I. Introduction
Could it be that constitutional disenchantment at the European level is a corollary of constitutional disenchantment at the national level, and that this national constitutional disenchantment itself emerges precisely in the context of treaty ratification, when that process circumvents national constituent power? Although the entire European project is majestically oriented towards the “ever closer union of the peoples of Europe”,Footnote 1 the place that the peoples in whom the Preamble delights hold in the political and constitutional architecture of the Union is still quite unsettled. A myriad of initiatives designed to vivify Europe's citizenry, including highlights as diverse as direct elections to the European Parliament, the effort to establish a European constitution, and the introduction of the European Citizens’ Initiative, have all fallen flat. Turnout in European Parliament elections is consistently low and getting lower,Footnote 2 popular referenda on the European Constitution resulted in spectacular defeats in France and the Netherlands, and the European Citizen's Initiative, promoted as the great herald of participatory democracy, has, in the six years since the entry into force of the Treaty of Lisbon, produced only three “successful” initiatives, none of which has resulted in legislation.Footnote 3 In such a context, it cannot be surprising that although scholarly interest in constituent power is strong and growing,Footnote 4 it is no longer as enthusiastic as it once was about the prospect of identifying a European demos that could be mobilised as a European pouvoir constituent.Footnote 5 Indeed, for some, the whole notion of constituent power should be considered simply as a relic of the Westphalian past which is unsuited for an age in which government is not the exclusive preserve of the nation state.Footnote 6 Taking the opposite perspective, Loughlin argues that “In a world in which government is both ubiquitous and increasingly remote from ordinary people, a concept of constituent power that conjoins right and interest – the symbolic representation of all with the concerns of the many – must not disappear from political thought”.Footnote 7
Sharing Loughlin's concern, this article is keen to investigate the resilience of constituent power in the European context, but not in the usual way. Typically, assessments of the existence or non-existence of a European constituent power and the need therefor begin and end with consideration of the link between European citizens, on the one hand, and the European rules and rulers which govern them, on the other hand. Taking a contrary position, this article proposes that any serious assessment of the role of the peoples of Europe contains a blind spot if it fails to earnestly consider the mediating import of national constituent power. Since participation in the EU is still mediated by and through national constitutional law and institutions, constitutional disenchantment at the national level will conduce to constitutional disenchantment at the European level, resulting in a freezing effect on the prospects for developing an ever closer union between Europe's citizens. Consequently, whilst this article is ultimately concerned with the increasing remoteness of the peoples of Europe, its particular focus will be on the resilience of national constituent power.
This resilience must, of course, be measured precisely in the circumstances of increasing European integration. It is trite to say that the concomitant transfer of competences from the national level to the European level is necessarily accompanied by an erosion of the competences and powers of the nation state such that, although a national constitution remains standing, we can no longer think of it as it once was meant to be – a foundational and comprehensive social contract that exclusively governs the relations between citizens and establishes the institutions of state, allocating all legislative, executive, and judicial authority exclusively between them. What is less clear is the impact that integration has on national constituent power. Can we assume that national constituent power remains secure of due recognition and respect even amidst the explicit, implicit, and sometimes quite considerable changes wrought national constitutions in order to accommodate deepening integration? Or is there a risk that the constituent power, even when it has been exercised and demonstrated in the constitution of a social contract that governs all of the people in the name of all of them, becomes compromised by these subsequent alterations? The point of departure here is the hypothesis that we cannot blithely assume that national constituent power is not appreciably affected by European integration.
In order to investigate the truth of this hypothesis, a particular methodology must be adopted. The methodology offered here is one which has its roots in a particular insight of the French constitutional theorists who developed the concept and theory of constituent power, and who emphasised a close link between constituent power and limits on the power of amendment. The key insight is that, since the Constitution prescribes the procedure that must be used to amend the Constitution, including limits on the power of amendment, failure to respect the limits on constitutional amendment not only compromises the values enshrined by those limits, but also constitutes a failure to respect the constituent power which adopted those provisions. Contrariwise, amendments achieved through the legitimate use of the amendment procedure and respecting limits on the power of amendment not only honour the constitutional content protected by the limits, but also the original constituent power which adopted them. Since contemporary literature tends to consider limits on the power of amendment in terms of the substantive content that they protect, rather than the constituent power that they defend, the argument here cuts across the grain in this respect also. To be clear, then, the broad scope of this article is a reflection on European constituent power and the aspiration of ever closer union between the peoples of Europe; its specific claim is that European constituent power will necessarily be enfeebled if national constituent power is weak; and the methodology used to investigate this claim is based on an insight recovered from the theories of the French constitutional scholars who argue that respect for constituent power is intimately connected to respect for the limits on the power of amendment. The specific kind of “remoteness” with which this article is concerned is that which results when amendments of national constitutions are achieved in contravention of the limits on the power of amendment, thus alienating the national constituent power and, in turn, having a debilitating effect on the prospects for ever closer union between the peoples of Europe.
