Law & McGowan: 2007: Colloquy: Northwestern University Law Review

Northwestern University Law Review : Colloquy : 2007 : Law & McGowan

There Is Nothing Pragmatic About Originalism

By David S. Law [*] and David McGowan [**]

John McGinnis and Michael Rappaport propose that originalist constitutional adjudication produces better consequences than competing approaches. They say they have “sketch[ed] the main elements of a pragmatic defense of originalism.”[1] We disagree. Pragmatism is about how well things work in practice, not how they should work in theory.[2] McGinnis and Rappaport’s argument turns pragmatism on its head: it rests not on any evidence of originalism’s actual superiority to other approaches, but rather on theoretical claims about the supermajoritarian character of originalism and the merits of supermajoritarian policymaking.

Part I of this essay explains why originalist judging does not honor the results of what McGinnis and Rappaport would consider “appropriate” supermajoritarian decisionmaking. Part II demonstrates that supermajority rules do not necessarily lead to the adoption of beneficial policies. We show that the actual effect of such rules depends upon a host of additional factors that cannot be analyzed in the abstract. In Part III, we review the ways in which McGinnis and Rappaport understate or ignore the costs of originalism, relative to those of other approaches. Finally, in Part IV, we present evidence that non-originalist judicial decisionmaking has, in fact, done a good job of enhancing social welfare, as measured by popular approval of the Court’s decisions.

I. The Constitution Was Not Adopted By A Supermajority

McGinnis and Rappaport claim that “appropriate” supermajority rules tend to result in the entrenchment of welfare-enhancing policies.[3] They also claim that the Constitution and its amendments have been passed (for the most part) under such rules.[4] It follows, they argue, that courts will maximize social welfare by applying only the original, supermajority-approved understanding of constitutional provisions.[5] The Constitution was never approved by a supermajority, however, and therefore does not satisfy the basic condition of their argument for originalism.

A. Only White Males Ratified the Constitution

With admirable candor, McGinnis and Rappaport concede that a “glaring defect” in the procedures used to adopt the Constitution was “their exclusion of African-Americans and women from the franchise.”[6] They concede, too, that the ratification of the Constitution by nothing more than a white male plurality goes to “the theoretical heart of the supermajoritarian argument.”[7]

To salvage their argument, McGinnis and Rappaport must explain how a supermajority at time t₂ could confer its imprimatur upon a constitution adopted earlier at time t₁. They do not. They simply observe that the Civil War amendments gave blacks the right to vote, and that women obtained the right to vote in 1920. But that would solve the problem only if blacks and women had been asked to ratify the Constitution retroactively.[8] They were not.

What McGinnis and Rappaport do say, instead, is that they see no “strong case that the Constitution would have been systematically different had these excluded groups been included.”[9] Absent “strong evidence” that such systematic differences would have existed, they argue, originalism is still welfare-enhancing because the provisions of the original Constitution “offer the best evidence of what good entrenchments would have resembled” under appropriate supermajority rules.[10]

There are several problems with this line of argument. First, McGinnis and Rappaport offer no support, either empirical or theoretical, for their conclusion that political participation by blacks and women would have neither defeated the Constitution nor changed it in any meaningful way. They offer only assertion, backed by an insistence that anyone wishing to prove them wrong must satisfy a heavy burden of proof.

Second, it is questionable whether a supermajority of the entire adult population would in fact have ratified the original Constitution. It seems doubtful, for example, that blacks would have endorsed the fugitive slave clause of Article IV,[11] much less the enshrinement of the slave trade found in Article I, Section 9.[12] To many people in 1787, however, a constitution without protections for the slave trade would no doubt have been “systematically different,” and in a highly unattractive way: the inclusion of such provisions may well have been the price that had to be paid to secure the support of white southerners in pivotal slave states.[13] The Constitution was a political agreement, and a political agreement is by definition a bundle of compromises. No one can say what kind of constitution would have been ratified in 1787—or, indeed, if a constitution would have been ratified at all—had the deal begun to unravel.[14]

Third, the argument is self-defeating. If McGinnis and Rappaport are right that the will of a supermajority can be correctly divined by thought experiment, then the actual use of supermajoritarian procedures becomes unnecessary. All constitutional decisionmaking and, indeed, all lawmaking could, in that case, be done faster and cheaper by thought experiment, while reaching the same result that actual supermajority voting would have produced. There would be no need to incur the time and expense of the constitutional amendment process; nor would there be any need for judges to adhere to the original meaning of the Constitution. Instead, judges could adopt up-to-date, welfare-enhancing policies in any given situation simply by asking themselves what a hypothetical supermajority would choose. What McGinnis and Rappaport are defending on pragmatic grounds is not originalism, but rather counterfactual speculation about the behavior of a supermajority that never existed as a historical matter.

B. A Supermajority of States Is Not a Supermajority of Voters [15]

McGinnis and Rappaport argue in favor of supermajority action by voters and legislators.[16] Yet the Constitution was not ratified by a supermajority of either group of people. It was ratified, instead, by a supermajority of states.

A supermajority of states is a far cry from a supermajority of actual people for several reasons. First, only a simple majority of state delegates or representatives was needed to commit a state in favor of ratification.[17] The delegates from Massachusetts, for example, ratified the Constitution by a vote of only 187 to 168,[18] while those from New York did so by an equally unimpressive margin of 30 to 27.[19] Second, only a simple majority of voters was required to elect those state delegates.

