To many Americans the Constitution, despite its authoritative status, seems to be a rather vague and open-ended document, whose meaning, as illustrated in judicial interpretations of its provisions, seems to change dramatically over the years. Many people are aware that the Constitution was once read to sanction slavery and the exclusion of women from many forms of public life, and is now read to prohibit slavery and to protect female American citizens from discrimination on the basis of their gender. They may also be aware that prayers in the public schools and prohibitions against desecration of the American flag coexisted with the Constitution for almost two centuries, but that in the late 20th century constitutional barriers to school prayer, and constitutional protections for flag-burners, have been established in Supreme Court decisions. Many have concluded, from these observations, that two oft-quoted observations about the Supreme Court’s changing interpretations of the Constitution ring true. One, by Charles Evans Hughes, was that “we live under a Constitution, but the Constitution is what the judges say it is”; the other, by newspaper columnist Finley Peter Dunne, was that “the Supreme Court follows the election returns.”
Because so many contested political and social issues have translated themselves into questions of constitutional law in the course of American history, it is tempting to conclude that Supreme Court decisions in constitutional cases are identical to the policy judgments rendered by elective branches of government. That conclusion is inaccurate. Although it is true that, in the long run, changes in the constitutional interpretations advanced by the Court can be seen to parallel changing attitudes in American society at large, such is not necessarily the case in the short run, for reasons that relate to the institutional role of judges as constitutional interpreters and the complicated relationship between constitutional claims brought to the Court and their cultural context. In particular, a very important ingredient in constitutional decision-making by the Court is often ignored in descriptions of changes in constitutional law.
That ingredient is the way in which the members of a Court, at a given point in time, define their posture in reviewing the actions of other branches of American government on constitutional grounds. It has made a great deal of difference, over the course of American constitutional history, whether the “approved” stance exercised by Supreme Court judges in reviewing constitutional challenges to legislation, or to executive or administrative policymaking, has been taken to be one of searching scrutiny, relaxed scrutiny, or some combination of those postures.
Too often the stances of judicial review adopted by Supreme Court justices have been given ideological labels, resulting in particular Courts being labeled as “liberal,” “conservative,” “progressive,” or “reactionary.” Judicial conceptions of the proper role of the Supreme Court in reviewing the policies of other branches on constitutional grounds cannot fruitfully be captured in such labels. Although such conceptions change over time, they tend to be shared by judges of quite different ideological convictions. Employing conventional political labels to characterize the constitutional review stance of the Court at any point in time only serves to obscure this dimension of the Supreme Court’s constitutional law decisions.
In the following survey of developments in American constitutional law between 1925 and 2000, I will be highlighting the relationship of changing doctrines to changes in the “mainstream” stance of constitutional review adopted by the Supreme Court. Although I will be discussing connections between particular stances of review and their historical setting, my primary purpose is to disabuse readers of an inclination to believe that the Court’s 20th-century constitutional law decisions have followed the election returns, or that they can be seen as the collective products of the idiosyncratic views of nine justices.
A brief glance at the universe of American constitutional law in 1925 reveals a doctrinal world dramatically different from that which existed at mid-century. In race relations cases, the Court’s articulation of the “separate but equal” doctrine in Plessy v. Ferguson (1896) had helped foster an intricate series of state laws requiring separate facilities for whites and other designated races. In free speech cases, the first two decades of the 20th century witnessed the absence of a single successful free speech challenge to legislation criminalizing membership in collectivist political organizations, or the expression of pacifist views in wartime. Thus in two areas where later Courts would be particularly aggressive in invoking the Equal Protection Clause of the Fourteenth Amendment to invalidate legislative discriminations, and in using the First and Fourteenth Amendments to protect various lands of unpopular expression, the early 20th-century Court passively tolerated legislative policies.
This passivity was not, however, a product of the stance of constitutional review adopted by the Court toward legislation making racial classifications or restricting freedom of expression. It was, rather, the result of established judicial interpretations of the meaning of the Equal Protection Clause and the First Amendment. Those interpretations presupposed that the Court had a good deal of leeway to scrutinize legislation on constitutional grounds. If it did not establish such limitations, that was a result of its own interpretations of constitutional provisions, not because of a general deference to legislatures.
