CONSTITUTIONS ARE INTERGENERATIONAL contracts between ‘we, the people’ and our future selves. However, some constitutions also aspire to be primarily monologues rather than dialogues between the constituent power and the future generations. They burden future generations by entrenching constitutional values against future decline,1 and limit constitutional change by institutionalising complex constitutional amendment procedures and including eternity clauses.2 The legitimacy of this inter-temporal binding as well as the attempt to shape and protect future generations from themselves has been thoroughly discussed in the literature.3 Contingent constitutional change, that is, the idea that the constituent power could defer the coming into effect of certain constitutional provisions and make them contingent upon the verification of factual or legal conditions (for example, a referendum, the enactment of a law as determined by the Constitution) has nonetheless remained undertheorised in the literature on comparative constitutional law.4
In the past decade, a growing number of scholars has analysed the relationship between time and law, in particular in the context of constitutional law.5 The legal literature has focused on the complexities of temporary legislation, timing laws and their effects,6 sunsetting judicial opinions,7 comparing temporary and permanent legislation and the incentives behind its adoption,8 lasting legislation,9 and, in the field of constitutional law, constitutional change and amendments by desuetude,10 and temporary constitutions.11 This strand of literature has focused on why, when, and how constitutional and legal dispositions should terminate. However, little has been said about the opposite phenomenon: the commencement of constitutional dispositions and whether sunrise clauses should be allowed to make it contingent upon a future event.12
Sunrise clauses have remained a highly overlooked topic in the literature.13 In this chapter I argue that for example in post-conflict and highly divided societies and in transitional contexts,14 sunrise clauses can be important tools to guarantee that the ‘future does not arrive too early’ and that important, transitional, and aspirational matters that cannot be realistically promised to the people at that time are nevertheless included in the constitutional agenda. As this chapter explains, this was the case of the former Directive 45 of the Indian Constitution that deferred the commencement of the right to education to a later date in the hope that the country would have the financial means to grant Indian citizens free access to education.
Sunrise clauses have been employed in the past century in different constitutions throughout the world so as to delay the commencement of some of their provisions to a certain date (automatic sunrise clauses) or make it contingent upon a certain event (contingent sunrise clauses). To illustrate, the Constitution of Georgia of 1995 delayed the coming into effect of some constitutional provisions to a later date. While the Constitution was due to enter into force ‘upon the date of recognizing the powers of the newly elected President and the Parliament of Georgia’ (Article 104[1], the commencement of Articles 49(1) [composition of Parliament] and 58(1) [formation of Parliamentary factions] was made dependent upon the enactment of ‘appropriate amendments and additions made to the organic law Election Code of Georgia’ (Article 104–1[1]).15 As this chapter explains, this type of provisions may not abound but it is far from uncommon.16
Although sunrise clauses are relatively unknown in the literature on comparative constitutional law, the idea of delaying the coming into effect of legal provisions or making it contingent upon a determined event is present in different fields of law and it can even be considered to be intrinsic to a legal system. When explicating the use of conditions in contracts, Arthur Corbin reminded us that ‘legal relations are merely mental concepts which are useful in enabling us to foresee the physical facts of the future’.17 Along with interpretation, the inclusion of conditions in contractual relations also promote the foreseeability of future facts since they allow a legal relation to start before all the contractual conditions (for example, payment) are fulfilled. As this chapter explains, sunrise clauses can play a similar role. The explicitly forward-looking nature of sunrise clauses in addition to the imposition of vague conditions for the coming into force of a constitutional provision might nonetheless raise important questions as to their constitutionality and the limits of the constituent power.
This chapter makes two contributions to the literature. First, it explores the meaning and functions of sunrise clauses as instruments of constitutional change. Secondly, it discusses their constitutionality in light of two judicial cases from South Africa and Ireland where this question was raised. This chapter relies on the literature on comparative constitutional law and offers a number of examples of different types of sunrise clauses included in different constitutions. While the Constitutions selected for this chapter are not always immediately comparable, these examples help operationalise the concept of ‘sunrise clause’ and show the instrumental diversity of contingent constitutional change.18
This chapter proceeds in three parts. In Section II I define and analyse the concept of ‘sunrise clause’ and provide an overview of different types of constitutional sunrise clauses. I refer not only to the use of sunrise clauses in the constitutional context but also in legislation and long-term contracts, where these provisions appear to be more common. This section offers a broad definition of sunrise clauses in order to encompass a number of sunrise variations, including automatic and contingent sunrise clauses. I also categorise sunrise clauses according to the type of contingency employed to determine their commencement. In this section, I also distinguish between sunrise clauses and by-law clauses.
In Section III, I explore the analogy of the Constitution to an intergenerational contract, which is required to achieve a compromise between short-term and long-term commitments by adopting either a backward- or forward-looking approach to constitutional change.19 I suggest that sunrise clauses should be employed to operationalise this forward-looking perspective and consider future contingencies that given the history of the country can be predicted by the constituent power.
In Section IV, I analyse a judicial decision of the Irish Supreme Court where the constitutionality of sunrise clauses was challenged. In light of this case, I argue that sunrise clauses are valuable instruments of constitutional change and ‘self-amendment’, which find their legitimacy directly in the constituent power. The constitutional intergenerational dialogue does not imply that all provisions should enter into effect at the same time. Rather, the constitutional will formulated in some provisions may remain dormant, ‘sunrising’ at a later date or upon the verification of a certain condition.
Sunrise clauses can be broadly defined as dispositions providing that a constitutional provision only comes into force on a specific date or that its coming into effect is contingent upon the verification of specific conditions.20 Sunrise clauses are a form of legal condition which have been employed in constitutions, primary and secondary legislation, and international treaties.21 Sunrise clauses are forward-looking instruments which determine that a provision shall remain dormant for a certain period. Their rationale is determined by the need to take into account conditions that are not yet verified at that time.
Although contingent constitutionalism has been studied in the United States to explain the interdependence between constitutional rights and state law,22 the concepts of contingent constitutional change and sunrise clauses as instruments of constitutional change have remained underexplored.23 Contingent legal instruments have been historically employed in the legislative and constitutional contexts not only in the United States but also in other common law (for example, India, Ireland) and, to a smaller extent, civil law countries (for example, Belgium). In this section, I start by providing an overview of the use of different types of sunrise clauses in legislation. I then distinguish between automatic and contingent constitutional sunrise clauses, and explain the difference between these provisions and by-laws.
A sunrise clause is a constitutional disposition that determines that a part of that constitution will only come into force after a specific date in the future or its commencement will be contingent upon the verification of specific conditions.24 Sunrise clauses bear thus a strong resemblance to contingent legislation and tie-barring provisions since these three legislative instruments share a common feature: the coming into effect of certain provisions is delayed and it is conditional. Until a future event occurs, the dispositions lie dormant.