To see just how far a supermajority of states can diverge from a supermajority of actual people, imagine that the Constitution were to be submitted to the states for ratification today under the same formula employed by the Framers. Under Article VII, nine out of thirteen states were required to ratify the Constitution.[20] Today, that nine-thirteenths formula would require ratification by thirty-five out of fifty states (rounding up to the nearest whole number of states). The population of the United States is approximately three hundred million people,[21] but the fifteen most populous states are home to over two-thirds of that total.[22] This means that, under the Article VII formula used by the Framers, the Constitution could be ratified by a 70% supermajority of states containing less than one-third of the nation’s voters.[23] Moreover, because only a bare majority of state voters is needed for any given state to ratify, the nationwide proportion of voters whose support would be needed to ratify the Constitution under the Framers’ formula must be further reduced by nearly half, to less than 17% of the electorate.

As a historical matter, it is more than a mere mathematical possibility that the supermajority of ratifying states represented less than a supermajority of voters. John Marshall himself observed that “it is scarcely to be doubted that, in some of the adopting states, a majority of the people were in opposition. In all of them, the numerous amendments that were proposed, demonstrate the reluctance with which the new government was accepted.”[24] New York exemplifies this reluctance. According to Hamilton, two-thirds of its ratifying convention and four-sevenths of its populace were opposed in principle to the new Constitution, but the convention nevertheless voted to ratify out of a widespread fear that failure to do so might lead to geographic partition of the state.[25] It is thus highly questionable whether the Constitution was favored by a supermajority of the white male minority that was eligible to vote, much less by a supermajority of the entire adult population.

We are not the first to make the point that a supermajority of states is not truly a supermajority. Ethan Leib makes a similar point in his own criticisms of McGinnis and Rappaport’s thesis.[26] In reply, McGinnis and Rappaport argue that “the inclusion of a majority component” within a supermajority voting system does not transform the entire system into one of simple majority rule.[27] If a rule requiring ratification by nine of thirteen states is only a majority rule, they ask, what would a rule requiring seven out of thirteen states be?[28]

This response betrays too great a fascination with the formal characterization of voting rules, and too little concern for what pragmatists are supposed to care about—namely, the welfare of actual people. People, not states, bear costs and reap benefits. The number of states required for ratification is therefore a red herring, as is the rhetorical question of how a seven-state requirement would differ from a nine-state requirement. The first concern for any pragmatist ought to be the extent to which actual people, not geographically defined political units, were for or against the Constitution.

II. Supermajoritarian Lawmaking Does Not Necessarily Enhance Welfare

The basis of McGinnis and Rappaport’s argument is their claim that supermajority rules entrench desirable policies.[29] Neither logic nor experience supports this claim very well.

First, there is no logical basis to conclude that supermajority rules entrench good policies, for a very simple reason: preferences vary in intensity. Even if the beneficiaries of a policy outnumber the losers, the losers may suffer to a much greater extent than the beneficiaries gain. The sheer proportion of people who approve of a policy tells us how widely the costs and benefits of the policy are dispersed, but it does not show that the policy increases net social welfare. Nothing prevents a supermajority from adopting policies that reduce net welfare as long as the costs of such policies are sufficiently concentrated. Slavery provides an obvious example: a society in which three-quarters of the citizenry are free and one-quarter are slaves can muster supermajority support for slavery, simply because the benefits of slavery are widely dispersed, while the costs are narrowly concentrated. Yet it would be wrong to conclude on this basis that the gains to the slaveholders exceed the losses to the slaves.

Second, bargaining may produce supermajority support for pernicious measures. In lawmaking, there is always more than one issue on the agenda, and some people inevitably care more about some issues than about others. The result will be bargaining between those who favor a certain policy on issue A and those who favor a certain policy on issue B. Such bargaining lies at the very heart of politics, and it frequently results in the adoption of widely opposed policies (such as the everyday pork-barrel spending bills that supermajority rules do nothing to prevent).

Our slavery example illustrates both points. Imagine a nation consisting of three equally populous regions: Oldland, Newland, and Southland. All residents of Oldland oppose slavery but favor tariffs. All residents of Southland wish to maintain slavery but oppose protectionism. All residents of Newland oppose both slavery and protectionism. Thus, a two-thirds majority opposes both slavery and protectionism. Nevertheless, a single statute (or constitution) that entrenches both slavery and tariffs could muster a two-thirds majority, as long as Oldland is prepared to accept slavery in exchange for tariffs and Southland is prepared to do the opposite.

Our point is not that supermajorities invariably adopt bad policies. Our point, rather, is that there is no theoretical basis for arguing—as McGinnis and Rappaport do—that supermajority rules tend to result in the adoption of better policies. That claim can only be made on the basis of additional factual evidence, which they do not offer.

III. Tipping the Scales: The Relative Costs and Benefits of Originalism and Pragmatism

In this Part, we identify four flaws in McGinnis and Rappaport’s cost-benefit analysis, all of which serve to disguise the actual costs of originalism and to inflate its benefits relative to other approaches. First, supermajority approval of amendments is very costly, to the point that beneficial amendments may never be adopted. Second, originalism is subject to significant error costs. Third, a court may pick better policies than a supermajority. Fourth, the costs of originalism accumulate during the inevitable delay that attends our supermajoritarian process of constitutional amendment.

A. The High Cost of Supermajority Action

McGinnis and Rappaport are dismissive of both the costs entailed by flaws in the original Constitution, and the costs involved in any effort to correct them via formal amendment.[30] In their reply to Professor Leib, for example, they assert that any defects in the original Constitution “are not likely to be all that costly—or else they would be eliminated.”[31] That assertion is neither responsive nor correct. First, small costs are still costs, which any pragmatist or utilitarian would avoid if possible. Second, even the costliest of defects can endure indefinitely if the costs of correcting the defect are sufficiently high.

