Tamara El Khoury
- Supremacy — Legitimacy
Published under the direction of the Max Planck Foundation for International Peace and the Rule of Law.
General Editors: Rainer Grote, Frauke Lachenmann, Rüdiger Wolfrum.
1. Pouvoir constituant (constituent power) is the power to establish the constitutional order of a nation. From the perspective of legal theory, it is the source of production of constitutional normsie, the power to make a constitution and therefore dictate the fundamental norms that organize the powers of the state (Negri 1). Indeed, the classical theory of pouvoir constituant rests on the distinction between a will that predates the constitution and is superior to it (the constituent power) and the positive constitutional forms created by the constituent subject (the constituted powers, including the legislative, executive, and judicial), which determine how public power is to be exercised and ordinary laws created (Colón-Ríos (2010) 205). In short, pouvoir constituant can be understood as the generative principle of modern constitutional arrangements (Loughlin 2004).
2. The concept of pouvoir constituant remains nevertheless elusive. Constitutional theorists have argued that since, paradoxically, it both rises from nowhere and organizes the hierarchy of powers and the law (Negri 1), any legal approach to pouvoir constituant is bound to fail. Indeed, pouvoir constituant is often dealt with as an extra-legal or transcendent source that has no place in constitutional theory. In this regard, Hans Kelsen argues that pouvoir constituant has nothing to do with the formal process of the production of norms, which follow the rules of the legal form (Negri 4). The idea of pouvoir constituant, which has its basis outside of the legal order, is thus avoided, and focus placed on the question of the constitution’s authority as completely internal to the legal order (Dyzenhaus). In contrast, there are those who defend that constitutional legality is not self-generating, and hence, pouvoir constituant cannot be dismissed. According to Carl Schmitt, constitutional authority cannot escape from its origins in pouvoir constituant, and that means that pouvoir constituant can reassert itself from within the constitutional order (Dyzenhaus 129). It is likewise argued that absorbing pouvoir constituant into some basic norm (ie, conflating it with constituted power) would be to deny the continuity of the political dynamic that is at the basis of the state, as well as the open nature of constitutions and their ability to respond to social change (Loughlin (2004) 113).
3. Modern constitutionalism requires that a constitution be authored by the people (Preuss (2007) 211). Put differently, to speak of pouvoir constituant is to speak of democracy (Negri 1); pouvoir constituant is thus identified with ‘the people’ or ‘the nation’. Indeed, the notion that all power originates from the people is now explicitly expressed in various constitutions (representative democracy). For example, according to the German Basic Law, ‘all state authority is derived from the people’ (Basic Law of the Federal Republic of Germany: 23 May 1949 (as Amended to 13 July 2017) Art. 20(2) (Ger)). Likewise, the Constitution of Belgium asserts that: ‘All powers emanate from the Nation’ (Coordinated Constitution of the Kingdom of Belgium: 17 February 1994 (as Amended to 31 January 2014) Art. 33 (Belg)).
4. Though prevalent, the identification of pouvoir constituant with popular sovereignty is by no means encompassing. Thus, it is the conception of constituent power as vested in a sovereign parliament that lies at the heart of British constitutionalism. The idea of parliament as the bearer of pouvoir constituant pits constitutional supremacy / primacy against parliamentary sovereignty. Nowhere has it played out more openly than in India, which experienced a struggle between parliament and the courts as to the extent of the former’s power to amend the constitution (Colón-Ríos (2017) 9). However, the identification of pouvoir constituant with the people does not necessarily mean that the latter can engage in exercising it outside of the context of a political revolution. Thus, even though Art. V of the United States (‘US’) Constitution contemplates the calling of a Constitutional Convention that could arguably exercise constituent power, such a convention can only be triggered through a supermajority in Congress (Constitution of the United States of America: 17 September 1787 (as Amended to 7 May 1992) (US)). To the contrary, and despite the absence of direct references to constituent power, the obiter dicta in a number of recent English cases have pointed to the existence of implicit limits to parliament’s constitutional amendment power. This approach, sometimes referred to as ‘common law constitutionalism’, if ever accepted by the courts, would amount to a rejection of parliament as the bearer of constituent power (Colón-Ríos (2017) 10). Thus, the boundary between the two conceptions is becoming increasingly porous.
5. From the perspective of constitutionalism, pouvoir constituant immediately raises the question of whether it disappears once it has turned into authority, ie, constitutional form, or lingers on, threatening to disrupt its creation (Dyzenhaus 129). It is precisely because of the expansive democratic principle it expresses that pouvoir constituant rests uneasily with constitutionalism, which is essentially a theory of limited government. If, however, it is conceded that pouvoir constituant, rather than express ‘the power of the multitude’, necessarily assumes a representational form, then, thus contained, it can come to express a relation of interdependence between democracy and constitutionalism, rather than a mere contradiction (Loughlin (2003) 113).
6. The manner in which that interdependence plays out is a matter of constitutional practice. From this viewpoint, the interest of pouvoir constituant mainly lies in its significance for the binding force of the constitution (Preuss (2007) 211), and, one might add, its capacity to generate, act from withinand even uponthat constitution. If pouvoir constituant is to be identified with ‘the people’, then the ‘rules for the recognition of the people as constituent power’ (Preuss (2007) 212) acting in any of the above-stated capacities, become the main object of our comparative analysis. This is, in order to understand how pouvoir constituant and the constitutional order interact, one ought to explore the different instances in which the former acts upon the latter, and the extent to which it does so. More specifically, it is the analysis of the rules of constitution-making, particularly from within the constitutional order, as well as constitutional amendment, and the limits thereof, that can allow to ‘test’ the strength of pouvoir constituant against constitutional form.
7. What do contemporary constitutionsthe expressions of constitutional formtell us about the nature and scope of pouvoir constituant? Pouvoir constituant is not simply reflected in the notions of popular and national sovereignty that the vast majority of constitutions allude to as a source of legitimacy. It is also expressed through its agents, which oft give form to the constitution-making and constitutional amendment processes they enclose, such as constituent assemblies, popular initiatives, and referendums (direct democracy). It further manifests itself in the limits placed on constitutional amendment (amendment or revision of constitutions), be it at the level of the text itself or through the judicial review of ‘unconstitutional constitutional amendments’. Whereas the ability of ‘the people’ to re-constitute themselves at any time is directly linked to democracy (Colón-Ríos (2010)), the extent to which pouvoir constituant can continue to express itself from within the constitutional order pits constitutionalism against democratic innovation.
