A Constitution I Am Used To? Constitutional Endurance and Replacement in Democratic Latin America ^*

Abstract

It is well known that during times of social unrest, replacing the constitution is one of the first proposals to emerge, at least in Latin America, as if this constitutional shift could resolve any problem. Constitutional substitution is not a new phenomenon in Latin America; it actually began after the region’s redemocratization. However, it is striking that, despite similar conditions of social unrest, many other countries in the region have not undergone such an encompassing constitutional reform. Our main focus here is to analyze the persistence of constitutions in Latin American countries. We thus ask: What conditions lead democratic countries to keep their constitutions? Using a configurational approach, our preliminary results are promising. Together, four pathways explain why some countries have not replaced their constitutions in democratic contexts, with special emphasis on the number of rights enshrined in these fundamental laws.

Constitution-making; written constitutions; political institutions; Latin America; QCA

Constitutional replacement is not a new phenomenon in Latin America. Since the third wave of democratization, when the military regimes fell in the region, almost all countries (14 out of 18) have replaced their constitutions at least once. The only exceptions are Chile ¹ , Colombia, Mexico, and Panama.

Indeed, studies have shown that regime change is an important predictor for constitutional replacement (NEGRETTO, 2012, 2008). Still, some countries have rewritten their constitutions even after they had already reformed their authoritarian constitutions. For instance, Bolivia, the Dominican Republic, Ecuador, and Venezuela have all undergone this process. On the other hand, there are cases where countries facing similar contexts and trends did not seek to change their constitutions. Here, we focus on this second set of cases – our goal is to understand why these countries did not see the need to change their constitutions. Hence, the central research question guiding our study is: What conditions lead to constitutional preservation?

There is a significant body of research that has already addressed the opposite research subject, which is constitutional change (ELKINS, GINSBURG and MELTON, 2009; MONTERO, 2017; NEGRETTO, 2012, 2008). However, we assume that despite similar trends and contexts there likely exists a particular configurational condition that hinders, or prevents the political inclination for constitutional change. Unlike previous studies that focused on measuring the average effect of individual conditions on constitutional survival, we are interested in examining how various conditions interact to maintain the constitution, even after regime transitions in Latin American countries. Thus, for our purpose, the Qualitative Comparative Analysis (QCA) appears to be the most appropriate methodological approach. To do so, we will analyze 17 Latin American countries to compare cases that have undergone similar democratization processes and to test the theory formulated on this matter.

The rest of the article is structured as follows. In the next section, we provide the theoretical framework for constitutional replacement, focusing on previous studies of Latin American cases. This section will offer valuable insights for theorizing the opposite outcome, i.e. constitutional survival, and establish our hypotheses. Then, we present the research design using QCA and explain how the cases were selected in the third section. After that, we outline the calibration process of our conditions. In the fifth section, we unveil the results by analyzing the necessary and sufficient conditions for explaining constitutional endurance, which is followed by the discussion of the results in light of illustrative cases in the sixth section. We conclude the article by discussing some intriguing outcomes.

Theoretical framework

When examining political institutions, constitutions serve as the fundamental source of institutional design, dictating the political system, government structure, administrative organization, and social and civil rights. As the supreme law of a political system, constitutions, however, are not immutable – even the Constitution of the United States, renowned for its endurance, has undergone amendments since its inception.

As the political, social, and economic landscape evolves over time, the higher law must also adapt (NEGRETTO, 2012). Typically, these revisions occur through textual alterations, such as amendments or even the drafting of a new constitution. Changes can also be brought about through judicial interpretation by the supreme courts, where the constitutional text itself remains unchanged but its interpretation is revised.

Negretto (2012) highlights that the decision for constitutional change involves a two-step process. First, it is necessary to determine whether there is a demand for constitutional change. If the answer is affirmative, the second step involves selecting the mechanism for change, whether through amendment or rewriting the constitution. While an amendment alters specific sections of a constitution, often detailing a process or extending the original text, a new constitution represents a complete shift in the state’s structure. Consequently, the latter process is much costlier than the former, as it involves time- and energy-consuming negotiations among key political players (ELKINS, GINSBURG and MELTON, 2009). Thus, the risks associated with constitutional replacements are higher than those accompanying reforms through amendments and/or judicial review.

Of note is that the distribution of power between institutions is indicative of stability, thereby shaping political environments that influence the survival of the rules of the game (PRZEWORSKI, 1999). Whether the distribution of power is inclusive or exclusive can affect politicians’ decision to revise the constitution (MONTERO, 2017; NEGRETTO, 2012; 2008).

According to the rational choice theory, a power-concentrating institution creates two positions for political players: winners and losers. Winners naturally concentrate more power than losers. However, these positions can shift in the future, leading former winners to become disadvantaged. As a result, winners may seek an equilibrium in power-sharing, giving rise to a more pluralistic institutional setting. Notably, major institutional reforms often take place when drafting a new constitution. Exclusive institutions, in particular, tend to motivate political actors to replace the current constitutional design.

In this context, a frustrated executive with limited powers may call for a constitutional change. Consequently, even after the end of an authoritarian regime, a constitution may be rewritten if political players disapprove of the concentration of power outlined within it. Therefore,

H1. posits that: When the constitution stipulates high power-sharing among the executive and legislative branches, the constitution tends to remain in place.

Regarding the mechanisms of constitutional reform, there are three avenues: replacement, amendment, or judicial revision. Among these, replacement is the most radical approach, and politicians usually prefer more convenient methods to reform their constitutions. However, in cases where making textual alterations through amendments is not feasible due to high thresholds or when the supreme court lacks sufficient independence to interpret the constitutional text, pressure for a new constitution that aligns with current demands remains constant. In contrast, if a constitution is regularly adapted by amendment or judicial review in response to political or social claims, the demand for a new constitution becomes weaker (MONTERO, 2017; NEGRETTO, 2012; 2008).

Notably, judicial review is the only reform instrument that does not require changes to the constitutional text (MONTERO, 2017; NEGRETTO, 2008). In a political system where constitutional amendment through the legislative process faces high hurdles and the judiciary lacks sufficient independence, the likelihood of constitutional replacement increases (ELKINS, GINSBURG and MELTON, 2009). Thus,

H2. posits: If a constitution is easily amended, then it is more likely to survive.

Furthermore, institutions are shaped over time, and their endurance depends on their adaptability. This principle also applies to constitutions. If a constitution stands the test of time, the likelihood of its replacement diminishes. As Elkins, Ginsburg, and Melton (2009, p. 90) emphasize, when a constitution becomes ingrained in the national culture and politics, its consensual nature and enforceability increase.

Conversely, in countries with a history of frequent constitutional enactments, indicating past political instability, a distinct pattern is found among political actors (NEGRETTO, 2008). In other words, when a constitution fails to address social, political, or economic demands, legislators, based on past experience, are less inclined to defend the original text. In such circumstances, constitutional change becomes less costly, and political actors may call for it. Thus,

H3 . posits that: The absence of a ‘culture’ of constitutional change affects the willingness and ability to pursue further constitutional changes.

Moreover, while Elkins, Ginsburg, and Melton (2009) primarily focus on the gains and losses of political actors during the bargaining process of constitutional maintenance or reformulation, there is a significant element that scholars have only superficially explored: the social demand for constitutional reform. As mentioned earlier, constitutions are rewritten in response to political, economic, and social demands. Consequently, politicians and legislators may propose not only amendments but also complete constitutional overhauls in response to these demands.

