon that day. Moreover, early morning that same day, Morales had learned that the OAS report, referring to relevant irregularities during the electoral process, was to be released that morning and sought to prevent its publication. Prior to Kaliman’s afternoon appearance, Morales had already instructed his spokesman to contact advisors of the main oppositional candidate to inform them that Morales had decided to step down,¹⁶⁸ and wanted to inquire whether the president of the Senate—a partisan of Morales and the constitutionally foreseen successor—could count on the opposition’s support. In his autobiography Morales himself confirmed he had decided to resign well before Kaliman’s declaration.¹⁶⁹ Notably, in his letter of resignation Morales did not cite Kaliman’s proposal as the reason for his decision. Apart from the fact that general Kaliman’s decision was backed by military law,¹⁷⁰ the Bolivian military’s cautious behavior is actually comprehensible given the thirty-five-year-long evolution that this institution underwent after Bolivia’s return to democracy. It is true that Latin American “radical leftist presidents,” like Morales, have implemented policies that in the past would have triggered military coups but these leaders have now been able to successfully pursue a strategy to secure the military’s loyalty.¹⁷¹ This historical background of military support to civilian rule since 1982 provides the larger context needed for approaching the Bolivian case.¹⁷² If history is only used to remind us that Bolivia holds a record in military putsches, there is a risk of taking what happened on November 10, 2019 as a continuation of a sort of genetic logic—one in which the only strategy for the military to react to a political crisis is a coup. This does not consider that “praetorian legacies can be broken,” and the risk of falling into the coup trap can be avoided.¹⁷³ In short, it is delusive to isolate a bizarre suggestion of a military general from the narrower and broader context.

4.2. Legal uprising of non-state actors

Additional aspects that enhance the plausibility of a restoration are the following. For Powell and Thyne, on the one hand, the illegality of a coup must be distinguished from the unseating of a ruler based on the liberty of citizens to protest against a government.¹⁷⁴ While civil society mobilizations against a government may also be illegal,¹⁷⁵ or of disputable virtue,¹⁷⁶ some that are not necessarily illegal may end in the resignation of a ruler.¹⁷⁷ In fact, legal removals of heads of state, such as impeachments or popular recalls, should not be distorted in their nature by calling them coups.¹⁷⁸ On the other hand, in a coup, the perpetrators must come from within the state apparatus.¹⁷⁹ Neither the condition of illegality nor the condition that perpetrators must be part of the state are met in the Bolivian case,¹⁸⁰ nor is the condition fulfilled that the military has to be active, if one agrees that it does not suffice for the military to deny its support to the incumbent in order to declare the army as coup plotter.¹⁸¹ In November 2019, the Bolivian military either remained largely passive,¹⁸² since the few actions it undertook were not determinant (i.e. Kaliman’s conference),¹⁸³ or it behaved arguably in a constitutional manner as when the military activated itself to launch operations against the illegal use of arms on November 9.

4.3. The restoration plausibility test

Let us finally check the three restoration-plausibility parameters mentioned earlier. First, Morales was not only the victim of an interruption of his mandate, but for years he himself was the author of massive legal violations against the PLO that underlies democracy.¹⁸⁴ Under such circumstances, the very removal of such a ruler could be the beginning of ending these violations. Second, the filling of the political vacuum that arose in Bolivia after Morales’s resignation was performed by and large in line with the foreseen constitutional procedures of succession.¹⁸⁵ President Añez was installed following Article 160/I of the Bolivian Constitution. Third, the first statute issued by the constitutionally backed transitory new government entailed concrete measures aiming at reinstating the broken order.¹⁸⁶ This statute went as far as to reposition the outcome of the 2016 referendum that Morales refused to accept, by explicitly ruling that candidates elected for two presidential periods were not eligible to run again, and its enforcement proves that this restoration was not sham but genuine.

5. Conclusions

The controversies deriving from the events in Bolivia hint at deeper theoretical concerns. The concept of restoration reminds us that political science sometimes defends too rigidly the axiom “democracy is the only game in town,” without considering that this ideal may at times be paradoxically protected through measures that are barely concordant with democratic mechanisms of conflict solution but are still constitutional. Restoration implies the change of the head of the executive through means other than free and fair elections. Yet, replacing an elected ruler while bypassing the constitutionally provided democratic mechanism does not necessarily mean this process is unconstitutional.¹⁸⁷ Some may fear that considering restoring a (democratic) regime by non-democratic albeit constitutional means may open us up too much to an undemocratic culture of thought, with these skeptics seeming to condone the despotic nature of those in power by recalling that these leaders were democratically elected before they became despots.

This fear becomes clear in article 9 of the 1948 OAS Charter, which condemns a forcible removal of a “democratically elected” president.¹⁸⁸ However, one might ask whether this article could not be supplemented as follows: unless the democratically elected president has systematically subverted the country’s legal democratic order,¹⁸⁹ provided the nondemocratic interruption of this democratically elected ruler is carried out by legal and non-violent means. This spirit was in fact advanced in the 2001 Inter-American Democratic Charter by going beyond the coup and confronting undemocratic actions of elected officials.¹⁹⁰ While Wolff may be right when stating that the transition from Morales to Áñez looks “particularly undemocratic in substantive terms,” because it “openly and drastically contradicts the popular will as reflected in the party affiliation of the elected president and the composition of parliament,”¹⁹¹ I would add that sometimes even such undemocratic ways can be an option if the priority is to protect the constitution (and the democratic system therein), provided that some minimum guarantees are respected. Otherwise, societies could be condemned to tolerate the unlawful practices of rulers only because they are democratically elected,¹⁹² instead of, in emergency situations, resorting to methods which may not be strictly speaking democratic but are still constitutional.

