Rigid (Entrenched) / Flexible Constitutions
- Amendment or revision — Form and substance of constitution — Types of constitutions
Published under the direction of the Max Planck Foundation for International Peace and the Rule of Law.
General Editors: Rainer Grote, Frauke Lachenmann, Rüdiger Wolfrum.
1. Core Meaning
1. The terms rigid/flexible constitution are opposites signifying how constitutions can be revised and the extent to which they are able to adjust to changing circumstances (see amendment or revision of constitutions).
2. A modern constitution is a set of written provisions or norms enacted by political actors, which define the framework and structure of the governmental machine, regulate the allocation of powers in the state, and, in most cases, contain the citizens’ fundamental rights, which the political actors pledge to respect and enforce (Pasquino (2012) 1008–10). Within Hans Kelsen’s ‘hierarchy of norms’ or the ‘pyramid of norms’, the constitution is located in the highest position in a state’s legal system, described as the ‘basic norm’. Constitutional norms are usually considered superior and more rigid than the other norms; they usually describe a special procedure and more rigorous conditions for their amendment (Kelsen 221–2) (see also theories concerning the hierarchy of norms; supremacy / primacy).
3. In the past, scholars were used to distinguish between ‘written’ constitutions and ‘unwritten’ constitutions (which consist mainly of customs and established practices) (codified / uncodified constitutions). In his Studies in History and Jurisprudence, Lord James Bryce criticized this classification for the reasons that first, many unwritten constitutions that have originated in custom later include statutes, and once a custom is being written it can hardly be designated as unwritten, and that second, all written constitutions, which are explicitly set forth in a special document or documents, have an element of unwritten usage (Bryce (1901) 127). Therefore, Bryce suggested replacing those terms with the alternative distinction between ‘rigid and flexible constitutions’.
4. A ‘rigid’ legal norm is a superior norm which the legislature is prohibited to modify through majority rule. This superiority, which makes the rigid constitution more difficult to amend, is the effect and the consequence of the fact that the constitution is a creature of a superior creator—what the Founding Fathers of the eighteenth century constitutions called ‘constituent power’ (pouvoir constituant). Rigid constitutions, such as that of France or Switzerland, are thus the work of the superior constituent power, not the work of the ordinary legislative authority which holds inferior constituted power and thus cannot be changed by the latter. Rigid constitutions stand above the other laws of the country and their amendment procedure is more complicated than the one employed for ordinary laws. In contrast, flexible constitutions, such as that of ancient Rome and modern England, are at a similar normative level to other laws of the country, or are also in the form of recorded decisions defining and confirming a custom. They derive from the same authorities which make the ordinary laws and are promulgated, repealed, or amended by the same procedure as ordinary laws (Bryce (1901) 129–130; Bryce (1905) 7–8). Therefore, Bryce’s classification of constitutions refers to the amendment procedure but is fundamentally based on the relationship between the constitution and the state’s ordinary laws.
5. Flexible and rigid constitutions carry, according to Bryce, diverse strengths and weaknesses. According to Bryce’s metaphor, a rigid constitution is like ‘an iron railway-bridge, built solidly to resist the greatest amount of pressure by wind or water that is likely to impinge upon it’ (Bryce (1901) 189). Therefore, rigid constitutions provide stability for political institutions, and additionally carry the advantages of explicit confirmation in a written text. Of course, this rigidity can also be regarded as a weakness. For example, rigid constitutions cannot necessarily provide a decisive solution to all constitutional problems which may arise, as there may be lacunae or obscurities (Bryce (1901) 184–189). Moreover, rigid constitutions carry an element of danger at certain times. As Bryce states:
[w]hen a party grows up clamo[u]ring for some reforms which can be effected only by changing the Constitution, or when a question arises for dealing with which the Constitution provides no means, then, if the Constitution cannot be amended in the legal way, because the legally prescribed majority cannot be obtained, the discontent that was debarred from any legal outlet might find vent in a revolution or a civil war. The Constitution of the United States illustrates this danger on so grand scale that no other illustration is needed (Bryce (1905) 67–68).
6. This is, in contrast, the greatest strength of flexible constitutions. Flexible constitutions, which are elastic by nature, may provide adaptation, resilience, and balance to the constitutional norms ‘just because their form is not rigidly fixed, a temporary change is not felt to be a serious change’ (Bryce (1905) 22). By more easily allowing change, flexible constitutions minimize the risk of extra-constitutional or revolutionary change. Nonetheless, this flexibility may become a weakness as it provides greater opportunity for abuse of its changeability.
7. In setting the ‘rules of the game’, constitutions should be sufficiently stable in order to allow participants to anticipate their acts’ consequences, but, simultaneously, sufficiently flexible to allow future generations to respond to political, social, and legal developments (Mueller 387). Rigid constitution may hinder current majorities from updating the constitution with the changing times and circumstances (Dixon in Ginsburg and Dixon (eds) 97–98). In their study on constitutional endurance, Elkins, Ginsburg, and Melton propose that ‘constitutions are more likely to endure when they are flexible, detailed, and able to induce interest groups to invest in their processes’ (Elkins, Ginsburg, and Melton 89). Of course, there should be a balance between flexibility and rigidity since extreme flexibility undermines constitutionalism and carries an increased risk of demise (Elkins, Ginsburg, and Melton 82, 100, 140).
8. The main distinctions between rigid and flexible constitutions are therefore: first, in rigid constitutions, there is usually a special procedure to amend the constitution, which is lacking in flexible constitutions that can be amended as ordinary legislation. Second, a rigid constitution carries a higher normative status than ordinary laws, as opposed to a flexible constitution which has a similar status as ordinary laws. Third, a rigid constitution can usually be changed through a cumbersome procedure while a flexible constitution can be easily changed.