In other areas of constitutional law the early 20th-century Court had established severe constitutional limitations on legislatures. Two years before The Virginia Quarterly was launched, the Court’s Adkins v. Children’s Hospital decision found a Congressional statute establishing a minimum wage for female workers in certain occupations in the District of Columbia constitutionally invalid as an invasion of the “liberty” of those workers, and their employers, to bargain for the terms of employment, which the Court read into the Fifth Amendment’s Due Process Clause. Even more revealing than the result in Adkins was the Court’s assumption that nearly every piece of state or federal legislation that attempted to implement regulatory or redistributive economic policies raised constitutional issues. The Court did not invalidate all such statutes: where it found that a piece of legislation bore a reasonable connection to public health, safety, or morals, it was prepared to sustain it. But it assumed that every legislative effort to regulate the hours, wages, or working conditions of American enterprise could be scrutinized by judges on constitutional grounds.
Federal legislation which attempted to police the flow of goods and services in interstate commerce was treated as comparably susceptible to judicial oversight. Although the Constitution gives Congress the authority to regulate “commerce among the several states,” early 20th-century judges treated this power as being carved out of a residuum of state power to regulate commercial activity within state borders. This meant that they regularly labeled Congressional efforts to apply the federal commerce power to “local” enterprises invalid encroachments into state prerogatives. Early 20th-century judges also defined “commerce” as distinct from other activities connected with the process of producing and distributing commercial products and services, treating efforts on the part of the federal government to regulate “manufacturing” or “production” as constitutionally suspect. A 1918 decision by the Court, Hammer v. Dagenhart, reaffirmed those limitations on the federal commerce power; they remained in place until the early 1940’s.
Historians, looking back at this sample of cases from a late 20th-century perspective, have been quick to contrast the Court’s response to legislation affecting civil liberties with its response to legislation restricting economic activity, and to conclude that the Court was biased in favor of one set of freedoms and against another. This conclusion is a distortion of the Court’s stance. The Court’s constitutional tendencies in race relations and free speech tracked those of other governmental branches of government at the same time: legislation passed by Congress and the states in those areas between 1900 and 1925 perpetuated racial discrimination and restricted various forms of expression. Rather than illustrations of specific biases, the tendencies can be seen as reflecting a governing judicial attitude toward the Constitution itself.
The majority of early 20th-century judges and legal commentators assumed, as Chief Justice John Marshall had assumed in the 1819 case of McCulloch v. Maryland, that the Constitution was “intended to endure for ages to come, and, consequently, to be adapted to the various crises in human affairs.” In making this comment Marshall was not suggesting that the meaning of the Constitution was expected to change with the times. On the contrary, he was suggesting that the Constitution embodied fundamental principles of American law and government which would be “adapted,” by judges, to new controversies as they arose.
A recognition that the attitudes about constitutional interpretation embodied in Marshall’s comment in McCulloch v. Maryland had not been fully abandoned by judges in the early 20th century goes far to explain the persistence, through the 1930’s, of limited constructions of the federal commerce power, aggressive judicial scrutiny of regulatory or redistributive economic legislation, and a disinclination in the Supreme Court to find constitutional objections to legislation restricting free speech or segregating persons on racial grounds.
The cases invalidating proposed federal legislation based on the commerce power assumed that the Constitution envisaged a federal government of limited, enumerated powers. The “liberty of contract” cases assumed that hours or wages laws were particularistic redistributions of the income of certain employers to certain employees, not general health measures. The free speech cases assumed that government had ample power to suppress expressions critical of its officials or threatening to its security, so long as it did not censor such expressions in advance. The racial segregation cases assumed that the Equal Protection Clause did not prevent the separation of persons of different races unless that separation resulted in unequal treatment based on race.