Section II reviews the link between constituent power and constitutional amendability as established in French constitutional theory, in the context of a distinction between two kinds of limits on the power of amendment: enumerated limits (Section II(A)) and essential limits (Section II(B)). The purpose of this section is to re-sensitise contemporary literature on constitutional amendment to the insight of the French theorists that these limits exist not only – and perhaps not even primarily – to protect specific material content but to defend the people in their original exercise of constituent power. Section III is a doctrinal analysis of how enumerated and essential limits have been judicially interpreted in the Member States. Since each Member State has its own Constitution with all its own foibles, and since the precise procedure for constitutional amendabilityFootnote 8 is determined after consideration of the idiosyncrasies of the polity in question,Footnote 9 the investigation can only proceed on a case-by-case basis. Section III(A) examines how enumerated limits on the power of amendment have been interpreted in Germany and Spain in the context of judicial decisions challenging the ratification of European treaties, whereas Section II(B) proceeds with a discussion of how the Irish and French constitutional courts have dealt with the idea of essential limits on the power of amendment in similar cases. Each of these jurisdictions has made a unique and noteworthy contribution to this question. In France, the home of the concept of pouvoir constituant, the Conseil constitutionnel has ruled several times on the need for amendments to the French Constitution in order to facilitate ratification of European treaties. Ireland is also an obvious choice because of the (in)famous Crotty decision, which has led to popular referenda in the context of each treaty ratification since 1987, and therefore to the popular rejections of the Nice and Lisbon Treaties, which were reversed by second referenda. Spain is included because of the unique circumstances in which the Tribunal Constitucional ruled on the compatibility of the European Constitutional Treaty and the Spanish Constitution, the Spanish people having voted in favour of this constitutional bigamy. Finally, the German Bundesverfassungsgericht has been an outspoken defender of the right of the German people to self-determination, even in the context of European integration, and links that right of self-determination to the limits placed on the power of amendment.
II. Pouvoir Constituant and Limits on the power of Amendment
A. Enumerated Limits as Defenders of the Pouvoir Constituant
The term pouvoir constituant surfaces amidst an emerging constitutional theory which exalted the role of the people as the creative force behind the constitutional order and therefore the foundation upon which this comprehensive social contract was to rest. Although Baron de Montesquieu (1689–1755) did not himself use the term pouvoir constituant, the theory of separation of powers that he did propound is built on his foundational premise that the people must be allowed to organise their living-in-common, and that delegation of authority to instituted powers is legitimate only because it supports, rather than supplants, their ultimate authority.Footnote 10 Raymond Carré de Malberg (1861–1935), who wrote quite extensively himself on the subject, attributed to Montesquieu the origins of the concept of pouvoir constituant because, he believed, the theory of separation of powers fundamentally implied the necessity of a theory of pouvoir constituant, since “to logically explain such a division of powers, it was necessary to arrive at the idea of a primordial and superior power … the theory of the separation of powers paves the way for the theory of constituent power”.Footnote 11 The force of logic in this reading of Montesquieu is compelling: a separation of powers that are themselves created powers cannot be achieved except by the exercise of an uncreated and creative power; hence, there must be a pouvoir constituant. Nonetheless, since Montesquieu did not himself coin the term pouvoir constituant, we must turn attention to the one who did, Emmanuel Joseph Sieyès (1748–1836). In a paper presented to the Constitution Committee in July 1789, Sieyès defended the position that all of the various created powers established by the Constitution “proceed from the general will, from the people, which is to say, from the nation”, and thus concluded, with great pith, that “above all, a constitution presupposes a pouvoir constituant”.Footnote 12 Following Montesquieu, the distinction between the legally unlimited and illimitable power of the nation to adopt a constitution, namely the pouvoir constituant, and the legally limitable and limited powers of the persons and bodies vested with power by the Constitution is one that Sieyès is keen to emphasise. Thus, he considered the two types of powers – constituent power and constituted powers – to be qualitatively different, the latter being inferior to the former and therefore necessarily limited.
Writing more than a century later, Carré de Malberg's Contribution à la théorie générale de l’État takes a different view, since his theory contemplates that the power to establish a constitution and the power to amend a constitution are two modes of pouvoir constituant. The first, le pouvoir constituant dans les circonstances révolutionnaires, is operative at the moment of the creation of a new constitutional state, while the second, le pouvoir constituant dans les circonstances d'une Constitution en vigueur, is operative during the lifetime of that state. Recognising that the creation of the state cannot be accomplished according to existing legal rules since the moment of creation precedes the establishment of the rules, Carré de Malberg insists that “the formation of a state is not constrained by an existing legal order: it is the condition for law, not conditioned by the law”.Footnote 13 The revolutionary circumstances ensure that the new Constitution is not adopted according to any existing laws or procedures of its precursor, of which nothing now remains. Indeed, the constitutional moment is governed not by law at all, but only by the brute fact of force,Footnote 14 as the pouvoir constituent dans les circonstances révolutionnaires is simply grabbed by the strongest pair of hands.Footnote 15
Conversely, according to Carré de Malberg, when the constitutional state comes into being, le pouvoir constituant dans les circonstances révolutionnaires that had existed outside of the domain of law is then brought within the purview of the constitutional state (“étatisé”). Henceforth, the nation shall express its will in accordance with the mechanisms and rules laid down by the Constitution, with the result that what was formerly done in a legal vacuum will now be entirely governed by law; it is the “essence” of the state that it should possess a legal order, which legal order regulates even the “repair” (“la refection”) of the constitutional state itself.Footnote 16 This power to amend the Constitution, circumscribed by the terms of the Constitution, is what Carré de Malberg terms le pouvoir constituant dans les circonstances d'une Constitution en vigueur. Importantly, this power may effect changes of any magnitude: it is possible, at one end of the scale, to tinker with the detail of the some particular provision of the Constitution or, at the other end of the scale, to repeal, and replace it in its entirety.Footnote 17 The sole requirement is that le pouvoir constituant dans les circonstances d'une Constitution en vigueur must follow the written rules laid down in the provision mandating constitutional amendment and conditioning its exercise. This key contribution of Carré de Malberg is that he supposes, countering Sieyès, that the constituent power and constitutional amendability procedure are fundamentally the same kind of power, the former being the raw and legally unlimited expression of the will of the people in the establishment of their Constitution and the latter being the domesticated version of that power which exists now within the Constitution with the same strength and effect but subject to limits which the former has placed on its exercise. On this view, the reason to respect the limits is that they are all that is left of le pouvoir constituant dans les circonstances révolutionnaires; everything else in the Constitution exists through the indulgence of le pouvoir constituant dans les circonstances d'une Constitution en vigueur.