9. Bryce’s distinction between flexible and rigid constitutions remains paradigmatic in constitutional theory (see eg Edward 63, 73) yet it has somewhat been fading away. Today, other than the United Kingdom (‘UK’) and New Zealand, who lack formal constitutions, the category of flexible constitutions, as described by Bryce, is hardly useful in practice. Therefore nowadays, the concepts of rigid and flexible constitutions have been somewhat channelled to focus on the amendment formula itself. Constitutional rigidity is now mainly used to describe stringency of the amending procedure and the existence of eternity clauses (entrenched clauses), the rarity or difficulty of change, or to indicate slow-moving processes or procedural dysfunctionalities (Contiades and Fotiadou (2013) 459).
10. Comparative constitutional design demonstrates that there is no single unified method or process for amending constitutions. Constitutions feature dissimilar degrees of amenability in terms of the rigidity/flexibility scheme. For example, some are considered ‘flexible’ in that the amendment process is relatively easy, such as in the case of ordinary legislative majorities. Other constitutions are more ‘rigid’ in that they require high barriers, such as a supermajority threshold in parliament, higher quorum requirements, time delays, or additional requirements, including state ratification in federal systems (federalism), constituent assemblies, intervening elections, and popular referenda (direct democracy), to name a few (Albert (2014) 913; Lijphart 47–48; Elster 1010; Lane 114). All these hurdles may influence the rigidity or flexibility of a constitution. Of course, constitutional amendment procedures are not only functional—by balancing between rigidity and flexibility—but also symbolic, both by stipulating who has the legal sovereign authority to change the constitution, ie to exercise constituent power, and by revealing the deepest constitutional values of the polity (Albert 2017).
2. Delineation of Overlapping Concepts
(a) Written and Unwritten Constitutions
11. Although criticizing the old, classical distinction between written and unwritten constitutions, Bryce’s distinction between rigid and flexible constitutions bears some resemblance to it (Ludwikowski 3). A written constitution is sometimes a rigid document which defines citizens’ rights and the organization of the government, as well as constituting the sole source of constitutional law in a state. In contrast, unwritten constitutions are comprised of the country’s laws—such as constitutional statutes—enacted over time, integrated with an emphasis on political precedents. In the latter, there is no formally written document that refers to the power or limits of the government or that guarantees the protection of civil rights (Shigong 12–13). Of course, no system of government is embodied solely in written rules or solely in unwritten rules (Wolf-Phillips 12–13).
(b) Supreme and Subordinate Constitutions
12. In his book ‘Modern Constitutions’, Sir Kenneth Wheare extended Bryce’s criteria and suggested different lenses in order to examine and classify constitutions. Similarly to Bryce’s criteria, Wheare suggests that rigid or flexible relates to the method by which constitutions may be amended: a rigid constitution requires a unique method to be amended, whereas a flexible one does not require any special process in order to be amended. However, Wheare suggested another classification of supreme and subordinate constitutions, a subdivision of ‘rigid’ constitutions which is concerned with distinguishing between constitutions which can be amended by the legislature, meaning they are ‘subordinate’ to it, and those which cannot be amended at all by the legislature, and in this respect, have supremacy over it (Wheare 24−25; see also Wolf-Phillips 12–13).
B. List of Analysed Constitutions
13. Today, almost all states have formal written constitutions and, within them, constitutions amendment formulas are a universal component. About 96 per cent of the world’s constitutions include amendment provisions (Lutz (1994) 356; van Maarseveen and van der Tang 80). Very commonly, the legislature also possesses the amendment authority. In most amendment procedures, the legislature requires a greater majority to pass an amendment than the majority required for ordinary legislation. Various degrees of qualified majorities exist, ranging from an absolute majority (50 per cent of all members plus one) to a four-fifths majority of all members. The most common qualified-majority requirement are three-fifths or two-thirds of the membership (Böckenförde 6). There is a recent trend in constitutional design of enhancing the role of the people in constitutional change and formal constitutional amendments (Contiades and Fotiado (eds) (2017)). The constitutions analysed here would naturally be only a small number of possible constitutions for study, but they include constitutions of the following continents: America, Africa, Asia, Europe, and Australia, in order to give a broad perspective. The analysis also includes constitutions from both civil law and common law jurisdictions, and constitutions that involve the people through direct participatory mechanism and those that provide an exclusive role of the legislature in the process. Thus, for examples of rigid constitutions the focus is on the United States (‘US’) Constitution, which is often regarded as extremely rigid, on the French Constitution, and on the Australian Constitution. The UK and Israel—which both lack a single formal document described as the constitution—and Tanzania’s Constitution will be analysed as examples of flexible constitutions. India, Singapore, and Colombia will be analysed as states with a constitution which combines flexibility and rigidity.
C. Comparative Description
1. Rigid Constitutions
(a) The United States
14. The Constitution of the United States of America: 17 September 1787 (as Amended to 7 May 1992) (US) is considered one of the most rigid constitutions in the world (Levinson (2006) 21). The formal constitutional amendment process is stipulated in Article V of the US Constitution, which states two processes for its amendment: the first requires that two-thirds of both Houses of Congress propose an amendment, and that legislatures or special conventions in three-quarters of the states ratify it. The second method is when two-thirds of the legislatures of the states call for a convention to propose amendments, and three-quarters of the legislatures or special conventions in the states agree to the proposed amendments. This alternative constitutional convention route for amendments has never been used (Vile (2016)).
