Countering the Majoritarian Difficulty

32 Const. Comment. 61

Our Republican Constitution: Securing the Liberty and Sovereignty of We the People. By Randy E. Barnett.[1] New York: HarperCollins Publishers. 2016. Pp. xiv + 283. $26.99 (cloth).

Amy Coney Barrett[2]

In Our Republican Constitution,[3] Randy Barnett argues that the United States Constitution rests on a foundation of individual rather than collective popular sovereignty. Grounding the legitimacy of the government in the authority given it by each individual rather than by the People as a whole echoes the thesis, advanced in Barnett’s prior work, that the government must justify incursions upon individual liberty.[4] If the People as a body are sovereign and the Constitution is designed to facilitate democratic self-governance, legislation is presumptively legitimate because it represents the sovereign will of the democratic majority. If the individual is sovereign, by contrast, legislation does not represent the sovereign will but rather the work product of government officials who serve as the agents of individual sovereigns. The citizen is thus positioned to demand that his agents explain why legislation lies within the authority he has constructively given them to secure his natural rights.

Courts play an important role under Barnett’s Republican Constitution. They provide the forum in which citizens seek protection of their natural rights from legislative infringement. Like legislators, judges serve as agents of each individual sovereign, and judicial deference to democratic majorities is “misguided and inconsistent with the most basic premises of the Constitution” (p. 18). Rather than treating legislation as presumptively constitutional, they must treat the citizen’s challenge as presumptively correct. And on the merits, they must critically rather than deferentially assess the question whether the legislature has exceeded its authority, which is limited to regulation securing the “equal protection of the rights of each and every person” (p. 25). Barnett thus calls for, among other things, a return to the pre-New Deal approach to the Due Process Clause.

Constitutional scholars have long viewed judicial review through the lens of the countermajoritarian difficulty. Under the Republican Constitution, however, it is legislatures rather than courts that we should worry about.       In this essay, I begin by developing the connection between Barnett’s theory of the Constitution and his approach to judicial review. I then express doubt about the historical support for Barnett’s approach, contend that the task he would give courts fails to account for the realities of the legislative process, and argue that he overestimates the institutional capacity of courts. I conclude by praising Barnett’s attention to the often-misunderstood concept of judicial restraint. That is a point on which many can agree with Barnett, regardless whether they accept his republican take on our Constitution.

  1. The Majoritarian Difficulty

Generations of constitutional scholars have grappled with the so-called countermajoritarian difficulty.[5] The power of judicial review enables courts to interfere with the majority’s preferences. Because the baseline in our republic is set in favor of democracy, the argument runs, courts should generally defer to what the majority wants. Courts apply heightened scrutiny to statutes implicating fundamental rights or suspect classes, but outside of that context, they are reluctant to interfere with the outcome of the democratic process. They give federal and state legislatures wide berth in enacting social and economic legislation and apply only minimal scrutiny when evaluating federal statutes for consistency with the limits on federal power.

In attacking this state of affairs, Barnett starts with its premise: that we should be concerned about the countermajoritarian nature of judicial review. Instead, Barnett claims, we should be concerned about the majoritarian nature of legislation. Democratic majorities pose a consistent threat to minority rights.

Barnett points out that many of the Founders had reservations about democracy. Madison’s essay on “The Vices of the Political System of the United States,” which matured into Federalist No. 10, details the concerns. Every society contains factions that will pursue their own self-interest.[6] When a faction includes a majority of citizens, what is to stop it from unjustly infringing upon the rights of those in the minority? Majorities will give into the temptation to self-deal by, among other things, enriching themselves at the expense of the minority.[7]

A republican form of government was the Founders’ solution to the excesses of democracy. On a view of our Republic that Barnett dubs “the Democratic Constitution,” the Founders countered the risk of democratic excess by opting for indirect rather than direct democracy (pp. 18-19). Direct democracy carries a greater risk of runaway majorities and is in any event impracticable in a country the size of the United States, even as it existed at the time of the Founding. Thus the Democratic Constitution filters its commitment to majority rule through the senators and representatives whom the majority votes into office. Structural features like federalism, bicameralism, equal state representation in the Senate, and differing terms lengths in the House and Senate were among the mechanisms the Founders employed to mute the influence of faction. But the Democratic Constitution does not eschew the importance of majority rule; it aims simply to temper the risk that the majority will get carried away. The majority vote of those senators and representatives represents, albeit indirectly, the majority will of the People. Hence government regulation is legitimate as the product of majority rule.

Barnett rejects this view of the Constitution in favor of what he calls “the Republican Constitution.” On Barnett’s account, the Founders did not design our Republic to enable elected representatives to “re-present” the will of the majority.[8] For one thing, such an approach would be inadequate to counter the risk of factions and democratic excess. For another, the Founders’ mistrust of democracy indicates that preserving majority rule was not in fact their primary concern. Drawing on, among other things, the Declaration of Independence and the Virginia declaration that inspired it, Barnett claims that the Founders’ purpose in forming the United States was the preservation of the pre-existing natural rights of the People—each and every one.[9] These natural, inalienable rights include the rights to life, liberty, and property.[10]

In the design of the Republican, as opposed to the Democratic, Constitution, elected representatives serve to secure the natural rights of the individual sovereigns who comprise “We the People,” not to carry out the mandate of the majority that voted them into office. The legitimacy of government rests on the consent of the governed, and the Republican Constitution conceives of that consent as flowing from individuals rather than the people as a group. Given that the consent of these individuals is only constructive, it ought to extend no farther than that to which a rational person would consent. A rational person would give up his liberty interests only if doing so advanced the larger goal of securing his life, liberty, and happiness.[11] That line defines the scope of authority conferred by the People (the principals) to government officials (their agents).

Thus the republican vision of the Constitution counsels courts and constitutional scholars to worry less about preserving the product of the democratic process than about the way the democratic process is apt to trample the rights of individuals. Because the point of government is to secure the pre-existing natural rights of the People, legislation is not presumptively legitimate simply because it has majority sanction. On the contrary, regardless of that majority sanction, it is presumptively illegitimate to the extent that it infringes upon the natural rights of individual sovereigns. Courts should not give statutes a presumption of constitutionality when they review them; instead, the state should bear the burden of justifying legislation as lying within its limited authority to secure the life, liberty, and property of the People. Nor should courts be unduly deferential when reviewing the state’s proffered justification. They should return in Due Process and Equal Protection challenges to the more demanding form of “rational basis” review practiced by courts in the Lochner era. And because structural constraints are often more effective than substantive limits in preserving individual liberty, courts should put teeth in the doctrines that enforce limits on federal legislative power.[12] As Barnett explains, “when the liberty of a fellow citizen and joint sovereign is restricted, judges as agents of these citizens have a judicial duty to critically assess whether the legislature has improperly exceeded its just powers to infringe upon the sovereignty of We the People” (p. 25).

