Originalism and the good constitution

Georgetown Law Journal

Volumn 98, Issue 6, 2010, Pages 1693-1768

  McGinnis, John O ^a   Rappaport, Michael B ^b  

^a NORTHWESTERN UNIVERSITY   (United States)

^b UNIVERSITY OF SAN DIEGO   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

EID: 77951863634     PISSN: 00168092     EISSN: None     Source Type: Journal    
DOI: None     Document Type: Article

References (275)

1. See Robert H. Bork, Neutral Principles and Some First Amendment Problems, 47 IND. L.J. 1, 4, 16, 20 (1971).
2. See, e.g., Michael C. Dorf, Foreword: The Limits of Socratic Deliberation, 112 HARV. L. REV. 4, 26 (1998).
3. See, e.g., RANDY E. BARNETT, RESTORING THE LOST CONSTITUTION: THE PRESUMPTION OF LIBERTY 121 (2004) (construction helps address "the vagueness of language");
4. Lawrence Solum, Originalism and the Natural Born Citizen Clause, 107 MICH. L. REV. FIRST IMPRESSIONS 22, 30 (2008) (suggesting that some "new originalists" resort to construction to resolve "irreducible ambiguity").
5. Some of the most famous originalists have never addressed these exclusions in their analysis. See, e.g., ROBERT BORK, THE TEMPTING OF AMERICA (1997);
6. ANTONIN SCALIA, A MATTER OF INTERPRETATION: FEDERAL COURTS AND THE LAW (1997). Indeed, we cannot think of any extended analysis of these matters by any originalist.
7. Entrenchment is the process by which provisions are formally made unchangeable by the ordinary legislative process. In our federal system, constitutional provisions are the primary, if not the only, example of entrenched provisions, and thus, for the purposes of this paper entrenched provisions and constitutional provisions are the same.
8. Our consequentialist approach is a modern version of utilitarianism. For an accessible discussion, see WILLIAM H. SHAW, CONTEMPORARY ETHICS: TAKING ACCOUNT OF UTILITARIANISM 52-63 (1999). Two other aspects of our position are worth mentioning. First, we believe that settled rules and practices are generally the best way of promoting good consequences and, therefore, often support positions that seem rule consequentialist but are actually justified on act consequentialist grounds. In this, we take an approach similar to the two-level approach to consequentialism of R.M. Hare.
9. See, e.g., R.M. HARE, MORAL THINKING: ITS LEVELS, METHOD, AND POINT (1981). Second, although we say that constitutions should be assessed based on their consequences for "the welfare of the people of the nation," we do not mean to imply that foreigners do not count, but merely to put to the side the complicated issue of the correct weighing of foreigners in this context.
10. We should also note that here we are discussing a good constitution for liberal democracies. Just as John Rawls limits the extension of his theory of justice to liberal democracies, see JOHN RAWLS, A THEORY OF JUSTICE (1971), prescriptions here are limited to a society with the characteristics, material and cultural, that can sustain such democracies.
11. See Sotirios A. Barber, Professor Eisgruber, the Constitution, and the Good Society, 69 FORDHAM. L. REV. 2151, 2151-2152 (2001);
12. Steven P. Croley, The Majoritarian Difficulty: Elective Judiciaries and the Rule of Law, 62 U. CHI. L. REV. 689, 694-697 (1995).
13. See Barry Friedman & Scott B. Smith, The Sedimentary Constitution, 147 U. PA. L. REV. 1, 87 (1998) (discussing the importance of "fundamental, historically based, deeper commitment[s]" in our constitutional institutions over present preferences).
14. See HANNAH ARENDT, Civil Disobedience, in CRISES OF THE REPUBLIC 49, 76 (1972).
15. Although the supermajoritarian process is thus likely to produce a good constitution, it is important to recognize that the constitution it generates will not appear ideal to everyone. Even though some people may regard the constitution produced by the supermajoritarian process as less than ideal, that does not mean, for at least three reasons, that it is not a good constitution for governing the nation. First, the importance of establishing a constitution that is supported by a consensus of the nation places real limits on obtaining the constitution that any one person would desire. If the constitution that one favors is strongly opposed by any significant group in the nation, that constitution is unlikely to be normatively desirable for the nation. Second, it is also important that the people accept the constitution as legitimately adopted. The mere fact that one particular individual believes his proposed constitution is the best will not lead others to accept it, as they will prefer different constitutions. Because people recognize that not everyone can secure their ideal constitutions, they are likely to accept only a constitution that they believe has been adopted in a manner that assures it is a reasonable compromise among various groups. Because the supermajoritarian process is easily seen as a procedure for generating a reasonable compromise, people are likely to accept the results of that process (especially given its other virtues). Third, the various virtues of the supermajoritarian process that we discuss in this Article, which involve using the distinct knowledge and preferences of a large number of people to form a constitution, suggest that the results of that process may actually be superior to an individual's intuitions about what a good constitution is. In other words, if a good constitution must take account of the preferences and beliefs of the people as well as their objective interests, then no single individual would be in a position to know what would further those interests, preferences, and beliefs better than the supermajoritarian process.
16. If one does conclude that a constitution that has passed through an appropriate supermajoritarian process is not desirable, then how should one treat it? If the constitution is not desirable, then the primary reasons we give in this Article for following it are not applicable, but that does not mean that one should not follow it for other reasons. We believe that a constitution that passes through appropriate supermajority rules would be at worst what we call a fair constitution. We have defined a good constitution as one that is inferior to, but is still in the neighborhood of, an ideal constitution. A fair constitution, by contrast, is one that is quite a distance from the ideal constitution but is still significantly better than the status quo. A fair constitution is an imperfect one, but we believe that the best alternative is still to enforce that constitution. The other main alternatives-writing a new constitution or judicially departing from the original meaning in an attempt to correct the constitution- are unlikely to be preferable to enforcing the fair constitution. For a discussion of these alternatives in the context of a constitution produced by inappropriate supermajority rules, see infra Part V.
17. Our view of the Constitution thus parallels Paul Bator's view of the criminal trial. No human institutions can be certain of being transcendentally correct. The results obtained by the best available procedures have a claim to finality. See Paul M. Bator, Finality in Criminal Law and Federal Habeas Corpus for State Prisoners, 76 HARV. L. REV. 441, 447-448 (1963).
18. By democratic representatives, we include anyone elected to make decisions about legal enactments, including both legislators and members of conventions elected to decide on constitutional provisions.
19. The preference perspective is discussed further in John O. McGinnis & Michael B. Rappaport, Majority and Supermajority Rules: Three Views of the Capitol, 85 TEX. L. REV. 1115, 1152 (2007).
20. The accuracy perspective is discussed further in John O. McGinnis & Michael Rappaport, The Condorcet Case for Supermajority Rules, 16 SUP. CT. ECON. REV. 67, 73 (2008).
21. See generally McGinnis & Rappaport, supra note 15 (offering a preference model to explain the operation of supermajority rules); McGinnis & Rappaport, supra note 16 (offering an accuracy model to explain the operation of supermajority rules). While our normative justification for supermajority rules is welfare consequentialist, our arguments in this Part involve an efficiency standard. We assess legislative behavior in terms of voter preferences under the preference model and we assume that legislators pursue efficiency as the public interest under the accuracy model. The reason we focus on efficiency, rather than welfare consequentialism, is that our earlier work, on which we rely, did so for reasons of tractability. But our conclusions concerning efficiency are readily transferable to welfare consequentialism. The two normative approaches are similar. First, the accuracy model will apply to welfare consequentialism so long as the policies that promote the welfare of the people are a matter of fact (or are capable of being true or false), which we believe they are. Second, while the normative standard of maximizing efficiency is not the same as that of maximizing the welfare of the people, they are similar, especially under a view of welfare that emphasizes preference realization. It seems quite unlikely that our results comparing majority and supermajority rule would change if the standard were welfare rather than preference maximization.
22. The reason is that only laws that secure fifty-one votes or more produce net benefits, and only such laws pass.
23. In other writings, we have suggested that to complete the analysis of the difference between majority and supermajority rule, we must consider redrafted as well as marginal legislation. See, e.g., John O. McGinnis & Michael B. Rappaport, Our Supermajoritarian Constitution, 80 TEX. L. REV. 703, 734 n.136 (2002). Redrafted enactments are those enactments blocked by majority rule that are redrafted to get additional support under majority rule. In the context of entrenchment, we think this additional concept would simply underscore the advantages of majority rule. To gain supermajority support, constitutional provisions that could not succeed under majority rule would have to be redrafted to get consensus and more bipartisan support. Thus, almost by definition, redrafted legislation would represent an improvement, making supermajority rule better than majority rule for entrenchments even by a larger margin.
24. For instance, a nation with two large groups of different sizes divided by ethnicity or religion might want to choose a rule that will require substantial buy in from the minority group. Only in that way would the constitution-making rule create a constitution that is likely to obtain widespread allegiance.