8. The traditional distinction between pouvoir constituant originaire (original or primary) and pouvoir constituant dérivé (derived or secondary) rests on the identification of the former with a power that is exercised in revolutionary circumstances, outside the laws, and the latter with the power exercised under legal circumstances, according to rules established by the constitution. In constitutional practice, pouvoir constituant dérivé is primarily exercised through constitutional amendment. There are constitutions that establish different procedures for their amendment, and their complete revision, thus acknowledging the relation between the concept of the people as pouvoir constituant and the requirement for their immediate involvement in a total revision of the constitution, be it through the election of a constituent assembly or a mandatory referendum. For example, the Spanish Constitution distinguishes between an ordinary procedure for its amendment (Art. 167), and an exceptional procedure for its total or partial (concerning fundamental rights) revision that requires a mandatory referendum (Art. 168) (Constitution of the Kingdom of Spain: 6 December 1978 (as Amended to 27 September 2011) (Spain)). Other constitutions do not distinguish between the conditions that are required in each of the two instances. For example, according to the Fundamental Law of Hungary, parliament both enacts and amends the constitution on the basis of a qualified majority (Constitution of the Republic of Hungary: 18 April 2011 (as Amended to 26 September 2013) Art. 1 (Hung)). Still, a growing number of constitutions establish express limits regarding specific constitutional subjects that cannot be changed. Thus, the Romanian Constitution forbids the amendment of its provisions regarding the republican form of government, political pluralism, and the official language among other subjects (Constitution of Romania: 21 November 1991 (as Amended to 29 October 2003) Art. 152 (Rom)).
9. While the definition of the nature of constitutional amendment power remains highly divisive amongst legal theorists, the question of the limits placed on that power in constitutional practicebe it at the level of the constitution itself or through the activity of the courtshas important, practical consequences for the degree of continuity and the preservation of the ‘identity’ of the constitution, ie, for the manner in which pouvoir constituant effectively relates to the constitutional order. In a way, the degree of entrenchment of a constitution will depend on the extent to which the distinction between pouvoir constituant originaire and pouvoir constituant dérivé is upheld (rigid (entrenched) / flexible constitutions). In this regard, the incorporation into contemporary constitutions of ‘eternity clauses’ (entrenched clauses)provisions prohibiting the amendment of certain constitutional subjectsplays a key role in the establishment of that distinction, and raises the question of the capacity of pouvoir constituant to continue to act from within the constitutional order. In particular, it is the comparative analysis of the question of whether courts hold the power to check the pouvoir constituant dérivé against such provisions, that is, in substantive terms, that can best depict the nature of the relation between pouvoir constituant and constitutional form, and the extent of the capacity of the former to keep operating despite the existence of the latter.
10. It thus becomes clear that notwithstanding its complex nature as a concept in constitutional theory, pouvoir constituant is far from irrelevant in constitutional practice. Whether from the perspective of constitutional amendments and the possibility for courts of reviewing their constitutionality, or from that of the openness of constitutions to pouvoir constituant and its agents, ie, democratic innovation or constitution-making from within the constitutional order, the battle between pouvoir constituant and constitutionalism is on.
11. This entry begins with a brief overview of the origins of pouvoir constituant and the different forms it assumes. It then moves on to provide a comparative description of the manner in which different constitutions and constitutional courts have conceived of those different forms and conceptualized the relationship between them. A brief exploration of the constitutional incorporation of pouvoir constituant agents will ensue. Lastly, an overall assessment of the role that pouvoir constituant plays in constitutional practice will be offered.
B. Origins of Pouvoir Constituant
12. Pouvoir constituant is a modern concept with medieval roots, which nevertheless distinctly emerges with the establishment of the modern state (Loughlin (2014) 219). The appearance of the concept of pouvoir constituant is concomitant to that of the belief that the authority of government rests on the consent of the people. While ‘the people’ invoked at first are an inert and symbolic collectivity, they are eventually transformed, following the revolutionary upheavals of the seventeenth and eighteenth centuries, into some form of active political agency. It is precisely in the process through which ‘the people’ are converted from a passive to an active force that the origins of pouvoir constituant can be found. This is particularly true of the American and French revolutions, which resulted in the overthrow of established legal orders. ‘The people’ were accordingly recognised as the originating source of political power, not simply as a ‘tenet of belief’, but also as a matter of ‘political practice’’ (Loughlin (2003) 100–108).
13. The origin of the expression pouvoir constituant is attributed to Abbé Emmanuel Joseph Sieyès, although the concept itself can at the least be traced back to the social contract theorists. Sieyès famously declared that ‘a constitution presupposes above all a pouvoir constituant’. His too is the assertion according to which ‘in each of its parts, a constitution is not the work of a constituted power but a constituent power’ (Gözler 8). Thus, pouvoir constituant is pitted at the outset against pouvoir constitué. Earlier, John Locke had claimed that ‘there remains still in the people a supreme power to remove or alter the legislative, when they find the legislative act contrary to the trust reposed in them’ (Locke 77–78). To the contrary, according to Thomas Hobbes, the bearer of constituent power, ‘the multitude’, not only vanishes with the creation of the sovereign but is reconstituted as individual subjects, each with an absolute obligation of obedience to the sovereign (Dyzenhaus 138). By asserting its unlimited nature, Sieyès both supersedes Locke’s confinement of pouvoir constituant within the right of resistance, and Hobbes’ subjugation of it. Sieyès further considered the problem of the exercise of pouvoir constituant, and conceded that it necessarily requires some form of representation. He was however careful to emphasize the extraordinary nature of that representation, which, contrary to ordinary representation, is bound by no pre-existing constitutional limits. Thus, pouvoir constituant is, since its inception, attached to representation.