15. The indication for the difficulty of the amendment process of the US Constitution is the rate of failed proposed amendments. Since 1789, over 11,500 constitutional amendments were proposed, but Congress had only voted on 33 by the necessary two-thirds majorities, and the states have ratified only 27 of those, an astonishingly low percentage (Vile (2013) 396).
16. The rigidity of Article V was criticized for being undemocratic and for exacerbating the ‘dead hand’ difficulty (Dixon 655; Griffin (1995) 173). On the other hand, some have claimed that this rigidity created by the entrenchment of Article V was meaningful in helping the Constitution to survive during the early decades of the republic (Huq (2014) 1168).
17. With such hurdles on making formal constitutional changes, some fundamental constitutional changes were the result of judicial decisions by the Supreme Court of the United States, such as the New Deal cases regarding the power of the national government to spend money and regulate the economy (United States v Butler (1936) (US)), and fundamental rights cases, for example concerning the banning of racial segregation (racial discrimination) (Brown v Board of Education for Topeka Case (US) (1954)) (Griffin (2011) 363–364).
18. The Constitution of the French Republic: 28 September 1958 (as Amended to 23 July 2008) (Fr) is considered to be rigid, given its restrictive amendment clause (Mastor and Icher 115). Article 89 of the French Constitution involves four actors within the amendment procedure—the president, the National Assembly, the Senate, and the people:
[t]he President of the Republic, on the recommendation of the Prime Minister, and Members of Parliament alike shall have the right to initiate amendments to the Constitution. A Government or a Private Member’s Bill to amend the Constitution must be considered within the time limits set down in the third paragraph of article 42 and be passed by the two Houses in identical terms. The amendment shall take effect after approval by referendum. However, a Government Bill to amend the Constitution shall not be submitted to referendum where the President of the Republic decides to submit it to Parliament convened in Congress; the Government Bill to amend the Constitution shall then be approved only if it is passed by a three-fifths majority of the votes cast. The Bureau of the Congress shall be that of the National Assembly.
19. Additionally, the provision also includes temporal and substantive limits on constitutional amendment power: ‘[n]o amendment procedure shall be commenced or continued where the integrity of national territory is placed in jeopardy. The republican form of government shall not be the object of any amendment’ (Art. 89). The protection of the republican form of government from amendments existed already in older French Constitutions since 1884, and the temporal limitation is intended to avoid constitutional revision under pressure of occupation, like the experience of 10 July 1940 when constitutional power was handed to General Pétain. Article 7 adds an additional temporal limit: ‘[A]rticle 89 of the Constitution shall be implemented during the vacancy of the Presidency of the Republic or during the period between the declaration of the permanent incapacity of the President of the Republic and the election of his successor’.
20. In 1992, the Constitutional Council of France (Conseil Constitutionnel) declared that, subject to the temporal and substantive restrictions provided in the Constitution, the constituent power is sovereign (Decision No. 92-312 DC Maastricht II (1992) (Fr)). From this declaration it seemed clear that constitutional amendment must observe the substantive unamendability imposed in Article 89, and that such limits would be enforceable in the courts. However, in 2003 the Constitutional Council laconically declared that it has no competence to conduct judicial review of constitutional amendments (Decision No. 2003-469 DC (2003) (Fr)). The Constitutional Council therefore took a restrictive approach towards constitutional unamendability (Baranger).
21. Notwithstanding its rigidity, and in contrast with the US Constitution, the French Constitution, which was promulgated on 1958, has since been modified twenty-four times (Mastor and Icher 115). The number and extent of constitutional amendments make evident that there is no constitutional disablement to constitutional change in the French Fifth Republic (Boyron 115).
22. When considering constitutional amendment rates, the US is often regarded as having one of the most rigid amendment rules in the world. If that is so, then the Australian Constitution should also be regarded as an extremely rigid constitution, as cross-national data identified it as the only country with a lower amendment rate than the US (Lutz (2006) 170) (Commonwealth of Australia Constitution Act: 9 July 1900 (as Amended to 31 October 1986) (Austl)). Donald Lutz, comparing the Australian Constitution to that of another parliamentary country—the New Zealand Constitution, which is probably the easiest in the world to amend as it can be amended by an act of a unicameral parliament—states that ‘Australia … largely because it is federal, has probably the most difficult constitution in the world to amend because of its much stronger separation of powers as a whole’ (Lutz (2006) 125). The formal amendment procedure is prescribed in Section 128 of the Australian Constitution, which is consistent with the federal framework and reflects a popular notion in which the people should have the final say on formal constitutional change. Accordingly, the procedure divides the process between the federal Parliament, the people of the states, and the Australian people. According to Section 128, the Parliament has the exclusive authority to initiate amendments, which must be approved by both Houses of Parliament by an absolute majority, or twice in one House after an interval of three months. After the proposal successfully passes the Parliament (in ‘not less than two nor more than six months’), the proposal is then submitted to the electors qualified to vote in federal elections, for approval by popular referendum. In order to be adopted, the proposed amendment must be approved by a majority of the electors in the majority of the states, and by an overall majority of all electors. If approved at a referendum, it shall take effect upon receiving the Queen’s assent from the Governor-General. Whereas the Australian Constitution is over 115 years old, there were only eight successful referendums, out of forty-four proposals put to referendum (Kenny; Weis 255–6). Accordingly, this double-majority process, which is not as onerous as super-majority decision-making rules, is still quite rigid.
