The ORE of Politics will be available for subscription in late September. Speak to your Oxford representative or contact us to find out more.

Dismiss
Show Summary Details

Page of

PRINTED FROM the OXFORD RESEARCH ENCYCLOPEDIA, POLITICS (politics.oxfordre.com). (c) Oxford University Press USA, 2016. All Rights Reserved. Personal use only; commercial use is strictly prohibited. Please see applicable Privacy Policy and Legal Notice (for details see Privacy Policy).

date: 17 August 2017

Constitution-Making in Comparative Perspective

Summary and Keywords

Constitution-making has been a central political activity in the modern era. Enacting a new constitution was an essential ingredient in the foundation of republics, the creation of new states, the inauguration of democratic regimes, and the reequilibration of democracies during or after a political crisis. Constitution writing has also become a crucial part of the process of overcoming a legacy of violent internal conflict and a component of authoritarian regimes that seek to gain legitimacy by emulating the formalities of representative democracies. This article surveys the most important concepts and issues related to the comparative analysis of constitution-making. Although it draws examples from constitutions made in a wide variety of settings, special attention is paid to constitutional texts adopted or implemented under competitive conditions.

Keywords: Constitution-making, written constitutions, constituent power, democracy, representation, popular participation

Introduction

Constitution-making has been a central political activity in the modern era. Enacting a new constitution was an essential ingredient in the foundation of republics, the creation of new states, the inauguration of democratic regimes, and the reequilibration of democracies during or after a political crisis. Constitution writing has also become a crucial part of the process of overcoming a legacy of violent internal conflict and a component of authoritarian regimes that seek to gain legitimacy by emulating the formalities of representative democracies. This article surveys the most important concepts and issues related to the comparative analysis of constitution-making. Although it draws examples from constitutions made in a wide variety of settings, special attention is paid to constitutional texts adopted or implemented under competitive conditions.

The text begins by defining the concept of constitution-making and discussing its relationship with the normative theory of constituent power. The following sections map the events that have triggered constitution-making in the world, as well as the key procedural and organizational features that characterize this process. The article will then discuss the potential effects of constitution-making on constitutional choice, constitutional endurance, and democracy. Finally, the international dimensions of constitution-making will be examined. A brief conclusion analyzes the importance of constitution-making in the contemporary world and the need to bring together legal, political, and historical approaches in the study of this process.

Making and Amending Constitutions

Constitution-making can be defined as a temporally limited process in which a group of political actors engage in the drafting, discussion, and approval of a formal document that intends to regulate the basic structure of the state and the political regime. This written text, which in most countries is called “the constitution,” includes provisions defining the channels of access to principal government positions, the allocation of powers among the branches and levels of government, and fundamental individual rights. Most written constitutions also contain rules establishing procedures for amending them and the conditions under which constitutional provisions can be suspended. This definition is somewhat restrictive, though. In a broad sense, constitution-making may include rules adopted via amendment or constitutional adjudication. It also may cover the emergence of or changes to unwritten conventions that are part of the constitution in the sociological sense. Such a comprehensive understanding, however, creates obstacles for the comparative analysis of constitution-making as a distinctive form of constitutional transformation.

Similar to the adoption of formal amendments, the production of a new constitutional text can be observed and compared directly and systematically. This analysis is almost impossible if one wants to trace the origins and content of informal conventions and judicial rulings in a relatively large number of cases. At the same time, given their political and symbolic importance, it is also convenient to distinguish new constitutions from mere amendments or revisions to an existing constitutional text. This distinction, however, demands some elaboration.

A classical typology attempts to distinguish amendments (or partial reforms) from new constitutions by their content. Amendments are supposed to be minor modifications that do not change the nature of the existing constitutional order. Replacements, by contrast, are said to alter the fundamental structure of the state or the political system (Levinson, 1995; Murphy, 2007, p. 506; Widner & Contiades, 2013). Based on this distinction, some constitutions declare that certain principles or provisions are unamendable, and constitutional courts occasionally rule that revisions that alter the basic structure of the state are invalid (Jacobsohn, 2006). Empirically, however, whereas a new constitution may reproduce verbatim the content of the previous one, an amendment may alter many fundamental aspects of the existing constitutional structure. For this reason, and perhaps only in extreme cases, it is difficult to determine in an objective way whether a given constitutional change is fundamental.

An alternative way of distinguishing between amendments and replacements is based on procedures. From this perspective, it has been argued that a formal constitutional change is an amendment when its adoption is regulated by the constitution in force, and a replacement when it is not (Elkins, Ginsburg, & Melton, 2009, p. 55). This distinction rightly takes into account the fact that most constitutions in the world have been adopted extralegally because there was no preexisting constitution, or there was a constitution, but either it did not regulate its own replacement or its replacement procedure was not adhered to. As we will see, however, several constitutions in the world have been enacted following established procedures. A different perspective, also focused on procedures, looks at the approval body. New constitutions are usually adopted by a popularly elected constituent convention, whereas amendments tend to be passed by ordinary legislatures operating under special procedures. New constitutions, however, can be adopted by an ordinary legislature, and partial revisions may require the election of a special convention.1

Ultimately, the distinction between replacing and amending constitutions is political. It is true that whereas amendments preserve the legal continuity of the existing constitution, constitutional replacements imply its abrogation. This legal effect, however, is not primarily based on the content of the reforms or on the procedures by which they are adopted. Rather, it reflects the will and authority of reformers to make their decisions accepted as a new constitutional order at a given historical juncture.2 It is for this contingent political reason that constitution-making is associated with a fundamental legal change that must be differentiated from minor revisions of existing structures.

Constitution-Making Power

The notion of constitution-making presupposes that constitutions are human artifacts and, more specifically, written documents. This idea is distinctly modern and was born with the great revolutions of the late 18th century. Some of the central political goals of the American Revolution were crystallized in instruments containing a set of formal rules, such as the 1776 Declaration of Independence and the Articles of Confederation adopted in 1781. The same can be said of the French Revolution, with the Declaration of the Rights of Man and of the Citizen of 1789 and the 1791 constitution. However, the key feature of these documents was not simply their written form, but the fact that they were supposed to be above ordinary legislation and binding on state authorities. This superior legal status originally derived from the claim that constitutions are made by the people, as the only legitimate holders of constituent power.3

The theory of democratic constituent power was developed during the 16th century in reaction to the doctrine that the right to make and repeal both ordinary and fundamental laws belongs to the prince or king (see Bodin, 1992, pp. 56–58).4 One of the first thinkers to give clear formulation to this theory was Althusius, who argued that only the people or the associated members of the commonwealth have the power to establish the fundamental law of the realm. He proposed that constituent and constituted powers must be strictly distinguished by their position in the polity: “the people is prior to, more important than, and superior to its governors, just as every constituting body is prior and superior to what is constituted by it” (Althusius, 1995, pp. 72–73).

Whereas the notion of the constituent power of the people was initially developed to oppose the power of kings, it was later affirmed against the power of parliaments. Because collective bodies such as estates and parliaments opposed the power of kings during the initial struggles against royal absolutism, there was a thin line between the idea that those bodies represented the interests of the people and the claim that they incarnated the sovereign’s authority. The Levellers’ proposal that the constituent power of the people should not be exercised by parliament, and their attempt to limit its powers in the 1647 Agreement of the People, broke with the association between parliaments and popular power (see Loughlin, 2007, pp. 36–37). George Lawson (2004, pp. 47–48) summarized the Levellers’ political program in his Politica Sacra et Civilis, where he argued that only the community has the power to create and alter the constitution and that this power is “above the power of a parliament.” According to this view, neither the executive nor the legislature should be able to alter the fundamentals of the constitution.

The constituent power theory evolved in parallel with theories of resistance and revolution. Whereas the initial opponents of absolutism placed the right to resist oppressive government in the hands of magistrates or bodies that were representative of the people, this right was later attributed to the governed. As is well known, John Locke popularized the idea that when the government has forfeited its rightful ends and arbitrarily takes away the lives, liberties, or fortunes of the people, the latter have a right to “appeal to heaven” and overthrow a tyrannical government (see Second Treatise, XIX, 221). Locke’s view became the foundation of the American Revolution, whose Declaration of Independence established that the people have the right “to alter or to abolish” their governments whenever they become destructive of their rightful ends. A similar right was established during the French Revolution.

The constituent power theory is clearly not descriptive; it seeks to provide a normative standpoint from which to evaluate actual practices of constitution-making (Tushnet, 2014, p. 15). Historically, most constitutions in the world have been made by elites, with various levels of popular participation that were not always genuine or meaningful. From a comparative perspective, the constituent power theory has not been adopted in several legal systems that enable the legislature to adopt a new constitution. The influence of the theory has also varied by historical period and region of the world. Nevertheless, the notion that fundamental constitutional changes should only derive from the people as the sovereign authority is still strong and has both theoretical and practical implications.

Many important concepts and views about constitution-making in political and constitutional theory derive, implicitly or explicitly, from the revolutionary doctrine of the constituent power. Such is the case with the distinction between ordinary and constitutional lawmaking, the difference between new constitutions and constitutional amendments, the need for direct popular participation in fundamental constitutional changes, and the role of special conventions in the creation or replacement of constitutions. These notions are not merely theoretical constructions; they often shape the rulings of constitutional courts and affect policy debates about the design of constitution-making procedures.5

The main problem with the use of the constituent power theory in assessing or designing actual episodes of constitution-making is that the institutional implications of this theory are deeply ambiguous and contradictory. Even within the revolutionary tradition, there was no single theory or practice of the constituent power. Whereas in the United States, “the people” was understood as a plural association of preconstituted entities, in France, “the nation” was conceived as a single collective subject that always remained in a state of nature with regard to existing institutions (Madison, Jay, & Hamilton, 1987, pp. 255–259; Sieyes, 2003, pp. 137–138; Arendt, 1963). Whereas for the American federalists, constitution-makers were different and separate from the people, for members of the French Third Estate, delegates in a constituent assembly were extraordinary representatives that stood for the nation itself.6 Finally, there was no consensus in the revolutionary tradition about the nature and powers of the constituent body. Although the Federal Convention proposed a new constitution instead of reforms to the existing one, it never absorbed legislative tasks or claimed to have sovereign powers. By contrast, the French National Assembly doubled as a constituent and legislative body in addition to exercising executive functions (Elster, 1994).

The institutional consequences of the theory are even more confusing outside the revolutionary context. The idea that the power to make a constitution preexists and is superior to the legal order in force is potentially disruptive in an established democracy, where citizens regularly express their will through legally regulated acts. Can this power be resumed any time a majority of the people desires to introduce fundamental changes, or can it be invoked only in the event of violent overthrow of the political system? Is some form of constitutional regulation necessary for this power to be executed, or are constitutional provisions unnecessary or irrelevant given its extralegal nature? Classical and contemporary authors in constitutional theory provide radically opposite answers to these questions.7 Historical examples, on the other hand, may lend support to either position.8

Causes of Constitution-Making

The adoption of a new constitution usually demands time, resources, and information. In a politically competitive environment, constitution-making generally requires a popularly elected constituent assembly and multiparty negotiations. In addition, changing constitutional structures impose strong informational requirements to anticipate the effects of different rules under changing political conditions. These costs suggest that constitutions should be adopted only as a consequence of major political events.

Profound political changes, such as the founding of a state or a regime transition, usually require a new legality. New states typically mark their birth by enacting a constitution. The independence processes that took place in the Americas from the late 18th century to the first half of the 19th century, in Africa during the 1960s and 1970s, and in Eastern Europe from 1990 to 1995 have all coincided with cycles of constitution-making. Something similar tends to occur during regime transitions, although in these cases the scope for variation is greater. Authoritarian regimes may simply suspend an existing democratic constitution. At their inauguration, democratic regimes may opt to restore a preauthoritarian constitution, maintain a constitution enacted during the authoritarian period, or introduce amendments to adapt an authoritarian constitution to new political conditions. The choice depends on which constitution is considered most capable of effectively and legitimately organizing the new democratic regime, as well as on the balance of forces between the outgoing authoritarian regime and democratic parties (Geddes, 1990). This variation may be seen in Latin America between 1978 and 1994 and in Eastern Europe between 1989 and 1995, where more than one-third of the transitions to democracy did not lead to the adoption of a new constitution.