14. In the US, the Declaration of Independence presented itself as the supreme act of pouvoir constituant, which established both a new framework of government and a novel meaning of constitution (Loughlin (2003) 108). The latter was now understood, as Thomas Paine put it, not as ‘the act of [a country’s] government, but of the people constituting a government’ (Paine). Pouvoir constituant thus made it possible to distinguish between ‘constitutional’ and ‘legal’, or, in Alexander Hamilton’s words, between ‘the representatives of the people’ and ‘their constituents’ (Hamilton, Federalist No. 28). This distinction lay at the origin of the defence of republican government, which, unlike democratic government, inscribes the pouvoir constituant of the people in the very framework of constituted power (Loughlin (2003) 108).
C. Forms of Pouvoir Constituant
1. Original Pouvoir Constituant v Derived Pouvoir Constituant
15. Original pouvoir constituant can be identified with a pre-constitutional political capacity that emerges in revolutionary constitutional moments (constitutional moment) and is subsequently absorbed by the constitutional system, in contrast with the derived pouvoir constituant or power of constitutional reform, which is exercised by constituted entities through legally qualified procedures. The rigidity of such distinction is nevertheless increasingly challenged, both in theory and practice. Alternative modes of pouvoir constituant rest on the distinction between revolution and augmentation. The latter describes a mode that lies between revolution and reform, as it neither aims at replacing the constitutional order nor proceeds in accordance with the pre-defined procedures for piecemeal amendment (Patberg 9). Augmentation thus enables the differentiation between extraordinary and ‘extra-in-the-ordinary’ pouvoir constituant (Patberg 9); while the former is extralegal, since it precedes the constitutional order, the latter is illegal, given that it ‘ruptures a given framework of norms, but nonetheless emerges from and remains explicable within its broader horizon of meaning and legitimacy’ (Wenman 163). Augmentation can thus be equated with (re)foundation. However, overemphasizing (re)foundation as the only outlet for the expression of pouvoir constituant in a global world that increasingly requires the introduction of new social and political forms has been criticized by more fervent proponents of an agonistic conception of democracy. Such conception considers both modes of pouvoir constituant as essential expressions of democracy (Patberg 8). As such, it begs a critical analysis of constitution-making processes as the primary expression of original pouvoir constituant. It further urges an exploration of the nature of amendment power and the degree to which it enables constitutional change, ie, the manner in which it relates to that original pouvoir constituant.
16. While constitution-making is the practice of original pouvoir constituant, constitution-making processes are the ultimate expression of the paradox it represents. Indeed, inasmuch as the latter involve the establishment of rules and procedures that can channel the people’s will (eg constituent assemblies or referendums), it is in fact a constituted powerthe institutional framework established to that effectthat enables the exercise of pouvoir constituant. In that sense, pouvoir constituant is never truly original.
17. The widespread adoption in recent decades of diversified instruments aimed at enhancing public participation in constitution-making processes has signalled a shift in pouvoir constituant from the realm of the imaginary to the realm of the historical. Inclusive constitution-making tools including the direct election of constituent assemblies, the organization of public consultations, whether face-to-face or online, holding opinion polls around constitutional matters, the publication of provisional articles for public comment in real-time, have been resorted to in a variety of processes ranging from South Africa’s democratic transition in the 1990s to Iceland’s experiment of a ‘crowdsourced constitution’ (Landemore) in 2013 to Tunisia’s consensus constitution in 2014, with the purpose of enabling a real, if not original, exercise of pouvoir constituant.
18. While the democratization of constitution-making processes has reinforced the exercise of original pouvoir constituant, the question of how subsequent constitutional amendment relates to that original constituent power continues to be met with different responses, both at the levels of constitutional theory and constitutional practice.
19. According to proponents of the constituent nature of amendment power, if ‘the people’ control the government through the constitution, then they necessarily control the authority to amend that constitution. From this perspective, pouvoir constituant is understood to be shared by the original constitution-makers with future generations, precisely by means of the amendment power (Roznai (2017) 111) This, it has been argued, is the prevailing approach of US constitutionalism, for which following the establishment of the US Constitution, Art. 5, through which ‘the people’ may amend the Constitution, contains the constituent power (Griffin 66). There are on the other hand those who hold that amendment power is a constituted power. Accordingly, it is argued that if within a constitutional polity all powers originate in and are defined by the constitution, then the amending power must be a constituted power just like the legislative powers, judicial powers, or executive powers (Preuss (2011) 430). Lastly, there are those who place amendment power between pouvoir constituant and pouvoir constitué. According to Schmitt, a distinction has to be made between the basic substantive principles of the constitution that embody its identity, and the constitution’s procedural rules, which are designed to protect its essence, and cannot be employed to annihilate the constitution (Roznai (2013) 675).
20. From a constitutional practice perspective, the question of the nature of the amendment power determines its scope and limits, with subsequent consequences on the degree of entrenchment of the constitution and the capacity of pouvoir constituant to continue operating from within. Attempts have been made at differentiating between a monist and a dualist approach to amendment power. According to the former, amendments are dealt with as an expression of pouvoir constituant, albeit dérivé, whereas the latter differentiates between constitution-making (or the complete overhaul of the constitution) and constitutional amendment, and only attributes the status of pouvoir constituant to the first of the two processes. Whether a given constitution includes ‘eternity clauses’ (or otherwise identifies the core elements of the constitution) is likely to favour the justiciability of constitutional amendments against such clauses, as has been the case in Germany or Turkey for example. However, it is also true that the monist approach has been maintained in instances where unamendable provisions do exist, as in France. In contrast, despite the absence of explicit immutable clauses sustaining a dualist approach, there are examples of constitutional practice where courts have construed implicit constitutional limits to the power of amendment, thus undermining or outright negating its constituent nature. This would most notably be the case of the Supreme Court of India. More recently, English courts have begun referring to the existence of ‘constitutional fundamentals’ that even a sovereign parliament cannot abolish, potentially paving the way for the emergence of an alternative constituent power (Colón-Ríos (2017) 10). Lastly, there are but a few constitutions that have expressly accorded constitutional courts the jurisdiction to review the constitutionality of constitutional amendments on substantive grounds (powers and jurisdiction of constitutional courts / supreme courts). This is the case for example of the Constitution of Romania (Art. 146(a)) and the Constitution of the Republic of Kosovo: 9 April 2008 (Kos) (Arts. 113(9) and 144(3), which specifically enable the court to examine whether proposed amendments diminish the rights and freedoms guaranteed by Chapter II of the Constitution. In the absence of unamendable provisions however, deference to the original pouvoir constituant can only be judicially inferred (judicial deference).