Securing the nationwide supermajoritarian action needed to adopt a constitutional amendment is notoriously difficult and costly. Attention must be drawn to the topic, people must be educated, and collective action problems must be overcome.[32] Even for seemingly uncontroversial issues, these can be formidable tasks, as the history of the most recent amendment to the U.S. Constitution aptly illustrates. It is difficult to find anyone who objects to the idea that members of Congress should have to wait an election cycle before reaping the benefit of a pay raise that they have awarded themselves.[33] Yet it took over two hundred years to ratify this simple rule in the form of the Twenty-Seventh Amendment.[34]

Other welfare-enhancing, supermajority-favored constitutional changes are likely to prove even more difficult to enact via the amendment process. Many of the silliest restrictions on personal freedom, such as prohibitions on contraception, criminal laws against sodomy, or even bans on interracial marriage, may exist in only a few states. Citizens in other states might understand that such laws are harmful, yet also rationally question why they should spend their scarce time and energy voting on constitutional amendments to permit conduct that their states do not ban in the first place. That such restrictions exist in relative isolation implies that they inflict costs on relatively few people and thus might never attract sufficiently widespread action to bring about a constitutional amendment to overturn them.

B. The Error Costs of Originalism

McGinnis and Rappaport argue that judges cannot be trusted to engage in pragmatism because “the judges’ own preferences would substantially interfere with the accuracy of their perceptions of social change.”[35] Yet the argument cuts both ways. “The judges’ own preferences” can also “interfere with the accuracy of their perceptions” of original meaning.[36]

Nor is bias the only obstacle to originalism done right. As historians have often complained, there is nothing inherently reliable about “law office history” performed by judges (or, more accurately, their law clerks) in the shadow of a particular case.[37] Meanwhile, ever-increasing scholarly disagreement over what “originalism” is, and how it ought to be done, raises the question whether it is even possible to identify the costs and benefits of the approach.[38]

Originalist-minded judges will make mistakes about history. And when they do, the supposed pragmatic benefits of originalism are lost, on McGinnis and Rappaport’s own account. Indeed, if the error rate is high enough—and much commentary by historians suggests it is very high—there would be no pragmatic reason at all to choose originalism. How often the Court will actually get it wrong, we cannot say. But neither can McGinnis and Rappaport, who do not even acknowledge the problem.

C. The Opportunity Costs of Originalism

McGinnis and Rappaport assert that supermajoritarian measures have distinctive beneficial properties such as helping people transcend ethnicity and geography.[39] The logic behind this assertion is not clear, but their premise appears to be that, in order to obtain supermajority assent, a measure must be acceptable to enough people that it embodies a truly national norm that is not very divisive.[40] They argue that non-originalist judicial decisions lack these salutary qualities both because judges might settle on some norm other than the one that a supermajority would endorse, and because judicial intervention preempts the operation of the amendment process.[41]

It is true that judges may fail to adopt a more welfare-enhancing norm that would otherwise have been embodied in a constitutional amendment. The reverse, however, is also true: pragmatic judges may adopt a more welfare-enhancing norm than the amendment process would select. Suppose the norm “separate educational facilities are inherently unequal” is better than “separate facilities can be equal if we just try harder.” Does anyone seriously think that a national supermajority in 1954 would have chosen the better of the two norms? Supermajorities and judges may choose differently, but McGinnis and Rappaport offer little reason to think that supermajorities will choose better than judges.[42]

In fact, there may be little or no room for supermajorities to choose better rules. What, exactly, were the alternatives to the rules adopted in Griswold v. Connecticut,[43] Eisenstadt v. Baird,[44] Loving v. Virginia,[45] or Lawrence v. Texas?[46] Such cases pose essentially binary policy choices. Married and single couples either may or may not use contraceptives; people of different races either may or may not marry; people of the same gender either may or may not have consensual sex. If there is only one choice that improves on the status quo, and judges are routinely in a position to make that choice more quickly and cheaply than the amendment process, what sort of pragmatism refuses to let them make that choice?

D. The Cumulative Costs of Originalism

McGinnis and Rappaport take the position that judges should refuse to alleviate suffering that they could alleviate. Only if judges do so, they argue, will society take notice of that suffering and choose whether to end it via the amendment process.

In the meantime, however, the costs of a bad rule will continue to accumulate over time. Where a rule generates costs that can be ameliorated or avoided, it is ordinarily better to fix the rule sooner rather than later, for the same basic reason that a dollar today is worth more than a dollar tomorrow: happiness deferred is happiness lost in the interim.

It is nevertheless worth the wait for an amendment, argue McGinnis and Rappaport, because the eventual benefits of supermajority decisionmaking will exceed the immediate payoff of judicial problem-solving. But there is not much reason to think that our patience and endurance will be rewarded in the end. It is uncertain that an ameliorating amendment will ever be adopted, thanks in part to the high cost of supermajority action.[47] It is even less certain that such an amendment would adopt a better policy than a pragmatic court would have chosen. In fact, as we have seen, there is plenty of reason to expect the opposite.[48] Still less certain is the adoption of a rule so much better that it compensates for all the suffering we have endured in the meantime.