Another area of future research concerns the fact that restoration of regime belongs to a peculiar transitional setting. There, a “broken, not yet replaced regime” coexists with a “given, not yet mature new regime” unable to impose itself. In other words, a breakdown does not always seem to immediately flow into a replacement and may lead to a constellation characterized by a stalemate of regimes.

Be that as it may, some scholars presented the term coup for describing the Bolivian events as a rather technical concept, pretending not to take sides. This makes sense, but only if the observers themselves are patient with fact-gathering, avoid undertheorizing, and are aware of the context.

Footnotes

1

Steven Levitsky & María Victoria Murillo, The Coup Temptation in Latin America, N.Y. Times (Nov. 26, 2019), https://www.nytimes.com/search?query=The+Coup+Temptation+in+Latin+America.

2

Evo pide 500 años de movimientos sociales, El Deber (Dec. 27, 2016), https://eldeber.com.bo/105796_evo-pide-500-anos-de-movimientos-sociales.

3

Fabrice Lehoucq, Bolivia’s Citizen Revolt, 31 J. Democracy 130, 137, 138 (2020); Jonas Wolff, The Turbulent End of an Era in Bolivia: Contested Elections, the Ouster of Evo Morales, and the Beginning of a Transition Towards an Uncertain Future, 40 Revista de Ciencia Política 163, 165 (2020); Linda Farthing, In Bolivia, the Right Returns with a Vengeance, 52 NACLA Report on the Americas 5, 5 (2020).

4

John Crabtree, Assessing Evo’s Bolivia: Inclusion, Ethnicity, and Class, 55 Latin Am. Res. Rev. 379, 383 (2020).

5

Lehoucq, supra note 3, at 132; Wolff, supra note 3, at 165.

6

Lehoucq, supra note 3, at 142; Crabtree, supra note 4, at 389.

7

Lehoucq, supra note 3, at 133; Wolff, supra note 3, at 164.

8

Santiago Anria & Kenneth Roberts, Bolivia after Morales: What Lies in Store for the Country?, For. Aff. (Nov. 21, 2019), https://www.foreignaffairs.com/articles/bolivia/2019-11-21/bolivia-after-morales.

9

See Aries Arugay, Saviors or Spoilers? Explaining “Civil Society Coups” among Democratizing Regimes, 16 Thammasat Rev. 167, 172 (2013) (discussing the activation of civil society mobilizations after the extenuation of constitutional options); Leon Zamosc, Popular Impeachment: Ecuador in Comparative Perspective, inShifting Frontiers of Citizenship: The Latin American Experience 237, 263 (Mario Sznajder, Luis Roniger, & Carlos Forment eds., 2013) (describing failures in vertical and horizontal accountability as causes for “direct non-institutional intervention of civil society to sanction the misconduct of a president”).

10

Leiv Marsteintredet, Explaining Variation of Executive Instability in Presidential Regimes: Presidential Interruptions in Latin America, 35 Int’l Pol. Sci. Rev. 173, 174 (2014).

11

Lehoucq, supra note 3, at 138.

12

On the illegality of Morales’s candidacy, see Franz Xavier Barrios-Suvelza, El golpe que no fue: La última crisis estatal boliviana y los límites del concepto de golpe de Estado, 191 Revista de Estudios Políticos 185, 201 (2021).

13

On reasons why the military in developing countries withdraws when facing mass protests, see David Pion-Berlin & Harold Trinkunas, Civilian Praetorianism and Military Shirking During Constitutional Crises in Latin America, 42 Comp. Pol. 395, 396 (2010).

14

See id. at 402. On the issue of endgame scenario in combination with mass mobilization, see Kevin Koehler & Holger Albrecht, Revolutions and the Military: Endgame Coups, Instability and Prospects for Democracy, 47 Armed Forces & Soc’y 148, 154 (2021).

15

The findings of this single case study will be reinforced by a review of selected Latin American countries and alternative concepts in Section 3.4. and in Table 1.

16

Leiv Marsteintredet & Andrés Malamud, Coup with Adjectives: Conceptual Stretching or Innovation in Comparative Research?, 68 Pol. Stud. 1014, 1016 (2020); Oisín Tansey, The Fading of the Anti-Coup Norm, 28 J. Democracy 144, 153 (2017).

17

Marsteintredet & Malamud, supra note 16.

18

Ryan Pitts, Rosemary Joyce, Russell Sheptak, Kregg Hetherington, Marco Castillo, & Rafael Ioris, 21st Century Golpismo: A NACLA Roundtable, 48 NACLA Report on the Americas 334, 345 (2016).

19

Marsteintredet & Malamud, supra note 16, at 2, 9.

20

The term partial legal order (Teilrechtssordnung) goes back to Kelsen, who used it to describe either the legal order embodied in a person, as opposed to the overall legal order, or the territorial differentiation of the legal order of the state. Hans Kelsen, Allgemeine Staatslehre 64, 66, 143 (1993).

21

David Collier & Steven Levitsky, Democracy with Adjectives: Conceptual Innovation in Comparative Research, 49 World Politics 430 (1997).

22

Marsteintredet & Malamud, supra note 16, at 16.