Constitutions also may be replaced within a democratic order, particularly when basic features of the current constitution become dysfunctional at a particular historical juncture. Constitutions are governance structures that organize electoral competition, enable representatives to provide public goods, and maintain citizen support for a political regime. When constitutions fail to perform these tasks, politicians are likely to have an incentive to replace them, usually under popular pressure for reform. For instance, the Algerian crisis in France and the widespread perception of lack of government effectiveness under the 1946 constitution led to its replacement in 1958 (Foyer, 1988). Similarly, a deep political crisis in Colombia during the late 1980s created popular demand for a constituent convention that eventually replaced the 1886 constitution in 1991 (Negretto, 2013).

Not all constitutional replacements, however, occur in response to dramatic political events. They also may take place as a result of significant shifts in party competition, as when established parties collapse or decline or when new parties and political leaders emerge. Alterations in the balance of power among parties should usually be addressed by means of amendment (Negretto, 2012). However, either because reformers lack congressional support to pass an amendment or because they want to signal a deeper form of institutional transformation, the change may result in a new constitution. In these cases, since the public justification for replacing a constitution cannot be self-serving, political actors often resort to arguments unrelated to changes in the distribution of partisan power to set in motion a constitution-making process. Although President Juan Perón adopted a new constitution in Argentina in 1949 to secure an electoral advantage for himself and his rising political party, he invoked the need to modernize the 1853 constitution (Negretto, 2013). Similarly, President Boris Yeltsin used the 1993 constitutional crisis in Russia as the background to justify the need for a new constitution that ended up strengthening his powers (Partlett, 2012).

Democracies also may replace their constitutions to simply codify existing practices, as well as to modernize, clarify, or unify constitutional norms. In 1974, Sweden adopted a new constitution to formalize the parliamentary democracy that had evolved over time in the country (Congleton, 2012). Switzerland also enacted a constitution in 1999 to update provisions related to its government structure and expand fundamental rights (Biaggini, 2011). Finland’s 2000 constitution was meant to legalize existing practices, rewrite provisions, and unify various acts of constitutional status in a single text (Suksi, 2011).

Procedural Features

Making a constitution entails setting up a relatively complex institutional structure with a wide variety of rules. I will focus on four procedural features: the legal framework of the process, the type of constitution-making body, the rules of representation, and the channels of citizen involvement.9 Empirically, these features capture the most important political decisions that adoption of a new constitution demands (Miller, 2010; Brandt, Cotrell, Ghai, & Regan, 2011). Normatively, they relate to the concerns of scholars who focus on the legality and legitimacy of constitution-making or on the rationality of the process as an act of collective choice (Arato, 2016; Fishkin, 2011; Elster, 2013).

Legal Framework

Once the decision to replace the constitution is made, the second most important choice is whether the existing constitution (in its original or amended form) or a norm of constitutional status will regulate basic aspects of the process. It determines the degree of influence that preexisting institutions and political actors will have during the drafting and approval stages, and thus whether substantive transformations will be moderate or radical.

It is often assumed that new constitutions require a legal break with the past. Since constitutions are often replaced when they fail to work as legitimate or effective governance structures, rupture with the preexisting institutions may appear to be necessary to provide a firm foundation to the new order (Kay, 1987, 2011). This idea reflects a common but not universal condition in constitution-making. From 1900 to 2015, 40 out of a sample of 126 constitutional replacements followed the rules established in a preexisting constitution or norm of constitutional status.10 Legal continuity thus characterizes a small but nonnegligible minority of episodes of constitution-making.

The extralegal enactment of constitutions is frequent at the creation of a new state or after a revolution. In these cases, there is no preexisting constitution, the one in force is considered illegitimate, or appropriate procedure for its replacement is lacking. When the American colonies adopted the Articles of Confederation in 1781, there was no preestablished constitution to direct the process. Later on, in 1787, the Federalists bypassed the amendment procedure of the Articles because it was considered inadequate under the circumstances (Ackerman & Katyal, 1995). In the wake of the December 1989 revolution in Romania, the 1965 constitution was suspended before being replaced in 1991 by a popularly elected constituent convention. The same may happen in a transition to democracy, when prodemocratic forces gain the upper hand and want to signal a new beginning. After the collapse of the military junta in Greece in 1974, the democratic opposition formed a government of national unity that suspended the 1968 constitution and called for the election of a constituent parliament (Contiades & Tassopoulos, 2013).

However, even in cases of deep political change, constitution-makers may feel the need to preserve continuity with the old legal order. For instance, when a territory enjoyed political autonomy and had a parliament before declaring independence, legal continuity could be a way for local powers to keep control or influence over the transition toward a new state. Immediately after independence in 1917, the Finnish parliament adopted the 1919 constitution, which was created following the reform procedures established in the 1906 Parliament Act (Suksi, 2011). In preparation for its independence from Denmark, the Icelandic parliament passed a Constitutional Act in 1942, which changed the amendment procedure of the 1920 constitution to regulate the adoption of a new constitution in 1944 (Thorarensen, 2011).

In a transition to democracy, continuity between the old and the new legal order is often part of a compromise between the outgoing authoritarian regime and democratic forces to make the process more predictable and changes less radical. This was the case in Brazil between 1986 and 1988. Whereas the most progressive sectors of the opposition wanted to replace the 1967 constitution and enact a new democratic constitution in an independently elected constituent convention, the military preferred keeping the 1967 constitution through a series of amendments implemented by the legislature elected in 1982. As neither actor could impose its preferred option, a compromise was reached so that an amendment to the 1967 constitution would legalize the making of a new constitution by the existing congress (Martinez Lara, 1992; Rosenn, 2010). Similarly, legal continuity in a democratic transition could be a strategy to build trust and consensus among the participants after a long period of conflict (Arato, 1995; Jackson, 2008). This was the role of the interim constitution enacted in South Africa in 1993, which provided both a temporary institutional structure of government and a framework for negotiating the final constitution.11

If the need for a peaceful transfer of power may induce reformers to avoid a legal break in a transition to democracy, there are stronger reasons to do so when constitutions are replaced within an already-existing democratic regime. In this context, it is likely that constituted powers would demand some role in the process by using the preexisting constitutional framework. One obstacle to this strategy is that most constitutions lack an explicit process for their replacement. In this situation, if government and opposition and state powers (in particular, the executive and the legislature) do not reach an agreement about the regulation of the process, then a legal break is likely. For instance, in the making of the 1999 Venezuelan constitution and the 2008 Ecuadorean constitution, the president established the rules of the process unilaterally, both transgressing the constitution and bypassing the congress (Brewer Carías, 2002, 2011; Negretto, 2015, 2017a). However, different solutions have been implemented to prevent this outcome.

Some democratic constitutions do have a procedure through which they can be either amended or replaced. This is the case of parliamentary constitutions that enable the legislature to amend or repeal an existing constitution or constitutional act. In 1953, Denmark adopted a new constitution following Article 88 of the 1915 Constitutional Act, which established a uniform parliamentary procedure of reform whenever the assembly passed a bill for the purpose of constitutional revision (Krunke, 2013). In 1974, the Swedish parliament replaced the 1809 instrument of government following a procedure established in Articles 81 and 82 of this document (Congleton, 2012). Similarly, in 2000, the Finnish parliament approved a new constitution following the procedure established in Section 67 of the 1928 Constitutional Act (Suksi, 2011).

A similar method is used when the constitution enables an elected legislature or convention to adopt either partial or total reforms. In Uruguay, all constitutions since 1934 have authorized the legislature to propose and pass total reforms. These rules provided the basis for the adoption of the country’s 1942, 1952, and 1967 constitutions (Negretto, 2017b).12 In Argentina, the same procedure applies for partial or total reforms, but in either case, a declaration made in the legislature of necessity of reforms must be followed by the election of a constituent convention responsible for enacting the reforms. This mechanism was used in the adoption of the 1949 and 1994 constitutions.

Some democratic constitutions regulate two different procedures for adopting partial or total reforms. For instance, articles 119 and 120 of the 1874 Swiss constitution established a special procedure of total reform (known as “Totalrevision”) by which Parliament was enabled to adopt a new constitution in 1999 (Biaggini, 2011). Comparatively few constitutions, most of which are located in Latin America, incorporate a distinction between amending and replacing the constitution and delegate the latter task to a special body. Such is the case of the 1991 constitution of Bulgaria, the 1991 constitution of Colombia, the 1999 constitution of Venezuela, the 2008 constitution of Ecuador, or the 2009 constitution of Bolivia.

Even in the absence of preexisting regulation, there are various ways of creating a bridge from one constitutional order to another in a democratic regime. One possibility is amending the existing amendment procedure to provide explicit rules for replacing the constitution. After the forced resignation of Gonzalo Sánchez de Lozada in Bolivia in 2003, a consensus emerged among traditional and new political forces that the country needed a new constitution. The 1967 Bolivian constitution, however, allowed only partial amendments. To channel the process in a legal manner, the constitution was amended in Congress in February 2004 to make total reform of the constitution possible. It was based on this reform that in 2006, the Bolivian Congress passed a law regulating the election of a constituent convention to adopt a new constitution (Böhrt Irahola, 2013). A similar strategy was followed in Thailand, where the 1997 constitution was drafted by a special body and then approved by parliament based on a 1996 amendment of the 1978 constitution (Sucharitkul, 1998).

Another way of achieving legal continuity is to regulate the process by means of an institutional agreement between the executive and the legislature. After the irregular impeachment of Ecuadorian president Abdalá Bucaram, an interim president, in agreement with Congress, convened a referendum asking for authorization to elect a constituent convention (Negretto, 2013). As a result of the popular support obtained in the referendum, Congress passed a transitory constitutional provision to regulate the election and tasks of the convention. In practical terms, although the existing amendment procedure was not formally amended, the addition of a transitory provision to the constitution had a similar effect. In both cases, the old constitution worked as a sort of interim or provisional legal framework to regulate the constituent process.

Legal continuity in the making of new constitutions within democratic orders is in fact quite common. From 1900 to 2015, a total of 25 constitutions have been created worldwide in the context of a democratic regime that was at least five years old.13 In the vast majority of these cases (20), the process that led to the new constitution was regulated by the existing constitutional text (whether in its original or amended version) or by an instrument of constitutional status, such as an interim constitution.14

Constitution-Making Body

The constitution-making body is the forum where a new constitutional text is drafted, approved, or both. As such, the nature, selection, and composition of this assembly shape the type of popular representation implemented during the process. It also affects the degree to which the writing of the new constitution is independent of and different from ordinary lawmaking procedures.

A variety of formal and informal collective bodies may participate at different stages of constitution-making; constitutional commissions, round tables, national conferences, constituent conventions, and constituent legislatures. The last two, however, are the most common instances that work as true constituent assemblies where constitutional texts are deliberated, negotiated, and finally voted on (Widner, 2008; Ginsburg, Elkins, & Blount, 2009).15 Executive bodies, in the form of presidential commissions or executive advisory councils, have also worked as approval bodies, but they are rarely observed in democratic or competitive settings.

Constituent conventions are assemblies created for the sole or primary purpose of adopting or proposing a new constitution.16 Given their special commission, constituent conventions are transitional in nature: they must be dissolved after the task is completed. An assembly exclusively focused on writing a constitution is very rare. It should exist only where an ordinary legislature or another government body is simultaneously responsible for enacting regular laws. Yet in the typical scenarios of constitution-making, such as the foundation of a new state, a revolution, or a transition to democracy, there is often no legislature, or else the legislature that existed was dissolved or ceased to function in a normal way. This means that unless an interim government is responsible for ordinary lawmaking, even a convention whose main goal is to draft and approve a new constitution may be forced to legislate or exercise other government functions, albeit temporarily.