  1. The Historical Case for Judicial Enforcement of the Republican Constitution

Given Barnett’s stature as an originalist, one might come to Our Republican Constitution expecting an originalist argument, and the book’s first chapter, which is devoted to founding-era history, gives it that flavor. Yet Barnett does not contend that the Constitution’s text demands acceptance of either the republican vision or the more searching form of judicial review for which he advocates. The book is less about what the Constitution’s original public meaning requires than about what is normatively attractive. Barnett claims it is desirable to understand the Constitution as a document designed to secure the natural rights of individual sovereigns, and that one accepting that view should find it similarly desirable for courts to play an active role in ensuring that the government not exceed the bounds of its authority. History, particularly founding-era history, is an important data point in his normative case: one reason we should find the republican vision attractive is that the founding generation did.

It is worth observing, however, that the history Barnett recounts is not entirely one-sided. He assembles evidence from the Declaration of Independence, early state Constitutions, and the Federalist Papers to support his argument that those who drafted and ratified the Constitution were committed to an individual rather than collective view of popular sovereignty. Vetting that claim would require independent study of the historical record, but even taken on its own terms, the evidence does not reflect unwavering insistence upon what Barnett describes as the republican conception. Instead, his account suggests that conflict between the republican and democratic views surfaced almost immediately. For instance, he indicates that the division is evident in the conflict between the Hamiltonian Federalists who favored broad national power and the Jeffersonian Republicans who stood for more limited federal government.[13] Barnett points to the opinions of Justices Jay and Wilson in Chisholm v. Georgia as support for an individual conception of popular sovereignty, but, as he acknowledges, Justice Iredell’s opinion clearly adopts the collective view.[14] The same divide exists between the opinions of Justices Chase and Iredell in Calder v. Bull.[15] If conflict existed that early, it is difficult to characterize Barnett’s republican view as one uniformly held in the founding era.

Competition, moreover, apparently persisted between these two views throughout American history. According to Barnett, the issue of individual versus collective popular sovereignty divided the pro-slavery Democratic Party (that emerged during Andrew Jackson’s presidency) from the abolitionist Republican Party (that emerged gradually in the years leading up to 1860).[16] Barnett describes the Reconstruction amendments as a triumph of the republican over the democratic view.[17] That triumph was short-lived, however, for the South used the democratic view as a justification for a white majority to impose the odious Jim Crow regime on an African American minority.[18] New Deal progressives succeeded in rendering the democratic view the dominant one, and modern conservatives as well as modern liberals are the heirs of the New Dealers insofar as they both profess commitment to the importance of majority rule and concern about the ability of judicial review to interfere with it.[19]

Barnett puts his finger on some of theoretical commitments that divide modern Americans. I do not here explore Barnett’s choice of the labels “Democratic” and “Republican” or the way he describes the political history; other contributions to this symposium take up those questions. [20] Here, I simply observe that even taken on Barnett’s own terms, competition between the republican and democratic views of sovereignty seems to be as old as the Constitution itself.

History is more complicated when it comes to Barnett’s argument about the role of the courts in protecting the natural rights of individual sovereigns against unauthorized government interference. As with the republican vision itself, Barnett does not claim that the Constitution’s original public meaning requires courts to review statutes as he suggests. On the contrary, he acknowledges that the constitutional text is silent on this point and maintains that the approach one takes “will depend on whether one holds a republican or democratic vision of the Constitution” (p. 111). Yet if founding-era commitment to the republican vision was as unwavering as Barnett maintains, and if the republican vision logically leads to greater reliance upon judicial review, one would expect to see many founding-era cases in which litigants came to the courts to enforce their rights to liberty and property against self-seeking democratic majorities. But Barnett does not identify federal or state cases in which litigants claimed—under either the federal or state constitutions—that statutes were invalid because they infringed upon the natural rights of the People.[21] To be sure, the lack of general federal question jurisdiction meant that federal courts, at least, would have had a limited opportunity to consider such claims, but that limited jurisdiction itself reflects an early view about the limited role of the federal courts.[22] Whatever support history gives to the case for a Republican Constitution, it appears to cut the other way with respect to the role of the courts in enforcing it.

III. The Rational Basis Test and the Legislative Process

The modern rational basis test instructs courts to uphold a statute if it can posit a permissible reason why the legislature might have enacted it. Barnett criticizes this test, insisting that courts must identify a statute’s actual purpose to evaluate its constitutionality. As he puts it, courts “should ferret out when [the legislature’s] ‘just powers’ are being invoked as a mere pretext to exercising powers that have not been—and cannot justly be—entrusted to a republican government, where the people are the ultimate sovereigns” (p. 112). Courts need to “realistically assess whether restrictions on liberty were truly calculated to protect the health and safety of the general public, rather than being the product of ‘other motives’ beyond the just powers of a republican legislature.”[23] This is necessary because “[r]equiring the government to identify its true purpose and then show that the means chosen are actually well suited to advance that purpose helps to smoke out illicit motives that the government is never presumed by a sovereign people to have authorized” (p. 232).

Barnett portrays the statutes at issue in many of the classic Fourteenth Amendment cases—including Lochner, Carolene Products, and Lee Optical—as illustrative of regulations actually designed to protect the economic interests of a powerful faction at the expense of a weaker minority rather than to advance any public interest in health or safety.[24] The anemic rational basis test permits such statutes to be characterized as reasonably calculated to serve a legitimate end, but Barnett maintains that anyone who believes that has been hoodwinked. The maximum-hours statute challenged in Lochner protected commercial bakeries from competition by smaller, ethnic bakeries;[25] the ban on filled milk challenged in Carolene Products insulated the makers of other dairy products from competition;[26] and the prescription requirement challenged in Lee Optical protected optometrists from competition by opticians who could sell cheaper glasses (pp. 222-223).

Barnett’s emphasis on the importance of recovering the legislature’s true purpose understates the complexity of identifying legislative intent. It is extraordinarily difficult—if possible at all—for a court to glean what was “really” going on behind the scenes of a statute.[27] A legislature is a multimember body, and different members may have different motives. Perhaps some legislators enacting a ban on filled milk were concerned about its health effects and others were beholden to a powerful dairy lobby. Whose intent controls? Is such a statute truly calculated to promote health and safety or is it the kind of rent-seeking statute that rational individual sovereigns would not countenance? Do the rent-seeking motives of some legislators corrupt the statute if other legislators act with the public welfare in view? Where, moreover, would a court look to discover the legislature’s true motive? Legislative history is unlikely to contain an express acknowledgement of illicit motive, and even if it did, floor statements and committee reports do not reliably reflect the views of the majority who supported the statute. Current doctrine accepts a possible, rational purpose—i.e., one that can be inferred from the statutory text—rather than engaging in a hunt for the actual, subjective purpose precisely because the latter is illusory.[28]

In addition to forcing identification of the government’s true purpose, Barnett calls for greater scrutiny of the fit between statutory ends and means. On the one hand, he offers a compelling case that modern courts have occasionally stretched even the existing rationality test too far. For example, it is indeed difficult to see the connection between safe casket-making and a funeral home director’s license.[29] A rational basis test ought not mean that courts are obliged to accept explanations that beggar all belief.