23

“Anomalous” is to be understood as both premature and extraordinary. See Michael Alvarez & Leiv Marsteintredet, Presidential and Democratic Breakdowns in Latin America: Similar Causes, Different Outcomes, inPresidential Breakdowns in Latin America 33, 34 (Mariana Llanos & Leiv Marsteintredet eds., 2010). To stay within the terminological tradition of coup d’état, my first inclination was to speak of restoration of state. However, what is at stake is something “within the frame of. . . the state,” and not necessarily the state itself. SeeRobert Kann, The Problem of Restoration 68 (1968). I borrow the term reinstatement from id. at 408. Kann distinguishes reinstatement (“of conditions”) from restoration (of a “system”): id. at 57, 3.

24

Juan Linz, Crisis, Breakdown & Reequilibration 91 (1978).

25

Pitts et al., supra note 18.

26

For studies on presidential interruptions, see Aníbal Pérez-Liñán, Presidential Impeachment and the New political Instability in Latin America (2007); Kathryn Hochstetler & David Samuels, Crisis and Rapid Reequilibration: The Consequences of Presidential Challenge and Failure in Latin America, 43 Comp. Pol. 127 (2011).

27

Urging to reassess the coup concept in order to clarify if “manipulated impeachment” is a “new form of coup,” see Aníbal Pérez-Liñán, Impeachment or Backsliding? Threats to Democracy in the Twenty-First Century, 33 Revista Brasileira de Ciencias sociais 1, 2 (2018).

28

Levitsky & Murillo, supra note 1; Michael Shifter, En términos muy técnicos lo de Bolivia es un golpe, La República (Nov. 17, 2019), https://larepublica.pe/domingo/2019/11/17/michael-shifter-en-terminos-muy-tecnicos-lo-de-bolivia-es-un-golpe/; Martha Chew, From the “Pink Tide” to “Soft Coup d’État” in Latin America: the Case of Bolivia, 19 Persp. on Global Dev. & Tech. 597, 611 (2020).

29

Steven Levitsky, Na Bolívia houve golpe, Globo (Nov. 12, 2019), https://glo.bo/3iu4xZj; Farthing, supra note 3, at 8; Anria & Roberts, supra note 8; Chew, supra note 28, at 611.

30

See Wolff, supra note 3, at 177 (skeptical of the military role). But see Lehoucq, supra note 3, at 132; Barrios-Suvelza, supra note 12, at 188 (against a prominent role of the military).

31

Jack Balkin & Sanford Levinson, Understanding the Constitutional Revolution, 87 Va. L. Rev. 1045, 1049 (2001).

32

Nancy Bermeo, On Democratic Backsliding, 27 J. Democracy 5, 6 (2016).

33

Michael Green, Legal Revolutions: Six Mistakes About Discontinuity in the Legal Order, 83 N.C. L. Rev. 331, 335 (2005).

34

Hans Kelsen, General Theory of Law and State 304, 373 (1949) (emphasis added).

35

Id. at 117 (emphasis added).

36

Tayyab Mahmud, Jurisprudence of Successful Treason: Coup d’État and Common Law, 27 Cornell Int’l L. Rev. 49, 107 (1994).

37

Kelsen, supra note 34, at 117.

38

Mahmud, supra note 36, at 129 (emphasis added).

39

Christopher Moore, Political and Military Coups, in21st Century Political Science: A Reference Handbook 124, 126 (John Ishiyama & Marijke Breuning eds., 2010). See especially David Rapoport, Coup d’État: The View of the Men Firing Pistols, inRevolution 53, 56 (Carl Friedrich ed., 1966).

40

Christoph Kotowski, Revolution. Untangling Alternative Meanings, inConcepts and Method in Social Science The Tradition of Giovanni Sartori 203, 218 (David Collier & John Gerring eds., 2009).

41

Charles Sampford, Coups d’État and Law, 13 Bull. Austl. Soc’y Legal Phil. 253, 257 (1989).

42

Rapoport, supra note 39, at 60; Samuel Huntington, Political Order in Changing Societies 218 (1968); Edward Luttwak, Coup d’État Practical Handbook 13 (1969); Naunihal Singh, Seizing Power 51 (2014); Eric Rittinger & Matthew Cleary, Confronting Coup Risk in the Latin American Left, 48 Stud. in Comp. Int’l Dev. 403, 405 (2013); Fabrice Lehoucq & Aníbal Pérez-Liñán, Breaking Out of the Coup Trap: Political Competition and Military Coups in Latin America, 47 Comp. Pol. Stud. 1105, 1109 (2014); Rafael Martínez, Subtipos de golpes de Estado: Transformaciones recientes de un concepto del siglo XVII, 108 Revista CIDOB d’afers internationals 191, 203 (2014); Deniz Aksoy, David Carter, & Joseph Wright, Terrorism and the Fate of Dictators, 67 World Pol. 423, 438 (2015); Milan Svolik, Which Democracies Will Last? Coups, Incumbent Takeovers, and the Dynamic of Democratic Consolidation, 45 Brit. J. Pol. Sci. 715 (2014); Bermeo, supra note 32, at 6. See also Pérez-Liñán, supra note 27, at 7 (the idea of a “legislative coup” for Honduras in 2009).

43

Monty Marshall & Donna Marshall, Coup d’État Events, 1946–2015: Codebook 1 (May 11, 2016), http://www.systemicpeace.org/inscr/CSPCoupsCodebook2015.pdf; Marsteintredet & Malamud, supra note 16, at 10.

44

Steven Barracca, Military Coups in the Post-Cold War Era: Pakistan, Ecuador and Venezuela, 28 Third World Q. 137, 138 (2007). Cf. also Tansey, supra note 15, at 151, 146.

