Polarization, Victimization, and Judicial Review
Table of Contents Show
Once upon a time, not that long ago, we had a pretty good explanation for why judicial review exists. It is the account generally associated with the Carolene Products footnote[1]—not necessarily the specific language of the footnote, but the understanding of judicial review reflected in the footnote. The premise is that, for the most part, important decisions in a nation like ours should be made by politically accountable officials, not by courts. Unavoidably, though, there will be defects in the democratic political process. The role of the courts is to correct those defects, to the extent they can. One such defect is that some groups may not have their fair share of political power. If that is true, then the courts should intervene to protect that group.[2]
There are many questions one might ask about that account. But before we even start with those questions: Isn’t the Carolene Products approach anachronistic? Maybe that approach to judicial review characterized the work of the Warren Court from the early 1950s to the late 1960s.[3] Brown v. Board of Education,[4] of course, is the centerpiece. No other institutions were protecting Black people in the Jim Crow South, so the Supreme Court did. Then when Congress responded by enacting the Civil Rights Act of 1964 and the Voting Rights Act of 1965, the Court upheld those statutes and, by expanding Congress’s power, opened the door to more legislation along the same lines.[5] That is the Carolene Products model for courts: intervene to protect a group that is not getting fair treatment in the political process, but step back when the political process is working.
But even if the Warren Court was a Carolene Products Court, the Warren Court is long gone. For the last half-century, the Supreme Court has shown no interest in the Carolene Products footnote. So whatever one might say about Brown and the Warren Court, the Carolene Products approach, it seems, no longer has much to do with judicial review.[6]
My claim is that this account is not quite right, and it is incorrect in a way that suggests a deep problem with judicial review today. Carolene Products, the decision, is no longer cited. But the Carolene Products approach has not disappeared. Instead, it has, in a sense, mutated. The view of judicial review associated with Carolene Products—specifically, the idea that courts are there to protect groups that are not being treated fairly in politics—has actually taken hold across the political spectrum. Even though no one cites Carolene Products, we are, in important ways, still living in a Carolene Products world.
The problem is that the Carolene Products approach to judicial review has degenerated into a competition among groups for victim status, playing out in court. To put the point crudely: Other groups want to claim that they are today’s version of Black people in the Jim Crow South—they are not getting fair treatment in politics—so the courts should protect them. The claims of these groups make up a significant part of what is going on today in constitutional law. Nonminority “victims” of affirmative action overtly align themselves with the Civil Rights movement. Some religious groups, again often explicitly, make similar claims of being victims of discrimination. Critically, these claims are made not just by obviously discriminated-against religious minorities, who do have a good claim to be the kinds of groups that courts should protect, but by religious groups that seem to have more than enough political power to fend for themselves without help from the courts. Advocates for gun rights sometimes use similar rhetoric, although they more often appeal, vaguely and implicitly, to the idea that they are victims of disrespectful treatment by elites. And, finally, a more complicated example—one that involves a group that does have a legitimate claim that their situation resembles, closely enough, that of the paradigmatic case of Black people in the Jim Crow South—is the movement for LGBTQ+ rights.
There is a trope in political theory that good forms of government tend to degenerate into deviant forms: Monarchy becomes tyranny, aristocracy becomes oligarchy, a republic becomes mob rule.[7] So, as a rough analogy, the Carolene Products form of judicial review exemplified by Brown has, today, degenerated into a litigation-focused status competition—specifically, a competition for victim status. Status competitions tend to be, by their nature, zero-sum: One group’s gain in status comes at the expense of another group. Litigation reinforces that tendency: A group’s victory does not just bring whatever relief the group was seeking; it seems to vindicate the claim that the group has been wronged. It follows that others are not the only ones with legitimate claims. In fact, maybe others’ claims are not that legitimate after all, by comparison.
For those reasons, among others, a competition for victim status seems like an especially bad way to resolve constitutional issues. But to the extent that kind of competition characterizes constitutional law today, it is, ironically, the perverse byproduct of one of the great success stories of U.S. constitutional law: the rejection of the regime of state-enforced racial discrimination.
Why has this happened? To some extent, it is attributable to the theoretical problems of the Carolene Products approach itself. But despite those problems, that approach did work well enough for a while. Nearly everyone thinks the Court did the right thing in Brown, and the Carolene Products approach guided the Warren Court more generally, in its decisions about legislative apportionment, freedom of speech, freedom of religion, and criminal procedure—areas in which the work of the Warren Court, contrary to some of its critics, has stood up well over time.[8]
The deeper source of the problem is that the preconditions for the success of Carolene Products, in the era that gave us Brown and the Civil Rights Movement, have eroded or even disappeared. Those preconditions were a consensus in elite opinion about certain central issues and, related, a relative lack of partisan political polarization. Once the consensus fragmented, as both a cause and effect of polarization, the problems with the Carolene Products approach became impossible to avoid. But in the meantime, Carolene Products, by virtue of Brown and the Civil Rights Movement, had become a model for what courts should do. That opened the door to a competition for victim status.
The Carolene Products Disequilibrium
The idea that judicial review exists to correct the weaknesses of the democratic political process is a natural way of dealing with what is often called the counter-majoritarian difficulty: Why should a political system that is generally committed to popular rule give significant power to courts that are, relatively speaking, politically unaccountable? Even so, the strength of the Carolene Products idea is, I think, often underestimated—maybe because the first fully explicit statement of the idea was dictum in a footnote, maybe because the theoretical justification seems too pat, and maybe because of genuine difficulties in the theory itself. But, to this day, it is not clear that we have a better justification for judicial review.
For one thing, the Carolene Products theory has deep roots; it was influential long before the Carolene Products decision itself. There is a clear statement of the basic idea in McCulloch v. Maryland, the foundational Marshall Court case that helped shape the power of the federal government.[9] Then the reappearance of the idea in 1938—in the Carolene Products opinion—was not an accident. By 1938, the Court had completed its rejection of what is often called the Lochner era, the period in which federal courts invalidated many social welfare and regulatory laws. The lesson was that those kinds of questions of social and economic policy were for the politically accountable branches of the government, not for the courts.
But at that same time, it was becoming more and more difficult for the courts to ignore the injustice of Jim Crow segregation. The first clear step toward ending “separate but equal” school segregation, in the progression that culminated in Brown, was Missouri ex rel. Gaines v. Canada,[10] which was decided in the same year as Carolene Products. The Carolene Products footnote approvingly cited earlier cases that had held unconstitutional racial discrimination in voting. And it was not just race discrimination: The Carolene Products footnote also referred favorably to cases that tried to hold the line against discrimination directed at religious minorities and against proto-McCarthyite attacks on free expression.[11]
Carolene Products was brought about, at least in part, by the collision of these tectonic plates. Rejecting Lochner meant deferring to politics; but the threats to civil rights and civil liberties required the courts to stand up to elected officials. Carolene Products was a way of reconciling those imperatives. In that way, Carolene Products was the result (to paraphrase Justice Holmes) not only of logic —the logic that the courts were in a position to address the failings of the political process—but of hard-won experience. Whatever their abstract virtues, other accounts of what the courts should be doing—originalism, for example, in its many varieties, or more radical theories of judicial restraint—cannot make a comparable claim.
But there are problems with Carolene Products, and they came home to roost. The principle that courts should protect minority groups that do not have their fair share of political power presupposes some way of identifying who those minority groups are. The footnote describes the minority groups in question as those that are, in a famous (although possibly poorly chosen[12]) phrase, “discrete and insular.”[13] Those terms suggest that the question of which groups should be protected is essentially empirical, a matter of political science; some groups are easily identified (“discrete”) but also isolated, lacking enough power on their own and unable to form alliances with others. But of course the question cannot be whether a group, in fact, lacks political power. Unavoidably, some groups will not have the power to get what they want. The question is whether a group lacks the share of political power that it should have. That question has an inescapably normative component, and it can be very difficult to answer.