Although the subjects were controversial, and Supreme Court decisions in cases raising constitutional issues related to those subjects were not always well received, the legal basis on which the Court’s decisions were grounded was consistent with orthodox constitutional jurisprudence at the time. Thus the Court’s decisions in race cases, free speech cases, and cases testing the constitutionality of governmental efforts to regulate the economy cannot satisfactorily be described as pro-business or anti-civil rights: the Court’s more significant bias was in support of the prevailing interpretive orthodoxies of early 20th-century constitutional jurisprudence.
By 1950 many observers of the state of American constitutional law described themselves as witnesses to a “constitutional revolution” that had profoundly changed the interpretive posture of the Supreme Court and the operative meaning of the Constitution. Scholars pointed to a series of decisions in free speech cases, many involving . members of the Jehovah’s Witnesses sect, in which the Court had curtailed the powers of states and municipalities to restrict inflammatory expressions. They noted that after 1937 a majority of the Court had consistently upheld state legislation, even if it infringed upon the “liberties” of economic actors, if it could be shown to be grounded on some rational public policy agenda. They pointed out that two commerce power decisions, United States v, Darby (1941) and Wickard v. Filburn (1942), had virtually obliterated the limitations on federal power that Hammer v. Dagenhart had extracted from the Constitution’s reservation of unenumerated powers in the states.
Finally, they noted that, despite the continued presence of officially mandated racial segregation and the entrenchment of the doctrine that “separate but equal” state facilities for different races did not violate the Equal Protection Clause, the Court’s attitude toward racial discrimination cases was showing signs of change. A group of Court decisions from 1938 through 1950, centering on higher education, began to insist that racially segregated state institutions achieve some functional measure of equality, a path that would lead to its abandonment of the “separate but equal” doctrine in Brown v. Board of Education (1954).
Arguably the most significant index of constitutional change in the years between 1937 and 1950, however, was not found in any line of cases. It was the Court’s altered definition of its role in reviewing the constitutionality of various types of legislation. In the years between 1900 and 1925, we have seen, the Court’s justices assumed that they would undertake the same level of scrutiny for all legislation that allegedly violated constitutional guaranties. Although the Court upheld far fewer challenges to laws affecting civil rights than to laws affecting economic relationships, its different degree of solicitude toward economic and civil rights was not a function of a different stance of constitutional review toward types of legislation.
By 1950, however, it was obvious that the Court was no longer adopting a posture of treating constitutional challenges to all forms of legislation as triggering severe judicial scrutiny. It had retained that posture in free speech and freedom of religion cases, and was showing signs of employing it in some racial discrimination cases. In all of the remaining areas of domestic constitutional law it had begun to adopt a deferential stance toward challenged legislation.
In the 1950’s and 1960’s the Court repeatedly deferred to the policies of state legislatures or Congress in matters governing the economy. In labor relations cases it consistently upheld Congress’s delegation of power to the National Labor Relations Board to impose collective bargaining requirements and to prescribe compulsory wage levels and limitations on the hours of work. Its labor relations cases were illustrative of a general tendency to approve the Congressional creation of a large number of federal regulatory and administrative agencies to govern the economy. In the same time period it regularly sustained state and federal legislation against “liberty of contract” challenges, and, as we have noted, permitted extensions of the federal commerce power into areas previously thought “local” in character.
The Court did not adopt this deferential approach to legislation regulating speech or establishing classifications on racial grounds. It subjected such legislation to a high level of constitutional scrutiny, in the process invalidating a number of state statutes restricting the speech rights of political dissidents or religious minorities, or denying white and black citizens equal access to public accommodations. The Court’s decisions, taken as a whole, suggested that it was abandoning its generically aggressive stance in reviewing legislation challenged on constitutional grounds for a stance in which it selectively deferred to, or vigorously scrutinized, particular classes of legislation.
In United States v. Carolene Products Corp. (1938), the Court gave a cryptic justification for its selectively deferential and aggressive stance of constitutional review. In that case the Carolene Products Company challenged the constitutionality of a 1923 statute prohibiting the shipment, in interstate commerce, of skimmed milk compounded with products other than milk. In an opinion sustaining the statute, Justice Harlan Fiske Stone took it upon himself to articulate a standard of constitutional review which should govern the Court’s scrutiny of most legislation then being subjected to constitutional challenges, and at the same time to outline instances in which the Court might continue to undertake heightened review.
