The complete independence of the courts of justice is peculiarly essential in a limited Constitution. By a limited Constitution, I understand one which contains certain specified exceptions to the legislative authority; such, for instance, as that it shall pass no bills of attainder, no ex post facto laws, and the like. Limitations of this kind can be preserved in practice no other way than through the medium of courts of justice, whose duty it must be to declare all acts contrary to the manifest tenor of the Constitution void. Without this, all the reservations of particular rights or privileges would amount to nothing.
Introduction
The United States was not constituted to give unlimited sway to popular sovereignty.Footnote 1 Political institutions designed in the founding era were designed to check, as well as give voice to, the will of majorities. At the end of the eighteenth century, the idea of a written constitution as fundamental law, whose provisions had a status higher and apart from ordinary statute law was new, yet the framers understood the new constitution to place certain matters beyond the purview of the legislative – or the executive – authority.Footnote 2 Delegates who met in Philadelphia in 1787 and who drafted the constitution expressed realmisgivings about what popular majorities might do.Footnote 3 They had before them examples of Daniel Shays’ rebellion and expressed worries about debtor relief and other provisions in newly enacted state constitutions. One prominent thesis about the founding era is that, in 1787, wealthy elites, concerned over what they viewed as a dangerous excess of democracy, managed a kind of coup against the Declaration and the democratic sentiment in the Articles of Confederation.Footnote 4 The House of Representatives, the body closest to the people, with members directly elected by them, was checked by the Senate and by the possibility of a presidential veto of legislation, yet one of Thomas Jefferson’s complaints to Madison about the new American Constitution was that it was too easy to pass legislation. He thought any bill should have to be engrossed one year and considered without amendment the next before it could be passed, with a two-thirds vote of both Houses required if greater speed were deemed necessary.Footnote 5
The federal judiciary, the Third Branch, was designed to play a role in curbing popular majorities. The scope of that power has been the subject of a great many scholarly treatments and disagreements; suffice it to say here that the power of judicial review, long established, is often justified today in terms of the protection of individual rights enshrined in the Bill of Rights and elsewhere in the constitution (e.g., habeas corpus) against majorities that may wish to trample these rights.Footnote 6 As we will see, the role of constitutional courts in regimes that celebrate popular sovereignty is complex, and invocation of language about judicial activism and restraint is not of much help.
The relationship between constitutional courts and popular sovereignty is potentially fraught in any nation that aspires to democracy. In some nations and under certain circumstances there are provisions that permit an override of the judiciary. In a few nations, a simple legislative majority can negate some kinds of decisions of constitutional courts. In Canada, the “notwithstanding clause,” found in Section 33 of the Canadian Charter of Rights and Freedoms permits both the federal and provincial legislatures to expressly declare by a simple majority that a law shall operate even if a constitutional court has ruled the law unconstitutional, when the matter involves Sections 2 and 7–15 of the charter, which contain many of that charter’s most important rights guarantees.Footnote 7 A Section 33 override lasts for five years, but can be renewed upon its expiration. After the provision was added to the Canadian Charter in 1981, then-Prime Minister Pierre Trudeau said: “[I]t is a way that the legislatures, federal and provincial, have of ensuring that the last word is held by the elected representatives of the people rather than by the courts.”Footnote 8 In Israel, a similar notwithstanding clause was added to the Freedom of Occupations chapter of its “Basic Law” in 1994, allowing a simple majority of the Knesset, to override certain judicial decisions.Footnote 9 The override can last for four years. In a different sort of constitutional arrangement, there may be “directive principles of public policy” that are committed solely and expressly to the legislature, unenforceable by courts.Footnote 10
The US Constitution does not provide for such overrides of the judiciary, and with its cumbersome Article V amendment process, “correcting” unpopular readings by use of formal amendments has been difficult.Footnote 11 However, Congress can sidestep and effectively erase court readings of ordinary statutes by passing new laws. On occasion, a displeased federal legislature has also stripped federal courts of jurisdiction to hear certain kinds of cases.Footnote 12