At most, their reasoning suggests that, in some cases, it might be utility-maximizing for judges to ignore evidence, to refrain from utility analysis, and to follow a blanket rule of originalism. In other words, they have argued for a form of rule utilitarianism, even if they do not call it that. But in what cases, and using what rule? Even if it is a good idea for judges to obey blanket rules of constitutional adjudication, there are alternatives to originalism that might perform better—say, a rule of obedience to public opinion polls that reveal a supermajority preference, or of conformity to a supermajority view among courts worldwide.[49]

Alternatively, it might be best to ditch rules altogether and opt for act utilitarianism, which in this context we might call case utilitarianism, or simply judicial pragmatism—namely, the case-by-case maximization of utility by judges on the basis of cost-benefit analysis. After all, cost-benefit analysis makes sense in a lot of situations, including judicial decisionmaking.[50] From a pragmatic perspective, it is not clear why constitutional cases should be any different.

IV. A Pragmatic Defense of Pragmatism

McGinnis and Rappaport’s argument about the relative superiority of originalism boils down to their assertion that the Supreme Court is bad at deciding constitutional cases in a pragmatic, utility-maximizing way.[51] The basis for their claim that the Court is bad at pragmatic utility-maximization is that (1) there are not many justices—far fewer than the nose-count needed for a constitutional amendment; (2) the Court decides things by majority vote, not supermajority vote; and (3) the justices are “elite lawyers” who “work in Washington.”[52]

It is hard to see what these observations prove about the Court’s aptitude for pragmatism. There are thousands of elite lawyers who work in Washington and make countless policy decisions. They do so, moreover, in small groups, or even individually, on a far-from-supermajoritarian basis. Some are agency officials and legislative aides; others are members of Congress and Cabinet secretaries. Very few wear black robes. Do McGinnis and Rappaport mean to suggest that they are all bad at policymaking?

Whether the Court is bad at maximizing utility on a case-by-case basis—and whether it would do a better job by sticking with originalism—depends instead upon a number of empirical questions that McGinnis and Rappaport do not even acknowledge. They never refer, for example, to the actual costs and benefits of pragmatic judicial decisions, or even to some proxy for those costs and benefits, such as public reaction. They never stop to ask whether judges are sophisticated or naive policymakers, or whether the judges’ preferences actually lead them to make bad decisions. In fact, they offer no account of how non-originalist justices decide cases. Are the justices indifferent to costs and benefits? Are they better or worse than a popular supermajority at ascertaining the consequences of their actions? Do their preferences coincide with those of the median voter?[53] If not, do they attempt to keep the median voter happy anyway?

Admittedly, there is no way to perform an actual head-to-head utility comparison of originalism and non-originalism. It is impossible to hook up the American population to utility-meters and measure the difference in overall utility between a hypothetical world in which the Court adjusts constitutional doctrine continuously as it deems best, and a hypothetical world in which the Court adheres to originalism.

Judged against what we do know, however, McGinnis and Rappaport’s wholesale indictment of judicial policymaking is wholly unwarranted. We know something about how actual non-originalist decisions have been received in the real world, and this information is obviously relevant to any choice between originalism and non-originalism on pragmatic grounds. If non-originalism tends to produce outcomes that people like, that fact alone is good reason to choose non-originalism on pragmatic grounds, and we should choose originalism only if it would yield even better outcomes that people would like even more.

Let us consider, therefore, how people actually feel about non-originalism. We can safely say, for starters, that the general public does not fetishize originalism for its own sake. Quite frankly, the general public would probably be hard-pressed to distinguish originalism from a hole in the ground. This is the same general public, after all, that thinks Clarence Thomas is the most liberal member of the Court.[54] To be sure, members of the public can be highly attentive to the Court’s work when their immediate interests are directly at stake [55]: it should come as no surprise, for example, that Orthodox Jews in the New York area were aware of the Supreme Court’s decision in Kiryas Joel.[56] To the extent that people know enough about a Supreme Court decision to be unhappy about it, however, it is safe to assume that their objection will be to the substantive policy result, not to the Court’s choice of a non-originalist approach per se.

We also know that many of the Court’s most obviously creative decisions have been warmly embraced over time, to the point that they are now considered beyond question. The doctrines announced in Brown v. Board of Education,[57] Griswold v. Connecticut,[58] Eisenstadt v. Baird,[59] and Loving v. Virginia[60] are as much a part of the American ethos as the document in which they cannot be found.[61] The Constitution that forbids racial apartheid in public schools and protects the right of unmarried couples to use contraception is the Constitution that Americans know and love, regardless of what anyone may have written or intended in 1789 or even 1868.

In fact, if public opinion is any indication of whether the Court is reaching desirable outcomes, there is much reason to think that a non-originalist Court gets things right—more so, indeed, than our elected officials do. We know that the public is generally satisfied with the Court’s performance, non-originalist warts and all. Study after study has found that the American people have greater, and more consistent, faith in the Court than in other institutions of government.[62] Not even Roe v. Wade[63] appears to have disturbed the public’s overall approval of the Court. In fact, Roe marked the middle of a period of sharply increasing public confidence in the Court.[64]

Finally, we know that the Court’s decisions are, whether by coincidence or design, remarkably consistent with public opinion.[65] Empirical studies have found repeatedly that the Court’s constitutional decisions have been largely in sync with public opinion.[66] Indeed, it appears that the Court’s actions are more often in sync with the general public than those of the elected branches.[67] The constitutional jurisprudence of the Warren and Burger Courts—surely a low point for originalism—was no exception to this general trend.[68]

If public satisfaction with the Court is any measure of whether the Court is any good at making constitutional law in a pragmatic, non-originalist, welfare-enhancing way, the answer seems fairly clear: yes, it is. McGinnis and Rappaport might demur that people could be even better off if the Court stopped trying to anticipate their wishes and instead let them work up the energy to pursue the amendment route. This is true. One cannot disprove a proposition about the path not taken. Our response is simple: yes, people might be better off under originalism. But the evidence suggests this is doubtful, because judicial pragmatism is hard to beat. The pragmatic thing to do, it turns out, is to stick with pragmatism.