Stone announced that a “presumption of constitutionality” would be afforded to “regulatory legislation affecting ordinary commercial transactions.” This “presumption of constitutionality” meant that the recital of a legislative purpose in connection with a statute regulating economic activity was sufficient to create a presumption that the legislation being challenged had a rational basis.
Having announced this “presumption of constitutionality” for most regulatory legislation affecting economic activity, Stone then, in a footnote, gave some reasons why it might be departed from in particular sets of cases. The first category of cases were ones in which the legislation being challenged “appear[ed] on its face to be within a specific prohibition of the Constitution.” The cases in that category, for the most part, were examples of the Court’s “incorporation,” between 1925 and 1938, of certain provisions of the Bill of Rights (primarily the First Amendment) into the Due Process Clause of the Fourteenth Amendment, so they applied against the states as well as the federal government. The judicial incorporation of such provisions posed direct conflicts between them and state legislation, suggesting, for Stone, that higher scrutiny was required.
Another category of cases listed in the footnote were those in which the Court had struck down efforts on the part of Southern states to disenfranchise blacks by disqualifying them from voting in primary elections. Stone used two of those cases, as well as “incorporated” free speech cases, in support of the general proposition that “legislation which restricts those political processes which can ordinarily be expected to bring about repeal of undesirable legislation” might be “subjected to more exacting judicial scrutiny under the general prohibitions of the Fourteenth Amendment than are most other types of legislation.”
The portion of Stone’s Carolene Products opinion in which he announced the presumption of constitutionality in cases involving ordinary commercial transactions, and which contained the footnote, was joined by only three other justices. Hugo Black declined to join it, Pierce Butler concurred only in the result, James McReynolds dissented, and Benjamin Cardozo and Stanley Reed took no part in the decision, leaving only Charles Evans Hughes, Louis Brandeis, and Owen Roberts subscribing to the “presumption of constitutionality” formula and Stone’s footnote. But the justifications Stone offered from departing from the “presumption of constitutionality,” and the almost casual elevation of that presumption into a norm of constitutional review, were to make Carolene Products an inspirational decision for the Court under the tenure of Earl Warren (1954—1969).
The Warren Court extended the logic of Carolene Productsreview. It relaxed its scrutiny of economic legislation even further, and at the same time extended the scope of its heightened scrutiny of non-economic legislation which fell into one of the categories Stone had outlined in the footnote. This tendency of the Warren Court was anticipated in its first major decision, Brown v. Board of Education.
Brown identified the Court as an institution that was prepared to employ its constitutional decisions as mandates to compel more majoritarian branches of government to conform their policies to the Court’s understanding of the requirements of equal justice. Congress, prior to the Brown decision, had shown no inclination to pressure Southern states to end enforced segregation, even though the United States military had been desegregated since the Second World War. Nor had any presidential administration, from the 1930’s through the early 1950’s, pressed for an end to racial segregation. The Court’s unarticulated premise in Brown had been that the legislatures which discriminated against black Americans had excluded them from participating in the political process, that the executive had reinforced that exclusion, and that the Court could act as a constitutional check on such activity.
For the remainder of Earl Warren’s tenure the Court continued to maintain the stance of an aggressive scrutinizer of non-economic majoritarian legislation on Carolene Products grounds. After Browncame the reapportionment cases of the 1960’s, in which systems of apportioning votes in state legislatures on the basis of other than population were struck down, even where a referendum of the state’s citizens had voted to retain proportional representation based on older settlement patterns. The early 1960’s also produced the school prayer cases, where the Court banned the practice of schools offering state-created, non-denominational prayers, devotional readings from the Bible, or recitations of the Lord’s Prayer as part of their regular curricula.