21. In comparative literature, provisions forbidding the amendment of certain constitutional subjects are referred to as ‘immutable’, ‘unchangeable’, ‘unalterable’, ‘irrevocable’, perpetual ordrawing from the German term Ewikeitsgarantie‘eternal’. It has nevertheless been argued that those expressions are misleading, for such provisions are not beyond the reach of the primary pouvoir constituant, and should therefore be called ‘unamendable provisions’ instead (Roznai (2017) 16). Examples of unamendable provisions abound, and generally concern fundamental rights (eg, Art. 212(5) of the Constitution of the People’s Democratic Republic of Algeria: 28 November 1996 (as Amended to 6 March 2016) (Alg); Art. 441 of the Constitution of the Republic of Ecuador: 28 September 2008 (as Amended to 7 May 2011) (Ecuador)), the form of government (eg, republican: Art. 139 of the Constitution of the Italian Republic: 22 December 1947 (as Amended to 20 April 2012) (It); eg, monarchical: Art. 120(c) of the Constitution of the State of Bahrain: 14 February 2002 (Bahr)), the state’s political structure (eg, federal: Art. 60(§4°) of the Constitution of the Federative Republic of Brazil: 5 October 1988 (as Amended to 15 September 2015) (Braz); Art. 79(3) of the German Basic Law) and the democratic order (eg, multi-party system: Art. 165 of the Constitution of Burkina Faso: 11 June 1991 (as Amended to 5 November 2015) (Burk Faso); eg, presidential term limits: Art. 281 of the Constitution of the Republic of Guatemala: 31 May 1985 (as Amended to 17 November 1993) (Guat)) among other matters (Roznai (2017) 23–25).
2. Regular Pouvoir Constituant v Irregular Pouvoir Constituant
22. Irregular pouvoir constituant may refer to primary pouvoir constituant in the sense of a non-regulated power, which has traditionally been considered not to be subject to legal rules, and which is increasingly understood to be regulated by public international law and the principles of justice (Sagüés 151). It more often refers to derived constituent power, which is exercised in violation of existing constitutional rules (Sagüés 151). Nowhere is the tension between constitutionalism and democracy made more evident than in the exercise of irregular constituent power in this latter sense. In response to Quebec’s desire to secede unilaterally (secession), the Supreme Court of Canada (Cour suprême du Canada) circumvented such irregular exercise of pouvoir constituant by overcoming the constraints of constitutionalism with democratic arguments that it read into ‘the larger context of other constitutional values’. It thus held that although ‘secession of a province ‘under the Constitution’ could not be achieved unilaterally ... a clear majority vote in Quebec on a clear question in favor of secession would confer democratic legitimacy on the secession initiative which all of the other participants in Confederation would have to recognize’ (Reference Re Secession of Quebec Case (Can) (1998)). In contrast, the Constitutional Court of Spain (Tribunal Constitucional de España) has consistently confined democratic expression within constitutional form as far as the ‘reconsideration of the constituted order’ is concerned (Judgment 103/2008 (2008) (Spain)). In a highly controversial decision involving the exercise of irregular constituent power, the Venezuelan Supreme Court of Justice pronounced itself on the establishment of a new constitutional order through a procedure that was not contemplated under the amendment rules of the Constitution of the Republic of Venezuela: 16 January 1961 (Venez) that was in force at the time. The government had called for a referendum on whether the Venezuelans wished to convene a constituent assembly with the objective of adopting a new constitution. In a decision that explicitly acknowledged the tension between democracy and constitutionalism, the court held that amendment rules only applied to government, whereas the exercise by the people of constituent power included the ability to alter the constitutional regime through extra-constitutional means, inasmuch as it is ‘prior and superior to the established legal order’ (Decision ‘La Roche’ 1999). Interestingly enough, the constitution that was subsequently approved introduced, perhaps with the intent of averting such irregular exercise in the future, the possibility for that original pouvoir constituant to be activated from within the constitutional order itself.
D. Comparative Description
1. Constitutional Amendment is Pouvoir Constituant: The Case of France
23. Notwithstanding the existence of an ‘eternity clause’, the Constitutional Council of France (Conseil Constitutionnel) has consistently refused to substantially review constitutional amendments by emphasizing the sovereignty of the French people. In doing so, it has equated amending power with pouvoir constituant. Though not uncontested, the prevailing stance in French doctrine is best expressed through the successive double amendment thesis according to which the provision forbidding the amendment can first be amended; subsequently, the content of that prohibition can be so (Le Divellec, Levade and Pimentel).
24. In 1962, the Conseil constitutionnel was requested to pronounce itself on the constitutionality of the referendum on direct presidential elections, which effectively amended the 1958 Constitution. In interpreting its mandate, the Council distinguished between laws that are voted by parliament and over which it asserted its competence, and laws that are ‘adopted by the People following a referendum’ and which ‘constitute the direct expression of national sovereignty’, thus excluding its competence (Decision 62–20 DC (1962) (Fr) para. 19). Although the referendum was procedurally faulty and hence unconstitutional, the Council still considered that it amounted to an expression of the nation’s will, over which it had no jurisdiction. The Council’s identification of popular referendums with the pouvoir constituant of the people contrasts sharply with the stance of the Constitutional Court of Colombia (Corte Constitucional de Colombia) which, faced with a question regarding the constitutionality of a referendum aimed at amending the constitutional presidential two-term limit, declared that a referendum ‘as a mechanism of constitutional reform is, always, an expression of derived constituent power, and not even the intervention of the electorate to vote the proposal… has sufficient legal force to transform the referendum into a constituent foundational act, primary or original’ (Decision C-141/10 (2010) (Colom)).
