I. Weak judicial review

The main characteristic of weak judicial review is that, while courts are given the power to review the consistency of legislation with recognized rights, the legislature retains the final word on the validity of all laws. Although it is usually maintained that weak-form judicial review was invented in the late twentieth century, and that its first institutionalization is to be found in the Canadian Bill of Rights of 1960 (CBOR),^{Footnote 4} weak judicial review was the first form of judicial review of legislation that existed.^{Footnote 5} Section 2 of the CBOR established that ‘every law of Canada shall, unless it is expressly declared by an Act of the Parliament of Canada that it shall operate notwithstanding the Canadian Bill of Rights, be so construed and applied as not to abrogate, abridge or infringe … any of the rights and freedoms herein recognized and declared’.

Fifteen years before the CBOR was adopted, the Political Constitution of the Republic of Ecuador (1945) had already institutionalized a variant of the weak-form model. Article 160 of that constitution created a special tribunal, the Tribunal de Garantías Constitucionales, charged with the duty of ensuring the consistency of primary and secondary legislation with the provisions of the constitution.^{Footnote 6} Among its powers, the tribunal could (if requested by a final appeals court), temporarily suspend a law or regulation ‘until the Congress makes a decision on its validity’.^{Footnote 7} Prior to the adoption of this tribunal in Ecuador, several nineteenth-century Latin American constitutions also institutionalized a weak-form model of judicial review.^{Footnote 8} These (entrenched) constitutions usually established that legislation inconsistent with their provisions was to be considered invalid, but at the same time gave Congress the final word on any questions of constitutional interpretation.^{Footnote 9} The consequence of this approach, it has been stated, was that courts could declare the unconstitutionality of a law, but these declarations would only have any legal effect upon the action of Congress.^{Footnote 10}

Weak judicial review represents a movement away from ‘pure’ parliamentary sovereignty, but without giving judges the final word on the legality of rights-inconsistent legislation: courts can make, at most, declarations of inconsistency, but not declarations of invalidity (at least not of the type that cannot be overridden by ordinary legislative majorities). The main current examples of systems of weak judicial review are provided by Canada, New Zealand and the United Kingdom. In 1982, Canada replaced the approach of the CBOR with a system in which judges are given the power to strike down legislation, but where simple majorities at the federal and provincial legislatures are able to override those decisions^{Footnote 11} by following a requirement as to the form of legislation. That is, by making a declaration that they wish a particular act to apply ‘notwithstanding’ a relevant provision of the Canadian Charter of Rights and Freedoms (CORAF).

New Zealand followed suit with the adoption of the Bill of Rights Act in 1990 (BORA). Unlike its Canadian counterpart, which forms part of an entrenched constitution subject to a special amendment procedure, the BORA is a statutory bill of rights. Not surprisingly, it does not provide judges with a strike-down power (in fact, it explicitly denies them this power).^{Footnote 12} Instead, it creates a judicial duty to give legislation a rights-consistent interpretation.^{Footnote 13} If that is not possible, then judges have no choice but to apply the rights-inconsistent legislation (although courts seem to have asserted jurisdiction to make non-binding declarations of inconsistency).^{Footnote 14} In 1998, the Westminster Parliament adopted the Human Rights Act (HRA), which created in the United Kingdom a system of weak judicial review similar to that of New Zealand, but expressly attributing courts with the power of making non-binding ‘declarations of incompatibility’ between legislation and protected rights.^{Footnote 15} Once such a declaration is made, the relevant minister is allowed (but not required) to amend the legislation through a remedial order.^{Footnote 16}

II. Strong judicial review

As noted earlier, the main characteristic of a system of weak judicial review is that it does not give judges the power to strike down legislation or that, if it does, it allows an ordinary legislative majority to override the relevant judicial determination (without the need of formally amending the constitution). Naturally, the main characteristic of strong judicial review is the opposite: it does give judges the power to strike down legislation, and these judicial determinations may only be overridden by the legislature through the formal amending process (which would normally require that body to meet a qualified majority requirement and/or, in some cases, to seek the approval of the electorate in a referendum).^{Footnote 17} Strong judicial review is much more common than its weak counterpart: it forms an important part of the constitutional systems of the United States and most Latin American and European countries. Naturally, these countries are characterized by having rigid constitutions, that is, constitutions that are relatively difficult to amend when compared to ordinary legislation.