The same decade witnessed the flowering of a line of cases tightening the constitutional requirements of federal, and especially state, criminal procedure. As early as 1957 the Court had insisted that criminal defendants be brought before a magistrate before being indicted in federal trials. Between 1961 and 1969 it revolutionized state criminal procedure in a two-step process, the first step being a series of decisions incorporating Bill of Rights protections accorded to defendants in criminal trials, such as the Fourth Amendment’s protection against unreasonable searches and seizures, the Fifth Amendment’s protections against self-incrimination and double jeopardy, and the Sixth Amendment’s guarantee of a jury trial in criminal cases. Against this backdrop of enhanced constitutional scrutiny of state legal and police procedures affecting criminal defendants, the Court held, in a trio of decisions emanating from state courts between 1963 and 1966, that persons accused of crimes were entitled to legal representation, even if they could not afford it; that a person interrogated by the police was entitled to a lawyer if one was requested; and that police were required to give a series of constitutionally mandated warnings to persons whom they had taken into custody, including the warnings that they could remain silent and that they were entitled to legal representation.
Accompanying these decisions, for virtually the entire span of Warren’s tenure, were a variety of cases in which the Court expanded the scope of constitutional protections for freedom of expression. They included a line of obscenity decisions, beginning in 1957, in which the Court, after years of struggling to find a species of purportedly “obscene” expression which could be suppressed, finally, in one case, adopted a procedure in which any obscenity conviction could be overturned if five sitting justices concluded, for one reason or another, that the work in question was not obscene. They also included cases significantly narrowing the category of unprotected “fighting words” and “subversive expressions.” In addition, the Court’s freedom of expression decisions widened the category of constitutionally protected “speech” to include symbolic acts, such as burning a draft card or desecrating a flag.
Finally, in the most doctrinally revolutionary of all of its free speech decisions, the Court constitionalized the law of defamation, discovering a First Amendment privilege to make false and damaging factual statements about “public figures,” which include most public officials and especially visible private citizens, so long as the statements were not made with “malice,” a constitutional term of art which included deliberate falsehood or reckless indifference to the truth or falsity of a statement. The theory animating the Court’s defamation decisions was that in a democratic society a constitutional privilege for even false and damaging statements was necessary to maintain a “breathing space” for comment about public affairs.
By 1975 Warren Burger had replaced Earl Warren as Chief Justice, and Richard Nixon, who had openly criticized the Warren Court in his 1968 presidential campaign, had appointed four new justices to the Court. But the Court’s constitutional review stance was still inspired by Carolene Products.In the years from 1969 to 1975 the Court not only maintained its commitment to most of the major Warren Court constitutional innovations, it continued to extend the range of its scrutiny of majoritarian legislation, venturing into areas where the Warren Court’s review posture had been cautious. Two major examples were gender discrimination and a species of “liberties” in the Due Process Clauses that the Court associated with privacy, personhood, or family relations.
Between 1971 and 1982 the Court regularly struck down statutes making gender-based classifications, invalidating an Idaho statute establishing a preference for male over female family members as administrators of the estates of persons who had died without making a will; Congressional legislation providing that male service members could assert that their wives were financial dependents for the purposes of receiving medical benefits, but female service members were required to prove their husbands’s dependency; an Oklahoma statute permitting females between the ages of 18 and 21 to drink 3.2 beer, but preventing males in the same age group from doing so; an Alabama statute requiring divorcing husbands, but not wives, to pay alimony to their ex-spouses; and a Mississippi statute excluding men from the state’s nursing schools. In each case the Court’s opinion emphasized the role of gender stereotyping in producing the legislative classification, which robbed the statute of its presumptive rationality.
Under the criteria for constitutional review set forth in Carolene Products, gender-based legislative classifications did not obviously fall into the types of statutes triggering heightened review. The first type of statute mentioned in Carolene Products had been one which facially contradicted a provision of the Constitution. There did not appear to be a facial contradiction between the Equal Protection Clause and laws classifying persons on the basis of gender. In some areas of human conduct, such as childbearing, gender was an essential, functionally significant characteristic, and the Equal Protection Clause had consistently been interpreted as not prohibiting all legislative discriminations among classes of persons, simply those that could not be seen as reasonable because they tracked some essential, functional differences among the classes receiving different treatment.