These written limits on the power of amendment that must be respected as a matter of respect for the original constituent power fall into two classes. They might identify circumstances in which the amendment power cannot be exercised, such as limits prohibiting the exercise of the amendability procedure during a time of war or national emergency. Alternatively, they might identify articles of the Constitution which cannot be made the subject of an amendment, such as limits which proscribe amendments to the provisions establishing the form of government or protecting fundamental rights. It is this second class that tends to receive more attention in the literature. For example, Yaniv Roznai's comprehensive global study of “unamendable provisions” reveals a marked increase in the use of these provisions in national constitutions: of the constitutions enacted between 1789 and 1944, only 17% contained unamendable provisions, compared with 27% of those enacted between 1945 and 1988, and 53% of those enacted between 1989 and 2013.Footnote 18 Grouping these provisions into six different categories according to material content, he identifies unamendable provisions that shield the form and system of government, the political or governmental structure, fundamental ideology or identity, basic rights, integrity or sovereignty, and idiosyncratic constitutional provisions (e.g. immunities, peace agreements, or rules on nationality).Footnote 19 Judicial enforcement of these provisions is increasingly uncontroversial – Barak, for example, argues that a constitutional court should ensure that these provisions are respected in the same way that they ensure that all the other provisions of the Constitution are respectedFootnote 20 – and even lauded for its defence of constitutional values.Footnote 21
Such contemporary scholarship, sophisticated and valuable as it is in its identification of categories of limits and defence of their judicial enforceability, is nonetheless missing precisely the sensitivity to the work of the French constitutional theorists that this article seeks to re-awaken. The focus on material content described in the limits tends to lull us into an amnesia which disremembers the insight that enumerated limits on the power of amendment must be respected as a matter of respect for constituent power. This is especially evident in the scholarship which discusses the purposes of these provisions, without any reference to constituent power. Richard Albert, for example, who uses the term “entrenchment clauses” co-extensively with Roznai's “unamendable provisions”, argues that entrenchment may have three possible purposes.Footnote 22 It may be “preservative”, i.e., designed to protect certain features of the Constitution; it may be “transformational”, i.e., designed “to transform the state by helping to paint a portrait of the state not as it is, but as it could be”Footnote 23; or it may be “reconciliatory”, i.e., designed to “advance the cause of reconciliation between … previously warring factions which have joined together in peace to form a new or reconstructed state”, including amnesties and pardons.Footnote 24 Having broadly endorsed Albert's classification, Roznai adds that limits on the power of amendment sometimes function as “gag rules”, silencing public debate on contentious issues for fear that such debate would be dangerously divisive in the fledgling constitutional state.Footnote 25 In identifying such purposes, they assume that these provisions are only ordered towards the protection of certain kinds of constitutional content for the sake of certain kinds of content-related purposes. Indeed, the very fact that these provisions are referred to as “unamendable provisions” and “noli me tangere provisions” (Roznai), “entrenchment provisions” and “constitutional handcuffs” (Albert), and, generally, “eternity clauses” serves, terminologically, to reinforce the idea that what is most important about these provisions is the specific legal content which is protected, rather than the defence of constituent power which is entailed. Since the purpose of this section is to re-sensitise contemporary scholarship to the insights of the earlier French theorists, these terms must be eschewed. Rather, accommodating the insights from both strands of literature, the term enumerated limits on the power of amendment is proposed here as the descriptor of those written restrictions which limit the susceptibility of a constitution to amendment, in order to underscore that the power to amend is a derived and limited power which thereby contrasts with the original and unlimited constituent power. A distinction may be drawn between enumerated procedural limits, which proscribe amendment in particular circumstances such as a state of emergency, and enumerated substantive limits, which foreclose the use of the amendability procedure in relation to specific constitutional provisions. “Enumerated substantive limits” accommodates both the material focus of recent scholarship and the teleological focus of the earlier theories, making it less likely that Carré de Malberg's insight – that written limits on the power of amendment are the permanent sign of the existence of the constituent power – will be obscured.
B. Essential Limits as Defenders of the Pouvoir Constituant
Once enumerated limits on the power of amendment have been respected, Carré de Malberg argues that anything and everything in the Constitution can be amended. That position is not endorsed either by Sieyès or by Olivier Beaud, who argue that the defence of constituent power requires respect not only for enumerated limits, but also for an essential limit on the power of amendment. In his celebrated pamphlet, Qu'est-ce que le Tiers État?, Sieyès emphasises that “the constitution is not the work of a constituted power, but of the pouvoir constituant”.Footnote 26 Since only the natural law exists before or above the exercise of constituent power by the nation, such exercise of constituent power is unimpeachable (“toujours légale”) as a matter of positive law.Footnote 27 Beyond the descriptive claim that constituent power is not subject to the Constitution, Sieyès defends two further claims. He holds both that the nation cannot be subject to the Constitution (because the nation cannot alienate or deny its constituent power)Footnote 28 and that the nation should not be subject to the Constitution, since constituent power is the source and supreme authority of all positive law.Footnote 29 Taking Sieyès's vision of this permanently inalienable and constitutionally illimitable pouvoir constituent as his starting point, Olivier Beaud (1958–) objects to Carré de Malberg's theory that the power to establish a constitution and the power to amend a constitution are two kinds of pouvoir constituant.