Although sunrise clauses have not been widely discussed in the comparative constitutional law literature, this does not mean that they have never been analysed by prominent scholars25 or considered by constitutional drafters as an option for accommodating constitutional change. In the literature, Akhil Amar recently suggested that the framers should have adopted this approach for issues such as secession and slavery.26 They did not and thus erred, surrendering the future, ‘giving slavocrats extra political clout in every election in perpetuity’. Amar argues:
just as the Constitution allowed the transatlantic slave trade to continue for twenty years but provided that Congress could ban this odious traffic in 1808 and thereafter, so the document should have allowed slave states to get extra credit in the House and electoral college (…) but not thereafter.27
Since in 1789 slavery was already perceived by many as being ‘morally wrong’, Amar contends that the framers should also then have agreed that after 1808 antislavery rules would sunrise.28 While the proposition can only remain at the theoretical level in the case of slavery, the use of sunrise clauses can have a more practical dimension in other issues, such as that of self-determination or socioeconomic rights. From a normative perspective, the promise of the sunrise of the appreciation of self-determination rights on a specific date or upon the verification of specific conditions may guarantee enhanced consensus, avoid the deferral of important decisions with constitutional value to the ever-changing legislature, and complement sunset clauses by providing a timed follow-up.
A.Automatic and Contingent Sunrise Clauses
Sunrise clauses can be automatic when their commencement is determined by objective events, that is, only the effective date of a constitutional provision is delayed. In a certain sense, automatic sunrise clauses bear a resemblance to the so-called commencement clauses, which define an exceptional date of coming into effect in order to avoid the retroactive application of a new law which would occur if general rules would be applied.29
Contingent sunrise clauses determine that the entering into force of a provision is dependent on a subset of factual or legal conditions. The commencement of a clause can therefore be ‘tied’ to a legal action which requires the intervention of the Parliament or the Executive. This is the case of the enactment of a statute or the results of a referendum organised at a later date (for example, a sunrise clause on the future secession of a province or state).
Automatic sunrise clauses merely delay the commencement of a constitutional provision. This form of sunrise clause determines when a provision constitution should begin in order to avoid its premature coming into effect. In addition, its main goal is to avoid immediate commencement according to general rules on publication and commencement of laws or delay the commencement of a law for a longer period until legal or administrative problems regarding the new act are solved. The difference between this first type of sunrise clause and other forms of sunrise clauses can be at times almost imperceptible as the latter can also grant some flexibility regarding the date of coming into effect of a law.30 The correct timing of the coming into effect of a constitutional provision is relevant since it may determine the success of the operationalisation of the constitutional change.
Within this category of sunrise clauses, we can also find sunrise clauses which might be associated with sunset clauses in two different ways. First, we may find constitutional dispositions that sunrise naturally, for example, after the termination of an interim constitution or set of dispositions. Secondly, some clauses may also have a contingent sunrise, if they do not commence within a certain period, the provisions expire. An example of the latter can be found in Section 2 of the Twenty-second Amendment of the US Constitution,31 which reads:
This article shall be inoperative unless it shall have been ratified as an amendment to the Constitution by the legislatures of three-fourths of the several states within seven years from the date of its submission to the states by the Congress.
Contrary to automatic sunrise clauses, contingent sunrise clauses determine that the coming into effect of a provision or constitutional amendment is dependent upon an objective event such as the ratification of a number of states, a subjective condition (for example, in the case of aspirational contingencies as explained in Section III) or the adoption of operationalising legislation. An example of the first type of contingency can be identified in Article XIV of the Constitution of Maryland which determines that: ‘any Constitution, or change, or amendment of the existing Constitution, which may be adopted by such Convention, shall be submitted to the voters of this State, and shall have no effect unless the same shall have been adopted by a majority of the voters voting thereon’.
The establishment of dependency relations between dispositions have been present both on legislative and constitutional levels. The practice of making the coming into effect of one law dependent on the enactment of another one has been called tie-barring at the legislative level and has raised a number of legal issues. A similar rationale has been applied on the constitutional level with contingent sunrise clauses. Article 35 of the Belgian Constitution is an example of this type of sunrise clauses, since the transitional disposition included in this article determines that the constitutional provision on communal and regional powers will only enter into effect after a statute listing the exclusive powers of federation is enacted. Article 35 of the Belgian Constitution reads:
The federal authority only has competences in the matters that are formally assigned to it by the Constitution and the laws passed by virtue of the Constitution itself. The Communities and the Regions, each in its own field of concern, have competences for the other matters, under the conditions and in the terms stipulated by the law. This law must be adopted by a majority as described in Article 4, last paragraph.
Transitional provision: The law referred to in the second paragraph determines the date on which this Article comes into force. This date cannot precede the date of the entry into force of the new Article to be inserted in Title III of the Constitution, which determines the competences exclusive to the federal authority. (emphasis added)
Since the subject underlying Article 35 has proven to be highly controversial in Belgium, this constitutional provision continues to lie dormant at the time of writing.
Contingent sunrise clauses can become problematic when constitutional sunrise clauses imply discretionary decisions of the Executive on highly subjective concepts employed as contingencies (for example, the gradual concretisation of rights). In a number of developing countries, the right to education, for example, is presented as an aspirational right or made contingent upon ‘the economic capacity and development’ of the country (for example, Article 41 of the Indian Constitution). One of the challenges of contingent sunrise clauses refers to the identification of the correct condition or the risk that the contingency is too vague to be ever triggered. Sunrise clauses that use vague language and subjective conditions for example in the context of aspirational constitutionalism may therefore be devoid of meaning.
B.Sunrise Clauses and the Use of Conditions
Although sunrise clauses are underanalysed instruments, the use of conditions is inherent to law and it should hence not be overlooked in the context of constitutional law. In this section, I support this claim by analogising constitutions with long-term contracts which often require condition-precedent and condition-subject clauses to meet the concerns of the parties regarding future changes.32 I draw upon a 1996 decision of the South African Constitutional Court which examined the validity of the sunrise clauses included in the KwaZulu-Natal constitution.33
A constitution is a long-term contract which, according to empirical research, typically lasts 19 years,34 which might be shorter than many of our financial and personal contractual obligations. In most long-term contracts, the central contractual object is deemed to endure, but, in order to fit changing circumstances and accommodate foreseeable issues, the parties may agree to include condition-subsequent clauses, in common law, or the so-called ‘resolutory clauses’ in civil law.35 These clauses update and shape the original content of the contract, by determining the termination of certain contractual elements. In the concrete case of constitutions, sunset clauses determine the termination of obsolete constitutional dispositions or guarantee smooth constitutional transitions, for example from dictatorial to democratic regimes. A much less common instrument in the constitutional context is, what in contractual terms one would call ‘condition-precedent’ clauses (or ‘suspensive clauses’ in civil law). These clauses determine that a certain disposition will be inactive until a specific date or its coming into effect will be contingent upon a set of facts or upon the verification of a specific condition.36 These ‘sunrise clauses’ are a form of ‘contingency law’ since they establish that the effects of one disposition will be dependent on the verification of certain conditions.
In 1996, the South African Constitutional Court was asked to validate a provincial constitution containing a number of sunrise clauses. In March 1996, the IFP37 announced that the provincial constitution would include a sunrise clause reflecting the provincial powers this party wished to secure for the province of KwaZulu-Natal.38 These provisions would allegedly guarantee that this province would benefit from a higher degree of autonomy after and if the definitive national constitution would come into effect.39 Until then, a number of sections would lie dormant, waiting for that automatic trigger—the enactment of the new Constitution—to be pulled. The South African Constitutional Court did not validate this provincial constitution, since these clauses violated the Interim Constitution valid at that time.