There are, however, important differences in the specific institutionalizations that strong judicial review has taken. The almost obligatory point of departure in this type of analysis is the United States, where, in the famous case of Marbury v Madison,^{Footnote 18} the US Supreme Court asserted its jurisdiction to strike down federal legislation inconsistent with the constitution (even though the constitutional text did not explicitly give it the power to do so).^{Footnote 19} Writing the opinion of the court and echoing an argument presented by Alexander Hamilton in the Federalist No 78, Justice Marshall maintained that ‘the Constitution is either a superior, paramount law, unchangeable by ordinary means, or it is on a level with ordinary legislative acts, and, like other acts, is alterable when the legislature shall please to alter it’.^{Footnote 20} Marshall agreed with the first proposition and concluded that ‘an act of the Legislature repugnant to the Constitution is void’ and it should be declared as such by the courts.^{Footnote 21} According to this view, the only way the legislature could surpass the limits established by a constitution (as interpreted by the courts) would be through constitution-amending legislation which, in the United States, involves a number of super-majority requirements at the federal and state level.

This is the basic premise shared by all systems of strong judicial review of legislation (with one exception, discussed below). It is, for example, the basic premise in those constitutional orders that, unlike the United States, have explicitly institutionalized systems of strong-form judicial review in their national constitutions. The Constitution of Spain of 1978, for instance, created a centralized^{Footnote 22} system of strong judicial review in which a special court (Tribunal Constitucional) was given the exclusive power to strike down legislation inconsistent with the constitution.^{Footnote 23} At the same time, the constitution creates a special amendment process that involves the legislature and the electorate, and which could be used to override the effects of a decision of the Tribunal Constitucional.^{Footnote 24} As with the US Supreme Court, decisions of this tribunal about the unconstitutionality of a law are binding on all branches of government.

In Europe and Latin America, as a result of the historical distrust of the judiciary in the civil law world (which had traditionally appeared as a conservative institution, attempting to block the effects of progressive legislation through interpretation), many countries attribute the power to strike down legislation to a special court (such as Spain’s Tribunal Constitucional), usually seen as separate from the judicial branch.^{Footnote 25} There are, in fact, many differences about the specific institutionalizations of strong judicial review: differences about the type of procedures that can trigger its exercise, differences about whether the court has jurisdiction to review the validity of laws in the abstract or only when their constitutionality is raised in the context of the facts of a specific case, or (as mentioned above) about whether the power to make declarations of invalidity is given to all judges or only to those sitting in special courts (there is also the possibility of a mixed system, best exemplified by the so-called Colombo-Venezuelan model).^{Footnote 26}

Despite these differences, one may say that what makes a model of judicial review ‘strong’ is that it gives judges the final word on the constitutionality and validity of ordinary legislation, and that those differences do not change this. That statement is accurate except that, as noted earlier, courts operating under a system of strong judicial review are not really given the final word about the validity of legislation, as the legislature normally has the ability to amend the constitution in order to render valid an otherwise unconstitutional act.^{Footnote 27} But there are exceptions. Some constitutions not only explicitly provide judges with the ability to strike down ordinary laws, but implicitly attribute them with the ability to strike down constitution-amending legislation inconsistent with certain constitutional provisions. The most famous example is that of Article 79(3) of the Basic Law of Germany (1949). This provision establishes an ‘eternity’ clause according to which amendments ‘affecting the division of the Federation into Länder, their participation on principle in the legislative process, or the principles laid down in Articles 1 and 20 shall be inadmissible’.^{Footnote 28} Clauses like this, which prohibit amendments contrary to particular clauses (or contrary to the principles expressed in certain provisions), have been present in different constitutions since early in the nineteenth century.^{Footnote 29}