25. In its 1992 Maastricht Treaty decision, the Conseil was even more explicit in its qualification of amendment power: ‘The constituent authority is sovereign; it has the power to repeal, amend or amplify constitutional provisions in such manner as it sees fit; there is accordingly no objection to insertion in the Constitution of new provisions which derogate from a constitutional rule or principle; the derogation may be express or implied’ (Decision 92–312 DC (1992) (Fr) para. 19). If the Conseil expressed no doubt regarding the constituent nature of the amendment power, it nevertheless declared it ‘subject to the provisions governing the periods in which the Constitution cannot be revised (Arts 7, 16 and 89(4)) and to compliance with the ‘eternity clause’ contained in Art. 89’ (Decision 92–312 DC para. 19).
26. The question henceforward was whether the obligation of compliance with Art. 89 could open the door for the substantive review of an amendment against that provision, ie, the characteristics inherent to the republican form as originally determined by ‘the people’. Seized with regard to the constitutionality of the 2003 constitutional amendment that established the decentralized organization of the Republic, the Council laconically replied that it ‘does not draw the power to decide on a constitutional revision either on the basis of Art. 61 or of Art. 89 or of any other provision of the Constitution’ (Decision 2003–469 DC (2003) (Fr) para. 2).
27. Interestingly in the French instance, the question has been raised regarding the limits of the amending power in effecting successive inroads into ‘the essential conditions for the exercise of sovereignty’ (as a result of European Integration), insofar as ‘the French constitutional order is constructed around the central notion of national sovereignty’ (Decision 92–312 DC para. 44). Put differently, it is the question of whether the existence of implicit limits willed by the pouvoir constituant originaire can be inferred from its very centralityexpressed through the notion of sovereigntyto the constitutional order. The Conseil has so far refrained from responding.
2. Original Pouvoir Constituant is Superior to Amendment Power: The Cases of Turkey, Germany, Austria, the Czech Republic and Brazil
28. The Constitution of the Republic of Turkey: 7 November 1982 (as Amended to 21 January 2017) (Tur), like its French counterpart, contains an unamendable provision concerning its republican form (Art. 1) as well as the characteristics of the Republic, which are that it is ‘a democratic, secular and social state governed by the rule of law’ (Art. 2). Unlike the French Constitution, however, it explicitly foresees the possibility of the Constitutional Court to formally review the constitutionality of constitutional amendments (Art. 148.1), which is specifically restricted to the requisite majorities and the prohibition on debates under urgent procedure (Art.148.2). Notwithstanding this constitutional limitation of its jurisdiction, the Constitutional Court of Turkey (Anayasa Mahkemesi) opted for a decidedly dualist approach in the 2008 Headscarf Decision, in which it upheld the superiority of the ‘primary constituent power’ over the amendment power of parliament on the basis of the constitution’s unamendable provisions (Headscarf Decision (2008) (Turk)). In a highly controversial decision, the court struck down a number of constitutional amendments aimed at allowing headscarves in universities by arguing that lifting the ban ‘indirectly changes and makes nonfunctional the basic features of the republic’. In addressing the nature of the amendment power, the court established a distinction between ‘original’ and ‘derived constituent power’ to conclude that the constituted powers created by the original constituent power derive their prerequisite legality from the constitution and thus must act within the limits defined in the constitution (Roznai (2012) 184). According to the court, the constitution’s unamendable provisions reflect the constituent power’s basic choice of political system, and the tangible reflections of this choice come into existence in the other articles of the Constitution (Roznai (2012) 185). In contrast with the French Council, the Turkish Court thus establishes a hierarchy within the constitution itself, which effectively encloses pouvoir constituant within constitutional form.
29. In what is perhaps the most renowned of ‘eternity clauses’, the German Basic Law prohibits amendments affecting the division of the Federation into Länder, human dignity (dignity and autonomy of individuals), the constitutional order, or basic institutional principles describing Germany as a democratic and social federal state (Art. 79(3)). Set against the Weimar Republic and adopted in the aftermath of the Second World War, it has been claimed that the Basic Law broke new ground by distinguishing amendable and immutable components (Preuss (2011) 439). However, the German Basic Law did not expressly grant the Federal Constitutional Court of Germany (Bundesverfassungsgericht) the power to review constitutional amendments. Despite the absence of an explicit review mandate, the court early on asserted that it could review constitutional amendments that conflict with higher law or fundamental principles of the Basic Law and declare them unconstitutional (Südweststaat (1951) (Ger)). Faced with a question reminiscent of that which was addressed to the French Conseil, the court had to determine whether the Lisbon Treaty affected the principle of democracy referred to in Art. 79(3) given that it might entail Germany to join a federal European state. In an explicit reference to pouvoir constituant, the court stated that the ‘eternity clauses’ have more authority than the amending power of the legislature because such power has founded the structure and substance of the Basic Law, which cannot be amended’ (Lissaboner Vertrag Case (Ger) (2009)). Unlike the Conseil constitutionnel that refused to consider the matter, the German court stated that the irrevocable transfer of sovereignty to a new subject of legitimisation … is reserved to the directly declared will of the German people alone [ie to its constituent power] (Preuss (2011) 442). ‘It is the constituent authority alone, and not the constitutional authority emanating from the constitution, which is entitled to release the state constituted by the Basic Law’ (Lissaboner Vertrag 179).
30. The distinction made by the court between pouvoir constituant and its constituted powers has nevertheless been underplayed. Since the will of the pouvoir constituant is embodied in the constitution itself, ‘respect for the constituent power’s privilege to unbind the constituted powers from the bonds of the ‘eternity clause’ is simply respect for the constitution and for its stop rule for constitutional change laid down in Art. 79.3’ (Preuss (2011) 442). Although the court acknowledges the legitimacy of the exercise of the pouvoir constituant of the people, by defending the pouvoir constituant embodied in the constitution, it reduces the pre-constitutional freedom of the people to an ‘eternally tacit and inactive’ force (Preuss (2011) 433). This analysis is in line with the assertion according to which Art. 146, which refers to the exercise of primary constituent power, is a reflection of the ‘immunization of German constitutional law against the idea of a constituent power’ (Möllers 97). Indeed, a considerable part of German doctrine assumes that even that pouvoir constituant of the people is bound by the fundamental standards of Art. 79(3). Although this approach is not uncontested, it reflects the virtually unlimited legalisation of the constituent power, and the extent to which the discourse between power and form leans toward form (Möllers 97). In the Lissaboner Vertrag Case, the court further emphasized form by arguing that ‘it may remain open due to the universal nature of dignity, freedom and equality alone [whether the Basic Law’s eternity clause] even applies to the constituent power’ (Lissaboner Vertrag 217). It has been argued that the ‘mystification of constituent power’, ie, the recognition of the people’s power to adopt a new and radically modified constitution without providing any institutional means for that power to be exercised [as does Art. 146] amounts to doing away with it (Colón-Ríos (2010) 233). Significantly, the German Federal Constitutional Court has nevertheless so far refrained from declaring any constitutional amendment unconstitutional.