The Court considered that the provincial constitution was fundamentally flawed on three grounds: usurpation of national powers,40 consistency clauses and ‘suspensive conditions’.41
The Constitutional Court considered that the ‘suspensive clauses’ could not be qualified as unwritten or inexistent just because they would commence later and were dependent upon a certain condition. Comparing these clauses to contractual ones, the Court argued that:
it is well established that in the field of contract an agreement subject to a suspensive condition is already a binding agreement, that its terms are clearly established and that, for example, a provisional creditor may, even before the condition precedent has been fulfilled, institute proceedings to protect such creditor’s provisional right.1[9] But what is clear is that merely to suspend part of the text of a provincial constitution that is inconsistent with the interim Constitution, cannot save the constitution from the consequence of such inconsistency.
The Court added that:
A suspended provision is part of the text, and it does not cease to be such simply because its operation is suspended until a future date, or is made contingent upon the happening of a future event. The text of the provincial Constitution is to be evaluated and certified as an integrated whole, because the meaning and effect of one particular clause can be crucially dependent on that of another. If certain clauses of the text come into operation after others, then the fact that certain clauses are inoperative for a period of time may well influence the effect and meaning of those parts of the text which do come into operation immediately upon certification in the absence of the suspended clauses.42
In this case, the inclusion of sunrise clauses by itself was not contested. Rather, due to the lack of other rules to assess its validity, the South African Court solved the controversy by referring to the common use of conditions in contract law and interpreting these sunrise clauses according to the analogy, that is, in light of the ‘entire contract’. In this specific case, more was at stake than an inter-temporal problem. The sunrise clauses under analysis were a disguised attempt to circumvent the Interim Constitution, making the coming into effect of the provincial powers contingent upon the enactment of the definitive constitution, which according to the predictions of that time, would be potentially more favourable to the self-determination of the province than the former.
Constitutional dispositions often delegate constitutional decisions to Parliament (‘by-law clauses’) or determine that a certain disposition shall be regulated as ‘determined by law’.43 While the enactment of such ‘law’ may hinder the practical concretisation of a constitutional right, the coming into effect of these dispositions is not at stake. As this chapter explains, sunrise clauses are only effective upon the verification of a certain condition and until that day arrives, they lie dormant.
By-law clauses do not delay constitutional decision-making to the future and in particular to future legislatures.44 Instead of a deferral, by-law clauses operationalise delegation. They decide who should have the power to legislate on a certain matter and determine that the legislature is the most competent body to do so. Tom Ginsburg and Rosalind Dixon have argued that ‘by-law clauses’ can avoid the ambiguity of constitutional silences, minimise design costs and error costs in the constitutional-making process.45 These clauses can either ‘require the legislature to decide certain constitutional issues in the future, or else explicitly empower the legislature to decide such issues’.46 Ginsburg and Dixon contend that by-law clauses consist in explicit delegations to future decision-makers to decide on a specific constitutional question. However, once we look closely, by-law clauses do not ‘defer the right to decide’ to the legislature, as the title of Ginsburg and Dixon’s work suggests. Rather, they identify actions that must be taken by the legislature through legislation, rather than by the Executive. They delegate the right to decide to the legislature. Contrary to by-law clauses, sunrise clauses make a constitutional decision and defer the coming into effect of the provision. In the case of contingent sunrise clauses, the verification of a condition might rest upon the legislature or the executive government, but also here, the central constitutional decision has been previously made by the constituent power.
III.CONTINGENT CONSTITUTIONAL CHANGE BETWEEN RETROSPECTIVE AND FORESIGHT
As democracies change over time, either as a result of a succession of gradual or disruptive events, so do constitutions.47 Constitutional change was traditionally thought to operate exclusively according to formal amendment procedures and to be limited by supermajoritarian requirements.48 More recently, the less institutionalised paths of constitutional amendment and the problem of ‘unconstitutional constitutional amendments’ initiated an intense debate in the literature.49 The idea of contingent or delayed constitutional change has nonetheless remained overlooked in the comparative legal literature.
Constitutional change is a problem for the legal enforcement of constitutions enacted in the wake of conflicts or other unsettling events.50 On the one hand, constitutional change is antithetic to the most fundamental desire to guarantee constitutional stability and avoid the high costs of constitutional instability at the political, economic and social levels.51 On the other hand, constitutions do not live as long as we would like them to, they ‘do not age well’, and tend to obsolesce rapidly.52 Unsurprisingly, constitutions are often amended by desuetude, that is, obsolete dispositions are silently amended because they are no longer implemented.53
Obsolete dispositions create nonetheless an undesirable feeling of uncertainty, which could open the door to abusive litigation.54 In addition, while there is the traditional belief that constitutional longevity and stability is particularly important for countries emerging from post-conflicts,55 there seem to be exceptions to the rule, proving that short-lived constitutions (such as the 1997 Thai Constitution) can also produce an important impact in the constitutional future of the country.56 Striking the balance between stability and constitutional change implicates firstly an analysis of the relationship between constitutions and time, secondly, the assessment of the instruments employed in national law to confer a certain degree of stability to constitutions without impeding change; and thirdly, allowing constitutions to take into consideration unpredictable or future circumstances. It is in this last context that the idea of contingent constitutional change as operationalised by sunrise clauses emerges.
Constitutions are compromises between the political forces present at the time of the drafting, the desire to shape posterity, and potential contingencies that may occur in the meanwhile. The evolution of democracy might be highly dependent on present and future contingencies.57 These contingencies are not only interpreted in the sense of ‘emergencies’ but rather more broadly as ‘future and sometimes unpredictable circumstances’, which have not been analysed in the context of the constitutional change. The hermeneutics of constitutional change require not only an analysis of past and present facts, but also a predictive theory of constitutionalism, which can tell us ‘what is occurring or, better, what will occur’.58 Traditional constitutionalism is often backward-looking placing considerable emphasis on the drafting moment, expecting the Constitution to become entrenched as it is.59 However, particularly in the contexts of transition, the Constitution should also be allowed to play a predictive role, allowing the constitutional project to deal with potential legitimacy deficits in a more gradual way.60
A theory of constitutional change must thus encompass a backward-looking and a forward-looking approach to this intergenerational approach. In the next subsection I explain why timing clauses such as sunset and sunrise provisions can promote a forward-looking approach to constitutional change.
A forward-looking approach to constitutional change can have multiple expressions. Constitutions can use sunset clauses to terminate constitutional provisions or interim constitutions; promote the openness of the constitutional text, for example, by employing vague constitutional language,61 indeterminate concepts, constitutional silences,62 or as mentioned in Section II by deferring further decisions to the legislator; or by including sunrise clauses.
Temporality, by determining the automatic termination of constitutions at the end of a fixed period, is meant to be an expression of a forward-looking approach to constitutional change. Thomas Jefferson famously argued in 1789 that: ‘No society can make a perpetual constitution, or even a perpetual law. The earth belongs always to the living generation.’63 Jefferson pleaded for temporary constitutions that would reign over one generation and then expire automatically. Sunset clauses can connect our idea of justice with that of ‘our posterity’ by acknowledging that some constitutional dispositions that respond to specific concerns of our generation (for example, ensuring transition from an authoritarian regime to democracy) should not last beyond ‘ourselves’. For example, in January 1776, New Hampshire adopted a simplistic Constitution which might not have been intended to endure but rather to timely reject the colonial charters under which it had been governed.64 In the 1990s, Hungary and South Africa drafted Interim constitutions to respond to transitional problems, and more recently, Thailand also adopted a temporary constitution.65
Sunset clauses66 are dispositions that determine the termination of a law, constitution or parts of it within a beforehand determined period.67 These provisions are conceived to automatically ‘erase’ a provision or a constitution which is no longer necessary either because it has fulfilled its function or because it is no longer effective.