On the other hand, the strength of the “rational basis” test can vary according to the perspective of the beholder, and Barnett favors one with more bite. In calibrating the strength of the test, it is important to keep in mind—especially with respect to the kind of complex legislation that emerges at the federal level—that courts cannot seek too much perfection from the often-chaotic legislative process.[30] Modern textualists in particular have emphasized the ways in which the battle between competing interests shapes legislation. In the federal system, the process of bicameralism and presentment forces compromise between the House and the Senate, as well as between both houses and the President. But even within each house, “[b]ills are shaped by a process that entails committee approval, the scheduling of a floor vote, logrolling, the threat of filibuster, the potential for presidential veto, and an assortment of other procedural obstacles.”[31] Passing these veto gates requires proponents to compromise with opponents, and compromise can produce awkward language.[32] For example, it may be necessary to narrow or broaden language in order to bring others on board.[33] As Justice Thomas wrote for the Court in refusing to apply the absurdity doctrine to awkwardly drawn provisions in a pension statute:

[N]egotiations surrounding enactment of this bill tell a typical story of legislative battle among interest groups, Congress, and the President. Indeed, this legislation failed to ease tensions among many of the interested parties. Its delicate crafting reflected a compromise amidst highly interested parties attempting to pull the provisions in different directions. As such, a change in any individual provision could have unraveled the whole.[34]

Reaching agreement about how to handle a particular social or economic problem requires give and take from parties who have not only conflicting self-interests but also conflicting ideas about what best serves the public interest.

No statute, moreover, pursues its purpose at all costs. A legislature must draw the line somewhere, and deferential rational-basis review acknowledges that line-drawing is often awkward.[35] Take, for example, the statute at issue in Lochner. Barnett insists that it was irrational for the New York legislature to cap the hours of bakery employees but not bakery owners (p. 130). Even assuming that it would have better served the legislative purpose to cap the hours of the owners too, must the legislature do everything at once?[36] Perhaps the legislature drew the line at bakery employees because it thought they were, all things considered, likely to benefit more than owners from fewer hours. Or perhaps owners would have vehemently opposed limits on their hours, and the bill might have failed if they were included. A statute is not necessarily irrational or contrary to the public interest because it would have been reasonable to do more.

In sum, the Court’s current approach, which accepts hypothetical purposes and demands only minimal rationality, accounts for the normal functioning of the legislative process. The legislature is not an idealized body that acts with one mind, but a multimember body that produces legislation through a complex and even chaotic process. Any bill that runs the gamut of this process represents compromises made along the way, sometimes to resolve the competing desires of different constituencies and sometimes because the legislature has drawn a line somewhere.

  1. The Institutional Capacity of the Courts

Barnett proposes to alter the current regime of highly deferential judicial review of social and economic legislation in two respects. He would both put more bite into rationality review and reverse the presumption of constitutionality. Under current doctrine, courts assume that statutes are constitutional unless the challenger shows otherwise. When applied to federal legislation, this presumption reflects respect for a coordinate branch, and when applied to a state statute, it reflects respect for the states. Barnett implicitly rejects departmentalism, inter-branch comity, and federalism as good reasons for the presumption; he argues that the presumption is always inappropriate because it favors the servant over the principal.[37] A constitutional challenge to a statute is a mechanism by which an individual sovereign calls the legislature to account. The presumption, he says, should thus run the other way—the individual sovereign is entitled to an explanation for why the legislature has acted within the scope of its limited authority to secure the People’s natural rights. Reallocating the burden, particularly when combined with more searching substantive review, better preserves the sovereignty of the People.

Barnett’s call for greater judicial willingness to invalidate statutes reflects confidence in the ability of the courts to protect individual liberty, particularly relative to the legislatures he describes as so easily corrupted by faction. Yet while he offers a fulsome explanation of why we should mistrust legislatures, he spends less time defending the institutional capacity of the courts.

Highly deferential judicial review reflects the judgment that a more searching inquiry would pull judges into terrain they are not good at navigating. Rational-basis review under the Due Process and Equal Protection Clauses is a case in point. The current, deferential regime reflects humility about the capacity of judges to evaluate the soundness of scientific and economic claims; Barnett’s approach, by contrast, reflects confidence that they are up to the job. Is that confidence warranted? Are judges well suited to assess competing claims about the nutritional value of filled milk or a complex environmental policy?[38] To be sure, Barnett acknowledges that judges are not perfect; he observes that “[a]cross-the-board skepticism about the rationality of a restriction on liberty does not guarantee that prejudice bolstered by junk science will lose” (p. 148). He does not address the opposing concern, however, that judges will reject scientific or economic claims that they ought to accept.

Moreover, nearly every government regulation comes at some price to individual liberty. Determining whether a government regulation truly serves the public interest, therefore, requires determining whether the price is worth paying. Would courts be good at identifying and re-weighing relative costs and benefits of decisions like a ban on the use of medicinal marijuana?[39] They attempt to identify and reweigh the costs and benefits of state regulation in the context of the Dormant Commerce Clause, and there is longstanding controversy about whether they are equipped to do it well.[40]

Apart from institutional capacity, Barnett does not talk about the features of Article III that have traditionally provoked worry about over-zealous exercises of the power of judicial review. Consider life tenure. On the one hand, life tenure may make judges brave enough to stand up to the majority. On the other hand, life tenure means that judges are unaccountable for bad decisions so long as they are not insane or corrupt. If judges get it wrong, the People have no way to remove them from office. While one can doubt whether the People can effectively discipline legislators at the ballot box for any given policy choice,[41] legislators must still ultimately face the People to keep their jobs. Article III judges—particularly Supreme Court justices, whose decisions are not subject to reversal by a higher authority—answer to no one. The feeling of infallibility that accompanies finality is a force to be guarded against.[42] Properly understood, a commitment to judicial restraint is a commitment to resist the temptation to exceed the bounds of the judicial power.