25. See, e.g., Rick A. Swanson & Albert P. Melone, The Partisan Factor and Judicial Behavior in the Illinois Supreme Court, 19 S. ILL. U. L.J. 303, 305 (1995) (arguing that partisan identification helps voters cast more informed votes).
26. Legislators will have the support of citizens if citizens favor the partisan measures for strategic reasons. Legislators will lack the support of citizens if citizens do not favor the entrenchments for strategic reasons, but legislators can exploit agency costs to pursue a measure citizens do not favor.
27. The problem of partisanship entrenchment has been recognized in the legal literature largely in the context of stacking the Court with partisans of one party. See, e.g., Michael J. Gerhardt, The Constitution Outside the Courts, 51 DRAKE L. REV. 775, 789 (2003). But majoritarian entrenchment by the legislature would present a more extreme version of the same problem because it is easier for legislators to coordinate on a present entrenchment than for the President and his party to assure that judges they appoint will do so in the future.
28. It might be argued that parties could avoid the prisoner's dilemma created by majoritarian entrenchment simply by entrenching a prohibition on matters that the other party would entrench when it came to power. One difficulty with this strategy is that a party cannot necessarily predict the full range of measures the other party will want to entrench and thus faces far more uncertainty in determining what entrenchments to prohibit than in determining what entrenchments to make. For instance, one party may seem to be interested in entrenching health care entitlements. Although that entrenchment could be prohibited, a party coming to power might also desire to make a different entrenchment. Moreover, it is often harder to motivate people to prohibit potential entrenchments of another party. Because the other party is not in power, their potential entrenchments are actions to be taken in the future, making them appear less salient in the politics of the present.
29. See Michael A. Fitts, Can Ignorance Be Bliss? Imperfect Information as a Positive Influence in Political Institutions, 88 MICH. L. REV. 917, 922-923 (1990).
30. It might be argued that a majority entrenchment rule would also establish a limited veil because entrenched provisions passed by a majority would still require a supermajority to amend. Although a majority entrenchment rule would also help to establish a veil of ignorance, it would not function as well as supermajority rule because it would not address the problem of majoritarian threats. A supermajority entrenchment rule would be superior because the majority entrenchment rule would produce the undesirable consequence of allowing a majority to threaten to pass problematic entrenchments in order to secure the repeal of an existing entrenchment that it disliked. To illustrate this problem, assume that the existing majority lacks the supermajority of votes needed to repeal an existing constitutional provision. That majority, however, does have the power to pass a new entrenchment. To secure the additional votes necessary to repeal the constitutional provision from those who are opposed to its repeal, the majority could threaten to entrench another measure that is disliked even more than the repeal of the existing entrenchment. If this threat succeeded, it would reduce the extent to which constitutional provisions are entrenched. But even if it did not always succeed, it would create a situation where the possible passage of constitutional amendments could be used as bargaining chips in an effort to secure repeal of existing amendments. Such political games can only undermine the public's allegiance to the constitution.
31. See Adrian Vermeule, Veil of Ignorance Rules in Constitutional Law, 111 YALE L.J. 399, 417 (2001). A high discount rate suggests that future concerns are given substantially less weight than present concerns in decision making.
32. Id. at 418.
33. Although a ruling coalition could not easily pass constitutional amendments when existing constitutional provisions no longer serve its interests, it is possible that judges could depart from the original meaning of those provisions when desired by the ruling coalition. We discuss ways for inducing judges to follow the original meaning below, arguing that the best method involves developing a legal culture that values originalism. See infra notes 169-171 and accompanying text. But even if one ignores this cultural effect and assumes that judges may sometimes not follow the original meaning, that would not allow a ruling coalition to dictate constitutional results because the Supreme Court's behavior is neither predictable nor easily controlled. 30. Id.
34. See DANIEL A. FARBER & PHILIP P. FRICKEY, LAW AND PUBLIC CHOICE: A CRITICAL INTRODUCTION 31 (1991) (describing evidence that "one factor in determining how a legislator votes is simply that legislator's view of the public interest").
35. For a more technical defense of the Condorcet Jury Theorem's application to legislative voting, see McGinnis & Rappaport, supra note 16.
36. Prominent among legal theorists supporting this view are Jeremy Waldon and Frank Michelman. See, e.g., David M. Estlund & Jeremy Waldron, Democratic Theory and the Public Interest: Condorcet and Rousseau Revisited, 83 AM. POL. SCI. REV. 1317, 1317-1318 (1989) (applying the Condorcet Jury Theorem to Rousseau's theory of the general will to justify majority rule);
37. Frank I. Michelman, Why Voting?, 34 LOY. L.A. L. REV. 985, 996 (2001) (stating that, in an epistemic theory of justice, the Condorcet Jury Theorem serves as justification for majority rule).
38. See, e.g., DENNIS C. MUELLER, PUBLIC CHOICE III 129-133 (2003) (describing the Condorcet Jury Theorem).
39. The Condorcet Jury Theorem has been generalized to analyze more than two alternatives. See Christian List & Robert E. Goodin, Epistemic Democracy: Generalizing the Condorcet Jury Theorem, 9 J. POL. PHIL. 277, 285 (2001) (extending Condorcet Jury Theorem to a case with k options). We look at two alternatives in this section to simplify the analysis.
40. See id. at 285.
41. One might think that legislators do evaluate factual states of the world as part of their assessment of legislation. In that case, the Condorcet theorem may apply more directly.
42. By efficiency we mean to suggest an independent, objective metric for the evaluation of legislation. For instance, the Office of Management and Budget undertakes a kind of efficiency review when it engages in cost-benefit analysis of regulations. See Elena Kagan, Presidential Administration, 114 HARV. L. REV. 2245, 2277-2283 (2001). Under our assumptions, one might think of members of Congress as engaging in a similar kind of analysis of legislation, voting for legislation where benefits would exceed costs and voting against legislation where costs would exceed benefits.
43. Willingness to pay does not require normative judgments but objective assessments of what individuals are willing to pay under various circumstances. It involves the evaluation of facts, not moral principles. See Avery Wiener Katz, Positivism and the Separation of Law and Economics, 94 MICH. L. REV. 2229, 2248 (1996) (arguing that willingness to pay is an observable fact that is "empirically measurable and hence objective").
44. We are not arguing and do not believe that legislators consider only the efficiency of legislation in deciding whether to vote for a bill. But we do believe that considerations of efficiency are an aspect of almost any legislator's vote. Thus, we are deploying an accuracy model to consider only an aspect of legislative voting. Of course, it may well be that the legislators have some other metric by which all evaluate legislation, like increasing equality of incomes. Given any agreed-upon criterion which can be measured objectively, the Condorcet framework can be used to assess the accuracy rates of legislators in determining whether legislation meets this criterion.
45. This point is explicit in our discussion of the Condorcet Jury Theorem, but it has also been specifically made in the law review literature. See Lewis A. Kornhauser, Adjudication by a Resource-Constrained Team: Hierarchy and Precedent in a Judicial System, 68 S. CAL. L. REV. 1605, 1626 n.43 (1995) (arguing this point in the context of judicial decision making).
46. See infra sections II.B.3, II.B.5.
47. The "representativeness" heuristic tends to make people extrapolate overconfidently about predicted characteristics of a class based upon a small sample size of which they happen to be aware. See Amos Tversky & Daniel Kahneman, Belief in the Law of Small Numbers, in JUDGMENT UNDER UNCERTAINTY: HEURISTICS AND BIASES 23, 24-25 (Daniel Kahneman et al. eds., 1982). If the sample consists of events rather than objects, the heuristic should tend to make people extrapolate in a similarly irrational manner from events of which they are aware to uncertain future events. Thus, individuals will tend to think that future events will resemble past events more than probability warrants.
48. For an important present-day application to people's mistaken attitudes about the stock market, see ROBERT J. SHILLER, IRRATIONAL EXUBERANCE 144 (2000).
49. See SHILLER, supra note 43.
50. Cf. U.S. CONST. art. I, § 10, cl. 1 (providing an example of a constitutional provision that might restrain such legislation by prohibiting states from "impairing the obligation of contracts").
51. See supra section II.A.3.
52. See supra notes 23-24 and accompanying text (arguing that there may be a race to entrench constitutional provisions).
53. A related danger is that aberrational elections can make the collective judgment of the legislature less reliable. A party may take office because of a scandal, like Watergate or some other unique event, and proceed to entrench legislation on matters that are unrelated to the reason for its election. An aberrational majority is likely to have a bias in favor of entrenching bad legislation. Because it recognizes that it may lose power in the future when the fallout from the scandal subsides, an aberrational majority is likely to focus on entrenching matters that an ordinary majority would not entrench.
54. The veil of ignorance guards against biases based on partiality (as opposed to cognitive failures, such as the heuristic of assuming that the future will be like the present). In particular, it guards against a bias that afflicts all decision makers-favoring the interests of oneself, one's family, friends, and socioeconomic class.
55. In fact, under the U.S. constitutional system, "[o]nly thirty-three proposals have been approved [by Congress] and sent to the states for ratification." See 1 CONG. QUARTERLY, GUIDE TO CONGRESS 360 (5th ed. 2000).