Although there are multiple examples of temporary constitutions (for example, interim constitutions in South Africa, Hungary, Thailand) and temporary constitutional dispositions (for example, state constitutions in the United States), Jefferson’s plea for temporary constitutions was not widely embraced in most countries, where the idea of a temporary constitution appears to be contrary to the dominant view of enduring constitutions. Temporary constitutions seem to have served primarily transitional purposes and not generational ones. Although Jefferson considered temporality as a form of unchaining the future generations from past commitments, sunset clauses also constrain the constituent power, by determining when certain clauses should automatically ‘sunset’ without relying on the future generations’ judgement on whether the disposition should be maintained or not.68
B.Sunrise Clauses and Aspirational Constitutionalism
A forward-looking approach to constitutional change includes an explicit acknowledgement of the need to construct the constitutional project gradually. Constitutions might confer the power to operate this gradualism to legislators (for example, Article 242 of the Constitution of Angola) or predict these contingencies in their texts, delaying the coming into effect of certain constitutional provisions. An example of the latter has emerged in the context of aspirational provisions.
In post-conflict societies and developing countries, the drafters may resort to sunrise clauses to allow certain rights to sunrise when the economic and political conditions are more auspicious. Even when such sunrise is timed, the sunrise of rights or provisions containing vague language may however imply a great amount of discretion. This was the case of Directive 45 of the Indian Constitution, which read, in the first version of the Constitution:69
Provision for free and compulsory education for children: The State shall endeavor to provide, within a period of ten years from the commencement of this Constitution, for free and compulsory education for all children until they complete the age of fourteen years.
The ‘endeavours’ of this country took however much longer than the initial ten-year period and the right to education was only to sunrise many decades later, with the 2002 constitutional amendment. The potential late sunrise of provisions is thus a risk present in vaguely formulated clauses that in this case might not have given the State sufficient incentives to ‘pull the trigger’ and initiate the sunrise of the provision.
Aspirational constitutionalism is a clear form of contingent constitutionalism which tries to bridge the wishes of past and future generations, as well as the different social, economic, and political circumstances they may be confronted with. Aspirational constitutionalism refers to ‘a process of constitution building (…) in which constitutional decision makers understand what they are doing in terms of goals that they want to achieve and aspirations that they want to live up to’.70 These aspirations can be present in the preambles of constitutions, in the choice of institutional design and arrangement of lists of rights.71 Sunrise clauses can also concretise such aspirational goals. This forward-looking approach is often contingent upon economic, social, and political circumstances of a country. The Constitution of South Africa includes a number of forward-looking and aspirational clauses that translate the different gradation of this perspective. This is the case of Article 29: ‘(1) Everyone has the right—(a) to a basic education, including adult basic education; and (b) to further education, which the state, through reasonable measures, must make progressively available and accessible.’
In South Africa, sunset and sunrise clauses appear to have played an important role in conflict resolution both in constitutional and lawmaking areas. These clauses embody the acknowledgement that not ‘all aspirations can be immediately met without destroying the demands made by adversaries and these aspirations might not also be met at the same time or in short-term’.72 In the post-apartheid context, the gap between rich and poor was trying to be solved by the inclusion of numerous socioeconomic rights in the Constitution such as the right of access to education, housing, health care. Although these rights could not be immediately met, using sunset and sunrise clauses ‘could ensure that [such] aspirations do not fall off the national agenda’.73
Sunrise clauses formalise this forward-looking approach by predicting future contingencies (for example, a referendum, enactment of a statute). The inclusion of these dispositions does not make the constitutional project incomplete but in permanent dialogue with the future. Both sunset and sunrise clauses impose this dialogue. Sunrise clauses address however a different side of inter-temporality: They reflect the acknowledgement of the need to decide a constitutional matter by determining determine that society is not yet ready to embrace it or might only be, when and if certain conditions are satisfied. Therefore, the past generation attempts to make predictions about the future generation wishes, the same way sunset clauses assume that our future selves might not wish to be bound by the exceptional powers contained in an interim constitution.
IV.THE CONSTITUTIONALITY OF SUNRISE CLAUSES
This section starts with the discussion of an Irish case involving a less visible form of contingent constitutional change but which, at the resemblance of the mentioned South African decision, also had self-determination roots: Riordan v An Taoiseach. The applicant objected to the fact that the Irish Constitution was only to be amended once certain other events—in this case, a referendum, had taken place to the ‘satisfaction of the Government’. He contended that the constitutional amendment procedure of Article 46 of the Irish Constitution had been violated.
This section also delves into the legitimacy of constitutional sunrise clauses and their ability to operate ‘self-amendments’ when they are not dependent on subjective contingencies or events that imply the intervention of other institutions.
Riordan v Taoiseach was a case decided in 1999 by the Irish Supreme Court,74 in the wake of the 1998 Belfast Agreement, which involved three interdependent components: first, an agreement between the political parties in Northern Ireland as to the internal structures for Northern Ireland; secondly, an agreement between the Northern Ireland parties and the Irish Government about the establishment of a number of all-Ireland institutions; thirdly, an agreement between the United Kingdom and Ireland. In this context, the Irish government committed to seeking the repeal of Articles 2 and 3 of the Constitution, which made a territorial claim to Northern Ireland.
According to the formal constitutional amendment laid down in Article 46 of the Irish Constitution, a referendum was compulsory to effectuate such an amendment. In this particular case, the success of the amendment was nonetheless dependent on the result of another referendum in Northern Ireland on the same issues.75 First, the IRA claimed that it was the legitimate government for the whole island of Ireland, as it was the successor of the second Dáil (Assembly), the last body to be elected on the basis of an all-Ireland vote. Secondly, since the 1998 Belfast Agreement was a negotiated agreement, simultaneous agreement was required from the different parties. The only exit under these circumstances was to establish a conditional amendment to Articles 2 and 3, relying on the results of the referenda taking place on the same day. These facts triggered the case Riordan v Taoiseach, where the applicant challenged this constitutional amendment, arguing that the procedure of Article 46 had been violated since the amendment was not only dependent on a constitutional referendum, but also on external events, that is, the referendum in Northern Ireland. The two decision-making moments were therefore ‘tie-barred’ (see Section II). As the Supreme Court explained:
before any amendment to Articles 2 or 3 can occur, two conditions precedent must be met. They are the coming into effect of the British/Irish Agreement (…) and the declaration of the Government to the effect that the State has become obliged, pursuant to the multi-party agreement, to give effect to the amendment to the Constitution.
The Supreme Court rejected the applicant’s arguments. Previously the High Court had already explained that:
The applicant ha[d] failed to understand the problem which confronted the Government in relation to the Belfast Agreement. That agreement imposed reciprocal obligations on the various parties to it and each party wished to be reassured that the other parties would carry out their respective obligations. For instance the Irish Government undertook to have Articles 2 and 3 of the Constitution amended but only on the basis that the British Government and the unionist parties to the Agreement would establish the power sharing executive and the cross-border bodies contemplated in the Agreement. On the other hand, the British Government and the Unionist parties did not wish to establish the power sharing executive and the cross-border bodies only to find that the proposal to amend Articles 2 and 3 was defeated in a constitutional referendum.