31. Unlike its German counterpart, in 2011, the Constitutional Court of Austria (Verfassungsgerichtshof) annulled a constitutional provision in what was conceivably the first such decision in Europe (Pfersmann 83). In contrast, the substantive review of constitutional amendments is an established practice is Brazil. According to the Supreme Federal Tribunal of Brazil (Supremo Tribunal Federal), the explicit material limitations defined in the Constitution (Art. 60(§4°)) ‘constrain the reforming power conferred upon the legislative’ due to the ‘immutability of such thematic nucleus’ which renders its review legitimate (Decision 466/91 (Braz)). In a similar vein, the Czech Constitutional Court has deemed that Art. 9(2) of the Constitution of the Czech Republic: 16 December 1992 (as Amended to 14 November 2002) (Czech), which states that ‘any changes in the essential requirements for a democratic state governed by the rule of law are impermissible’, is an enforceable constitutional provision that protects the constitution’s material core (Decision Pl. ÚS 27/09 (2009) (Czech)). Increasingly, pouvoir constituant is congealed in form.
3. Is Original Pouvoir Constituant Implicitly Superior? The Cases of India, South Africa, Georgia and Hungary
32. Despite the fact that the Constitution of the Republic of India: 26 January 1950 (as Amended to 28 May 2015) (India) contains no unamendable provisions, the Supreme Court has developed a distinct dualist approach on the basis of implied substantive limits. In its landmark decision, Kesavananda Bharati v State of Kerala (1973) (India), the court held that since it was doubtful that the founders wished to authorize the destruction of the constitution by the process of internal amendment, the amendment in question, which purported to place a number of statutes beyond judicial review, was unconstitutional because it contravened the ‘basic structure’ of the constitution, which can only be changed by a constituent assembly. Although the Supreme Court has never attempted a comprehensive definition of the scope of ‘basic structure’, individual justices have included, among others, the ideas of the supremacy of the constitution, the basics of democratic government, the sovereignty of India, and judicial review (Neuborne 491). Parliament responded to the court’s basic structure doctrine by amending the constitution so that ‘no amendment of this Constitution … made or purporting to have been made under this article… shall be called in question in any court on any ground’ and further clarified that ‘it is hereby declared that there shall be no limitation whatever on the constituent power of Parliament to amend by way of addition, variation or repeal the provisions of this Constitution under this article’ (Art. 368(4) and (5)). In Minerva Mills Case (India), delivered in 1980, the court deemed the amendment unconstitutional since the approach according to which the authority to amend is restricted is a basic component of the Indian Constitution, which cannot be amended (Barak 327). Despite the absence of a textual basis to its basic structure doctrine, the court has nevertheless inferred the superiority of the original pouvoir constituant over its derived formthe amendment power.
33. At any rate, where unamendable provisions have not been foreseen by the constitutional text itself, the basic structure doctrine has not been readily received. For instance, despite the fact that the Constitution of the Republic of South Africa: 16 December 1996 (as Amended to 1 February 2013) (S Afr) allows for the review of all constitutional amendments (Art. 167.4), in the absence of ‘eternity clauses’, the court has so far limited itself to procedural considerations (Premier of Kwazulu-Natal and Others v President of the Republic of South Africa and Others (1995) (S Afr)). Following an explicit reference to the jurisprudence of the Indian Supreme Court, the court stated that if properly done, an amendment is ‘constitutionally unassailable’. However, it conceded that were a formally sound amendment to imply ‘radically and fundamentally restructuring and re-organizing the fundamental premises of the Constitution’, it ‘might not qualify as an ‘amendment’ at all’. Notwithstanding its acknowledgement of the possibility of the existence of ‘this kind of implied limitation to what can properly be the subject matter of an amendment to our Constitution’ (at para. 47) the court concluded that the amendments in question were not so basic as to abrogate or destroy the constitution. Hence, the scope of the basic structure doctrine in South Africa remains unclear. Similarly, the Constitution of Georgia: 24 August 1995 (as Amended to 15 October 2010) (Geor) contains no unamendable provisions. Interestingly, in Citizens of Georgia Irma Inashvili, Davit Tarkhan-Mouravi and Ioseb Manjavidze v Parliament of Georgia (2013) (Geor) the Constitutional Court did recognize the existence of constitutional principles that have the important role of checking the constitutionality of challenged provisions. However, it also emphasized that it had no legal foundation to deliver an opinion about their constitutionality based on only constitutional principles. It further noted that since the constitution does not include ‘eternity clauses’, any other interpretation of the court’s own competence would be a violation of the principle of separation of powers (Gegenava 403). Deference to the principle of separation of powers has equally been exercised by the Hungarian Constitutional Court. Inasmuch as the Constitution of Hungary does not distinguish between the procedures of constitution-making and constitutional amendment, the court has long held that amendments are an exercise of pouvoir constituant. More recently however, the court decided that while it is not competent to review the constitutionality of constitutional amendments, the obligations that Hungary has undertaken in its international or European treaties, along with jus cogens, constitute a unified system of values it will not disregard in the course of judicial review (Decision 45/2012 (2012) (Hung)). It could tentatively be inferred that in establishing its competence of finding the coherent interpretation of the Fundamental Law, the court has decided on the possibility of effecting substantive review, since it will have to interpret the constitution in light of potentially incoherent constitutional elements.