The paradigmatic example of a constituent convention is the Philadelphia Constitutional Convention of 1787. Similar cases can be found in the 1814 Norwegian Constituent Assembly, the 1848–1849 Danish Constituent Assembly, the 1948–1949 German Parliamentary Council, and the 1976 Portuguese Constituent Assembly. In general, however, this type of constituent assembly has been used in the history of relatively few countries or regions (Elster, 2006). From 1900 to 2015, only 38 out of 124 deliberative bodies created to approve a new constitution may be classified as constituent conventions.17

Most of the assemblies of this type that have existed in the world were created in the American states. According to Hoar (1917), the New Hampshire conventions of 1778 and 1781–1783 and the Massachusetts convention of 1780 initiated the constituent convention movement in the United States. The practice proliferated throughout the 19th century, and several states resorted to these bodies during the 20th century. Since the 1970s, however, there has been a sharp decline in the use of conventions for replacing or revising constitutions (Williams, 1996). Experts in U.S. state constitutional law have counted a total of 233 constituent conventions between 1776 and 2005 (Dinan, 2009, pp. 7–8). Most of these conventions were convened to draft or revise state constitutions at the time of important political events at the national or local level that took place between the late 18th century and the early decades of the 20th century, such as the independence movement, the inclusion of new states into the union, the civil war, the Reconstruction, or the inclusion of participatory institutions during the progressive era.

At the national level, constituent conventions also have been frequent in Latin America.18 Most of the first constitutions enacted after independence were made by assemblies that, along with framing a new constitution, were responsible for performing legislative and other government functions. The use of constituent conventions, however, emerged over time with the adoption of the 1830 constitutions of Colombia and Venezuela, the 1853 constitution of Argentina, the 1857 constitution of Mexico, and the 1870 constitution of Paraguay. The use of these bodies continued well into the 20th century and the first decade of the 21st century. According to Negretto (2017b), of a total of 83 constitution-making processes that took place in the region from 1900 to 2014, 26 (31%) used a special convention as the constituent body.19 Just as the conventions of the 19th century were convened to adopt the first state constitution or revise the existing one after a civil war or revolution, during the 20th century, most conventions were used to create new constitutions after regime change. More recently, special conventions also have been used to replace the existing constitutional framework within a democratic order.

Constituent legislatures are assemblies that work as ordinary legislatures both during the drafting and after the approval of a new constitution. In contrast to special conventions, they are the most common form of constituent body. From 1900 to 2015, 68 out of 124 deliberative bodies created to approve a new constitution qualify as constituent legislatures.20 These bodies differ from the point of view of the source of their authority to adopt a new constitution. Using this perspective, Jon Elster (2006, 2013) has distinguished among three types of mixed constitution-making bodies: mandated constituent legislatures, self-created constituent legislatures, and self-created legislating assemblies.

Mandated constituent legislatures are assemblies elected to enact a new constitution and pass ordinary legislation. They are frequently used in transitions to democracy, where there is either no legislature at the beginning of the process or the existing one lacks legitimacy to assume a constituent function. Examples of these bodies can be found in the making of the 1975 Greek constitution, the 1931 and 1978 Spanish constitutions, the 1992 Slovakian constitution, and the 1946 and 1988 Brazilian constitutions.

Self-created constituent legislatures are ordinary legislatures that decide (without an electoral mandate and in the absence of preexisting constitutional authorization) to transform themselves into a constituent body.21 These bodies have been common during revolutions and independence processes. For instance, the 1776 constitutions of North Carolina, South Carolina, Georgia, Virginia, and New Jersey, as well as the continuation of the charters of Massachusetts, Rhode Island, and Connecticut were decided by ordinary legislative bodies that were self-appointed as constituent assemblies (Hoar, 1917, p. 4). Similarly, we may find this type of body in the adoption of some independence constitutions in Eastern Europe, such as those of Croatia in 1990 and Estonia in 1992 (Mirth, 1992; Kiris, 1991). The drafting of a new constitution by a self-appointed constituent legislature also works as a nonlegal but politically expedient method for enacting a new constitution within an existing constitutional order, usually during authoritarian regimes. Constitution-making bodies of this type adopted most constitutions in Bolivia during the 20th century and in Venezuela from 1904 to 1925.

Self-created legislating assemblies are a hybrid type of mixed body that is intermediate between constituent conventions and ordinary legislatures. These are constituent assemblies that were supposed to dissolve after enacting a new constitution but decided on their own authority to continue as ordinary legislatures. The Indian constituent assembly of 1946–1950 and the 1934 constituent assembly of Brazil may fit this category.

Elster’s classification is not exhaustive, however. A missing category of mixed constituent bodies is what we might call constitutionally authorized constituent legislatures. These are ordinary legislatures that can turn themselves into constituent bodies following a special procedure established in the existing constitution or legal instrument of constitutional status. This occurs when the existing reform procedure enables the ordinary legislature to replace the constitution in force. Constitutionally authorized constituent legislatures have been used in some transitions to democracy where the old constitution was amended or an interim constitution created to enable parliament to enact a new constitution.22 Yet this mechanism is much more common in the context of established democracies. In particular, 16 of the 25 constitutions adopted between 1900 and 2015 in a democracy at least 5 years old were approved by a constitutionality authorized constituent legislature.

Rules of Representation

Rules of representation determine how inclusive or exclusive the composition of the constitution-making body is from a political and social point of view, thus affecting the degree of deliberation and consensus that can be observed during the drafting and approval of the text. The main rules in this regard are the methods for selecting members of the constituent body and the quorum and voting rules used during the process.

The mechanisms for the selection of members to a constituent body are popular election, appointment, selection by lot, or a mixed method. In most contemporary episodes of constitution-making in most regions of the world, members of the constituent body are elected by popular vote. For instance, 35 out of 41 and 39 out of 43 constitution-making bodies convened in Latin America and the Caribbean and in Western and Eastern Europe between 1900 and 2015 were popularly elected. Selection by other methods is, however, still very important in some regions of the world, such as sub-Saharan Africa, where only 6 out of 23 constitution-making bodies formed during the same period have been elected. Most have been appointed or selected through mixed methods.23

When the constitution-making body is popularly elected, proportional representation (PR) formulas are the most common rules used since the early decades of the 20th century to allocate constituent assembly seats among parties.24 This trend reflects a similar pattern in the evolution of systems to elect ordinary legislatures (Colomer, 2004). The degree of proportionality can vary a great deal depending on the formula, the size of the assembly, and the average magnitude of the districts from which delegates are elected. The 70-member constituent convention elected in Colombia in December 1990 by a Hare PR formula using the whole country as a single national district counts as the most inclusive. By contrast, the 1966 Dominican constitution was adopted by a 74-member constituent legislature elected by a d’Hondt PR formula in 27 districts, with a low mean magnitude of 2.7.25 Somewhere in between, the 421 members of the constituent legislature elected in Germany in 1919 were selected by a d’Hondt PR formula in 36 districts, with a relatively high mean magnitude of 11.7 (Nohlen & Stöver, 2010, pp. 745–753). As a consequence, whereas more than one party was necessary to pass reforms in the Colombian and German assemblies, a single party had control over the constituent body in the Dominican Republic.

In spite of the widespread use of PR, the constituent body has been elected by majoritarian rules in some cases. These rules often emerge from the desire of a dominant party or political group to secure an electoral advantage. In the June 1931 election of the Constituent Cortes in Spain, a highly majoritarian formula was used that on average provided almost 80% of the seats in a given electoral district to the party obtaining a plurality of votes (Nohlen & Stöver, 2010). A similar formula, which provided up to 66% of the seats to the party supported by a plurality of votes in the district, was used to elect the 1949 constituent convention in Argentina (Negretto, 2013). These formulas led to the predictable dominance of one political group (Republicans and Peronists, respectively, in these cases) over the assembly. Yet whether or not one party benefits from them, majoritarian rules also may be used out of tradition. The implementation of a first-past-the-post system to elect the 2007–2010 constituent legislature in Kenya had less to do with the purpose of controlling the assembly by a single party than with the fact that this system was inherited from the British, who governed the country from 1895 to 1963.

When the constituent body is elected, the election is often restricted to candidates fielded by parties. There are, however, experiences of delegates being elected on a nonpartisan basis. The 2010 Icelandic constituent convention was made up of 25 citizens elected from a pool of 522 candidates who had to gather signatures to appear on the ballot (Meuwese, 2013). Even if most delegates are elected on a partisan basis, the number of elections in the last two decades allowing independent candidates to compete in constituent assembly elections has increased. In addition, a recent but growing practice in the election of constituent bodies is the adoption of gender quotas to grant a more equitable distribution of seats between men and women.

The use of mixed methods of selection, such as those combining election and appointment, often seeks to include representatives of previously excluded social groups, to balance the presence of popular representatives with the participation of party or government delegates, or both. In Colombia, for instance, although the 70 members of the convention were elected, four seats were added (two with voting rights) for appointed members of guerrilla groups (Negretto, 2013). The 1998 Australian constituent convention was made up of 152 delegates—76 elected by the people, 40 appointed by parties from the federal and local legislatures, and 36 selected from outside parliament to reflect the interests of aboriginal groups, women, and territories (Renwick, 2014).

There are no experiences of citizen assemblies (i.e., assemblies where all or a majority of the delegates are ordinary citizens selected by lot) for the purpose of drafting and approving a wholly new constitution. Assemblies exclusively composed of citizens selected by lot have been formed in British Columbia in 2004, in the Netherlands in 2006, and in Ontario in 2007 (Fournier, Van Der Kolk, Kenneth Carty, Blais, & Rose, 2011). In all these cases, however, the assembly focused on electoral reform rather than on broader constitutional reform. In Ireland, a convention working between 2012 and 2014 had a composition in which two-thirds of its members were randomly selected citizens and one-third were politicians appointed by parties. This convention, however, was responsible for proposing specific amendments to be approved in parliament rather than approving or proposing a new constitutional text.26

Another set of key procedural rules that affects representative pluralism in constitution-making consists of the quorum and voting rules, particularly those for approving the final text. Just like ordinary legislatures, most constituent assemblies require an absolute majority (more than 50%) of the total membership in order to hold a valid session. In the case of constituent conventions, a similar rule has also been required for the approval of reforms in plenary sessions, although sometimes a simple majority (more than 50% of those present and voting) has been sufficient to make decisions. Qualified majority requirements have been rare for constituent assemblies in general, and for conventions in particular.27 When qualified majority voting is required, a deadlock-breaking device may be established, such as submitting the final approval of the constitution to a popular referendum.

In some cases, although only an absolute majority threshold was required for making final decisions, complementary procedures were adopted to reach an adequate level of consensus. In the 1991 Colombian constituent convention, for instance, the three main political parties agreed to share the presidency of the assembly, and in addition, they allowed members of minority parties to preside over the various committees responsible for making proposals on different parts of the constitution (Dugas, 1993).

Citizen Participation

The basic form of citizen participation in constitution-making usually takes place through the election of representatives who will propose or decide on the content of revisions. But this is an indirect form of citizen involvement, via the representative process activated by elections. To be precise, popular participation should be restricted to instances of direct citizen involvement, such as formal or informal public consultation processes, the capacity to submit reform proposals, and the right to vote on ratification or rejection of the new text. These are all mechanisms that attempt to enhance the democratic legitimacy of the process by providing ordinary citizens with the opportunity to express their preferences or to give actual consent to the changes that representatives propose.

Citizen participation before or during the drafting process usually takes the form of public consultations in open meetings, participatory forums, or public hearings. They can encompass a wide variety of activities, from the deliberate effort to truly engage citizens in public debate about the new constitution and take their views into account, to more formal and window-dressing practices intended to create an aura of public involvement within an essentially elitist process. It is often difficult to draw the exact line between genuine and sham participation and determine its exact impact on the final product, although the type, timing, and duration of participatory channels may serve as heuristically valuable clues (Wallis, 2014). From this perspective, an example of true and consequential participation may be the extensive process of public education and consultation that took place in South Africa from February 1995 to February 1996 before the final text of the constitution was drafted (Ebrahim & Miller, 2010). By contrast, an instance of window-dressing participation can be observed in the adoption of the 1995 Malawi constitution, where the only form of public involvement took place during a few days in February 1994, when deliberations of the drafting body were broadcast and listeners were invited to call in for comments (Chirwa, 2003).