If apex courts protect the separation of powers, they help prevent any one branch of government from claiming to be the sole voice of the popular sovereign.Footnote 13 When perceived as protecting constitutional barriers to raw democratic power, constitutional courts have often been targeted by contemporary populist attacks on liberal institutions. This is part of a pattern of attacks on the foundations of liberal constitutional order in the name of democratic equality and popular sovereignty flagged by Ewa Atanassow earlier in this volume.Footnote 14 Andrew Arato observes that “in the populist struggle against enemies, from Peron to Indira Gandhi, and from Fujimori to Orbán, the one constant is the attack against independent apex courts, that fully ceases only when such a body entirely loses its independence ….”Footnote 15 Since the time of Peron, populist regimes have attacked, or exhibited antagonism toward, independent apex courts.Footnote 16 When courts uphold a secular liberal democratic order, they may also be seen as facilitating the “spread of a cultural liberalism at odds with custom and religion.”Footnote 17
Support by political elites for courts could even be another factor contributing to the erosion of public faith in the project of legal neutrality.Footnote 18 There is evidence that elites have been turning to constitutional courts to entrench their most important policy and rights preferences (e.g., property rights) against waves of democratization.Footnote 19 Judicialization of political conflicts over the past half-century or more, removing political disputes from legislatures to courts, is a manifestation of this process. For some scholars, this behavior of constitutional courts represents a transition from judicial review to judicial supremacy.Footnote 20
American Support for Constitutional Courts
While the rise of populism has also affected American politics, the US Supreme Court still enjoys a good deal of legitimacy. Despite occasional attacks by former president Trump on the judiciary, blaming “Obama judges” for decisions with which he disagreed and expressing (as a candidate) the belief that a Mexican-heritage judge could not be impartial, the court has fared reasonably well, and the chief justice, a Republican appointee, has defended judicial impartiality.Footnote 21 In Gallup Polls, Americans express considerably more confidence in the Supreme Court than in Congress, although that poll registered an 11 point drop in confidence in the Court between 2021 and 2022.Footnote 22
Support for the court has waxed and waned. When populist and progressive reformers were frustrated with the court for invalidating state and federal legislation dealing with social and economic ills a little more than a century ago, there were numerous calls to end judicial review.Footnote 23 That progressive era tide subsided. Yet, it remains the case that “[a] central puzzle for the study of judicial review is identifying how judges are able to exercise the power of judicial review so successfully and so often.”Footnote 24
Current levels of support for the Supreme Court are affected by partisanship. At the same time that former president Trump was challenging judicial independence, Republicans were gaining additional seats on the Supreme Court and on the federal bench. While 46 percent, of the US citizens surveyed in August 2022 in the Annenberg Civics Knowledge Survey trusted the Supreme Court to work in the best interests of the American people (down from 68 percent in 2019, when the question was last asked), 70 percent of Republicans expressed a great deal or a fair amount of trust in the Court, while only 32 percent of Democrats (and 44 percent of independents) did. Only 40 percent of those surveyed in 2022 believed that Supreme Court justices “set aside their personal and political views and make rulings based on the Constitution, the law, and the facts of the case,” but ony 29 percent of Democrats, as opposed to 55 percent of Republicans (and 41 percent of independents) took this position.Footnote 25
There may be good reason to believe that the more politicized the court appears the less deference is accorded to court interventions in political controversies. The polarization of American politics has now become more pronounced on the Roberts Court. Major decisions that split justices along partisan lines and bitter, partisan confirmation battles (including the majority-Republican Senate’s refusal to hold confirmation hearings in the case of Merrick Garland (President Obama’s last nominee) fuel perceptions that the court is ideologically driven. “For the first time in the Supreme Court’s history, every Republican on the court is to the right of every Democrat,” observed Jeffrey Segal, who has used what are known as Martin–Quinn scores to plot the relative location of Supreme Court justices on an ideological continuum by looking at the votes they cast on cases.Footnote 26 A court that overturns long-standing precedent by narrow margins is likely to be perceived as simply another arena of partisan warfare.