Conclusion

Professors McGinnis and Rappaport are brilliant and thoughtful scholars. It is to their credit that they have even attempted the task of reconciling originalism and pragmatism, which have long been mortal enemies. They have tackled a difficult—we suspect impossible—challenge with creativity and originality. Even in the world of constitutional theory, however, speculation can only go so far.

There is nothing pragmatic about either originalism or their defense of it. Pragmatism is about paying attention to consequences, which in turn demands observation and measurement. Yet observation and measurement are precisely the things that McGinnis and Rappaport want our judges to avoid, and which, tellingly, they themselves avoid when it comes to how our judges have actually performed. When it comes to pragmatism, theory and conjecture are no substitute for empiricism.

————

*. Associate Professor of Law, University of San Diego; Assistant Adjunct Professor of Political Science, University of California, San Diego.

**. Professor of Law, University of San Diego. We are grateful to Dan Farber, Susan Franck, Ethan Leib, Robert Post, Mike Ramsey, Larry Solum, and David Zaring for their extremely valuable comments and suggestions.

1. John O. McGinnis & Michael B. Rappaport, A Pragmatic Defense of Originalism, 101 Nw. U. L. Rev. Colloquy 68, 69 (2007), http://www.law.northwestern.edu/lawreview/colloquy/2007/1/ (link).

2. See, e.g., Charles Saunders Peirce, How To Make Our Ideas Clear, in 1 The Essential Peirce: Selected Philosophical Writings 124, 132 (Nathan Houser & Christian Kloesel eds., 1992) (advocating an epistemological approach focused upon “effects, which might conceivably have practical bearings”).

3. Id. at 70.

4. See id. at 70, 73.

5. Id. at 69–70.

6. Id. at 73 n.22.

7. Id. at 79.

8. Professor Leib makes the same point in his own response to McGinnis and Rappaport. See Ethan J. Leib, Why Supermajoritarianism Does Not Illuminate the Interpretive Debate Between Originalists and Non-Originalists, 101 Nw. U. L. Rev. Colloquy 113, 117 n.19 (2007), http://www.law.northwestern.edu/lawreview/colloquy/2007/7/ (link).

9. McGinnis & Rappaport, supra note 1, at 81 n.55.

10. Id.

11. U.S. Const. art. IV, § 2, cl. 3 (link) (“No Person held to Service or Labour in one State, under the Laws thereof, escaping into another, shall, in Consequence of any Law or Regulation therein, be discharged from such Service or Labour, but shall be delivered up on Claim of the Party to whom such Service or Labour may be due.”).

12. Id. art. I, § 9 (“The Migration and Importation of such Persons as any of the States now existing shall think proper to admit, shall not be prohibited by the Congress prior to the Year one thousand eight hundred and eight, but a Tax or duty may be imposed on such Importation, not exceeding ten dollars for each Person.”). Perhaps such provisions are what McGinnis and Rappaport have in mind when they say that the Constitution would not have been “systematically different” had blacks and women voted: excise a few offending provisions, they might argue, and the essential elements of the Constitution (whatever those might be) would still have passed by a supermajority.

13. See, e.g., Christopher Collier & James Lincoln Collier, Decision in Philadelphia: The Constitutional Convention of 1787 137–52 (1986) (discussing how the issue of slavery was “enmeshed with a host of other problems” and “ran into everything,” and noting the awareness of northerners that “the Convention could not interfere with slavery very much, if at all, if they hoped to keep the Deep South states in the union”).

14. There is also a deeper conceptual reason why it makes no sense to say that blacks and women would have voted to ratify the Constitution in 1787. A world in which blacks and women could vote on such matters would have been a very different place. The rules that denied the franchise to blacks and women did not exist in isolation. They both reflected and perpetuated an entire system of social, political, and economic inequality. An eighteenth-century America in which blacks and women could vote on a proposed constitution would necessarily have been a radically different and more egalitarian place in other ways as well. This different, and better, version of America would presumably have experienced different needs and expressed different aspirations. These are the kinds of profound differences that can influence the substance of a nation’s constitution.

15. We are indebted to Dan Farber for prompting us to address this point.

16. See McGinnis & Rappaport, supra note 1, at 70–73.

17. See Akhil Reed Amar, The Consent of the Governed: Constitutional Amendment Outside Article V, 94 Colum. L. Rev. 457, 487 n.112 (1994); Leib, supra note 8, at 120.

18. See Michael Allen Gillespie, Massachusetts: Creating Consensus, in Ratifying the Constitution 138, 158 (Michael Allen Gillespie & Michael Lienesch eds., 1989).

19. See Cecil L. Eubanks, New York: Federalism and the Political Economy of Union, in Ratifying the Constitution, 300, 328 (Michael Allen Gillespie & Michael Lienesch eds., 1989).

20. U.S. Const. art. VII; see McGinnis & Rappaport, supra note 1, at 73.

21. U.S. Census Bureau, Population Estimates Data Sets, http://www.census.gov/popest/datasets.html (last visited Sept. 30, 2007) (link).