Beaud's argument is that, if, as Sieyès contends (and Carré de Malberg concurs), the power to establish a constitution is legally illimitable, then this power is profoundly non-equivalent to the power to amend the Constitution according to the legal rules established therein. He therefore refutes Carré de Malberg's nomenclature of equivalence on the basis that “one cannot simultaneously maintain the legally unconditioned character of the pouvoir constituant originaire, the legally authorised character of the derived power, and categorise them both as pouvoir constituant as if they had an identical nature”.Footnote 30 Instead, we should recognise that the two powers are fundamentally distinct and opposed to each other.Footnote 31 Having different natures because they have different purposes, he insists that they should bear a nomenclature that makes this incommensurability apparent, proposing the terms acte constituant and acte de révision.Footnote 32 Moreover, Beaud, relying broadly on a Kelsenian understanding of the hierarchy of norms,Footnote 33 emphasises that the acte de révision is inferior to the acte constituant (“infraconstituante”).Footnote 34 More than that, since a constitutional amendment only has legal validity because it is materially founded on the Constitution,Footnote 35 the power to amend the Constitution relies on the continuing strength of the acte constituant. For Beaud, therefore, it is not true to say that anything and everything in the Constitution can be amended. Constitutional amendment cannot logically repeal the Constitution because the acte constituant itself cannot be legally repealed (abrogé)Footnote 36 but only illegally abolished (aboli)Footnote 37; moreover, constitutional amendment cannot logically abolish the Constitution because the only power capable of overcoming an acte constituant will be a new acte constituant.Footnote 38 On this vision, an acte de révision, in order to be valid, must conform to the technical conditions laid down in the Constitution (i.e. the enumerated limits) but it must also respect an unwritten limit, by acknowledging its own essential nature as a power that is subordinate to the acte constituant.Footnote 39 This unwritten limit on the power of amendment supplements and complements any enumerated limits because it excludes the precarious possibility that a constitution could be replaced without the exercise of constituent powerFootnote 40; constituent power is thus protected by the insistence that constitutional amendments can only be valid if they acknowledge the subordinate quality of the power to amend vis-à-vis the superordinate quality of the constituent power.
In the same way that contemporary scholarship tends to appreciate enumerated limits on the power of amendment in terms of the material legal content they immunise from amendment, contemporary discussion of unwritten or implicit limits on the power of amendment both by judges and academics also focuses on the legal content which is thereby shielded. For example, the “basic structures doctrine” which was initially articulated in India in the case of Kesavananda Bharati v State of Kerala Footnote 41 and has since been adopted in various countries throughout the worldFootnote 42 is ordered towards the protection of the “framework”, “foundation or … basic institutional pattern”Footnote 43 of the Constitution. Similarly, the increasing preoccupation of certain European constitutional courts with “constitutional identity” seems to mimic the basic structure doctrine insofar as it protects fundamental features of the material content of their national constitutions.Footnote 44 Having endorsed the idea that there are implicit limits on the power of amendment, particular judges then list provisions which they take to be immune from constitutional amendment, which lists comprise, inter alia, democracy, the existence of Parliament, the secular character of the state, freedom of religion,Footnote 45 liberty of thought, expression, belief, faith and worship, equality, the socialist character of the state, the amendment provision itself,Footnote 46 the requirement of periodic elections,Footnote 47 the requirement that parliamentarians be elected, the separation of powers, the protection of fundamental rights,Footnote 48 and sovereignty.Footnote 49 Judicial review of essential limits on the power of amendment is more controversial for lacking a textual basis,Footnote 50 although it is becoming more and more commonplace,Footnote 51 justifiable precisely in terms of the content that is safeguarded. After all, if enumerated limits are judicially enforceable for the protection of certain content, and implicit limits protect those provisions that the judges deem to be the most fundamental, then implicit limits should also be judicially enforceable in order to defend (what the judges consider to be) the most important aspects of constitutional content. Throughout, the focus of contemporary engagement is once again on the specific content that these judicially created unwritten limits protect.Footnote 52
As a result, the insight present in the work of Beaud, relying on Sieyès, that an essential limit signals the inherently subordinate nature of the power of constitutional amendment, whose very existence serves as a reminder of the superior nature of the constituent power which must be defended, is also disremembered. The full extent of this amnesia can be understood through consideration of a recent turn in this scholarship: some scholars now include transnational and supranational law within the category of limits on the power of amendment that are not specifically written in national constitutions. Roznai, for example, argues that “supranational law seems to serve as an autonomous limitation to constitutional amendments”.Footnote 53 Rosalind Dixon arrives at the same conclusion that transnational norms become limits on the power of amendment in the national context.Footnote 54 Their conclusions are simply the logical culmination of the preoccupation with content – the intimate connection between constituent power and unwritten limits on the power of amendment has been systematically overlooked for so long that unwritten limits on the power of amendment are now being used to undermine, rather than to defend, national constituent power. In other words, they are used to precisely the opposite effect that Sieyès and Beaud had in mind! Here, too, new nomenclature is required if we are to recover Beaud's insight, since the terms “basic structures doctrine” and “constitutional identity” signal clearly the exclusive focus on content that is implied by these courts. The term essential limits on the power of amendment is proposed here, in order to accommodate the insights of both the modern judges and theorists who focus on material content and those of Sieyès and Beaud who remind us of the unwritten limit on the power of amendment which operates to defend constituent power.
The purpose of this section has been to re-sensitise our thinking on the topic of constitutional amendments to the insight of the French theorists of constituent power: limits on the power of amendment exist not only to defend specific constitutional content, but also to defend the constituent power. As far as Carré de Malberg is concerned, the constituent power is defended through the recognition of enumerated limits which testify to the continuing existence of the constituent power, whereas, for Sieyès and Beaud, constituent power is also defended through the recognition of an essential limit on the power of amendment which acknowledges the superordinate quality of the constituent power. Failure to respect enumerated or essential limits on the power of amendment undermines constituent power. While modern scholarship, as a result of its preoccupation with constitutional content, has a tendency to disremember this insight, the terminology of enumerated substantive limits and essential limits is defended here precisely for its capacity to accommodate both strands of literature, reuniting the concern for specific constitutional content with the concern for constituent power. Recovery of this insight makes it possible to investigate the resilience of national constituent power through an examination of how national courts interpret enumerated and essential limits on the power of amendment in the context of the ratification of European treaties. For now that we have become re-sensitised to the link between amendability limits and constituent power, we can recognise the limits of an approach that, cashing in the Solange legacy, simply assesses whether the treaties protect the same content as national constitutions, concluding that nothing is lost as long as there is no deficit of material constitutional content. Such an approach cannot measure remoteness because it cannot capture any losses that are endured by the constituent power.