Some legal scholars in the United States have been concerned (even prior to Bush v. Gore in 2000), about the propensity to defer to the court’s declarations about constitutional meaning. This deference occurs alongside the court’s willingness, in recent decades, to wade into politically charged controversies, sometimes declaring that it alone can determine the meaning of the constitution, and leading to charges that prudential judicial restraint is a thing of the past.Footnote 27 When other branches as well as the public are willing to accept that the court alone determines the meaning of the language in the constitution, deliberation about that document’s values and meaning – along with a sense of ownership – as Cass Sunstein and others have noted, declines.Footnote 28 Such a court may then be contributing to a democratic deficit.
In this vein, Mark Tushnet toyed with the idea of taking the constitution away from the courts, or at least providing instead for a form of weak judicial review as a way for American citizens to own the constitution again.Footnote 29 Weak-form systems of judicial review “openly acknowledge the power of legislatures to provide constitutional interpretations that differ from … [or] alter … the constitutional interpretations provided by the courts.”Footnote 30 This kind of review is a kind of dialogue between the court and the legislature, allowing for revision of the court’s constitutional judgments.Footnote 31 “Weak-form review combines some sort of power in courts to find legislation inconsistent with constitutional norms with some mechanism whereby the enacting legislature can respond to a court decision to that effect.”Footnote 32 However, Tushnet argues, legislative interventions cannot be too frequent or become routine if the system is not to morph into one of parliamentary supremacy.Footnote 33
While there are dangers to institutionally constrained, liberal popular sovereignty arising from attacks on constitutional courts, there are also dangers that flow from extreme deference to such courts if popular sovereignty is to be anything other than a fiction we invoke.Footnote 34 If the Supreme Court engages in decision-making that undercuts processes that help construct a democratic “people,” such decisions become highly problematic, even if they are accepted by a majority of “the people.”
Determining who are the sovereigns in the United States adds another complication. Atanassow, Bartscherer, and Bateman write in the Introduction to this volume that “the fundamental questions of popular sovereignty” concern questions of people: Who is or are the people, against whom is it defined, and who can rule in its name?Footnote 35 As Azari and Nemecek point out in this volume “there are many overlapping relevant political communities” in the United States; we will explore some dimensions of this particular problem below.Footnote 36
Situating the Us Court Among Popular Sovereigns
At the height of the Warren Court’s liberal rights jurisprudence, Alexander Bickel famously worried about an unelected tribunal whose members have lifetime appointments to insulate them from political pressure; identifying a “countermajoritarian difficulty,” he recommended the court exercise the “passive virtues.”Footnote 37
A major counterargument in this long-running discussion is that the court has largely worked in concert with other branches of the federal government and is much more aligned with dominant political coalitions than Bickel’s worry suggests.Footnote 38
The court can do harm not only by thwarting the wishes of popular majorities but also by acceding to them. An illustration that the latter problem poses for systems of popular sovereignty with limits is that some of the court’s most egregious decisions regarding indigenous peoples have been robustly majoritarian. Lone Wolf v. Hitchcock (1903), referred to as “the Indians’ Dred Scott decision” established congressional plenary power over Native Americans. The court held that Congress could legislate for American Indians just as they did for ordinary Americans, and no longer had to respect past treaties. In an act of what some would call judicial restraint, the court took itself out of the business of reviewing congressional decisions that established direct governance over Native Americans, stating (with great misrepresentation): “Plenary authority over the tribal relations of the Indians has been exercised by Congress from the beginning, and the power has always been deemed a political one, not subject to be controlled by the judicial department of the government.”Footnote 39 Supporting settler colonizers and also progressive reformers who wanted Indians to live under the law like other Americans the court followed public sentiment and dominant narratives.Footnote 40 This meant that Native Americans were generally legislated for without representation, since “Indians, not taxed” were still deemed ineligible to vote by some western states when they set voter qualifications. Even after the Indian Citizenship Act of 1924, Native Americans living on reservations were sometimes deemed under guardianship or not state residents and were not fully enfranchised until courts intervened in the years leading up to, and immediately following World War II.Footnote 41
While the McGirt decision (2020) was celebrated as a victory for Native American authority over reservation lands and a rebuff to states seeking to enforce their own criminal laws over crimes committed there, in fact the court’s 5–4 holding simply established that unless and until Congress explicitly abrogated a treaty promise, tribal authority over lands that were explicitly protected by treaty remained intact.Footnote 42
An issue that has too often been neglected in considerations of when popular majorities should be curbed is which majorities, if any, are being thwarted (although it is sometimes suggested that federal courts lean in the direction of supporting federal power).Footnote 43 Some make the argument that the majority the court should follow is the one that adopted the constitution – even against temporary latter-day majorities.Footnote 44 This claim, however, would require that constitutional meaning can be clearly ascertained, which does not help us very much if meaning often has to be constructed.Footnote 45
Popular sovereignty is exercised through an ensemble of institutions in the American constitutional structure. By constitutional design, each branch of the federal government makes efforts to represent “We the People.”Footnote 46 Popular sovereignty is exercised at the subnational as well as the national level. This makes an attempt fruitless to locate a specific guardian – or single institution policing the boundaries – of popular sovereignty. With multiple layers of governance, national power is not all-encompassing, and powers are sometimes concurrent or overlapping; there may frequently be more than one sovereign. Court can thwart popular majorities at one level of government while supporting popular majorities at another.