23. With rigid constitutions, the locus of constitutional change often shifts to the judiciary which, outside the formal amendment process, brings about constitutional changes that some might view as constitutional amendments. In the US context, for example, judicial decisions are commonly viewed as instruments of constitutional change and some scholars have even suggested that certain judicial interpretations of the US Constitution are best understood as constitutional amendments (Coudert; Levinson (1995) 33). Likewise, in Australia, albeit the lack of express guarantee of freedom of expression in the constitution, since 1992 the High Court has acknowledged a limited implicit guarantee of freedom of political expression (Australian Capital Television Pty Ltd v Commonwealth (1992) (Austl)). This judicial decision ‘made a substantial change to Australia’s system of government, which to many, was more like a constitutional amendment than the discovery of a genuine implication’ (Goldsworthy 704).
2. Flexible Constitutions
(a) The United Kingdom
24. The constitution of the UK is considered flexible (even overly flexible). The UK lacks a formal and entrenched document entitled ‘the constitution’, legal changes even to the basic aspects of the constitutional and political order can take place through the same process as ordinary legislation of Parliament, and conventions remain an informal means of changing the constitution. Thus, the term constitution in the UK context refers to ‘the set of laws, rules and practices that create the basic institutions of the state, and its component and related parts, and stipulate the powers of those institutions and the relationship between the different institutions and between those institutions and the individual’ (Blackburn 359). It is an uncodified constitution.
25. The procedures by which the constitution is formally amended involves a normal legislative process: a bill is being presented in one of the two Houses of the Parliament for a First Reading; then a Second Reading debate follows at a later date in which the general principles of the bill are debated and voted on; then a bill goes into a committee stage in which its clauses are scrutinized one by one and amendments may be made. A final Third Reading then follows, after which the bill goes to the House of Lords. The Royal Assent—which is a formality—then follows. No matter how fundamental the nature of a bill, there is no special majority requirement in either House and a simple majority is all that is required. Amendments to the law governing succession to the Crown are subject to a special process which is that the Commonwealth realms, where the monarch serves as Head of States, must be consulted and indicate their agreement to the proposed changes. Referendums are not a general requirement for constitutional amendments although few have taken place on an ad hoc basis and they are only consultative in nature (Blackburn 366–370; Oliver 329–355).
26. Indeed, in 1998, the enactment of the Human Rights Act had a profound effect on the law of the UK, most particularly on judicial review (Gordon and Ward). However, although the act now allows courts to issue declarations of incompatibility of legislation with the European Convention for the Protection of Human Rights and Fundamental Freedoms (1950), thereby entrenching, in a way, fundamental rights, courts cannot invalidate statutes, which retain their validity notwithstanding any declaration of incompatibility. Thus, in effect, parliamentary sovereignty remains a fundamental constitutional principle in the UK (Young).
27. The UK is therefore the ‘textbook example’ of a flexible constitution, where any Act of Parliament can change the constitution, there is no procedural or formal difference between constitutional and ordinary legislation or minor or major constitutional changes, and no practice has evolved as a convention for introducing constitutional changes (Contiades and Fotiadou (2013) 437).
28. The Israeli constitutional regime was based primarily on the constitutional structure of the British regime, since during the years before the establishment of the state of Israel the Westminster Parliament served as a model for leaders of the pre-state Jewish settlement. After the declaration of independence, there were far-reaching debates regarding the constitution. The opposition to the adoption of a constitution claimed that the legacy of the British tradition rendered a constitution unnecessary. The 1950 ‘Harari Resolution’ laid the foundations for a compromise according to which the constitutional process will continue as an incremental enterprise whereby the Knesset (the Israeli Parliament) would enact the constitution in stages through a series of distinct Basic Laws. The common denominator of all the Basic Laws was the addition of the word ‘basic’ to their title. Although the Knesset’s intention was to place the Basic Laws normatively above regular acts of law, the Knesset failed to make a clear distinction between a Basic Law and a regular law. Most of the Basic Laws were adopted in accordance with the procedure for adopting regular legislation. In other words, the Knesset holds both ordinary legislative and constituent powers, and both set of laws—ordinary and constitutional—are enacted and amended in a similar process. Until 1992, the normative status of the Basic Laws was equal to that of normal laws and, with one exception, the Basic Laws were not entrenched and did not require any special majority for their amendment. The only exception is Basic Law: The Knesset: 12 February 1958 (as Amended to January 2003) (Isr) which provided a special majority of sixty-one (out of 120) for amending certain sections. In 1992, two new basic laws on human rights were enacted: Basic Law: Human Dignity and Liberty, and Basic Law: Freedom of Occupation. Those two Basic Laws introduced two major changes in Israeli constitutionalism. First, the Basic Law: Freedom of Occupation included a formal requirement for a majority of sixty-one members of the Knesset (‘MKs’) in order to be amended. Second, and more importantly, the two new Basic Laws included a substantive ‘limitation clause’ that imposed restrictions on the possibility of violating human rights. Apart from these exceptions, the formal procedure for voting in Basic Laws is equal to that of a regular law and the formal process of enacting them does not demand any special procedure (Navot 191–209).
29. In 1995, the Supreme Court of Israel (Beit Ha Mishpat Ha Elyon) gave its decision in the matter of Mizrahi Bank Case (Isr) and ruled that the limitation clause is meant to limit the Knesset’s powers as a legislative authority. Moreover, this judgment declared the supremacy of Basic Laws by holding that a Basic Law may not be changed except by another Basic Law. Even if the Basic Law does not contain ‘rigidity’, it is not in a similar normative status as ordinary laws but carries a superior constitutional status and therefore requires a Basic Law for its amendment (United Mizrahi Bank Ltd v Migdal Cooperative Village (1995) (Isr)).