45

Nikolay Marinov & Hein Goemans, Coups and Democracy, 44 Brit. J. Pol. Sci. 799, 801 (2014).

46

Marsteintredet & Malamud, supra note 16.

47

Javier El-Hage, Under What Circumstances May the OAS Apply the Democracy Clause against a Member State? 4 (Hum. Rts. Found. General Counsel Working Paper on International Democracy Law, Mar. 2010), https://ssrn.com/abstract=2480086. As a factor for a coup, El-Hage identifies the “violation of the constitutional procedure to remove a president.” Yet, this may simply undergird the illegality aspect of the removal rather than the reach the PLO. Of course, a partial break of the PLO does not suffice to speak of a breakdown.

48

Jonathan Powell & Clayton Thyne, Global Instances of Coups from 1950 to 2010: A New Dataset, 48 J. Peace Res. 249, 250 (2011).

49

E-mail from Jack Balkin to Michael Green (June 16, 2004, 04:14 EST) (on file with the North Carolina Law Review), quoted in Green, supra note 33, at 405. See also Bush v. Gore, 531 U.S. 98 (2000).

50

For the complexity of legality as the main component of the tactics dimension besides “logistic” (speediness and suddenness), or “mechanics” (abandonment of the post or resignation) in the coup definition, see the Online Appendix.

51

Marsteintredet & Malamud, supra note 16, at 9.

52

Id. at 2 (emphasis added); Pérez-Liñán, supra note 27, at 7.

53

Marinov & Goemans, supra note 45, at 801. “Seizure” is a little less one-sided than “unseating,” since, despite being access-prone, it better conveys the ideas of “exit from” and “access to” the executive power. Although “unscheduled change” of the executive is even more symmetrical, I will opt for (anomalous) seizure. If exit and access are legal, no coup is possible; and it suffices that either exit or access is illegal to speak of illegal tactics.

54

Id. at 809.

55

Jens Bartelson, Making Exceptions: Some Remarks on the Concept of Coup d’État and Its History, 25 Pol. Theory 323, 324 (1997).

56

Marinov & Goemans, supra note 45, at 808.

57

Pitts et al., supra note 18, at 334.

58

Id.; Martínez, supra note 42, at 209.

59

Aksoy, Carter, & Wright, supra note 42, at 427, 428.

60

Marsteintredet & Malamud, supra note 16, at 10.

61

Marsteintredet and Malamud understand “regime” once as “rules”, and once as involving “all key state institutions.” Id. at 13.

62

Aksoy, Carter, & Wright, supra note 42, at 423; Aurel Croissant, Coups and Post-Coup Politics in South-East Asia and the Pacific: Conceptual and Comparative Perspectives, 67 Austl. J. Int’l Aff. 264, 274 (2013). Lehoucq assumes that the Bolivian crisis of 2019 disembogued in a “regime collapse,” see Lehoucq, supra note 3, at 140. Actually, the democratic regime “collapsed” many years before the ousting of Morales, and what Lehoucq probably means is that a governing elite was removed, which is different from removing a regime understood as a system of rules.

63

See Croissant, supra note 62, at 274 (mentioning “junta”).

64

Marsteintredet & Malamud, supra note 16, at 9.

65

Id. at 11; Pedro Nikken, Analysis of the Basic Conceptual Definitions for the Application of Mechanisms for the Collective Defence of Democracy Provided for in the Inter-American Democratic Charter, in Collective Defence of Democracy: Concepts and Procedures 29, 31 (Carlos Ayala Corao & Pedro Nikken eds., 2006) (arguing that the “interruption of the democratic order”—in Inter-American Democratic Charter (IADC) vocabulary— must not be “total,” but it suffices that the “regime has lost. . . the quality of being democratic.” Id. at 72. This might be true for “interruption” in IADC jargon, yet coup means a complete breakdown of the PLO, otherwise one has erosion that, even if severe, falls short of a regime collapse). See also Inter-American Democratic Charter, Sept. 11, 2001, www.oas.org/en/democratic-charter/pdf/demcharter_en.pdf.

66

Marsteintredet & Malamud, supra note 16, at 14.

67

Id. at 10.

68

But see Dexter Boniface, The OAS’s Mixed Record, inPromoting Democracy in the Americas 40, 43 (Thomas Legler, Sharon Lean, & Dexter Boniface eds., 2007). Boniface cautions against the “blending of illegality and constitutionality” in so-called “impeachment coups.”

69

Aníbal Pérez-Liñán & John Polga-Hecimovich, Explaining Military Coups and Impeachments in Latin America, 24 Democratization 839, 839 (2017).

70

Huntington, supra note 42, at 201; George Derpanopoulos, Erica Frantz, Barbara Geddes, & Joseph Wright, Are Coups Good for Democracy?, 3 Res. & Pol. 1, 1 (2016). Leiv Marsteintredet & Einar Bernzten, Reducing the Perils of Presidentialism in Latin America through Presidential Interruptions, 41 Comp. Pol. 83, 88 (2008).

71

Aksoy, Carter, & Wright, supra note 42, at 423. For Kotowski, too, a coup is only about removing officials and not about changing the political structure. Kotowski, supra note 40, at 229.

72

But see Marsteinredet & Malamud, supra note 16, at 10 (using the terms “victim” and “target” synonymously, which may mean “the state,” besides “government” or “head of government”).

73

Marinov & Goemans, supra note 45, at 808.

74

Marsteintredet & Malamud, supra note 16, at 10.