In this context, studies on social movements and public policies suggest that public mobilization through civil society, protests, or public demands plays a crucial role in instigating change (BAUMGARTNER and MAHONEY, 2005; MEYER, 2005). Mass mobilization effectively influences agenda-setting, as ordinary citizens can loudly express their frustrations, needs, and desires. In light of this, the Chilean protests that culminated in a constitutional referendum provide evidence that social mobilizations, combined with the aforementioned conditions, drive the enactment of a new constitution. Thus, we propose

H4. The absence of continuous social mobilization constitutes a sufficient condition for the survival of the constitution.

Additionally, there is a well-established tradition in Latin American politics of the 21st century to expand the scope of human rights listed in national constitutions (ELKINS, 2017). Elkins, Ginsburg, and Simmons (2013) demonstrate that the inclusion of new rights and the preservation of existing ones is a global trend that emerged after World War II, following the Universal Declaration of Human Rights. As a result, the number of rights included in national constitutions has increased over the years. In the context of Latin American countries, most national constitutions include a greater number of rights compared to the global mean (ELKINS, GINSBURG and SIMMONS, 2013). This regional diffusion of rights cannot be overlooked. Given the tradition of expanding the number of rights in Latin American constitutionalism (MARQUARDT, 2010) and the overall upward trend of including rights in national constitutions, we propose

H5. Constitutions with an increased number of rights are more likely to survive.

Finally, let us consider the post-World War II context, wherein seven out of ten countries transitioning to democracy did so under a constitution written by an authoritarian regime. Some countries replaced the constitutional text shortly thereafter, while others did not (ALBERTUS and MENALDO, 2018). The authors argue that this disparity is due to the effective control exerted by authoritarian elites during the transition process.

When a new constitution is not promptly enacted, public demands may intensify over time, as exemplified by Chile, where social movements and protests have long called for the inclusion of social issues in the country’s Constitution (PISCOPO and SIAVELIS, 2021). Although Chileans frequently pursued reform initiatives through amendments, Pinochet's Constitution, written during the dictatorship, no longer reflected the aspirations of the Chilean society (VERDUGO and PRIETO, 2021). Similarly, other Latin American countries, such as Bolivia, Colombia, and Ecuador, maintained constitutions written by dictators for a time after the authoritarian period. Based on these historical observations, we propose

H6. A democratic starting point constitutes a strong condition for the legitimacy and survival of a constitution.

Research design: QCA and case selection

To test our hypotheses, we employ a Crisp-Set Qualitative Comparative Analysis (henceforth, csQCA) to uncover the conditions under which Latin American democratic countries either maintain or replace their constitutions. In so doing, our approach differs from the existing literature in two significant ways. First of all, we focus on the storyline of constitutional endurance, specifically the non-replacement of constitutions, instead of predicting constitutional shifts, as most of the literature does. Secondly, most of the empirical scholarly work on constitution-making processes adopts a conventional statistical approach, aiming at measuring the net effects of concurrent variables on the likelihood of constitutional change within a given year and country (ELKINS, GINSBURG and MELTON, 2009; NEGRETTO, 2012; 2008). In contrast, we are more interested in investigating how different conditions interact with each other to account for constitutional persistence, without having to measure the magnitude of the effect of contrasting variables. Consequently, rather than adopting an ‘effects-of-cause’ approach, our study assumes a ‘cause-of-effects’ perspective, prioritizing the examination of specific cases (OANA, SCHNEIDER and THOMANN, 2021).

In simple terms, Qualitative Comparative Analysis (QCA) offers a distinctive perspective by examining the world through a set-theoretic lens. This means that QCA enables us to untangle complex causal relations between conditions and outcomes. Expressed in terms of asymmetry, conjunctural causation, and equifinality, QCA shows that explanations for an outcome may be substantially different from the explanation for the negation of that outcome (asymmetry). That certain conditions might be relevant only in conjunction with others (conjunctural causation) and that different combinations of conditions may lead to the very same outcome (equifinality) (RAGIN, 2008; SCHNEIDER and WAGEMANN, 2012).

In general, QCA is a comprehensive method comprising many variants (MELLO, 2022). Among these variants, csQCA ² is just one of the different ways to apply QCA. Notwithstanding the existence of several QCA variants, they all share the very same interpretation of causal relations, wherein conditions are categorized either as necessary or sufficient. A condition (or a configuration) is necessary when the outcome cannot occur in its absence, whereas it is deemed sufficient when it alone can yield the outcome.

The uniqueness of csQCA is that both conditions and outcomes must necessarily assume dichotomic values, denoting either the presence (01) or absence (0) in their respective sets. Therefore, unlike other QCA variants, conditions in csQCA are strictly limited to values 0 and 01, with no other values allowed. On the one hand, this implies that csQCA cannot take into account more fine-grained information since the method draws a relatively crude divide between the presence and absence of conditions by default (ALBALA and LEAL, 2022). On the other hand, csQCA enables a more straightforward calibration process and enhances research transparency by revealing the discretionary decisions made by researchers in more detail (OPPERMANN and BRUMMER, 2020).

Our analysis relies on a dataset that covers 17 Latin American countries, as presented in Table 01 . It is important to know that we are not conducting a general test of theory on constitution-making in our study. Instead, we intend to conduct a middle-range analysis. More specifically, our analysis is centered on democratic Latin American countries, both in terms of time and geography. Nevertheless, it is worth mentioning that the temporal coverage of our dataset varies across countries due to differences in the timing of their transition to democracy. For instance, Colombia and Venezuela returned to democratic normalcy by the late 1950s, while Panama and Paraguay transitioned away from their autocratic regimes only three decades later.

Table 01
Constitutional replacement in the post-authoritarian era in Latin America

Country	Transition Year	Constitutional Change Year
Argentina	1983	-
Bolivia	1982	2009
Brazil	1985	-
Chile	1990	- Soon?
Colombia	1958	1991
Costa Rica	1949	-
Ecuador*	1979	1993
El Salvador	1984	-
Guatemala	1986	-
Honduras	1982	-
Mexico	2000	-
Nicaragua	1984	-
Panama	1989	-
Paraguay	1989	-
Peru**	1980	-
Uruguay	1985	-
Venezuela	1959	1999

Albertus and Menaldo (2018)
• Note: Data on the transition year to democracy are sourced from Albertus and Menaldo (2018), whereas data on constitutional replacement come from Elkins, Ginsburg, and Melton (2009) and Negretto (2020).
• (*) Ecuador has replaced its constitution on five occasions under democratic rule, but we have considered only the first change – including the other Ecuadorian changes would introduce a bias into the analysis, as Ecuador would then represent more than 50% of the instances of constitutional replacement.
• (**) Peru replaced its constitution in 1993, but it was done from an autocratic perspective.

Still, one could arguably sneer at our research design, asking ‘Why focus so much on studying just one region?’ Indeed, comparative studies frequently suffer from a disappointing lack of justification for case selection (COLLIER, 1993; PETERS, 1998). However, within the context of our study, Latin America stands out as particularly compelling due to its abundance of constitution-making processes within democratic settings (LANDAU, 2019; NEGRETTO, 2020). This is illustrated by the fact that several countries have replaced their constitutions since the turn of the century, including Bolivia, the Dominican Republic, Ecuador, and possibly Chile in the near future. In addition, by focusing only on Latin America, we can control for another seemingly important condition: the diffusion of constitutional replacements across the region (NEGRETTO, 2012). Since we exclusively examine Latin America, the spread of constitution-making processes in neighboring countries remains consistent for all our cases. Similarly, our case selection also enables us to control for other unobservable conditions.

