Oliver Ellsworth Research Professor and Wallace Stevens Professor Emeritus, University of Connecticut School of Law. Anastasiya Collins provided essential research assistance.
The American Journal of Comparative Law, Volume 66, Issue suppl_1, July 2018, Pages 243–268, https://doi.org/10.1093/ajcl/avy019
31 May 2018Richard S Kay, Formal and Informal Amendment of the United States Constitution, The American Journal of Comparative Law, Volume 66, Issue suppl_1, July 2018, Pages 243–268, https://doi.org/10.1093/ajcl/avy019
Navbar Search Filter Mobile Enter search term Search Navbar Search Filter Enter search term SearchThe United States Constitution became effective in 1789. After 228 years, it is touted as “the world’s longest surviving written charter of government.” 1 More remarkably, notwithstanding that it is sometimes credited with inventing the very idea of machinery for modification of a constitutional text, 2 it has continued in much the same form as when ratified. There have been twenty-seven amendments, but really it has “been amended” only eighteen times. 3 The first ten amendments, dealing with individual rights, were presented to the states by the first Congress in a single package, fulfilling a pledge made to induce ratification of the unamended text. 4 In Article V, the Constitution sets out the following rules for its own amendment:
The Congress, whenever two thirds of both houses shall deem it necessary, shall propose amendments to this Constitution, or, on the application of the legislatures of two thirds of the several states, shall call a convention for proposing amendments, which, in either case, shall be valid to all intents and purposes, as part of this Constitution, when ratified by the legislatures of three fourths of the several states, or by conventions in three fourths thereof, as the one or the other mode of ratification may be proposed by the Congress; provided that no amendment which may be made prior to the year one thousand eight hundred and eight shall in any manner affect the first and fourth clauses in the ninth section of the first article; and that no state, without its consent, shall be deprived of its equal suffrage in the Senate. 5
These rules were proposed by the Philadelphia Convention of 1787 after considering an alternative providing that when two-thirds of the states called for amendment, the national legislature would “call a convention for that purpose.” That procedure intentionally excluded any substantive role for Congress because abuses by that body might be the reason that amendment was desired. After Alexander Hamilton pointed out that any need for modification would likely be first noticed by Congress, the Convention provided that Congress could propose amendments on the approval of two-thirds of its members or when it received applications from two-thirds of the state legislatures. Such proposals would then be ratified by three-fourths of the state legislatures or by state conventions with the choice of method left to Congress. Now, however, Congress’s role threatened to frustrate amendments aimed at its own oppressive behavior. The Convention, therefore, added a second method of proposing amendments: Congress would be obliged to call a national convention for that purpose when requested by two-thirds of the states. With the approval of this option, the final framework of Article V was established. 6 With two ways to propose amendments and two ways to ratify them, there were four separate sequences for enacting amendments. In fact, twenty-six amendments have been approved using only one method—proposal by Congress and ratification by state legislatures. Only the Twenty-First Amendment, repealing the Eighteenth (Prohibition) Amendment was proposed by Congress but approved by state conventions. There has never been a national constitutional convention, one for proposing amendments.
While Article V’s reference to “two-thirds of both houses” might mean two-thirds of the combined membership of the two bodies, Congress has uniformly treated it as calling for a two-thirds vote in each house. 7 The question of what a state legislature must do to make an effective ratification has been left to the legislators in each state. 8 As will be examined below, the Article tells us little about the kind of membership or procedures that would be required for a deliberative body to qualify as a national “convention for proposing amendments.” 9
Amendments to the Constitution are appended to the original text as additional articles. Article V merely states that after ratification, such amendments shall be “valid, to all intents and purposes, as part of this Constitution.” In fact, when James Madison presented the first versions of the proposed rights amendments to the House of Representatives in 1789, he specified places in the original text that would be changed or supplemented. He thought this would assure a certain “neatness and propriety,” allowing a reader to see the amended meaning “without references or comparison.” Others, however, objected that “interweaving” the changes distorted and confused the work of the original enactors. 10 Amendment, according to one delegate, was “the act of the state governments” and they lacked authority to alter the Constitution which was “the act of the people, and ought to remain entire.” 11 After some hesitation, the House agreed with this position, adopting the format that would be followed for all subsequent amendments. 12
Since amendment procedures create constitutional rules, they have often been considered in connection with the authority to make constitutions in the first place. In designing the amendment formula, the constitution makers might have aimed “to share some of their authority . . . with subsequent generations.” 13 In that case, the amendment procedure would reflect “roughly the same level of popular sovereignty as that used in the adoption of the Constitution.” 14 The identification of the sovereign constituent authority in the United States is complicated by the central role of federalism. Like almost every federal constitution, Article V amendment incorporates the division between central and local power. 15 The drafting history of Article V, outlined above, 16 shows the attempt to balance these interests. The result justifies James Madison’s description of it as “neither wholly federal nor wholly national.” 17 It provided one route to amendment that largely bypassed the national government: Congress would be obliged to call a national convention on the application of a sufficient number of states and to submit the proposals of that convention to the states for ratification. But the state governments might also be bypassed insofar as Congress could formulate a proposal and submit its ratification not to the state legislatures, but to specially elected state conventions. 18
The use of conventions, state and national, shows another way Article V borrows the assumptions associated with constitution making. Unlike many modern constitutions, the Article has no provision for direct recourse to the approval of the governed population. 19 When the United States Constitution was created, the use of plebiscites to measure the assent of “the people” was largely unknown. 20 It went more or less without saying for the American founders that “the people” would express themselves only in extraordinary conventions. Speaking of the people’s constituent act, Chief Justice Marshall said that “[t]he people acted upon it in the only manner in which they can act safely and effectively and wisely, on such a subject, by assembling in Convention.” 21 This was “a method corresponding to the original organization which proposed the Constitution itself.” 22 The availability of conventions as part of the amendment-making machinery is, therefore, a recognition of the underlying authority of the people as pouvoir constituent even if, when acting in the amendment process, it is also a pouvoir constituent dérivé. 23 At the Philadelphia convention, Alexander Hamilton declared congressional power to submit amendments to state conventions was safe insofar as “the people would finally decide.” 24 Madison, in The Federalist No. 49, likewise, referred to a convention as a “recurrence to” or an “appeal to” “the people.” 25