4. Can Original Pouvoir Constituant be Unconstitutional? The Case of Honduras
34. In 2015, the Constitutional Chamber of the Supreme Court of Honduras issued an unprecedented decision in comparative jurisprudence, in which the original pouvoir constituant was deemed unconstitutional. The provisions in question concerned the creation and protection of presidential term limits. The court heard the case using the action of unconstitutionality, which is available against constitutional reforms, whereas the effect of the decision was to hold inapplicable several provisions of the original constitution itself (Decision F-165 (2015) (Hond)). In doing so, it hinted at the doctrine of unconstitutional constitutional amendment. Contrary to the standard logic of this doctrine, however, which holds that the amendment power is limited in order to preserve the more fundamental pouvoir constituant of the people themselves, the court considered that the latter was unconstitutional given the change in circumstances (Landau).
35. While the practice of constitutional amendment offers much insight into the manner in which pouvoir constituant relates to the constitutional order, the capacity of pouvoir constituant to be exercised beyond mere constitutional amendment and from within the constitutional order can help elucidate the extent of openness of constitutionalism to democratic renewal.
5. Agents of Pouvoir Constituant: Constituent Assemblies and Popular Initiatives
36. Whether a given constitution incorporates agents of pouvoir constituant that can aid its exercise is a matter of democratic constitutional change, and hence of democratic legitimacy. Besides the mystification of pouvoir constituant, which is a common feature of modern constitutionalism, three other manners of neutralizing pouvoir constituant have been identified: displacing, legalizing, or hiding it (Colón-Ríos (2010) 233). A classical instance of traditionally displaced pouvoir constituant toward ordinary government occurs in the United Kingdom, where in the absence of a central, written constitution, the principle of parliamentary sovereignty effects an almost complete identification between ‘the people’ and ‘Parliament’ (Colón-Ríos (2010) 233). Israel, too, lacks a formalized constitution, ie rules of organization and function that are clearly distinguished from ordinary law-making. Though possessing a number of Basic Laws that could be considered formally constitutional because of the qualified majorities required for their adoption, or materially so because of the limits they set on ordinary laws, such laws, in the tradition of the Westminster Constitution, do not have a ‘derogatory superiority’ (Pfersmann 109). However, in the famous Mizrahi Bank Case (Isr), the Israeli Supreme Court held that those Basic Laws were indeed the chapters of the Israeli Constitution and thus entrenched, and concluded that it was competent to annul provisions violating norms laid down in them (United Mizrahi Bank v Migdal Cooperative Village (1995) (Isr)). From the perspective of pouvoir constituant, it can be claimed that its invocation by the court retroactively placed it at the origin of the polity.
37. Interestingly enough, it is held that both the quasi-identification of amendment power with pouvoir constituant (where amendment procedures are relatively loose), and the protection of pouvoir constituant through extremely rigid procedures, result in its displacement, since it rarely if ever can be truly exercised. Contrary to the view held by the French Conseil constitutionnel, it has been equally argued that the constitutional referendum is not an agent of pouvoir constituant since its primary role is to render amendment more difficult (Colón-Ríos (2010) 235).
38. The question of establishing an ‘outlet for the constituent power’ (Colón-Ríos (2010) 201) within the constitutional order has been taken up by a number of constitutions that allow for a constituent assembly to be convened ‘from below’, including Bolivia, Ecuador and Venezuela. For example, according to the Constitution of the Pluri-National State of Bolivia: 25 January 2009 (Bol), the total reform of the constitution or those modifications that affect its fundamental principles or the supreme nature and amendment procedure can take place through a ‘sovereign constituent assembly’ that can be activated, amongst other instances, by popular initiative, by the signature of at least twenty per cent of the electorate (Art. 411). Likewise, the incorporation of recall referendums into the Venezuelan and Peruvian constitutions has been highlighted as an instrument that has tapped into the pouvoir constituant of the people (Cameron and Sharpe 112).
E. Comparative Assessment
39. It has been asserted that the analysis of decisions taken by constitutional judges indicates that there is a constant and largely shared tendency to recognize the intangibility of certain constitutional principles or rules, often with the object of protecting fundamental rights (Le Divellec, Levade and Pimentel). Be it on the basis of the existence of unamendable provisions (eg Austria, Czech Republic, Brazil, Turkey, Germany) or in the absence thereof (eg India), numerous courts have upheld the superiority of pouvoir constituant originaire. Significantly however, the German Federal Constitutional Court has so far refrained from declaring any amendment unconstitutional. Likewise, restraint has been exercised by a number of courts, be it on the basis of the identification between sovereignty and amendment power (eg France) or through deference to the principle of separation of powers (eg Hungary and Georgia). From the perspective of pouvoir constituant, the question of whether the constitution can be contrary to the constitution is a question of the extent to which constituent power is superseded by constitutional form, ie the capacity of the latter to contain the former. Whilst the idea of democracy requires that pouvoir constituant be enabled to act, the identification of ‘intra-constitutional hierarchies’ (Pfersmann 93) within the constitution itselfthough aimed at upholding democracyhas the potential effect of neutralizing pouvoir constituant.
- Barak, A, ‘Unconstitutional Constitutional Amendments’ (2011) 44 IsLR 321.
- Cameron, MA, and Sharpe, KE, ‘Andean Left Turns: Constituent Power and Constitution-making’ in Cameron, MA, and Hershberg, E, (eds), Latin America’s Left Turns: Politics, Policies and Trajectories of Change (Lynne Rienner Publishers 2010) 61.
- Colón-Ríos, J, ‘Five Conceptions of Constituent Power’ (2017) 7.31 Victoria University of Wellington Legal Research Papers 127.
- Colón-Ríos, J, ‘The Legitimacy of the Juridical: Constituent Power, Democracy, and the Limits of Constitutional Reform’ (2010) 48 OsgoodeHallLJ 199.
- Colón-Ríos, J, Weak Constitutionalism: Democratic Legitimacy and the Question of Constituent Power (Routledge 2012).
- Dyzenhaus, D, ‘The Politics of the Question of Constituent Power’ in Loughlin, M, and Walker, N, (eds), The Paradox of Constitutionalism: Constituent Power and Constitutional Form (OUP 2007) 129.