My choice of the word “power” is deliberate. Barnett contrasts the “power” of Congress with the “duty” of judicial review. He says that Article I’s choice of the word “power” to describe the scope of Congress’s authority invokes “long-standing principles of agency law;” the government exercises its power “on behalf of and subject to the control of the principal,” the sovereign People who granted it power (p. 63, emphasis omitted). Power, he explains, goes hand in hand with limits. When he is talking about judicial review, by contrast, he stresses that it is not a “power” but a “duty” of the courts. In characterizing it as a duty, he stresses that Federalist No. 78 called it a “duty,” as did Marbury v. Madison in insisting that “[i]t is emphatically the province and duty of the judicial department to say what the law is.” (p. 60, emphasis omitted). The choice of the word “duty” rather than “power” is significant, Barnett argues, because “[p]owers can and should be exercised with discretion or ‘restraint,’ but we don’t speak the same way of our duties . . . [r]ather, we think these duties should be completely and fully honored. In contrast, the exercise of our powers can be characterized as a matter of discretion and moderation” (p. 126).

The Constitution, however, nowhere refers to a “duty of judicial review.” The ability to engage in judicial review exists by virtue of the “judicial Power” conferred by Article III, because that is the provision that enables federal courts to act. For constitutional purposes, then, judicial review is an exercise of power. If the distinction between “duty” and “power” matters in the way Barnett says—i.e., that the concepts of restraint and limits are inapplicable to the exercise of “duty”—then judicial review falls in the “power” category. Just as Congress must be cognizant of the limits upon its legislative power, the courts must be cognizant of the limits upon their judicial power.

Barnett is much less concerned about courts exceeding the limits of the judicial power than about their not doing enough. When he addresses the potential harms of judicial review, he focuses on the harm the Court inflicts when it fails to invalidate a statute that it should. To make the point, he invokes examples about which there is agreement that the Court should have intervened, such as Plessy v. Ferguson,[43] as well as those about which there is not, such as United States v. Carolene Products.[44] Barnett does not discuss, however, the harm the Court can do when the mistake runs the other way: when it invalidates legislation that it should let stand. Focusing on the danger of the Court’s doing too little rather than too much is consistent with Barnett’s generally libertarian approach, which is reflected in the presumption of unconstitutionality he would give statutes—regulation is presumed to be unauthorized unless the government can show otherwise.[45] If government should regulate less, it is better for the Court to err on the side of striking down too much regulation than on the side of letting too much stand.

Even under the republican vision that Barnett elaborates, however, wrongful invalidation of regulation should be concerning. He argues that the purpose of government is to secure our pre-existing rights and that the government does that through regulations that promote health and safety. If courts stand in the way of legitimate health and safety regulation, they cripple the ability of government officials (the agents) to pursue policies securing the rights of individual sovereigns (the principals). Barnett keeps his focus on the sovereigns who challenge legislation in court. But of course other sovereigns favor the challenged policy, and it is unfair (not to mention unconstitutional) for courts to prevent them from achieving permissible aims.

There are winners and losers in the battle over whether to pass any given statute. Barnett is not a big fan of democracy, but voting is undeniably part of our system. Even if our Constitution is the republican one for which Barnett advocates, when a majority of the elected representatives who serve the People supports a constitutionally permissible statute, that statute is binding even on those who dissent from it. Overall, Barnett too quickly dismisses concerns about judicial activism. While he is right to insist that courts ought not operate based on a distorted understanding of judicial restraint,[46] he overcompensates in the other direction. There is a risk that a faction can run away with the legislative process, but there is also a risk that a faction will conscript courts into helping them win battles they have already lost, fair and square.

When litigants challenge the constitutionality of a statute in the Supreme Court, the question is whether those who object to the statute are entitled to a national rule precluding such regulation. Barnett’s insights about federalism, while aimed at Congress, are relevant here. Barnett observes that nationalizing policy preferences risks increasing political polarization by entirely eliminating the possibility of dissent. “[T]he more important the issue,” he points out, “the more likely it will engender a political war of all against all to avoid having another’s social policy imposed on you” (p. 183, emphasis omitted).[47] While he is talking about Congress, the point is also applicable to the Supreme Court. Because the Court’s holding on a constitutional question stands as a national rule that precludes local variation, battles in high profile cases are incredibly pitched and their results can be politically polarizing.[48] The Court’s reluctance to disturb statutes that do not involve fundamental rights or suspect classifications limits the number of such battles that play out before it; insofar as state laws are concerned, the Court’s deferential approach errs on the side of permitting local variation. More vigorous enforcement of federalism might decrease the risk of over-nationalizing policy preferences at the hands of Congress; at the same time, more vigorous enforcement of the Due Process and Equal Protection Clauses may increase the risk of over-nationalizing policy preferences at the hands of the Supreme Court. Once the Supreme Court weighs in on a constitutional question, the entire nation is bound, and the opportunity for regional differences is extinguished.

Deferential judicial review of run-of-the-mill legislation is consistent with the reality that the harm inflicted by the Supreme Court’s erroneous interference in the democratic process is harder to remedy than the harm inflicted by an ill-advised statute. The Supreme Court’s constitutional mistakes are extremely difficult to correct; one can hope only for a change of heart, a change of personnel, or a change by constitutional amendment. By contrast, it is feasible, even if difficult, to repeal or amend bad statutes, and both Congress and state legislatures do it with varying levels of frequency. At the state level, moreover, the harm of an ill-advised statute is regionally confined. Even if one state legislature makes a mistake, the other forty-nine remain free to choose a different course. A Supreme Court constitutional error, however, applies nationwide.

When it comes to confidence in the courts, it is worth noting that Barnett’s examples highlight the ability of courts to curb the legislatures who enable rent-seeking majorities to unfairly restrict the liberty and property rights of minorities. He thus envisions the way that property-protective courts can curb the influence of progressive legislatures. But what if the tables are turned, and there are progressive courts and a property-protective legislature?[49] Would Barnett find it as attractive to have courts engaging in more aggressive judicial review? If one worries about the political inclinations of judges creeping in—and Barnett’s critique of progressive judges indicates that he does—then one might be warier about enhancing the risk that judges will confuse the demands of the Constitution with their own conception of the public interest.

Barnett characterizes courts as a refuge from the majoritarian excesses of the legislature, but they are only a refuge if they are untainted by the corrupting influences that Barnett sees in the democratic process. The history he recounts leaves one to wonder why he has such faith in courts. Would Prigg, Dred Scott, and Plessy have come out differently if courts had only applied the standard Barnett proposes? Was it really a misguided attachment to judicial restraint that drove those cases, or did the Court see through the same discriminatory lens as the legislature?[50] Courts are not always heroes and legislatures are not always villains. They are both capable of doing good, and they are both capable of doing harm.