However, in the nineteenth century, eternity clauses were usually found in constitutions that affirmed the principle of parliamentary sovereignty, established systems of weak judicial review, or in which any explicit limits to constitutional change were to be enforced politically. The best-known example is the Norwegian Constitution of 1814, which establishes explicit limits on the amending power, limits which have never been enforced by the judiciary.^{Footnote 30} What made the adoption of eternity clauses in the twentieth century different is that, by that time, most countries with written constitutions had adopted some version of strong judicial review. As a result, many courts quickly came to understand that their power to strike down ordinary laws also extended to constitution-amending legislation inconsistent with the eternity clause (even in the absence of a specific constitutional provision recognizing their power to review the constitutionality of constitutional amendments).^{Footnote 31} This amounted to the creation of what might appear as the strongest form of judicial review possible, giving judges the actual final word on the validity of any law. Nevertheless, judges in countries whose constitutions did not contain eternity clauses soon found themselves wanting of a mechanism that allowed them to protect the constitutional system as a whole. The doctrine of the basic structure allowed them to justify the judicial creation of such a mechanism. In the next part of the article, it will be argued that the adoption of the doctrine of the basic structure amounts to the development of a third model of judicial review.

III. Strong basic structure review

When a written constitution gives judges the power to strike down legislation and, at the same time, contains eternity clauses, strong judicial review becomes even stronger: constitution-amending legislation would not be enough to override all judicial declarations of invalidity. But just as a sovereign parliament could use a two step-process to avoid the limitations posed by a provision entrenching a statute, a legislature could adopt a constitutional amendment removing or altering the eternity clause, and then proceed to adopt the constitution-amending legislation that was, or is at risk, of being declared invalid.^{Footnote 32} To put it in the language of English constitutional theory, eternity clauses are usually not self-entrenched. Of course, it is not clear whether a court would enforce a constitutional amendment directed at removing an eternity clause from a constitution (and it is always possible that unwritten conventions would make such a removal unthinkable in practice).^{Footnote 33}

Nevertheless, the very possibility that an amendment removing an eternity clause could be accepted by the courts makes the ‘eternity clause approach’ a very strong version of strong-form judicial review rather than a new model of its own: qualified legislative majorities still retain the last word on the validity of ordinary (or constitution-amending) legislation, even if they must meet twice the stringent requirements of an amendment rule.^{Footnote 34} Some courts operating under constitutions that do not contain eternity clauses, however, have been able to assert a final jurisdiction to declare the invalidity of constitution-amending legislation. In so doing, they have given birth to a new model of judicial review which may be called strong basic structure review.^{Footnote 35} The adoption of such a model is connected in important ways to the nature of the amendment rule: in a relatively flexible constitution, courts are arguably more likely to develop the basic structure doctrine in order to protect the constitution from frequent fundamental constitutional changes adopted by the legislature (and will certainly have more opportunities to do so than in a system in which amendments are very rare as a result of a very difficult amendment process).^{Footnote 36}

The first court to adopt this approach was the Supreme Court of India. In Kesavananda Bharati v Kerala (1973),^{Footnote 37} a case which dealt with a series of land reforms that affected property rights, the court determined that while parliament had the power to amend any constitutional provision, it could not alter the basic structure of the constitution. For some of the judges, the constitution’s preamble and some constitutional clauses implicitly attributed a special importance to certain principles (among the principles mentioned in the different concurring opinions were those of constitutional supremacy, the republican form of government, federalism, the welfare state, individual liberty and secularism).^{Footnote 38} Other judges attempted to partially derive the doctrine of the basic structure from the text of the amendment rule itself. According to them Article 368 only gave parliament the power to amend the constitution, not to alter the very principles in which it is based, which would involve an act of constitution-making rather than a mere amendment.^{Footnote 39} In this sense, just as the adoption of strong judicial review in the USA, and the judicial assertion of jurisdiction to enforce eternity clauses in Germany, the adoption of the doctrine of the basic structure in India was based on a particular understanding of the constitutional system and not on a formal authorization found in the constitutional text.