Ideally, our case selection should cover all Latin American countries, excluding only the small Caribbean nations. Nonetheless, we had to remove the Dominican Republic from our original dataset since the years leading up to the 1994 constitutional replacement could not be considered minimally democratic according to Polity 05 (MARSHALL and GURR, 2020). Hence, since all other cases were drawn from minimally democratic settings, we purposely excluded the Dominican Republic from our analysis to maintain the coherence of our case selection as much as possible. This way, we avoid comparing cases with fundamentally distinct characteristics (SARTORI, 1970).

Calibration

Calibration is the warp and woof of QCA analysis. Put simply, calibration is the process by which researchers input theoretical and grounded knowledge into mathematical measurements of theoretical concepts (RAGIN, 2008; SCHNEIDER and WAGEMANN, 2012). In so doing, QCA can match concepts with close empirical indicators, thus avoiding the problem of having an analysis in which theory is disconnected from the empirical analysis (GOERTZ, 2020). Table 02 outlines the calibration process.

Table 02
Calibration process

Sets	Hypotheses/Question	Calibration	Source
Power-Sharing (POS)	Constitutions endure where there are power-sharing institutions	1 = POS > 0.742 0 = POS < 0.742	Coppedge et al. (2020)
Amendment (AME)	Constitutions endure where there exist minor hurdles to amending the current constitution	1 = High Degree of Judicial Review and Low Majority to pass a constitutional amendment in the legislature 0 = Low Degree of Judicial Review or High Majority to pass a constitutional amendment in the legislature	Negretto (2012); Comparative Constitutions Project (2021)
Legacy of Replacement (LEG)	Constitutions endure in countries with a small legacy of replacing constitutions	1 = LEG > 13 0 = LEG < 13	Comparative Constitutions Project (2021)
Continuous Mobilization (MOB)	The lack of unrelenting large social mobilizations is linked to constitutional endurance	1 = MOB > 1.86 0 = MOB < 1.86	Coppedge et al. (2020)
Number of Rights (RG)	Constitutions survive if they have enshrined a substantial number of rights	1 = RIG > 48 0 = RIG < 48	Comparative Constitutions Project (2021)
Autocratic Constitutions (ACO)	Constitutions persist when they are not a heritage from the previous authoritarian regime	1 = Autocratic Constitution 0 = Non-Autocratic Constitution	Albertus and Menaldo (2018)
Constitutional Endurance (OUT)	Do constitutions survive in the democratic period?	1 = Yes 0 = No	Comparative Constitutions Project (2021); Landau (2019); Negretto (2020, 2012), 2012)

Source: Elaborated by the authors.

To begin, the first condition examines how power is distributed among the three branches of government (H1). Previous research on constitutional replacement has measured power-sharing by making a composite index comprising three components: the extent of presidential veto power, the bicameral nature of the legislature, and the degree of judiciary independence from the executive (NEGRETTO, 2012). However, this measurement is at odds with the concept of power-sharing. In this context, power-sharing implies that each branch of the government is actively complying with its obligations. Thereby, the presence of both a bicameral legislature and presidential veto power alone does not necessarily indicate whether a country has effective power-sharing institutions. These are simply institutional features of each country, where a bicameral legislative arrangement, for example, does not encroach on the executive's activities.

To provide a better-suited measurement for this concept, we use data from Varieties of Democracy (V-Dem) (COPPEDGE et al., 2020). Basically, our measure of power-sharing institutions is based on the average extent to which the executive refrains from interfering with the judiciary’s decisions and the degree to which the executive is subject to oversight by the legislature and other government departments ³ . That being the case, Power-Sharing is assessed empirically by examining the extent to which the executive refrains from dominating the other branches, thereby preserving the institutional equilibrium of the polity.

The calibration follows the direct assignment method, where we distinguish between membership and non-membership in the set based on the natural distribution of the data.

The second condition is whether political actors face significant hurdles in modifying the rules and norms established by the existing constitution. Political actors may seek constitutional replacement when 01. amending the constitution is a challenging process or 02. there are major impediments in interpreting the constitution’s provisions. As such, this set encompasses two related but distinct concepts: to what extent the constitution is ‘amendable’ and the authority granted to judges to interpret the constitution.

To gauge the difficulty of amending constitutions, we examine the size of the parliamentary majority required to approve constitutional amendments, as documented by the Comparative Constitutions Project ⁴ . In general, Latin American countries either require a qualified majority (60%) or an even stricter majority in the legislature to amend their constitutions. Against this backdrop, countries are classified into cases with either easy or complex amendment procedures ⁵ . The former is composed of countries that demand up to 60% of the parliament to amend the constitution, whereas the latter comprises cases with higher thresholds.

In parallel, the ‘de jure’ index of judicial power provides the basis for measuring the extent to which judicial review is a prominent means for modifying the laws and rules established by the constitution (NEGRETTO, 2012; RÍOS-FIGUEROA, 2011). The index ranges from 0 to 8, with higher values indicating that judges have greater wiggle room to interpret constitutions. Despite having a maximum score of eight, no case can attain such a score due to theoretical unavailability (RÍOS-FIGUEROA, 2011, pp. 41-42). Therefore, countries with a historical median score higher than three are labeled cases where judicial review is a feasible mechanism for altering the ‘status quo’. Taken together, membership in the Amendment Set depends on whether the approval of amendments requires a ‘small’ majority in the legislature, and whether judges have substantial powers to change the way in which the constitution is interpreted at their discretion.

The third condition sheds light on the countries’ historical backgrounds. Institutions are sticky (CHEIBUB, 2007; ELKINS, GINSBURG and MELTON, 2009), so countries with a limited legacy of constitutional replacement face numerous obstacles when attempting to alter their existing constitutional order (H3). Based on the Comparative Constitutions Project and the natural break in the data, we assigned a code of 1 to cases with more than thirteen constitutional changes, and 0 to cases with fewer changes.

The fourth condition concerns the level of social mobilization in countries. Our approach contrasts with the prevailing perspective in the literature, which suggests that large demonstrations may trigger substantial institutional changes in a particular year. Instead, we theorize that institutional flip-flops arise from cumulative mass mobilizations over time (H4). Once again, we resort to data garnered by the V-Dem (COPPEDGE et al., 2020). Country experts were asked, ‘How frequent and large have events of mass mobilization been?’ ⁶ . The responses were given on a 5-point scale in which 0 corresponds to ‘there have been virtually no events’ and 4 to ‘there have been many large-scale and small-scale events’. Ideally, our threshold should be established at the midpoint, where countries have not witnessed many large-scale demonstrations across the years. Following V-DEM’s questionnaire, this point should be located at 02, where respondents replied that ‘there have been many small-scale events’. Nonetheless, given that the literature highlights the role of social mobilizations in the calling of a constituent assembly in Chile (PISCOPO and SIAVELIS, 2021; VERDUGO and PRIETO, 2021), we have slightly reduced the cross-over point to 1.86. As such, Latin-American countries with figures of demonstrations above 1.86 were coded as 1, and 0 otherwise.

The fifth condition focuses on the extent of rights enshrined in constitutions. The rationale is that the populace presses for a constitutional change if the existing constitution does not grant citizens a substantial number of rights. As democratic countries theoretically lack the same capability as authoritarian regimes to shut down demonstrations and repress the public, the cumulative demand for more rights would lead to the implementation of a more protective constitutional order. It bears noting that we are not interested in the different ‘types’ of rights provided by constitutional laws. In other words, constitutions are considered equivalent as long as they contain the same number of rights, even if there are differences when it comes to the ‘extent’ and ‘content’ of economic, social, and political rights in their texts. To calibrate this condition, we again use data from the Comparative Constitutions Project ⁷ . The calibration is based on grounded knowledge of constitutional texts. Given the significant contrast in rights between the 1999 Venezuelan Constitution and the previous 1961 Venezuelan Constitution (ELKINS, 2017), as well as criticism toward the 1980 Chilean Constitution for its limited social provisions (ANSALDI and PARDO-VERGARA, 2020), only constitutions with more than 48 rights are included in the reference set of Number of Rights.