Even if courts were always heroes, however, they could not offer us complete protection from legislative overreaching. The reality is that Congress and the President are frequently the only institutional actors with the opportunity to evaluate the constitutionality of legislation. In describing checks and balances, Barnett says, “Not only does such legislation have to pass two chambers of Congress, but it must be approved by the president and evaluated for constitutionality by the judiciary” (p. 211). But it doesn’t. Courts only get the opportunity to engage in judicial review when litigants with standing file complaints. Limitations upon standing—most notably, the general prohibition of “taxpayer standing”—mean that there are a great number of laws that federal courts never review. Courts do not, and in our constitutional system cannot, have as great a role in policing the political branches as Barnett would like them to.

  1. Judicial Restraint

Our Republican Constitution is animated in large part by Barnett’s frustration with what he regards as a misguided attachment to judicial restraint, particularly on the part of conservatives.[51] In NFIB v. Sebelius, the inspiration for Barnett’s book, Chief Justice Roberts pushed the Affordable Care Act beyond its plausible meaning to save the statute. He construed the penalty imposed on those without health insurance as a tax, which permitted him to sustain the statute as a valid exercise of the taxing power; had he treated the payment as the statute did—as a penalty—he would have had to invalidate the statute as lying beyond Congress’s commerce power.[52] Barnett vehemently objects to the idea that a commitment to judicial restraint—understood as deference to democratic majorities—can justify a judicial refusal to interpret the law as written.

Barnett is surely right that deference to a democratic majority should not supersede a judge’s duty to apply clear text. That is true, incidentally, even if one subscribes to the collective view of popular sovereignty, for a judge who adopts an interpretation inconsistent with the text fails to enforce the statute that commanded majority support. If the majority did not enact a “tax,” interpreting the statute to impose a tax lacks democratic legitimacy. Insofar as Barnett aims his critique of judicial restraint at the conservatives in his popular audience, it is worth considering why they might misunderstand the concept. It may be because they consider themselves originalists but misunderstand originalism.

Originalism is associated with judicial restraint in the popular consciousness because it emerged in the 1980s as a conservative response to the perceived activism of the Warren and Burger Courts.[53] Originalists insisted that the Court needed to be reined in so that the democratic process could function. They characterized originalism as a mechanism for stopping the minority of Supreme Court justices (and the elites who supported them) from imposing their will on the American majority. Originalism’s ability to restrain judges was trumpeted as its chief virtue. It “was thought to limit the discretion of the judge” and to promote “judicial deference to legislative majorities.”[54]

Originalists have refined their arguments in the intervening years, however, and they have abandoned the claim that one should be an originalist because originalism produces more restrained judges.[55] Originalism has shifted from being a theory about how judges should decide cases to a theory about what counts as valid, enforceable law.[56] The Constitution’s original public meaning is important not because adhering to it limits judicial discretion, but because it is the law. And because it is the law, judges must be faithful to it.[57] As Keith Whittington has explained, “The new originalism does not require judges to get out of the way of legislatures. It requires judges to uphold the original Constitution—nothing more, but also nothing less.”[58]

The measure of a court, then, is its fidelity to the original public meaning, which serves as a constraint upon judicial decisionmaking.[59] A faithful judge resists the temptation to conflate the meaning of the Constitution with the judge’s own political preference; judges who give into that temptation exceed the limits of their power by holding a statute unconstitutional when it is not. That was the heart of the originalist critique of the Warren and Burger Courts. At the same time, fidelity will inevitably require a court to hold some statutes unconstitutional.[60] When a statute conflicts with the Constitution, the fundamental law of the Constitution must take precedence, and the ordinary law of the statute must give way—because, properly understood, it is not law at all. A court does not overstep simply by holding a statute unconstitutional; it oversteps if it does so without constitutional warrant. Assessing whether the Court has been activist requires one to evaluate the merits of its decisions, not to tally the number of statutes it has held unconstitutional.

Given their commitment to textual fidelity, one would be hard-pressed to find many originalists who think that a court should find a way to uphold a statute when determinate text points in the opposite direction. That is certainly true if the relevant text is constitutional. The Constitution’s meaning is fixed until lawfully changed; thus, the court must stick with the original public meaning of the text even if it rules out the preference of a current majority. It is also true, however, if the relevant text is a statute. Most originalists in constitutional interpretation are textualists in statutory interpretation.[61] Textualists interpret statutes in accord with their original public meaning and maintain that their meaning is fixed until lawfully changed. Because textualists refuse to depart from clear statutory text, they would consider it wrong to twist statutory text beyond what its meaning will bear to avoid collision with a constitutional barrier.[62]

NFIB v. Sebelius might be explained by the fact that Chief Justice Roberts has not proven himself to be a textualist in matters of statutory interpretation. Even in straight-up statutory interpretation cases, Chief Justice Roberts has found himself on the opposite side of staunch textualists like Justices Scalia, Thomas, and Alito precisely because of his willingness to depart from ostensibly clear text to better serve the statutory purpose.[63] Indeed, Richard Re has dubbed the Roberts Court’s approach “the new Holy Trinity” after the case best known for openly prioritizing purpose over text. While the Roberts version does not expressly assert the power to depart from statutory text, Re observes that it accomplishes a similar result by considering “non-textual factors when determining how much clarity is required for a text to be clear.”[64] This methodology, when combined with Chief Justice Roberts’ devotion to constitutional avoidance,[65] has yielded cases like NFIB v. Sebelius.

To the extent that NFIB v. Sebelius expresses a commitment to judicial restraint by creatively interpreting ostensibly clear statutory text, its approach is at odds with the statutory textualism to which most originalists subscribe. Thus Justice Scalia, criticizing the majority’s construction of the Affordable Care Act in both NFIB v. Sebelius and King v. Burwell, protested that the statute known as Obamacare should be renamed “SCOTUScare” in honor of the Court’s willingness to “rewrite” the statute in order to keep it afloat.[66] For Justice Scalia and those who share his commitment to uphold text, the measure of a court is its fair-minded application of the rule of law, which means going where the law leads. By this measure, it is illegitimate for the Court to distort either the Constitution or a statute to achieve what it deems a preferable result.

All of this is to say that Barnett is not alone in his skepticism of either the Roberts Court’s conception of judicial restraint or its approach to statutory interpretation. Indeed, this is a point on which those who treat the original public meaning of text as a constraint might agree, regardless whether they embrace Barnett’s Republican Constitution.


        [1].    Carmack Waterhouse Professor of Legal Theory; Director, Georgetown Center for the Constitution.

        [2].    Diane & M.O. Miller, II Research Chair in Law, Notre Dame Law School. This essay was prepared for a roundtable on Our Republican Constitution hosted by the University of Illinois and the Georgetown Center for the Constitution. It benefited from the comments of the other participants, Jack Balkin, Randy Barnett, Jud Campbell, Kurt Lash, Sanford Levinson, Jason Mazzone, and Larry Solum, as well as of my colleagues Bill Kelley and John Nagle, who generously read an earlier draft.