We therefore must incorporate federalism when exploring the relationship between national-level constitutional courts and popular sovereignty in the United States. While Article V federal amendment procedures are cumbersome, state activity in constitutional amendment and rewriting is ongoing and dynamic, responding far more readily to popular pressures. In practice, as Robinson Woodward-Burns argues, states play an important role in constitutional development as nationally divisive controversies are pushed to the state level. Finding a positive association between the topics of federal and state constitutional proposals, Woodward-Burns argues that state constitutional reforms help stabilize national constitutionalism; the result is a patchwork of changes that are nonbinding on national actors. Many matters are left to the states, partly because the court accepts such a small percentage of cases that parties would bring before it. Most regulation of elections, police powers, referenda, and grassroots initiatives occur at the state level.Footnote 47
When thinking about matters best left to the voters and their elected representatives and matters that ought not be decided by majority vote for the sake of preserving the health of bounded popular sovereignty, constitutional courts sometimes have conflicting claims of both state and federal majorities to factor in alongside a founding document that seeks to limit certain kinds of harms that national majorities might inflict. As Keith Whittington observes, “State laws pose the question of not only whether the political majority should get its way but also which political majority should get its way.”Footnote 48
Several important cases decided in the recent past raise pointed questions about the role the Supreme Court has to play in maintaining the health of popular sovereignty in a liberal institutional order. In the Crawford decision (2008), the court upheld vigorous new state requirements for voter identification, imposed in the name of enhancing confidence in the integrity of the electoral system, without any evidence of in-person voter fraud, giving the green light to states wishing to impose such burdens on what had been considered an extremely important right. In Citizens United (2010), Justice Kennedy, writing for the court, found that only quid pro quo corruption warranted congressional intervention in election contributions by individuals or corporations; the appearance of corruption rationale the court held sufficient to justify the Federal Election Campaign Act in Buckley (1976, 26–30) (for undermining faith in the integrity of the electoral system) was no longer sufficient. In Shelby County v. Holder (2013), the court invalidated the congressional formula used in the preclearance provisions of the 1965 Voting Rights Act – reaffirmed by Congress most recently in 2006 – effectively erasing Section 5 and leaving the Department of Justice the recourse of initiating prosecution on a case-by-case basis under Section 2.Footnote 49 The court went further in Brnovich (2021) and made it much more difficult for the federal government to prevent states from implementing new voting restrictions under Section 2, absent proof of a racially discriminatory purpose.
In addition, in Schuette (2014), the court determined that state voters could instruct the state of Michigan not to take race into account in admissions and hiring. In Obergefell v. Hodges (2015), the court, nationalizing a right, held that majorities within states could not bar same-sex couples from marrying. And in Rucho v. Common Cause (2019), the court determined that extreme politically motivated gerrymandering in the states did not present a justiciable question.
This array of important decisions sometimes thwarted the will of national majorities expressed in statute law in favor of states, or a majority of voters in some states; sometimes upheld emerging national majority opinion and rights of gay couples against majorities in dissenting states; sometimes empowered state citizens to challenge state actors seeking to promote racial diversity in state institutions; or empowered members of one political party to reduce the effectiveness of the vote of the opposition party within a state. All these cases had invited Supreme Court interventions in the political process and raised, directly or indirectly, equal protection issues.