56. One might think that Article V's double supermajoritarian rule both in the Congress and in state legislatures increases the power of this effect. See U.S. CONST. art. V. Very few constitutional amendments are proposed by Congress and sent to the states for ratification. This restricted agenda assures substantial scrutiny of the proposals and a rich stream of information about their merits, raising the accuracy rate of the state legislators who are to consider them.
57. See McGinnis & Rappaport, supra note 16, at 100-103
58. See McGinnis & Rappaport, supra note 15, at 1167.
59. See McGinnis & Rappaport, supra note 16, at 100-103
60. See Steven G. Calabresi & Nicholas Terrell, The Fatally Flawed Theory of the Unbundled Executive, 93 MINN. L. REV. 1696, 1701-1702 (2009).
61. See McGinnis & Rappaport, supra note 15, at 1167-1168
62. We have also provided an insurance rationale for supermajority rule in the context of entrenchment. See McGinnis & Rappaport, supra note 16, at 67-68, 84-85. To summarize our argument: Uncertainty exists regarding the actual accuracy rates for constitutional proposals. But this very uncertainty provides additional support for a supermajority rule, particularly when citizens are risk averse. If we believe that average accuracy rates are likely to be higher than fifty percent but also recognize a nonnegligible chance that they will fall below fifty percent, it may be optimal to employ rule to reduce the risk of very bad results in the latter circumstances. Supermajority rule reduces risk because it reaches superior results when accuracy rates are less than fifty percent. It blocks more entrenchments where there are more bad entrenchments because of the legislators' faulty judgment. Of course, it will reach worse results when the accuracy rate is above fifty percent, as Condorcet itself suggests. But the assumption here is that accuracy rates are uncertain. Thus, even if supermajority rule may decrease the number of good entrenchments, it may still be beneficial because it reduces the range of outcomes and thus the possibility of really bad outcomes. Supermajority rule thus functions as insurance.
63. There is one other cost of using a multiple-voting-rule system-that is, a system that employs both supermajority and majority voting rules. That cost is substitution costs, but they are not large in this context. In this context, a constitutional system with multiple voting rules would permit legislators to substitute legislation that can be passed under majority rule for constitutional provisions that require a supermajority to enact when the majority-passed legislation would accomplish much the same function as the constitutional provision. Such substitution would reduce the benefits of supermajority rule. But such substitutions of legislation for constitutional provisions are difficult because entrenchment has unique characteristics that cannot easily be replicated through ordinary legislation. Only entrenched legislation takes priority over ordinary legislation and cannot be repealed through the ordinary legislative process. It would be impossible to use ordinary legislation to entrench a provision prohibiting interference with the freedom of speech or prohibiting future debt relief. Ordinary legislation could simply be repealed.
64. It has been observed that in some contexts even ordinary legislation can have some of the effects of entrenchment because it may create political interests that make it very hard to eliminate or even modify the original legislation. See Eric A. Posner & Adrian Vermeule, Legislative Entrenchment: A Reappraisal, 111 YALE L.J. 1665, 1687-1688 (2002). Thus, we cannot say that the substitution costs of a supermajority rule for entrenchment are zero, but nevertheless, such substitution is very unlikely to make supermajority rule less beneficial than majority rule for entrenchment, given the large benefits for supermajority entrenchment discussed in this Part.
65. See infra notes 68-90 and accompanying text. This defect includes the cost of blocking good entrenchments that would consist of overruling Supreme Court decisions, which we discuss below. See infra section VI.D.
66. We provide other reasons for the presumption of symmetry in John O. McGinnis & Michael B. Rappaport, Symmetric Entrenchment: A Constitutional and Normative Theory, 89 VA. L. REV. 385, 427-429 (2003).
67. It might be argued that early generations had more of an opportunity to add provisions, even though they were subject to largely the same formal enactment rules. Once the early generations added provisions that became popular, subsequent generations would not have an effective opportunity to replace those provisions. We discuss this objection below, where we argue that it is best for each generation to be subject only to the same formal enactment rules. See infra notes section IV.F and accompanying text.
68. See U.S. CONST. art. V.
69. Id.
70. Akhil Amar has suggested that the ratification process for the original Constitution was majoritarian because the state conventions ratified the Constitution by majority vote. See Akhil Reed Amar, The Central Meaning of Republican Government: Popular Sovereignty, Majority Rule, and the Denominator Problem, 65 U. COLO. L. REV. 749, 774 (1994). But properly characterizing the ratification vote requires a more discriminating analysis. From Amar's perspective, it may be proper to describe the ratification as majoritarian. His focus is on what was necessary at the state level to adopt the Constitution, and it is true that a majority of the convention in each state was all that was necessary. But our focus is on what was necessary at the national level to adopt the Constitution. It is clear that, from the perspective of the new nation, a nine-thirteenths supermajority rule was employed. Moreover, from the national perspective, that each state used a majority rule is not crucial. In a complex democratic system, rules can contain majoritarian elements, and yet their overall effect can be supermajoritarian. For instance, senators are elected by a majority in each state, and yet no one would doubt that the requirement that a treaty receive two-thirds vote for advice and consent is supermajoritarian. U.S. CONST. art. II, § 2, cl. 2. So it is with the ratification of the Constitution.
71. Five states demanded a Bill of Rights as part of the ratification process. See, e.g., JAMES W. ELY, JR., THE GUARDIAN OF EVERY OTHER RIGHT: A CONSTITUTIONAL HISTORY OF PROPERTY RIGHTS 52 (2d ed. 1998). If majority rule had been in place it would have been easier to ignore these states, but under the Constitution's requirement of nine of thirteen states for ratification of the Constitution, support from these states was necessary for approval.
72. It is well established that the doctrine of federalism emerged as a compromise between nationalist and more states-oriented forces. GORDON WOOD, THE CREATION OF THE AMERICAN REPUBLIC, 1776-1787, at 525-526 (1969). At the Convention, the Nationalists introduced the Virginia Plan; in response, the small states proposed the more states-oriented New Jersey Plan.
73. See RICHARD BEEMAN, PLAIN HONEST MEN: THE MAKING OF THE AMERICAN CONSTITUTION 88 (2009). Subsequently, it was unclear which side, given the shifting attendance of delegates, had a majority, id. at 172-173, and a committee was appointed to find a compromise that formed the basis of the eventual Constitution, id. at 189. While it was possible that either the nationalists or their opponents might have gained a majority in the Convention or among the ratifying states for their plans, the need to secure a consensus of the ratifying states (whatever the precise consensus requirement turned out to be) would have made such a strategy extremely risky. Instead, the Connecticut compromise allowed the Constitution to secure the unanimous support of the twelve states at the Convention and to move forward to the ratification process with significant momentum.
74. MICHAEL KAMMEN, A MACHINE THAT WOULD GO OF ITSELF: THE CONSTITUTION IN AMERICAN CULTURE 162 (1986) (quoting William Gladstone).
75. As James Madison himself stated, Article V "guards equally against that extreme facility, which would render the Constitution too mutable; and that extreme difficulty, which might perpetuate its discovered faults." See THE FEDERALIST No. 43, at 278 (James Madison) (Clinton Rossiter ed., 1961).
76. See id.
77. Our focus here is not to decide absolutely whether particular amendments are bad or good but rather to assess how Article V has performed overall in history. Our criteria for evaluating whether amendments are good or bad consist of two components, one abstract and one more specific. First, we consider whether the amendment comports with the outline of the liberal constitution set forth in Part I. Second, because an amendment must be consistent with the long term preferences of society, we evaluate the amendment in light of the popular preferences of its time as well as subsequent developments. Thus, we do not assess a provision by reference to some set of narrow political principles like socialism or libertarianism. Moreover, it generally becomes easier to evaluate amendments with the passage of time and with a perspective outside of the controversies that gave rise to these proposals than it would be to make a similar judgment on contested contemporary issues of constitutional law. We have offered strong theoretical reasons to believe that relatively stringent supermajority rule is a good way of generating a constitution and superior to other methods. But we would be foolish to refuse to consider the results of this process in our evaluation of the method. Architecture progresses by attention to the laws of physics and the heuristics of engineering, but in evaluating building processes, we must consider how buildings constructed by these processes have stood the test of time.
78. See 1 Pub. Res. 3, 1st Cong., 1 Stat. 97 (1789).
79. See AKHIL REED AMAR, THE BILL OF RIGHTS 15-16 (1998).
80. Id. at 16-17. Moreover, there does not seem any reason to believe that Congress is more likely to abuse discretion in setting the number of its members man in using its own powers.