As noted, ratification by state conventions has been adopted only once, in connection with the Twenty-First Amendment repealing the Eighteenth (Prohibition) Amendment. At that time, much improvisation was necessary with respect to the selection of delegates and the procedures followed. 26 Although there have been several occasions on which state legislatures have petitioned Congress for it, no national convention has ever been held. Numerous questions have been raised about the circumstances in which Congress would be obliged to call one and the extent of its powers once assembled. 27 It is unclear whether state petitions must specify a particular subject of amendment or call for a general re-examination of the Constitution. 28 It is similarly disputed whether or not the necessary state requests must be similar and if so, in what degree. There is disagreement as to how much control Congress or the states can exercise over a convention with respect to its subject matter or its procedure. 29 Even apart from these questions, the prospect of a national convention has caused considerable anxiety in political observers. 30 The biggest fear is that such a convention, sensing the plenary power of the sovereign people, would ignore any agreed limitations of procedure or subject matter. It would become a “runaway” convention, one not unlike the Philadelphia Convention of 1787. 31
Article V’s rules for constitutional change have been described as unusually demanding. 32 As early as 1788, in the Virginia ratifying convention, Patrick Henry—who thought the new constitution would need some prompt improvement—complained about the “destructive and mischievous” requirement that amendments be approved by three-fourths of the state legislatures. This would demand “genius, intelligence and integrity, approaching to miraculous.” 33 Woodrow Wilson, in his book, Congressional Government, published in 1885, argued that “no impulse short of the impulse of self-preservation, no force less than the force of revolution, can nowadays be expected to move the cumbrous machinery of formal amendment.” 34 A modern commentator calls the Article “almost comically complex.” 35 In fact, about 12,000 proposals for amendment have been introduced in Congress but only thirty-three have been submitted to the states for ratification. 36 A number of studies have confirmed this conclusion by assembling and comparing data on methods and experiences under constitutions in many jurisdictions. The United States is always at or near the top of the list of states with the most difficult amending procedures. 37
The very nature of an “entrenched” constitution requires that it be harder to change than ordinary law. It follows that constitutional decision making should require more than a mere majority vote in the legislature. The majority’s preference, therefore, can be frustrated by some minority and the harder it is to amend the greater will be the abridgment of the democratic power to choose the rules governing collective life. Patrick Henry’s objections to the amending procedures implicitly raise this point with respect to Article V. Not only may a minority block widely desired change, a minority of citizens residing in the smallest population states can impose new constitutional rules on a dissenting majority. 38 According to historian David Kyvig, however, “in reality . . . [t]he distribution of population in ratifying and non-ratifying states [has been], in the aggregate, close to the proportion that the Founders held to be satisfactory to establish or deny a supermajority consensus.” 39
It is also hard to evaluate the “difficulty” built into amendment rules. It is challenging to identify all, or even the most important, variables that might explain the ease or difficulty of constitutional amendment. Amendment formulas make use of many institutional and procedural devices. They call for the approval of designated officers and bodies and specify the forms in which their assents can be manifested such as approval by supermajorities or by repeated votes. They can build delays into the process allowing the relevant actors to have second thoughts. 40 Many of these factors may not lend themselves to comparison across legal systems. 41
In connection with this, it is worth reviewing the chronology of amendments to the United States Constitution. We have already noted the prompt adoption of the first ten amendments, the Bill of Rights, proposed by Congress in 1789 and ratified in 1791. As one commentator observed, they might reasonably be regarded, “as part of the original Constitution.” 42 In fairly short order, two more amendments were adopted, each correcting what were regarded as mistakes or oversights in the original document. The Eleventh, ratified in 1795, made explicit the immunity of state governments from suit in the federal courts. The Twelfth, ratified in 1804, rejiggered the voting procedure for president in the electoral college to prevent a repetition of the deadlocked election of 1800. 43
There followed a period of sixty-one years with no successful constitutional amendment. Then, from 1865 to 1870, in the aftermath of the constitutional crisis of the Civil War, three critical amendments were approved. The Thirteenth, Fourteenth, and Fifteenth Amendments abolished slavery, secured the civil rights of the freed slaves, and prohibited denying the vote on account of race. 44 Now followed another long period, from 1870 to 1913, with no amendments. In fact, at the beginning of the twentieth century, it looked to many observers as if the Article V obstacles were too great to permit approval of useful constitutional modifications. In 1912, Senator Robert La Follette introduced a proposed constitutional amendment to Article V that would have reduced the requirement for congressional proposal from two-thirds to a majority of each house and permitted the alternative of proposal by ten states. Ratification would be sufficient if approved by a majority of voters in a majority of states. 45 Notwithstanding the failure of this and similar plans, the next four years saw an explosion of constitutional amendment. Two amendments were approved in 1913, the Sixteenth, empowering the federal government to impose an income tax, and the Seventeenth, calling for direct election of senators. The Eighteenth Amendment, approved in 1919, created the national prohibition on the manufacture or sale of “intoxicating liquors,” and the Nineteenth, ratified in 1920, guaranteed suffrage for women in both state and federal elections. This amendment “surge” is associated with the success of the “progressive” movement in national politics. 46 Writing in 1920, Walter F. Dodd drily noted that “[a] few years ago it was thought that the difficulties of amending . . . were insurmountable except in times of grave crises . . . [but] at the present time the difficulties of federal amendment do not appear quite so great . . . .” 47
Two amendments in 1933 repealed the Prohibition amendment and reduced the delay between election and taking office for the President and the Congress. In 1951, the Twenty-Second Amendment limited presidents to two terms. Then, in the period from 1961 to 1971, there was another burst of amendment activity, with four amendments being approved. Three (Twenty-Three, Twenty-Four, and Twenty-Six) expanded the right to vote. The Twenty-Fifth Amendment provided new rules for presidential succession and disability. Apart from the Twenty-Seventh Amendment, discussed below, 48 no further amendment has been adopted since—a period of forty-seven years.