III. Remoteness Based on Transgression of Enumerated and Essential Limits
The distance between the Constitution as established by the constituent power (and amended without transgression of enumerated and essential limits) and the Constitution as altered by amendments achieved in violation of the enumerated and substantive limits on the power of amendment is, for the purposes of this article, the measure of “remoteness” of the people from their Constitution. As we examine judicial decisions which deal with enumerated and essential limits on the power of amendment in the context of the ratification of European treaties, the principal question is: do constitutional courts examining these treaties by reference to enumerated and essential limits on the power of amendment do so with a consciousness that national constituent power is at stake? Before probing their decisions, however, two particularities of EU-inspired amendments must be mentioned. First, most of the time, these amendments are implied. For example, a national constitution vests exclusive legislative power in the national parliaments but a European treaty which transfers legislative power in certain policy fields to European institutions is ratified without the adoption of an amendment stating that henceforth Parliament has neither exclusive nor superordinate law-making power in those specific fields. Instead, the doctrine of implied amendment, whereby constitutional provisions are implicitly amended by the treaties, but the exact terms of the amendment are unspecified and uncertain,Footnote 55 has become commonplace.Footnote 56 Second, even a significant re-allocation of national powers or adjustment in fundamental rights protection leaves untouched the basic principle that the Constitution is the touchstone for the interpretation and validity of all laws in the legal order, whereas, when an amendment is adopted to mandate ratification of a treaty that gives powers to supranational institutions, it is certain that the Constitution will not be the touchstone for the interpretation or validity of those powers.
A. Enumerated Limits and National Constituent Power in Germany and Spain
In Germany and Spain, enumerated substantive limits on the power of amendment have been relied on in cases challenging the capacity of the national legal order to accept European treaties, on the grounds that those treaties transgress the terms of those limits. The question therefore is whether the respective constitutional courts have interpreted those enumerated substantive limits in such a way as to honour national constituent power in Germany and Spain.
Article 79(2) of the German Basic Law provides that amendments thereto may be passed by a two-thirds majority of the Bundestag and a two-thirds majority of the Bundesrat. Three enumerated substantive limits on the power of amendment are articulated in Article 79(3), detailing that there can be no amendments to Article 1 (human dignity and inalienable rights), Article 20 (democracy), or the federal structure. At the time of the ratification of the Maastricht Treaty in 1992, several constitutional provisions were amended, including Article 23, which was revised to establish that Germany's participation in the EU should respect the principle of subsidiarity and the position of the Länder to mandate German participation in European integration on the basis that the EU “is committed to democratic, social and federal principles, to the rule of law and to the principle of subsidiarity and … guarantees a level of protection of basic rights essentially comparable to that afforded by this Basic Law”, and to establish that sovereign German powers could be transferred to the European institutions, but always subject to Articles 79(2) and 79(3).Footnote 57 In decisions concerning the ratification of the Maastricht and Lisbon treaties, the German judges seem to concur with the French theorists that there are important connections between constituent power and enumerated limits on the power of amendment.
In the case of Brunner and others v The European Union Treaty,Footnote 58 the complainants alleged a breach of their rights to democratic representation under Article 38 on the basis that the Maastricht Treaty transferred essential competences of the German Parliament to the European institutions even though the latter lacked democratic legitimacy.Footnote 59 Finding against the complainants, the reasoning of the Bundesverfassungsgericht centres on the fact that the Constitution, through Article 79(3), proscribes amendments to the democratic guarantees made in Article 20. The decision unequivocally recognises a connection between constitutional amendability and constituent power, holding that “Article 79(3) binds the development of the German state to the essential content of the constitutional order specified therein, and seeks thus to fortify the Constitution in force against a development aimed at a new order, without being able itself to lay down rules binding on the constitution-making power”.Footnote 60 Although this seems to imply that the constituent power will be robustly defended through a strict application of the enumerated substantive limits contained in Article 79(3), in fact, the Court concluded that the rights to democratic representation guaranteed by Article 38 were not infringed without engaging in a rigorous review.Footnote 61 The Court did seem to accept that Article 23 established limits on Germany's future participation in the European project, holding that “the development of the European Union … is subject to the principles of democracy and the rule of law, social and federal principles and the subsidiarity principle”.Footnote 62 Still, the decision is disappointing in that the Bundesverfassungsgericht did not use Article 79(3) to strictly review the legality of the amendment to Article 23, but rather to limit the interpretive potential of Article 23 as amended.
In 2008, a similar case challenging the ratification of the Lisbon TreatyFootnote 63 on the basis of alleged breaches of Article 38 and the principle of sovereign statehoodFootnote 64 came before the Bundesverfassungsgericht. Again, the Bundesverfassungsgericht acknowledged an intrinsic connection between the original exercise of constituent power by the German people and Article 79(3) which immunises Article 20 from constitutional amendmentFootnote 65: “… the violation of the constitutional identity codified in Article 79(3) of the Basic Law is at the same time an encroachment upon the constituent power of the people.”Footnote 66 This seems, at first glance, to be a clear echo of Carré de Malberg's insight that to fail to respect enumerated limits is to fail to respect constituent power. Note, however, that the Bundesverfassungsgericht refers to the content of Article 79(3) as the inalienable “constitutional identity” of the German Constitution,Footnote 67 thereby eliding the distinction between an enumerated substantive limit and an essential limit on the power of amendment. Since Article 79(3) first operates in defence of constitutional identity, it offers much interpretive potential to the Bundesverfassungsgericht, which the Court duly avails itself of, holding that “the principle of democracy … sets content-related limits to the transfer of sovereign powers”.Footnote 68 These judicially discovered limits were given in a non-exhaustive list comprising revenue law, criminal law and punition, language law, family law, the provision of education, decisions on the use of force, and the protection of freedom of conscience and belief.Footnote 69 Not only does the Bundesverfassungsgericht consistently emasculate its own limits, by failing to apply them in practice,Footnote 70 but this recasting of Article 79(3) as protector of amorphous constitutional identity weakens the link between Article 79(3) and constituent power because it consumes the Court's attention with considerations of material content. In sum, then, the decisions of the Bundesverfassungsgericht are commendable for their initial principled recognition of the connection between respect for enumerated substantive limits on the power of amendment and defence of the constituent power. However, since the Court re-commissions Article 79(3) as an interpretive limit on Article 23 as amended in the Brunner decision and recasts Article 79(3) as a defender of constitutional identity in the Lisbon decision, that principled recognition gives way to a preoccupation with content, which in turn means that Article 79(3) has not in fact been interpreted in such a way as to guard against remoteness.