        [3].    Randy Barnett, Our Republican Constitution: Securing the Liberty and Sovereignty of We the People (2016) [hereinafter Barnett].

        [4].    See, e.g., Randy Barnett, Restoring the Lost Constitution: The Presumption of Liberty (2004) (arguing, inter alia, that courts should replace a “presumption of constitutionality” with a “presumption of liberty”). Restoring the Lost Constitution has been described as “the best defense ever written of a libertarian or conservative/libertarian approach to constitutional law.” Steven G. Calabresi, The Originalist and Normative Case Against Judicial Activism: A Reply to Professor Randy Barnett, 103 Mich. L. Rev. 1081, 1081 (2005).

        [5].    See Alexander Bickel, The Least Dangerous Branch: The Supreme Court at the Bar of Politics (1962).

        [6].    See p. 56 (A faction is “a number of citizens . . . who are united and actuated by some common impulse of passion, or of interest, adverse to the rights of other citizens, or to the permanent and aggregate interests of the community.”) (quoting The Federalist No. 10, at 78 (James Madison) (Clinton Rossiter ed., 1961)).

        [7].    See p. 54 (observing that “[i]n a democracy, the debtors outnumber the creditors and the poor outnumber the rich”).

        [8].    See p. 27 (maintaining that under the republican approach, our representative government serves as “a popular ‘check’ on the servants of the people” rather than as “a practical way to ‘re-present’ the will of the sovereign people”); p. 23 (under a Republican Constitution, the “purpose [of government] is not to reflect the people’s will or desire—which in practice means the will or desires of the majority—but to secure the preexisting rights of the We the People, each and every one of us”).

        [9].    See pp. 33–41 (describing the origins of the Declaration and how it captured thinking about natural rights at the time of the Founding).

      [10].    See pp. 38–39; see also p. 69 (asserting that “the right to acquire, possess, and use property is a vital means to the pursuit of happiness”).

      [11].    Thus, for example, “any rational person” would consent to “the equal protection of their [sic] fundamental rights, including their [sic] health and safety” (p. 43); see also p. 75 (attributing this view of consent to John Locke).

      [12].    See pp. 169–84 (describing the Constitution’s structural and substantive constraints upon legislative power and arguing that the former are more effective in preserving individual liberty).

      [13].    See p. 86 (asserting that “in its early days, the Republican opposition to the Federalists was in defense of the constitutional limitations of national power that characterizes what I am calling our Republican Constitution”).

      [14].    See pp. 72–73 (describing the opinions in Chisholm v. Georgia, 2 U.S. 419 (1793)). Barnett disputes the claim that the ratification of the Eleventh Amendment represented an embrace of Justice Iredell’s view. See p. 80 (arguing that the Eleventh Amendment “said nothing to repudiate the underlying principle of individual popular sovereignty articulated by Jay and Wilson . . . [i]t merely changed the text of Article III to deny federal courts the jurisdiction to hear such cases”).

      [15].    See p. 73 (contrasting Justice Chase’s commitment to the sovereignty of each individual with Justice Iredell’s commitment to the sovereignty of the democratic majority in Calder v. Bull, 3 U.S. 386 (1798)).

      [16].    See p. 87 (asserting that the Democratic Party of Andrew Jackson and Martin Van Buren “often called itself ‘the Democracy’ because it presumed to speak for the people as a whole”); pp. 89-97 (maintaining that the Democratic Party was pro-slavery and that its view of popular sovereignty permitted the majority to enslave the minority); pp. 90–98 (describing the evolution of the Republican party from the antislavery movement and characterizing it as grounded in a commitment to the sovereignty of each individual).

      [17].    See pp. 106–11 (describing how the Reconstruction amendments led to a more Republican Constitution).

      [18].    See id. at 120–24 (describing the decline of the republican view in the post-Reconstruction era).

      [19].    See pp. 124–63 (describing the rise and influence of progressivism).

      [20].    See Jack Balkin, Which Republican Constitution? 32 Const. Comment. 31 (2016) (arguing that Barnett’s version of “republicanism” is closer to “natural rights liberalism” than to the “historical tradition of republicanism”); id. at 42 (arguing that Barnett’s imagined opposition “between the Republican and Democratic Constitutions is really a schematic or idealized version of the struggle between classical liberalism and progressivism at the beginning of the twentieth century”); id. at 43 (claiming that Barnett unfairly “lumps modern liberals together with progressives”); see also Sanford Levinson, Randy Barnett’s Disdain for Democracy (and John Marshall), 32 Const. Comment. 113, 114–16 (2016) (critiquing Barnett’s use of the term “republican”).

      [21].    Litigants could have raised such a challenge to federal legislation after the Fifth Amendment was ratified in 1791. See U.S. Const. amend. V (providing that no person shall be deprived of life, liberty, or property without due process of law). A claim that rent-seeking state legislation violated the United States Constitution could not have been made until after the Fourteenth Amendment was ratified. Nonetheless, one would expect to see such challenges made to state laws under state constitutions, particularly if those state constitutions were indeed committed to the republican vision. See p. 67 (contending that many state constitutions were so committed). Barnett does point out that Corfield v. Coryell, 6 Fed. Cas. 546 (C.C. Ed. Pa. 1823) invoked the concept of natural rights, but that case involved a claim by a nonresident that New Jersey was discriminating in violation of Article IV’s Privilege and Immunities Clause rather than a claim that a statute is invalid if it is not truly designed to secure the natural rights of the People.

      [22].    Congress did not enact a lasting grant of general federal question jurisdiction until 1875.

      [23].    See p. 125 (emphasis added) (praising the late-18th and early 19th century courts who took this approach).

      [24].    Barnett also characterizes the statute at issue in the Slaughter-House Cases, which required all butchers to use a particular facility, as one giving a private monopoly to the company that owned this facility. See pp. 115–16. Muller v. Oregon addressed a statute protecting white male union members from competition with women, and Nebbia v. New York involved a statute that helped large milk distributors avoid competition from small retailers operating in poor neighborhoods. See p. 223.

      [25].    See p. 138 (noting that the provision capping the hours of bakery employees at 60 per week benefited commercial bakeries that could schedule their workers in shifts at the expense of smaller, ethnic bakeries with fewer employees).

      [26].    See p. 156 (asserting that filled milk was healthier than fresh cow’s milk, which carried dangerous bacteria, and the “politically powerful dairy farm lobby” pushed Congress to ban it from the market).

      [27].    Cf. Kassel v. Consolidated Freightways Corp. of Delaware, 450 U.S. 662, 702-03 (1981) (Rehnquist, J., dissenting) (criticizing Justice Brennan’s argument that the Court should consider only the legislature’s actual purpose, rather than a possible purpose, in adjudicating a Dormant Commerce Clause challenge because, inter alia, “it assumes that individual legislators are motivated by one discernible ‘actual’ purpose . . .”).