75

Luttwak speaks of “palace revolution” and Gandhi of “palace putsch.” Luttwak,supra note 42, at 18; Jennifer Gandhi, Political Institutions under Dictatorship 7 (2008). Without this distinction, it is not surprising that scholars code the transfers of power in Ecuador in 2000 and Venezuela in 2002 as “coups” despite the fact of no breakdown (see the Online Appendix for the figure of “façade impeachment” as equivalent to palace revolution).

76

Barbara Geddes, Joseph Wright, & Erica Frantz, Autocratic Breakdown and Regime Transitions: A New Data Set, 12 Persp. on Pol. 313, 314, 316 (2014).

77

Aaron Belkin & Evan Schofer, Coup Risk, Counterbalancing, and International Conflict, 14 Sec. Stud. 140, 144 (2005) (emphasis added).

78

Marsteintredet & Malamud, supra note 16, at 10.

79

Kotowski, supra note 40, at 215.

80

Geddes, Wright, & Frantz, supra note 76, at 316.

81

Gabriel Naudé, Considérations politiques sur les coups d’état 108 (1712).

82

Id. at 77; Martinez, supra note 42.

83

Powell & Thyne, supra note 48, at 252.

84

A similar effect has been analyzed by Hochstetler and Samuels, who consider presidential interruptions as “quasiparliamentary procedures. . . that. . . allow the system to reequilibrate.” Hochstetler & Samuels, supra note 26, at 133.

85

See Katrin Hochstetler, The Fates of Presidents in Post-Transition Latin America: From Democratic Breakdown to Impeachment to Presidential Breakdown, 3 J. Pol. in Latin Am. 125, 139 (2011); Powell & Thyne, supra note 48, at 258. For the role of “hegemonic presidents,” see Pérez-Liñán, supra note 27, at 2. On measuring “president’s degree of undemocratic behavior” in relation to interruption, see Marsteintredet, supra note 10, at 182. For cases of alleged salutary interruptions of mandates, for example Alberto Fujimori in 1992, Joaquín Balaguer in the Dominican Republic in 1994/1996, Raúl Cubas in Paraguay 1999, and Lucio Gutiérrez in 2005 in Ecuador, see Marsteintredet, supra note 10, at 182, 188.

86

Milan Svolik, The Politics of Authoritarian Rule 4 (2012); Marsteintredet & Malamud, supra note 16, at 9; Aksoy, Carter, & Wright, supra note 42, at 429.

87

Gandhi, supra note 75, at 181.

88

Id. at 20, 35.

89

Geddes, Wright, & Frantz, supra note 76.

90

Svolik, supra note 42, at 4.

91

Some see in the combination of mass protests and leadership turnover not a coup but a “popular revolution.” See Tansey, supra note 16, at 153.

92

Hochstetler & Samuels, supra note 26.

93

Marsteintredet & Malamud, supra note 16, at 13.

94

Aníbal Pérez-Liñán, Instituciones, coaliciones callejeras e inestabilidad política: Perspectivas teóricas sobre las crisis presidenciales, 49 América Latina Hoy 105, 112 (2008); Wolff, supra note 3, at 177 (suggesting this approach be used in the Bolivian case).

95

Maeda declares the military as one “exogenous” source of democratic breakdown, which is misleading, since the military, while being outside the “government,” is inside the state apparatus and thus not extra-state. See Ko Maeda, Two Modes of Democratic Breakdown: A Competing Risks Analysis of Democratic Durability, 72 J. Pol. 1129, 1130 (2010).

96

Other authors consider a coup d’état as given when the victims are heads of branches other than the executive. See, e.g., Ángel Orellana, Golpe de Estado en Honduras: Un análisis jurídico, 93 Boletín Especial Centro de Documentación de Honduras 2, 6 (2009). See alsoPérez-Liñán, supra note 26, at 8.

97

Alvarez & Marsteintredet, supra note 23, at 34; Pérez-Liñán, supra note 94, at 108. Hochstetler and Samuels advise to expand the gamut of interruption possibilities to understand cases that “fall somewhere between total regime breakdown. . . and true presidential breakdown where the democratic regime is largely unscathed.” Hochstetler & Samuels, supra note 26, at 129.

98

Alvarez & Marsteintredet, supra note 23, at 34.

99

Hochstetler & Samuels, supra note 26, at 129.

100

Id. at 144; Svolik, supra note 42, at 730 (considering “popular uprising” as the fourth way to democracy breakdown, alongside coup, incumbent takeover, and civil war).

101

Mariana Llanos & Leiv Marsteintredet, Ruptura y continuidad: La caída de “Mel” Zelaya en perspectiva comparada, 55 América Latina Hoy 173, 175 (2010).

102

Barracca, supra note 44, at 138.

103

Id.

104

Marsteintredet & Malamud, supra note 16, at 13. In the family-resemblance approach to the construction of a concept, the principle of the classical model that all features of the concept are necessary and jointly sufficient for the definition of the subject in question does not apply (id. at 11). Marsteintredet and Malamud complain of a proliferation of “coup with adjectives,” in which the coup d’état is “frequently combined with a qualifying adjective” (id. at 2), this time as the result of changing “the concept structure of coups from classical to a family resemblance type” (id. at 8).

105

Moore, supra note 39, at 126.

106

Ozan Varol, The Democratic Coup d’État, 53 Harv. Int’l L.J. 219 (2012). Critically, Tansey, supra note 16, at 144.

107

Derpanopoulos et al., supra note 70, at 2.