Moving on to the last causal condition, Autocratic Constitutions refer to the origins of the constitutions. We depart from the principle that constitutions crafted during autocratic periods are prone to be replaced under democratic regimes in order to suppress the last remnants of the autocratic period. The data, sourced from Albertus and Menaldo (2018), simply reflect whether the constitution was written under an autocratic or a democratic period.

Finally, our outcome indicates whether the country has retained its constitution under a democratic regime or undergone a process of constitutional replacement. We have chosen to analyze the Chilean case, despite the process of replacing the constitution starting in 2019 and resulting in the recent rejection of a more progressive constitution in 2022. This case drew our attention to the issue in the first place, although it is worth noting that new constitutional law, may be promulgated shortly and therefore incorporated into Chile’s legal system.

Analysis of necessary and sufficient conditions

As discussed earlier, QCA involves the analysis of both necessary and sufficient conditions. In general, QCA studies follow a step-by-step procedure to get to their findings. In short, this scheme begins with the analysis of necessity, followed by the construction of the truth table, and concludes with the analysis of sufficiency. The major advantage of this neat process is that the likelihood of performing a misguided QCA analysis is greatly minimized (SCHNEIDER and WAGEMANN, 2012). To proceed, we use the QCA and SetMethods packages in R (DUŞA, 2019; OANA and SCHNEIDER, 2018).

Following that procedure, we start by carrying out the analysis of necessity. The literature highly recommends setting a consistency threshold of 0.9 in the analysis of necessity, as higher thresholds for necessity relations help avoid further problems when interpreting the final solution formula (RAGIN, 2006). Complementarily, it has also become more widely acknowledged that necessity relations should undergo a ‘double’ check to assess their degree of triviality (ALBALA and LEAL, 2022; SCHNEIDER and WAGEMANN, 2012). In practical terms, this implies considering the 0.9 consistency cut-off point along with the Relevance of Necessity (RoN) score, where a ‘relevant’ necessity relation is expected to have a RoN value above 0.6 (MELLO, 2022; OANA, SCHNEIDER and THOMANN, 2021). Based on these considerations, the results of the necessity test ⁸ are presented in Table 03 .

Table 03
Necessity test

Paths	Consistency	Coverage	RoN
RIG	0.923	0.857	0.600
AME + ~LEG	0.923	0.923	0.800
~LEG + ~ACO	0.923	0.857	0.600

Source: Elaborated by the authors.

However, QCA analysis does not operate in a ‘black-and-white’ manner, where conditions and/or disjunctions are considered necessary as long as they pass the required thresholds. Indeed, there are several criteria that conditions and disjunctions must meet to be validated as necessary, amongst which are 01. empirical consistency, 02. empirical relevance, and 03. theoretical soundness (OANA, SCHNEIDER and THOMANN, 2021; SCHNEIDER, 2018; SCHNEIDER and ROHLFING, 2013). While the necessary expressions identified in the necessity analysis account for a significant portion of the cases and are theoretically sound, visualizations of the necessary statements reveal that every path fails to secure full empirical consistency, as each one is plagued by one deviant case ⁹ .

Of particular interest to us is the fact that Chile is precisely the country that violates the necessary statement regarding the number of rights. Hence, even though the process is still unfinished, the argument that the number of rights serves as a necessary condition gains weight from the fact that Chile has taken the first steps toward displacing its 1980 Constitution. As a result, since it may only be a matter of time before Chile ceases to be a deviant case, we consider the number of rights as the sole necessary condition for constitutional endurance in the subsequent analyses ¹⁰ .

With the results of the necessity test in hand, our next step is to set out the truth table before analyzing sufficient conditions. In summary, the truth table lays out all the logically possible combinations of conditions that either lead to the outcome or its absence. The total number of configurations is directly determined by the number of conditions employed in the research. In our case, the truth table contains 64 rows ¹¹ . Table 03 presents the truth table for constitutional endurance in democratic Latin America ¹² , with rows without empirical instances omitted for better visualization.

Even before starting the minimization process of the truth table, a couple of characteristics draw our attention. Firstly, our theoretical framework did not yield any contradictory rows, where two cases share the very same conditions but display distinct outcomes. Second, our truth table is severely hindered by limited diversity. In other words, many rows are not covered by any case, as evidenced by the fact that Table 04 displays only 14 out of the 64 truth table rows. This means that logical remainders ¹³ are scattered throughout.

Table 04
Truth table for constitutional endurance in democratic settings in Latin America

POS	AME	LEG	MOB	RIG	ACO	OUT	n	incl.	Cases
0	0	0	1	1	1	1	2	1.000	MX, SV
1	0	0	1	1	0	1	2	1.000	CR, UY
0	0	0	0	1	1	1	1	1.000	GT
0	0	0	1	1	0	1	1	1.000	AR
0	0	1	0	1	0	1	1	1.000	HN
0	1	0	0	1	0	1	1	1.000	PY
0	1	0	0	1	1	1	1	1.000	PA
0	1	1	1	1	0	1	1	1.000	NI
0	1	1	1	1	1	1	1	1.000	PE
1	1	0	0	1	0	1	1	1.000	BR
1	1	0	1	0	1	1	1	1.000	CL
0	0	0	1	0	1	0	1	0.000	CO
0	0	1	0	1	1	0	1	0.000	EC
0	0	1	1	1	1	0	1	0.000	BO
1	0	1	1	0	0	0	1	0.000	VE

Source: Elaborated by the authors.

The analysis of sufficiency in QCA can be approached in three ways, and the differences among them lie in the assumptions made about the logical remainders. For those more familiar with qualitative research, the differences revolve around how researchers address counterfactual claims. One approach is to make no assumption about logical remainders and exclude them from the analysis (conservative solution). Conversely, it is possible to include all logical remainders and contend that both easy and difficult counterfactuals contribute to the outcome (parsimonious solution). Lastly, researchers can strike a balance between the two approaches and include only the counterfactuals that align with the theory in the analysis (intermediate solution).

Here, we opt for applying the intermediate solution, considering both the results of the necessity test ¹⁴ and the expectations of our conditions as outlined in Table 02 . The results are presented in Table 05¹⁵ .

Table 05
Enhanced intermediate solution for constitutional endurance

Paths	Consistency	Raw Coverage	Unique Coverage	Cases
AME*RIG	1.000	0.385	0.154	BR, NI, PA, PE, and PY
~LEG*RIG	1.000	0.692	0.462	AR, BR, CR, GT, MX, PA, PY, SV, and UY
~MOB*RIG*~ACO	1.000	0.231	0.077	BR, HN, PY
POS*AME*~LEG*MOB*ACO	1.000	0.077	0.077	CL
Overall Solution	1.000	1.000

Source: Elaborated by the authors.

The explanation for constitutional persistence resides in four alternative pathways, with the first two causal recipes being empirically more relevant than the others. First, one way to avoid constitutional replacement is to have an ‘amendable’ constitutional law with a significant number of rights. This path covers roughly 40% of the democratic Latin American countries that have not replaced their constitutions, and it uniquely explains the cases of Nicaragua and Peru. The second pathway leading to constitutional preservation is the combination of a limited history of constitutional replacement with a substantial number of rights enshrined in the constitution. This particular causal recipe gains prominence due to its robust empirical coverage, accounting for approximately 70% of the cases under study and particularly outlining the reasons why half of the countries have not entirely replaced their constitutions ¹⁶ .