Both the Supreme Court and the High Court agreed on the validity and necessity of the ‘condition-precedent’, or sunrise clauses. The High Court underlined:
The people have a sovereign right to grant or withhold approval to an amendment to the Constitution. There is no reason therefore why they should not, provided the matter is properly placed before them, give their approval subject to a condition. It is quite wrong to suggest that the people have delegated to the Government the right to amend the Constitution. This is not so. The people have consented to an amendment to the Constitution subject to the happening of a particular future event.
In this case, a contingency was necessary for the operation of constitutional change, adding an additional burden to the formal constitutional amendment process. Also, contrary to the South African decision on the sunrise clauses in the Constitution of KwaZulu Natal, the contingency included in the Irish case was not being used to circumvent the Constitution but rather to promote one of its core values: peace in Northern Ireland.
As this case demonstrates, the existence of a condition and connected contingencies taking place outside the Irish Constitution was not judged unconstitutional. This decision illustrates however a different dimension of constitutional sunrise clauses since the contingency is not imposed by the Constitution but by an international agreement between Ireland and the United Kingdom.
Although this relationship of ‘condition-precedence’ could be contested, the process of constitutional change and the contingency seem to be logically interconnected. This logic interconnection or tie-barring does not only result from the Irish Supreme and High Courts, but it has also been defended, under different circumstances, in US state and federal courts deciding on contingency-legislation and more specifically on tie-barring provisions. According to some states’ constitutions (for example, Florida and Michigan), the validity of one bill may not be made contingent upon the passage of another one. According to the case law of state courts, tie-barring does not violate the mentioned constitutional provision as long as there ‘is a reasonable relationship between the statutes which have been tie-barred to each other’ or a ‘direct and relative interdependence between them’.76 Although this criterion has been developed in a different setting, the logic interdependence between the sunrise of a provision and its contingency appear to be a reasonable rule applicable to any contingent sunrise clauses discussed in this chapter.
B.Sunrise Clauses and the Constituent Power
Following the Irish case, I explain in this section why the constituent power should be allowed to include sunrise clauses in the constitutional text, deferring the coming into effect of a constitutional provision to the future or requiring future generations to take further actions in order to determine its commencement.
While automatic sunrise clauses can be triggered automatically, a contingent sunrise clause imposes a deliberative burden on future generations. By doing so, sunrise clauses distinguish themselves from other timing provisions such as sunset clauses that determine the termination of a clause. Regardless of whether sunrise clauses leave some or no discretion to the Executive or Parliament as to when and how to allow a certain provision to sunrise, it is clear that they voice a constituent desire of constitutional change. Instead of allowing the future generation to make the central decision on such change, sunrise clauses enable the constitutional drafter to include it in the text and only defer its sunrise.
If such a constitutional change is operated automatically by simply delaying the effective date of a provision, a self-amendment or an ‘autopoietic amendment’ would occur.77 Such an amendment would operate regardless of existing formal constitutional amendment procedures. This might seem disconnected from reality but it is not. For example, in the case of a constitution enacted after a conflict, the constituent power may foresee that the country will only be prepared to accept or concretise certain provisions at a later date (for example, Articles 167 and 169 of the Weimar Constitution). In this sense, contingent constitutional change is susceptible of materialising the traditional concept of the constituent power as a rational and unlimited power (for a thorough discussion of the concept of constituent power see the chapter by Thomaz Pereira),78 which ‘even after the constitution is adopted does not lose the right to alter its decision’.79
When sunrise clauses are triggered automatically, the constituent power sunrises, resulting in a ‘self-amendment’ of the Constitution. This idea can seem at first problematic as it blurs the concepts of ‘original constituent power’, the sovereign power to ‘make the Constitution’, and ‘derived constituent power’, the power to amend the Constitution.80 However, the legitimacy of these constitutional sunrise clauses is found in the original constituent power, which comprehends the power to draft the Constitution and determine when it comes into effect.
The term ‘self-amendment’ might sound rather imprecise as on the one hand it suggests that sunrise clauses amend the Constitution; and on the other, it is antithetical as the constituent power only speaks once. Can sunrise clauses be qualified as alternative and legitimate instruments of constitutional change? As Akhil Amar explains in the context of sunrise clauses: ‘at first, it might be wondered whether anything truly constitutional can be said about which future amendments should be adopted. After all, we the People are free to adopt just about any amendments we like’.81
The Constitution is a project with unwritten parts, which can be enlightened not only by the legislature but also by the framers. Amar argues that, on a second thought, this enlightenment regarding future amendments might be entirely constitutional depending on the topics. Sunrise clauses can thus be interpreted as devices that shape the future trajectory of the Constitution. Amar suggests, for example, that a future amendment (or sunrise clause) regarding the criminalisation of flag-burning or the restriction of equality rights of same-sex couples should be rejected as such amendments ‘would do violence to the trajectory of the American constitutional project over the past two hundred years’. Amar suggests however different future amendments or sunrise clauses, such as those used by ‘various states in the Founding era to achieve the gradual abolition of slavery itself. Under these rules, existing slaves would not be liberated—but eventually their future children would walk free’.82
Sunrise clauses can thus potentially accommodate constitutional change without requiring predictable constitutional amendments. These clauses can also constrain future constitutional change by impeding the constituted powers to decide on the life of a certain disposition.83 These instruments induce a process of ‘self-amendment’ which is determined by the constituent power in an attempt to avoid the perpetuation of dispositions motivated by transitional concerns or insufficient information.84 A sunrise clause may also provide additional time for society to reach an agreement on controversial topics that the society might not have been prepared to accept at the time of the drafting (for example, the organisation of a referendum on secession or other self-determination rights) and concretise (for example, socioeconomic rights).
In this context, it is important to emphasise that, at the resemblance of sunset clauses and transitional provisions, the use of sunrise clauses should also be limited to constitutional provisions that do not consecrate fundamental rights or state institutions necessary for the normal functioning of the society at stake.85 This normative statement, based on the case-law of the German Constitutional Court, implies a casuistic analysis which should consider the political and economic situation of the country.
Time and constitutions are so intimately connected that it is often hard to separate them.86 The intertwining of past, present and future constitutional wills is personified in a constitution that is supposed to live beyond its framers. A constitution implies multiple commitments between the past, present, and future generations, and a more durable constitution is commonly regarded as a shield against the whims of political power, shifting political majorities and crises.87 However, ‘the past cannot fully control the future (…) [and] there are not enough rules in the world adequate to navigate centuries of change and crisis’.88 Sunrise clauses personify the challenges of this inter-generational dialogue but instead of deferring constitutional decisions to the future, they allow the framers to make them.
Sunrise clauses allow the constitution-making and amending process to be not only a backward-looking effort in the sense of ‘claiming our constitutional inheritance’ but also a form of ‘making a constitutional donation’ to ‘our posterity’.89 Nevertheless, sunrise clauses determine a priori the path to our posterity, by entrusting the constituent power with the responsibility of predicting an effective date for certain constitutional changes as well important contingencies that might consolidate the enforcement of a constitutional provision.
The added value of sunrise clauses when compared to constitutional amendments or by-law clauses resides namely first in the fact that here the constituent power decides to confer a constitutional character to such controversial topics and ensure that they are included in the constitutional agenda. Secondly, instead of delegating further decisions on these issues to the legislature using, for example, by-law clauses, the constituent power decides and only delays the coming into effect of such dispositions to a moment when, for example, all the necessary economic or political conditions will be verified.