81. See J. Res. 2, 11th Cong., 2 Stat. 613 (1810).
82. See J. Res. 13, 36th Cong., 12 Stat. 251 (1861).
83. See H.R.J. Res. 184, 68th Cong., 43 Stat. 670 (1924).
84. See H.R.J. Res. 208, 92d Cong., 86 Stat. 1523 (1972).
85. See H.R.J. Res. 554, 95th Cong., 92 Stat. 3795 (1978).
86. As has been proposed by the District of Columbia-Maryland Reunion Act, H.R. 1858, 110th Cong. (2007).
87. We could continue our assessment by discussing amendments that did not come quite as close to passage as these six, such as those that passed one house but foundered in the other. We believe, however, that the arguments above are sufficient. It is worth noting, though, that the proposed flag-burning amendment would have been proposed by the Congress had it received one more vote in the Senate. See H.R.J. Res. 10, 109th Cong. (2009) (passed by House); S.J. Res. 12, 109th Cong. (2006) (rejected in Senate by one vote). A slightly less stringent amendment process might have permitted this amendment to pass, despite the fact that it appears to be a largely symbolic piece of legislation rather than an attempt to solve a real social problem.
88. See U. S. CONST. amend. XVII. In 1910, the year the amendment was proposed, only fourteen of the thirty Senators had been designated by a popular vote before being formally appointed by their state legislatures. See GEORGE H. HAYNES, THE SENATE OF THE OF THE UNITED STATES: ITS HISTORY AND PRACTICE 104 (1938).
89. See U.S. CONST. amend. XIX.
90. Any mention of the Civil War Amendments raises the questions of whether these amendments were legal and whether they represented a national consensus. Scholars have vigorously disagreed about these matters. See 2 BRUCE ACKERMAN, WE THE PEOPLE: TRANSFORMATIONS 99-119, 207-234 (1998);
91. AKHIL REED AMAR, AMERICA'S CONSTITUTION: A BIOGRAPHY 366-368 (2005);
92. John Harrison, The Lawfulness of the Reconstruction Amendments, 68 U. Chi. L. Rev. 375, 451-452 (2001). If the Reconstruction Amendments were legal-either under the Amar or Harrison theories-this analysis would largely address the concerns about their consensus character. For example, suppose, as Amar suggests, that many of the confederate states were properly excluded from voting because they did not have the constitutionally required republican form of government. Then, the amendments would have been legal, having passed with the requisite constitutional supermajorities. It is true that the consensus behind the amendments would have excluded much of the former Confederacy, but the Constitution would have made the judgment that nonrepublican states should be excluded. When a portion of the country is no longer republican, there is a tradeoff between requiring an extended consensus and limiting input to those with republican values, and the Constitution would have chosen republican values over a full consensus. One might question whether the Constitution's decision was optimal, but it was at least plausible, and thus on this view, the Reconstruction amendments were legal.
93. Even if one did conclude that the Constitution's approach was problematic or that the Reconstruction Amendments had been illegal, following the amendments might still be justifiable under our theory as a normative matter. The amendments operated to correct a supermajoritarian failure-the original Constitution's exclusion of blacks from political participation-and therefore might be justified on that basis. See infra Part V. If the Judiciary enforced the amendments under this theory, that enforcement would represent a judicial correction of the original exclusion of African-Americans.
94. See, e.g., BRUCE ACKERMAN, WE THE PEOPLE: FOUNDATIONS (1991);
95. DAVID A. RICHARDS, CONSCIENCE AND THE CONSTITUTION: HISTORY, THEORY, AND LAW OF THE RECONSTRUCTION AMENDMENTS (1993).
96. See infra section VI.A.
97. Nor does the passage of a single amendment that is now widely thought to be a mistake discredit the supermajoritarian process. Our point is not that the process is perfect, merely that it is good overall and superior to other processes. Moreover, the Eighteenth Amendment provides another example of the ability of the amendment process to overcome vested interests. Women were not yet fully represented in the polity, and yet prohibition was widely seen as a response to women's concerns about the effects of excessive drinking, particularly the costs imposed on women through the alcohol's fostering of spousal violence and its dissipation of male earnings that would otherwise support the family. See William K. Eskridge Jr., Some Effects of Identity-Based Social Movements on Constitutional Law in the Twentieth Century, 100 MICH. L. REV. 2062, 2113 (2002).
98. See U.S. CONST. amends. XX, XXV.
99. See Donald S. Lutz, Toward a Theory of Constitutional Amendment, in RESPONDING TO IMPERFECTION: THE THEORY AND PRACTICE OF CONSTITUTIONAL AMENDMENT 237, 273 (Sanford Levinson ed., 1995).
100. Others have noted that lenient amendment processes are also used as a justification for judicial activism because the less strict process allows citizens to respond to judicial activism. See A.E. Dick Howard, State Courts and Constitutional Rights in the Day of the Burger Court, 62 VA. L. REV. 873, 939 (1976).
101. See infra section IV.C.2.
102. Compare Akhil Reed Amar, The Consent of the Governed: Constitutional Amendment Outside Article V, 94 COLUM. L. REV. 457, 461 (1994) (positing that Article V's unanimity requirement could potentially be bypassed through a series of two amendments),
103. With Stephen M. Griffin, The Nominee Is ⋯ Article V, 12 CONST. COMMENT. 171, 173 (1995).
104. The Balanced Budget Amendment is an example of an amendment that probably failed on account of this feature of the Constitution because that amendment would have restricted the power of Congress to run deficits and run up debt, a not inconsiderable power and one that aids members of Congress in rewarding interest groups, thereby advancing electoral prospects.
105. See Michael B. Rappaport, Reforming Article V: The Problems Created by the National Convention Amendment Method and How To Fix Them, 96 VA. L. REV. (forthcoming 2010). There is an argument that much of the fear of a runaway convention is not rational, because the proposals of any convention would still have to surmount the hurdle of three-quarters of state legislatures to be enacted.
106. See Ronald D. Rotunda & Stephen J. Safranek, An Essay on Term Limits and a Call for a Constitutional Convention, 80 MARQUETTE L. REV. 227, 239 (1996). But an amendment not favored by state legislatures could pass in several different ways, including because the proposed amendment developed momentum, changed circumstances made the amendment more popular, Congress chose to have the ratification decision made by state conventions rather than state legislatures, or the runaway convention chose to employ a less than the three-quarters ratification rule and this decision was accepted by the nation.
107. Notable recent examples of constitutional amendments that would have constrained Congress but were not passed were term limits, the line item veto, and a balanced budget amendment. See Rappaport, supra note 93.
108. See JOURNALS OF THE CONTINENTAL CONGRESS, 1774-1789, at 74 (Worthington Chauncey Ford et al. eds., 1904-1937), available at http://memory.loc.gov/cgi- bin/query/r?ammem/hlaw:@field (DOCID+@lit(jc03225)) (noting approval of the Congress's recommendation for a convention). It is unclear exactly how many states were present and voting, but it appears that all states that voted ultimately supported the recommendation;
109. See also EDMUND CODY BURNETT, THE CONTINENTAL CONGRESS 768-769 (1941) (describing approval of recommendation).
110. See MAX FARRAND, THE FRAMING OF THE CONSTITUTION OF THE UNITED STATES 11-12 (1913) (discussing Rhode Island's unique failure to attend).
111. See CLINTON ROSSITER, 1787: THE GRAND CONVENTION 54-55 (1966).
112. See S. Doc. No.170, at 13 n.m (1953) (indicating that thirty-nine of the fifty-five delegates who attended signed the document).
113. See CATHERINE DRINKER BOWEN, MIRACLE AT PHILADELPHIA: THE STORY OF THE CONSTITUTIONAL CONVENTION, MAY TO SEPTEMBER, 1787, at 13 (2d ed. 1986).
114. Another way of looking at the Framers' choice is that they chose nine states because that was the number of states the approval of which was required for important decisions under the Articles of Confederation. See AMAR, supra note 83, at 311. In that case the three-quarters requirement in Article V was chosen as the closest round fraction to reflect the proportion of states required for ratification in Article VII rather than the other way around.
115. There is a third possible asymmetry between the constitutional origination and the amendment processes. In the former, only those states who agreed to the Constitution were bound, but in the amendment process those states who decline ratification are nevertheless bound if the amendment obtains ratification by three-quarters of the states. One might argue that this asymmetry will tend to make the original Constitution less good than amendments because in the origination process, states were under pressure to agree to the Constitution; the alternative was being cast out of the union with possible dire consequences. But state decisions in the negative were understood not to be final, and North Carolina and Rhode Island in fact assented after they originally rejected the Constitution. See BEEMAN, supra note 66, at 391-392, 405. Thus, states did not in reality face such a harsh choice. Moreover, the sponsors of the Constitution wanted all the states to ratify it because the Union would then be stronger. As a result, the incentives to propose a good original constitution in this respect may have been even greater than those to propose a good amendment because in the latter case it does not matter if the ultimate support is not unanimous. Thus, our judgment is that the difference in the binding nature of the origination and the amendment processes does not overall detract significantly from the rough symmetry between the two.
116. Cf. Kent Greenawalt, The Rule of Recognition and the Constitution, 85 MICH. L. REV. 621, 639 (1987) (noting that officials of small states predisposed to ratification might nevertheless have been wary of taking the Constitution as governing law if certain major states had failed to ratify).
117. See infra section VI.C.
118. Both of these problems arguably affected the constitutional enactment process. When Congress and the states supported a convention, they may have expected that the results would be a true amendment of the Articles and would go through the Articles' amendment process. Moreover, when the Philadelphia Convention chose a nine-thirteenths ratification requirement, delegates could have done so with an eye towards the degree of support they expected for the constitution that they were writing.