For one thing, it is not easy to measure political power or even to individuate groups. Political power may derive from numbers, but it does not always; it is a familiar insight from public choice theory that large groups can be out-organized, and therefore defeated politically, by smaller groups whose members are less numerous but have more at stake.[14] Or political power may derive from wealth, but again not always; it is not exactly unheard of for wealthy people to claim that they are unfairly victimized by popular majorities (and, in that sense, to claim that they are actually a discrete and insular minority). Or political power may come from other sources, like a group’s exceptional degree of cohesion or commitment to certain issues. A group may be powerful in certain localities but not nationally, or vice versa. Or the extent to which a group is able to exercise political power may depend on the specific issue. More fundamentally, some groups may deserve to be politically weak; people who have committed serious crimes may constitute a “discrete and insular” group that lacks political power, but they are not in the same category as historically discriminated-against racial or religious minorities.[15]
All of those are difficulties with the Carolene Products theory, and they are generally pretty well-known. Sometimes, though, theoretical problems like those do not have to be resolved in the abstract if there is a compelling concrete example. In the mid-twentieth century, there was an indisputable example of a minority group that was not treated fairly: African Americans in the Jim Crow South. Elite opinion in the North, especially, recognized that to a degree; several Supreme Court Justices were more or less explicit about it when they were deciding Brown.[16] The point is not that the Court in Brown was rigorously following the logic of Carolene Products. Many other factors were at work. But to the extent that members of the Supreme Court, and other people whose opinions were influential with the courts, were aware of the tension between the campaign against Jim Crow and the supposed lessons of the Lochner era—and of the general concern that the courts not usurp the role of elected politicians[17] —the Carolene Products logic helped resolve the tension, precisely because racial segregation was increasingly unacceptable.
One can imagine an alternative history of the succeeding decades in which the Brown era was a one-off—a unique chapter in the history of constitutional law with no important lessons for the future. The idea might be that the courts had to wrap up what the Civil War started but Reconstruction did not finish. Just as the Civil War required the federal government to take extraordinary and, in some instances, arguably unlawful measures, so, on this account, Brown was an example of the Supreme Court stepping outside its generally proper role to deal with a peculiarly intractable, historically rooted problem.[18]
But that is not how the Brown era has been treated. It is more nearly the opposite. In mainstream legal thinking, Brown has become not just an icon—as the Supreme Court’s finest hour—but a template. The celebratory account of Brown may overrate the significance or the virtues of what the Supreme Court did at that time. But it remains true that the mainstream view in the legal culture is that in Brown the Court did the kind of thing it is supposed to be doing. There are disputes over just what Brown did, but little disagreement about that premise.
A predictable consequence is that now other groups want their own version of Brown v. Board of Education. Other groups want to get the same kind of recognition that Brown gave to African Americans in the Jim Crow South—a recognition that they are not getting treated fairly in politics and society generally. The rewards of being recognized in that way are both tangible and symbolic. When nonminorities succeed in challenging affirmative action in higher education, they improve their chances of being admitted to the colleges and universities of their choice. Religious groups gain subsidies or are relieved of regulatory burdens. People can get the firearms they want. But those tangible benefits are only part of the story, and possibly not the most important part. Getting the benefit of special judicial protection from the otherwise normal operations of politics is a way that groups can assert that their claim to having been treated unfairly has been vindicated. And there is a competitive element: If you think you are being treated unfairly, the natural tendency is to look at other groups whose parallel claims have been vindicated, and to be aggrieved that yours have not been.
The customary argument for judicial restraint is that a group should not try to win in court what it lost in the legislature. But groups that litigate about constitutional issues sometimes want something that they cannot get from the legislature. Legislative victories—a piece of the proverbial pie, or favorable regulatory treatment—are not enough. They want the status that comes from being recognized as a victimized group, a group that should be protected by the courts irrespective of normal politics. Necessarily you cannot get that through the political process. By the same token, when the courts deny special protection to a group—when they tell a group that it will have to accept the outcome of the political process, even though they are not completely happy with it—they seem to be devaluing the group, denying its claim to victim status, and treating it (or allowing it to be treated) with less respect than other groups get. This is the way in which constitutional litigation becomes a competition for status.
In a sense, the problem is that Brown was too easy a case. That is an odd thing to say, considering how controversial it was at the time. But African Americans in the Jim Crow South were an exceptionally clear example of a group that could not get fair treatment in politics. That is why there was something like a consensus among the legal elite on the issue even at the time of Brown, and an overwhelming consensus in retrospect. But once the Court moved on from Jim Crow, it became clear that what Carolene Products actually requires is a controversial judgment about fairness. When a group has not gotten everything it wants from politics, does it have a legitimate reason to complain about its relative lack of political power, or has it just lost a political battle, fair and square? What the elite consensus about the treatment of African Americans obscured is that, in other contexts, there will routinely be deep disagreements about that question. And because that question can involve matters that are especially important to people, like religion and deeply held political views (both of which were mentioned in the Carolene Products footnote) and, today, sexuality—matters that are, one might loosely say, associated with people’s identities—those disagreements are more likely to be especially intense, and to break down along political or cultural lines.
The Carolene Products approach not only understates the potentially controversial nature of the question; it also does not take into account how the stakes become greater when that question is the focus of litigation. The implicit theory of Carolene Products is pluralist in the mid-twentieth-century sense. Politics is a matter of competition among groups for concrete benefits. The groups will form alliances and end up, in all probability, with a compromise. Each will get a part, but not all, of what it wanted. All of that is not the courts’ business. In fact, one plausible account of the rejection of Lochner in the New Deal era is that Lochner-era courts had tried to keep politics free of the influence of interest groups, but later courts came to recognize that politics is inevitably an interest group competition.[19] The central idea of Carolene Products is that the Supreme Court’s role is to let that competition run its course, so long as no groups are wrongly excluded.
But when the Supreme Court says that a group has a constitutional right to something—that it does not have to accept the outcome of the political process—the Court does not just correct an error. It also confers a status on that group. This may partly be a result of the aura that now surrounds the Court’s decisions from the Civil Rights era. Once Brown v. Board of Education (and the Civil Rights era more generally) became such a visible icon, it was probably inevitable that other groups would want to say that their treatment is not so different from African Americans’. They, too, in their own eyes, are the victims of unfair treatment of the kind that can be rectified only by a Brown-like declaration from the Supreme Court.
To the extent this is true, Carolene Products has produced not a stable role for the Court but a disequilibrium. The more some groups seem to be winning, the more others will feel they have been treated unfairly, even disrespectfully, and the more intensely they will try to control the courts. This makes the courts not just a forum for relitigating political decisions—what they were not supposed to be, post-Lochner—but, worse, a forum for a particularly unproductive kind of politics: a zero-sum competition over status in which once some groups succeed, others will feel they have been devalued and will want to secure recognition for themselves.
Partisan Polarization
The current state of partisan polarization makes it even less likely that the Carolene Products approach will succeed.[20] In the mid-twentieth century—when Congress’s ability to enact the Civil Rights and Voting Rights Acts made Carolene Products seem like a more-than-workable approach—the views of Democratic and Republican members of Congress overlapped far more than they do today.[21] Partisan polarization in more recent decades has undermined the roles that Carolene Products assigns to both legislatures and the courts.