Sunrise clauses remain a relatively overlooked instrument in the constitutional practice. This chapter has sought to fill this gap, opening the door to future research on both sunrise clauses and contingency constitutionalism. This perspective on contingent constitutional change complements the ‘one-way communication’ or monologue between the past and future generations, and acknowledges a basic fact of life that has permeated most fields of law: in long-term contracts, change is inevitable but constitutional drafters might also want to ensure that the validity of the constitutional contract remains contingent upon it.
*I would like to thank Richard Albert, Jack Balkin, David Grewal, Oran Doyle, Eric Fish, Jurgen Goossens, Nicholas Robinson, Yaniv Roznai, Mark Tushnet, and Ozan Varol, and the participants of the Boston College Law School Workshop on Constitutional Amendments for insightful comments on earlier versions of this chapter. I would also like to extend my gratitude to the Information Society Project and the Niels Stensen Fellowship.
1M Dorf, ‘The Aspirational Constitution’ (2008) 77 Geo Wash L Rev 1631; N Barber, ‘Why Entrench’ (2016) Int’l J Const L (forthcoming).
2Y Roznai, ‘Towards a Theory of Unamendability’ (2015) Working Paper 15-12/2015 <http://www.academia.edu/11053169/TOWARDS_A_THEORY_OF_UNAMENDABILITY> accessed 1 August 2016. See also Y Roznai, ‘Unamendability and the Genetic Code of the Constitution’ (2015) New York University Public Law and Legal Theory Working Paper 514/2015 <http://lsr.nellco.org/nyu_plltwp/514> accessed 1 August 2016; UK Preuss, ‘The Implications of “Eternity Clauses”: The German Experience’ (2011) 44 Israel L Rev 429; A Friedman, ‘Dead Hand Constitutionalism: The Danger of Eternity Clauses in New Democracies’ (2011) 4 Mexican L Rev 77, 80.
3AC Desai, ‘What a History of Tax Withholdings Tells US about the Relationship Between Statutes and Constitutional Law’ (2014) 108 Nw U L Rev 859; MJ Klarman, ‘Majoritarian Judicial Review: The Entrenchment Problem’ (1997) 85 Georgetown L J 492, 496; R Dixon and T Ginsburg, ‘Deciding Not to Decide: Deferral in Constitutional Design’ (2011) 9 Int’l J Const L 636, 637; see also J Balkin and R Siegel (eds), The Constitution in 2020 (Oxford University Press, 2009); J Balkin, Constitutional Redemption: Political Faith in an Unjust World (Harvard University Press, 2011).
4In the US constitutional law, sunrise clauses have been discussed primarily by Akhil Amar. See A Reed Amar, America’s Unwritten Constitution: The Precedents and Principles We Live By (Basic Books, 2012) 474-476; America’s Constitution: A Biography (Random House, 2010). See also EB Foley, ‘The Posterity Project: Developing a Method for Long-Term Political Reform’ (2013) 66 Okla L Rev 1; DE Herz-Roiphe and D Singh Grewal, ‘Make Me Democratic, But Not Yet: Sunrise Lawmaking and Democratic Constitutionalism’ (2016) 90 N Y L Rev 1975.
5Specifically, on time and law, and the effects of ‘temporary de-juridification’, see A Kouroutakis and S Ranchordás, ‘Snoozing Democracy: Sunset Clauses, De-juridification, and Emergencies’ (2016) 25 Minn J Int’l L 29.
6JE Gersen and E A Posner, ‘Timing Rules and Legal Institutions’ (2007) 121 Harv L Rev 543.
7N Katyal, ‘Sunsetting Judicial Opinions’ (2004) 79 Notre Dame L Rev 1237.
8See F Fagan, Law and The Limits of Government: Temporary Versus Permanent Legislation (Edward Elgar, 2013).
9For a critical approach to temporary legislation see RM Kysar, ‘Lasting Legislation’ (2011) 159 U Pa L Rev 1007; RM Kysar, ‘The Sun also Rises: The Political Economy of Sunset Provisions in the Tax Code’ (2006) 40 (2) Ga L Rev 335.
10See R Albert, ‘Constitutional Amendment by Constitutional Desuetude’ (2014) 62 Am J Comp L 641.
11O Varol, ‘Temporary Constitutions’ (2014) 102 Cal L Rev 409.
12On commencement clauses, see M Gobbi, ‘When to Begin: A Study of New Zealand Commencement Clauses with Regard to those Used in the United Kingdom, Australia, and the United States’ (2010), 31 (3) Stat L Rev 153.
13In the context of Constitutional Law, see Amar (n 4) 474–76.
14S Choudhry (ed), Constitutional Design for Divided Societies: Integration or Accommodation? (Oxford University Press, 2008). See also the special 2013 issue of the International Journal of Constitutional Law on Constitutional Transition (vol 11(3)); see S Choudhry, ‘Constitutional Transitions in the Middle East: Introduction’, 11 (3) Int’l J Const L 611.
15According to Art 107(2) of the Constitution of Georgia, Arts 18(2) and (3) of the Constitution would also only enter into force after the respective criminal procedural legislative acts are adopted.
16Another example is Art 148 of the Constitution of Tunisia (2014) that also delays the commencement of multiple dispositions, making them for example contingent upon ‘the definitive results of the first general elections’, the direct election of the President, the election of the Assembly of Representatives, and the enactment of a number of operationalising statutes.
17A Corbin, ‘Conditions in the Law of Contract’ (1919) 28 Yale L J 731.
18For a reflection on the methodology of comparative constitutional law, see R Hirschl, ‘The Question of Case Selection in Comparative Constitutional Law’ (2005) 53 Am J Comp L 125.
19See A Sajó, ‘Preferred Generations: A Paradox of Restoration’ (1993) 14 Cardozo L Rev 847.
20See M Freeman, Necessary Evils: Amnesties and The Search for Justice (Cambridge University Press, 2009) 142.
21For a historical perspective on the use of contingent legislation and contingent provisions in interstate commerce, see JP Comer, Legislative Functions of National Administrative Authorities (Columbia University Press, 1927) 63–70. For some types of treaties, such as humanitarian law conventions (eg, Geneva Convention of 1949), the effective date of the withdrawal from a treaty can be made contingent upon external events, such as the cessation of an existing armed conflict, see LR Helfer, ‘Terminating Treaties’ in D Hollis (ed), The Oxford Guide to Treaties (Oxford University Press, 2014) 634, 642.
22See WA Logan, ‘Contingent Constitutionalism: State and Local Criminal Laws and the Applicability of Federal Constitutional Rights’ (2009) 51 Wm & Mary L Rev 143; M Manheimmer, ‘The Contingent Fourth Amendment’ (2015) 64 Emory L J 1229.
23As this chapter explains below, in the United States, Akhil Amar has regretted the limited use of sunrise clauses in the Constitution, see Amar, America’s Unwritten Constitution (n 4) 474–76; A Reed Amar and S Levinson, ‘What Do We Talk About When We Talk About the Constitution?’ (2013) 91 Tex L Rev 1119, 1143.
24See M Freeman, Necessary Evils: Amnesties and The Search for Justice (Cambridge University Press, 2009) 142.