The basic structure doctrine abolishes the last remnant of parliamentary sovereignty in systems that previously operated under a model of strong judicial review: it prevents a special parliamentary majority from amending the constitution in order to render ineffective a judicial declaration of invalidity (the legislature can always render ineffective a judicial invalidation of an ordinary law through a constitutional amendment, but the constitutional amendment itself may be invalidated). The doctrine was developed further in Minerva Mills v Union of India (1980), where the court expressed that the amending power could not be used to ‘destroy the identity’ of the existing constitution. In that case, the court struck down a constitutional amendment that explicitly attributed to parliament an unlimited amending power so as to prevent the future applicability of the doctrine.^{Footnote 40} In other words, the court made clear that it possessed the very final word on the validity of ordinary and constitution-amending legislation: under the basic structure doctrine, there is no existing legal process that can be used to give parliament the right to prevail over a judicial invalidation of a constitutional amendment.^{Footnote 41}

The Supreme Court of Belize, in a series of recent decisions, has adopted and applied the doctrine of the basic structure, openly relying on Indian jurisprudence. The first of these decisions is Bowen v Attorney General (2008).^{Footnote 42} In that case, the court declared the invalidity of the Sixth Constitutional Amendment Bill, which sought to exclude certain natural resources (including petroleum) from the protections of the right to property. The court expressed that the amendment, by impeding access to the courts in the context of alleged infringements of property rights, offended ‘the basic structure of the Constitution of Belize regarding the principle of the separation of powers and … the rule of law and the protections of fundamental rights especially those relating to the ownership and protection of property from arbitrary deprivation’.^{Footnote 43} More recently, in 2012, the same court struck down part of the Eight Amendment, which (as the amendment invalidated by the Supreme Court of India in Minerva Mills) declared that the amending power of the National Assembly was not subject to any other limitations (‘substantive or procedural’) than those contained in the constitution’s amendment rule.^{Footnote 44}

Attributing a special role to the constitution’s preamble (‘the root of the tree from which the provisions of the Constitution spring’),^{Footnote 45} the court expressed that ‘the basic structure doctrine holds that the fundamental principles of the Preamble of the Constitution have to be preserved for all times to come and that they cannot be amended out of existence’.^{Footnote 46} ‘The framers of the Preamble could not have intended’, the court continued, ‘that the National Assembly with the required majorities … could make literally any amendment to the Constitution to, for instance, abolish the judiciary, or expropriate private property without compensation, or imprison its enemies without trial.’^{Footnote 47} The doctrine of the basic structure assumes that the limits on constitutional change are absolute limits, that is, limits that no legally relevant entity can surpass. No constitution-amending legislation would be enough to reassert the legislature’s power to override a judicial declaration that a constitutional amendment is invalid, as any attempt to do so would be struck down by the courts.

Strong basic structure review, unlike strong judicial review, allows judges to identify themselves (even if they do it by reference to the constitution’s preamble, through an interpretation of the phrase ‘to amend’, or based on principles reflected by the constitution as a whole) the values that will act as a legally insurmountable limit to the legislature’s law-making power. Interestingly, if ever accepted in actual constitutional practice, common law constitutionalism (as currently proposed by some academics and judges in the United Kingdom), could be characterized as a version of strong basic structure review operating in the context of an unwritten constitution.^{Footnote 48} According to common law constitutionalism’s proponents, there is a ‘common law constitution’ that rests on certain principles (such as democracy, the rule of law, or the protection of basic human rights), and judges have the right to identify and enforce these principles against the decisions of parliament.^{Footnote 49} In short, common law constitutionalism, as the doctrine of the basic structure, would create a true system of constitutional (or judicial) supremacy, in which the permanency of the constitution is protected from the elected legislature.