      [28].    See John F. Manning, The Absurdity Doctrine, 116 Harv. L. Rev. 2387, 2450 (2003) (arguing that a court that looks for “‘actual’ legislative purpose risks attributing unwarranted coherence to the legislative process, which may entail logrolling or other strategic voting, making concessions to strongly felt but outlying interests, or papering over disagreements to ensure the legislation’s passage” and that respect for this “inherently unruly legislative process” requires judges to “focus only on the rationality of the legislative outcomes themselves, not on whether those outcomes further some actual or likely legislative purpose.”); see also Edwards v. Aguillard, 482 U.S. 578, 636–37 (1987) (Scalia, J., dissenting) (“For while it is possible to discern the objective ‘purpose’ of a statute (i.e., the public good at which its provisions appear to be directed) . . . discerning the subjective motivation of those enacting the statute is, to be honest, almost always an impossible task.”).

      [29].    See p. 233 (relating “story of the Benedictine monks of St. Joseph Abbey in Louisiana who were barred by the Louisiana State Board of Embalmers & Funeral Directors from selling caskets without a funeral home director’s license”).

      [30].    See Manning, supra note 28, at 2446 (“[T]he rational basis test . . . . starts from the premise that a properly functioning legislative process often produces imperfect legislation, rough accommodations, and uneven compromises.”).

      [31].    Id. at 2417.

      [32].    See Landgraf v. USI Film Prods., Inc., 511 U.S. 244, 286 (1994) (“Statutes are seldom crafted to pursue a single goal, and compromises necessary to their enactment may require adopting means other than those that would most effectively pursue the main goal.”); Preseault v. I.C.C., 494 U.S. 1, 19 (1990) (“The process of legislating often involves tradeoffs, compromises, and imperfect solutions, and our ability to imagine ways of redesigning the state to advance one of Congress’ ends does noes not render it irrational.”).

      [33].    See Manning, supra note 28, at 2417 (pointing out that imperfect statutory language “may well reflect an unrecorded compromise or the need to craft language broadly or narrowly to clear the varied veto gates encountered along the way to enactment”).

      [34].    Barnhart v. Sigmon Coal Co., 534 U.S. 438, 461 (2002).

      [35].    Cf. U.S. R.R. Retirement Bd. v. Fritz, 449 U.S. 166, 179 (1980) (observing that line-drawing “‘inevitably requires that some persons who have an almost equally strong claim to favored treatment be placed on different sides of the line,’ and the fact [that] the line might have been drawn differently at some points is a matter for legislative, rather than judicial, consideration”) (quoting Matthews v. Diaz, 426 U.S. 67, 83-84 (1976)).

      [36].    Cf. Lochner v. New York, 198 U.S. 45, 75 (1905) (Holmes, J., dissenting) (“Men whom I certainly could not pronounce unreasonable would uphold [the hours cap] as a first instalment [sic] of a general regulation of the hours of work.”).

      [37].    See p. 229 (asserting that a judge who “simply ‘presumes’ that the legislature is acting properly, or ‘defers’ to the legislature’s own assessment of its powers, then that judge is not acting impartially”). In any suit, someone has to bear the burden of proof. Here, either the challenger must bear the burden of showing that the statute is unconstitutional or the government must bear the burden of showing that it is. Despite Barnett’s wording, his attack on the presumption of constitutionality seems better understood as a claim about where the burden should be placed than as a claim that placing a burden reflects judicial bias.

      [38].    See p. 234 (criticizing the Carolene Products Court for relying on “junk science” to sustain a ban on the sale of “filled milk”). Barnett says that it was easy for the lower court in Lee Optical to conclude that operating a lensometer does not require the expertise of a specialist (like an optometrist) but can be operated by any “reasonably intelligent person” (like an optician) (p. 238). Regardless whether that conclusion would be easy to reach, many statutes would present far more complex questions.

      [39].    See p. 188 (suggesting that prohibiting medicinal marijuana would violate either the Due Process Clause or the Ninth Amendment).

      [40].    See, e.g., Camps Newfound/Owatonna, Inc. v. Town of Harrison, 520 U.S. 564 (1997) (Thomas, J., dissenting) (insisting that “any test that requires us to assess,” inter alia, “whether a particular statute serves a ‘legitimate’ local public interest” and “whether there are alternative means of furthering the local interest that have a ‘lesser impact’ on interstate commerce, and even then makes the question ‘one of degree,’ surely invites us, if not compels us, to function more as legislators than as judges.”); Southern Pacific Co. v. Arizona, 325 U.S. 761 (1945) (Black, J., dissenting) (“Whether it is in the interest of society for the length of trains to be governmentally regulated is a matter of public policy. . . . this Court [should] leave that choice to the elected legislative representatives of the people themselves, where it properly belongs both on democratic principles and the requirements of efficient government.”).

      [41].    See pp. 176–77 (arguing that it is a fiction to believe that voters can discipline the policy choices of legislators through elections).

      [42].    Cf. Brown v. Allen, 344 U.S. 443, 540 (1953) (Jackson, J., concurring) (“We are not final because we are infallible, but we are infallible only because we are final.”).

      [43].    163 U.S. 537 (1896).

      [44].    304 U.S. 144 (1938).

      [45].    In other work, Barnett has argued that courts should replace the presumption of constitutionality with a presumption of liberty. See supra note 4. He does not make that argument express in Our Republican Constitution.

      [46].    See infra Part V.

      [47].    See also pp. 183–84 (“[T]he more important the issue, the less it is fit to be decided at the national level.”).

      [48].    Cf. William Eskridge, Pluralism and Distrust: How Courts Can Support Democracy by Lowering the Stakes of Politics, 114 Yale L.J. 1279, 1313 (2005) (asserting that “Roe essentially declared a winner in one of the most difficult and divisive public law debates of American history,” and it “was a threat to our democracy because it raised the stakes of an issue where primordial loyalties ran deep”).

      [49].    Cf. John Copeland Nagle, Newt Gingrich, Dynamic Statutory Interpreter? 143 U. Pa. L. Rev. 2209 (1995) (pointing out that proponents of dynamic statutory interpretation may be less enthused about the project of updating statutes to reflect current political preferences when the congressional majority is conservative).

      [50].    A court opining that “we cannot say that a law which authorizes or even requires the separation of the two races in public conveyances is unreasonable,” Plessy v. Ferguson, 163 U.S. 537, 550–51 (1896), does not seem the kind of court likely to reach a different result even if the standard of review were more exacting.