Nor did the other justification for enhanced review offered in Carolene Products—keeping open the political processes that might produce repeal of legislation discriminating against minorities— easily apply to gender discrimination cases. The overwhelmingly male legislative majorities who established classifications based on gender might have been acting out of a sense that in many respects males and females were essentially and functionally different.
The Burger Court’s gender discrimination opinions suggested that its justices implicitly recognized the difficulty of fitting legislation establishing classifications on the basis of gender into one of the Carolene Products rationales for heightened scrutiny. But they also suggested that the Court did not find such classifications presumptively rational. The opinions eventually produced the first appearance of an intermediate standard of Carolene Products constitutional review, in which the Court came to insist that some legislative classifications needed to serve “important governmental objectives” and be “substantially related” to those objectives. This standard made it relatively easy for the Court to expose legislation based on gender stereotypes, since treating men and women differently could only be an “important” purpose of legislation if the different treatment tracked some essential gender difference with functional significance.
Meanwhile the early Burger Court began to wrestle with another line of cases that bore an awkward relationship to the blueprint for constitutional review laid out in Carolene Products.The cases, which the Warren Court had originally conceptualized as involving constitutional “privacy” claims, were eventually treated by the Burger Court as another category of “liberty” cases under the Due Process Clauses.
The sequence of cases had begun with two Warren Court cases, Poe v. Uttman (1962) and Griswold v. Connecticut (1965), involving challenges to a Connecticut statute prohibiting the distribution of contraceptives or information about birth control.Poe and Griswoldraised a number of problems for the Carolene Products framework of constitutional review. There was no clear textual basis for protecting “privacy” or the “liberty” to make intimate sexual or procreational choices. The designation of “liberty” in the Due Process Clauses as encompassing such choices, and the characterization of that liberty as “fundamental,” was being made by judges. It was just this sort of judicial glossing of open-ended constitutional provisions that Carolene Products jurisprudence had deemed to be inconsistent with democratic theory in cases reviewing economic legislation. Nor was it easy to see married couples as classes of persons who were excluded from the processes which produced legislation or who were stigmatized by legislative prejudice.
Initially a Court majority demonstrated its reluctance to identify sexual and procreative liberties as “fundamental” protections afforded by the Due Process Clauses by endorsing an opinion, by Justice William O.Douglas in Griswold, which argued that “penumbras and emanations” from specific Bill of Rights provisions could be combined to derive a constitutional right of privacy. By the early 1970’s Burger Court majorities had abandoned this approach, and concluded that cases involving sexual and procreative choices were more accurately described as “liberty” rather than “privacy” cases. In Roe v. Wade, the Burger Court’s most visible and controversial decision, the constitutional basis for a limited right to an abortion was placed squarely in the Due Process Clauses’s protection for “liberties.”
The Burger Court’s efforts to modify Carolene Products review had not been accompanied by any general theoretical justification, and the awkwardness of its posture was highlighted in one of its most controversial decisions, the 1986 case of Bowers v. Hardwick, a challenge to a Georgia statute criminalizing oral and anal sex as “unnatural,” enforced against consenting homosexual males.
Bowers v. Hardwick presented the Burger Court with an opportunity to clarify the scope of “fundamental” liberties under the Due Process Clauses, or, alternatively, to decide whether classifications on the basis of sexual orientation might trigger heightened scrutiny. The Court declined to embrace either opportunity. In a narrow opinion written by Justice Byron White, it held only that the choice to engage in consensual homosexual sodomy was not a “fundamental right” for due process purposes. It came to this conclusion because proscriptions against sodomy had a long history in America, with all states having criminalized the practice as late as 1961, and because it felt disinclined “to take [an] expansive view of our authority to discover new fundamental rights imbedded in the Due Process Clause,” since the Court was “most vulnerable … when it deals with judge-made constitutional law having little or no cognizable roots in the language or design of the Constitution.”