The Spanish Constitution, meanwhile, adopts a tiered approach to constitutional amendment. Most provisions of the Constitution may be amended using Article 167, which requires that proposals to amend the Constitution must be approved by three-fifths of the deputies in each House of Parliament (Congreso de los Diputados and Senado), after which one-tenth of the deputies in either House may request that the proposal become the subject of a referendum.Footnote 71 However, an extraordinary amendment procedure, detailed under Article 168, is required whenever either “a total revision of the Constitution is proposed” or where there is a proposal affecting issues of sovereignty, nationhood, the rule of law, or the independence of political parties, trade unions, or the Armed Forces (Preliminary Title), fundamental rights and freedoms (chapter II, section 1 of Title I) or the position of the King (Title II). The extraordinary procedure requires that the amendment proposal be carried first by a two-thirds majority of both Houses of Parliament, then, after their dissolution and re-constitution following a general election, be carried a second time by a two-thirds majority in both Houses, before being submitted to the people for their approval in referendum. In making such amendments so politically treacherous albeit not actually proscribed, Article 168 functions, to all intents and purposes, as an enumerated substantive limit on the power of amendment.
In Declaration 1/1992, the Tribunal Constitucional ruled on the compatibility of Article G of the Treaty on the European Union (Maastricht Treaty) with Articles 1.2, 13.2, and 23 of the Spanish Constitution.Footnote 72 In its judgment, the Court upheld the principle that the doctrine of implied amendment has no place in Spanish law: although Article 93 allows that powers derived from the Constitution may be transferred to international institutions,Footnote 73 the Tribunal Constitucional emphasised that it could not be used to nullify or contradict the provisions of the Constitution, nor as a mechanism for “implicit or tacit reform”.Footnote 74 Nonetheless, the Court held that the Treaty was incompatible only with Article 13.2Footnote 75 and not Articles 1.2 or 23, despite the substantive overlap between Articles 13.2 and 23. Could it be that the Court was influenced by the fact that inconsistency with Articles 1.2 or 23 would have required the use of the extraordinary amendment procedure, whereas inconsistency with Article 13.2 could be remedied through the ordinary procedure? On 20 February 2005, Spain became the first Member State to secure popular support for the Constitutional Treaty in a consultative referendum, after which the Congreso de los Diputados and Senado voted in favour of ratification. The Government then requested the Tribunal Constitucional to issue a declaration on the compatibility of the Constitutional Treaty, specifically Article I-6 on the primacy of European law and Articles II-111 and II-112 on the Charter of Fundamental Rights, with the Spanish Constitution, and, in the event of incompatibility, to state whether the extraordinary procedure was required. Given that Declaration 1/1992 declared incompatibility and evinced a principled stance against implicit constitutional reform, the expectation must have been that the Tribunal Constitucional would once more declare incompatibility. Given that such incompatibility would concern sovereignty and human rights, it would be very difficult to avoid the need for the extraordinary procedure.
The decision on the Constitutional Treaty, Declaration 1/2004 was issued on 13 December 2004.Footnote 76 Rather than interpreting Article 93 consistently with its own Maastricht decision as a prohibition on implied amendment, the Tribunal Constitucional was keen to underscore instead that Article 93 has “a substantive or material dimension, which should not be disregarded”.Footnote 77 The implication is that “the Constitution requires that the Legal System accepted as a result of the transfer shall be compatible with its basic principles and values”.Footnote 78 These basic principles and values comprise the unenumerated “material limits” on the transfer of powers to the European institutions, which are “implicitly derived from the Constitution and its essential meaning” and which include “due respect for the sovereignty of the State, for our basic constitutional structures and the fundamental value system and principles enshrined in our Constitution, in which fundamental rights acquire their own legal status”.Footnote 79 It is impossible not to notice that these newly minted unwritten material limits are co-extensive with some of the enumerated substantive limits referred to in Article 168, although the Court did not note the similarity. In any case, the Court concluded after a discussion that was less than robust, that these essential limits were “scrupulously adhered to” in the Constitutional Treaty and therefore that no constitutional amendment was required.Footnote 80 It its decisions, the Tribunal Constitucional considers limits on the power of amendment only insofar as they affect constitutional content, and seems either not to be aware or not to accept the insight of the French theorists that limits on the power of amendment must be respected as a matter of respect for constituent power. Indeed, the concept of constituent power is almost completely overlooked throughout the judgment, with one unlikely exception. In the passage in which the Tribunal Constitucional discusses its own role as a court of last resort with the unlikely-to-be-exercised power to review the constitutionality of European law, it justifies that role by reference to “the interests of maintaining the sovereignty of the Spanish people and the supremacy of the Constitution adopted by the Spanish people”.Footnote 81 Thus, instead of enforcing limits on the power of amendment out of respect for the constituent power, the Tribunal Constitucional leans on the constituent power in order to prop up its own reviewing authority.