Moreover, the third and fourth paths are of lesser practical importance due to their low coverage. While the former suggests that constitutional endurance results from a minimal level of continuous mobilization and a substantial depth of rights in a constitution established during democratic rule, the latter stands out as a unique recipe specifically applicable to the Chilean case. Hence, according to our theoretical framework, the fact that Chile has not yet changed its constitution diverges from the explanation for all other cases in democratic Latin America ¹⁷ .

Robustness tests

Recent studies applying QCA have subscribed to the idea of conducting a series of tests to probe the soundness of their analysis, aiming to determine whether the results are influenced by the researchers’ decisions in the earlier stages of the research (e.g., IDE, 2015; JANZWOOD, 2020). In our study, these tests include changing the initial case selection and altering the cut-off point of half of the conditions.

To begin, we include the Dominican Republic in the analysis and change the coding for the outcome of Chile from 1 to 0, thus providing an analysis for the Chilean constitution-making process ending with the adoption of a new constitution. The results are available in Table 19 in the Appendix.

By doing so, the fourth pathway of the original sufficiency analysis for constitutional endurance no longer exists. This suggests that Chile is indeed an outlier for not replacing its constitution so far under democratic rule within the Latin American context. On the other hand, the results for constitutional replacement are more streamlined, suggesting that the overall solution is not overly complex. Interestingly, a configuration formed between MOB and ~RIG (MOB*~RIG) covers most of the cases of constitutional replacements that occurred after the democratic rebound in Latin America.

Yet, even if these paths may seem more appealing for their apparent theoretical contribution, it is worth noting that these findings are based on questionable case selection. This is evident from the fact that contrary to the other countries under study, the Dominican Republic changed its constitution from an autocratic standpoint. Moreover, these results were only obtained under the assumption that the ongoing Chilean constitution-making process would definitively replace the constitution inherited from the Pinochet Era. Notably, the ultimately unsuccessful Icelandic constitution-writing process reminds us that implementing a constitution-making body does not necessarily imply that a constitutional replacement will occur (ELSTER, 2016).

Turning to changes in the calibration process, we slightly changed the thresholds of Legacy of Replacement, Continuous Mobilization, and Number of Rights. Regarding the historical background of promoting constitutional replacements, the cut-off point was raised from 13 to 15 to test whether a more stringent criterion would somehow change the findings. Complementarily, since the calibration process for Continuous Mobilization was guided by the events unfolding surrounding the Chilean constitution-making process, we test whether our cross-case analysis may have been contaminated by an excessive focus on the Chilean reality. Thus, we set the cross-over point at 2.0 in line with the V-Dem’s questionnaire, instead of relying on the Chilean experience. Finally, despite the skyrocketing increase of rights from one constitution to another in Venezuela, we lower the threshold of Number of Rights from 48 to 46. As a result, while there are minor differences, the findings remain practically the same, contrasting with changes in case selection. These results are available in Table 21 and Table 22 in the Appendix.

Illustrative cases

Invariably, QCA is wedded to a more qualitative approach toward empirical research, despite growing scholarly interest in integrating QCA with large-N analysis (e.g., GRECKHAMER, MISANGYI and FISS, 2013; RUTTEN, 2022). Frequently, this means that QCA research prioritizes the cases themselves. Thus, we follow this trend, revisiting cases and delving deeper into how the results from previous analyses speak to them. Due to the complex and particularistic findings explaining why constitutions have been replaced, we focus our case-orientated discussion on the process of constitutional persistence.

As seen previously, the explanation of why constitutions endure under democratic Latin America resides in four pathways. The first pathway of the solution indicates that constitutions that are ‘amendable’ and include provisions for a considerable number of rights (AME*RIG) are not replaced in the long run. The prominence of this combination is evident in the maintenance of the 1987 Nicaraguan and 1993 Peruvian constitutions, albeit the autocratic origins of the latter. Even before the emergence of the Bolivarian model of constitutions (ELKINS, 2017), the 1987 Nicaraguan Constitution, in particular, stands out for its extensive inclusion of rights across various areas, including the environment, health, social security, and protection of indigenous communities (JONAS and STEIN, 1990). By contrast, the 1993 Peruvian constitutional law is designed to limit the state’s role in providing social goods (VERGARA and ENCINAS, 2016). However, this does not mean that the Constitution is depleted of rights. The 1993 Peruvian Constitution offsets the lack of social provisions by extending the rights granted to indigenous populations in the country (SEGADO, 1994; FAJARDO, 2002). Moreover, constitutional reforms occur sporadically in both countries, either through the statutory legislative process or judicial review (ANDERSON, 2006; DARGENT, 2009; WALKER and WILLIAMS, 2010). Taken together, debate over a new constitution is largely absent in Nicaragua. Meanwhile, although calls for constitutional replacement have persisted over the years in Peru, the combination of rights and constitutional flexibility has guaranteed their failure thus far.

The second path emphasizes a historical approach to explain why some Latin American countries have not replaced their constitutions in democratic settings. The causal mechanism proposed here follows a path-dependent explanation for institutional change (see PIERSON, 2004). Just as in a typical path-dependent process, the costs of replacing a constitution increase over time, thereby making the writing and approval of an alternative constitutional text much more burdensome. Interestingly, the lack of a tradition of replacing constitutions further increases the costs of engaging in constitution-making processes, given the unpredictability of whether the outcome would ultimately be desirable. This is seen in Mexico, where the long-standing 1917 Constitution is believed to endure because it is thought that a brand-new constitution would hardly generate higher payoffs (ELKINS, GINSBURG and MELTON, 2009, Chap. 08). Also, it is worth considering that the absence of historical precedents for constitutional changes raises concerns regarding the viability of embarking on a complete rewrite of the constitutional law in case of dissatisfaction with the constituent assembly’s outcome ¹⁸ . Furthermore, the inclusion of numerous rights provides an additional compelling reason to stick with an existing constitution, even if it is imperfect. The inclusion of social rights in the constitutional laws of countries such as Argentina, Costa Rica, and Uruguay (GARGARELLA, 2017), coupled with a limited tradition of constitutional replacements, might prevent political actors from embracing the uncertainty and risks surrounding constitution-making processes.

The next pathway to constitutional endurance highlights the combination of three features: comprehensive constitutional texts with a significant number of rights, democratic origins of constitutional law, and the absence of street protests. This particular configuration effectively explains the case of Brazil, where despite large demonstrations during some periods, notably in 2013 and 2015-16, there have not been significant challenges to the constitution. Indeed, a constitution that emerged from the efforts of redemocratisation, filled with social rights, proves effective in resisting calls for the use of constituent power. Likewise, Paraguay has not witnessed calls for a complete constitutional overhaul, even amid the political turmoil surrounding the impeachment of former President Fernando Lugo. The explanation for the Honduran case, however, is slightly different, as the country was marked by a constitutional crisis. Unlike Brazil and Paraguay, the process of replacing the existing constitution began with initial steps in 2006, when then-President Zelaya vigorously tried to hold a referendum to gauge citizens’ support for a constitutional replacement (LLANOS and MARSTEINTREDET, 2010). Ultimately, despite the executive’s efforts, a constituent assembly was not convened in Honduras, and there have been no significant calls from civil society for a broad constitutional reform since then, despite the emergence of new constitutional controversies (LANDAU, DIXON and ROZNAI, 2019).

Lastly, a unique combination of conditions sheds light on constitutional endurance in Chile. In sum, social unrest and a constitution inherited from the dictatorship era suggest a move toward constitutional replacement. On the other hand, the existence of power-sharing institutions with means to amend the existing constitutional text and the absence of a history of constitutional replacement point in the opposite direction. At the time of writing, two constitutional drafts have been rejected by an overwhelming majority of votes, effectively halting the process of constitutional replacement, thereby giving more credence to the latter part of this configuration, at least for now.