IV. Weak basic structure review

It is true, of course, that most courts that have adopted the doctrine of the basic structure (such as that of India and Belize), implicitly or explicitly accept that the people (as opposed to the legislature acting through the formal amendment procedure), possess an unlimited constitution-making faculty that could be used to change elements of the constitution’s basic structure.^{Footnote 50} As Aaron Barak has stated, the doctrine ‘does not block off the people’s ability to change the basic structure of the constitution’, although in such a case a ‘completely different route must be chosen: the route of establishing a new constitution’.^{Footnote 51} Nevertheless, what makes those systems forms of ‘strong basic structure review’ (and not ‘weak basic structure review’, the model that will be discussed below) is that regardless of any implicit or explicit attribution of an unlimited constituent power to the people, they are based on a constitution that does not provide a formal mechanism for that unlimited power to manifest. In Barak’s words, they do not provide a ‘route’ for the establishment of a new constitution. In that sense, and absent a radical transformation of the constitutional order, courts that operate under the strong basic structure review model have the legal final word on the validity of a constitutional amendment. Moreover, even if the legislature attempts to create a mechanism (through ordinary or constitution-amending legislation) to channel the expression of the unlimited constitution-making power of the people in those situations in which an alteration of the basic structure is sought, there is always the possibility that the court would refuse to accept that the institutions called to express that power truly represent the people.^{Footnote 52}

But as suggested above, there is another form of basic structure review that, while also protecting the constitutional system from the legislature, does not give judges the final word on the validity of constitution-amending legislation. Weak basic structure review, present in several Latin American countries, seeks to protect the constitution from the legislature by the formal establishment of limits to constitutional change. However, at the same time, it recognizes that those limits do not apply to the people’s ultimate constitution-making power and provides the means for that power to be exercised. The countries that currently exhibit such a system moved in the nineteenth- and twentieth century from systems of weak judicial review to strong judicial review, but in the late twentieth- and twenty-first century, replaced the latter system with the weak basic structure review model. In Colombia, this move was the result of a judicial decision that, while asserting the Constitutional Court’s jurisdiction to invalidate constitutional amendments inconsistent with the fundamental principles in which the constitution rested, left open the possibility for those decisions to be overridden by a constituent assembly authorized to exercise the constituent power of the people (a mechanism provided by the constitution itself).^{Footnote 53}

In that judgment, rendered in 2003, the court followed a similar line of reasoning to the one adopted by the Supreme Court of India in Kesavananda (combined with a strong reliance on the distinction between constituent and constituted powers), deciding that although the Colombian Constitution of 1991 did not contain eternity clauses, the amending power could not be used to adopt constitutional changes so fundamental that amounted to the creation of a new constitution.^{Footnote 54} In that case, the court was asked to review the substance of a set of proposed amendments, even though the text of Article 241 of the constitution only gave it the authority to review constitution-amending legislation for formal or procedural defects.^{Footnote 55} The court’s reasoning took the following form. First, it was argued that when Article 241 of the Constitution of 1991 restricts the review power of the court with respect to constitutional amendments to that of identifying procedural or formal defects, it is necessarily conferring that body the power to examine if the institution promoting the constitutional changes is acting ultra vires ^{Footnote 56} – that is, if the legislature is attempting to use the amending procedure to adopt constitution-amending legislation that it is not authorized to adopt.

Second, the court maintained that there are some constitutional amendments that fall outside the scope of the legislature’s amending power, even in the absence of eternity clauses. It noted that legal scholars and courts around the world have recognized that under any democratic constitution the power of constitutional reform is subject to certain limits.^{Footnote 57} These limits emerge from the nature of the power of constitutional reform as a constituted, rather than a constituent, power.^{Footnote 58} That is, there are certain changes that are so fundamental that, while only altering part of the constitutional text, amount to the creation of a new constitution and can thus only be adopted by the constituent people. The absence of eternity clauses only meant that any provision of the Constitution of 1991 could be reformed and modified, not that one could use the amendment rule to, instead of reforming the constitution, replace it with a different one.^{Footnote 59} Nowhere in the amendment rule,^{Footnote 60} the court stated, is there an authorization ‘to eliminate or substitute the existing Constitution with a different one, something that can only be done by the constituent power’.^{Footnote 61} If the legislature were to use the ordinary amendment process to alter the constitutional text in a way that transgressed these limits, in a way that for all practical effects brought into existence a new constitution, it would be invading the terrain of the constituent people.