This brief summary of the history of amendment in the United States suggests that the requirements of constitutional amendment do not present insuperable obstacles. While amendment is rare in American history, there have been periods where the alignment of political forces has been sufficient to accomplish important changes. The quick adoption of the Twenty-Sixth Amendment provides an example. In December 1970, in a challenge to a federal statute setting a minimum voting age of eighteen in both state and federal elections, the United States Supreme Court held that the law was valid with respect to federal elections but unconstitutional insofar as it applied to state elections. 49 Since federal and state officers were generally chosen in single, state-run elections, this risked acute administrative problems. (At the time, eighteen-year olds could vote in only three states.) A constitutional amendment setting a uniform age of eighteen for both levels of government was approved by the necessary majorities in Congress at the end of March 1971. The necessary ratifications were made and the amendment went into effect on July 1, 1971, making it the fastest ratification in United States history.
It has been pointed out that while only thirty-three amendments have survived the proposal stage in Congress, only six of these failed ratification. This suggests that the bottleneck in the Article V process is the need for approval of two-thirds majorities in the houses of Congress. But Vicki Jackson has pointed out that supermajorities are common in other constitutional systems. In the United States, moreover, the same majorities are required to override presidential vetoes of proposed legislation. Such overrides are infrequent but, as Jackson notes, they have occurred 110 times since 1789. 15.9% of vetoes since 1961 have been overridden. 50 In sum, the strict procedures required for constitutional amendment in Article V do not make amendment impossible or near impossible. It is uncertain if the latest constitutional drought indicates that “the Article V process is ‘dead,’ or simply quiescent . . . .” 51
The frequency of constitutional amendment must depend in some measure on factors external to the bare amendment procedure. According to Bjørn Rasch and Roger Congleton, the frequency of amendment “cannot be understood by focusing on the number of veto players and degree of required consensus alone.” It also depends on “economic, political, and cultural circumstances, as well as the magnitude of unresolved problems.” 52 Tom Ginsburg and James Melton treat these and other intangible factors together under the caption of “amendment culture.” 53 A reluctant amendment culture, moreover, may be self-perpetuating. As Vicki Jackson points out, the failure of amendment proposals may convince subsequent actors that such attempts are futile. 54 In the United States, there is reason to believe that the political environment is unfriendly to explicit constitutional change. The constitutional text has developed an aura of sanctity placing a heavy burden of persuasion on anyone suggesting that it is imperfect and needs improvement. In The Federalist No. 49, Madison asserted that “a constitutional road to the decision of the people ought to be marked out and kept open” but only “for certain great and extraordinary occasions.” Such occasions should be infrequent since “every appeal to the people would carry an implication of some defect in the government, frequent appeals would, in a great measure, deprive the government of that veneration which time bestows on everything, and without which perhaps the wisest and freest governments would not possess the requisite stability.” 55
He need not have worried. David Kyvig concluded that by the time of its centennial, “Americans referred to the Constitution as ‘the Ark of the Covenant,’ and Independence Hall as ‘the holiest spot of American earth.’” “[G]lorification of the Constitution,” he concluded, had become a “formidable foe to advocates of political reform . . . .” 56 That attitude appears to continue in the twenty-first century. 57
At various times, questions have arisen as to the permissible scope of the Constitution’s amendment power. Article V itself expresses explicit limits:
[P]rovided that no amendment which may be made prior to the year one thousand eight hundred and eight shall in any manner affect the first and fourth clauses in the ninth section of the first article; and that no state, without its consent, shall be deprived of its equal suffrage in the Senate.
Unlike their counterparts in modern constitutions, these provisions do not reflect the most fundamental features of the constitutional order. They are the result of compromises that were essential to secure the agreement of various states at the time of constitutional drafting. 58 The provisions exempted from amendment were safeguards for property and commerce in slaves. At the time of drafting, differences over slavery were—and continued to be—the most difficult and serious threats to national viability. Southern states demanded protection for these clauses in return for their assent to the new government.