25See Amar (n 4) at 474–76.
26ibid 474.
27Amar and Levinson (n 23) 1143.
28ibid.
29M Gobi, ‘When to Begin: A Study of New Zealand Commencement Clauses with Regard to those Used in the United Kingdom, Australia, and the United States’ (2010) 31(3) Stat L Rev 153.
30ibid 153, 186.
31See The Constitution of the United States, Amendment XXII. This type of sunrise/sunset clause is also present in other amendments of the Constitution of the United States, see, for example, s 3 of the Twenty-first Amendment which repealed the Eighteenth Amendment.
32The contractual terminology was also used by the Irish Supreme Court in the case analysed in Section IV.
33Constitutional Court of South Africa, Certification of the Constitution of the Province of KwaZulu, Case CCT 15/96, 6 September 1996.
34Z Elkins, T Ginsburg and J Melton, The Endurance of National Constitutions (Cambridge University Press, 2009).
35For the clarification of this terminology, see, eg, B Fauvarque, ‘Position Paper on Conditions, Unidroit, Working Group for the Preparation of the Principles of International Commercial Contracts’, UNIDROIT (2007) 10, <http://www.unidroit.org/english/documents/2007/study50/s-50-103-e.pdf> accessed 1 August 2016.
36This analogy was also made by the Constitutional Court of South Africa in the assessment of the constitutionality of the provincial constitution of KwaZulu-Natal in 1995 (see below for more information on this example).
37Inkatha Freedom Party, still nowadays one of the largest political parties in South Africa.
38O Malley, ‘Increased Repression and Continued Attack in 1995’ <https://www.nelsonmandela.org/omalley/cis/omalley/OMalleyWeb/03lv02424/04lv03275/05lv03336/06lv03344/07lv03354/08lv03358.htm> accessed 1 August 2016.
39The negotiation of constitutional principles was particularly complex in the post-apartheid period when the different parties involved diverged as to the objectives and structure of the transition. A problematic case was KwaZulu-Natal, where there was significant resistance to the national government. The Zulu king was at the time struggling to demonstrate his authority and reject central government and its influence. The local legislature began working on a provincial constitution design to maintain a kingdom within the South African republic, establish a constitutional monarchy and guarantee independent control over security forces in the province. See Africa Confidential, vol 37, 5 July 1996, p 64 (‘a series of provincial powers would remain inactive until the national constitution came into force’). According to s 160(1) of the 1993 Constitution of South Africa (the ‘Interim Constitution’), the provincial legislatures were entitled to pass a constitution for their provinces. The provincial constitution had to be certified by the then recently established Constitutional Court of South Africa that had the task of guaranteeing that none of its provisions was inconsistent with the Interim Constitution. In order to pass the test of the Constitutional Court, KwaZulu-Natal decided an ingenious plan which included making the coming into effect of most controversial constitutional provisions contingent upon the enactment of the new and permanent South African Constitution. See Th Eloff, ‘Managing Negotiations: Lessons and Pitfalls 1990–1994’, in B de Villiers and J Sindane (eds), Managing Constitutional Change (HSRC Publishers, 1996) 1, 8.
40Constitutional Court (South Africa), Certification of the Kwazulu-Natal Constitution (CCT15/96) [1996] ZACC 17; 1996 (11) BCLR 1419; 1996 (4) SA 1098 (6 September 1996) [14].
41South Africa is a bi-juridical or a mixed system of law with Roman Dutch civil law and common law, which explains the civil law terminology here.
42Constitutional Court (South Africa), Certification of the Kwazulu-Natal Constitution (CCT15/96) [1996] ZACC 17; 1996 (11) BCLR 1419; 1996 (4) SA 1098 (6 September 1996) [39] [41].
43See Dixon and Ginsburg (n 3).
44On by-law clauses, see Dixon and Ginsburg (n 3).
45ibid 639.
46ibid.
47For a historical perspective on constitutional and legislative change, see M Schwartzberg, Democracy and Legal Change (Cambridge University Press, 2009); see also B Friedman and S B Smith, ‘The Sedimentary Constitution’ (1998) 147 U Pa L Rev 20.
48For an overview and comparison of the majority requirements for constitutional amendments in a number of countries, see X Contiades (eds), Engineering Constitutional Change: A Comparative Perspective on Europe, Canada and the USA (Routledge, 2013).
49On other paths to amend the Constitution beyond Art V of the Constitution of the United States, see generally B Ackerman, We the People: Foundations (Harvard University Press, 1993); more specifically, on constitutional amendment procedures and rules, see R Albert, ‘Amending Constitutional Amendment Rules’ (2015) 13(3) Int’l J Const L 655; ‘Constitutional Amendment by Constitutional Desuetude’ (2014) 62 Am J Comp L 641. There is a vast literature on unconstitutional constitutional amendments, see, eg, A Barak, ‘Unconstitutional Constitutional Amendments’ (2011) 44 Israel L Rev 321; J Mazzone, ‘Unamendments’ (2005) 90 Iowa L Rev 1746; R Albert, ‘Nonconstitutional Amendments’ (2009) 22(1) Canadian Journal of Law & Jurisprudence 5; C Bernal, ‘Unconstitutional constitutional amendments in the case study of Colombia: An analysis of the justification and meaning of the constitutional replacement doctrine’ (2013) 11 Int’l J Const L 339.
50SM Griffith, ‘The Problem of Constitutional Change’ (1996) 70 Tul L Rev 2121.
51Ph A Joseph and G R Walker, ‘A Theory of Constitutional Change’ (1987) 7(2) Oxford Journal of Legal Studies 155, 175.
52A Vermeule, ‘Constitutional Amendments and the Constitutional Common Law’ in RW Bauman and T Kahana (eds), The Least Examined Branch: Legislatures in the Modern State (Cambridge University Press, 2007) 229; On the ‘life expectancy’ of constitutions, see Elkins, Ginsburg and Melton (n 34).
53See R Albert, ‘Constitutional Amendment by Constitutional Desuetude’ (2014) 62 Am J Comp L 641.
54See MJ Mitchell, ‘Cleaning up the Closet: Using Sunset Provisions to Clean up Cluttered Criminal Codes’ (2005) 54 Emory L J 1671.
55See J Hatchard, ‘Establishing Popular and Durable National Constitutions in Commonwealth Africa’, in M Andenas (ed), The Creation and Amendment of Constitutional Norms (The British Institute of International and Comparative Law, 2000) 1.
56See T Ginsburg, ‘Constitutional Afterlife: The Continuing Impact of Thailand’s Postpolitical Constitution’ (2009) 7(1) Int’l J Const L 83: Ginsburg describes the context in which the 1997 Interim Thai Constitution was adopted and examines how some of its institutional innovations (such as the introduction of judicial review) were passed on to the Constitution that superseded it.
57See A Przeworski, ‘Democracy as a Contingent Outcome of Conflicts’ in J Elster and R Slagstad (eds), Constitutionalism and Democracy (Cambridge University Press, 1988) 64–66.
58Ph A Joseph and GR Walker, ‘A Theory of Constitutional Change’ (1987) 7(2) Oxford Journal of Legal Studies 155, 171.
59C Dupré and J Yeh, ‘Constitutions and Legitimacy over Time’, in M Tushnet, Th Fleiner and Ch Saunders, Routledge Handbook of Constitutional Law (Routledge, 2013) 45, 52.