A premise of Carolene Products is that Congress will enact legislation addressing pressing and controversial issues. When Carolene Products was decided (in the wake of a productive period of New Deal legislation) and when it most closely corresponded to what the Supreme Court was doing (when the Warren Court, having decided Brown, emphatically upheld civil rights legislation that Congress had finally enacted)—that premise seemed entirely realistic. It does not seem to be anymore, at least in part because of partisan polarization. Partisan polarization in the U.S. system, among other things, makes it more difficult for Congress to enact legislation.[22]
The polarization is not confined to Congress. Unsurprisingly, polarized parties produce polarized judicial nominees and a polarized judiciary[23]—undermining the elite consensus, presupposed by Carolene Products, about the minority groups that most need protection from the courts. In large part because of partisan polarization, judicial nominations have become more controversial in recent decades.[24] That further enmeshes judges and justices in partisan politics: Nominees will see the President’s party as their champions and the opposing party as their opponents. Of course there is evidence of the partisan polarization of the judiciary in the news routinely: Judges and justices are identified with the President who appointed them, because that provides information about their likely views on issues, and justices, in particular, are routinely identified as “liberal” or “conservative” appointees of, respectively, Democratic and Republican Presidents. Neither the Court that decided Brown nor the later Warren Court—Carolene Products Courts—was like that. Chief Justice Earl Warren himself had been the Republican candidate for Vice President in 1948; Justice William Brennan, a central figure in the later Warren Court, was appointed by President Eisenhower, a Republican, although Justice Brennan himself was a Democrat; Justice Byron White, one of the more frequent dissenters from Warren Court decisions, was a Democrat, appointed by Democratic President John Kennedy.
The polarization of the judiciary extends further, to social and political networks. Judges and justices belong to, and respond to, groups that have defined political and ideological dispositions.[25] It is possible that the consensus that underlay Brown, on the need for the courts to do something about racial segregation, was historically unique, given the role of race in U.S. history. But even if it was not, the polarization that has spread to the courts makes it unlikely that there will emerge a consensus comparable to the one that existed from around the time of Carolene Products through the Warren Court.
The Status Competition
In these ways, political polarization exposed the inherent instability of Carolene Products. Carolene Products presupposed a relatively well-functioning political process. The later Warren Court years, when Congress enacted the Civil Rights and Voting Rights Acts, conformed to that model. After that, though, partisan polarization changed things. And Carolene Products also presupposed a consensus on the question of which groups were not treated fairly in the political process, because that was a more divisive question than the Carolene Products reference to “discrete and insular” groups acknowledged. Since the Brown era, for many reasons—partisan polarization among them—a consensus of that kind has not developed in any important way, at least about any group other than African Americans in the Jim Crow South.
Those developments set the stage for a competition for victim status. The iconic status of Brown made it a model for groups claiming that they were treated unfairly. The fragmentation of legal elites—and their responsiveness to different social and political networks—made it possible for groups to find a receptive audience among some subset of judges and justices. And the way constitutional law developed in the areas I mentioned earlier—affirmative action, religious freedom, gun rights, and, in a qualified way, the rights of LGBTQ+ individuals—suggests that a competition for victim status is what in fact resulted. Several aspects of those developments, in particular, are markers of that kind of competition.
First, the rhetoric accompanying challenges to government action in these areas often made claims of victimhood—of being treated unfairly or disrespectfully—sometimes with explicit comparisons to the Civil Rights Movement. Second, and related, the symbolic stakes of the decisions seem inordinately prominent, suggesting that the issues concerned status at least as much as more concrete concerns. Third, to put the point colloquially, if the Court, in these areas, is not recognizing claims of victim status along the lines of the mutated version of Carolene Products, what exactly is the Court doing?
To be specific, critics of the Warren Court, for example, called for more judicial restraint, in the form of deference to the political branches of government; for greater respect for federalism, so that constitutional principles would accommodate differences in local circumstances; and for adherence to precedent. In the areas I mentioned—affirmative action, the claims of religious groups, gun rights, and LGBTQ+ rights—the Supreme Court has done none of those things. Instead it has invoked, implicitly or explicitly, claims about victimization. The result of all this is that the doctrines the courts have devised in these areas have significant weaknesses that are the result of the Court’s implicit view of the groups in question as victims entitled to special judicial protection from politics.
Affirmative Action
There is a straightforward argument in defense of the constitutionality of affirmative action that, under the original version of Carolene Products, would have carried the day. The argument is that whatever the merits of affirmative action as a matter of policy, there is no reason to remove the issue from ordinary politics. Affirmative action is discrimination in favor of a minority group. Why can’t the majority protect itself through the normal political processes, without courts getting involved? A leading defender of the Carolene Products approach to judicial review made just this argument at the time of the Court’s earliest affirmative action decisions.[26]
The Supreme Court did not accept this argument, of course. Its response—and, in general, the response of opponents of affirmative action—has been to claim, in effect and sometimes literally, that nonminorities who lose out because of affirmative action are victims of essentially the same kind of discrimination that Brown condemned: that there is no substantial difference between affirmative action measures that favor minority groups and discrimination against historically disadvantaged minorities. Students for Fair Admissions, Inc. v. President and Fellows of Harvard College (SFFA)[27]—the Supreme Court’s most recent affirmative action decision, and the one that is most unequivocally hostile to affirmative action—illustrates the point.[28]
The Court in SFFA treated Brown as the crucial decision.[29] And Brown, the Court said, dictated that the “time for making distinctions based on race had passed.”[30] The problem with this characterization of Brown—and this is hardly a new insight—is that Brown dealt with a specific kind of “distinction[] based on race”: one that was used to subordinate African Americans. That was not a coincidental feature of Brown, and it is not something that is just attributed to Brown as an after-the-fact way of making Brown consistent with affirmative action. It is integral to Brown that it involved not just any “distinction based on race” but that specific kind of distinction based on race.
The controversy that surrounded Brown makes this clear; in a sense that controversy was about this very point. The defenders of Jim Crow segregation did not, of course, deny that segregation made distinctions on the basis of race. Their claim was that distinctions based on race could be compatible with equality. Plessy v. Ferguson,[31] which upheld “separate but equal” segregation, asserted, in a notorious passage, that if “enforced separation stamps the colored race with a badge of inferiority,” that was only because African Americans “choose to put that construction upon it.”[32]
The central issue in Brown was how to rebut that claim—that there was nothing inherently unequal about the distinctions based on race that were characteristic of Jim Crow. Today it might be hard to take Plessy’s defense of separate but equal seriously. But at the time, even some dedicated opponents of racial segregation were not convinced that the Brown Court had successfully responded to the Plessy opinion.[33] That is why the controversy surrounding Brown was about when distinctions based on race precipitated an unacceptable form of racial inequality.
The answer, of course, is that racial segregation of the kind that was outlawed by Brown did, in fact, “stamp . . . a badge of inferiority” on a group; it was part of a system of racial subordination.[34] Some answer of that kind had to be provided in order to justify Brown. It was no response to Plessy just to assert—as SFFA does—that “distinctions based on race” are incompatible with equality. The question is why, and the answer is that those particular distinctions based on race, the ones that were characteristic of the system of segregation that existed in parts of the United States, treated African Americans as inferior.
SFFA—by claiming that the problem was just with “distinctions based on race”—elided all of that. To be clear, there might be reasons to say that all distinctions based on race are unacceptable. Perhaps there is an argument that those distinctions are in some sense divisive or otherwise harmful to social cohesion, to a degree or in a way that makes them unconstitutional. But that is not obvious, and SFFA did not try to make such an argument; it just invoked Brown. The SFFA majority enlisted Brown as a basis for placing laws that disadvantage nonminorities in the same category as Jim Crow segregation—in other words, that nonminority “victims” of affirmative action are, in relevant respects, like the plaintiffs in Brown. SFFA accepted that superficial account of victimization without acknowledging that it is inconsistent with the principal justification of Brown itself.
There are other ways in which the litigation against affirmative action illustrates competitive claims of victimization. The treatment of Asian American applicants in the SFFA majority opinion, and elsewhere,[35] is an example. The Court in SFFA emphasized that the affirmative action measures that Harvard College used in its admissions decisions had reduced the number of Asian Americans who were admitted. “Harvard’s consideration of race has led to an 11.1% decrease in the number of Asian-Americans admitted to Harvard.”[36] But if Harvard’s policy was otherwise lawful, the fact that it had this effect on Asian American applicants does not matter. That follows from the holding of Washington v. Davis:[37] A government action that disadvantages a minority group is not unconstitutional unless it was adopted with discriminatory intent.