22. In descending order of population, those states are California, Texas, New York, Florida, Illinois, Pennsylvania, Ohio, Michigan, Georgia, North Carolina, New Jersey, Virginia, Massachusetts, Washington, and Indiana. See id.

23. For the sake of simplicity, we assume that the proportion of people in each state who can vote is the same.

24. II John Marshall, The Life of George Washington 127 (1834).

25. See Letter from Alexander Hamilton to James Madison (June 8, 1788), in 5 The Papers of Alexander Hamilton 2, 3 (Harold C. Syrett & Jacob E. Cooke eds., 1962) (reporting the fear among New Yorkers that “separation of the Southern district from the other part of the state . . . would become the object of the Fœderalists and of the two neighbouring states”); see also Marshall, supra note 24, at 28 n.1 (observing that New York was “apparently dragged into [ratification] by a repugnance to being excluded from the confederacy”). Forrest McDonald put the vote for electors to the New York convention at about 16,000 to 7,000 against ratification, producing a convention of 46 opponents to ratification and 19 advocates. Forrest McDonald, Alexander Hamilton 114 (1979). Scholars have concluded—perhaps on the basis of Hamilton’s letter—that New York ratified because Virginia’s ratification a month earlier made it inevitable that the Constitution would be adopted, and because Jay and Hamilton threatened to take New York City out of the state and into the Union if the convention voted against the Constitution. See id. at 115; Eubanks, supra note 19, at 325–30.

26. See Leib, supra note 8, at 120.

27. John O. McGinnis & Michael B. Rappaport, Originalism and Supermajoritarianism: Defending the Nexus, 102 Nw. U. L. Rev. Colloquy 18, 24 (2007), http://www.law.northwestern.edu/lawreview/colloquy/2007/18/ (link)

28. See id. The gist of McGinnis and Rappaport’s position seems to be that more is better: the greater the consensus of any kind needed to adopt a policy, the better the resulting policy. For example, the requirement of a two-thirds vote of both houses of Congress is, in their view, a desirable supermajority rule because it “requires a greater consensus than a mere majority, even though a mere majority of the people in each district can put those legislators in office.” Id. at 24. By the same logic, a rule requiring a supermajority of states rather than people would still be a welfare-enhancing supermajority rule simply because it requires more consensus than a rule requiring a majority of states.

“More is better” is an appealing slogan, but one that ultimately leads to ruinous results. If McGinnis and Rappaport truly mean to argue that greater consensus is always better than lesser consensus, then logic dictates that they should favor total consensus in the form of unanimity requirements, which would of course lead to complete gridlock. They do not go that far; neither, however, do they qualify their “more is better” view or recognize the existence of a potentially thorny tradeoff between degree of consensus and gridlock.

29. See McGinnis & Rappaport, supra note 1, at 70.

30. It may be more accurate to say that they are inconsistent. At one point they say “entrenched norms cannot be easily eliminated.” McGinnis & Rappaport, supra note 1, at 71.

31. McGinnis & Rappaport, supra note 27, at 24.

32. See, e.g., Mancur Olson, The Logic of Collective Action: Public Goods and the Theory of Groups 1–65 (rev. ed. 1971) (explaining the inherent difficulties of collective action); Jack L. Walker, Mobilizing Interest Groups in America: Patrons, Professions, and Social Movements 41–55 (1991) (discussing the role of institutions and policy entrepreneurs in overcoming collective action problems and mobilizing large groups of people on questions of policy).

33. See U.S. Const. amend. XXVII (“No law, varying the compensation for the services of the Senators and Representatives, shall take effect, until an election of Representatives shall have intervened.”).

34. See Richard B. Bernstein, The Sleeper Wakes: The History and Legacy of the Twenty-Seventh Amendment, 61 Fordham L. Rev. 497, 498 (1992). Nor is the Twenty-Seventh Amendment the only example of how difficult it can be to secure constitutional change that enjoys supermajority support. Consider also Franklin Roosevelt’s New Deal: in the 1936 election, Roosevelt won over 60% of the vote, carried every state except Maine and Vermont, and secured a whopping three-quarters Democratic majority in Congress. See David E. Kyvig, The Road Not Taken: FDR, the Supreme Court, and Constitutional Amendment, 104 Pol. Sci. Q. 463, 465 (1989). Yet even Roosevelt, a master politician with a massive electoral and legislative majority at his disposal, despaired at the difficulty of having to secure a constitutional amendment in order to overcome the Court’s resistance to the New Deal—so much so that he chose instead to embark upon his now-infamous campaign to pack the Court. See id. at 464–68 (noting FDR’s perception of the constitutional amendment process as “impossibly difficult,” notwithstanding his firsthand experience with earlier, successful amendment efforts). The eventual and effective remedy for constitutional doctrine that had fallen badly out of sync with pressing needs and circumstances was a change of course on the part of the Court itself—the so-called “switch in time that saved nine.” See, e.g., Michael Comiskey, Can A President Pack—or Draft—the Supreme Court? FDR and the Court in the Great Depression and World War II, 57 Alb. L. Rev. 1043, 1046–47 (1994).

35. McGinnis & Rappaport, supra note 27, at 23.

36. Id.; see, e.g., Leonard W. Levy, Original Intent and the Framers’ Constitution 299–300 (1988) (deeming it a “notorious fact” that the Justices have routinely “abus[ed] historical evidence in a way that reflects adversely on their intellectual rectitude as well as on their historical competence”); Alfred H. Kelly, Clio and the Court: An Illicit Love Affair, 1965 Sup. Ct. Rev. 119, 155 (surveying the Court’s long tradition of twisting history to its own purposes, and assessing the results as “very poor indeed,” from a “professional point of view”).