30. Therefore, like the UK, Israel has a flexible constitution. Basic Laws can be amended through a similar process like ordinary legislation and without a special procedure. However, unlike the UK, Basic Laws carry a higher normative status than ordinary laws, and can be amended only by other Basic Laws. Moreover, Basic Law: Freedom of Occupation and a limited number of sections within other Basic Laws require a majority of MKs for their amendment, ie a special procedure for change.
31. Many of the African modern constitutions require, in their amendment procedure, that amendments are approved by a supermajority vote of two-thirds, three-fifths, or four-fifths of a parliament. The Constitution of the United Republic of Tanzania: 26 April 1977 (as Amended to 2005) (Tanz), for example, requires for initiating constitutional amendments, ordinary law-making procedure, and for adoption of amendments. It includes two different procedures for two categories of provisions: for one category, two-thirds majority in Parliament; and for the second category, two-thirds majority in Parliament and two-thirds members of Parliament from Zanzibar (Art. 98). It therefore makes a procedural distinction for amending constitutional provisions considered as particularly important or fundamental (Fombad (2007) 58).
32. Regarding such procedural restrictions on constitutional amendments, one would place almost all African constitutions in the semi-rigid category (Fombad (2011) 1018). Nevertheless, many African countries are ruled by dominant parties which have more than two-thirds of the seats in parliament. In such cases, the two-thirds majority requirement for constitutional amendments is not a major obstacle to amending the constitution, and in many of the cases even a three-quarters majority requirement for constitutional amendment can be easily fulfilled by dominant parties (Fombad (2013) 403–404).
33. This is the case in Tanzania. For example, in the elections of 2005, the ruling party (Chama Cha Mapinduzi) occupied 186 of the total 239 number of seats in Parliament, ie, the ruling party occupied 77.8 per cent of Parliament’s seats (Fombad (2013) 403–404), easily overcoming the two-thirds requirement, which makes the Tanzanian Constitution de-facto very flexible (on the history of Party Supremacy in Tanzania see Ong’wamuhana).
34. This flexibility is evident in the high amendment rate. Although since Tanzania’s independence the Chama Cha Mapinduzi has operated under one constitution, without replacement, from 1977 the Constitution, which consists of 152 articles, had 100 amendments (Okubasu 332). This caused Fombad to raise the following question: ‘with such frequent and radical piecemeal amendments, one wonders whether there can be either constitutional stability or coherence in the text’ (Fombad (2013) 411).
3. Combination of Rigidity and Flexibility
35. In spite of the dichotomy between the terms ‘rigid’ and ‘flexible’ constitutions, in some cases constitutions may contain a combination of rigid and flexible procedural requirements and may be seen as ‘flexible-rigidity’ constitutions. The Indian Constitution is a good example for this type of constitution (Constitution of the Republic of India: 26 January 1950 (as Amended to 28 May 2015) (India)).
36. India’s constitution is very sophisticated as it includes a variety of procedures for amendment based upon the careful selection of different matters. Thus, it can be seen as rigid and also as flexible, depending on the subject matter of the change. Most amendments require the support of a majority of all members and a two-thirds supermajority of those members ‘present and voting’ in both Houses of Parliament. Nevertheless, the Constitution also establishes an additional requirement of ratification by a majority of state legislatures for amendments affecting representation of the states in Parliament, and provisions concerning the election of the president, executive and legislative power, the union judiciary, High Courts in union territories and the states, and the lists of federal, state, and concurrent responsibilities found in the constitution’s Seventh Schedule (Constitution of India, Art. 268; Singh (2011)). Moreover, the Supreme Court of India has created an additional judge-made limit to constitutional amendment through its ‘basic structure doctrine’ according to which the power to amend the constitution does not include the power to alter the basic structure, or framework, of the constitution so as to change its identity (Kesavananda Bharati v State of Kerala (1973) (India); Krishnaswamy (2009)). Since its adoption in 1950, and as of 2016, the Indian Constitution had been formally amended 101 times. In their study on constitutional endurance, Elkins, Ginsburg, and Melton note that India’s amendment procedure appears to be ‘just right’ on their amendment flexibility scale (Elkins, Ginsburg, and Melton 141).
37. Another example of a combination of rigidity and flexibility is the Constitution of the Republic of Singapore: 9 August 1965 (as Amended to 2016) (Sing) (Tan 45–47). Singapore is a former colony of Great Britain with a written constitution. Article 90 of the State of Singapore Constitution of 1963 required the support of no less than two-thirds of the total number of Members in order to amend the constitution. When Singapore was separated from Malaysia and became independent in 1965, it adopted a flexible amendment regime so that a simple majority can amend the constitution. In 1979, the amendment formula changed—and was actually restored—to a two-thirds majority (Singapore Const. Art. 5(2)). However, the amendment formula is more complex than that. While constitutional provisions can be amended through a two-thirds majority, some provisions require a higher threshold. In 1972, the constitution was amended, and a provision was inserted requiring a national referendum, by no less than two-thirds of the total number of votes, in order to amend Part III of the Constitution, dealing with surrender of state sovereignty and armed forces (Singapore Const. Art. 8(1)). In 1992 and 1996 respectively, Article 5(2A) and Article 5A were inserted into the constitution and introduced complex amendment procedures regarding core constitutional provisions, such as fundamental liberties and the office of the president and its discretions; yet both Articles have never been brought into force (Neo and Lee 162–3).
38. A final example of a complex and combined amendment formula may be found in Colombia, the 1991 constitution of which includes an ‘escalating structure’ of amendment with multiple procedures, in addition to a judicially-made implicit unamendability (Constitution of the Republic of Colombia: 5 July 1991 (as Amended to 4 November 2011) (Colom)).