108

The OAS secretary general concluded in an official briefing that Morales himself perpetrated a coup by rigging the elections of 2019. See Durante el consejo permanente extraordinario convocado para estudiar la situación en Bolivia, Org. Am. States (Nov. 12, 2019), www.oas.org/es/acerca/discurso_secretario_general.asp?sCodigo=19-0064.

109

Marsteintredet & Malamud, supra note 16, at 13. For incumbent takeover, see Svolik, supra note 42, at 730.

110

Maeda, supra note 95, at 1130 (describing the case of “democratically elected leaders” that “ended the democratic process themselves”). A conceptual precedent was continuismo as explained by Rapoport, supra note 39, at 63.

111

Schmitter Philippe, Dangers and Dilemmas of Democracy, 5 J. Democracy 57, 59 (1994).

112

Maeda, supra note 95, at 1130; Svolik, supra note 42, at 715.

113

Marsteintredet & Malamud, supra note 16, at 13.

114

SeePérez-Liñán, supra note 26, at 50; El-Hage, supra note 47, at 7.

115

Sampford, supra note 41, at 196.

116

Powell & Thyne, supra note 48, at 250.

117

Chew, supra note 28, at 604; Octavio Moreno & Carlos Figueroa, Golpismo y neogolpismo en América Latina, 3 Iberoamerica Social 98, 98 (2019); Lorena Soler, Golpes de Estado en el siglo XXI: Un ejercicio comparado Haiti (2004), Honduras (2009) y Paraguay (2012), 14 Cuadernos Prolam 77, 81 (2015).

118

The laxity is evident when neo-coup thinkers not only describe the ousting of Rousseff as a “parliamentary coup,” but characterize the social protests that accompanied her overthrow as “fascist.” Avelar describes this approach as “sloppy partisan political science.” Idelber Avelar, A response to Fabiano Santos and Fernando Guarneri, 26 J. Latin Am. Cultural Stud. 341, 348 (2017).

119

To be sure, some authors limit coup to an action carried out by the military, police, and/or security forces. SeeSingh,supra note 42, at 51; Lehoucq & Pérez-Liñán, supra note 42, at 1109; Svolik, supra note 42, at 730. But see Powell & Thyne, supra note 48, at 251. See also Marinov & Goemans, supra note 45, at 801 (declaring that the military is not a necessary component of a coup).

120

But see Pérez-Liñán & Polga-Hecimovich, supra note 69, at 851 (analyzing the case of Rousseff in Brazil).

121

See David Almagro, ¿Juicio legítimo o golpe de Estado encubierto?, 42 Revista Derecho del Estado 25, 47 (2019) (questioning the legality in Rousseff’s case). But see Pérez-Liñán, supra note 27, at 3 (speaking of Rousseff’s impeachment as “quite respectful of procedure”).

122

See Moreno & Figueroa, supra note 117, at 114 (denouncing the attack on progressive governments as the main target of soft coups). Contra Pérez-Liñán, supra note 27, at 7 (alerting against “blurring moral distinctions” between impeachment (legal) and coup (non-legal)).

123

Soler, supra note 117, at 78; Moreno & Figueroa, supra note 117, at 114.

124

Id. at 113; but see Pérez-Liñán, supra note 27, at 7.

125

Moreno & Figueroa, supra note 117, at 113. But see Pérez-Liñán, supra note 27, at 11.

126

Soler, supra note 117, at 81.

127

Moreno & Figueroa, supra note 117, at 109; Soler, supra note 117, at 81.

128

Chew, supra note 28, at 605. See also Moreno & Figueroa, supra note 117, at 99 (on Ecuador in 2010 and Haiti in 2005). See similarly Soler, supra note 117, at 80.

129

Soler, supra note 117, at 80.

130

But see Id.

131

Guillermo O’Donnell, Transitions, Continuities, and Paradoxes, inIssues in Democratic Consolidation 17, 19 (Scott Mainwaring, Guillermo O’Donnell, & Samuel Valenzuela eds., 1992).

132

See also Steven Levitsky & James Loxton, Populism and Competitive Authoritarianism in the Andes, 20 Democratization 107, 115 (2013).

133

See recently Steven Levitsky & Lucan Way, The Myth of Democratic Recession, 26 J. Democracy 47 (2015).

134

Reverting decay comes close to the idea of (early) “re-equilibration” for cases of severe regime crisis, reverted before regime breakdown. SeePérez-Liñán, supra note 26, at 47.

135

Adam Przeworski, Latin American Political Regimes in Comparative Perspectives, inRoutledge Handbook of Latin American Politics 542, 546 (Peter Kingstone & Deborah Yashar eds., 2011).

136

Guillermo O’Donnell, Delegative Democracy, 5 J. Democracy 55, 56 (1994).

137

Ariel Armony & Hector Schamis, Babel in Democratization Studies, 16 J. Democracy 113, 125 (2005).

138

Kelsen, supra note 34, at 118.

139

O’Donnell, supra note 136, at 19.

140

Marsteintredet & Malamud, supra note 16, at 9.

141

Marsteintredet & Bernzten, supra note 70, at 92, point out that because impeachments are difficult societies have tried other ways for “removing unpopular presidents.”

142

Ramiro Sanchez & Alberto Escamilla, La interrupción del mandato presidencial en América Latina, 13 Polis 47, 69 (2017), Llanos & Marsteintredet, supra note 101, at 184.