119. Finally, we should also note one clear departure from symmetry that we have already discussed in the section on reasonable stringency-the provision that prohibits amending the states' equal representation in the Senate. Whether this provision can be eliminated with two amendments or not, the process for eliminating it is certainly stricter than was the process for enacting it.
120. See, e.g., William E. Scheuerman, Constitutionalism in an Age of Speed, 19 CONST. COMMENT. 353, 354-355 (2002).
121. For a discussion of the Framers' original methods, see John O. McGinnis & Michael B. Rappaport, Original Methods Originalism: A New Theory of Interpretation and the Case Against Construction, 103 Nw. U. L. REV. 751, 788-793 (2009) (describing in general terms the content of the original methods of interpretation).
122. Our defense of using the original enactors' interpretive rules is thus not based either on the circular argument that we should follow the intentions of the Framers, because that is what they wanted, or on the more subtly circular
123. 128 S. Ct. 2783 (2008).
124. U.S. CONST. amend. II.
125. See, e.g., Heller, 128 S. Ct. at 2790 n.3 (citing Copeman v. Gallant, (1716) 24 Eng. Rep. 404 (Ch.)). One reason for the venerable canon is that "the remedy often extends beyond the particular act or mischief which first suggested the necessity of the law." JOEL PRENTISS BISHOP, COMMENTARIES ON THE WRITTEN LAWS AND THEIR INTERPRETATION § 51 (Boston, Little, Brown, & Co. 1882) (quoting Rex v. Marks, (1802) 3 East 157, 165 (K.B.)).
126. 128 S. Ct. at 2789-2790
127. A difficulty for Justice Stevens's dissent is that he cannot rebut this rule. See id. at 2822, 2825 (Stevens, J. dissenting) (relying on indications of congressional intent other than the prefatory clause canon). Thus, he can only deploy the prefatory language if he can show that it clarifies an ambiguity in the phrase "the right of the people to keep and bear arms."
128. See William K. Kelley, Avoiding Constitutional Questions as a Three-Branch Problem, 86 CORNELL L. REV. 831, 897-898 (2001).
129. See infra section IV.C.1.
130. For an example of a view that we should understand the Framers' meaning through the prism of modern philosophy, see Lawrence B. Solum, Constitutional Texting, 44 SAN DIEGO L. REV. 123, 150 (2007) ("[O]ur understanding of constitutional meaning should be modeled on Grice's conception of sentence meaning ⋯.").
131. McGinnis & Rappaport, supra note 107, at 788-793
132. See McCulloch v. Maryland, 17 U.S. (4 Wheat.) 316, 406-408 (1819).
133. See Philip A. Hamburger, The Constitution's Accommodation of Social Change, 88 MICH. L. REV. 239, 287 (1989).
134. See Robert A. Goldwin, Comment, Original Intent and the Constitution, 47 MD. L. REV. 189, 194-195 (1987).
135. Moreover, the passage of time may strengthen rather than weaken the rationale for some provisions of the Constitution. The Framers did not understand competition among the states to be a benefit sustained by federalism, but in the modern world that has become a strong argument.
136. See Hamburger, supra note 119, at 300.
137. One also wants a system that prevents any entity from having a veto on the amendments. We discussed Article V's imperfections in that regard: Congress must effectively propose amendments, potentially giving it a stranglehold on the process. See supra notes 92-94 and accompanying text.
138. See John O. McGinnis, Reviving Tocqueville's America: The Rehnquist Court's Jurisprudence of Social Discovery, 90 CAL. L. REV. 485, 566 (2002). On Condorcet grounds, small numbers are less likely to reach accurate results.
139. See Adrian Vermeule, Common Law Constitutionalism and the Limits of Reason, 107 COLUM. L. REV. 1482, 1493 (2007).
140. See Vermeule, supra note 124, at 1499-1500 (making a similar point about common law constitutionalism).
141. See William M. Landes & Richard A. Posner, Rational Judicial Behavior: A Statistical Study 15-16 (John M. Olin Law & Econ. Working Paper, Paper No. 404, 2008), available at http://ssrn.com/abstract=1126403 (noting that Justices have recently become more polarized and susceptible to block ideological voting).
142. It is a fundamental premise of the constitutional separation of powers that each branch will seek to aggrandize its own powers. See, e.g., THE FEDERALIST NO. 51, at 322 (James Madison) (Clinton Rossiter ed., 1961) (suggesting that tendency of aggrandizement in the branches is rooted in human nature);
143. See also Jonathan R. Macey, Judicial Preferences, Public Choice, and the Rules of Procedure, 23 J. LEGAL STUD. 627, 631 (1994) (arguing that judges will have an inherent preference for discretion-maximizing rules).
144. See, e.g., Tracey George, Court Fixing, 43 ARIZ. L. REV. 9, 38 (2001) (judges deploy discretion in part to pursue their own preferences).
145. It is important to note that this tendency is not malevolent. Judges naturally want to reach a result that is fair, and their preferences are one window into the often obscure landscape of fairness.
146. Compare Griswold v. Connecticut, 381 U.S. 479, 530 (1965) (Stewart, J., dissenting) (arguing that there is no right to privacy in the Constitution grounded in penumbras), with Seminole Tribe of Fla. v. Florida, 517 U.S. 44, 72 (1996) (Rehnquist, C.J.) (state sovereign immunity is protected under the Constitution through penumbras).
147. Compare Lochner v. New York, 198 U.S. 45, 75-76 (1905) (Holmes, J., dissenting) (arguing that the Fourteenth Amendment does not "embody a particular economic theory" and therefore does not protect an absolute right to contract from state interference), with Roe v. Wade, 410 U.S. 113, 152-53 (1973) (finding an unenumerated right to abortion in the Constitution).
148. Some scholars have suggested that a virtue of an informal amendment process, or what we call judicial updating, is the fluidity it permits. See, e.g., Reva B. Siegel, Constitutional Culture, Social Movement Conflict and Constitutional Change: The Case of the De Facto ERA, 94 CAL. L. REV. 1323, 1327-1328 (2006). We have shown why just the opposite is the case. It is the formal constitution-making process that brings the consensus that will establish affection for the Constitution. Even apart from this point, claims like that of Professor Siegel fail to show how such an informal process is likely to lead to high quality constitutional provisions.
149. See supra notes 127-128 and accompanying text.
150. See Heather K. Gerken, The Hydraulics of Constitutional Reform: A Skeptical Response to Our Undemocratic Constitution, 55 DRAKE L. REV. 925, 939 (2007).
151. Recently, scholars have argued that judicial updating constitutes part of a larger social process in which political actors other than the Judiciary help advance the ideas that are ultimately used to update the Constitution. See id. at 935. Although this is undoubtedly true as a factual matter, it does not change our normative critique. The existence of this social process does not erase the Judiciary's necessary discretion to choose which ideas to incorporate into the Constitution from the many advanced to them.
152. For instance, the Justices have acted to protect rights of minorities only when they have significant support. When minorities do not have the requisite level of political support, the Court has been unwilling to incur the attacks that protecting the minorities would occasion. See Michael J. Klarman, Rethinking the Civil Rights and Civil Liberties Revolutions, 82 VA. L. REV. 1, 16-18 (1996).
153. Some law professors and political scientists have recently argued that the Court follows public opinion. See generally, BARRY FRIEDMAN, THE WILL OF THE PEOPLE: HOW PUBLIC OPINION HAS INFLUENCED THE SUPREME COURT AND SHAPED THE MEANING OF THE CONSTITUTION (2009) (arguing that the Supreme Court is a political actor which responds to constituencies). We are skeptical of this claim. That the Court is an institution constrained by other institutions should not be confused with the claim that it follows popular will in the manner of a legislature. First, the legislature is structured to reflect popular opinion, while the Judiciary is insulated from it.
154. See JESSE H. CHOPER, JUDICIAL REVIEW AND THE NATIONAL POLITICAL PROCESS: A FUNCTIONAL RECONSIDERATION OF THE ROLE OF THE SUPREME COURT 3-70 (1980). Second, consistent with this analysis, some political scientists have suggested that there is no convincing proof of the majoritarianism of the Supreme Court.
155. See JEFFREY A. SEGAL & HAROLD J. SPAETH, THE SUPREME COURT AND THE ATTITUDINAL MODEL 331-332 (1993). Finally, there are whole areas of the law in which the Judiciary is patently out of step with the public. Dramatically, for instance, even a Supreme Court made up of Republican appointees offers little or no support for prayer in public schools, although huge majorities support the policy.
156. See Michael Comiskey, The Rehnquist Court and American Values, 77 JUDICATURE 261, 265-266 (1992). The current Court appears regularly to act against the weight of popular sentiment.