For liberal legal scholars such as the late John Hart Ely, the strongest rationale for judicial review is to correct a democratic deficit and pursue representation reinforcement where the political process had failed. In Democracy and Distrust, Ely famously claimed that the constitution embodied a textual commitment to democracy as a process for resolving issues.Footnote 50 The judiciary played an important role in upholding that commitment, by protecting discrete and insular minorities who had been systematically barred from achieving their goals through the political process (via discrimination). When Congress chose to impose some costs on white Americans by enacting affirmative action, the court had warrant to uphold such measures; since African Americans had been systematically excluded from participation in the political process, they could not rely upon the political process to achieve redress (Ely, 1980). As one scholar restates Ely’s formulation, judges “should try to make representative democracy more democratic. They should try to make democracy work according to its own underlying principles.”Footnote 51
Somewhat similarly, Akhil Amar, for whom the underlying and legitimating principle of America’s constitution is popular sovereignty, contends that statutes passed when suffrage has become more inclusive may trump the earlier written constitution, especially when a modern Congress is protecting citizens at risk of being systematically injured on the basis of their birth status. If a post-Nineteenth Amendment statute protective of women’s rights conflicted with an older statute or even a constitutional provision that restricted women’s rights or interests, the recent one should be preferred since women were not excluded from the decision-making process or the electorate – correcting a retrospective democratic deficit.Footnote 52
Below, we take a deeper dive into two of the cases mentioned above that are less well mined than the others, posing questions about the relationship between constitutional courts and popular sovereignty in the context of equal access to the political process. Such access is essential if the voice of the “people” is to be credited in a liberal political order. In both cases, the response of the court is troubling.
Partisan Gerrymandering: Rucho v. Common Cause (2019)
Since Carolene Products (1938, footnote 4), the court has frequently reiterated that it closely scrutinizes cases in which government impinges on the rights of discrete and insular minorities. Race, the court recognized, is different, and the Thirteenth, Fourteenth, and Fifteenth Amendments are read as conferring special court responsibility to closely scrutinize classifications based on race. Certain classifications are inherently suspect (Yick Wo v. Hopkins, 1886; Korematsu v. United States, 1944). The constitution says nothing about parties; the framers viewed parties as factions inimical to the general interest.Footnote 53 And yet parties quickly became essential to organizing political preferences, and the party system became integral to American politics without formal amendment to the constitution. What about the rights of persons who identify with a political party – are there any equal protection rights, or associational rights?
Inherent in the notion of free and fair popular elections is that, when the voice of the people is authoritative in determining political outcomes, votes should be counted equally. That principle is sometimes modified or violated in the United States. The Electoral College and the election of US senators modify this principle. The Electoral College design provided for no direct election of the president; each state made provision for selection of electors, but since South Carolina allowed for their direct popular election (1868), all states allowed voters to select electors.Footnote 54 Since 1832, almost all states have allocated their electors on a winner-take-all basis. Some scholars complain that the method of election of US senators provided for in the US Constitution – with sparsely populated states (e.g., Wyoming) having the same number of senators as heavily populated states (e.g., California) – is highly undemocratic, yet a clause in the constitution stipulates that no state can be deprived of its equal representation in the Senate without its own consent, effectively making change impossible.Footnote 55 States, not American citizens at large, were to be represented in the Senate.
The right to vote – and to have one’s legitimate vote counted – has often been thought of as having special status in American democracy. “The political franchise of voting” was deemed a “fundamental political right, because preservative of all rights” in an important 1886 Supreme Court case (Yick Wo, 370), and quoting that case in a 1960s apportionment case (Reynolds v. Sims, 1964, 361–62), the court stated:
Undoubtedly, the right of suffrage is a fundamental matter in a free and democratic society. Especially since the right to exercise the franchise in a free and unimpaired manner is preservative of other basic civil and political rights, any alleged infringement of the right of citizens to vote must be carefully and meticulously scrutinized.