143

As Wolff, supra note 3, at 177, states it is about processes that constitute “within-regime change, not regime breakdown.” This corresponds to Pérez-Liñán’s “presidential crisis” with “removal of the president” but without breakdown: seePérez-Liñán, supra note 26, at 49.

144

Hochstetler & Samuels, supra note 26, at 130 (emphasis added), state “large protest movements. . . can directly force presidents to resign.” Congress, then, is “marginal,” notes Pérez-Liñán, supra note 26, at 10. Pion-Berlin & Trinkunas, supra note 13, at 398, mention “at times street protests alone have single handedly evicted a president.”

145

Non-state nature does not necessarily mean grassroots movements. Presidential interruptions can draw on “lower-class” (Ecuador, 2000), be “broad-based” (Pakistan, 1999), or draw on “upper and middle class” groupings (Venezuela, 2002). See Barracca, supra note 44, at 143.

146

Zamosc, supra note 9, at 237; Pion-Berlin & Trinkunas, supra note 13, at 395 (preferring to talk of “civilian coup” when not the military but “mass mobilization” displaces a ruler). However, the term “civilian coup” does not tell whether a breakdown of the PLO has necessarily occurred, overlooks that coup implies intra-state actors, and assumes that any illegal removal has to be a coup. But see Table 1 for the category of pseudo-impeachment, or popular pseudo-coup if a breakdown occurs, although legally.

147

Hochstetler & Samuels, supra note 26, at 129; Pérez-Liñán, supra note 26, at 46.

148

Id. at 204; Zamosc, supra note 9, at 262.

149

Lehoucq, supra note 3, at 134. For a typology of extra-state actors, see Alvarez & Marsteintredet, supra note 23, at 34.

150

Marsteintredet & Bernzten, supra note 70, at 93; Pérez-Liñán, supra note 26, at 206. In the Bolivian case of 2003, counting mobilizations as perpetrators does not preclude the military from playing a role, not by military quartering or rebelling, but by repressing. See Pion-Berlin & Trinkunas, supra note 13, at 404.

151

Arugay, supra note 9, at 179 (showing the normal feature of the perpetrator status in the classical impeachment, “without civil society participation”). However, that in a classical impeachment the perpetrator is embodied in an intra-state actor holds for the factor of the perpetrator, otherwise this could be misleading, since even impeachment may involve civil society participation.

152

Moreno & Figueroa, supra note 117, at 109; Pérez-Liñán & Polga-Hecimovich, supra note 69, at 842. The largest protests in Brazilian history took place “demanding Rousseff’s ousting” even before the political class began campaigning for impeachment. See Avelar, supra note 118, at 346.

153

Pérez-Liñán, supra note 26, at 183.

154

Boniface, supra note 68, at 43. Pseudo impeachment is close to Boniface’s idea of an “impeachment coup,” which is an illegal act of unseating an elected president (even if the successor is constitutional). However, as Boniface himself concedes, his idea does not imply the breakdown of the regime, hence the term coup is not felicitous. For Honduras, see Omar Huertas & Victor Cáceres, Los golpes de Estado constitucionales en Latinoamérica: Una amenaza emergente para el principio democrático, 10 Justicia Juris 28, 31 (2014).

155

Id., at 32.

156

Pérez-Liñán, supra note 26, at 3. Pérez-Liñán advises addressing the distorted use of impeachment instead of calling it a coup, without ignoring the risks of manipulated impeachment. Id. at 11.

157

Pérez-Liñán & Polga-Hecimovich, supra note 69, at 840.

158

Pérez-Liñán, supra note 27, at 7.

159

But see Marsteintredet, supra note 10, at 17.

160

But see Detlef Nolte, Leiv Marsteintredet & Mariana Llanos, Paraguay and the Politics of Impeachment, 24 J. Democracy 110, 118 (2013).

161

Moreno & Figueroa, supra note 115, at 110; but see the Online Appendix.

162

El-Hage, supra note 47, at 11. El-Hage argues that when erosion is “systematic,” it may trigger OAS sanctions leading to a “corrective” towards “restoration of democracy.” It is, however, important to distinguish restoration of regime from “restoration of democracy.” The former implies the chance of reinstating a long-broken PLO (in the wake of an anomalous seizure of executive power) which then leads to regime restoration no matter if the regime, whose PLO has been affected, is democratic or not. In contrast, El Hage’s concept of “restoration of democracy” is based either (i) on an “interruption” of (only) democracy, which must not imply a regime breakdown but its severe deterioration (id. at 72); or (ii) on a presidential interruption, in which democracy as a regime was not actually broken, but some of its processes were briefly disrupted and restored within hours or a few days, as in Ecuador in 2000 and Venezuela in 2002.

163

Tribunal Constitucional Plurinacional [TCP] [Constitutional Tribunal], Apr. 25, 2013, Declaracion constitutional 003/2013 (Bol.).

164

Ley del Regimen Electoral, June 30, 2010, art. 15 (Bol.).

165

Tribunal Constitucional Plurinacional [TCP] [Constitutional Tribunal], Nov. 28, 2017, Sentencia 0084/17 (Bol.).

166

American Convention on Human Rights, Nov. 22, 1969, in force July 18, 1978, 1144 U.N.T.S. 123 [hereinafter ACHR].

167

Levitsky, supra note 29.

168

Raul Peñaranda, La pacificación del país se operó tras bambalinas, inLa Revolución de las Pititas 235, 237 (Mary Vaca ed., 2019).

169

Evo Morales, Volveremos y seremos millones 43–9 (2020).

170

Ley 1405 Orgánica de las FFAA, Dec. 30, 1992, art. 20/b (Bol.).