The courts in these two jurisdictions have dealt with enumerated substantive limits on the power of amendment in the context of challenges to the constitutionality of European treaties. At a theoretical level, the Tribunal Constitucional fails to make the connection between limits on the power of amendment and defence of constituent power, while the Bundesverfassungsgericht clearly echoes Carré de Malberg in recognising the principle that enumerated limits must be honoured as a matter of respect for constituent power. Nonetheless, when faced with the practical question of enforcing limits on the power of amendment, both courts become preoccupied with the material legal content to be protected – rather than the constituent power to be defended – by those limits, as if the provenance and pedigree of the legal rules were irrelevant! Furthermore, the conclusions reached are unconvincing because the courts fail to rigorously demonstrate (in the German case) that European law has not crept into the areas of revenue law, criminal law and punition, family law, etc. and (in the Spanish case) that the primacy clause and the Charter of Fundamental Rights would have no impact on Spanish sovereignty and human rights protection.Footnote 82 If the French theorists are right that failure to respect limits on the power of amendment is failure to respect constituent power, then national constituent power has been weakened in Germany and Spain as a result of a clear lack of regard for their enumerated substantive limits on the power of amendment.
B. Essential Limits and National Constituent Power in Ireland and France
In Ireland and France, challenges to the ratification of European treaties have been made in circumstances where the litigant does not have the luxury of relying on enumerated substantive limits on the power of amendment. The question therefore is whether the courts will acknowledge the existence of essential limits on the power of amendment and consider the possibility that a treaty transgresses those limits, as a matter of respect for the national constituent power.
Articles 46 and 47 of the Irish Constitution provide that “any provision of this Constitution may be amended … by way of variation, addition, or repeal”,Footnote 83 such amendment having been approved by a simple majority of both Houses of Parliament (the Dáil and the Seanad) and also by a simple majority of the people in referendum. There are no enumerated limits on the power of amendment, and so the only question is whether the Irish Supreme Court is prepared to recognise essential limits on the power of amendment.Footnote 84 Of all of the Member States, Ireland is the one that apparently best protects the role of the people in relation to the ratification of European treaties because of the 1987 case of Crotty v An Taoiseach.Footnote 85 In this case, the Supreme Court decided that the Government was entitled to ratify amendments to the treaties without holding a referendum “so long as such amendments do not alter the essential scope or objectives” of the treaties as they had been already approved by the people in a previous referendum,Footnote 86 but referenda would be required for those that exceeded that essential scope or those objectives. Since the Crotty case arose in the context of the Single European Act, the Supreme Court concluded that those aspects thereof which allowed for co-operation in foreign policy would require an amendment of the Constitution, and hence a referendum. In justifying the decision, the judges leaned heavily on Article 6 of the Constitution, which asserts that it is the right of the people “in final appeal, to decide all questions of national policy, according to the requirements of the common good”, interpreting the provision to mean that the Government “cannot, within the terms of the Constitution, agree to impose upon … the people the contemplated restrictions upon freedom of action”.Footnote 87 On the contrary, such an agreement would, the Court held, “require a recourse to the people”.Footnote 88 At first glance, of course, the Crotty judgment seems to be a strong endorsement of the power of the people, and indeed it does defend the role of the people in referenda, but does it defend them in their capacity as constituent power? Since the judgment did not countenance any limits on the power of amendment, since later authorities explicitly reject the proposition that such limits exist,Footnote 89 and since Crotty has been recently reaffirmed as a decision that supports the unconditioned power of the people to amend the Constitution,Footnote 90 the Supreme Court stands squarely behind the proposition that there are no essential limits on the power of amendment. Sieyès and Beaud, we recall, argue that constituent power is only properly defended when an essential limit is imposed on the power to amend the Constitution. Paradoxically, then, although the Supreme Court defends the position of the Irish people in their capacity to amend the Constitution, it also undermines the position of the Irish people in the capacity as constituent power because it refuses to countenance any limits whatsoever on the power of amendment. This is the deeper and darker truth which lurks between the lines when the Court proclaims that “In the last analysis, it is the people themselves who are the guardians of the Constitution”.Footnote 91 Perhaps the Court considered that, since the power to amend the Irish Constitution necessarily involves the people, there is no great need for a limit on the power of amendment, but that would be to conflate constituent power with the power to amend, supposing that, when the people vote in a referendum to change one provision of the Constitution, they consider themselves to be acting in their capacity as constituent power and once more giving themselves an entire constitution de novo. It would also be to suppose that the Irish people do actually have the option to decline to ratify a European treaty – a supposition that has not been borne out by experience to date! Moreover, this conclusion seems to be contradicted by Article 1: “The Irish nation … affirms its inalienable, indefeasible, and sovereign right to choose its own form of Government”, which envisages an inalienable kind of “fighting self-determination” and could have would have provided a solid textual basis upon which to establish an essential limit such as Beaud had in mind: the power of amendment cannot undermine the inalienable right of the Irish people to constitutional self-government.
Article 89 of the French Constitution regulates constitutional amendment, according to which the French President has discretion to submit an amendment proposal to the people in referendum once it has been approved by the Assemblée Nationale and the Sénat (Article 42). If he declines to do so, the proposal must be passed by a three-fifths majority of votes cast at a joint sitting of the two Houses. Article 89.5 is an enumerated substantive limit proscribing amendments to the republican form of government, but it has not been raised in relation to the ratification of European treaties. In 1992, the Conseil constitutionnel was asked three times to pronounce on the compatibility of the Treaty on the European Union with the French Constitution. In the first case of the trilogy, Maastricht I, Decision 92–308 DC of 9 April 1992,Footnote 92 the Conseil constitutionnel found that the French Constitution did not preclude France from participating in “a permanent international organisation enjoying legal personality and decision-making powers on the basis of transfers of powers”.Footnote 93 However, the Court held that a constitutional amendment would be required prior to ratification in the event that such international co-operation would involve provisions of Treaty law “conflicting with the Constitution or jeopardising the essential conditions for the exercise of national sovereignty”.Footnote 94 It is no surprise that the Court would seek to remedy specific incompatibilities between the Constitution and a treaty, and it is no real surprise that the proposed remedy should be constitutional amendment, which was passed (without a referendum) as the Constitutional Act 92–554 of 25 June 1992. But was something more contemplated by the notion that a treaty might compromise national sovereignty even if there were no specific incompatibilities and even after any incompatibilities had been remedied by constitutional amendment? Could it be that this was an essential limit on the capacity of the Constitution to ratify a treaty, and therefore an essential limit on the power of amendment?