157. See, e.g., Boumediene v. Bush, 128 S. Ct. 2229, 2240 (2008) (granting habeas rights for terrorist suspects held at Guantanamo); Kennedy v. Louisiana, 128 S. Ct. 2641, 2646 (2008) (prohibiting death penalty for child rapists); Kelo v. City of New London, 545 U.S. 469, 489-90 (2005) (permitting taking of homes for commercial development); Grutter v. Bollinger, 539 U.S. 306, 343 (2003) (upholding racial and ethnic preferences for admission to universities);
158. See also Lawrence Baum & Neal Devins, Why the Supreme Court Cares About Elites, Not the American People, 98 GEO. L. J. 1515 (forthcoming Aug. 2010) (arguing that Justices are more influenced by elite opinion than by public opinion).
159. For similar reasons, a jurisprudence suggesting that the Court should simply defer to legislative majorities would not maintain a good constitution over time. For an argument in favor of such a jurisprudence, see Jeffery Rosen, Originalism and Pragmatism: False Friends, 31 HARV. J.L. & PUB. POL'Y. 937 (2008). As we argued above, legislative majorities would not entrench good constitutional principles. Thus, the Court cannot sustain a good constitution on the basis of a political theory of deference. Moreover, unless the Court was to defer absolutely to the Legislature and thus abnegate judicial review, a jurisprudence of deference is not a complete jurisprudence. The Judiciary must first determine the meaning of a constitutional provision before it can decide whether the Legislature's interpretation is close enough to the original meaning to warrant deference. The Legislature, in fulfilling its own duty to pass only constitutional legislation, also must determine the Constitution's meaning. It cannot simply defer to its own authority. Thus, whatever one's view of the proper scope of deference, a meaningful theory beyond simple deference is necessary to constitutional jurisprudence. The original methods approach supplies that normative theory for a good constitution. If the Judiciary simply defers to the Legislature and the Legislature defers to its own principles, the original meaning and the benefits it provides will be lost over time.
160. For example, while we argue below that the nation would have passed an amendment expanding the Commerce Clause during the New Deal, see infra section IV.C.2.b., once the Supreme Court broadened the Clause on its own, there was unlikely to be enough support for an amendment to cut it back, even though the nation would not have expanded the Clause as broadly as the Supreme Court did.
161. See, e.g., Wickard v. Filburn, 317 U.S. 111, 128-129 (1942) (holding that Congress may regulate local activities not regarded as commerce if those activities have a substantial economic effect on interstate commerce).
162. See DAVID E. KYVIG, EXPLICIT AND AUTHENTIC ACTS: AMENDING THE U.S. CONSTITUTION, 1776 - 1995, at 305, 314 (1996) (arguing that Roosevelt could have passed constitutional amendments expanding federal power).
163. For discussion of why the pre-New Deal Court was correct in its construction of the Clause, see RANDY BARNETT, RESTORING THE LOST CONSTITUTION: THE PRESUMPTION OF LIBERTY 274-318 (2004);
164. Richard A. Epstein, The Proper Scope of the Commerce Power, 73 VA. L. REV. 1387 (1987).
165. U.S. SENATE, PARTY DIVISION IN THE SENATE, 1789-PRESENT, http://www.senate.gov/pagelayout/history/one-item-and-teasers/partydiv.htm (last visited Jan. 15, 2010) (referring to the 75th Congress).
166. OFFICE OF THE CLERK OF THE U.S. HOUSE OF REPRESENTATIVES, PARTY DIVISIONS OF THE HOUSE OF REPRESENTATIVES (1789 TO PRESENT), http://clerk.house.gov/art- history/house-history/partyDiv.html (last visited Jan. 15, 2010) (referring to the 75th Congress).
167. See MICHAEL J. DUBIN, PARTY AFFILIATIONS IN STATE LEGISLATURES: A YEAR BY YEAR SUMMARY, 1976-2006, at 28, 36, 40, 68, 82, 94, 122, 129, 137, 144, 154, 167, 174, 189 (2007) (Kansas, Maine, Massachusetts, New Hampshire, North Dakota, South Dakota, and Vermont were entirely controlled by Republicans, but the party only partially controlled California, Connecticut, Delaware, New Jersey, New York, Oregon, and Rhode Island). Iowa had one house controlled by Democrats and the other evenly divided. Id. at 65. In Wisconsin, the Progressives were the largest party in each house. Id. at 202.
168. Wallace S. Sayre, Major Party Platforms of 1936, CURRENT HIST., Aug. 1936, at 52.
169. See Rafael Gely & Pablo T. Spiller, The Political Economy of Supreme Court Constitutional Decisions: The Case of Roosevelt's Court-Packing Plan, 12 INT'L REV. L. & ECON. 45, 63 (1992).
170. For a discussion of the extent to which the Court-packing plan and related external pressures were responsible for the Court's switch, see Mark Tushnet, The New Deal Constitutional Revolution: Law, Politics, or What?, 66 U. CHI. L. REV. 1061, 1075-1076 (1999) (reviewing BARRY CUSHMAN, RETHINKING THE NEW DEAL COURT (1998)).
171. For an argument that internal, doctrinal factors were responsible, see CUSHMAN, supra, at 3-7. Whether the Supreme Court decisions of 1937 can be reconciled with prior doctrine, later decisions like Wickard v. Filburn, 317 U.S. 111 (1942), cannot be so reconciled. Thus, we believe that judicial failure to enforce the Constitution resulted from political changes, including judicial appointments, even if there can be legitimate arguments about when that judicial failure occurred.
172. Thus, our argument that the original meaning of the Commerce Clause accommodates change should not be confused with the spurious argument that cases like United States v. Darby, 312 U.S. 100 (1942), and Wickard v. Filburn, 317 U.S. 111 (1942), accommodated change within the Constitution's original meaning. Although the meaning of the Commerce Clause is static, it accommodates change by permitting the regulation of more activities as commerce increases with the growth of the economy. The main problem' with the New Deal cases is that they changed the original meaning of the term "commerce." The difference between the originalist vision and that of New Deal Justices is that between a constant principle that changes the scope of its application as the facts of the world change and a variable principle that itself mutates in meaning.
173. Gerard N. Magliocca, The Child Labor Amendment and the Court-Packing Plan 3-4 (Nov. 21, 2009) (unpublished manuscript), available at http://papers.ssrn. com/sol3/papers.cfm?abstract-id=1510865.
174. Id. at 6.
175. Id. at 48-50.
176. All states had child labor laws at the time. See Hammer v. Dagenhart, 247 U.S. 251, 275 (1918). Many were equally stringent or more stringent than federal law. See S. Comm. on the Judiciary, 68th Cong., Child Labor Amendment 23 (Comm. Print 1924) (noting that the labor standards of thirteen states are equal to or higher than the standards of federal law).
177. See supra note 146 and accompanying text.
178. See 2 ACKERMAN, supra note 83, at 338 (suspecting the President would have wanted an amendment "'empowering Congress to legislate for the general welfare'" in the absence of effective regulation by the states).
179. See JOSEPH P. LASH, FROM THE DIARIES OF FELIX FRANKFURTER 58 (1975) ("[H]e feared that an amendment to the Constitution would diminish the intrinsic character of a document that was intended to endure and that he profoundly believed had ample resources within its original terms to meet the changing needs of successive generations.").
180. See JANE MCCARTHY WITH ALICE SHORETT, NEGOTIATING SETTLEMENTS: A GUIDE To ENVIRONMENTAL MEDIATION xi (1984) ("The 1970s, tagged the Decade of the Environment, created unprecedented awareness of the need to protect and preserve our natural resources.").
181. Consistent with this hypothesis, there has been stronger interest group participation in the Supreme Court nomination process since the 1950s. See MICHAEL GERHARDT, THE FEDERAL APPOINTMENT PROCESS 219 (2000).
182. Cf. Jeremy Paul, Commentary, Campaign Reform in the Twenty-first Century: Putting Your Mouth Where the Money Is, 30 CONN. L. REV. 779, 799 (1998) (suggesting that some of our legal culture is simply opposed to constitutional amendments).
183. See Ronald Dworkin, Comment, in ANTONIN SCALIA, A MATTER OF INTERPRETATION: FEDERAL COURTS AND THE LAW 115, 119-123 (Amy Gutmann ed., 1997) (suggesting that general provisions may have dynamic application over time).
184. See JOHN R. VILE, ENCYCLOPEDIA OF CONSTITUTIONAL AMENDMENTS, PROPOSED AMENDMENTS, AND AMENDING ISSUES, 1789 - 2002, at 177 (2d ed. 2003).
185. See Frontiero v. Richardson, 411 U.S. 677, 688 (1973) (Brennan, J., plurality opinion) (applying strict scrutiny to strike down a law that discriminated against women while the amendment was pending); Reed v. Reed, 404 U.S. 71, 75-76 (1971) (striking down discriminatory laws and effectively applying intermediate scrutiny while the Equal Rights Amendment was being considered). In Frontiero, Justice Powell concurred on other grounds but argued against strict scrutiny to allow the amendment a chance to settle the issue. Frontiero, 411 U.S. at 691-692 (Powell, J., concurring in judgment).
186. 404 U.S. at 72-73.