Both slavery protections lapsed in 1808. 59 This left just one explicit restriction on the amendment power. “[N]o state, without its consent, [could] be deprived of its equal suffrage in the Senate.” This does not make equal representation of states in the Senate “unamendable.” It creates an additional requirement for enacting such an amendment: It must be assented to by any state whose proportional representation is diminished. Since no such state is likely to give its approval, however, it is fair to say that equal representation in the Senate is effectively outside the reach of Article V. State equality in the Senate—balancing population-based representation in the House of Representatives—was crucial in securing the agreement of “small states” to the new settlement. 60 Another instance of unamendability was proposed in February 1861 after the election of Abraham Lincoln and the first secessions of southern states. Both houses of Congress approved an amendment providing that “[n]o amendment shall ever be made to the Constitution which will authorize or give to Congress power to abolish or interfere within any State, with the domestic institutions thereof, including that of persons held to labor or service by the laws of said State.” This amendment was ratified by three states. The secession of the remaining southern states and the commencement of the Civil War made any further action moot. 61
Textual restrictions, however, do not exhaust possible limits on the amendment power. In the twentieth century, courts in several jurisdictions have found amendment power to be intrinsically bounded. 62 Thomas Cooley, probably the pre-eminent American constitutional authority of the late nineteenth century, published an article in 1893 insisting that since the Constitution was intended to create a permanent union of states and since the purpose of the amendment power was to enable such adjustments as would ensure the Constitution’s survival, only amendments that were “harmonious with the original structure” were authorized. 63 This argument was renewed in the law reviews and in the courts in response to the flurry of amendment activity in the first twenty years of the twentieth century.
One variation was directly premised on the effectively unamendable quality of equal representation of states in the Senate. It supposed that there were certain powers which could not be taken away from states without destroying their status as “states” as that term was used in Article V. Such an amendment would then eliminate the state’s “equal suffrage” in the Senate. The Eighteenth Amendment, forbidding “the manufacture, sale, or transportation of intoxicating liquors” was attacked on this ground insofar as it transferred part of the states’ ability to make laws for their own citizens to the federal government. 64 The extensions of the right to vote in state elections mandated by the Fifteenth and Nineteenth Amendments were likewise alleged to alter the definition of the state since states with different bodies politic had been recognized in the original Constitution. 65
This kind of argument has been framed even more generally. In the twentieth century, several constitutional courts followed the lead of the Supreme Court of India in Kesavananda Bharati v. State of Kerala. 66 That court held that constitutional amendments could be invalid if they contradicted certain fundamental principles making up the “basic structure” of the Indian Constitution. Very similar arguments were articulated in the nineteenth and twentieth centuries by American observers and were argued (though unsuccessfully) in American courts. On this view, there was a critical distinction between the amendment and the replacement of a constitution. To qualify as a permissible amendment, the result must leave the pre-amendment constitution basically in place. 67 During the debate on the proposed Thirteenth Amendment in the House of Representatives in 1865, one Congressman argued that “the very term ‘amendment’ [was] a word of limitation,” disallowing a “plenary, omnipotent, unlimited power over every subject of legislation.” 68 Advocates of this limited view of amendments disagree on the extent of change that disqualifies something as a permissible amendment. A modern commentator on Article V has insisted that it is limited to “fine tuning what is already in place.” 69 Others have allowed more latitude. Yaniv Roznai quoted John Calhoun, the defender of states’ rights in the antebellum period, denying only that an amendment could “radically change the character of the Constitution or the nature of the system.” 70 More recently, Laurence Tribe has offered the example of “an amendment repealing the Article IV guarantee of a ‘republican’ form of government and simultaneously making membership in Congress a matter of heredity, rather than election by ‘the People’. . . .” 71
The United States Supreme Court has entertained a number of cases challenging the validity of constitutional amendments, but it has never overturned the decisions of the political branches of government recognizing an amendment as valid. In Leser v. Garnett 72 in 1922, a group of male voters attempted to have the names of female voters struck from the Maryland voter rolls, the constitution of that state restricting the franchise to men. The Nineteenth Amendment to the United States Constitution, prohibiting denying the right to vote “on account of sex” had been proclaimed ratified in 1920, though without the approval of the Maryland legislature. Relying on the kind of arguments just summarized, plaintiffs contended that when an amendment “changes the electorate, the original State is destroyed and a new State created.” The amending power could not defeat the central point of the Constitution, to create a permanent union of states. 73 The Court, in a short unanimous opinion, dismissed this proposition in a single conclusory paragraph. This amendment was indistinguishable from the Fifteenth, eliminating racial qualifications. The validity of the Fifteenth Amendment had been unquestioned for “half a century,” and plaintiffs’ assertion that the Fifteenth Amendment was a “war measure . . . validated by acquiescence, cannot be entertained.” 74 In United States v. Sprague 75 in 1931, the Eighteenth (Prohibition) Amendment was challenged on the basis that the Tenth Amendment required amendments limiting personal rights be approved by “the people,” that is, by the convention method. As in Leser, the Court wasted little time on this point. The appellees “ask us to hold that Article V means something different from what it plainly says.” 76 The suggested construction is a “complete non sequitur.” The choice of ratification method was committed to Congress by Article V. 77
The failure of these claims is understandable in light of the absence of Article V language limiting the scope of amendment. Perhaps more surprising is the Supreme Court’s reluctance to review the validity of amendments against claims that the process of adoption failed to comply with the procedures set out in the Constitution. By its terms, Article V calls for the approval of certain bodies and designates how those approvals should be manifested. Its general terms might have been expected to be worked out in litigation. The Supreme Court has shown a distinct reluctance to grapple with these questions.