Conclusion

This paper presents an original methodological approach to studying national constitutional replacements or lack thereof. So far, previous studies have focused on the impact of separate, independent variables on the lifespan of written constitutions (ELKINS, GINSBURG and MELTON, 2009; MONTERO, 2017; NEGRETTO, 2012, 2008). However, this paper takes a different approach by analyzing how combinations of conditions either facilitated or hindered the enactment of new constitutions after regime transitions in Latin American democracies. By doing so, this work contributes to scholarly discussions and research in comparative constitutionalism and constitution-making processes, particularly in the Latin American context. It is worth noting that while our analysis has focused on Latin America, our research framework can be successfully applied to other regions worldwide, especially those that have witnessed democratic flourishing after the third wave of democratization.

The empirical results reveal that the following conditions contribute to the endurance of constitutions within a democratic context: the extension of constitutional rights, the degree to which the constitution is ‘amendable’, a cultural tendency to replace constitutions, and the lack of sustained popular unrest. While the extension of constitutional rights serves as a necessary condition, the remaining factors function as INUS conditions. In sum, in different ways, these conditions play an important role in explaining constitutional endurance in democratic Latin America.

On the other hand, our understanding of constitutional replacements in democratic Latin American countries remains unclear. This is because the findings provide quite complex causal recipes, making it challenging to discern meaningful shared patterns among the cases. Nevertheless, a common factor emerges across all pathways leading to constitutional replacement: the absence of streamlined procedures to change the constitutions’ provisions. That is, previous constitutions were marked by either a low degree of judicial review or the requirement for more than a ⅔ majority to pass constitutional amendments.

Future research in comparative constitutionalism would gain valuable insights by examining the different types of rights enshrined in constitutional texts. In our study, we focused on the total number of rights, regardless of their content. However, different rights may convey different causal mechanisms for either replacing or preserving constitutional laws. Additionally, future investigations should consider how the systematic undermining of democratic pillars can impact constitution-making processes. As autocrats adopt subtler tactics to hold onto power, gradually eroding democratic institutions instead of resorting to overt coups (LÜHRMANN and LINDBERG, 2019), constitutions are potentially entangled in this process. Therefore, scholars interested in constitutional theories should acknowledge that democratic backsliding potentially affects both the endurance and replacement of constitutions.

Acknowledgments: Previous versions of this paper were presented at the VII Brazilian Forum of Graduate Studies in Political Science (FBCP), the 13th Meeting of the Brazilian Political Science Association (ABCP), and the LAPCIPP Meetings at the Universidade de Brasília. We thank the discussants and audiences at the talks for their helpful comments and suggestions. We are also grateful to anonymous reviewers for their recommendations, which immensely helped us to improve the manuscript. We also thank the BPSR's editorial team for carefully handling the manuscript and Karin Blikstad for meticulously proofreading the final text.