Even if one accepts the line of reasoning succinctly outlined above, there is at least one possible objection to this approach: what if a great majority of the people, the constituent power, favours a constitutional amendment that amounts to the creation of a new constitution? In other words, how to avoid the potentially dangerous situation that could take place under the traditional formulation of the strong basic structure review, where regardless of the intensity of popular support for fundamental constitutional change, there is no legal mechanism readily available to alter the basic principles of the constitution? In the opinion of the court, the framers of the Constitution of 1991 attempted to ‘solve’ this problem, to ease the tension between popular sovereignty and constitutional supremacy, by creating an opening for constituent power to manifest from time to time through an extraordinary constitution-making body. According to Article 376 of the Colombian Constitution, a legislative majority may ask electors whether they wish to convene a constituent assembly, and for the court this assembly could be authorized to adopt constitutional changes that amount to the creation of a new constitution, even if those changes have the effect of overriding a previous judicial invalidation of a constitutional amendment.

For the court, such an assembly would allow the political community to exercise its constituent power ‘in order to revise or modify its fundamental political decisions and to give its juridical institutions new forms and content’.^{Footnote 62} Such a mechanism is absent from the constitutions of India and Belize, and this is why those constitutions are examples of strong, rather than weak, basic structure review.^{Footnote 63} The new constitutions of Bolivia and Ecuador have taken a further step in institutionalizing the weak basic structure review model. These new constitutions establish that constitutional amendments that touch on the constitution’s ‘fundamental principles, its recognized rights, duties, and guarantees, or the supremacy of the constitution and the process of constitutional reform’^{Footnote 64} or that ‘set constraints on rights and guarantees … [or] change the procedure for amending the Constitution’^{Footnote 65} must take place through a constituent assembly convened by referendum.

The referendum can be triggered by popular initiative, that is, through the collection of a number of signatures from registered electors (20 per cent in Bolivia, 12 per cent in Ecuador).^{Footnote 66} Once convened, the assembly would draft a new constitution (or a radically transformed one) which would only become valid after being directly approved by the electorate in an additional referendum. Since these provisions explicitly negate the legislature’s right to alter the amendment procedure, they rule out the possibility of the legislature regaining an unlimited power of constitutional change through an ordinary constitutional amendment that alters the amendment rule. Under weak basic structure review, the legislature retains the power of adopting ordinary constitution-amending legislation in order to render ineffective judicial invalidations of ordinary laws, and courts retain the power to declare ordinary and constitution-amendment legislation invalid, but neither the legislature nor the courts are given the final word on the validity of fundamental constitutional changes.

Under this model, judicial declarations of invalidity are always susceptible of being overridden by the people, who retain an unlimited power of constitutional change that must be exercised outside the legislature. This does not mean that judicial declarations of invalidity are seen as something negative: on the contrary, judges are attributed with the responsibility of protecting the constitution, as well as the fundamentals of the constitutional order, from the legislature. That is to say, they are called to determine which type of constitutional changes are so fundamental that can only be adopted through the exercise of constituent power (a role that is expressly attributed to the constitutional court by the constitutions of Bolivia and Ecuador – Articles 202(10) and 443 respectively). In the countries that exhibit this approach (with the exception of Colombia),^{Footnote 67} judges have not yet used their power to strike down constitution-amending legislation and thus a constituent assembly has never been convened with the purpose of overriding a judicial invalidation of a constitutional amendment. However, since instances of fundamental constitutional change would be expected to be an exceptional occurrence, this does not mean that the weak basic structure model is not working.

V. The possible emergence of a fifth model

These four models of judicial review, as we saw, involve a particular relationship between courts, legislatures and constitutional amendments. Weak judicial review allows judges to strike down or suspend legislation inconsistent with rights, to only declare this inconsistency or, when possible, to prefer a rights-consistent interpretation. The legislature, however, is able to ignore or override judicial decisions, and in the latter case, it can do so through a law adopted by an ordinary legislative majority. Strong judicial review, on the contrary, always allows judges to strike down rights-inconsistent legislation, but gives legislative super-majorities (or the legislature and the electorate acting together) a final override power (and, at least theoretically, a final power with respect to the validity of constitutional amendments even in those cases in which the constitution contains eternity clauses).