There were, in fact, allegations in SFFA that Harvard had deliberately discriminated against Asian Americans. If those allegations were true, then Harvard had acted illegally, not because it engaged in affirmative action, but because it discriminated against a minority group in a way that contravened a principle established long ago.[38] But intentional discrimination against Asian Americans was not the basis of the SFFA decision—for one thing, the district court found that Harvard had not engaged in such discrimination[39]—and the effect of an otherwise lawful policy on a group does not, by itself, establish a constitutional violation. Despite that, the effects of affirmative action measures on Asian Americans played a significant role in the litigation.
In challenging affirmative action, the advocates’ reflex, and the Court’s, was to identify groups that it could portray as victims—nonminorities who were denied the benefits of affirmative action, or minorities, like Asian Americans, who were adversely affected—and to rely on the alleged parallel to Brown. But Brown was crucially different, and under established principles the effects on other minority groups should not have mattered to the question the Court was addressing.
Religion
The Carolene Products footnote itself approvingly cited cases in which courts protected religious groups, and, in principle, the Carolene Products approach should apply in a relatively straightforward way to claims raised by religious groups. Prejudice against some religious groups—even, at times, the persecution of religious minorities—is far from unknown in U.S. history. In this respect, the treatment of religious groups presents issues that are unlike those concerning affirmative action. There is an argument that questions about the constitutionality of affirmative action should simply be left to politics,[40] and that the courts’ involvement in affirmative action cases reflects the degeneration of Carolene Products into competitive victimization claims. The same is not true of the claims of religious groups: Under core Carolene Products logic, there is an important role for courts.
In fact, though, the law in this area has evolved in a way that almost inverted Carolene Products. Courts have protected religious minorities only to some degree; in some instances, they have conspicuously failed to do so.[41] In the meantime, mainstream religious groups that plausibly can look after their own interests in the political process have fared well in court. The central Carolene Products idea is that when the political process is functioning well, the courts should stay out of the way; the Warren Court was conspicuously deferential in its review of civil rights statutes. But when Congress enacted a statute designed to protect religious free exercise, the Religious Freedom Restoration Act, the Court sharply narrowed Warren Court precedents and held the statute unconstitutional as applied to state governments.[42] This evolution revealed both the internal problems with the Carolene Products approach—including the difficulty, in the absence of an elite consensus, of identifying the groups that should be protected—and, again, that Carolene Products is vulnerable to competing claims of victimization.
The Warren Court, in the Carolene Products heyday, established a principle that was designed to protect religious minorities. Sherbert v. Verner[43] held that a state violated the Free Exercise Clause of the First Amendment (as incorporated in the Fourteenth Amendment) when it denied unemployment compensation to a Seventh Day Adventist who was fired because she would not work on Saturday, her church’s Sabbath. The Court ruled the state’s action had burdened the free exercise of religion by treating the employee’s observance of her religious obligations as the equivalent of a voluntary refusal to work that made her ineligible for unemployment benefits.[44] And a burden on religious exercise, the Court held, was unconstitutional unless it was justified by a compelling state interest.[45] Sherbert v. Verner seems to conform to the Carolene Products model: The employment compensation system, the product of ordinary political and bureaucratic processes, had a blind spot about a minority religious group with nonconforming views. The Court protected that group by requiring that its interests be accommodated, as long as that could be done at reasonable cost.
The inversion of the Carolene Products approach took hold after the Warren Court era. A recent statistical study of cases involving the claims of religious groups—under the Free Exercise Clause, the Establishment Clause, and the Religious Freedom Restoration Act—concluded that the claims of religious groups prevailed more often in the Burger and Rehnquist Courts than they had in the Warren Court, and prevailed much more often in the Roberts Court.[46] But—the important point for Carolene Products purposes—the principal beneficiaries of this shift were mainstream Christian groups, not minority religions like Seventh Day Adventists.[47] In fact, one of the relatively few losses for a religious group in the Roberts Court was Trump v. Hawaii,[48] which rejected a claim that a ban on travel to the United States discriminated against Muslims—a group that certainly qualifies as a Carolene Products minority, especially in the context of immigration after the September 11 attacks.
This shift corresponded to victimization claims that are reflected in both rhetoric and doctrine. The rhetoric cannot be discounted, because it came from, among others, Supreme Court Justices.[49] Both the rhetoric and the doctrine treat mainstream religions, or religion itself, as being threatened in a way that requires special protection: Members of mainstream religions, or religious believers generally, should be treated as if they are a “discrete and insular” Carolene Products minority.
The rhetorical claims are fairly explicit.[50] The role of victimization claims in the doctrine is less obvious but nonetheless central. There are at least two important examples. One is the Supreme Court’s apparent decision to adopt what is usually called the “most favored nation” principle for determining when a government regulation infringes the free exercise of religion. The Court did so in a Covid-era decision, Tandon v. Newsom.[51] California residents who wanted to meet in a private home to engage in Bible study challenged a state regulation that limited such gatherings to three households. They argued that larger numbers of people were allowed to gather for other purposes—for example, on city buses. Tandon held that the regulation was unconstitutional:
[G]overnment regulations . . . trigger strict scrutiny under the Free Exercise Clause, whenever they treat any comparable secular activity more favorably than religious exercise. It is no answer that a State treats some comparable secular businesses or other activities as poorly as or even less favorably than the religious exercise at issue.[52]
There are obvious problems in identifying what secular activities are “comparable” to religious exercise.[53] To some extent those problems will inhere in any principle that protects religious exercise against discrimination, because religious claims by their nature are generally not comparable to secular claims. If the issue were discrimination against a particular minority religion, courts or other government institutions would have to deal with the problem of comparability as best they could. But the logic of the “most favored nation” test is that the problem is not discrimination against a disfavored group or creed but discrimination against religion in general—that religious believers, as a class, are comparable to a Carolene Products “discrete and insular” minority.
The second example concerns the regulation of speech. Under the Free Speech Clause of the First Amendment, certain kinds of government property—so-called “limited public forums”—may be opened to some forms of speech but not others.[54] The established rule is that the government may exclude speech with a certain content, as long as the exclusion is based on the subject matter of the speech but not on the viewpoint expressed. So, for example, the government may allow commercial advertising in public transportation but not political advertising;[55] and it may allow speeches on military bases but exclude political speeches.[56] But it cannot, for example, permit some political views but not others.
In a series of cases, the Supreme Court has ruled that, in public property of this kind, the exclusion of all religious speech constitutes viewpoint discrimination and therefore violates the First Amendment.[57] Religious expression—not speech expressing a particular creed or set of religious beliefs, but religious speech in general—is treated as speech that expresses a distinct viewpoint. One might have thought that, by analogy to political speech, religious speech as a whole could be excluded, as long as there was no discrimination between or among religious sects or religious views. The Court’s refusal to accept the analogy—its conclusion that religious speech per se embodies a point of view—is of a piece with the idea that religious believers as a class, whatever the specific content of their beliefs, are a Carolene Products minority that is victimized by discrimination. For much of human history, including today, people, all of them deeply committed religious believers, have nevertheless treated each other as mortal enemies because they held different religious views. Those warring religious believers might be surprised to learn that, according to the Supreme Court, they all share a single point of view—the “religious” point of view.