Thus Bowers finessed all of the constitutional issues it could have resolved. Would a sodomy statute enforced against consenting heterosexuals be upheld in light of Griswold and Roe? Would such a statute rest on legislative stereotypes, and would it affect “fundamental” rights? And if it were constitutionally invalid as applied to heterosexuals, why was it valid as applied to homosexuals? In a 1985 case the Burger Court had given some hints about how it might have responded to the last of those questions.
The case was City ofCleburne v. Cleburne Living Center, in which a zoning ordinance which prevented the construction of a facility housing mentally retarded persons in a residential neighborhood was challenged on equal protection grounds. White’s majority opinion was illustrative of the Burger Court’s modification of Carolene Products criteria to produce strict, intermediate, and deferential standards of review. No one suggested in Cleburne that statutes establishing classifications on the basis of mental disability should be subjected to strict scrutiny, such as classifications based on race; the question before the Court was whether those who enacted the ordinance should be held to a rational basis or an intermediate standard. This was because mental disability was not a facially inessential characteristic, such as skin color. It was potentially, but not invariably, a characteristic that had functional consequences.
After Cleburne, it was clear that strict scrutiny was only available, in equal protection cases, when a legislative classification was based on a facially inessential characteristic. When the classification was based on an apparently essential characteristic with functional consequences, rational basis scrutiny appeared appropriate unless the legislation in question did not seem to be tracking the essential differences between the classes signaled out for different treatment. If it did not, the possibility of legislative stereotyping was raised, and a heightened level of scrutiny, employing additional Carolene Products criteria, came into play.
The cases in which the Burger Court struggled to articulate differing levels of constitutional review revealed that Carolene Products criteria continued to dominate the Court’s approach to legislation challenged on constitutional grounds during the middle and late years of the 20th century. The Court’s continued acceptance, and refinement, of the assumptions driving Carolene Products review go far to explaining why so many of its decisions befuddled observers who expected it to reverse the Warren Court’s stance of constitutional solicitude for allegedly powerless minorities.
William Rehnquist’s succession to the Chief Justiceship on Warren Burger’s retirement in 1986 brought forth predictions that the Court would now make a decisive turn to the right end of the political spectrum. But in four areas where Warren and early Burger Court decisions had generally been identified as “liberal” and had drawn opposition from right-wing groups—abortion, school prayer, gender discrimination, and First Amendment protection for flag descecration —the Rehnquist Court reaffirmed and extended those decisions. It endorsed Roe v. Wade’s declaration of a limited constitutional right to abortion, refused to allow a non-denominational prayer at a middle school graduation ceremony, invalidated the exclusion of women from a state-supported college, and struck down ordinances prohibiting the burning of the American flag as a political protest.
The various rationales advanced by the Rehnquist Court in those decisions convey some sense of the difficulties of equating the Court’s constitutional law decisions with those made by elected officials. In upholding Roe v. Wade the Court stressed the importance of not lightly overruling a precedent that had significant implications for the procreational choices of numerous Americans. In the middle school prayer case the Court reaffirmed a constitutional principle articulated by the Warren Court’s school prayer decisions in the early 1960’s, that any symbolic affiliation of public schools with religious ceremonies cannot be squared with the Establishment Clause’s prohibition against state endorsement of religion. In the case preventing Virginia from excluding women from a state school whose educational orientation was patterned along the lines of military service academies, the Court invoked a line of gender discrimination cases in concluding that the absence of an “exceedingly persuasive” rationale for the exclusion suggested that it may have been based on gender stereotypes. And in the flag burning cases Court majorities referred to the “bedrock principle underlying the First Amendment” that “the Government may not prohibit the expression of an idea simply because society finds the idea . . .offensive or disagreeable.” All of the decisions reflected the justices’s belief in a sharp distinction between their constitutional or institutional obligations as members of the Court and the political consequences of their decisions.