37. E.g., Levy, supra note 36, at 299–300, 313–21, 388–89 (bemoaning the quality of “law office history,” and observing that “examples of the historical illiteracy of the Supreme Court can be multiplied ad nauseum”); Jack N. Rakove, Original Meanings: Politics and Ideas in the Making of the Constitution 11 (1996) (remarking upon the poor quality of “law office history”); Kelly, supra note 36, at 132 (same).

38. To pick one exchange involving McGinnis and Rappaport, see Jack M. Balkin, Abortion and Original Meaning, 24 Const. Comment. (forthcoming 2007); John O. McGinnis & Michael B. Rappaport, Original Interpretive Principles as the Core of Originalism, 24 Const. Comment. (forthcoming 2007); and Jack M. Balkin, Original Meaning and Constitutional Redemption, 24 Const. Comment. (forthcoming 2007).

39. McGinnis & Rappaport, supra note 1, at 71–72. We find this assertion surprisingly ahistorical. For one thing, it begs the question of when, exactly, America has ever transcended its ethnic or geographic differences, with or without the help of supermajority rules. For another, history demonstrates that supermajority requirements can easily perpetuate ethnic and geographic divisions. To take one example, the Senate’s supermajoritarian filibuster rules thwarted welfare-enhancing measures such as anti-lynching laws and thereby helped prolong Southern apartheid for decades. See, e.g., Robert A. Caro, Master of the Senate: The Years of Lyndon Johnson 93, 218 (2002).

40. McGinnis & Rappaport, supra note 1, at 71–72.

41. See id. at 78.

42. What little reason they do give is discussed below in Part IV.

43. 381 U.S. 479 (1965) (link) (holding that a state law prohibiting the dispensing of contraceptives to married people violated a constitutionally protected “zone of privacy”).

44. 405 U.S. 438 (1972) (link) (extending the rule of Griswold v. Connecticut to unmarried couples on equal protection grounds).

45. 388 U.S. 1 (1967) (link) (striking down a state law against interracial marriage on equal protection grounds).

46. 539 U.S. 558 (2003) (link) (invalidating a state law against same-sex sodomy on substantive due process grounds).

47. See supra Part III.A.

48. See supra Parts II & III.C.

49. Other possibilities include rules that do not displace, but instead reduce the burden of, full-blown judicial cost-benefit analysis. For example: only issue decisions you expect to be enforced; never order taxes raised; keep your hands off foreign relations. The choice is not between originalism and impossible calculation. The choice is between originalism and pragmatic analysis informed by a long institutional history and, typically, the experience evinced by the record of a concrete case. It might indeed be hard to specify, for example, the utility-maximizing rule of sexual behavior in general, but it is not nearly as hard to decide the much narrower question whether a ban on prosecution of consensual same-sex acts increases net social welfare.

50. Suppose, for example, that an originalist cannot decide on the historical record whether the First Amendment condemns prosecutions for seditious libel. Presumably McGinnis and Rappaport would want a judge to do his best to work through the consequences of a decision allowing or forbidding such prosecutions. Judges do not magically lose this ability when the historical record is clearer.

51. See, e.g., McGinnis & Rappaport, supra note 27, at 23 (arguing that the judiciary is unlikely to “capture true social change” because “the judges’ own preferences would substantially interfere with the accuracy of their perceptions of social change”).

52. McGinnis & Rappaport, supra note 1, at 75.

53. See, e.g., Neal Devins, The D’Oh! of Popular Constitutionalism, 105 Mich. L. Rev. 1333, 1347–50 (2007) (link) (arguing that the appointments and confirmation process has produced “a Court whose preferences generally track the median voter”); infra notes 65–68 and accompanying text (arguing that, contrary to the premises of most contemporary constitutional theory, the Court is largely a majoritarian institution).

54. When asked to name the most liberal member of the Court, respondents in a 2006 survey picked Justice Thomas twice as often as Justice Stevens and three times as often as Justice Breyer. See Kevin T. McGuire, The Judicial Branch: Judging America’s Judges, in A Republic Divided 194, 198 fig.1 (Annenberg Democracy Project ed., 2007); see also, e.g., Devins, supra note 53, at 1340–41 (mustering a spate of statistics to illustrate that Americans know “[n]ext-to-nothing” about the Constitution).

55. See Gregory A. Caldeira & Kevin T. McGuire, What Americans Know About the Courts and Why It Matters, in The Judicial Branch 262, 266 (Kermit L. Hall & Kevin T. McGuire eds., 2005).

56. Board of Educ. of Kiryas Joel Sch. Dist. v. Grumet, 512 U.S. 687 (1994); see Valerie J. Hoekstra, The Supreme Court and Local Public Opinion, 94 Am. Pol. Sci. Rev. 89, 91–93 (2000) (discussing awareness of the case within New York’s Hasidic Jewish community).

57. 347 U.S. 483 (1954). We are of course aware of originalist efforts to bring Brown into the fold. See, e.g., Michael W. McConnell, Originalism and the Segregation Decisions, 81 Va. L. Rev. 947, 952–53 (1995). Nevertheless, the view of Brown as contrary to nineteenth-century understanding is the standard, and best-supported, originalist view. See William N. Eskridge, Jr., Some Effects of Identity-Based Social Movements On Constitutional Law in the Twentieth Century, 100 Mich. L. Rev. 2062, 2365 (2002) (“Virtually no one has been persuaded by McConnell’s learned account.”). It was, in any event, the view held by the justices who decided Brown that they were departing from the original understanding of the Fourteenth Amendment. See Michael J. Klarman, From Jim Crow to Civil Rights 302–08 (2004).