Ordinary citizens and civil society organizations may also be involved during the drafting process by being allowed to submit reform proposals. Processes of this type took place, for instance, in Peru between 1978 and 1979, Brazil between 1986 and 1987, El Salvador in 1983, Nicaragua between 1985 and 1986, Colombia in 1991, Bolivia between 2007 and 2008, and Ecuador in 2008. In Brazil, procedural rules allowed the submission of popular amendments supported by 30,000 voter signatures. During the process, 122 reform proposals of this type were formally submitted to the constituent legislature. In Nicaragua, 1,800 citizens submitted comments on the first draft of the constitution (Miller, 2010). Submission of reform proposals has also been part of the drafting process in recent episodes of constitution-making in sub-Saharan Africa, such as those of South Africa in 1993–1993, Ghana in 1991–1992, and Kenya in 2009–2010. Unlike public consultations, submission processes are supposed to induce citizens to take a more proactive role in expressing their needs and preferences about the new constitution. Yet the extent to which these submissions are taken into account in the final outcome is a matter of controversy.

One of the main criticisms of previous forms of citizen involvement is precisely that they may work as information-gathering mechanisms, without visible effects in terms of the final decisions made by representatives (Blount, 2011). From this perspective, direct popular voting on specific proposals of constitutional change or on the ratification or rejection of the whole text is often seen as the most important instance of mass involvement in giving actual consent to the constitution. Since information on the reform proposals is typically limited and public debate on their merits is usually superficial, this form of popular participation is obviously questionable on normative grounds (Lenowitz, 2015). Nonetheless, it is clear that at least in competitive environments, referendums have an important influence on the outcome. Whereas a referendum held at the beginning of the process creates an “upstream” constraint on the institutions that reformers will adopt, a ratification referendum works as a “downstream” constraint on reform alternatives that representatives can chose from.28

Before the actual text is drafted and approved, a referendum may be used to decide particularly important issues. For instance, on December 1974, immediately after the election of a constituent parliament, Greeks voted on a referendum about whether the head of state should be hereditary or elected. Since 70% of the voters supported an elected head of state, the republican form was incorporated into the new constitution enacted in June 1975 (Tsatsos, 1988).29 Similarly, a referendum was held in Ecuador in 1997 to consult voters about a series of electoral reforms to be later adopted by a constituent convention. In addition to or instead of asking citizens to decide about specific issues, a referendum could be used to authorize replacement of the constitution or to elect a convention when the latter is not regulated by the constitution in force. This type of referendum was recently implemented in Colombia in 1990, in Venezuela in 1998, and in Ecuador in 1997 and 2007 (Negretto, 2017a).

A more common form of referendum is one held to ratify or reject the proposed constitutional text. As a matter of practice, the first constitutions to be ratified by referendum were the 1778 Massachusetts constitution, the 1792 New Hampshire constitution, and the 1793 French constitution. During the 19th century, the people ratified relatively few national constitutions. However, the use of this type of referendum has increased significantly during the 20th century.30

Ratification referendums have been used in both authoritarian and democratic periods, which is one of the reasons why they are not a guarantee per se of genuine popular participation. For example, referendums were held to ratify the 1980 Chilean constitution, the constitution of Ghana of 1992, and the 1993 Peruvian constitution. In cases like these, support for the new constitution appears high, but the validity of the vote is usually not credible. Ratification referendums during competitive years tend to enjoy a higher degree of democratic legitimacy. Yet they have also been questioned, particularly when a single party or political group controlled the process. These referendums are criticized not so much for the existence of direct government control or coercion over the vote, as for the power asymmetry between the dominant party and the opposition and the systematic exclusion of the latter during deliberations and negotiations of the text. This has been the case of the referendums held to ratify the 1993 Russian constitution, the 1999 Venezuelan constitution, and the 2008 Ecuadorean constitution (Partlett, 2012; Negretto, 2017a).

Constitution-Making as an Explanatory Variable

Arguments about the benefits or risks of particular forms of constitution-making are often based on principle. In this perspective, scholars support a particular set of procedures from a normative point of view that usually takes popular sovereignty, democracy, or political autonomy as its starting point (Colon-Rios, 2012). A less common but also noninstrumental view evaluates the procedural rules of constitution-making based on their ability to remove obstacles to rational decision-making (Elster, 2012, 2013).31

Most analyses on constitution-making, however, start with the premise that the choice of a particular procedure (or set of procedures) would make some outcomes more likely than others. Proponents of parliamentary constitution-making usually justify their position on the pragmatic need to include existing parties in the negotiations or prevent institutional conflicts between the ordinary legislature and an independently elected convention (Holmes & Sunstein, 1995). Those who favor participatory processes of constitution-making often postulate that citizen involvement would strengthen citizen rights, reduce levels of conflict, or promote democratization (Samuels, 2006; Widner, 2008). As we will see, one also can find implicit causal arguments involved in noninstrumental perspectives about the normatively desirable features of a constitution-making process.

Process choices are certainly not irrelevant. Selecting one particular procedure rather than another may matter for normative or strategic reasons that are likely to be extremely important at the time of drafting and adopting a constitution. In the short term, and depending on the circumstances, choosing one particular procedural alternative may increase or decrease the legitimacy of the process and benefit some actors more than others. It may also stabilize or jeopardize a democratization process. However, causal claims about the effects of process choices face several analytic and empirical problems.

Given interactive effects among various procedural rules, as well as between the latter and nonprocedural variables, one obvious challenge is to select a specific rule or set of rules that can be isolated as a causal factor. An equally demanding requirement is to single out an outcome that can be observed and identified as a consequence of specific rules. In this respect, the vast majority of hypotheses about the impact of procedures look for an outcome that somehow signals success in constitution-making. Success, in turn, is usually related to the realization of one or more of the core values of constitutionalism, such as limited government and rights protection, constitutional endurance, or constitutional enforcement. Yet these may be different and not always mutually consistent outcomes.

An advocate of liberal constitutionalism may choose as an indicator of success the creation of a constitution whose design promotes the ideal of limited government. This perspective would rule out as examples of success such cases as the 1936 Soviet Constitution, whose Article 126 infamously granted the Communist Party a leading role in the state, thus providing legal foundation for the single-party dictatorship that the Soviet system in fact was (Ginsburg & Simpser, 2014). It also would disqualify the 1967 Paraguayan Constitution, whose Article 182 invested the president with the prerogative to dissolve Congress if events occurred that, according to the president’s interpretation, could endanger the normal functioning of the constitution.32

Beyond its content, however, a constitution must last a considerable time to stabilize the expectations of citizens and elites about the actions of state actors. A constitution that is replaced a few years after being promulgated (or that is not implemented at all) can hardly count as a successful constitution. From this perspective, then, a constitution that supposedly has the right content also must endure to produce the beneficial effects associated with that content (Elkins et al., 2009; Negretto, 2012).

Durability is not, however, the ultimate goal of constitutionalism. A constitution that lasts but does not matter cannot qualify as a successful constitution (Elster, 1991). In a recent important book on the determinants of constitutional endurance, Elkins, Ginsburg, and Melton (2009, p. 77) consider that maintenance of the constitution through amendments (or judicial interpretation) is in itself an act of enforcement. This view is debatable, however. Political actors can use existing amendment procedures to adapt the constitution even though its most basic provisions remain a dead letter. Several amendments were legally incorporated into the 1917 Mexican constitution during the PRI hegemony and to the 1924 Dominican Republic constitution during the Rafael Trujillo era. Yet those constitutions were largely unenforced. Constitutional durability may be a necessary condition for constitutionalism, but it is certainly not a sufficient one.

Given the variety of phenomena associated with a successful constitution-making process, it is quite possible for the same procedural rule or organizational form to have strictly opposite effects in practice across equally relevant outcomes. For instance, removing the influence of institutional and group interests in a nonpartisan constituent convention may produce an impartial and balanced design in the distribution of powers between branches of government. This design could potentially strengthen governability and generate a suitable level of rights protection. At the same time, however, the same process may provide organized political elites with an incentive to resist a constitutional structure in which they had no influence, thus compromising its chances of future survival and enforcement (Voigt, 2004).

Some authors have tried to avoid these problems by using a single outcome, such as the level of democratization, which may combine various notions of constitutional success (Carey, 2009; Mendez & Wheatley, 2013; Eisenstadt, LeVan, & Maboudi, 2015). In effect, if the constitution outlines some scheme of separation of powers and rights protection, lasts a considerable time, and is enforced, then we may end up having something close to a full constitutional democracy. The problem is that the concept of democracy has different components, and not all of them relate to the authority and binding effects of the constitution. Most current notions and measurements of representative democracy give more emphasis to how representatives gain access to power through fair and competitive elections than to how they exercise power after being elected. Some gross violations of the constitution may be taken into account, but important variations in constitutional enforcement usually go unnoticed.33 Another possibility is to use a notion of success that suits particular cases, such as conflict reduction in divided societies (Widner, 2008; Lerner, 2013). This approach would facilitate analysis of a single country or a limited set of countries, but it would be less useful for a more ambitious comparative study.

Even if one focuses on a single dimension of success in constitution-making, it is important to keep in mind the potential tradeoffs involved in the expected effects of any procedural rule. Legal continuity may guarantee a peaceful transition to democracy at the cost of maintaining authoritarian structures that could undermine the democratic regime in the long run. On the other hand, a clean legal break with the past may make deep constitutional transformations possible at the cost of creating severe political conflicts that can also erode democratic institutions. Inclusive representation may facilitate acceptance of the constitution by a larger number of political actors. At the same time, however, representative pluralism may create obstacles to agreement. To be sure, one may think of a theoretical optimum in which one outcome can be achieved without jeopardizing the others. In practice, however, it is not clear where this optimum lies.

With these cautionary notes, I will now discuss how constitution-making has been causally linked to three relevant outcomes: constitutional choice, constitutional endurance, and democracy. I will pay particular attention to the underlying logic of the association and whether the hypothesized causal link is empirically supported.

Constitutional Choice

Perhaps the most intuitive association is between constitution-making and constitutional choice. If the process through which a constitution emerges matters at all, it should matter in explaining its particular design. As we will see, however, the logic behind this intuition is not always solid, and the evidence to support it is in some cases weak.

In several works over the years, Jon Elster (1995, 2006, 2012, 2013) has argued that just as an executive constitution-making body would write an important role for itself in the constitution, so would a constituent legislature give a preponderant importance to the legislative branch at the expense of the executive and the judiciary. Elster portrays this outcome as undesirable because in his view, personal, group, or institutional self-interest distorts the impartial long-term perspective that constitution-makers should have in mind when designing a constitution. Based on this perspective, one could also argue that an unbalanced distribution of powers may turn the constitution into an ineffective governance structure. The rationale behind these arguments is, however, debatable.

The idea that a legislature would be tempted to benefit itself at the expense of other institutions is based on a dubious assumption about the influence of the type of constitution-making body on reformers’ choices. In most democratic contexts, popular representation in both constituent conventions and constituent legislatures is channeled through political parties. This means that the institutional preferences of constitution-makers are more likely to be shaped by the concrete interests of their parties than by the abstract interests of the collective body in which they gather as representatives of the people. As politicians, reformers tend to defend the institutional interests of their parties because doing so benefits them individually by helping them to win office and have influence over important decisions. For this reason, constitution-makers who in a particular institutional setting have a partisan link with the legislature or the executive are prone to making constitutional choices that favor these branches, regardless of whether the constituent body is a legislature, an executive commission, or a convention. For instance, in a presidential system, members of a party that expects to win the presidency, but not a majority in the legislature, are likely to favor an executive with strong legislative powers (Negretto, 2013).

Because the interests of existing institutions are usually represented through the parties that control or expect to control them, one way to induce impartiality in constitutional choice would be to forbid parties to participate in the constituent body. This would be the case of conventions made up of randomly selected citizens or delegates elected on a nonpartisan basis. This solution would imply, however, that removing the influence of group and institutional interests does not depend on the type of body per se, but on its composition. Another possibility is to allow political parties to field candidates for election as delegates, but postpone the implementation of the new constitution until some time after it was adopted. This would increase the level of uncertainty of constitution-makers as to what institutions would benefit them most, thus inducing a more impartial constitutional design (Elster, 1995, 1997).34 However, if this impartial perspective develops effectively, it would derive from the delay in implementing the constitution, not from the nature of the constituent body.