187. Id.
188. Frontiero, 411 U.S. at 682.
189. See Suzanne Sangree, Title IX and the Contact Sports Exemption: Gender Stereotypes in a Civil Rights Statute, 32 CONN. L. REV. 381, 412 (2000).
190. See Richard A. Posner, Professionalisms, 40 ARIZ. L. REV. 1, 7 (1998) (describing the Warren Court era as a period of judicial activism).
191. See Griffin, supra note 91, at 172 (citing a study supporting many commentators' belief that the Constitution is relatively difficult to change and therefore has been amended in most cases via judicial interpretation).
192. See McGinnis & Rappaport, supra note 107, at 776-777 (citing the lack of evidence of the use of construction in the early republic).
193. For a discussion of the strength of the originalist interpretative methods at the time of the founding, see id. at 788-93. Originalism has been undergoing a revival since the 1970s. JONATHAN O'NEILL, ORIGINALISM IN AMERICAN LAW AND POLITICS 111-160 (2005) (discussing the revival of originalism). Both liberals and conservatives adhere to this doctrine.
194. See, e.g., Akhil Amar, The Supreme Court 1999 Term, Foreword: The Document and the Doctrine, 114 HARV. L. REV. 26 (2000) (providing a sustained argument for privileging original meaning over mere Supreme Court doctrine).
195. See Richard A. Posner, What Do Judges and Justices Maximize? (The Same Thing Everybody Else Does), 3 SUP. CT. ECON. REV. 1, 28-29 (1993) (arguing that judges maximize their utility in part by following the rules). Therefore, if originalism becomes the prevailing rule, judges will want to follow it.
196. See Adrian Vermeule, Foreword: System Effects and the Constitution, 123 HARV. L. REV. 4, 56 (2009).
197. Id.
198. See id. at 55-56.
199. For the original statement of the theory of the second best, see R.G. Lipsey & Kelvin Lancaster, The General Theory of Second Best, 24 REV. ECON. STUD. 11, 11-12 (1956-1957).
200. For the original statement of the second best in the constitutional context, see Peter B. McCutchen, Mistakes, Precedent, and the Rise of the Administrative State: Toward a Constitutional Theory of the Second Best, 80 CORNELL L. REV. 1, 17 (1994) (arguing that the legislative veto, although inconsistent with the original meaning of the Constitution, compensates for the divergence from the constitutional baseline caused by the Court's allowance of open-ended delegations).
201. See id. ("[T]wo wrongs may not quite make a right ⋯."); see also Lipsey & Lancaster, supra note 175, at 12, 28-31 (stating that "nothing can be said about the direction or the magnitude of the secondary departures from optimum conditions made necessary by the original non-fulfillment of one condition" and that making such judgments is extremely difficult).
202. As our discussion of judicial correction in Part V shows, we are open to consequentialist considerations that would overcome our rule-consequentialist arguments for originalism. We thus agree with Professor Vermeule that rule-consequentialist arguments must be tested by such calculations. See Vermeule, supra note 172, at 55, 62. We disagree that second-best arguments offer a reason to do so.
203. See supra section IV.C.1.
204. We have offered a theory of the precedent that would encourage originalist Justices to overrule many nonoriginalist precedents but would still authorize Justices to keep precedents that are necessary for the stability and desirable operation of our system. See John O. McGinnis & Michael B. Rappaport, Reconciling Originalism and Precedent, 103 Nw. U. L. REV. 803, 805 (2009). Professor Vermeule, however, appears to think a theory for precedent in a consequentialist originalism is impossible because it will either allow too many nonoriginalist precedents, thus vitiating originalism, or too few nonoriginalist precedents, thereby requiring the overturning of fixed features of the legal landscape. Vermeule, supra note 172, at 56-57. But we have already answered this type of criticism in our published work, where we argue in favor of an intermediate approach to precedent that avoids overturning a class of essential precedents, but allows the overturning of other precedents. See McGinnis & Rappaport, supra, at 835-836
205. See Vermeule, supra note 172, at 60-63.
206. Vermeule overlooks such possible differences between originalism and other jurisprudences in this strategic calculus. See id.
207. See, e.g., Jeffrey Rosen, Conservatives v. Originalism, 19 HARV. J.L. & PUB. POL'Y 465, 465 (1996) (criticizing Scalia for substituting conservative values for originalism).
208. Cf. John O. McGinnis, Impeachable Defenses, 95 POL. REV. 27, 27-29 (1999) (showing that under the intense public scrutiny of congressional hearings even nonoriginalist scholars resort to originalism).
209. See Vermeule, supra note 172, at 61.
210. See Richard H. Fallon, Jr., "The Rule of Law" as a Concept in Constitutional Discourse, 97 COLUM. L. REV. 1, 48 (1997).
211. See supra section IV.C.1.
212. See Gerken, supra note 134.
213. See Michael W. McConnell, Textualism and the Dead Hand of the Past, 66 GEO. WASH. L. REV. 1127, 1127 (1998).
214. Id.
215. As we have noted elsewhere, the dead hand argument in its strongest form seems inconsistent with all constitutionalism that prevents an existing majority from taking action it desires. See McGinnis & Rappaport, supra note 19, at 796. Even most living constitution approaches do not confer unlimited discretion on the current Court to ignore previous decisions but instead accept some limitations that decisions of earlier generations have made. See, e.g., RONALD DWORKIN, LAW'S EMPIRE 355-99, 412-413 (1986);
216. Jack M. Balkin, Abortion and Original Meaning, 24 CONST. COMMENT. 291, 293-294 (2007).
217. Thus, our argument provides a functional answer to the argument that the passage of time undermines the case for applying originalism, see Michael C. Dorf, Integrating Normative and Descriptive Constitutional Theory: The Case of Original Meaning, 85 GEO. L.J. 1766, 1819-1820 (1997). A supermajoritarian process designed for entrenchment compensates for the passage of time. Moreover, it is better to enforce the results of the past supermajoritarian process than to enforce the decisions of current majorities in the legislature or Supreme Court.
218. See McGinnis & Rappaport, supra note 19, at 797.
219. The third alternative, departing from some of the provisions of an imperfect constitution, can take several different forms, depending on which institution does the departing. Here we focus on departure by the Judiciary because it is the most commonly proposed alternative.
220. Akhil Amar, for instance, believes that the Constitution could be amended outside of Article V by a national referendum. See Akhil Reed Amar, Philadelphia Revisited: Amending the Constitution Outside Article V, 55 U. CHI. L. REV. 1043, 1044 (1988). Others have critiqued the historical support for that view.
221. See, e.g., Henry Paul Monaghan, We the People[s], Original Understanding, and Constitutional Amendment, 96 COLUM. L. REV. 121, 128 (1996). In addition, such a national referendum lacks the supermajoritarian features that make for beneficial constitutionalism.
222. See supra notes 97-100 and accompanying text.
223. For a defense of the process by one who has so suggested, see SANFORD LEVINSON, OUR UNDEMOCRATIC CONSTITUTION: WHERE THE CONSTITUTION GOES WRONG (AND HOW WE THE PEOPLE CAN CORRECT IT) 24, 165, 173 (2006).
224. Many of these judges, of course, would not be members of the excluded groups.
225. While we do not have space to provide a full theory of judicial correction, we can list five obstacles such a theory faces. First, there must be agreement on the nature of the imperfection in the original constitutional enactment process. Although some imperfections are obvious, such as the exclusion of African-Americans, others are more subtle, like the representation through states rather man individuals. Second, for there to be a successful correction, there must be agreement on the substance of the constitutional provision that would have emerged but for the imperfection. Again some consequences may be obvious. African-Americans would likely have insisted on antidiscrimination provisions, assuring their right to vote and participate in civil society. But other possible changes in the Constitution are harder to assess, either because it is unclear what an excluded group would have wanted or whether it would have had the leverage to entrench its desires. Third, introducing such a corrected provision must advance the interests of those harmed by the imperfection. Before the Civil War, a correction that ended slavery would almost certainly have split the union. It is not at all clear that a retreat to sectarian regional governments would have left African-Americans better off. Fourth, the corrective provision must be an important one. Introducing unimportant principles through correction creates more costs than benefits by destabilizing the Constitution. Fifth, the introduction of even an important principle must not undermine the legitimacy of the Constitution. A correction that obviously ignored significant textual portions of the Constitution, such as equal state representation in the Senate, would be such a blow to legitimacy that it is hard to imagine the circumstances in which it would be justified.
226. One other potential defect is that property restrictions on the franchise in the original constitution-making process excluded some poor individuals. But Akhil Amar has recently shown that states substantially widened their franchises to include most free males, see AMAR, supra note 72, at 48, and thus we think this is a less significant defect than those we discuss. We would respond to it as we have responded to the other failures by suggesting that judicial action to correct it would have more costs than benefits.