While the court occasionally protected the rights of blacks denied the right to vote because of white-only primaries and grandfather clauses (Smith v. Allwright, 1944; Guinn v. United States, 1915), beginning in the 1960s, the political branches of the federal government weighed in more vigorously to protect voting rights of African Americans, and the court supported these efforts for nearly half a century. Because of a legacy of black vote denial and voter intimidation in the South and the purpose of the Civil War Amendments, the power given to Congress to enforce those amendments (e.g., §5 of the Fourteenth Amendment and §2 of the Fifteenth) and the 1965 Voting Rights Act with subsequent amendments, race has generally been treated differently in the court than other kinds of interference with counting votes equally. In 1960, the court held that when Tuskegee, Alabama, redrew its boundaries to become an irregular twenty-eight-sided city, removing all but about five black voters and no white voters, it violated the Fifteenth Amendment (Gomillion v. Lightfoot, 1960). When it comes to drawing electoral districts on a racial basis, the court has held that voters should have an equal chance to elect the candidate of their choice, and no changes in a VRA-covered state or district can lead to retrogression in the right to vote.
Since 1993, the court has demanded that, even when race is a predominant factor in redistricting based in a desire to remedy past racial discrimination, such plans must be viewed using the highest level of scrutiny: The government must demonstrate a compelling interest in using the race classification and the remedy has to be narrowly tailored to achieve this compelling purpose (Shaw v. Reno, 1993).
The Supreme Court also began to intervene in districting that was not simply race based in the 1960s, determining that the equal protection clause of the Fourteenth Amendment and Article I §2 of the constitution required that districts be drawn so that they were as closely equal in population as possible. This rule governed both state legislative districts and US congressional districts (Baker v. Carr, 1962; Reynolds v. Sims, 1964; Westberry v. Sanders, 1964). Justice Felix Frankfurter, objecting to the court’s intervention to equalize the population in state districts, cautioned that “there is not under our Constitution a judicial remedy for every political mischief.”Footnote 56
When the court issued its decision in Rucho v. Common Cause in 2019, it declared state legislative gerrymandering of electoral districts designed to advantage a specific political party a political question. Political questions are matters the court refuses to engage in, either because they are thought to be textually consigned by the constitution to a coordinate branch of the federal government or, for prudential reasons, the court thinks it cannot find a judicial remedy or manageable standards for resolving them.Footnote 57 Infrequently used in recent years, invocation of the political questions doctrine is often applauded as an indicator of the court’s wisdom and self-restraint. But “when the Court refrains from engagement because it cannot devise or identify appropriate tests or standards, it means that constitutional guarantees may not yield judicially enforceable rights.”Footnote 58
The issue of partisan gerrymandering as an equal protection issue under the Fourteenth Amendment was not new to the court. Since 1986, the court has been asked to determine that gerrymandering for purely partisan purposes constituted vote dilution, impeding the ability of voters to elect the candidate of their choice. That year, the court held that cases of partisan gerrymandering could, at least in some cases, be heard. But “unconstitutional discrimination occurs only when the electoral system is arranged in a manner that will consistently degrade a voter’s or a group of voters’ influence on the political process as a whole,” and “plaintiffs were required to prove both intentional discrimination against an identifiable political group and an actual discriminatory effect on that group.” And furthermore, the constitution does not mandate proportional representation; legislatures do not have to “draw district lines to come as near as possible to allocating seats to the contending parties in proportion to what their anticipated statewide vote total will be.”Footnote 59
In Rucho, the court found no judicially manageable standard for intervention in even the most extremely partisan gerrymanders; dissenters said the tools and standards were there to be used. Blatant partisan gerrymanders are troubling; they may contribute to polarization, they threaten democracy, and may undermine faith in the democratic process. But the federal judiciary declines to intervene. The remedies the court left were state constitutions and statutes, congressional legislation, or possibly independent redistricting commissions (which the conservative minority would have barred under Article I §4 because establishing them took power away from the legislature in Arizona State Legislature v. Arizona Redistricting Commission in 2015).