Since Maastricht I did not resolve these questions, a second challenge, Maastricht II, Decision 92–312 DC of 2 September 1992,Footnote 95 was taken on the grounds, inter alia, that the Maastricht Treaty had jeopardised the essential conditions for the exercise of sovereignty. The applicants specifically asked the Conseil constitutionnel to specify how far it is possible to go with revisions of the Constitution to effect successive inroads into “the essential conditions for the exercise of sovereignty”Footnote 96; in effect, they were inviting the Court to articulate an essential limit on the power of amendment which would foreclose the possibility of a thorough routing of national sovereignty. The Conseil constitutionnel declined to do so, holding that it had jurisdiction only to consider questions of substantive incompatibility.Footnote 97 Had the Court stopped there, the gist of the two judgments would have been that, whilst the Conseil constitutionnel had flirted with the idea of recognising an essential limit on the power of amendment in principle, it ultimately decided that it lacked jurisdiction to consider such a question. But the Court went on to hold that “[T]he 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 [sic] of new provisions which derogate from a constitutional rule or principle”.Footnote 98 Quite plainly, in this passage, the Conseil constitutionnel conflates constituent power with the power to amend the Constitution and denies the principle that there are essential limits on the power of amendment. The final case, Maastricht III: Decision 92–313 DC of 23 September 1992,Footnote 99 was taken following the referendum in which the Maastricht Treaty was approved by 51% of voters. The Conseil constitutionnel once more denied its own jurisdiction, specifying this time that its power to review legislation could not be exercised in respect of a law adopted by the French people in referendum.Footnote 100
Not much has changed in the meantime. In 2004, the Conseil constitutionnel ruled on the compatibility of the European Constitutional Treaty with the French Constitution in Decision 2004–205 of 19 November 2004,Footnote 101 repeating that, when European treaties “contain a clause running counter to the Constitution, call into question constitutionally guaranteed rights and freedoms or adversely affect the fundamental conditions of the exercising of national sovereignty”, then a constitutional amendment is required prior to ratification.Footnote 102 Here again, specific incompatibilities between the Constitutional Treaty and the French Constitution were identified by the Court.Footnote 103 The resulting referendum was held and defeated on 29 May 2005, but the Constitutional Treaty was repackaged as the Lisbon Treaty on 13 December 2007 and presented for ratification. In Decision 2007–560 DC of 20 December 2007,Footnote 104 the Conseil constitutionnel ruled that the same material incompatibilities which had existed between the Constitutional Treaty and the French Constitution existed between the Lisbon Treaty and the French Constitution. This time, in the exercise of his discretion, President Sarkozy declined to hold a referendum to effect the necessary amendments.
The most striking feature of the decisions handed down by both the Irish Supreme Court and the Conseil constitutionnel is that these courts put the power to amend the Constitution on an equal footing with the unlimited and sovereign constituent power. This is clearly reminiscent of Carré de Malberg's theory insofar as he describes both powers as pouvoir constituant and differentiates them only on the basis that they are exercised in different circumstances.Footnote 105 The logical result of the failure to recognise that constituent power and the power to amend are incommensurable, non-equivalent powers is that both courts position themselves against Beaud and Sieyès, flatly denying that there are essential limits on the power of amendment despite its intrinsically subordinate position vis-à-vis constituent power. If Beaud and Sieyès are right that failure to recognise the essential limit on the power of amendment constitutes failure to respect constituent power, then national constituent power has been weakened in Ireland and France as a result of the refusal to acknowledge the existence of essential limits on the power of amendment.
IV. Conclusion
At first glance, it might seem counter-intuitive to suggest that the chasm between Europe and her citizens is partially caused by the weakening of constituent power at the national level. After all, nationalism is largely considered as an obstacle for the European project to overcome, if not a disease that the EU has been designed precisely to cure. But the argument of this paper is that, so long as participation in the EU is mediated through national constitutions, then the strength of “ever closer union” depends partly on the resilience of the national constituent power. Conversely, as national constituent power is undermined, remoteness between the people and their constitutions at the national level will necessarily conduce to ever closer remoteness among the peoples of Europe.
Having recovered an insight from French constitutional theory – that respect for national constituent power is closely related to respect for limits on the power of amendment – this article has measured remoteness in terms of the failure of national constitutional courts to enforce enumerated substantive limits and essential limits in the context of the ratification of European treaties. In Germany and Spain, where enumerated substantive limits exist in the respective constitutions, the courts have considered those limits only in terms of the content that they protect and not the constituent power that they defend and, without justifying their conclusions in a satisfactory manner, they have discovered no transgressions of those limits. In Ireland and France, meanwhile, the courts have consistently refused to recognise essential limits on the power of amendment, holding, to the contrary, that there is no real difference between the power to amend and constituent power. These failures to take enumerated substantive limits seriously much less to acknowledge the existence of essential limits demonstrate the lack of resilience of national constituent power and, since they distance the peoples of Europe from even their own constitutions, represent an impediment to ever closer union. The fact that these patterns of ever closer remoteness are iteratively precipitated by the need to ratify European treaties is one of the many ironies of European integration. Such general constitutional disenchantment can only be remedied, it is contended, by the effort to re-sensitise ourselves to the insight of the French theorists that enumerated and essential limits on the power of amendment are important, and that they exist not only for the sake of the constitutional content that they protect, but also for the sake of the constituent power that they defend.