Having said that, though, there is a certain logic to the Court’s idea that there is a cleavage between, on the one hand, views based in religious belief, whatever the particular commitments of the adherents, and, on the other hand, a nonreligious viewpoint. In a society in which much of the culture is secular and many institutions are nonreligious—the government, public schools, many private sector enterprises—it is understandable that people with strong religious beliefs would feel a kinship despite having theological differences. This illustrates both the instability of the Carolene Products idea and its vulnerability to competing claims of victimization. On the one hand, it seems implausible to say that in a society in which the great majority of individuals not only hold religious beliefs but consider religion important,[58] religious belief, in itself, marks one as a member of a “discrete and insular” minority that requires special protection from the courts. On the other hand, it is not difficult to understand why strong secular currents in society might cause religious believers, as a class, to think of themselves as “misunderstood, dismissed as unreflective and, even worse, scorned by much of society today”[59]—that is, the kind of minority that, according to Carolene Products, should be protected.
Gun Rights
In 2008, in District of Columbia v. Heller,[60] the Supreme Court held for the first time that the Second Amendment gives individuals the right to possess firearms. McDonald v. City of Chicago[61] incorporated that right in the Due Process Clause of the Fourteenth Amendment, so that it applied to the states. The decisions—like the decisions concerning affirmative action, religion, and LGBTQ+ rights—were the products of a sustained litigation campaign, in this instance by gun rights advocates.[62] And since those decisions, the Court has defined and expanded Second Amendment rights, although it has also qualified those rights in some respects.[63]
The victimization claims may be less obvious in these cases than in the others, but they are part of the story. Issues about firearms regulation have become components of what are often called culture wars, and they divide people along pre-existing lines.[64] People who live in rural areas, for example, are more likely to oppose gun control than people who live in cities. But beyond that, a significant (although contested) body of social science research suggests that individuals’ attitudes toward gun control are associated with deep-seated cultural views, in ways that suggest that opponents of gun control believe that the regulation of firearms treats them disrespectfully.[65] For example, according to this view, opposition to gun control is fostered by a belief that it would be “cowardly and dishonorable—a gesture of individual impotence—for society to disarm citizens for their own protection.”[66]
The campaign for gun rights has other things in common with victimization claims. As with affirmative action, there was no obvious reason why the matter could not have been left to the political process. Gun control has been a prominent political issue for years. It is difficult to identify any way in which the political process has been distorted (at least against the interests of groups favoring gun rights) or any groups that have been systematically excluded. In addition, this is a question that seems exceptionally well-suited to being settled at the state and local level. Firearms obviously present different issues in urban and rural areas. Even a minimal commitment to federalism, one might have thought, would have caused the Court to stay on the sidelines.
Beyond that, the doctrine seems difficult to rationalize, an indication that its development is the result not of a principled view of the interests that should be protected but instead of a general sense of grievance. The measures that the Court invalidated in Heller and McDonald were both unusually restrictive laws adopted by major U.S. cities.[67] In areas of the country where gun ownership is common and regulation is unpopular, the laws were nothing like those involved in Heller or McDonald, and, as a matter of politics, there was no chance that they would be. In addition, Heller itself explicitly said that a wide range of regulations would remain constitutional.[68] So for more than a decade after Heller was decided, it was not clear that gun rights advocates had won much more than a symbolic victory.[69] Of course, if you are just trying to vindicate a claim that you are a victim of laws that treat your cultural views disrespectfully, then a symbolic victory is very much a victory.
The Court returned to the subject in a very significant way—one that was not at all symbolic—in 2022, in New York State Rifle & Pistol Ass’n v. Bruen.[70] Bruen invalidated a New York regulation that gave the government discretion to decide whether to issue a license to carry a concealed handgun in public. The Supreme Court rejected an approach that had crystallized in the lower courts: Those courts first decided if the challenged measure infringed on Second Amendment rights at all. If it did not, it was, of course, constitutional. If it did, the question would become whether the government’s interest was sufficient to overcome what Heller had identified as the core purpose of the right recognized in Heller, self-defense in the home. That approach is similar to the one that is used to enforce other constitutional rights.[71]
Bruen rejected that approach: It was not enough that a regulation “promotes an important interest.”[72] Instead, in order to sustain a regulation of conduct that is “cover[ed]” by “the Second Amendment’s plain text,” the government has to show that the regulation is “consistent with the Nation’s historical tradition of firearm regulation.”[73] That meant deciding whether the challenged regulation is “relevantly similar” to one found in history. The Court said that it would not provide an “exhaustive survey” of how to make that determination, but that it involved “two metrics: how and why the regulations burden a law-abiding citizen’s right to armed self-defense.”[74]
There are obvious problems with this approach.[75] For example, there is the question of how prevalent the “similar” historical regulation must be: How many jurisdictions, and which jurisdictions, must have had such a regulation in order for it to count? More fundamentally, though, to say that the historical regulation must be “similar” begs any number of questions. What historical regulation is relevantly similar to a prohibition on firearms on, for example, commercial airplanes? Is regulation of the firearms made by artisans similar to the regulation of mass-produced firearms that are shipped interstate? At what level of generality should the “how” and “why” be characterized? And so on. On these points, the Bruen opinion, it seems fair to say, resorted to unhelpful metaphors.[76]
The complexity and indeterminacy of the approach that Bruen adopted—and the Court’s rejection of the more conventional approach of weighing the government’s interests against the individual’s—raises the question of just what was animating the Court’s approach to this area. The Court announced its approach based on historical analogies without explaining why that was the correct way to define the Second Amendment right. Heller, the foundational case, purported to be originalist: to be determining the original public meaning of the Second Amendment. But Bruen left originalism behind—for good reason, in fact; the world has changed in so many ways since the Second Amendment (or the Fourteenth Amendment) was adopted that originalism would be at least as indeterminate as the tradition-centered approach that the Court adopted.
One possible account of Bruen—speculative, but it is not clear that there is a better one—is that, to invoke the cliché, the Court was a dog that caught the car and did not know what to do next. The claims of gun rights advocates, that they were victims in the culture wars, resonated with some of the Justices—a result of the fragmentation and polarization of political and legal elites. The text of the Second Amendment provided a vehicle. Heller’s recognition of a constitutional right to possess a weapon provided a symbolic victory that acknowledged the victimization. But because that right rested on that sense of victimization, rather than on a more secure foundation, there was no coherent way to proceed.
Marriage Equality
LGBTQ+ people—particularly transgender individuals—are, perhaps, as good a candidate as we have today for a group whose situation resembles that of African Americans in the Jim Crow South. LGBTQ+ people have been subject to discrimination and prejudice throughout history. Until relatively recently they were, in many places, not welcomed as allies in interest groups in politics—a key element of the Carolene Products “discrete and insular” formation. But things changed relatively quickly in the last decade or so. Discrimination against gay and lesbian individuals, at least, became much less acceptable. Laws precluding same-sex marriage seemed like the most difficult form of anti-gay legislation to overcome, but well before Obergefell, same-sex marriage had significant popular support—more support than interracial marriage did at the time of Loving v. Virginia.[77]
So on the logic of Carolene Products, the Court might have decided to stay out of the picture and allow the political process to determine whether (and where) there would be a right to same-sex marriage. The principal dissent in Obergefell urged precisely that course. The dissent noted that eleven states had recognized same-sex marriage through their political processes and five state supreme courts had done the same. Moreover, the dissent suggested, plausibly, “the democratic ‘momentum’” favored proponents of same-sex marriage.[78]
Against this background, Obergefell may be the best evidence for the proposition that members of the Court, and judges more generally, now see judicial review as a way of protecting groups that are being unfairly treated even if those groups could get most of what they wanted through politics. The opinion of the Court in Obergefell explicitly rejected the dissent’s argument that same-sex marriage should be left to the political process and, notably, explicitly invoked the risk of dignitary interests to the status of gay individuals as a reason for rejecting it. “Of course,” the Court acknowledged, “the Constitution contemplates that democracy is the appropriate process for change”; but the Court immediately added “so long as that process does not abridge fundamental rights.”[79] For that reason, the Court said, “It is of no moment whether advocates of same-sex marriage now enjoy or lack momentum in the democratic process.”[80] And the Court explicitly spoke of the “[d]ignitary wounds” that “[a] ruling against same-sex couples” would inflict.[81]
The opinion is ambiguous about whether the dignitary harm it was referring to would be inflicted by the inability to marry or by an adverse decision itself. So while it is clear that Obergefell rejected the Carolene Products logic—the Court was going to decide for itself whether there was a right to same-sex marriage, irrespective of how well the political process was functioning—it is less clear that the majority was attuned to claims of victimization and the resulting symbolic importance that the decision itself would have. But only a little less clear. In its rhetoric the opinion announces itself as something more than just a legal conclusion about same-sex marriage; the majority pretty clearly considered LGBTQ+ people to be the kind of minority group that it was the Court’s responsibility to protect.