39. The multiple procedures are prescribed in Articles 374–378. According to Article 374 of the Colombian Constitution: ‘[t]he Political Constitution may be reformed by Congress, a Constituent Assembly, or by the people through a referendum’. The legislative majority required by Article 375 in order to pass a constitutional amendment is not demanding to obtain; an amendment ‘must be approved in two ordinary and consecutive periods. Following approval in the first period by the majority of those present, the proposal will be published by the government. In the second period, approval will require the vote of the majority of the members of each chamber’. Article 376 provides a mechanism of a referendum by stating that: ‘[b]y means of a law approved by the members of both chambers, Congress may stipulate that the people decide by popular vote if a Constituent Assembly should be called with the jurisdiction, term, and members determined by that same law’. The people can then convoke the constituent assembly by an approval of at least one-third of the electoral rolls. And Article 377 stipulates that if a constitutional reform refers to rights recognized in Chapter I of Title II and to their guarantees, to the procedures of popular participation (public participation), or to Congress, they ‘must be submitted to a referendum approved by Congress’. A referendum can also be called by Congress to approve a constitutional reform, through a law which mandates the approval of the majority of the members of both Chambers (Art. 378).
40. Although the text of the Constitution does not include explicit limits to constitutional amendments, and expressly empowers the Constitutional Court to review constitutional reform ‘only for errors of procedure in their formation’ (Arts 241 and 379), the Constitutional Court of Colombia (Corte Constitucional de Colombia) developed the ‘constitutional replacement’ doctrine, according to which:
[t]here is a difference, then, between the amendment of the Constitution and its replacement. Indeed, the reform that is incumbent upon Congress may contradict the content of constitutional norms, even drastically, since any reform implies transformation. However, the change should not be so radical as to replace the constitutional model currently in force or lead to the replacement of a ‘defining axis of the identity of the Constitution’, with another which is ‘opposite or completely different (Decision C-1040/2005 (2005) (Colom)).
Accordingly, the Constitutional Court has held that it possesses the power to review constitutional reforms in order to examine whether the constitutional amendment authority is acting within its jurisdiction (ie nearly modifying the constitution) or in fact replacing the constitution with a new one (Bernal; Colon-Rios; Ramirez-Cleves; Cajas-Sarria). This form of implied unamendability doctrine adds an element of rigidity to the constitutional amendment process that is non-existing in the text.
D. Critical Assessment
41. Since most countries now have formal written constitutions, the concept of flexible/rigid constitutions is now mostly used to describe only one of Bryce’s features: the relative ease or difficulty of constitutional change. However, there are various complications in describing a constitution as rigid or flexible. Two of the complications are selective rigidity and amendment culture.
1. Selective Rigidity
42. Constitutions have different procedures for constitutional amendments. Not only do entire constitutions differ from one another in the mechanisms, actors, and procedures involved in the amendment process, but also a specific constitution might incorporate dissimilar procedures for amending different provisions and principles. Like the aforementioned Indian example, some constitutions incorporate different procedures for constitutional amendments of different constitutional subjects. For example, in Canada, the Constitution Act, 1982 creates an escalating structure of five formal amendment rules that impose increasingly demanding thresholds (Part V of the Canada Constitution Act, 1982 (Constitution of Canada: the Constitution Acts 1867 to 1982 (Unofficial consolidation current as of September 2008) (Can)); Albert 2015). Likewise, the South African Constitution stipulates that constitutional provisions can be amended with the support of two-thirds of the National Assembly in addition to six provisional legislature—if the amendment relates to a matter that affects the National Council, alters provincial boundaries, powers, functions or institutions, or deals specifically with a provincial matter. However, amending the ‘Founding Provisions’ of the constitution requires a 75 per cent majority as well as six of provisional legislatures (Constitution of the Republic of South Africa: 16 December 1996 (as Amended to 1 February 2013) (S Afr) Art. 74; Corder 270).
43. The idea behind selective rigidity is that provisions that are deemed more fundamental or protection-worthy are more difficult to amend and enjoy special protection. It allows greater protection of the core parts of the democratic order and thereby reduces the possibilities of abusing the amendment process. These procedures may include increased qualified majorities in parliament, referendums or intervening elections, or the convening of a special constituent assembly. Many even provide that certain provisions cannot be amended at all. In other words, they are unamendable. At the same time, non-fundamental principles or rules can be easily amended (Albert (2010) 707–10; Landau; Roznai).
44. Dixon and Landau term this ‘tiered constitutional design’. Under tiered constitutionalism, the default rule of constitutional amendment is a flexible one as most provisions are made fairly easy to change. However, certain constitutional provisions or principles are placed on a higher tier, are given higher levels of entrenchment, and are therefore made more difficult to amend. This ‘tier’ constitutional design can combine the best of both forms of constitutionalism. Since most provisions can be changed easily, the constitution can be updated as needs arise, a virtue of flexible design. At the same time, enhanced protection of a core set of provisions can provide stability for the core of the constitution, and protect against anti-democratic or other destabilizing forms of constitutional change (Dixon and Landau).
2. Amendment Culture
46. Rigid/flexible distinction focuses on amendment procedure. Political scientists have attempted to measure constitutional rigidity by focusing on amendment rates and amendment procedures and hurdles (Lutz (1994); Lijphart; Lorenz).