Strong basic structure review not only provides judges with the power to strike down legislation, but also gives them the final word with respect to the validity of constitution-amending legislation. As noted earlier, under the strong basic structure review model, there is always the possibility of an extra-constitutional exercise of constituent power that could insert any content into the constitution. However, one of the distinguishing features of this model is that it operates under a constitution that does not provide the means for such exercises of popular power to occur. The result is that, as a matter of constitutional law, courts have the true final word as to the content of the constitution. Weak basic structure review is similar, but exists only in constitutional systems that contain mechanisms that facilitate the exercise of constituent power. These mechanisms attempt to reproduce the original episode of constitution-making, and could be used to reauthorize judicially invalidated constitutional amendments.

In a certain way, however, both strong and weak basic structure review (and arguably the traditional strong and weak judicial review models as well), are entirely consistent with what may be called a ‘voluntarist’ conception of constitutional authority, one in which the people, as the bearer of the constituent power, can ultimately insert any content into the constitution. A question that arises here is whether a decision attributed to the constituent power itself could ever be seen as susceptible of being set aside by courts. That is to say, whether the (voluntarist) conception mentioned above could be replaced by a ‘supra-constitutionalist’ approach according to which there are certain changes that are always constitutionally impermissible even if willed by the people and that, in such cases, it is the responsibility of the courts to review the acts of the constituent subject itself. In such a scenario, courts would be extending their review power beyond that envisaged under the strong and weak basic structure models. Such a conception would reflect not simply a different way of institutionalizing judicial power: it would be based on the idea that the authority of a constitution is not to be derived from ‘the sovereign people’, but from its adherence to certain supra-constitutional norms that require some principles to be always part of a constitution.^{Footnote 68}

The idea that constituent power could be subject to certain limits is not new, and can at least be traced back to seventeenth-century social contract theorists, whose reliance on natural law meant that, for them, there were things that even the sovereign people could not do. For example, Locke insisted that since an individual in the state of nature had ‘no arbitrary power over the life, liberty, or possession of another, but only so much as the law of nature gave him for the preservation of himself and the rest of mankind, this is all he doth or can give up to the commonwealth’.^{Footnote 69} Accordingly, no institution or collective entity could ever be authorized by individuals to exercise an arbitrary power over life, property and liberty. Sieyès himself offered a similar view, when he wrote that ‘prior and above’ constituent power ‘there is only natural law’, and that it was ‘impossible to imagine a legitimate association whose object would not be the common security, the common liberty and, finally, the common welfare’.^{Footnote 70} Of course, one thing is to say that ‘natural law’ poses limits on constituent power, and quite another that those limits are to be enforced by judges (after all, Thomas Hobbes and Jean Bodin also thought that sovereign princes were limited by natural law, even if those limits were not to be judicially enforced).^{Footnote 71}

But the judicial enforcement of this type of limits could have been in the minds of the German judges who, in the famous Southwest Case, stated that there are constitutional principles ‘so fundamental and to such an extent an expression of a law that precedes even the constitution that they also bind the framer of the constitution’.^{Footnote 72} Since the Basic Law contains an eternity clause (Article 79.3), the idea would be that there are some natural law principles (identified by some authors as ‘super-positive norms’) that stand above it, and that bind both the amending and constituent powers.^{Footnote 73} This view seems to find indirect support in some more recent judgments, where the references to ‘natural law’ are weaker or entirely absent, and have been replaced by appeals to fundamental human rights or universal principles.^{Footnote 74} The Lisbon Case, in which the German Federal Constitutional Court ruled on several constitutional issues surrounding the adoption of the Treaty of Lisbon, offers one of the most recent examples.