In any event, the jubilant response to the actual decision suggests that the Court understood the significance of its decision. If the Court had refused to establish a right to same-sex marriage, that would have been viewed, by supporters of same-sex marriage, not as a neutral deferral to the political process but as a rejection of their claim that they were a group that had been subject to the kind of unfair treatment that, in a different context, Brown had condemned. That gave the Obergefell majority a reason to dismiss out of hand the Carolene Products principle that ordinarily politics, not the courts, should determine whether a fundamental right exists.
The upshot, though, is that the Court is in the business of refereeing, and thereby encouraging, the competition for victim status. The dissents in Obergefell backhandedly made this point: They relentlessly pressed the question why the logic of that decision does not extend to polyamorous relationships. In United States v. Skrmetti,[82] the Court upheld a Tennessee law restricting sex transition treatments for minors that was challenged on the ground, among others, that it unconstitutionally discriminated against transgender individuals. But the Court asserted that the case did “not raise th[e] question” whether “transgender individuals are a suspect or quasi-suspect class” because the statute “does not classify on the basis of transgender status” but only “removes one set of diagnoses—gender dysphoria, gender identity disorder, and gender incongruence—from the range of treatable conditions.”[83] The Court then agreed to hear two cases that squarely present the question whether states may discriminate against transgender individuals.[84]
That question will, once again, identify the fault line in Carolene Products: In the absence of a consensus, the courts will have to decide which groups are being treated unfairly in the political process. And the effects of its decisions will extend beyond that Carolene Products-inspired question. After Obergefell, if the Court declines to protect transgender individuals, or, for that matter, people in polyamorous relationships, that will be seen—by the individuals in those groups and by others—as a comment on their status in society.
Is Victim Status Competition Unavoidable?
If a competition for victim status is, in fact, a significant feature of what the Supreme Court does today, is there anything that can be done about that? It is hard to argue that this is a good situation. Of course one might applaud certain decisions. For many of us, the world is a better place because of Obergefell (and would have been a better place if the Court had reached a different result in Skrmetti). Others might applaud the other decisions on this list, or wish the Court had gone further, for comparable reasons: some gun owners, people with deep religious commitments that they perceive to be at odds with a governing secular ethos in society, and nonminorities who feel aggrieved about affirmative action. People in those categories may not be familiar with the details of Supreme Court decisions, but the news that the Court has vindicated the rights of people like them might make them feel more fully recognized in society. It seems unlikely, though, that anyone would assign this kind of role to a court as a matter of first principles. The role envisioned by Carolene Products—of paying attention to groups that are genuinely excluded—is a sensible role for a court to play. But a competition for victim status among groups that are not excluded, and that can get a reasonable share of what they seek through politics, does not seem like something you would want at all, much less want in court.
Restraint and Passive Virtues
One way to avoid status competition would be for the Court to revert to the anti-Lochner horn of the dilemma that Carolene Products tried to avoid and become more deferential across the board. Brown and the civil rights era cases might, as I suggested earlier, be seen as a unique episode. This seems to be only a theoretical option, though. The Court has shown no tendency to move in this direction. Since Justice Frankfurter left the Court in 1962, no justice has even nominally embraced that approach (as opposed to invoking it opportunistically in dissents). There is also the problem of unilateral disarmament: There is no guarantee that a Court can change the culture so fundamentally that future Courts, with a different orientation, will be kept from going back to protecting their favored groups. So that form of judicial restraint does not seem like a plausible solution, even apart from the question whether, at this point, the Court should leave behind the core Carolene Products idea that was so central to its work in the last century.
A more plausible approach might be for the Supreme Court to use some version of the “passive virtues” more often[85]—that is, for the Court to try to avoid the most divisive issues that implicate the competition for victim status. That competition focuses on Supreme Court decisions, not (at least to nearly the same degree) on decisions of the lower courts. So perhaps if the Court avoided intervening, instead allowing more issues to be resolved in the lower courts— perhaps even when the lower courts disagreed—the effects of the competition could be lessened. The idea would be to avoid a decisive Supreme Court resolution of a conflict as long as it seemed to be playing out reasonably without the Court—that is, in politics and in the lower courts—and, in that way, to avoid encouraging a competition for victim status. The Court may have been doing that when it initially declined to review lower court decisions upholding marriage equality. Or, related—and again, as the Court initially did in the marriage equality cases—the Court might rule on narrow grounds and leave to the lower courts the decision whether to proceed further.[86]
Embracing the Inevitable?
It may be that some form of competition for recognition by the Supreme Court is inevitable, given the legacy of the civil rights era. Supreme Court decisions are high-profile events. They can establish clear winners and losers (at least nominally, but that is the point of the status competition). Many groups have become adept at organizing long-term litigation campaigns designed to produce a Supreme Court victory years in the future—again, using Brown as a template. And claims of victim status can be a very useful way of enlisting support for those campaigns.
Besides that, there are reasons to think that Warren Court era in which the Carolene Products approach worked reasonably well was an outlier, the product of an unusual era in which the parties were not polarized. If that is so, then the current situation—in which legal elite opinion is fragmented—may be the norm. In that situation, people appointed to the Supreme Court will be responsive to claims of victim status and will not be inclined to the kind of restraint that is needed to avoid the status competition. In particular, claims of victim status are particularly likely to be made by members of political coalitions that are losing power and, for that reason, trying to entrench themselves in the judiciary. The Federalists famously did this after losing the election of 1800, and there are less conspicuous examples as well.[87] Groups that have lost political power will tend to be disproportionately represented on life-tenured courts. That may even be what is happening now. The groups that are most concerned about gun rights, religious exercise (at least of certain kinds), and affirmative action seem to belong to a political coalition that, according to some accounts, is losing power to a more secular, cosmopolitan, and urban coalition.
Something analogous might happen on the other side, too—the side that is ascendant in politics but not yet dominant. It would not be surprising if the people appointed to courts by an emerging coalition are more aggressive and impatient than ordinary democratic politics allows. Especially if the coalition includes groups that have been excluded and treated unfairly in the past—or who perceive themselves to be the victims of unfair treatment—judges who are sympathetic to those groups will want to establish the status of those groups sooner rather than later. Even if the political trends are in their favor, they will not want to wait for political victory.
If, in fact, this kind of status competition is an inevitable response to the role that the Supreme Court plays, it might be worth identifying some value in it after all. For the groups whose power is waning, the Court—with its tendency to lag the political consensus—might provide status victories that cushion the loss of power. So long as the victories remain symbolic, you could even see this as a fortuitous compromise built into our system: As one coalition begins to dominate politics and achieve concrete gains, the losing side can keep something. On the other side, to the extent ascendant groups are represented, status victories might help siphon off some of the impatience that is generated by the slow working of democratic politics. But all of this is giving the most optimistic account of the competition for victim status, a competition that, in less favorable circumstances, will have very damaging effects. The visibility of Supreme Court decisions, the iconic status of what the Court did in the Civil Rights era, the cross-generational composition of the Court, and the increasing sophistication of litigation campaigns have made the Court a natural forum for those competing claims to victimhood. For now, a contest for victim status seems to be an aspect of the role of the Supreme Court that is not fully recognized and that is an unintended consequence of a justifiably celebrated era in the Court’s history.