47. Yet, formal amendment rules are not the only determining factor with regard to constitutional change. What may matter even more to the rigidity or flexibility of a constitution than voting thresholds or temporal limits is the amendment culture in a constitutional tradition, meaning ‘the set of shared attitudes about the desirability of amendment, independent of the substantive issue under consideration and the degree of pressure for change’ (Ginsburg and Melton 699). In other words, constitutional rigidity is not merely institutional but also attitudinal and factual (Contiades and Fotiadou (2013) 458–459). Take, for example, the UK, which is considered a flexible constitution in light of the amendment procedure. It is in fact a rigid constitution, and its rigidity stems from the constitutional culture which obliges actors to show self-restraint in handling constitutional matters (Contiades and Fotiadou (2016)). At the end, measuring amendment difficulty is a problematic task and formal amendment rules do not sufficiently explain the rigidity or flexibility of a constitution.
Albert, R, ‘Constitutional Handcuffs’ (2010) 42 Arizona State Law Journal 663.
Albert, R, ‘The Difficulty of Constitutional Amendment in Canada’ (2015) 53 Alberta Law Review 85.
Albert, R, ‘The State of the Art in Constitutional Amendment’ in Albert, R, Contiades, X, and Fotiadou, A, (eds), The Foundations and Traditions of Constitutional Amendment (Hart Publishing 2017) 1.
Albert, R, ‘The Structure of Constitutional Amendment Rules’ (2014) 49 Wake Forest Law Review 913.
Baranger, D, ‘The Language of Eternity: Constitutional Review of the Amending Power in France (Or the Absence Thereof)’ (2011) 44 IsLR 389.
Bernal, C, ‘Unconstitutional Constitutional Amendments in the Case Study of Colombia: An Analysis of the Justification and Meaning of the Constitutional Replacement Doctrine’ (2013) 11 ICON 339.
Blackburn, R, ‘Constitutional Amendment in the United Kingdom’ in Contiades, X, (ed.), Engineering Constitutional Change: A Comparative Perspective on Europe, Canada and the USA (Routledge 2013) 359.
Böckenförde, M, Constitutional Amendment Procedures: International IDEA Constitution-Building Primer 10 (International IDEA 2017).
Boyron, S, ‘France’ in Oliver, D, and Fusaro, C, (eds), How Constitutions Change: A Comparative Study (Hart Publishing 2011) 115.
Bryce, J, ‘Flexible and Rigid Constitutions’ in Studies in History and Jurisprudence vol. I (1901) 124.
Bryce, J, Constitutions (OUP 1905).
Cajas-Sarria, MA, ‘Judicial Review of Constitutional Amendments in Colombia: A Political and Historical Perspective, 1955–2016’ (2017) 5 The Theory and Practice of Legislation 245.
Colon-Rios, J, ‘Beyond Parliamentary Sovereignty and Judicial Supremacy: The Doctrine of Implicit Limits to Constitutional Reform in Latin America’ (2013) 44 Victoria University of Wellington Law Review 521.
Contiades, X, and Fotiadou, A, (eds), Participatory Constitutional Change: The People as Amenders of the Constitution (Routledge 2017).
Contiades, X, and Fotiadou, A, ‘Models of Constitutional Change’ in Contiades, X, (ed.), Engineering Constitutional Change: A Comparative Perspective on Europe, Canada and the USA (Routledge 2013) 417.
Contiades, X, and Fotiadou, A, ‘The Determinants of Constitutional Amendability: Amendment Models or Amendment Culture? (2016) 12 EuConst 192.
Corder, H, ‘The Republic of South Africa’ in Oliver, D, and Fusaro, C, (eds), How Constitutions Change: A Comparative Study (Hart Publishing 2011) 261.
Coudert, FR, ‘Judicial Constitutional Amendment’ (1904) 13 YaleLJ 331.
Dixon, R, ‘Constitutional Amendment Rules: A Comparative Perspective’ in Ginsburg, T, and Dixon, R, (eds), Comparative Constitutional Law (Edward Elgar Publishing 2011) 96.
Dixon, R, ‘Partial Constitutional Amendments’ (2011) 13 University of Pennsylvania Journal of Constitutional Law 643.
Dixon, R, and Landau, D, ‘Tiered Constitutional Design’ (2018) 86 GWashLRev (forthcoming 2018).
Edward, D, ‘The Community’s Constitution—Rigid or Flexible? The Contemporary Relevance of the Constitutional Thinking of James Bruce’ in Curtin, DM, and Heukels, T, (eds), Institutional Dynamics of European Integration: Essays in Honour of Henry G Schermers (Brill 1994) 63.
Elkins, Z, Ginsburg, T, and Melton, J, The Endurance of National Constitutions (CUP 2009).
Elster, J, Ulysses Unbound: Studies in Rationality, Precommitment, and Constraints (CUP 2000).
Fombad, CM, ‘Constitutional Reforms and Constitutionalism in Africa: Reflections on Some Current Challenges and Future Prospects’ (2011) 59 Buffalo Law Review 1007.
Fombad, CM, ‘Limits on the Power to Amend Constitutions: Recent Trends in Africa and Their Potential Impact on Constitutionalism’ (2007) 6 University of Botswana Law Journal 27.
Fombad, CM, ‘Some Perspectives on Durability and Change under Modern African Constitutions’ (2013) 11 ICON 382.
Ginsburg, T, and Melton, J, ‘Does the Constitutional Amendment Rule Matter at All? Amendment Cultures and the Challenges of Measuring Amendment Difficulty’ (2015) 13 ICON 686.
Goldsworthy, J, ‘Constitutional Cultures, Democracy, and Unwritten Principles’ (2012) University of Illinois Law Review 683.
Gordon, R, and Ward, T, Judicial Review and the Human Rights Act (Routledge 2013).
Griffin, SM, ‘The Nominee Is … Article V’ (1995) 12 Constitutional Commentary 171.
Griffin, SM, ‘The United States of America’ in Oliver, D, and Fusaro, C, (eds), How Constitutions Change: A Comparative Study (Hart Publishing 2011) 357.