58. 381 U.S. 479 (1965) (holding that a state law prohibiting the dispensing of contraceptives to married people violated a constitutionally protected “zone of privacy”).

59. 405 U.S. 438 (1972) (extending the rule of Griswold v. Connecticut to unmarried couples on equal protection grounds).

60. 388 U.S. 1 (1967) (striking down a state law against interracial marriage on equal protection grounds).

61. The fact that “the Constitution” transcends the words set to paper centuries ago has long been evident to acute observers and has only grown more obvious over time. See, e.g., Karl N. Llewellyn, The Constitution as an Institution, 34 Colum. L. Rev. 1 passim (1934) (distinguishing “the Document” from “the Constitution”).

62. See, e.g., Caldeira & McGuire, supra note 55, at 264 (“[O]ver and over again, polls show that Americans have more confidence in the Court than either the president or the Congress.”); McGuire, supra note 54, at 206 (reporting the results of a 2006 Gallup poll revealing public esteem for the Court to be “at its highest level in several years, well above that of the Congress and the president”); Jeffery J. Mondak & Shannon Ishiyama Smithey, The Dynamics of Public Support for the Supreme Court, 59 J. Pol. 1114, 1116 (1997) (observing that “aggregate support for the Supreme Court consistently exceeds levels for Congress and the executive branch”).

63. 410 U.S. 113 (1973) (link).

64. See Gregory A. Caldeira, Neither the Purse Nor the Sword: Dynamics of Public Confidence in the Supreme Court, 80 Am. Pol. Sci. Rev. 1209, 1213 & fig.1 (1986).

65. See, e.g., Barry Friedman, Mediated Popular Constitutionalism, 101 Mich. L. Rev. 2596, 2606 (2003) (“[T]he wealth of existing evidence suggests that most of the time judicial decisions fall within the range of acceptability that one might expect of the agents of popular government.”); id. at 2607 (“[I]n the main the results of Supreme Court decisionmaking comport with the preferences of a majority or at least a strong plurality, something that many political scientists now take as a given.”); Jeffrey J. Mondak & Shannon Ishiyama Smithey, The Dynamics of Public Support for the Supreme Court, 59 J. Politics 1114, 1120 (1997); William Mishler & Reginald S. Sheehan, The Supreme Court as a Countermajoritarian Institution? The Impact of Public Opinion on Supreme Court Decisions, 87 Am. Pol. Sci. Rev. 87, 91–95 (1996) (reporting a significant correlation between public opinion and the outcome of Supreme Court decisionmaking, subject to a five-year lag); William Mishler & Reginald S. Sheehan, Public Opinion, the Attitudinal Model, and Supreme Court Decision Making: A Micro-Analytic Perspective, 58 J. Pol. 169, 186–96 (1996) (finding a significant correlation between public opinion and the voting behavior of approximately half of all justices over the preceding forty years).

66. See, e.g., Thomas R. Marshall, Public Opinion, Representation, and the Modern Supreme Court, 16 Am. Pol. Q. 296, 297–98 (1988) (same); Thomas R. Marshall & Joseph Ignagni, Supreme Court and Public Support for Rights Claims, 78 Judicature 146, 148–49 (1994) (reporting that the Court is much more likely to rule in favor of constitutional claims that enjoy public support, than constitutional claims that do not); Alan D. Monroe, Public Opinion and Public Policy, 1980–1993, 62 Pub. Opinion Q. 6, 12–13 & 13 tbl.1 (1998) (reporting that overall federal policy is consistent with majority preference only about 55% of the time); Terri Peretti, An Empirical Analysis of Alexander Bickel’s The Least Dangerous Branch, in The Judiciary and American Democracy: Alexander Bickel, The Countermajoritarian Difficulty, and Contemporary Constitutional Theory 123, 140 (Kenneth D. Ward & Cecilia R. Castillo eds., 2005) (observing that “the Court’s decisions match majority opinion more often than not and about as often as the other branches”).

67. Compare Thomas R. Marshall, Public Opinion and the Supreme Court 78–81 (1989) (reporting that the Court’s decisions on constitutional issues are in sync with public opinion approximately 60% of the time) with Alan D. Monroe, Public Opinion and Public Policy, 1980–1993, 62 Pub. Opinion Q. 6, 12–13 & 13 tbl.1 (1998) (reporting that overall federal policy is consistent with majority preference only about 55% of the time).

68. See Marshall & Ignagni, supra note 66, at 148–49 (finding that, from 1953 through 1992—a period that spans the entirety of both the Warren and Burger Courts—the Court ruled in favor of civil rights claims 73% of the time when supported by public opinion, but only 40% of the time when not supported by public opinion).

To be sure, the public does not endorse everything that the Court does. There is significant evidence, for example, that invalidation of federal statutes erodes public support for the Court, at least in the short term. See Caldeira, supra note 64, at 1219 tbl.1, 1222–23. Yet this finding does not imply that the Court would make people happier by adhering to originalism: as it has repeatedly demonstrated, the Court is perfectly capable of striking down federal statutes in the name of originalism. See, e.g., United States v. Lopez, 514 U.S. 549 (1995) (striking down the federal Gun-Free School Zones Act on the grounds that it exceeds an originalist understanding of Congress’s powers under the Commerce Clause); United States v. Morrison, 529 U.S. 598 (2000) (striking down the federal Violence Against Women Act of 1994 for similar reasons).