227. For a discussion of the legality of the Reconstruction amendments, see supra note 83.
228. African-American freedmen, however, were eligible to vote under the same rules as whites in five states at the time of the ratification: New Hampshire, New York, New Jersey, North Carolina, and New Jersey. See Dred Scott v. Sandford, 60 U.S. (19 How.) 398, 572-73 (1856) (Curtis, J., dissenting). We know that African-Americans generally voted for the Federalist party in the antebellum period and often had a high opinion of both George Washington and Alexander Hamilton. See LEON F. LITWACK, NORTH OF SLAVERY: THE NEGRO IN THE FREE STATES, 1790-1860, at 80-81 (1961). Therefore, it seems likely that most of the few African-Americans who did participate in voting for delegates to the Convention supported ratification.
229. Justice Marshall famously called attention to this central problem for American constitutional law. See Thurgood Marshall, Commentary, Reflections on the Bicentennial of the United States Constitution, 101 HARV. L. REV. 1, 2 (1987).
230. Nevertheless, even here it is not clear what alternative was better than adhering to the Constitution. Junking the Constitution over the slavery issue likely would have led to a retreat to sectional governments. The southern governments may well have treated African-Americans even worse. Perhaps importantly, the failure of the Constitution would have retarded the progress of a liberal social order based on markets that made slavery ideologically anomalous. Trying to change the Constitution on these essential matters through nonoriginalist interpretations in the early Republic would likely have had a similar result. The different regions would have promptly spiraled into disunion. See McGinnis & Rappaport, supra note 180.
231. U.S. CONST. amend, XIII.
232. U.S. CONST. amend. XIV.
233. U.S. CONST. amend. XV.
234. See AMAR, supra note 83, at 349-403.
235. See, e.g., Voting Rights Act of 1965, Pub. L. No.89-110, 79 Stat. 437 (codified as amended at 42 U.S.C. §§ 1971, 1973 to 1973aa-6 (2006)).
236. The strongest argument for concluding that the Reconstruction Amendments did not correct the supermajoritarian failure is that they failed to prohibit a variety of forms of discrimination against African-Americans, such as discrimination in matters that would not have been treated as civil rights in 1868. The original meaning of the Reconstruction Amendments is hotly contested, but even under the most optimistic interpretations, it is quite possible that the Amendments were insufficient. Significantly, many of the ways that the Amendments might have left African-Americans underprotected have now been addressed by judicial interpretation. These precedents should generally be followed, even if they are inconsistent with the original meaning, because they correct a supermajoritarian failure and do not involve any additional departures from the original meaning. See McGinnis & Rappaport, supra note 180, at 841-843 (discussing the concept of corrective precedent).
237. We recognize that some historians believe that the subordination of African-Americans in the post-Reconstruction South was largely created by social custom and private violence and that Jim Crow laws, thus, were largely unnecessary to achieve subordination. See MICHAEL J. KLARMAN, FROM JIM CROW TO CIVIL RIGHTS: THE SUPREME COURT AND THE STRUGGLE FOR RACIAL EQUALITY 59-60 (2004). But we agree with critics who see legal restrictions on African-Americans as greatly facilitating their subordination by preventing defections of Southerners for whom customs would not have been in their economic and personal interest.
238. See David E. Bernstein & Ilya Somin, Judicial Power and Civil Rights Reconsidered, 114 YALE L.J. 591, 603-04 (2004) (reviewing KLARMAN, supra). Moreover, without discrimination as to voting, African-Americans would have gained much more control over the government, which would have allowed them to act against private violence and social custom. Finally, the Equal Protection Clause obligates the states to act against such private violence. See U.S. CONST. amend XIV, § 1.
239. See KLARMAN, supra note 211, at 30.
240. Id.
241. Id. at 31.
242. See Giles v. Harris, 189 U.S. 475, 487 (1903) (refusing to compel registration when blacks were being discriminated against in a good character test because compelling registration would make the Court complicit in the scheme).
243. See AMAR, supra note 83, at 395.
244. 163 U.S. 537 (1896).
245. Id. at 550-551
246. See John Harrison, Reconstructing the Privileges or Immunities Clause, 101 YALE. L.J. 1385, 1458-1459 (1992).
247. Id. at 1459.
248. See, e.g., AMAR, supra note 72, at 162 (understanding the Fourteenth Amendment as an attempt to prevent restoration of racial caste).
249. See KLARMAN, supra note 211, at 93.
250. For a comparison of the views of supporters of the Fourteenth Amendment, like Bingham and Stevens, see JACOBUS TEN BROEK, EQUAL UNDER LAW (Collier Books 1965) (1951).
251. For Charles Crisp's views on civil rights, see Raphael O'Hara Boyd, Service in the Midst of a Storm: James Edward O'Hara and Reconstruction in North Carolina, 2001 J. NEGRO HIST. 319, 329 (2001).
252. Act of February 8, 1894, ch. 25, 28 Stat. 36, 36 (1894).
253. For a discussion that emphasizes the importance of Brown in civil rights jurisprudence as an imposition of cultural values rather than an enforcement of the original meaning of the Constitution, see Robert Post, Law and Cultural Conflict, 78 CHI.-KENT L. REV. 485, 500 (2003). Our point here is scholars like Post miss a larger truth. Whatever the correctness of Brown as a matter of the original meaning, the original meaning granted such a large measure of legal equality to African-Americans that if it had been enforced, Brown either would not have been necessary or would not have been a cultural watershed.
254. See Michael W. McConnell, Originalism and the Desegregation Decisions, 81 VA. L. REV. 947, 1131-1139 (1995). McConnell's is admittedly a controversial thesis.
255. Compare Michael J. Klarman, Brown, Originalism, and Constitutional Theory: A Response to Professor McConnell, 81 VA. L. REV. 1881 (1995),
256. With Michael W. McConnell, The Originalist Justification for Brown: A Reply to Professor Klarman, 81 VA. L. REV. 1937 (1995);
257. See also Harrison, supra note 219, at 1462-1463 (arguing that the original application of the Privileges or Immunities Clause would have resulted in invalidating the segregation laws at issue in Plessy and Brown).
258. The close connection between women and male relatives may mean that women's suffrage at the time of ratification would not have made a large difference. Indeed, we have some evidence to this effect. Women who owned the requisite amount of property were able to vote in New Jersey at this time, and politicians acted as if they would vote in the manner that others did in their area of the state. It was a Federalist movement to enfranchise women, but they limited the scope of this act to the seven most Federalist counties. See Judith Apter Klinghoffer & Lois Elkis, "The Petticoat Electors": Women's Suffrage in New Jersey, 1776 - 1807, 12 J. EARLY REPUBLIC 159, 172 (1992).
259. U.S. CONST. amend. XIX.
260. Moreover, we argue elsewhere that these judicial precedents should be protected because they are both entrenched and operate to correct a supermajoritarian failure. See McGinnis & Rappaport, supra note 180.
261. See supra Part IV.C.2.b.
262. See JOSEPH W. DELLAPENNA, DISPELLING THE MYTHS OF ABORTION HISTORY 958 (2006).
263. See BEEMAN, supra note 66, at 82, 386-411 (describing state ratification conventions).
264. U.S. CONST. art. V.
265. Id.
266. We model the influence of the Senate on judicial appointments in John O. McGinnis & Michael B. Rappaport, The Judicial Filibuster, the Median Senator, and the Countermajoritarian Difficulty, 2005 SUP. CT. REV. 257.
267. See RICHARD A. POSNER, BREAKING THE DEADLOCK: THE 2000 ELECTION, THE CONSTITUTION, AND THE COURTS, 228-31 (2001).
268. Given the loyalty of citizens at the time of the framing to their states rather than to a larger national entity that had not yet fully come into being, it seems fanciful that some element of representation by state could have been avoided. Even in light of that reality, it was still better to have a supermajoritarian than majoritarian process for proposing and ratifying the Constitution. Cf. McGinnis & Rappaport, supra note 19, at 748-749 (arguing that in spite of any problems resulting from overrepresentation of individuals in small states, majority and supermajority rules are still the best voting rules for the Senate).
269. See supra note 102 and accompanying text.
270. Small states may have disproportionate power to pass legislation in the ordinary legislative process. Even though legislation also has to be passed by the House, small states can get some items passed by trading their disproportionate power in the Senate to block legislation that large states might desire. As a result, small states tend to get more pork barrel spending per capita. See FRANCIS L. LEE & BRUCE I. OPPENHEIMER, SIZING UP THE SENATE: THE UNEQUAL CONSEQUENCES OF EQUAL REPRESENTATION 158-186 (1999). But, given that constitutional amendments for the last two centuries have concerned a single subject, such trades do not seem to have much affected the constitutional amendment process.
271. We have already explained why state applications for a convention are largely a dead letter in the convention process. See supra notes 92-93 and accompanying text.
272. See John A. Ragosta, Trade and Agriculture, and Lumber: Why Agriculture and Lumber Matter, 14 KAN. J.L. & PUB. POL'Y 413, 439 (2005).
273. See Keith Poole, 105th Congress, http://voteview.com/c105/c105.htm (last visited Jan. 21, 2010).
274. See supra note 91 and accompanying text. Even if the Constitution did not include such a provision, a constitutional amendment to eliminate equal representation would not likely pass given that more than a quarter of the states-the smaller ones-benefit from that equal representation.
275. See supra section III.B.