Huq, AZ, ‘The Function of Article V’ (2014) 162 UPaLRev 1165.
Kelsen, H, Pure Theory of Law (tr. from the 2nd edn by Knight, M, The Lawbook Exchange 2009).
Kenny, S, ‘Evolution’ in Saunders, C, and Stone, A, (eds), The Oxford Handbook of the Australian Constitution (OUP 2018) 119.
Krishnaswamy, S, Democracy and Constitutionalism in India: A Study of the Basic Structure Doctrine (OUP 2009).
Landau, D, ‘Selective Entrenchment in State Constitutional Law: Lessons from Comparative Experience’ (2016)69 Arkansas Law Review 425.
Lane, J-E, Constitutions and Political Theory (Manchester University Press 1996).
Levinson, S, ‘How Many Times Has the United States Constitution Been Amended? (A) < 26; (B) 26; (C) 27; (D) > 27: Accounting for Constitutional Change’ in Levinson, S, (ed.) Responding to Imperfection: The Theory and Practice of Constitutional Amendment (Princeton University Press 1995) 13.
Levinson, S, Our Undemocratic Constitution (OUP 2006).
Lijphart, A, Patterns of Democracy (2nd edn Yale University Press 2010).
Lorenz, A, ‘How to Measure Constitutional Rigidity—Four Concepts and Two Alternatives’ (2005)17 Journal of Theoretical Politics 339.
Ludwikowski, RR, ‘‘Mixed’ Constitutions: Product of an East-Central European Constitutional Melting Pot’ (1998)16 BostonUIntlLJ 1.
Lutz, DS, ‘Toward a Theory of Constitutional Amendment’ (1994)88 AmPolSciRev 355.
Lutz, DS, Principles of Constitutional Design (CUP 2006).
Mastor, W, and Icher, L, ‘Constitutional Amendment in France’ in Contiades, X, (ed.), Engineering Constitutional Change: A Comparative Perspective on Europe, Canada and the USA (Routledge 2013) 115.
Mueller, DC, ‘On Amending Constitutions’ (1999)10 Constitutional Political Economy 385.
Navot, S, ‘Israel’ in Oliver, D, and Fusaro, C, (eds), How Constitutions Change: A Comparative Study (Hart Publishing 2011) 191.
Neo, JL-C, and Lee, YCL, ‘Constitutional Supremacy—Still a Little Dicey?’ in Thio, L-a, and Tan, KYL, Evolution of a Revolution: Forty Years of the Singapore Constitution (Routledge 2014) 153.
Okubasu, D, ‘The Implication of Conflation of Normal and ‘Constitutional Politics’ on Constitutional Change in Africa’ in Albert, R, Contiades, X, and Fotiadou, A, (eds), The Foundations and Traditions of Constitutional Amendment (Hart Publishing 2017) 327.
Oliver, D, ‘The United Kingdom’ in Oliver, D, and Fusaro, C, (eds), How Constitutions Change: A Comparative Study (Hart Publishing 2011) 329.
Ong’wamuhana, K, ‘Party Supremacy and the State Constitution in Africa’s One-Party States: The Kenya-Tanzania Experience’ (1988) Third World Legal Studies 77.
Pasquino, P, ‘Classifying Constitutions: Preliminary Conceptual Analysis’ (2012)34 Cardozo Law Review 999.
Pasquino, P, ‘Flexible and Rigid Constitutions: A Post-Kelsenian Typology of Constitutional Systems’ in López-Guerra, C, and Maskivker, J, (eds), Rationality, Democracy, and Justice: The Legacy of Jon Elster (CUP 2015) 85.
Ramirez-Cleves, GA, ‘The Unconstitutionality of Constitutional Amendments in Colombia: The Tension between Majoritarian Democracy and Constitutional Democracy’ in Bustamante, T, and Gonçalves Fernandes, B, (eds) Democratizing Constitutional Law: Perspectives on Legal Theory and the Legitimacy of Constitutionalism (Springer 2016) 213.
Roznai, Y, Unconstitutional Constitutional Amendments—The Limits of Amendment Powers (OUP 2017).
Shigong, J, ‘Written and Unwritten Constitutions: A New Approach to the Study of Constitutional Government in China’ (2010)36 Modern China 12.
Singh, MP, ‘India’ in Oliver, D, and Fusaro, C, (eds), How Constitutions Change: A Comparative Study (Hart Publishing 2011) 169.
Tan, KYL, The Constitution of Singapore: A Contextual Analysis (Bloomsbury Publishing 2015).
van Maarseveen, H, and van der Tang, G, Written Constitutions: A Computerized Comparative Study (Brill 1978).
Vile, JR, ‘Constitutional Revision in the United States of America’ in Contiades, X, (ed.), Engineering Constitutional Change: A Comparative Perspective on Europe, Canada and the USA (Routledge 2013) 389.
Vile, JR, Conventional Wisdom: The Alternate Article V Mechanism for Proposing Amendments to the US Constitution (University of Georgia Press 2016).
Weis, LK, ‘Constituting ‘the People’—The Paradoxical Place of the Formal Amendment Procedure in Australian Constitutionalism’ in Albert, R, Contiades, X, and Fotiadou, A, (eds), The Foundations and Traditions of Constitutional Amendment (Hart Publishing 2017) 253.
Wheare, K, Modern Constitutions (2nd edn Oxford University Press 1966).
Wolf-Phillips, L, Comparative Constitutions (Palgrave Macmillan 1972).
Young, AL, Parliamentary Sovereignty and the Human Rights Act (Bloomsbury Publishing 2008).