Probably no amendment has been more problematic in this regard than the Fourteenth, mandating that states respect the “privileges and immunities of citizens of the United States” and forbidding a “state depriv[ing] any person of life, liberty, or property, without due process of law . . . [or] deny[ing] . . . any person within its jurisdiction the equal protection of the laws.” It is arguably the most important of all the constitutional amendments. It was, along with the Thirteenth (abolishing slavery) and Fifteenth (guaranteeing the right to vote without regard to race), enacted in the aftermath of the Civil War as part of an effort to secure the legal equality of emancipated African-American slaves. The constitutional theory of the Union was that secession was illegal, that there were and continued to be thirty-six states. Yet, when Congress proposed the amendment in 1866 by two-thirds vote of each house, the seceding states were unrepresented in Congress. The ratification was equally worrying. Congress (which, it should be recalled, has no role in ratification) had enacted the Military Reconstruction Act conditioning restoration of political rights to ten southern states on their ratification of the amendment. This was very far from the kind of voluntary and deliberate agreement that Article V contemplated. According to one modern commentator it was “plain coercion”; 78 according to another, the amendment was “added only at the point of a gun.” 79 Still, as a proponent of intrinsic limits on amendment conceded in 1920, there has never been a challenge to the amendment’s validity, “although infinite opportunities to contest [its] validity had presented themselves . . . . [I]t must be conceded that no court in the world could be blamed for declining to consider objections to [its] validity after such a long period of universal assent . . . .” 80
It is hard to characterize the Supreme Court’s approach in the relatively few cases where it has dealt with claims of defective procedure in adoption of other amendments. In an early decision, a party maintained that the Eleventh Amendment limiting actions against states in federal courts had never become law because the President had not assented to it. In argument, Justice Chase firmly rejected the claim: “The negative of the President applies only to the ordinary cases of legislation: He has nothing to do with the proposition, or adoption, of amendments to the Constitution.” 81 That conclusion has not been questioned since. On the other hand, in 1939 the question of whether the Lieutenant Governor of a state could participate in the ratification vote of a state senate was left unanswered by the Supreme Court, its members being equally divided on whether or not it presented a nonjusticiable “political question.” 82 The Court has, however, decided that a state may not condition its ratification on the result of a referendum. The reference to “legislatures” in Article V could only mean “the representative body which made the laws of the people.” 83 This raises additional questions. What makes a legislature is necessarily determined by the state’s constitution. So, ratification in forty-nine states appears to require the agreement of two houses, but in Nebraska, which has no upper house, only one. 84 Likewise, the sittings of legislative bodies and the procedures for bringing matters to a vote differ from state to state. It is unclear at what point these variations might disqualify a state’s institutions from being an Article V legislature. A federal district court has held that a state could require that ratification be by a three-fifths vote of each chamber. 85
The Supreme Court has given mixed signals on the existence of implicit time limits on ratification. When Congress proposed the Eighteenth (Prohibition) Amendment to the states, it included a section requiring ratification to be completed within seven years. The Court rejected an argument in Dillon v. Gloss 86 in 1921 that inclusion of this limit invalidated the amendment. Agreeing that Article V itself expressed no time limit, the Court also found nothing “suggest[ing] that an amendment once proposed is to be open to ratification for all time.” Rather, the “natural inference” of the ratification scheme is that state approvals “are not to be widely separated in time.” Ratification “must be sufficiently contemporaneous in that number of states to reflect the will of the people in all sections at relatively the same period . . . .” The Court illustrated this proposition by noting that there were four long outstanding congressional proposals, including two left over from the original submission of 1789. These proposals had been ratified in some states “by representatives of generations now largely forgotten.” Without a time limit, those proposals could become part of the Constitution by adding enough ratifications to the eighteenth-century ones to reach the three-fourths threshold. “To that view few would be able to subscribe, and in our opinion, it is quite untenable. We conclude that the fair inference or implication from Article V is that the ratification must be within some reasonable time after the proposal.” 87 The Court thus held that Congress was acting within its power in specifying seven years for ratification.