Copyright © 2026 David A. Strauss* & sdf**
* Gerald Ratner Distinguished Service Professor of Law, The University of Chicago. This Essay is a version of the Jorde Lecture at the University of California, Berkeley, School of Law, and at Yale Law School, part of the 2024–25 Brennan Center Jorde Symposium. I am grateful to the Brennan Center for Justice for the invitation to give the Jorde Lecture; to John Kowal of the Brennan Center for his help; to Justin Driver, Doug NeJaime, Ryan Doerfler, Amna Akbar, and Moshe Halbertal for their responses at the Jorde Symposium events; and to workshop participants at Harvard, Duke, and the University of Chicago for their comments on earlier versions of this Essay.
[1]. See United States v. Carolene Prods. Co., 304 U.S. 144, 152 n.4 (1938).
[2]. See id. (citations omitted):
It is unnecessary to consider now whether legislation which restricts those political processes which can ordinarily be expected to bring about repeal of undesirable legislation, is to be subjected to more exacting judicial scrutiny . . . than are most other types of legislation. [citing cases involving “the right to vote,” “restraints upon the dissemination of information,” “interferences with political organizations,” and “prohibition of peaceable assembly.”]
Nor need we enquire whether similar considerations enter into the review of statutes directed at particular religious or racial minorities [or] whether prejudice against discrete and insular minorities may be a special condition, which tends seriously to curtail the operation of those political processes ordinarily to be relied upon to protect minorities, and which may call for a correspondingly more searching judicial inquiry.
[3]. This is the central argument of John Hart Ely, Democracy and Distrust (1980). See also Geoffrey R. Stone & David A. Strauss, Democracy and Equality: The Enduring Constitutional Vision of the Warren Court (2020).
[4]. 347 U.S. 483 (1954).
[5]. See, e.g., Katzenbach v. McClung, 379 U.S. 294 (1964) (upholding the constitutionality of the Civil Rights Act of 1964); South Carolina v. Katzenbach, 383 U.S. 301 (1966) (upholding the constitutionality of the Voting Rights Act of 1965).
[6]. See generally David A. Strauss, Is Carolene Products Obsolete?, 2010 U. Ill. L. Rev. 1251 (2010); Daniel A. Farber & Philip P. Frickey, Is Carolene Products Dead? Reflections on Affirmative Action and the Dynamics of Civil Rights Legislation, 79 Calif. L. Rev. 686, 691–97 (1991); Nicholas O. Stephanopoulos, The Anti-Carolene Court, 2019 Sup. Ct. Rev. 111 (2019).
[7]. I am grateful to Aziz Huq for this analogy.
[8]. For an argument in support of this claim, see generally Stone & Strauss, supra note 3. In 2024, Justice Kavanaugh commented that although the Warren Court justices “were unpopular basically from start to finish from ‘53 to ‘69,” “a lot of those decisions . . . are landmarks now that we accept as parts of the fabric of America, and the fabric of American constitutional law.” Jim Vertuno, Justice Kavanaugh Says Unpopular Rulings Can Later Become ‘Fabric of American Constitutional Law,’ Associated Press (May 10, 2024), https://apnews.com/article/supreme-court-kavanaugh-roe-wade-aa49e89ea0f35442ee1042b5bdcd9d69 [https://perma.cc/7XKK-CFPB]. I am grateful to Richard Re for this reference.
[9]. The Carolene Products footnote itself cited McCulloch. See United States v. Carolene Prods. Co., 304 U.S. 144, 152 n.4 (citing McCulloch v. Maryland, 17 U.S. (4 Wheat.) 316, 428 (1819)). McCulloch held that a state tax on the Bank of the United States was unconstitutional. The Court, anticipating Carolene Products, reasoned that while “in general [it is] a sufficient security against erroneous and oppressive taxation” that “[i]n imposing a tax the legislature acts upon its constituents,” federal instrumentalities “have no such security, nor is the right of a State to tax them sustained by the same theory.” See McCulloch v. Maryland, 17 U.S. (4 Wheat.) 316, 428 (1819).
[10]. 305 U.S. 337 (1938).
[11]. See sources cited supra note 3.
[12]. See, e.g., Bruce A. Ackerman, Beyond Carolene Products, 98 Harv. L. Rev. 713, 738 (1985).
[13]. See Carolene Prods. Co., 304 U.S. at 152 n.4 (1938).
[14]. The canonical discussion is Mancur Olson Jr., The Logic of Collective Action (1965).
[15]. This is not to say that convicted offenders’ lack of political power is unproblematic, and it is arguably a strength of the Carolene Products approach that it alerts us to, for example, the possibility that excessively severe criminal sentences are the result of a political process failure—possibly one that courts should, in principle, do something about, for example in applying a principle of lenity in interpreting statutes that impose severe sentences. But those difficult issues cannot be solved just by invoking Carolene Products.
[16]. See, e.g., Michael J. Klarman, From Jim Crow to Civil Rights 308–11, 450–51 (2004).
[17]. See, e.g., id. at 308.
[18]. For a discussion, see David A. Strauss, The Neo-Hamiltonian Temptation, 123 Yale L.J. 2676, 2695–97 (2014).
[19]. See Howard Gillman, The Constitution Besieged 101–46 (1993).
[20]. On partisan polarization, see generally, for example, Ezra Klein, Why We’re Polarized (2020); Richard H. Pildes, Why the Center Does Not Hold: The Causes of Hyperpolarized Democracy in America, 99 Calif. L. Rev. 273 (2011).
[21]. See, e.g., Drew DeSilver, The Polarization in Today’s Congress Has Roots that Go Back Decades, Pew Rsch. Ctr. (Mar. 10, 2022), https://www.pewresearch.org/short-reads/2022/03/10/the-polarization-in-todays-congress-has-roots-that-go-back-decades/ [https://perma.cc/NQ4D-AY24] (asserting that “[f]ive decades ago, 144 House Republicans were less conservative than the most conservative Democrat, and 52 House Democrats were less liberal than the most liberal Republican”).
[22]. See, e.g., Jody Freeman & David B. Spence, Old Statutes, New Problems, 163 U. Pa. L. Rev. 1, 14, 82–98 (2014) (citing sources).
[23]. This is a central theme of Neal Devins & Lawrence Baum, The Company They Keep: How Partisan Divisions Came to the Supreme Court (2019).
[24]. See, e.g., id. at 121–30.
[25]. See id. at 21–26; see also id. at xvi (“[T]he orientation of Supreme Court justices toward elite networks whose approval is important to them [has] helped to create party-line divisions on the Court.”).
[26]. See John Hart Ely, The Constitutionality of Reverse Racial Discrimination, 41 U. Chi. L. Rev. 723 (1974).
[27]. 143 S. Ct. 2141 (2023).
[28]. SFFA involved the affirmative action policies of two universities, Harvard and the University of North Carolina. The University of North Carolina is a public institution, so it is subject to the Equal Protection Clause of the Fourteenth Amendment. Harvard is private, so the Constitution, of its own force, does not apply to it. But the Court has held that Title VI of the Civil Rights Act of 1964, which forbids racial discrimination by recipients of federal funds (of which Harvard is one), mirrors the constitutional standard. So the Court discussed the Constitution in connection with Harvard. Id. at 2161–62.
[29]. Id. at 2159–61.
[30]. Id. at 2160.
[31]. 163 U.S. 537 (1896).
[32]. Id. at 551–52.
[33]. The best-known example is Herbert Wechsler, Toward Neutral Principles of Constitutional Law, 73 Harv. L. Rev. 1, 33–34 (1959). On Wechsler’s opposition to segregation, and his role in helping the lawyers in Brown attack the constitutionality of segregation, see Richard Kluger, Simple Justice: The History of Brown v. Board of Education and Black America’s Struggle for Equality 529–30 (1976).