Whether constituent legislatures engage in self-dealing has been tested empirically by Ginsburg, Elkins, and Blount (2009), based on a sample of 411 episodes of constitution-making around the world from 1789 to 2005. Correlating the use of pure- and mixed-constituent assemblies with an index of parliamentary powers, they conclude that there is no evidence to sustain the claim that constituent legislatures are more likely to strengthen the powers of the legislature. These results may be questionable, however. The sample used by Ginsburg, Elkins, and Blount includes both democratic and authoritarian episodes of constitution-making, so the units of observation are too heterogeneous to permit a reliable comparison (see Elster, 2013). The results do not change, however, in tests restricted to constitutions made under competitive conditions. Based on an analysis of 31 episodes of competitive constitution-making in Latin America from 1900 to 2014, Negretto (2017b) shows that choosing a special convention or a constituent legislature has no significant impact on the allocation of powers between the different branches of government.

Inclusive representation, measured by the presence of a multiparty reform coalition, has been associated with certain types of constitutional choices. For instance, according to Pozas-Loyo and Rios-Figueroa (2010), whereas multiparty constitution-making bodies are likely to adopt power-sharing institutions, those under the control of a single party tend to opt for power-concentrating designs. In particular, they argue that plural constitution-making bodies are likely to select institutions that strengthen the power and independence of the judiciary. The rationale of this hypothesis is that in an inclusive constituent body, weaker actors would have veto power to prevent the preferences of stronger actors from being adopted, so that only power-sharing arrangements would be acceptable. Carey (2009, p. 160) has also proposed that inclusive constitution-making should lead to more constraints on government authority. By inclusion, however, he means not the internal diversity of the constitution-making body, but the number of individual or collective actors with a formal role in the drafting and approval of the constitution, including the participation of citizens at the ratification stage.

The main problem with these hypotheses is that they do not take into account that in the presence of multiple actors (whether within the constitution-making body or in the process as a whole), the only way to arrive at decisions is usually by reaching a compromise between opposite institutional preferences. Inclusive constituent bodies that have both government and opposition parties as partners in the reform coalition are more likely to combine power-concentrating and power-sharing institutions than to adopt a coherent power-sharing design. Based on this logic, Negretto (2013) has proposed that in the context of presidential regimes, multiparty constitution-making bodies are likely to opt for hybrid designs that reduce the electoral and government powers of presidents while strengthening their powers in the lawmaking process.

The available evidence about the impact of representative pluralism on various outcomes of interest is mixed. In a bivariate analysis of 48 episodes of constitution-making around the world between 1990 and 2005, Carey (2009) finds preliminary support for the proposal that greater inclusiveness in constitution-making leads to more constraints on the executive. As already mentioned, it is not clear how this finding should be taken, given that the analysis mixes channels of representation with mechanisms of popular participation. In any case, however, this result is inconsistent with statistical analyses performed using Latin American cases, which show that multiparty constituent bodies tend to opt for designs mixing electoral rules that weaken the partisan powers of presidents, with an allocation of legislative powers that increase their influence in policy-making (Negretto, 2013, 2017b).

The various forms of direct involvement of citizens in constitution-making have also been linked to constitutional choice. Samuels (2006) has argued that participatory processes tend to expand the rights of previously excluded groups. Ginsburg, Blount, and Elkins (2008) proposed that participatory processes are likely to increase the number and extent of constitutional rights. Several other authors (e.g., Ginsburg, Blount, & Elkins, 2008; Ginsburg, Elkins, & Blount, 2009; Voigt, 2004) have argued that the use of popular referendums to ratify a constitution is likely to produce a text incorporating participatory institutions in ongoing governance. These hypotheses make sense, particularly in relation to the use of ratification referendums. If reformers know or anticipate the preferences of those who have the power to accept or reject their proposals, they have every incentive to satisfy those preferences beforehand. Both the expansion of citizen rights and the inclusion of participatory institutions in the new constitution can be a safe strategy to secure the approval of proposals.

Popular participation in constitution-making can also have an impact on other aspects of the constitution, such as the distribution of powers between presidents and assemblies. As already argued, Carey (2009) proposes that participatory constitution-making, including the use of referendums, is likely to produce a design where more constraints on the executive are created. However, the mechanism underlying this hypothesis is not apparent. For instance, in a context of economic or political crisis, where people tend to support strong executive authority, there is no reason to suppose that citizen involvement in constitution-making would lead to more executive constraints. Looking at Latin American countries that have experienced such crises, Gargarella (2013) argues that recent radical participatory processes in the region have increased the powers of the executive, particularly his or her capacity to run for consecutive reelection.

The evidence of different tests on the impact of participatory processes on constitutional choice seems to confirm the relationship between levels of participation during constitution-making and the number of participatory institutions incorporated into the constitution (Ginsburg, Blount, & Elkins, 2008; Ginsburg, Elkins, & Blount, 2009; Negretto, 2017b). With regard to the expansion of rights, evidence is more mixed. In a limited study of 12 constitution-writing processes, Samuels (2006) claims support for the expansion of rights in participatory processes. However, in a larger study on the impact of process features in 194 cases of constitution-making between 1975 and 2002, Widner (2008) finds no correlation between levels of public consultation and rights protection. The problem here, however, is not only the difference in the number of cases, but also in what each author codifies as a measure of popular participation.

The impact of popular participation on checks and balances is also mixed. Carey (2009) finds no clear association between the use of referendums in themselves and greater or lesser constraints on the executive. Looking at Latin American cases, Negretto (2017b) shows that higher levels of popular participation in constitution-making have some impact on strengthening the legislative powers of presidents. These results are not strictly comparable, however, because they are based on different measures of citizen involvement in constitutional change and executive powers.

Constitutional Endurance

Constitutional endurance is an old intellectual concern in political philosophy, constitutional law, and comparative legal studies. Yet it has only recently become an object of systematic empirical analysis. Most of the hypothesized determinants of constitutional durability belong to the particular design of the constitution or to features of the political environment from which the constitution emerges. A few, however, are related to the constitution-making process.

Some normative arguments about the type of constitution-making body can be read as implicit proposals about the durability of constitutions. For instance, Jon Elster (2006, p. 185) has suggested that “constitutions produced by conventions tend to have greater legitimacy and hence tend to enjoy greater stability.” The logic of this argument is that since a constituent convention is elected for the specific task of writing a constitution, it makes it possible to organize elections with a view toward representing a variety of interests and programs rather than maximizing the representation of a particular group or ensuring stable governance, as might occur with constituent legislatures. For similar reasons, a special convention would provide voters with the opportunity to elect delegates based on their reform proposals and not on other considerations (Elster, 2006, pp. 186–189). From this perspective, if constituent conventions strengthen the legitimacy of the constitution, they also should generate wider public support for the constitution in the future.

This argument assumes that conventions secure a degree of representative pluralism that is absent from constituent legislatures. Yet there is no reason why an electorally mandated (or constitutionally authorized) constituent legislature could not be widely representative. At least in democratic contexts, securing diversity in representation is an important concern in the election of most constituent bodies in the contemporary world. In addition, voters may elect members of a constituent legislature on the basis of their positions on constitutional reform. This is likely to be the case when making a constitution is the first task of the legislature after it has been elected during a transition to democracy or when, in a democratic regime, the constitution authorizes the legislature to enact a total reform after a new election of its members. Finally, even if it were true that one type of constituent body leads to stronger support for the constitution at the foundational moment, it is not apparent how this support is maintained over time.

There is no cross-regional test available on the durability of constitutions made by different constituent bodies. However, the finding of recent empirical works that more detailed constitutions survive longer (e.g., Hammons, 1999; Elkins et al., 2009) suggests as a plausible hypothesis that legislatures might be better constituent bodies than special conventions. Being concerned with reelection, legislators are likely to pay more attention to the interests of the particular groups that support them than would delegates of a special convention. If this is the case, then legislators may be more inclined to write longer and more detailed provisions into the constitution to show their constituents that their interests have been protected. These constituents, in turn, would have more incentive to support the constitution in the long run. In a more direct analysis of the effect of pure constituent assemblies vis-à-vis constituent legislatures in Latin America, Negretto (2017b) finds that there is no significant difference in the impact of either body on the durability of constitutions.

Somewhat related to the previous argument is the proposal that a higher level of inclusion during the constitution-making process leads to more durable and stable constitutional texts (Voigt, 2004; Carey, 2009). One problem with this hypothesis, again, is the confusion between the degree of representative pluralism and the level of direct citizen involvement. In this view, constitutions are more likely to endure as representation becomes more diverse, a larger number of citizens become involved in the process, or both. While it is plausible that both representative diversity and popular participation may lead to more durable constitutions, it is not clear what the outcome would be if representation is restricted and popular participation is high, or vice versa. Would a high level of citizen involvement offset the potentially negative effects of a constitution adopted unilaterally? Or is an elite-driven process preferable so long as all the relevant organized interests are included in constitutional deliberations and negotiations?

Other authors have focused more explicitly on the relationship between direct popular participation and constitutional durability. Although there are several formulations of this hypothesis, they are all based on the idea that involving citizens in constitution-making would make them perceive the constitution as their collective work, and they would thus be more inclined to defend it in the future (see Widner, 2008). Using this logic, Elkins et al. (2009) have argued that breadth of participation (conceptualized as “inclusion”) in the formulation and ongoing enforcement of the constitution is likely to make constitutions more durable because these measures would invite participants to invest in their maintenance. This hypothesis is problematic because it mixes mechanisms of representation, such as the election of the constituent body, with mechanisms of popular participation, such as ratification of the constitution, which need not be a manifestation of a single procedural variable on either theoretical or empirical grounds.35 At the same time, it combines participation during the making of the constitution with mechanisms of popular participation in the constitution, which may be endogenously related.

The evidence in support of these hypotheses is not conclusive. Carey (2009) finds a correlation between the number of institutional actors participating in the constitution-making process and the subsequent stability of the constitution, but no clear association between the use of referendums alone and this outcome. Elkins et al. (2009) show that the election of the constituent body, along with its ratification and the inclusion of mechanisms of participation in the constitution, increase the durability of the constitution. One problem in interpreting this finding is that it seems to hold for both authoritarian and democratic constitutions (Elkins et al., 2009, p. 227). Yet the election of the constituent body and the participation of citizens in referendums should not have the same effects across democratic and authoritarian conditions.

Democracy

In the contemporary world, most constitutions adopted at the founding of a new state or during a transition to democracy seek to inaugurate a new era of political liberty and government accountable to the people. They do not always achieve this goal, however. Whereas in some cases, constitutions generate levels of democratization greater than those of the preceding period, in others, they produce no change or are followed by a relative deterioration of the preexisting political conditions. Legal continuity, inclusiveness, and popular participation in constitution-making have been associated with these outcomes.

Perhaps one of the most repeated lessons derived from the study of transitions to democracy has been that moderation and accommodation are needed to prevent authoritarian regression (O’Donnell & Schmitter, 1986). From this perspective, even though a radical break with authoritarian institutions might create a stronger sense of democratic legitimacy for the new order, it could also endanger its stability. Based on an analysis of historical and contemporary cases of constitution-making, Arato (1995) also has suggested that preserving legal continuity in a transition to democracy is important both to provide security to the different actors involved and to signal to the population that constitution-makers, as well as future rulers, are subject to the law. These arguments imply that using existing revision procedures to create a new constitution or reform the previous one could enhance the survival of both the constitution and democracy.

In the case of intrademocratic constitutional change, radical attempts at democratic refoundation through constitutional rewrites have been linked to the breakdown or erosion of democracy (Levitsky & Loxton, 2013). In a similar vein, but looking at the procedural features of these events, some authors (e.g., Brewer-Carias, 2011; Partlett, 2012; Landau, 2013; Negretto, 2017a) have argued that populist presidents may be tempted to use the extralegal creation of a new constitution as part of their plebiscitarian tactics to produce a constitutional crisis and concentrate power in their own hands. This research would suggest that using existing institutions and procedures to replace constitutions within a democratic regime might favor the future preservation of representative institutions.