In Coleman v. Miller, 88 decided in 1939, the Supreme Court was asked to apply this “reasonable time” requirement to an amendment that had been submitted to the states thirteen years previously and which, unlike the Eighteenth Amendment, had named no time limit. This time the majority held that the existence as well as the length of a ratification period was “essentially political and not justiciable.” 89 The Equal Rights Amendment that would have mandated recognition of gender equality by both federal and state governments was proposed by Congress in 1972. The authorizing resolution—but not the text of the amendment—contained a seven-year ratification deadline. When that deadline expired, it was still three states short of adoption and Congress by majorities—but not two-thirds majorities—voted to extend the deadline for three years. The validity of this action became moot when no new ratifications were forthcoming in the additional time. 90
The Twenty-Seventh Amendment represents the reductio ad absurdum of the choice not to enforce a requirement of reasonable contemporaneity in ratifications. This provision, postponing any increased compensation for members of Congress until after an intervening election, was one of the original amendments approved by the first Congress and submitted to the thirteen states in 1789. Six states ratified before 1800 and a seventh in 1872. A new wave of ratifications began in 1978 and, along with the prior seven approvals, these reached thirty-eight in 1992, that being three-fourths of the by now fifty states. Congress promptly declared the text properly ratified and part of the Constitution. 91
It will be observed that, notwithstanding many unresolved questions about the meaning and application of Article V’s procedures, the Supreme Court has exhibited a certain skittishness in cases involving the amending power. Many of its opinions suggest that the issues might present “political questions” for which there are no reliable legal standards and which, therefore, should be left to the other branches of government. Even when the Court has decided these cases it has shown some stinginess in its explanations. Its 1920 opinion for a consolidated group of cases challenging the validity of the Eighteenth (Prohibition) Amendment omitted the usual discursive statement of reasons. Justice Van Devanter simply “announced the conclusions of the court.” He then set down eleven conclusory statements dealing with the various issues raised by the parties. 92 In a concurring opinion, Chief Justice White rebuked the Court for refusing to provide “an exposition of [its] reasoning” “in a case of this magnitude.” 93 In Coleman v. Miller, 94 the opinion for the Court pronounced two of the claims about state ratification to be political questions left to Congress. Four justices made an even more thoroughgoing justiciability objection to judicial intervention. “The process itself is ‘political’ in its entirety, from submission until an amendment becomes part of the Constitution, and is not subject to judicial guidance, control or interference at any point.” 95 Although lower courts have very occasionally weighed in on amendment questions, 96 Coleman is the last serious examination of the subject by the Supreme Court.
I began by noting that the United States Constitution is both the world’s oldest and one of the least changed. Yet, if we examine the institutions and procedures that the Constitution purports to define and regulate, it is clear that they are now drastically different from what they were in 1789. Large categories of constitutional authority have been radically transformed. Perhaps most obvious is the rebalancing of authority in the federal system. The national government, which the original text contemplated playing a limited and exceptional part in the day-to-day tasks of government, has vastly expanded its field of operation. The range of all government activity, state or federal, moreover, has markedly enlarged. Governments at the end of the eighteenth century may have been more than “night watchman states,” but they were not expected to carry out the regulatory and welfare functions which are routine today. The Constitution has also become the cited basis for important new categories of individual rights, both economic and personal, and actionable against both state and federal governments. Some of these changes are attributable to the adoption of formal amendments, but the greater part has emerged from political and judicial processes that are difficult to justify by reference to any positive law. 97
It is fair to say that most of what now goes under the caption “constitutional law” in the United States is attributable to extraconstitutional, “off-the-books” developments. 98 It is impossible here to do more than to describe, in broad terms, the principal devices by which such change has been accomplished and to consider the extent to which they amount to an acceptable substitute for formal amendment. It is usual to identify the judiciary as the principal agent of constitutional change outside of Article V, but first it is worth noting the contributions made by the political departments of the government. 99 In an ambitious undertaking, combining legal theory and constitutional history, Bruce Ackerman, has described two major reconstructions of the American Constitution. 100 The first, emerging in the wake of the Civil War, entrenched a national regime of individual rights. The second, appearing in and after the New Deal administrations of President Franklin Roosevelt, laid the foundations for the modern regulatory and welfare state. These transformations did not employ the Constitution’s amendment rules. They initiated new constitutional regimes in the same way as did the 1787–1789 convention and ratification. Ackerman posits a rough outline of the things that have to happen for such reworkings to succeed. They are, first and foremost, political events in which the new arrangements are proposed and critically examined in a period of “constitutional politics” marked by intense reflection, debate, and engagement by a large proportion of the population. While the resulting settlement is reflected in judicial interpretations of the (new) constitution, the judges are not the prime movers. Ackerman’s analysis yields many new insights about American constitutional history but, not surprisingly, it has also attracted powerful criticisms, challenging its history, its consistency with prevailing attitudes about the Constitution, and the extent of its explanatory power. 101
Commentators have pointed out more prosaic ways in which the effective Constitution has changed outside of Article V. One example concerns the relative powers of the President and the Congress in foreign affairs. Article II of the Constitution grants the President the power to make treaties but only “with the advice and consent of the Senate . . . provided two-thirds of the Senators present concur.” With increasing frequency, international agreements have been concluded by the President with the approval of only ordinary majorities in both houses of Congress. In the twentieth and twenty-first centuries, “executive agreements, especially congressional–executive agreements, have come to represent the vast majority of international agreements made by the United States.” Leading authorities, moreover, take the position that these agreements are “fully interchangeable under U.S. domestic law with Article II treaties.” 102 This is a case where Congress has successfully used “ordinary legislative means to update constitutional meaning.” 103