171

Rittinger & Cleary, supra note 42, at 409.

172

This matches the trend of post-Cold War subordination of the military under civilian power in Latin America. See Pion-Berlin & Trinkunas, supra note 13, at 397; Pérez-Liñán, supra note 26, at 203; Marsteintredet & Bernzten, supra note 70, at 91; Rittinger & Cleary, supra note 42, at 407. The military’s actions or inaction between November 7 and 12, 2019 should also be seen in light of the army’s constitutional duties to guarantee national integrity. While it is true that “military quartering” during times of citizen revolt is “serious” (see Pion-Berlin & Trinkunas, supra note 13, at 397, 398), I believe this view is incomplete without considering these duties within a broader constitutional frame. For instance, a constitution may regulate the right of civil resistance against a wrongful ruler, which could imply an exceptional reversal of the normal chain of command between civil and military power, leading the latter to at least not repress civil mobilization, while still adhering to the constitutionality of this kind of state of emergency.

173

Lehoucq & Pérez-Liñán, supra note 42, at 1122.

174

Powell & Thyne, supra note 48; but seePérez-Liñán, supra note 26, at 212.

175

Marsteintredet & Malamud, supra note 16, at 14.

176

Pérez-Liñán, supra note 26, at 211; Arugay, supra note 9, at 167.

177

Kathryn Hochstetler, Rethinking Presidentialism: Challenges and Presidential Falls in South America, 38 Comp. Pol. 401, 411 (2006) (stating that “most of the millions of street protesters over the decades marched peacefully,” though adding that “incivility” was also a “regular part of civil society mobilization”). But see Pion-Berlin & Trinkunas, supra note 13, at 398 (equating “street politics” with non-“democratic channels,” whereas street politics can actually be legal, constitutional, and democracy-restoring). To be clear: to assert that a resignation forced by the streets may still be legal does not mean endorsing the idea that “protests should trump votes.” SeePérez-Liñánsupra note 26, at 211; Pérez-Liñán, supra note 27, at 3 (skeptical about the legality of popular protests leading to interruption).

178

Marsteintredet & Malamud, supra note 16, at 9.

179

Powell & Thyne, supra note 48; Aksoy, Carter, & Wright, supra note 42, at 423.

180

Lehoucq, supra note 3, at 140.

181

Varol, supra note 106, at 297. See Croissant, supra note 62, at 265, who, in a crisis, distinguishes between military defection as an expression of military discontent and military coup. On the frequency of the military’s passivity in Latin America, see Pion-Berlin & Trinkunas, supra note 13, at 395.

182

The passivity of the Bolivian military is consistent with the trend observed since 2000 that the militaries in Latin America “have chosen to stay quartered more often than not in the face of civilian uprisings.” Id. at 408.

183

On the difficulties for knowing when the military has or has not coerced a president, see Tansey, supra note 16, at 153.

184

Marsteintredet, supra note 10, at 182 (considering as an extreme case of violation a president’s illegal attempt to extend terms limits, and suggesting that the removal of such a president could be better for democracy).

185

Wolff, supra note 3, at 177. But see Lehoucq, supra note at 3, at 141 (asserting that the succession was “constitutionally suspect” because of the “absence of legislative quorum” at the moment Añez assumed the presidency). While this process of succession was not explicitly regulated by the constitution in the case of a throughout resignation of the officers in the chain of succession, the succession was eventually constitutional and followed a model of what has been called a “rule-based succession.” See Marsteintredet & Bernzten, supra note 70, at 94; Zamosc supra note 9, at 237. It so happens that nowhere in article 40(a) of the standing rules of the Bolivian senate (Camara de Senadores-Reglamento General as modified Feb. 6, 2012), which regulates the automatic replacement of an absent authority, is the summoning of the chambers stipulated as a condition, nor is it stated that the replacing person must belong to the majority faction. Thus, after the Senate president and the first vice president resigned in those hours, that body was automatically headed by Añez, then second vice president of the Senate, moving her up the chain of succession. Moreover, by leaving the country, Morales unwittingly overrode the resignation clause with the absenteeism clause. According to the Bolivian constitution, absenteeism also triggers succession, again, without the need of a quorum of whatever assembly.

186

Ley de régimen excepcional y transitorio para la realización de elecciones generales no. 1266, art. 19/II (Nov. 24, 2019).

187

See Pion-Berlin & Trinkunas, supra note 13, at 395 (ambivalent about constitutionality, assuming a tendency toward “nonconstitutional transfers of power” by “mass action of civilian sectors” who carry out a “civilian coup” to change “governments outside the normal democratic processes established by the constitution”).

188

See Charter of the Organization of American States, art. 9, Apr. 30, 1948, art. 9, http://www.oas.org/dil/1948%20charter%20of%20the%20organization%20of%20american%20states.pdf (in addition to over-emphasizing the “elected” nature, it focuses on “overthrowing” and leaves the issue of incumbent takeover beyond scope). See also Nikken, supra note 65, at 33.

189

Cf. Yatama v. Nicaragua, Inter-Am. Ct. H.R. (ser. C) No. 27, ¶ 23 (June 23, 2005).

190

Boniface, supra note 68, at 45.

191

Wolff, supra note 3, at 177.

192

See Hochstetler & Samuels, supra note 26, at 133; Nolte, Marsteintredet & Llanos, supra note 160, at 120; but see Pion-Berlin & Trinkunas, supra note 13, at 409 (rather rigid on this point, underscoring that “the public must know that their votes count”).