Deference to “we the people” does not seem to promote the health of democratic self-governance. Elected leaders have incentives to entrench themselves (either collusion among the elected or collusion to maximize the strength of their parties in Congress) at the expense of voters. Vote dilution, the ability to elect candidates of one’s choice, associational rights, and equal protection now depend, in this context, on elected branches. The invocation of the political questions doctrine to extricate the court from this controversy arguably undermines faith in the democratic process, discourages voters whose candidates have been deliberately engineered out of contention, and depresses participation in the political process. The court’s nod to restraint by deferring to legislative redistricting practices deliberately aimed at minimizing the other major party’s electoral chances tends to undermine free, open, and fair elections – much as the court’s more interventionist Citizens United decision in 2010 did.
Rights of Minorities and Schuette v. Coalition to Defend Affirmative Action (2014)
This case is not about how the debate about racial preferences is resolved. It is about who may resolve it. There is no authority in the Constitution of the United States or in this Court’s precedents for the Judiciary to set aside Michigan laws that commit this policy determination to the voters.Footnote 60
Justice Kennedy, author of the fractured holding in Schuette v. Coalition to Defend Affirmative Action (2014), also penned the major gay rights and gay marriage decisions in Romer (1996), Lawrence (2003), Windsor (2013), and Obergefell (2015). Only two other justices supported the language in Kennedy’s Schuette opinion; three other justices wrote concurrences. It is useful to compare Schuette (2014) and Romer v. Evans (1996) as we consider why popular majorities in the states are permitted to make some decisions that have meaningful consequences for minorities but not others. Michigan voters adopted a constitutional amendment by ballot initiative, prohibiting state universities, employers, and contractors from discriminating or from giving any kind of racial preferences.Footnote 61 Schuette held that this amendment did not violate the Fourteenth Amendment equal protection clause. The court had never maintained that the federal or state governments were required to use affirmative action to remedy society-wide inequalities – at best, for a few years, such remedies were permitted;Footnote 62 and recognition of race for remedial purposes is now subjected to strict scrutiny and is acceptable only in very narrow circumstances.Footnote 63 Kennedy reasoned – and the effect of the decision means – that if the voters of Michigan choose to bar the state from recognizing race in their university admissions decisions, they may do so.
Colorado voters attempted to exclude LGBTQ individuals from the protections of antidiscrimination legislation in employment or housing by constitutional amendment (Proposition 2). Passage of Proposition 2 also retroactively invalidated measures that had been passed (including by municipalities) to extend antidiscrimination protections to gays and lesbians.Footnote 64 In Romer v. Evans, Kennedy saw no rational basis – only animus against a group of people based on their sexual orientation – and sided with liberals to strike down Proposition 2. “A law declaring that in general it shall be more difficult for one group of citizens than for all others to seek aid from the government is itself a denial of equal protection of the laws in the most literal sense.”Footnote 65 Kennedy’s reasoning in Romer was that the Colorado Amendment “exclud[ed] sexual minorities from accessing political remedies that were freely available to others.”Footnote 66 The perspective that treating gays and gays alone differently violated equal protection (and privacy) without having to consider them a suspect class paved the way for the larger same-sex marriage decision in Obergefell (2015). In the case of gays and lesbians, a national majority was rather quickly moving toward reading equal protection more generously, and the court moved to endorse that position.
Perceived animus toward LGBTQ citizens invalidated Colorado voters’ determination not to extend antidiscrimination protection to gays and lesbians, whose identities (or behaviors) they did not wish to affirm; there was no perception of animus in Michigan voters’ determination to bar state actors from taking race into account to enhance diversity. Yet, placing racial minorities outside the normal political process (by making it impossible for underrepresented minorities to pursue greater access to employment and educational opportunities through the normal legislative process)Footnote 67 should arguably be more closely scrutinized than a voter decision such as Colorado Proposition 2, because a lower level of scrutiny (rational basis plus) was applied when the issue was sexual orientation. The result is somewhat ironic, given that strict scrutiny for classifications based upon race is now the norm. Today, “affirmative action” and “preferences” are dog whistles about race, especially in the context of Michigan’s earlier experience with Grutter and Gratz. Justice Kennedy saw race-neutral language in the amendment.
John Hart Ely himself would probably have concluded that if white voters chose not to disadvantage themselves in pursuit of greater racial equality once barriers to African American voting fell, they were not required to do so. But when the court endorsed what Michigan voters chose, they stepped away from claiming constitutional grounds for measures supporting diversity or racial inclusiveness.