More dramatic has been the change in the power to make war. According to Article I, Congress has the power “to declare war.” This decision has now been effectively transferred to the President. Congress last expressly declared war on December 8, 1941. More than 90,000 American military personnel have died in hostile actions in various conflicts in which the United States has been involved in military actions since that day. 104 The line between Congress’s power and the President’s constitutional status as “commander in chief” of the armed forces is not perfectly clear. Likewise, certain forms of congressional authorization might serve as implicit declarations of war. Several serious instances of military action, however, exceed even a generous reading of the President’s power. The consistent acquiescence of Congress in these enterprises is reasonably understood as an effective shift of constitutional authority. 105
An equally striking instance of modification of the Constitution’s scheme is the creation of the modern administrative state. The federal government, as outlined in the Constitution, exercised authority through three rather simply organized branches. The Constitution contemplated the President appointing and managing “officers” of the United States who were organized in “executive departments.” 106 All public business, that is, would be conducted by people ultimately answerable to the political forces that fill the positions described in the Constitution. Congressional and executive actions since the end of the nineteenth century have established a radically different government. The regulatory activity of the federal government is carried out by scores of agencies. These are staffed by people, some of whom are appointed by the President alone and some only with consent of the Senate. Some are removable by the President at any time and some only after a fixed term. 107 They legislate and enforce rules that have only the scantest connection to policies enacted by Congress. 108 These agencies adjudicate disputed questions about compliance with those rules, and their decisions are subject only to strictly limited and highly deferential review by the ordinary courts. 109 The force of the federal government is, therefore, exercised in a way that departs dramatically from the eighteenth century constitutional scheme. 110
The most clear-cut modification of constitutional rules outside Article V, however, has been worked by the courts in the course of constitutional litigation. While the developments outlined above have resulted in very substantial changes in the practical operation of constitutional government, they have been, for the most part, completed without the relevant actors devoting serious attention to their consistency with the text of the Constitution. Judicially wrought changes, on the other hand, are usually accomplished in the course of an explicit act of constitutional interpretation. To the extent these “interpretations” are perceived as changing the meaning of the relevant text, their effect is indistinguishable from an Article V amendment of that text.
The undisputed fact that judicial interpretation has substantially departed from the original constitutional rules has often been related to the supposed difficulty of the Article V procedure. If we think that changing economic, technical, and political facts make essential a certain amount of change in the state’s governing rules and that change cannot be accommodated under the rules of Article V, we might think it must necessarily occur in some other way. 111 In a book on American government, published in 1885 before he entered electoral politics, Woodrow Wilson wrote:
The process of formal amendment of the Constitution was made so difficult by the provisions of the Constitution itself that it has seldom been feasible to use it; and the difficulty of formal amendment has undoubtedly made the courts more liberal, not to say more lax, in their interpretation than they would otherwise have been. 112
He was ready to describe the Supreme Court as “a kind of Constitutional Convention in continuous session.” 113 More recent observers have come to similar conclusions. 114
It is inarguable that “the two most prominent sources” “of new constitutional law” “are the amendment process, on the one hand, and judge-made constitutional law on the other.” 115 It is impossible to measure their relative contributions to the current set of enforceable constitutional rules. It does not exaggerate, however, to say that judicial decisions have “alter[ed] the polity quite radically.” 116 Walter Murphy reasonably argued that “Marbury v. Madison (1803), McCulloch v. Maryland (1819), Gibbons v Ogden (1824), Brown v. Board of Education (1954), and even Dred Scott v. Sandford (1857) were far more important than many amendments in (re)shaping the American nation.” 117 The Supreme Court, with the help of a single litigant, may recognize new constitutional rights, significantly reforming the relationship between states and individuals. But the judges also play a critical role in the establishment of the extra-constitutional procedures and institutions initiated by the political branches discussed earlier. In these situations, as Adrian Vermeule has observed, one might see the judges’ contribution as simply “getting out of the way.” But since, for example, in the case of the “New Deal” reforms of the 1930s, people were able and willing to challenge their legality in the courts, the judges could really get out of the way only by formulating a “capacious interpretation of the national government’s [constitutional] powers—which was itself just another type of judicial updating.” “[S]tructural changes developed by the political branches are always constitutionally insecure until the judges put an affirmative stamp of approval on them . . . . In that sense, there is no such thing as nonjudicial constitutional change outside the Article V process.” 118
Notwithstanding this critical judicial role, it is unusual for judges to acknowledge that constitutional adjudication may be a form of amendment. Nothing, in fact, “is more alien, at least to our conventional notion of judicial (and judicious) analysis of the Constitution, then a [court] decision that is described as an ‘amendment.’” 119 The key word in negotiating this tension is “interpretation.” If it is possible to conceive of the development of novel rules as an interpretation of the Constitution, the change can be legitimated by the same factors that give the Constitution its status as fundamental law. “To designate something as interpretation . . . is to accord it a certain legal dignity . . . .” 120 It is on this ground that the principal political and intellectual disputes about the Constitution have been fought. In two speeches delivered in 1985, President Reagan’s Attorney General, Edward Meese, warned about Supreme Court judgments purporting to interpret the Constitution. He objected to decisions based on an evaluation of competing policies and not on “articulations of constitutional principle.” The resulting jurisprudence, once unmoored from the constitutional text, resulted in a “jurisprudence of idiosyncracy [sic].” 121 He called for the Court to pursue a “jurisprudence seriously aimed at the explication of original intention.” 122 By this he meant the original intentions animating the Constitution, the intentions of the people who enacted its provisions.