References

• ALBALA, Adrián and LEAL, Leonardo Prates (2022), Introdução à 'Qualitative Comparative Analysis' (QCA). In: Política comparada: teoria e método. Edited by PERISSINOTTO, Renato; BOTELHO, João Carlos Amoroso; BOLOGNESI, Bruno; BATISTA, Mariana, and SANTOS, Manoel Leonardo. Rio de Janeiro: EdUERJ. pp. 113-140.
• ALBERTUS, Michael and MENALDO, Victor (eds) (2018), Constitutions as elite deal making. In: Authoritarianism and the elite origins of democracy Cambridge: Cambridge University Press. pp. 63-98.
• ANDERSON, Leslie. E. (2006), The authoritarian executive? Horizontal and vertical accountability in Nicaragua. Latin American Politics and Society Vol. 48, Nº 02, pp. 141-169.
• ANSALDI, Octavio and PARDO-VERGARA, María (2020), What Constitution? On Chile's constitutional awakening. Law and Critique. Vol. 31, Nº 01, pp. 07-39.
• BAUMGARTNER, Frank. R. and MAHONEY, Christine (2005), Social movement, the rise of new issues and the public agenda. In: Routing the opposition: social movements, public policy, and democracy. Edited by MEYER, David S.; JENNESS, Valerie, and INGRAM, Helen. Minneapolis: University of Minnesota Press. pp. 65-86.
• BLAKE, William D.; COZZA, Joseph Francesco; ARMSTRONG II, David A., and FRIESEN, Amanda (2023), Social capital, institutional rules, and constitutional amendment rates. American Political Science Review. Vol. 118, Nº 02, pp. 1075-1083.
• CHEIBUB, José Antônio (2007), Presidentialism, parliamentarism, and democracy . Cambridge: Cambridge University Press. 202 pp..
• COLLIER, David (1993), The comparative method. In: Political science: the state of the discipline II. Edited by FINIFTER, Ada W. Washington: American Political Science Association. pp. 105-119.
• COPPEDGE, Michael; GERRING, John; KNUTSEN, Carl Henrik; LINDBERG, Staffan I.; TEORELL, Jan; ALTMAN, David; BERNHARD, Michael; FISH, M. Steven; GLYNN, Adam; HICKEN, Allen; LÜHRMANN, Anna; MARQUARDT, Kyle L.; McMANN, Kelly M.; PAXTON, Pamela; PEMSTEIN, Daniel; SEIM, Brigitte; SIGMAN, Rachel; SKAANING, Svend-Erik; STATON, Jeffrey K.; CORNELL, Agnes; GASTALDI, Lisa; GJERLØW, Haakon; MECHKOVA, Valeriya; RÖMER, Johannes von; SUNDTRÖM, Aksel; TZELGOV, Eitan; UBERTI, Luca Jacopo; WANG, Yi-Ting; WIG, Tore, and ZIBLATT, Daniel (2020), V-Dem Codebook v10. Varieties of Democracy (V-Dem) Project. Available at ? https://doi.org/10.23696/vdemds20 ?. Accessed on March 24, 2022.
  » https://doi.org/10.23696/vdemds20
• COUTO, Cláudio and ARANTES, Rogério (2022), Constitutions and public policies. In: The Oxford Handbook of Constitutional Law in Latin American. Edited by MENDES, Conrado Hübner; GARGARELLA, Roberto, and GUIDI, Sebastián. Oxford: Oxford University Press. pp. 496-514.
• DARGENT, Eduardo (2009), Determinants of judicial independence: lessons from three 'cases' of Constitutional Courts in Peru (1982-2007). Journal of Latin American Studies Vol. 41, Nº 02, pp. 251-278.
• DUSA, Adrian (2019), QCA with R: a comprehensive resource. Cham: Springer International Publishing. 289 pp..
• ELKINS, Zachary (2017), Constitutional revolution in the Andes? In: Comparative constitutional law in Latin America. Edited by DIXON, Rosalind and GINSBURG, Tom. Cheltenham: Edward Elgar Publishing. pp. 108-126.
• ELKINS, Zachary; GINSBURG, Tom, and MELTON, James (2014), Imagining a world without the Universal Declaration of Human Rights. Working paper.
• ELKINS, Zachary; GINSBURG, Tom, and MELTON, James (2009), The endurance of national constitutions. New York: Cambridge University Press. 272 pp..
• ELKINS, Zachary; GINSBURG, Tom, and SIMMONS, Beth A. (2013), Getting to rights: treaty ratification, constitutional convergence, and human rights practice. Harvard International Law Journal. Vol. 54, Nº 01, pp. 201-234.
• ELSTER, Jon (2016), Icelandic constitution-making in comparative perspective. In: Iceland's financial crisis: the politics of blame, protest, and reconstruction. Edited by INGIMUNDARSON, Valur; URFALINO, Philippe, and ERLINGSDÓTTIR, Irma. New York: Routledge. pp. 187-202.
• FAJARDO, Raquel Yrigoyen (2002), Peru: Pluralist constitution, monist judiciary. In: Multiculturalism in Latin America: indigenous rights, diversity and democracy. Edited by SIEDER, Rachel. Houndmills: Palgrave Macmillan. pp. 157-183.
• FASSIO, Fernando Castilla (2023), Luisa González: educación, seguridad y, tal vez, Asamblea Constituyente. Radio Centro. July 01, 2023. Available at https://radiocentro.com.ec/luisa-gonzalez-educacion-seguridad-y-tal-vez-asamblea-constituyente . Accessed on July 17, 2023.
  » https://radiocentro.com.ec/luisa-gonzalez-educacion-seguridad-y-tal-vez-asamblea-constituyente
• GARGARELLA, Roberto (2017), Equality. In: Comparative Constitutional Law in Latin America. Edited by DIXON, Rosalind and GINSBURG, Tom. Cheltenham: Edward Elgar Publishing. pp. 176-197.
• GINSBURG, Tom and MELTON, James (2015), Does the constitutional amendment rule matter at all? Amendment cultures and the challenges of measuring amendment difficulty. International Journal of Constitutional Law. Vol. 13, Nº 03, pp. 686-713.
• GOERTZ, Gary (2020), Social science concepts and measurement. Princeton: Princeton University Press. 288 pp..
• GRECKHAMER, Thomas; MISANGYI, Vilmos F., and FISS, Peer C. (2013), The two QCAs: from a small-N to a large-N set-theoretic approach. Research in the Sociology of Organizations. Vol. 38, pp. 49-75.
• IDE, Tobias (2015), Why do conflicts over scarce renewable resources turn violent? A qualitative comparative analysis. Global Environmental Change . Vol. 33, pp. 61-70.
• JANZWOOD, Amy (2020), Explaining variation in Oil Sands Pipeline Projects. Canadian Journal of Political Science. Vol. 53, Nº 03, pp. 540-559.
• JONAS, Susanne and STEIN, Nancy (1990), The construction of democracy in Nicaragua. Latin American Perspectives Vol. 17, Nº 03, pp. 10-37.
• LANDAU, David (2019), Constituent power and constitution making in Latin America. In: Comparative constitution making. Edited by LANDAU, David and LERNER, Hanna. Northampton: Edward Elgar. pp. 567-588.
• LANDAU, David E.; DIXON, Rosalind, and ROZNAI, Yaniv (2019), From an unconstitutional constitutional amendment to an unconstitutional constitution? Lessons from Honduras. Global Constitutionalism. Vol. 08, Nº 01, pp. 40-70.
• LATIN NEWS DAILY REPORT (2017), Lasso moots constituent assembly for Ecuador. Latin News Daily Report. March 07, 2017.
• LLANOS, Mariana and MARSTEINTREDET, Leiv (eds) (2010), Epilogue. In: Presidential breakdowns in Latin America. Causes and outcomes of Executive instability in developing democracies New York: Palgrave Macmillan. pp. 229-238.
• LÜHRMANN, Anna and LINDBERG, Staffan I. (2019), A third wave of autocratization is here: what is new about it? Democratization. Vol. 26, Nº 07, pp. 1095-1113.
• MARQUARDT, Bernd (2010), La cuarta fase del constitucionalismo Iberoamericano: el ascenso del constitutionalismo social, 1917-1949. Pensamiento Jurídico. Vol. 28, pp. 119-164.
• MARSHALL, Monty G. and GURR, Ted Robert (2020), Polity 05: political regime characteristics and transitions, 1800-2018: dataset users' manual. Louisville: Center for Systemic Peace. 85 pp..
• MELLO, Patrick A. (2022), Qualitative comparative analysis: an introduction to research design and application. Washington: Georgetown University Press. 260 pp..
• MEYER, David S. (2005), Introduction. Social movements and public policy: eggs, chicken, and theory. In: In: Routing the opposition: social movements, public policy, and democracy. Edited by MEYER, David S.; JENNESS, Valerie, and INGRAM, Helen. Minneapolis: University of Minnesota Press. pp. 01-26.
• MONTERO, José Luis Castro (2017), Constitutional replacement in five countries of South America: an empirical comparative analysis. Estudios de Derecho. Vol. 74, Nº 164, pp. 69-88.
• NAVARRETE, Bernardo and TRICOT, Victor (2021), The social outburst and political representation in Chile. Cham: Springer. 206 pp..
• NEGRETTO, Gabriel L. (2020), Constitution-making and liberal democracy: the role of citizens and representative elites. International Journal of Constitutional Law. Vol. 18, Nº 01, pp. 206-232.
• NEGRETTO, Gabriel L. (2012), Replacing and amending constitutions: the logic of constitutional change in Latin America. Law and Society Review Vol. 46, Nº 04, pp. 749-779.
• NEGRETTO, Gabriel L. (2008), The durability of constitutions in changing environments: explaining constitutional replacements in Latin America. Working Paper Nº 350. The Helen Kellogg Institute for International Studies.
• OANA, Ioana-Elena; SCHNEIDER, Carsten Q. (2018), SetMethods: an Add-on R package for advanced QCA. The R Journal Vol. 10, Nº 01, pp. 507-533.
• OANA, Ioana-Elena; SCHNEIDER, Carsten Q., and THOMANN, Eva (2021), Qualitative comparative analysis using R: a beginner's guide. Cambridge: Cambridge University Press. 180 pp..
• OPPERMANN, Kai and BRUMMER, Klaus (2020), Who gets what in foreign affairs? Explaining the allocation of foreign ministries in coalition governments. Government and Opposition. Vol. 55, Nº 02, pp. 241-259.
• PETERS, B. Guy (1998), Comparative politics: theory and method. Basingstoke: MacMillan. 262 pp..
• PIERSON, Paul (2004), Politics in time: history, institutions and social analysis Princeton: Princeton University Press. 208 pp..
• PISCOPO, Jennifer M. and SIAVELIS, Peter M. (2021), Chile's constitutional moment. Current History. Vol. 120, Nº 823, pp. 43-49.
• PRZEWORSKI, Adam (1999), La democracia como resultado contingente de conflictos. In: Constitucionalismo y democracia. Edited by ELSTER, Jon and SLAGSTAD, Rune. México: Fondo de Cultura Económica. pp. 91-112.
• RAGIN, Charles C. (2008), Redesigning social inquiry: fuzzy sets and beyond Chicago: University of Chicago Press. 240 pp..
• RAGIN, Charles C. (2006), Set relations in social research: evaluating their consistency and coverage. Political Analysis Vol. 14, Nº 03, pp. 291-310.
• RAGIN, Charles C. (2000), Fuzzy-set social science Chicago: University of Chicago Press. 370 pp..
• RAGIN, Charles C. (1987), The comparative method: moving beyond qualitative and quantitative strategies. Berkeley: University of California Press. 218 pp..
• RÍOS-FIGUEROA, Julio (2011), Institutions for constitutional justice in Latin America. In: Courts in Latin America. Edited by HELMKE, Gretchen and RÍOS-FIGUEROA, Julio. New York: Cambridge University Press. pp. 27-54.
• RUTTEN, Roel (2022), Applying and assessing Large-N QCA: causality and robustness from a critical realist perspective. Sociological Methods & Research. Vol. 51, Nº 03, pp. 1211-1243.
• SARTORI, Giovani (1970), Comparing and miscomparing. Journal of Theoretical Politics Vol. 03, Nº 03, pp. 243-257.
• SAZO, Diego (2023), Chile 2022: from great expectations to rising pessimism. Revista de Ciência Política. Vol. 43, Nº 02, pp. 193-222.
• SCHNEIDER, Carsten Q. (2018), Realists and idealists in QCA. Political Analysis Vol. 26, Nº 02, pp. 246-254.
• SCHNEIDER, Carsten Q. and ROHLFING, Ingo (2013), Combining QCA and process tracing in Set- theoretic multi-method research. Sociological Methods & Research Vol. 42, Nº 04, pp. 559-597.
• SCHNEIDER, Carsten Q. and WAGEMANN, Claudius (2012), Set-theoretic methods for the Social Sciences: a guide to qualitative comparative analysis. New York: Cambridge University Press. 350 pp..
• SEGADO, Francisco Fernandez (1994), El nuevo ordenamiento constitucional del Perú: aproximación a la constitución de 1993. Revista de Estudios Políticos Vol. 84, pp. 27-67.
• TSEBELIS, George (2022), Constitutional rigidity matters: a veto player approach. British Journal of Political Science. Vol. 52, Nº 01, pp. 280-299.
• VERDUGO, Sergio and PRIETO, Marcela (2021), The dual aversion of Chile's constitution-making process. International Journal of Constitutional Law. Vol. 19, Nº 01, pp. 149-168.
• VERGARA, Alberto and ENCINAS, Daniel (2016), Continuity by surprise: explaining institutional stability in contemporary Peru. Latin American Research Review. Vol. 51, Nº 01, pp. 159-180.
• WALKER, Lee Demetrius and WILLIAMS, Philip J. (2010, The Nicaraguan constitutional experience: process, conflict, contradictions, and change. In: Framing the state in times of transition: case studies in constitution making. Edited by MILLER, Laurel E. Washington: United States Institute of Peace. pp. 483-504.