      [51].    See p. 17 (asserting that with NFIB v. Sebelius, “[t]he chickens of the conservative commitment to judicial restraint had thus come home to roost.”); see also p. 81 (asserting that “the tragedy of the Supreme Court’s decision in the Obamacare case was made possible by modern-day ‘judicial conservatives’ accepting as valid the progressive attack on our Republican Constitution.”); p. 248 (“The visibility of our Obamacare challenge and the way a Republican-nominated, conservative chief justice snatched defeat from the jaws of victory, may prove to be a political inflection point.”).

      [52].    See NFIB v. Sebelius, 132 S. Ct. 2566, 2593–2600 (2012) (characterizing the “penalty” imposed by the individual mandate as a “tax”). The other four justices in the majority on this issue would not have needed to construe the penalty as a tax to save the statute, because they thought that the Commerce Clause authorized Congress to impose the mandate. See id. at 2609 (Ginsburg, J., concurring in part, dissenting in part) (“Unlike the Chief Justice, however, I would hold, alternatively, that the Commerce Clause authorizes Congress to enact the minimum coverage provision.”). The four dissenting justices objected that “[w]e have never held that any exaction imposed for violation of the law is an exercise of Congress’ taxing power—even when the statute calls it a tax, much less when (as here) the statute repeatedly calls it a penalty.” See id. at 2651 (joint opinion of Scalia, J., Kennedy, J., Thomas, J., and Alito, J., dissenting).

      [53].    See Keith Whittington, The New Originalism, 2 Geo. J.L. & Pub. Pol. 599, 601 (2004) (“[O]riginalism was a reactive theory motivated by substantive disagreement with the recent and then-current actions of the Warren and Burger Courts . . . .”). By saying that originalism “emerged in the 1980s,” I do not mean that it was entirely new. It is simply that before the 1980s, it was “not a terribly self-conscious theory of constitutional interpretation, in part because it was largely unchallenged as an important component of any viable approach . . . .” Id. at 599. It emerged “in its modern, self-conscious form” after it was attacked. Id.

      [54].    Id. at 602.

      [55].    Id. at 608-09 (explaining that the new originalists have largely abandoned the emphasis on judicial restraint).

      [56].    See Mitchell N. Berman & Kevin Toh, On What Distinguishes the New Originalism from the Old: A Jurisprudential Take, 82 Fordham L. Rev. 545, 559 (2013) (asserting that it is accurate to say that the new originalism “is principally a theory about ‘what counts as law.’”) (quoting Steven D. Smith, Reply to Koppelman: Originalism and the (Merely) Human Constitution, 27 Const. Comment. 189, 193 (2010).

      [57].    Keith Whittington, Originalism: A Critical Introduction, 82 Fordham L. Rev. 375, 393 (2013) (“There is now a widespread emphasis on the centrality of constitutional fidelity to the originalist project, rather than the centrality of judicial restraint.”).

      [58].    See Whittington, supra note 53, at 609.

      [59].    See Thomas Colby, The Sacrifice of the New Originalism, 99 Geo. L. J. 713, 751 (2011) (distinguishing between “judicial restraint—in the sense of deference to legislative majorities” and “judicial constraint—in the sense of promising to narrow the discretion of judges” (emphasis omitted)).

      [60].    See Whittington, supra note 57, at 393 (asserting that “the tendency of the judiciary to uphold or strike down political actions must be purely contingent” because it is “[t]he stringency of constitutional requirements and the decisions of political actors [that] will determine the extent to which an originalist court will actively strike down legislation”).

      [61].    See John F. Manning, Textualism and the Equity of the Statute, 101 Colum. L. Rev. 1, 8 (2001) (noting that “statutory textualists are originalists in matters of constitutional law”).

      [62].    Thus modern textualists have backed away from the absurdity doctrine, which justifies textual departures when the application of a statute’s plain text would produce an inequitable result. See Manning, supra note 28, at 2485-86 (arguing that the absurdity doctrine is inconsistent with the premises of modern textualism). They have also expressed doubt about the legitimacy of those canons that arguably permit courts to depart from a text’s ordinary meaning. See Amy Coney Barrett, Substantive Canons and Faithful Agency, 90 B.U. L. Rev. 109, 164 (2010) (asserting that substantive canons permitting courts to alter the language of a statute conflict with “[t]he bedrock principle of textualism,” which “is its insistence that federal courts cannot contradict the plain language of a statute, whether in the service of legislative intention or in the exercise of a judicial power to render the law more just”).

      [63].    Compare King v. Burwell 135 S. Ct. 2480, 2496 (2015) (holding in an opinion written by Chief Justice Roberts that the phrase “Established by the State” in the Affordable Care Act allows tax credits for insurance purchased on an exchange established by the federal government) with id. at 2496 (Scalia, J., dissenting) (insisting in an opinion joined by Justices Thomas and Alito that an exchange established by the federal government does not qualify as an exchange “Established by the State”); compare Bond v. United States, 134 S. Ct. 2077 (2014) (holding in an opinion written by Chief Justice Roberts that the defendant’s use of toxic chemicals to injure her husband’s lover did not constitute use of a “chemical weapon,” defined by the relevant statute as “any chemical that can cause death, temporary incapacitation, or permanent harm to humans or animals”) with id. at 2094 (Scalia, J., dissenting) (insisting in an opinion joined by Justices Thomas and Alito that “it is clear beyond doubt that [the Chemical Weapons Convention Implementation Act of 1998] covers what Bond did; and we have no authority to amend it. So we are forced to decide—there is no way around it—whether the Act’s application to what Bond did was constitutional”). Chief Justice Roberts has also joined opinions that reflect disagreement with the Court’s textualists. In Yates v. United States, 135 S. Ct. 1075 (2015), Chief Justice Roberts joined the majority to hold that a fish is not a “tangible object” for purposes of the Sarbanes-Oxley Act; Justice Alito concurred only in the judgment, and Justices Scalia and Thomas joined Justice Kagan’s dissent. In American Broadcasting Co., Inc. v. Aereo, 134 S. Ct. 2498 (2014), Chief Justice Roberts joined the majority’s interpretation of the word “perform” in the Copyright Act; Justices Scalia, Thomas, and Alito dissented. See id. at 2515.

      [64].    Richard M. Re, The New Holy Trinity, 18 Green Bag 2d 407, 408 (2015).

      [65].    See Neal Kumar Katyal & Thomas P. Schmidt, Active Avoidance: The Modern Supreme Court and Legal Change, 128 Harv. L. Rev. 2109, 2112 (2015) (arguing that the Roberts Court has applied the avoidance canon so aggressively that it is willing to rewrite statutes to avoid addressing constitutional questions).

      [66].    See King, 135 S. Ct. at 2507 (Scalia, J., dissenting). King, of course, is a case about statutory rather than constitutional interpretation, but insofar as the Court strained the language to avoid a holding that would have gutted the statute, the opinion reflects the same impulse animating NFIB v. Sebelius.