- Gegenava, D, ‘Unconstitutional Constitutional Amendment: Three Judgments from the Practice of the Constitutional Court of Georgia’ (2014) 5 South Caucasus Law Journal 396.
- Gözler, K, Le pouvoir de révision constitutionnelle (Presses universitaires du Septentrion 1997).
- Griffin, SM, ‘Constituent Power and Constitutional Change in American Constitutionalism’ in Loughlin, M, and Walker, N, (eds), The Paradox of Constitutionalism: Constituent Power and Constitutional Form (OUP 2007) 49.
- Hamilton, A, ‘Idea of Restraining the Legislative Authority in Regard to the Common Defense Considered (continued)’ (26 December 1787) Federalist No. 28, available at http://www.constitution.org/fed/federa28.htm (27 December 2017).
- Landau, D, ‘Honduras: Term Limits Drama 2.0—How the Supreme Court Declared the Constitution Unconstitutional’ (27 May 2015) Constitutionnet, available at http://www.constitutionnet.org/news/honduras-term-limits-drama-20-how-supreme-court-declared-constitution-unconstitutional (27 December 2017).
- Landemore, H, ‘Inclusive Constitution-Making: The Icelandic Experiment’ (2015) 23 The Journal of Political Philosophy 166.
- Le Divellec, A, Levade, A, and Pimentel, CM, ‘Le contrôle de constitutionnalité des lois constitutionnelles : Avant-propos’ (2010) 27 Cahiers du Conseil constitutionnel, available at http://www.conseil-constitutionnel.fr/conseil-constitutionnel/francais/nouveaux-cahiers-du-conseil/cahier-n-27/cahiers-du-conseil-constitutionnel-n-27.42524.html (27 December 2017).
- Locke, J, Second Treatise of Government (1689; Hackett Publishing Company 1980).
- Loughlin, M, ‘The Concept of Constituent Power’ (2014) 13 European Journal of Political Theory 218.
- Loughlin, M, The Idea of Public Law (OUP 2004).
- Möllers, C, ‘‘We Are (Afraid of) the People’: Constituent Power in German Constitutionalism’ in Loughlin, M, and Walker, N, (eds), The Paradox of Constitutionalism: Constituent Power and Constitutional Form (OUP 2007) 87.
- Negri, A, Insurgencies: Constituent Power and the Modern State (University of Minnesota Press 1999).
- Neuborne, B, ‘The Supreme Court of India’ (2003) 1 ICON 476.
- Paine, T, Rights of Man (1791) available at http://www.ushistory.org/Paine/rights/ (27 December 2017).
- Patberg, M, ‘Agonistic Democracy: Constituent Power in the Era of Globalisation’ Review (2015) 14 Contemporary Political Theory 8.
- Pfersmann, O, ‘Unconstitutional Constitutional Amendments: A Normativist Approach’ (2012) 67 Zeitschrift für öffentliches Recht 81.
- Preuss, U, ‘The Exercise of Constituent Power in Central and Eastern Europe’ in Loughlin, M, and Walker, N, (eds), The Paradox of Constitutionalism: Constituent Power and Constitutional Form (OUP 2007) 211.
- Preuss, U, ‘The Implications of Eternity Clauses: The German Experience’ (2011) 44 IsLR 429.
- Roznai, Y, ‘Unconstitutional Constitutional Amendments—The Migration and Success of a Constitutional Idea’ (2013) 61 AmJCompL 657.
- Roznai, Y, Unconstitutional Constitutional Amendments: The Limits of Amendment Powers (OUP 2017).
- Roznai, Y, and Serkan, Y, ‘An Unconstitutional Constitutional Amendment—The Turkish perspective: A Comment on the Turkish Constitutional Court Headscarf Decision’ (2012) 10 ICON 175.
- Sagüés, N, ‘Notas sobre el poder constituyente irregular’ (2009) Año XV Anuario de derecho constitucional latinoamericano 151.
- Wenman, M, Agonistic Democracy: Constituent Power in the Era of Globalisation (CUP 2013).
- Citizens of Georgia Irma Inashvili, Davit Tarkhan-Mouravi and Ioseb Manjavidze v Parliament of Georgia (5 February 2013) N1/1/549 (Geor).
- Decision ‘La Roche’ Supreme Court of Venezuela (19 January 1999) (Venez).
- Decision 2003–469 DC Constitutional Council of France (26 March 2003) (Fr).
- Decision 45/2012 Constitutional Court of Hungary (28 December 2012) XII. 29 (Hung).
- Decision 466/91 ADIMC (3 April 1991) (Braz).
- Decision 62–20 DC Constitutional Council of France (6 November 1962) (Fr).
- Decision 92–312 DC Constitutional Council of France (2 September 1992) (Fr).
- Decision C-141/10 Constitutional Court of Colombia (26 February 2010) (Colom).
- Decision F-165 Supreme Court of Honduras (22 April 2015) (Hond).
- Decision Pl. ÚS 27/09 Constitutional Court of the Czech Republic (10 September 2009) (Czech).
- Headscarf Decision E. 2008/16, K. 2008/116 Constitutional Court of Turkey (5 June 2008) Official Gazette 22 October 2008-27032 (Turk).
- Judgment 103/2008 Constitutional Court of Spain (11 September 2008) ECLI:ES:TC:2008:103 (Spain).
- Kesavananda Bharati v State of Kerala (1973) 4 SCC 225 (India).
- Lissaboner Vertrag 2 BvE 2/08 (30 June 2009) BVerfGE 123, 267 ECLI:DE:BVerfG:2009:es20090630.2bve000208 (Ger).
- Minerva Mills v Union of India (31 July 1980) AIR SC 1789 (India).
- Premier of Kwazulu-Natal and Others v President of the Republic of South Africa and Others (CCT36/95)  ZACC 10 (S Afr).
- Reference Re Secession of Quebec  2 SCR 217 (Can).
- Südweststaat 2 BvG 1/51 (23 October 1951) BVerfGE 1, 14 (Ger).
- United Mizrahi Bank v Migdal Cooperative Village CA 6821/93 (9 November 1995) 49(4) PD 221 (Isr).
- VfSlg 16.327/2001 (11 October 2001) (Austria).