[34]. For the most prominent account, see generally Charles L. Black, Jr., The Lawfulness of the Segregation Decisions, 69 Yale L.J. 421 (1960).
[35]. See, e.g., Fisher v. Univ. of Tex. at Aus., 579 U.S. 365, 410–15 (2016) (Alito, J., dissenting).
[36]. 143 S. Ct. 2141, 2168 (2023) (citation omitted).
[37]. 426 U.S. 229 (1976).
[38]. See Yick Wo v. Hopkins, 118 U.S. 356 (1886); case cited supra note 28.
[39]. Students for Fair Admissions, Inc. v. President & Fellows of Harvard Coll., 397 F. Supp. 3d 126, 158 (D. Mass. 2019).
[40]. See supra note 26 and accompanying text.
[41]. See, e.g., infra note 48 and accompanying text (discussing Trump v. Hawaii).
[42]. See City of Boerne v. Flores, 521 U.S. 507 (1997).
[43]. 374 U.S. 398 (1963).
[44]. Id. at 403.
[45]. Id.
[46]. See Lee Epstein & Eric A. Posner, The Roberts Court and the Transformation of Constitutional Protection for Religion: A Statistical Portrait, 2021 Sup. Ct. Rev. 315, 315–22 (2022).
[47]. See id. at 326 (“One way to think about this pattern is that the Roberts Court extended the Warren Court’s protections for minority religions so as to encompass majority religions as well. The Roberts Court is pro-mainstream religion, and more pro-mainstream Christian than the Warren Court, but not exclusively pro-mainstream.”).
[48]. 585 U.S. 667 (2018).
[49]. See generally Lawrence Hurley, Secret Recording Puts Spotlight on Alito’s Strong Conservative Views on Religious Issues, NBC News (June 11, 2024), https://www.nbcnews.com/politics/supreme-court/secret-recording-puts-spotlight-alitos-strong-conservative-views-relig-rcna156535 [https://perma.cc/PEM7-NXG5]; Michael Stokes Paulsen & Steffen N. Johnson, Scalia’s Sermonette, 72 Notre Dame L. Rev. 863 (1997).
[50]. See, e.g., Paulsen & Johnson, supra note 49, at 863–64 (“The gist of Scalia’s remarks, according to press accounts, was that traditional Christian beliefs are often misunderstood, dismissed as unreflective and, even worse, scorned by much of society today.” (footnote omitted)).
[51]. 141 S. Ct. 1294 (2021).
[52]. Id. at 1296 (citing Roman Cath. Diocese of Brooklyn v. Cuomo, 141 S. Ct. 63, 67–68 (2020)).
[53]. For an extended discussion, see Zalman Rothschild, The Impossibility of Religious Equality, 125 Colum. L. Rev. 453 (2025); see, e.g., id. at 500 (“Consider the example of an Orthodox Jewish person rushing to make it home before the Sabbath begins at sundown on a Friday who exceeds the speed limit and runs a red light. According to [the most favored nation] test, this person should be immune from the state’s traffic laws if the state provides even a single exemption—including, say, for entourages of foreign dignitaries or emergency vehicles—that undermines the law’s interest in traffic safety.”).
[54]. See generally Cornelius v. NAACP Legal Def. & Educ. Fund, Inc., 473 U.S. 788 (1985); Perry Educ. Ass’n v. Perry Loc. Educators’ Ass’n, 460 U.S. 37 (1983).
[55]. See Lehman v. City of Shaker Heights, 418 U.S. 298 (1974) (plurality opinion).
[56]. See Greer v. Spock, 424 U.S. 828 (1976).
[57]. See, e.g., Good News Club v. Milford Cent. Sch., 533 U.S. 98 (2001); Rosenberger v. Rector & Visitors of the Univ. of Va., 515 U.S. 819 (1995); Lamb’s Chapel v. Moriches Union Free Sch. Dist., 508 U.S. 384 (1993).
[58]. In a 2023 Gallup poll, 45 percent of the people surveyed said they considered religion “very important” in their lives; an additional 26 percent described religion as “fairly important.” See In Depth: Topics A to Z: Religion, Gallup, https://news.gallup.com/poll/1690/religion.aspx [https://perma.cc/A4C7-TWM4].
[59]. See Paulson & Johnson, supra note 49, at 864.
[60]. 554 U.S. 570 (2008).
[61]. 561 U.S. 742 (2010).
[62]. See, e.g., David Cole, Engines of Liberty 95–148 (2016).
[63]. See N.Y. State Rifle & Pistol Ass’n v. Bruen, 142 S. Ct. 2111 (2022); United States v. Rahimi, 144 S. Ct. 1889 (2024).
[64]. See, e.g., Mark V. Tushnet, Out of Range: Why the Constitution Can’t End the Battle over Guns, at xiv (2007) (“Disputes over gun policy have become deeply enmeshed in the culture wars between liberals and conservatives, between people who live in cities and people who live in the country.”).
[65]. See generally Dan M. Kahan & Donald Braman, More Statistics, Less Persuasion: A Cultural Theory of Gun-Risk Perceptions, 151 U. Pa. L. Rev. 1291 (2003). For a criticism, see generally Philip J. Cook & Jens Ludwig, Commentary, Fact-Free Gun Policy?, 151 U. Pa. L. Rev. 1329 (2003).
[66]. Kahan & Braman, supra note 65, at 1324; see Tushnet, supra note 64, at 115–16.
[67]. See generally Cass R. Sunstein, Second Amendment Minimalism: Heller as Griswold, 122 Harv. L. Rev. 246 (2008).
[68]. See District of Columbia v. Heller, 554 U.S 570, 626–28 (2008).
[69]. See also Caetano v. Massachusetts, 577 U.S. 411 (2016) (per curiam) (reaffirming Heller in the face of nearly open defiance by the lower court).
[70]. 142 S. Ct. 2111 (2022).
[71]. See generally Joseph Blocher, Categoricalism and Balancing in First and Second Amendment Analysis, 84 N.Y.U. L. Rev. 375 (2009).
[72]. 142 S. Ct. at 2126.
[73]. Id. at 2128–30.
[74]. Id. at 2131–33.
[75]. For a thorough analysis, see Darrell A.H. Miller & Joseph Blocher, Manufacturing Outliers, 2022 Sup. Ct. Rev. 49 (2022).
[76]. See, e.g., 142 S. Ct. at 2133 (“historical twin”; “regulatory straightjacket”; “blank check”).
[77]. See generally Justin Driver, The Consensus Constitution, 89 Tex. L. Rev. 755, 828 (2011); Loving v. Virginia, 388 U.S. 1 (1967). This is especially striking because Loving was not decided until thirteen years after Brown, and the conventional understanding is that the Supreme Court waited until public opinion had changed before taking the potentially inflammatory step of requiring states to allow interracial marriage. See Driver, supra.
[78]. Obergefell v. Hodges, 576 U.S. 644, 693 (2015) (Roberts, C.J., dissenting).
[79]. Id. at 676 (majority opinion).
[80]. Id. at 677.
[81]. Id. at 678.
[82]. 145 S. Ct. 1816 (2025).
[83]. Skrmetti, 145 S. Ct. at 1832–33.
[84]. Hecox v. Little, 104 F.4th 1061 (9th Cir. 2024), cert. granted, 145 S. Ct. 2871 (2025); B.P.J. v. W. Va. State Bd. of Educ., 98 F.4th 542 (4th Cir. 2024), cert. granted, No. 24-43, 2025 WL 1829164 (U.S. July 3, 2025).
[85]. See Alexander M. Bickel, The Least Dangerous Branch 111–98 (1962).
[86]. See, e.g., Hollingsworth v. Perry, 570 U.S. 693, 704–15 (2013); United States v. Windsor, 570 U.S. 744, 763–75 (2013).
[87]. See, e.g., Gillman, supra note 19, 31–38.