In that case, the court stated that ‘the constituent power of the Germans which gave itself the Basic Law wanted to set an insurmountable boundary to any future political development’ through an eternity clause that ‘prevents a constitution-amending legislature’ from affecting the principles laid down in Article 1 and Article 20 (principles that reflect the constitution’s identity). However, it then added the following obiter statements: ‘[i]t may remain open whether, due to the universal nature of dignity, freedom and equality alone, this commitment even applies to the constituent power, i.e. to the case that the German people, in free self-determination, but in a continuity of legality to the rule of the Basic Law, gives itself a new constitution’.^{Footnote 75} The phrase ‘in a continuity of legality to the rule of the Basic Law’ suggests that in order for the court to have jurisdiction to invalidate an act of the constituent power, the exercise of this power must not involve a legal revolution (that is to say, a break in the chain of legal continuity that may come accompanied with the abolition of the Constitutional Court and the emergence of a new juridical order, based on a different Grundnorm).^{Footnote 76} In other words, this qualification may be understood as recognition that the effectivity of any judicial attempt to enforce substantive limits on a constitution-making body, ultimately depends on purely political considerations.

In Venezuela, which has a system of judicial review that falls under the weak basic structure model, the country’s highest court seems to share the views of its German counterpart. In a 2006 judgment that concerned the eligibility to office of certain state officials, the Constitutional Chamber of the Supreme Court of Justice stated in passing that the people’s constituent power is subject, like any political power, to limits that arise ‘from the rights inherent to all human persons and derived from their own dignity’.^{Footnote 77} It is not clear whether this statement necessarily implies that the limits to the constituent power are judicially enforceable, that is, that in the event of the convocation of a constituent assembly under Article 348 of the constitution, the court would be prepared to assert its jurisdiction to invalidate newly adopted provisions if they are determined to be contrary to the unspecified fundamental rights mentioned above. On the contrary, it might be that these limits are to be enforced politically, an approach that is suggested by the constitutional text itself. While Article 347 of the constitution describes the people as the bearer of the ‘original constituent power’ and states that in the exercise of that power it may ‘create a new juridical order’, Article 350 expresses that ‘the people of Venezuela … will not recognise any regime, legislation, or authority contrary to democratic values and principles, or that affects human rights’.^{Footnote 78}

Common law constitutionalism, which was previously identified as a possible example of strong basic structure review, may nevertheless be seen as the closest approximation to a model in which the judiciary is called to impose legal limits on a constituent people.^{Footnote 79} It is in a way strange that countries that currently operate under the weakest form of judicial review appear as candidates to adopt the strongest version of judicial review imaginable (if the obiter contained in cases such as Jackson ever becomes ratio). The reason for this is directly connected to the fact that in those constitutional systems (particularly in that of the United Kingdom), parliament was traditionally conceived as a sovereign law-making body, as a constituent assembly in permanent session.^{Footnote 80} Accordingly, there is no extra-constitutional people to whom an ultimate constitution-making faculty is to be attributed: the people exercises its constitution-making power through parliament, and if this power is to be subject to legal limits, then those limits apply to the people as well.^{Footnote 81} Moreover, the common law constitution that serves as the basis for judges’ strike-down power under this model is not understood as having been the product of a discrete political act, but a result of historical practices. In short, there is no obvious reason why judges would conclude that the law-making power of parliament is limited by the common law constitution and at the same time suggest that the principles protected by that constitution could be superseded by a popular exercise of (extra-parliamentary) constitution-making.

This is why Rivka Weill has recently written that ‘[u]nder common law constitutionalism, certain rights and values are too fundamental for even the people or the original constituent assembly to alter’.^{Footnote 82} However, even in judgments that contain strong statements in favour of common law constitutionalism, there are some passages that evidence the influence that the voluntarist conception of constitutional authority may have even over judges apparently sympathetic to this ‘fifth’ model. For example, in Jackson, Lord Hope stated that the principle of parliamentary sovereignty, which for him was a creature of the common law, ‘is built upon the assumption that Parliament represents the people whom it exists to serve’ and that it ‘depends upon the legislature maintaining the trust of the electorate’.^{Footnote 83} These statements are not in any way conclusive, but by referring to the citizenry as somehow superior to parliament, they suggest that the idea that judges would ever enforce substantive limits on the constitution-making power of the people might never become a constitutional reality.^{Footnote 84}