In the years thereafter, the American legal academy has engaged in an apparently endless and ever more complicated debate over the propriety of “originalism” as a technique of judicial constitutional decision making. 123 The originalists object to methods of “interpretation” that permit a series of decisions giving different meanings to the same constitutional language. This objection is premised on a certain understanding of the purpose of constitutional government. Specifically, it assumes that constitutionalism is not simply a device to limit government power. It limits that power by application of fixed rules. It therefore offers individuals a reliable advance indicator of activities that they may undertake with confidence that they will be able to act without public interference. 124 The stringent procedures for amendment laid down in Article V comfortably fit into this picture of the constitutional state. The Constitution can change, but it can change only infrequently, and since it changes by a prospective lawmaking process, it provides prior notice of the changes. The very broad consensus that the formal amendment procedure demands may also prevent any radical revision of the existing system. 125 Admittedly, judicial management of constitutional change proceeding according to the “common law” method of decision making, might also be expected to foster gradual change. 126 On the other hand, judicial changes in constitutional law arrive on no predictable schedule depending, as they do, on the happenstance of litigation. Furthermore, the history of American constitutional law suggests that a change of personnel on the Supreme Court (something that mainly depends on actuarial chance) may establish or eliminate major features of the constitutional regime. By design, moreover, judges, unlike the political institutions involved in Article V amendment, do not need to take account of the lumbering shifts in public opinion before changing the rules in important ways. In addition, given the pretense that the new policy is merely an enunciation of existing law, such holdings are presumptively retroactive, possibly unsettling past private actions. 127
Vermeule has argued that reliance on constitutional litigation rather than on explicit amendment to reform the Constitution is more likely to produce confusing and inconsistent rules:
The question is whether piecemeal amendment produces greater incoherence then piecemeal judicial updating, carried out in particular litigated cases, by judicial institutions whose agenda is partly set by outside actors. There is little reason to believe the latter process more conducive to coherence than the former, and much evidence to suggest that judicial decision-making produces a great deal of doctrinal incoherence. We should disavow any implicit picture of judge-made constitutional law as an intricately crafted web of principles whose extension and weight has been reciprocally adjusted. Precisely because judicial updating requires overrulings, reinterpretations, and other breaks in the web of prior doctrine, a system that relies on judicial updating to supply constitutional change . . . generates internal pressures toward incoherent doctrine. 128
In terms of the values of constitutionalism, therefore, the combination of originalist constitutional adjudication and exclusive use of the formal amendment procedure has distinct advantages. Those advantages, however, need to be balanced against the costs of relying on a formal procedure that, as a matter of hard fact, will often be unsuccessful. Notwithstanding the issues raised in section A about metrics of amendment difficulty, the historical record leaves little doubt that use of Article V has not itself been sufficient to keep the constitutional text adequate to the facts and values of the twenty-first century. For many commentators, the obstacles to change created in Article V are at the root of the “dead hand” problem. 129 Later constitution makers in both the states and in other countries have chosen less demanding amendment procedures, suggesting that the 1787 founders erred on the side of excessive entrenchment. 130 It is impossible to say for sure how the amendment power might have been used in the absence of what might have been preemptive judicial revision. 131 It is clear, however, that sometimes, even a broadly and intensively held opinion has a hard time translating itself into the consensus required by Article V. According to one observer, faithful adherence to the original text’s “rules for issues such as congressional power and executive discretion would likely have invited national calamity and constitutional failure . . . .” 132 As noted, the required ratification by three-fourths of state legislatures might, in theory, allow opposition in the smallest states to frustrate adoption of proposals urgently desired by an overwhelming portion of the population. In 2016, this meant that single legislative chambers in states with less than 5% of the population could block amendments. 133
The fate of the Equal Rights Amendment on gender equality, mentioned above, may illustrate the capacity of Article V requirements to thwart widely supported change. It passed both houses of Congress by overwhelming votes (354 to 24, and 84 to 8) in 1971 and 1972. Within one year of submission, thirty states had ratified, but ratification stalled at thirty-five despite congressional extension of the initial seven-year deadline. 134 During the time that the amendment was pending, public opinion polls showed a solid and consistent national majority in its favor. 135 In the same period, the Supreme Court decided several cases holding that the equal protection clause of the Fourteenth Amendment provided some of the rights that proponents of the Equal Rights Amendment hoped it would secure. While these judicial interventions may be viewed as providing a desirable alternate route to accomplish the changes stymied by the Article V barriers, they may also be seen as taking the steam out of the ratification drive and thus as partially responsible for the amendment’s failure. 136
These questions about whether or not judicial interpretation is a legitimate alternative method of constitutional change present another variation of the standard opposition of stability and flexibility in constitutional law. The value of constitutionalism arises from its ability to settle the basic rules governing the operation of the state. Constitutional rules need to be harder to change than ordinary law if they are going to perform that steadying function. But there is a point beyond which the inability to deal with new circumstances can put the very survival of the state in jeopardy. It was something of a “breakthrough” in constitutional theory when the framers of the Constitution provided a method by which its inadequacies could be cured. 137 Those founders well understood that the amendment formulas had to compromise these values. In defending Article V, James Madison spoke directly to the point: “It guards equally against that extreme facility, which would render the Constitution too mutable; and that extreme difficulty, which might perpetuate its discovered faults.” 138 The undeniable development since the founding of additional avenues for constitutional change is some evidence that the balance struck at that time, if faithfully adhered to, would have proved unsatisfactory. Its supplementation with a crazy quilt of alternate methods of constitutional change, however, carries the risk that the Constitution, which is supposed to provide clear and stable limits on the state, will itself become impossible to identify.