Racial discrimination may currently be more subtle than sexual orientation discrimination, but it is engrained in the social fabric.Footnote 68 The difference between categorization based on sexual identity or orientation (Romer, 1996; Obergefell, 2015) and racial categorization (Schuette, 2014; Fisher v. University of Texas I, 2013) may be that the court is much better at seeing – and addressing – overt discrimination than more subtle forms. This may help explain what Russell Robinson (2016, 153) terms “LGBT exceptionalism” within the court – advantages LGBT people enjoy relative to other contemporary civil rights constituencies. The Bostock decision (2020), in which the court extended Civil Rights Act Title VII protections (“because of … sex”) to those experiencing employment discrimination because of sexual orientation is a case in point.
Just as the Rucho majority declined to intervene in extreme cases of motivated partisan gerrymandering, and the Brnovich majority declined to invalidate measures passed by Arizona’s legislature that would make it harder for some citizens (including Native Americans, rural Latinos, and blacks) to cast ballots, the court passed the baton to Michigan voters in Schuette, saying: “The holding in the instant case is simply that the courts may not disempower the voters from choosing which path to follow” (Schuette, 2014, 314). Is it best that the court leave these kinds of matters to the democratic process in the several states?
Since Michigan voters were allowed to determine what equal protection meant or required, this did allow citizens to deliberate and own the meaning of the constitution in an area where affirmative action has not been mandated. They were constraining, by constitutional amendment, state actors. However, the court is most certainly not making clear which classifications it is willing to turn over to the elective branches, which it is not, and why. The constitution presumably took some contentious issues off the table, removing them from resolution through the democratic process.Footnote 69
Conclusion
Systems of robust legislative supremacy, even with a written constitutional framework, rely on internalized norms for boundary maintenance – either self-policed by legislators or vigilantly policed by an attentive citizenry. Notable examples since the early twentieth century have shown that such systems run risks of fascism, authoritarianism, or totalitarianism.Footnote 70 American judicial review, within a written constitutional framework, is one significant structural impediment to full-throated majoritarian rule. The results are imperfect.
It is hardly certain that robust constitutional courts could serve as bulwarks against radical popular sovereignty on the left or the right, against authoritarianism or rising ethnic and racial animus, or even against strong-willed majorities. However, there remains considerable support for a US Supreme Court that thwarts some majorities some of the time. Balancing the claims of national majorities, state majorities, and even occasionally making borderline antimajoritarian decisions, the court could probably maintain legitimacy while doing a better job that it has of late in expanding the scope for participation in the electoral process, enhancing faith in the fairness of elections, reinforcing representation, extending principles of equal protection, and thereby advancing a more inclusive version of the “people.” Perhaps a “no backsliding” principle for rights, equal protection, and access to the democratic process would be consistent with a commitment to enhancing popular sovereignty, although by defending a particular vision of popular sovereignty, it would discourage some deliberation.
Article III courts could be more vigilant in promoting the health of democracy as process, while limiting the harm majorities impose on minorities. These are roles courts can perform reasonably well – probably better than other branches. Mariah Zeisberg urges us to think about “distinctive governance capacities” of the branches. Though writing in the context of war powers, her point is applicable here: “the Constitution fails to provide for one authoritative institution to settle” many controversies.Footnote 71 We can evaluate branches’ competing claims of authority “in terms of how well they bring their special institutional capacities to bear on the problem of interpreting the Constitution’s substantive standards …”Footnote 72 They “exercise distinctive capacities that predictably generate distinctive perspectives on both policy and constitutional meaning.”Footnote 73
This perspective seems akin to that of Stephen Breyer (2010): Judges need to consider comparative institutional expertise and specialization when they think pragmatically about the constitution. For Breyer, a workable constitution is one that allows problems to be solved in a way that the public, and other governmental institutions, can find acceptable. Constitutional Courts have an important role to play in persuading other participants in the political system that their comparative institutional expertise and specialization includes promoting the health of the democratic process, limiting the harm majorities sometimes impose on minorities, and by doing so, preserving the health of popular sovereignty.
Open access