• 1
  In 2019, Chile experienced a massive wave of social mobilization that swept the country from north to south. Although riots had initially emerged over a rise in public transportation fares in Santiago, the true source of discontent among Chileans stemmed from the precarious socioeconomic situation experienced that year (NAVARRETE and TRICOT, 2021). Interestingly, the roots of Chile’s socioeconomic problems do not lie in 2019; rather, many of these issues stem from the democratization process itself – the constitution-making process resulted in a constitution marked by neoliberal tenets. Despite the return of democracy in 1990, Chile’s current constitution dates back to General Pinochet’s authoritarian rule. That is to say, the autocratic constitution persisted even after Chile transitioned to democracy at the end of the twentieth century. Against this backdrop, despite facing a global pandemic, Chileans chose to organize a referendum in which an overwhelming majority voted in favor of rewriting the constitution. Although this process is still ongoing, it initially resulted in the rejection of the first proposed draft (SAZO, 2023).
• 2
  As a matter of fact, following its initial release in the 1980s, csQCA was the only way to conduct QCA research for over a decade (RAGIN, 2000, 1987).
• 3
  The two questions used to develop our measure were: 01. ‘To what extent does the executive respect the constitution and comply with court rulings, and to what extent is the judiciary able to act in an independent fashion?’ 02. ‘To what extent are the legislature and government agencies e.g., comptroller general, general prosecutor, or ombudsman capable of questioning, investigating, and exercising oversight over the executive?’ Both responses were standardized to range from 0 to 1 on a ratio scale. As the V-DEM project asked both questions on an annual basis, we have also averaged the responses as a means to suit our uneven temporal coverage.
• 4
  Available at ˂ https://comparativeconstitutionsproject.org/download-data/˃ . Accessed on June 04, 2023.
• 5
  It is worth mentioning that there has been an ongoing discussion in the literature regarding whether constitutional rigidity plays a role in the frequency of amendments (GINSBURG and MELTON, 2015). On the one hand, Tsebelis (2022) argues that constitutional rules for amendments serve as a necessary but not sufficient condition to enable constitutional reforms through amendment in democratic regimes. In other words, even if other factors also matter, such as civic activism and political trust, the formal thresholds are still an important part of the process of amending the constitutional law (BLAKE et al., 2023). On the other hand, Couto and Arantes (2022) claim that constitutional amendments depend less on the constitutional requirements ‘per se’, but mainly on the legislators’ intention to achieve effective policy-making through amendments. While resolving this debate exceeds the scope of our work, we maintain the formal condition – amendment rule rigidity – in our analysis.
• 6
  Once more, we averaged the responses to suit our temporal coverage.
• 7
  Available at ˂ https://comparativeconstitutionsproject.org/˃ . Accessed June 05, 2023. We listed 116 rights following Elkins, Ginsburg, and Melton (2014).
• 8
  The complete results for the analysis of necessity, including the presence and absence of each condition, are available in Table 06 in the Appendix.
• 9
  The respective three 2x2 tables for each pathway of the necessity test are available in the Appendix, specifically in Table 07, Table 08, and Table 09, in the Appendix.
• 10
  As subsequent robustness tests demonstrate, this decision does not affect our overall results.
• 11
  The number of logically possible combinations is calculated through a simple factorial of 2^n, where n is the number of conditions in the analysis.
• 12
  Despite Chile’s involvement in a constitution-making process, it remains uncertain when, or even if, a new constitution will be enacted. At the time of writing, as of June, 2023, Chileans have just rejected the constitution proposed by a Constituent Assembly. Hence, as we cannot predict what will happen next, we categorized Chile as a country that has not undergone a process of constitutional replacement under democratic rule.
• 13
  Remainders are combinations of configurations that do not cover any empirical instances.
• 14
  In other words, the logical remainders that contain the negation of RIG (the necessary condition) have been excluded from the analysis, as they represent a sort of implausible counterfactuals (MELLO, 2022; SCHNEIDER and WAGEMANN, 2012).
• 15
  Given the fairly individual nature of the findings, we chose to report the analysis for constitutional replacement in the Appendix. More specifically, the necessity analysis, the 2x2 table, the truth table, and the solutions are in Table 12 through Table 18 (Appendix).
• 16
  Namely, Argentina, Costa Rica, Guatemala, Mexico, El Salvador, and Uruguay.
• 17
  Chile continues to diverge from the rest of the cases even when all plausible logical remainders are included in the analysis. The results based on the parsimonious and conservative solutions are available in the Appendix, specifically in Table 10 and Table 11.
• 18
  Conversely, a history of continuous rewriting of constitutional law creates opportunities for future replacements. For instance, Luisa González, the presidential candidate for the ‘correísta’ movement in the 2023 Ecuadorian presidential contest, has taken a firm stance in support of drafting a new constitution in the country (FASSIO, 2023). Previously, former Ecuadorian President Guillermo Lasso had expressed his intention to establish a constituent body while still in the midst of the electoral campaign (LATIN NEWS DAILY REPORT, 2017). The bottom line is that it might be difficult to reverse the trend when constitutional replacements become the norm.
• *
  This article was presented at the 13th Meeting of the Brazilian Political Science Association (ABCP), which took place from September 19 to September 23, 2022, in an online format, and was named as the best paper presented in the subfield of "Comparative Politics".
• Data replication: https://doi.org/10.7910/DVN/DOR817

Edited by

Revised by Karin Blikstad

Data availability

Data replication: https://doi.org/10.7910/DVN/DOR817

Publication Dates

• Publication in this collection
  09 Sept 2024
• Date of issue
  Jan-Apr 2025

History

• Received
  01 Dec 2022
• Accepted
  02 Nov 2023