60Dupré and Yeh (n 59) 45, 53.
61On constitutional language, see, eg, F Schauer, ‘An Essay on Constitutional Language’ (1981) 29 UCLA L Rev 803. On constitutional indeterminacy in the Constitution of the United States, see M Klarman, ‘Fidelity, Indeterminacy, and the Problem of Constitutional Evil’ (1997) 65 Fordham L Rev 1739.
62Constitutional silences are not necessarily cartes blanches for Parliament to legislate on constitutional matters not listed in the Constitution. See DN Hoffman, Our Elusive Constitution: Silences, Paradoxes, Priorities (SUNY Press 1997); L Tribe, ‘Toward a Syntax of the Unsaid: Construing the Sounds of Congressional and Constitutional Silence’ (1982) 57 Ind L J 524.
63S Straub ‘Th Jefferson, Letter to James Madison, September 6, 1789’ (The Federalist Papers Project) <http://www.thefederalistpapers.org/founders/jefferson/thomas-jefferson-september-6-1789>, accessed 19 August 2013: this letter referred to the possibility ‘of the representatives of a nation validly engage debts beyond what they may pay in their own time, that is to say, within thirty-four years of the date of the engagement’.
64J Dinan, ‘American Constitutions: Natural Law and Constitution-Making in the Founding Era’ (American Founding and Constitutionalism) <http://www.nlnrac.org/american/founding-era-constitution-making> accessed 1 August 2016.
65For a thorough analysis of temporary constitutions, see Varol (n 11) 409.
66It is important to distinguish between ‘sunset clauses’ and ‘sunrise clauses’: While sunset clauses determine the termination of a law or some of its dispositions, sunrise clauses, on the contrary, only determine that a law will come into effect later on a certain date. Until that period, the clause will be inactive. See M Freeman, Necessary Evils: Amnesties and the Search for Justice (Cambridge University Press, 2009) 142.
67On the definition of ‘sunset clause’, see Parliament of the United Kingdom, Glossary, [‘sunset clause’] <http://www.parliament.uk/site-information/glossary/sunset-clause/> accessed 1 August 2016. (‘A provision in a Bill that gives it an “expiry date” once it is passed into law. “Sunset clauses” are included in legislation when it is felt that Parliament should have the chance to decide on its merits again after a fixed period.’)
68Elkins, Ginsburg and Melton (n 34) 13, (‘a sunset clause arguably constrains future generations as much, and perhaps more, than does presumed permanence. Why should citizens artificially abandon a well-functioning constitution simply because it has reached a certain age?’).
69India amended its Constitution in 2002 to include a right to education.
70KL Scheppele, ‘Aspirational and Aversive Constitutionalism: The Case for Studying Cross-constitutional Influence through Negative Models’ (2003) 1(2) Int’l J Const L 296, 299.
71ibid 299.
72Ch Villa-Vicencio, Walk with Us and Listen: Political Reconciliation in Africa (Georgetown University Press, 2009) 85.
73ibid 85.
74Denis Riordan Applicant v An Taoiseach Bertie Ahern, The Government of Ireland, (No 2), 1998 No 213 JR, SC No 202 of 1998.
75I thank Oran Doyle for the contextual information on this case.
76See Re Advisory Opinion to the Governor, 239 So 2d 1, 9 (Fla 1970). See also Gaulden v Kirk [1950] 47 So 2d 567. Town of San Mateo City v State ex rel. Landis, 117 Fla. See also the Cargo Brig Aurora regarding the revival of an act by proclamation of the President, and Field v Clark, where the constitutionality of broad contingent legislation was challenged. In Brig Aurora, the Court denied that the President was exercising unwarranted discretion in reviving a law, although the reasons to revive the law were not established in the statute at the time. See Cargo of the Brig Aurora v United States [1813] 11 US 382. However, the enactment of contingent legislation has also been criticised in the United States, particularly when the legislative acts are found to be incomplete and uncertain until another agent acts. For a critique of contingent legislation, see Th M Cooley, Constitutional Limitations (Legal Classics Library, 1987) 121. See also F Kameny, ‘Are Inseverability Clauses Constitutional?’ (2005) 68 Alb L Rev 957, 1016, citing 16 CJS Constitutional Law ß 166, at 532 (1984): ‘In any case, as a general rule, the enactment of the statute itself may not be made contingent on the action of officers or people; the act must be complete in itself, must be made law by the legislature, and only its effect and operation may be made dependent on the contingency.’ For a more recent study of severability and conditions in constitutional law, see E Fish, ‘Severability as Conditionality’ (2015) 64 Emory L J 1298.
77I borrow the term from Niklas Luhmann and Gunther Teubner, but not entirely with the meaning developed by these scholars, see G Teubner, Autopoietic Law: A New Approach to Law and Society (Walter de Gruyter, 1988).
78See E Sieyès, What is the Third Estate? (Hackett Publishing Company, Inc, 2003)
79See J Colon-Rios, Weak Constitutionalism: Democratic Legitimacy and the Question of Constituent Power (Routledge, 2012) 8. See also M Loughlin, ‘The Concept of Constituent Power’ (2014) 13(2) Eur J Pol Theory 218.
80Referring to the distinction in the French literature between original constituent power, as the power to make the Constitution, and derived constituent power, as the power to enact and amend (‘établir la constitution’) and arguing that the power to amend constitutions is ‘sui generis’ constituent power, see Rosnai, ‘Towards a Theory of Unamendability’ (n 2). On the traditional definitions of original constituent power and derived constituted power in the French literature, see, eg, G Burdeau, Droit constitutionnel (LGDJ 1988) 76–77; K Gözler, Le pouvoir de révision constitutionnelle (Presses Universitaires du Septentrion, 1997) 9.
81Amar (n 4) 451.
82ibid 446.
83See House of Lords, Select Committee on the Constitution, The Process of Constitutional Change, 15th Report Session 2010–2012, 21 (‘even without a codified constitution, it is possible to provide for specific procedural requirements within constitutional bills, including such measures as (…) sunrise and sunset clauses, and parliamentary super-majorities’.)
84Temporary instruments are often used to gather information and overcome the cognitive bias that often characterises the lawmaking process, see JE Gersen, ‘Temporary Legislation’ (2007) 74 U Chi L Rev 247.
85See my previous work S Ranchordás, Constitutional Sunsets and Experimental Legislation: A Comparative Perspective (Edward Elgar, 2014). Deciding on the constitutionality of transitory law, the German Constitutional Court decided in 1976 that ‘laws that are indispensable for the legal capacity and [normal] functioning of a state’ and the laws that are required for the concretisation of fundamental rights guarantees (eg, media and broadcasting laws in that specific case) are not compatible with a temporary or transitory nature. See 1 BvR 79/70 of 09.11.1976.
86Dupré and Yeh (n 59) 45.
87For a thorough discussion of the value of temporary constitutions as a reality opposed to the traditional perception that constitutions should be rigid and long-enduring, see Varol (n 11) 411 (‘a durable constitution is thought to constrain political majorities in moments of irrational fear or passion. A constitution represents a powerful acknowledgement by a society of its own weaknesses and its ability to fall prey to pernicious majoritarian impulses’).
88J Balkin, Constitutional Redemption: Political Faith in an Unjust World (Harvard University Press, 2011) 9.
89Amar (n 4) 477.