One can also find, however, radically opposite hypotheses. Viciano Pastor and Martinez Dalmau (2010) derive from the doctrine of the constituent power of the people that only a clean legal break with the past can create a constitution that can promote an effective democratic transformation. In a similar vein, but based on the successful American experience with constitution-making, Bruce Ackerman (1994) has argued that to have a firm foundation, new democracies should avoid using the existing amendment procedures to create a new legality. Although these arguments are mainly normative, they clearly imply that breaking with the former legal order is likely to deepen democracy.

To determine which of these hypotheses has more support in reality, we would need a large systematic comparative analysis on the link between legal continuity and levels of democracy before and after the enactment of a new constitution. There are no studies of this kind using a cross-regional database.36

Carey (2009) has stated that more inclusive constitution-making processes should strengthen democracy after the enactment of the new constitution. Eisenstadt et al. (2015), in turn, argue that popular participation, particularly at the drafting stage, is likely to improve levels of democracy. The logic of these arguments is related to the stronger legitimacy associated with constitutions adopted with the inclusion of a larger number of institutional actors and with the involvement of citizens in the process. These features would enhance the sense of collective ownership over the constitution among both political elites and citizens, thus leading to an overall improvement of levels of democratization after the enactment of a new constitution.

Within his limited sample, Carey (2009) finds that as more institutional actors participate in the process, subsequent democracy improves, although democratization levels are only marginally greater than with fewer actors. At the same time, he finds that the presence of elected constituent assemblies is associated with improvements in democracy. Using a larger database and a more sophisticated statistical technique, Eisenstadt et al. (2015) show empirical support for the hypothesis that popular participation in general and participation during the drafting stage in particular leads to higher levels of democratization. Several factors affect the interpretation of these findings.

In the first place, constitution-making is often a byproduct of democratization rather than the other way around. From this perspective, constitution-making after the collapse of a brutal dictatorship would obviously be linked to democratic improvements in the country, at least in the short term. But this improvement is unrelated to the procedural features of constitution writing.37 A comparison between democracy levels before and after the promulgation of a constitution drafted within an established democratic regime would lead to more credible results. There are, however, too few observations for these cases.

Another problem is that of distinguishing between genuine and sham citizen participation. Dictators and democrats alike have used similar public participation procedures, which suggest that their effects derive from the intentions and bargaining power of designers, not from the formal aspects of the process.38 A second problem, already mentioned, is how citizen participation relates to representation. What should be the net expected effect of constitution-making on democracy when the process is participatory at the citizen level, but exclusionary at the level of elected political representatives? Finally, even if a short-term improvement in democracy levels is verified, it is not clear how this effect persists over time. Popular support for a constitution may erode for reasons that are independent of the initial constitutional moment.

Domestic and Foreign Actors

Constitution-making has always had an international dimension. Both the process by which a constitution is made and the options of design that reformers consider at a particular historical juncture are subject to a process of diffusion or imitation. It is not by chance that Latin America is the region where special conventions have been used most often as constituent bodies, and where almost all countries opted for separation-of-powers systems in the mid-19th century. Although the adoption of these processes and designs always contained idiosyncratic adaptations, there is no doubt of the influence that the 1787 Philadelphia convention and the U.S. constitution had in Hispanic America after independence (Negretto, 2003; Gargarella, 2010). Similarly, in Anglophone Africa after independence, many countries followed the British tradition of using parliament as a constitution-making body and drafted constitutions modeled on a Westminster-style parliamentary system (Kirkby & Murray, 2016).

A different form of internationalization of constitution-making is the participation of foreign governments and international organizations during the process. This type of international involvement has become more frequent since the mid-20th century, starting at the end of World War II and the beginning of the period of decolonization. The United States had a direct influence on the replacement of the Meiji constitution in Japan and on the drafting of the 1947 text, and the Allied forces participated in the organization of the constitution-making process and design of the 1949 German constitution (Arato, 2009). Representatives of Greece, Turkey, the Greek Cypriot community, and the Turkish Cypriot community signed the 1960 constitution of Cyprus. During the decolonization process in Africa from 1945 to 1960, colonial powers often participated in the design and ratification of the constitutions of newly independent countries.

In more recent years, international agencies have provided assistance and the international community has been involved in the enforcement of peace agreements that preceded or were part of constitution-making in postconflict settings. On December 14, 1995, the constitution of Bosnia-Herzegovina was adopted as part of international peace negotiations in which the United States and various European countries were involved (Al-Ali, 2011). The 2005 constitution of Burundi was adopted following the guidelines of the 2000 Arusha Peace and Reconciliation Agreement, which was signed by warring factions in the country thanks to the mediation of South Africa and support from the United Nations (UN). In 2000, the UN mission in Kosovo facilitated the adoption of a provisional constitutional structure by sponsoring negotiations between Serb leaders and Kosovar Albanians (Miller, 2010).

There are today several international agencies, such as the Institute for Democratic and Electoral Assistance, the U.S. Institute of Peace, Interpeace, and the United Nations, which have programs whose main objective is supporting constitution-building processes around the world. These programs provide advice and support to governmental, international, and nongovernmental organizations on procedural and substantive issues related to constitution-making, primarily but not exclusively in postconflict settings.

The involvement of international actors in the actual drafting and implementation of new constitutions has varied from case to case. Nonetheless, the mere existence of this phenomenon raises several important normative and empirical questions to be addressed in the comparative study of constitution-making. One obvious question is the influence of the particular mode of intervention on the capacity of local actors to negotiate an acceptable constitutional agreement. Another is the impact of external actors on the diffusion of particular models or practices of constitution-making.

A Multidimensional Phenomenon

Constitution-making is a prolific political activity. During 2015 alone, 19 countries in the world were involved in processes of significant constitutional change. Wholesale replacements are more common in transitional and postconflict settings than in established democracies. Yet the latter also may experience a debate or initiate a process oriented to replacing the constitution in force, particularly when citizens have a low level of trust in current institutions and the performance of the political regime is suboptimal.

Today, constitution-making is central to several important research agendas in political theory, comparative politics, political history, and comparative constitutional law. Constitution-making is analyzed from a normative perspective to propose the best conditions for achieving democratic legitimacy, popular participation, or a rational design; it is studied empirically to test the causal effect of procedural rules on important outcomes such as constitutional choice, constitutional durability, or democracy; it is incorporated as an important aspect of the study of critical historical junctures; and it is examined from a legal point of view.

This variety of perspectives is necessary because constitution-making is a multidimensional phenomenon. In the absence of internal dialogue, however, a plurality of views may prevent knowledge sharing. There is significant collaboration today between legal scholars and social scientists for the comparative study of constitutions. Yet there is still much more to be done to transform the analysis of constitutionalism and constitution-making into a truly interdisciplinary research enterprise.

References

Ackerman, B. (1994). The future of the liberal revolution. New Haven, CT: Yale University Press.Find this resource:

Ackerman, B., & Katyal N. (1995). Our unconventional founding. University of Chicago Law Review, 62(2), 478–573.Find this resource:

Al-Ali, Z. (2011). Constitutional drafting and external influence. In T. Ginsburg & R. Dixon (Eds.), Comparative constitutional law (pp. 77–95). Cheltenham, U.K.: Edward Elgar.Find this resource:

Althusius, J. (1995). Politica Methodice Digesta. Indianapolis, IN: Liberty Fund.Find this resource:

Amar, A. (1988). Philadelphia revisited: Amending the Constitution outside Article V. University of Chicago Law Review, 55(4), 1043–1104.Find this resource:

Arato, A. (1995). Forms of constitution-making and theories of democracy. Cardozo Law Review, 17(2), 191–232.Find this resource:

Arato, A. (2009). Reediming the still reedemable: Post sovereign constitution making. International Journal of Politics, Culture, and Society, 22(4), 427–443.Find this resource:

Arato, A. (2011). Arato on Constitution Making in Hungary and the 4/5 Rule. Blog of the International Journal of Constitutional Law and ConstitutionMaking.org. Retrieved from http://www.iconnectblog.com/2011/04/arato-on-constitution-making-in-hungary-and-the-45-rule/.Find this resource:

Arato, A. (2016). Post sovereign constitution making: Learning and legitimacy. New York: Oxford University Press.Find this resource:

Arendt, H. (1963). On revolution. New York: Penguin.Find this resource:

Bejarano, A. M., & Segura, R. (2013). Asambleas constituyentes y democracia: Una lectura crítica del nuevo constitucionalismo en la Región Andina. Colombia Internacional, 79, 19–48.Find this resource:

Biaggini, G. (2011). Switzerland. In C. Fusaro & D. Oliver (Eds.), How constitutions change. Portland, OR: Hart Publishing.Find this resource:

Blount, J. (2011). Participation in constitutional design. In R. Dixon & T. Ginsburg (Eds.), Comparative constitutional law (pp. 38–56). Cheltenham, U.K.: Edward Elgar.Find this resource:

Böhrt Irahola, C. (2013). El proceso constituyente Boliviano. In C. Böhrt Irahola & N. Wray Reyes (Eds.), Los procesos constituyentes Boliviano y Ecuatoriano: Analisis comparativo y prospectiva. Stockholm, Sweden: IDEA International.Find this resource:

Bodin, J. (1992). On sovereignty: Four chapters from six books of the commonwealth. In J. Franklin (Ed.), J. Bodin-On sovereignty. commonwealth. New York: Cambridge University Press.Find this resource:

Brandt, M., Cotrell, J., Ghai, Y., & Regan, A. (2011). Constitution-making and reform: Option for the process. Geneva, Switzerland: Interpeace.Find this resource:

Brewer Carías, A. (2002). Golpe de estado y proceso constituyente en Venezuela. Mexico City, Mexico, UNAM.Find this resource:

Brewer Carías, A. (2011). Dismantling democracy in Venezuela: The Chávez authoritarian experiment. New York: Cambridge University Press.Find this resource:

Carey, J. (2009). Does it matter how a constitution is created? In Z. Barany & R. G. Moser (Eds.), Is democracy exportable (pp. 155–177). New York: Cambridge University Press.Find this resource:

Carré de Malberg, R. (2012). Teoría general del estado. Mexico City, México: Fondo de Cultura Económica.Find this resource:

Colomer, J. M. (2004). Handbook of electoral system choice. New York: Palgrave.Find this resource:

Chirwa, D. M. (2003). Democratisation in Malawi 1994–2002: Completing the vicious circle? South African Journal on Human Rights, 19, 316–338.Find this resource:

Colon Rios, J. (2011). Carl Schmitt and constituent power in Latin America: The cases of Venezuela and Colombia. Constellations, 18(3), 365–388.Find this resource:

Colon Rios, J. (2012). Weak constitutionalism. Democratic legitimacy and the question of constituent power. New York: Routledge.Find this resource:

Combellas, R. (2003). El proceso constituyente y la Constitución de 1999. Politeia, 26(30), 183–208.Find this resource:

Congleton, R. D. (2012). Improving democracy through constitutional reform. Some Swedish lessons. Boston: Kluwer Academic Publishers.Find this resource:

Contiades, X., & Tassopoulos, I. (2013). Constitutional change in Greece. In X. Contiades (Ed.), Engineering constitutional change: A comparative perspective on Europe, Canada, and the USA (pp. 151–178). Oxon, U.K.: Routledge.Find this resource:

Dinan, J. (2009). The American state constitutional tradition. Lawrence, KS: University Press of Kansas.Find this resource:

Dugas, J. (1993). La Constitución de 1991. ¿Un pacto político viable? Bogotá, Colombia: Departamento de Ciencia Política, Universidad de los Andes.Find this resource:

Ebrahim, H., & Miller, L. E. (2010). Creating the birth certificate of a new South Africa: Constitution making after apartheid. In L. E. Miller (Ed.), Framing the state in times of transition: Case studies in constitution making (pp. 111–157). Washington, DC: U.S. Institute of Peace Press.Find this resource:

Eisenstadt, T., LeVan, C., & Maboudi, T. (2015). When talk trumps text: The democratizing effects of deliberation during constitution-making, 1974–2011. American Political Science Review, 103(3), 592–612.Find this resource:

Elkins, Z., Ginsburg, T., & Melton, J. (2009). The endurance of national constitutions. New York: Cambridge University Press.Find this resource:

Elster, J. (1991). Born to be immortal: The constitution-making process. Ann Arbor: University of Michigan Press.Find this resource:

Elster, J. (1994). Constitutional bootstrapping in Philadelphia and Paris. In Michel Rosenfeld (Ed.), Constitutionalism, identity, and legitimacy: Theoretical perspectives (pp. 57–83). Durham, NC: Duke University Press.Find this resource:

Elster, J. (1995). Forces and mechanisms in constitution-making. Duke Law Review, 45, 364–396.Find this resource:

Elster, J. (1997). Ways of constitution-making. In A. Hadenius (Ed.), Democracy’s victory and crisis (pp. 123–142). Cambridge, U.K.: Cambridge University Press.Find this resource:

Elster, J. (2006). Legislatures as constituent assemblies. In R. W. Bauman & T. Kahana (Eds.), The least examined branch: The role of legislatures in the constitutional state (pp. 181–197). New York: Cambridge University Press.Find this resource:

Elster, J. (2012). Clearing and strengthening the channels of constitution-making. In T. Ginsburg (Ed.), Comparative constitutional design (pp. 15–30). New York: Cambridge University Press.Find this resource:

Elster, J. (2013). Securities against misrule: Juries, assemblies, elections. New York: Cambridge University Press.Find this resource:

Fishkin, J. (2011). Deliberative democracy and constitutions. In E. Frankel, F. Miller, & J. Paul (Eds.). What should constitutions do? (pp. 242–260). New York: Cambridge University Press.Find this resource:

Franklin, J. (1992). Introduction. In J. Franklin (Ed.), J. Bodin-On sovereignty: Four chapters from six books of the commonwealth (pp. ix–xxvii). New York: Cambridge University Press.Find this resource:

Fournier, P., Van Der Kolk, H., Kenneth Carty, R., Blais, A., & Rose, J. (2011). When citizens decide. New York: Oxford University Press.Find this resource:

Foyer, J. (1988). The drafting of the French Constitution of 1958. In R. A. Goldwin & A. Kaufman (Eds.), Constitution makers on constitution making (pp. 7–46). Washington, DC: American Enterprise Institute for Public Research.Find this resource:

Fritz, C. (1997). Alternative visions of American constitutionalism: Popular sovereignty and the early American debate. Hatings Constitutional Law Quarterly, 24(2), 287–357.Find this resource:

Gargarella, R. (2010). The legal foundations of inequality. Constitutionalism in the Americas, 1776–1860. New York: Cambridge University Press.Find this resource:

Gargarella, R. (2013). Latin American constitutionalism, 1810–2010: The engine room of the constitution. New York: Oxford University Press.Find this resource:

Geddes, B. (1990). Democratic institutions as a bargain among self-interested politicians. Paper presented at the 1990 meeting of the American Political Science Association, San Francisco, August 30–September 2.Find this resource:

Ginsburg, T., Blount, J., & Elkins, Z. (2008). The citizen as founder: Public participation in constitutional approval. Temple Law Review, 81(2), 361–382.Find this resource:

Ginsburg, T., Elkins, Z., & Blount, J. (2009). Does the process of constitution-making matter? American Review of Law and Society, 5, 201–223.Find this resource:

Ginsburg, T., & Simpser, A. (Eds). (2014). Authoritarian constitution-making: The role of the military in Latin America. New York: Cambridge University Press.Find this resource:

Hammons, C. W. (1999). Was James Madison wrong? Rethinking the American preference for short, framework-oriented constitutions. American Political Science Review, 93(4), 837–849.Find this resource:

Hoar, R. S. (1917). Constituent conventions: Their nature, powers, and limitations. Boston: Little, Brown, and Co.Find this resource:

Holmes, S., & Sunstein, C. (1995). The politics of constitutional revision in Eastern Europe. In S. Levinson (Ed.), Responding to imperfection: The theory and practice of constitutional amendment (pp. 275–306). Princeton, NJ: Princeton University Press.Find this resource:

Jackson, V. C. (2008). What’s in a name? Reflections on timing, naming, and constitution-making. William & Mary Law Review, 49(4), 1249–1305.Find this resource:

Jacobsohn, G. F. (2006). An unconstitutional constitution? A comparative perspective. International Journal of Constitutional Law, 4(3), 460–487.Find this resource:

Kalyvas, A. (2005). Popular sovereignty, democracy, and the constituent power. Constellations, 12(2), 223–244.Find this resource:

Kay, R. (1987). The illegality of the constitution. Constitutional Commentary, 4, 57–80.Find this resource:

Kay, R. (2011). Constituent authority. American Journal of Comparative Law, 59, 715–762.Find this resource:

Kiris, A. (1991). Restoration of the independence of the republic of Estonia: Selection of legal acts, Ministry of Foreign Affairs of the Republic of Estonia. Tallinn, Estonia: Estonian Institute for Information.Find this resource:

Klein, C., & Sajó, A. (2012). Constitution-making: Process and substance. In M. Rosenfeld & A. Sajó (Eds.), Oxford handbook of comparative constitutional law (pp. 419–441). London: Oxford University Press.Find this resource:

Krunke, H. (2013). Formal and informal methods of constitutional change in Denmark. In X. Contiades (Ed.), Engineering constitutional change: A comparative perspective on Europe, Canada and the USA (pp. 73–92). Oxon, U.K.: Routledge.Find this resource:

Landau, D. (2013). Abusive constitutionalism. UC Davis Law Review, 47, 1–58.Find this resource:

Lawson, G. (2004) Politica sacra et civilis. New York: Cambridge University Press.Find this resource:

Lehoucq, F. (2008). Bolivia’s constitutional breakdown. Journal of Democracy, 19(2), 46–60.Find this resource:

Lenowitz, J. (2015). A trust that cannot be delegated: The invention of ratification referenda. American Political Science Review, 109(4), 803–816.Find this resource:

Lerner, H. (2013). Making constitutions in deeply divided societies. New York: Cambridge University Press.Find this resource:

Levinson, S. (1995). How many times has the United States Constitution been amended? In S. Levinson (Ed.), Responding to imperfection: The theory and practice of constitutional amendment (pp. 13–36). Princeton, NJ: Princeton University Press.Find this resource:

Levitsky, S., & Loxton, J. (2013). Populism and competitive authoritarianism in the Andes. Democratization, 20(1), 107–136.Find this resource:

Locke, J. (1988). Two treatises of government. New York: Cambridge University Press.Find this resource:

Loughlin, M. (2003). The idea of public law. New York: Cambridge University Press.Find this resource:

Loughlin, M. (2007). Constituent power subverted: From English constitutional argument to British constitutional practice. In M. Loughlin & N. Walker (Eds.), The paradox of constitutionalism: Constituent powers and constitutional form (pp. 27–48). New York: Oxford University Press.Find this resource:

Madison, J., Jay, J., & Hamilton, A. (1987). The federalist papers. London: Penguin. Originally published in 1788.Find this resource:

Martínez-Lara, J. (1992). Building democracy in Brazil: The politics of constitutional change, 1985–95. Basingstoke, U.K.: Macmillan Press.Find this resource:

Mendez, F., & Wheatley, J. (2013). Patterns of constitution making over time and space. In J. Wheatley & F. Mendez (Eds.), Pattern of constitutional design: The role of elites and citizens in constitution making (pp. 21–47). Farnham, U.K.: Ashgate.Find this resource:

Meuwese, A. (2013). Popular constitution making. The case of Iceland. In D. Galligan & M. Versteeg (Eds.), The social and political foundations of constitutions (pp. 469–496). New York: Cambridge University Press.Find this resource:

Miller, L. E. (2010). Designing constitution-making processes: Lessons from the past, questions for the future. In L. E. Miller (Ed.), Framing the state in times of transition: Case studies in constitution making. Washington, DC: U.S. Institute of Peace.Find this resource:

Mirth, K. (1992). Croatia’s independence. Journal of Croatian Studies, 32–33, 5–9.Find this resource:

Murphy, W. (2007). Constitutional democracy: Creating and maintaining a just political order. Baltimore: John Hopkins University Press.Find this resource:

Kirkby, C., & Murray, C. (2016) Constitution-making in Anglophone Africa: We the people? In M. Ndulo & M. Gazibo (Eds.), Growing democracy in Africa: Elections, accountable governance, and political economy (pp. 86–113). Cambridge Scholars Publishing.Find this resource:

Negretto, G. (2003). Diseño constitucional y separación de poderes en América Latina. Revista Mexicana de Sociología, 1(1), 41–76.Find this resource:

Negretto, G. (2012). Replacing and amending constitutions: The logic of constitutional change in Latin America. Law and Society Review, 46(4), 749–779.Find this resource:

Negretto, G. (2013). Making constitutions: Presidents, parties, and institutional choice in Latin America. New York: Cambridge University Press.Find this resource:

Negretto, G. (2015). Procesos constituyentes y refundación democrática: El caso de Chile en perspectiva comparada. Revista de Ciencia Política, 31(1), 201–215.Find this resource:

Negretto, G. (2017a). Constitution making in democratic constitutional orders. The challenge of citizen participation. In S. Ruth, Y. Welp, & L. Whitehead (Eds.), Let the people rule? Direct democracy in the twenty-first century (pp. 21–39). Essex, U.K.: ECPR Press.Find this resource:

Negretto, G. (2017b). Constitution making and constitutionalism in Latin America: The role of procedural rules. In T. Ginsburg & R. Dixon (Eds.), Comparative constitutional law in Latin America (pp. 17–56). Northampton, MA: Elgar.Find this resource:

Neuman, L., & McCoy, J. (2001). Observing political change in Venezuela: The Bolivarian constitution and 2000 elections. Atlanta, GA: The Carter Center.Find this resource:

Nohlen, D., & Stöver, P., Eds. (2010). Elections in Europe. Baden-Baden, Germany: Nomos.Find this resource:

O’Donnell, G., & Schmitter, P. (1986). Transitions from authoritarian rule: Tentative conclusions about uncertain democracies. Baltimore: John Hopkins University Press.Find this resource:

Partlett, W. (2012). The dangers of popular constitution-making. Brooklyn Journal of International Law, 8, 193.Find this resource:

Pozas-Loyo, A., & Ríos-Figueroa, J. (2010). Enacting constitutionalism. The origins of independent judicial institutions in Latin America. Comparative Politics, 42(3), 293–311.Find this resource:

Renwick, A. (2014). After the referendum: Options for a constituent convention. London: Constitution Society.Find this resource:

Ríos-Figueroa, J. (2011). Institutions for constitutional justice in Latin America. In G. Helmke & J. Rios-Figueroa (Eds.), Latin America (pp. 27–54). New York: Cambridge University Press.Find this resource:

Rosenn, K. S. (2010). Conflict resolution and constitutionalism: The making of the Brazilian constitution. In L. E. Miller (Ed.), Framing the state in times of transition (pp. 435–466). Washington, DC: U.S. Institute of Peace.Find this resource:

Samuels, K. (2006). Constitution-building processes and democratization: A discussion of twelve case studies. Stockholm: International IDEA.Find this resource:

Schmitt, C. (2008). Constitutional theory. Durham, NC: Duke University Press.Find this resource:

Sieyès, E. J. (2003). What is the third estate? In M. Sonenscher (Ed.), Political Writings: Including the debate between sieyès and Tom Paine in 1791 (pp. 93–162). Cambridge, U.K.: Hackett.Find this resource:

Sucharitkul, S. (1998). Thailand: Constitutional developments since Amendment No. 5 of 10 February 1995 to the constitution of 9 December 1991. Publications, Paper 565, 1–13.Find this resource:

Suksi, M. (2011). Finland. In C. Fusaro & D. Oliver (Eds.), How constitutions change: A comparative study. Portland, OR: Hart Publishing.Find this resource:

Thorarensen, B. (2011). Constitutional reform process in Iceland. Paper presented at the Oslo-Rome International Workshop on Democracy, Viale April 30, Rome. Oslo, Norway: University of Oslo. Retrieved from https://www.uio.no/english/research/interfaculty-research-areas/democracy/news-and-events/events/seminars/2011/papers-roma-2011/Rome-Thorarensen.pdf.Find this resource:

Tsatsos, C. D. (1988). Making the constitution of Greece. In R. A. Goldwin & A. Kaufman (Eds.), Constitution makers on constitution making (pp. 69–93). Washington, DC: American Enterprise Institute for Public Research.Find this resource:

Tushnet. M. (2014). Advanced introduction to comparative constitutional law. Cheltenham, U.K.: Edward Elgar.Find this resource:

Viciano Pastor, R., & Martínez Dalmau, R. (2001). Cambio político y proceso constituyente en Venezuela (1998–2000). Valencia, Spain: Tiranto Lo Blanch.Find this resource:

Viciano Pastor, R., & Martínez Dalmau, R. (2010). Los procesos constituyentes Latinoamericanos y el nuevo paradigma constitucional. Revista del Instituto de Ciencias Juridicas de Puebla, 25, 7–29.Find this resource:

Vile, J. (1993). Contemporary questions surrounding the constitutional amending process. Westport, CT: Praeger.Find this resource:

Voigt, S. (2004). The consequences of popular participation in constitutional choice: Towards a comparative analysis. In A. van Aaken, C. List, & C. Luetge (Eds.), Deliberation and decision: Economics, constitutional theory and deliberative democracy (pp. 199–229). Aldershot, U.K.: Ashgate.Find this resource:

Wallis, J. (2014). Constitution making during state building. New York: Cambridge University Press.Find this resource:

Wheatley, J., & Mendez, F. (2013). Patterns of constitutional design: The role of citizens and elites in constitution-making. London: Ashgate.Find this resource:

Widner, J. (2008). Constitution writing in post-conflict settings: Overview. William and Mary Law Review, 49(4), 1513–1541.Find this resource:

Widner, J., & Contiades, X. (2013). Constitution writing processes. In M. Tushnet, T. Fleiner, & C. Saunders (Eds.), Routledge handbook of constitutional law (pp. 57–70). New York: Routledge.Find this resource:

Williams, R. F. (1996). Are state constitutional conventions things of the past? The increasing role of the constitutional commission in state constitutional change. Hofstra Law & Policy Symposium, 1(1), 1–26.Find this resource:

Wray Reyes, N. (2013). El proceso constituyente Ecuatoriano. In C. Böhrt Irahola & N. Wray Reyes (Eds.), Los procesos constituyentes Boliviano y Ecuatoriano: Analisis comparativo y prospectiva. Stockholm: IDEA.Find this resource:

Zulueta-Fülscher, K. (2015). Interim constitutions. Peacekeeping and democracy-building tools. IDEA. Retrieved from http://www.constitutionnet.org/files/interim-constitutions-peacekeeping-and-democracy-building-tools-pdf.pdf.

Notes:

(1.) The constitution of Uruguay, for instance, allows Congress to pass a total revision of the constitution (with subsequent approval in a popular referendum), and this revision may count as a new constitution. On the other hand, there are constitutions such as the 1853 constitution of Argentina, which require the election of an independent constituent convention to amend even a single article.

(2.) From this perspective, it may be argued that a constitution should be considered new only when its drafters claim that it is new and when state institutions acknowledge the text as such. See Negretto (2012).

(3.) For the historical and conceptual evolution of the constituent power theory, see Loughlin (2003), Kalyvas (2005), and Colon-Rios (2012).

(4.) Although he did not distinguish between ordinary and fundamental laws, Bodin clearly included what would later be called the constituent power or right to change the constitution as a mark of sovereignty. See Franklin (1992, p. xvi).

(5.) See, for instance, the influence of Carl Schmitt´s concept of constituent power on constitutional court rulings in Latin America, as described in Colon-Rios (2011).

(6.) Compare Madison´s view on the power of delegates at the convention in Federalist 40 with Abbé Sieyès’s notion of “extraordinary” representatives in What Is the Third Estate? For this reason, whereas the proposal of the Federal Convention needed indirect ratification by the people represented in elected conventions to be enacted, the National Assembly did not submit the constitution to any form of popular ratification.

(7.) To have a sense of these conflicting perspectives, one can contrast the views of Carl Schmitt (2008) and Carré de Malberg (2012), or Amar (1988) and Vile (1993) on whether all aspects of the constituent power are inherently extralegal and on the institutional channels through which it has been exercised historically.

(8.) In fact, opposite examples can even come from the same country, as is the case with the different experiences of constitution-making in the United States at the federal and state level. See Fritz (1997).

(9.) Empirical data on these features have been collected by the author and belong to an ongoing research project that will be released in the future as the Comparative Constitution Making Database. This database covers all constitutions adopted in the world between 1900 and 2015 that have been implemented during competitive years for the largest portion of their lives. As of January 2017, it contains complete observations on 126 of 130 episodes that satisfy this definition. Specific observations can be obtained from the author upon request.

(10.) Data obtained from the Comparative Constitution Making Database.

(11.) Interim constitutions can fulfill a similar consensus-building goal, even in contexts where violent conflicts did not take place in the immediate past, as was the case with the Small Constitution enacted by the Polish Parliament in 1992 to prepare for the adoption of the final constitution. See Zulueta-Fülscher (2015).

(12.) The 1942 constitution was derived from an initiative made by two-fifths of the whole legislature, the 1952 constitution from agreement between the main parliamentary parties, and the 1967 constitution from a text proposed by the Colorado and Nacional parties and passed by the legislature.

(13.) These constitutions are from Denmark (1915), Denmark (1953), Finland (2000), France (1958), Greece (1952), Iceland (1944), Ireland (1937), Sweden (1974), Switzerland (1999), Hungary (2011), Poland (1997), Ukraine (1996), Kenya (2010), Thailand (1997), Nepal (2015), Sri Lanka (1972), Argentina (1994), Bolivia (2009), Ecuador (1998), Ecuador (2008), Trinidad and Tobago (1976), Colombia (1991), Uruguay (1942), Uruguay (1952), Uruguay (1967), and Venezuela (1999). Democratic years are coded based on Boix, Miller, and Rosatto’s database (2010).

(14.) It should be noted that on occasion, political and institutional actors might disagree about whether legal continuity was truly preserved. This was the case for the making of Hungary’s 2011 constitution. See Andrew Arato, at http://www.iconnectblog.com/2011/04/arato-on-constitution-making-in-hungary-and-the-45-rule/

(15.) In an analysis of 411 episodes of constitutional replacement that took place around the world between 1789 and 2005, Ginsburg, Elkins, and Blount (2009, p. 213) found that in the vast majority of cases, the approval body was either a constituent convention (103 observations) or a constituent legislature (178 observations). As reported in the text, I found a similar proportion of these bodies using a sample restricted to the creation of constitutions implemented during competitive years between 1900 and 2015.

(16.) Some authors use the term constitutional convention, common in the U.S. constitutional tradition, to denote assemblies whose specific task is to adopt a new constitution. See Elster (2006). However, since special conventions may or may not be authorized by the existing constitution, it would be odd to distinguish between constitutional and extraconstitutional (or unconstitutional) constitutional conventions. For this reason, throughout this text, I use the generic term constituent convention or simply convention to refer only to the nature of the task of this body.

(17.) Data from the Comparative Constitution Making Database, described in note 9.

(18.) In a sample of 160 constitution-making episodes around the world from 1780 to 2012, Mendez and Wheatley (2013, p. 29) find that 12 of the 23 events using constituent conventions were located in Latin America. I found a comparable share of Latin American cases among the 38 conventions reported in the text.

(19.) All 26 assemblies were created to enact new constitutions, were dissolved after completion of the task, and worked in parallel with an ordinary legislature or were succeeded by one. Some of them, however, remained in place until the new legislature was elected and in some cases had to pass ordinary laws. This legislative function was not always seen beforehand, and it lasted for a limited time that in total was shorter than the work of the assembly as a constituent convention.

(20.) Data from the Comparative Constitution Making Database, described in note 9.

(21.) Jon Elster (2006, p. 183) defines self-created constituent legislatures as bodies that “begin as legislatures and turn themselves into constituent assemblies.” He does not clarify on what basis a legislature can make this decision, except that it is not derived from an electoral mandate. To make the adjective self-created as precise as possible, I think the term should not apply to legislatures that turn into constituent bodies by virtue of a preexisting constitutional authorization. This means that self-created constituent legislatures should be restricted to legislative assemblies that assume a constituent task after a political decision made by the incumbent government or existing legislative parties without electoral or legal authorization.

(22.) The enactment of Slovenia’s 1991 constitution could be used as an example of the former, and the 1998 Albanian constitution of the latter.

(23.) Data collected by the author (see note 9). The use of appointment and mixed methods of selection in sub-Saharan Africa is due both to the lack of preexisting election practices and to the history of civil strife that prevailed in many countries of this region. Against this background, elections can be polarizing and have a negative impact on the need for negotiation and consensus building in these societies.

(24.) I have found that 59 out of a sample of 92 popularly elected constitution-making bodies between 1900 and 2015 used a PR electoral formula.

(25.) See Latin American Constitutional Change Database, at http://la-constitutionalchange.cide.edu/.

(27.) I have found that only 33 of 124 approval bodies created between 1900 and 2015 required a qualified majority threshold to pass the constitution. Of these, 7 correspond to constituent conventions and 26 to constituent legislatures.

(28.) The idea of “upstream” and “downstream” constraints on constitution-making comes from Elster (1995).

(29.) A similar referendum on the continuity of hereditary monarchy was held in Italy in 1946 as part of the constitution-making process.

(30.) Using a sample of 413 episodes of constitutional change between 1800 and 2000, Ginsburg, Blount, and Elkins (2008, p. 377) find that promulgation of the constitution through a popular referendum has become the most common method of constitutional enactment during the 20th century, particularly after 1950.

(31.) According to Elster (2013), procedural rules should be designed with an eye to removing obstacles to good decisions without pursuing an ideal of what those decisions should be. In his view, a negative form of design is required because there are conflictive normative perspectives about what a good outcome is, and even if we choose one of them, it is uncertain what institutions can reliably promote the chosen conception of goodness (p. 3).

(32.) To be sure, a scholar who considers “political stability” or “effective government” as the most important indicators of success might disagree with this assessment.

(33.) Some well-known indexes of liberal democracy, such as from Polity IV and Freedom House, do have some components related to the rule of law. These components, however, mix and aggregate several factors of widely different natures.

(34.) Prohibiting constitution-makers from competing in future elections may reduce the influence of their personal interests in constitutional choice, but it is not likely to prevent them from advancing the interests of their parties.

(35.) Using a sample of 160 constitution-making events, Mendez and Wheatley (2013, p. 36) find that having direct elections for the constitution-making body and popular ratification of the constitution are negatively and significantly correlated. It is not clear, however, whether this negative correlation holds equally well for the authoritarian and democratic cases included in the sample.

(36.) A recent analysis limited to episodes of constitution-making in Latin America during competitive years shows that there is no significant correlation between legal continuity and postpromulgation levels of democratization. See Negretto (2017b).

(37.) To disentangle potential endogenous effects between democratization and constitution-making, Eisenstadt et al. (2015, p. 31) use general strikes as a purportedly exogenous instrument for levels of participation. The logic is that while general strikes correlate with the level of participation in constitution-making, they affect democracy only through features of the drafting process. This is not convincing because strikes also can force a transition to democracy and affect subsequent levels of democratization outside the constitution-making process, or even in the absence of one.

(38.) Eisenstadt et al. (2015, pp. 21–22) distinguish between “decreed,” “mixed,” and “poliarchic” participation, depending on whether the executive, an existing legislature, or a fairly elected constituent assembly controlled the process and whether public debate existed and had an influence on the draft content. While the distinctions may seem clear theoretically, the authors do not explain their criteria to determine in practice when the views of citizens and civil society organizations are truly taken into account.