By 1980 the Burger Court had approved race-based remedies, the most conspicuous of which being affirmative action programs, in higher education, private employment, and public contracting, stressing that each of the remedial programs it sustained were responses to past discrimination. But a 1976 decision, Washington v. Davis, had insisted that race-specific remedies needed to be connected to previous race-specific legislative policies: a facially neutral policy could not trigger a race-conscious remedy even if its administration disadvantaged racial minorities. Building on the distinction between specific past discriminations affecting the claimants in a lawsuit and general past discrimination against minorities, the Court, in cases decided between 1984 and 1986, found race-based remedies constitutionally invalid if they merely represented a response to previous discrimination in society at large.
Beginning with a 1989 decision, Rehnquist Court majorities have taken the Burger Court’s race cases to have insisted that all race-based classifications be subjected to strict scrutiny. This is a creative reading of the Burger Court racial discrimination cases, some of which contained language suggesting that the Court’s review of certain race-conscious remedies had been less than strict. It suggested that in the future only race-conscious remedies tailored to relieve specific past discriminations that burdened specific claimants would pass constitutional muster.
But a majority of the Rehnquist Court nonetheless seems to have settled on an approach to race cases hauntingly similar to that employed by the Warren Court immediately after Brown: any classification based on race smacks of racial stereotyping and requires the highest level of justification. It is not even clear, at this point, whether one of the established justifications for race-conscious admissions programs in institutions of higher education—that of maintaining diversity in the student body—could surmount the Court’s conclusion that any form of racial discrimination is highly suspect. Two federal Court of Appeals decisions striking down affirmative action admissions policies of state universities have been let stand by the Court.
In seeking to explain the Rehnquist Court’s decisions in race relations cases, political labels would seem to be little help. If the test for being labeled a “liberal” or “conservative” judge is limited to racial discrimination cases, and based on a judge’s attitude toward policies designed to benefit racial minorities, the Burger Court was collectively more liberal than the Warren Court, and the apogee of the Burger Court’s liberalism came in the early 1980’s, when the national electorate ostensibly tilted rightward. But if the test is put in terms of standards of constitutional review, the Burger Court, more than either the Warren Court or the Rehnquist Court, was willing to relax the principle that any legislative classification based on race required exacting scrutiny.
This survey of American constitutional law in the 20th century has suggested that no variable has had a greater impact on the course of that body of law than the changing standards employed by successive Supreme Courts in the review of legislation challenged on constitutional grounds. From this point of view, the most important thing one can say about the Court before 1925 was that it treated all forms of legislation as candidates for aggressive constitutional review; the most significant constitutional development of the years between 1925 and 1950 was the gradual replacement of uniformly aggressive review with selective, Carolene Products review; the dominant tendency of the Court between 1950 and 1975 was the extension of the categories of heightened scrutiny under Carolene Products review; and the defining feature of the Court’s constitutional law decisions in the last 25 years has been the playing out of the logic of Carolene Productsreview to the point where it may have reached a dead end.
The meaning of the last comment may be clarified if we recall that the dominant stance of constitutional review adopted by the Court at a point in time can be seen as an index of attitudes toward the role of Supreme Court judges as constitutional interpreters, and also as a signal of the constitutional claims to which Americans, at a period in their history, attach great significance. A stance of aggressive review toward all forms of legislation signifies that the justices on a Court treat themselves as having a special responsibility to define and declare the meaning of constitutional provisions, and have no particular anxiety about the “countermajoritarian” implications of that responsibility. When one finds that stance held by the Court in the first decades of the 20th century, but at the same time very few cases in which the Court engaged in aggressive scrutiny of legislation making racial classifications, the message is clear: most Americans, including the Court’s justices, did not regard a claim based on racial discrimination as pressing and important.
When, in the early 1940’s, one finds the Court relaxing its standard of review in cases challenging legislation regulating economic activity, maintaining a heightened standard of review in free speech cases, and striking down several state statutes and municipal regulations on First Amendment grounds, a different message is equally clear. The Court was less confident about its role as a constitutional interpreter in the realm of political economy, possibly because its justices no longer believed that the Constitution gave definitive answers to questions of economic policy. At the same time, because of an enhanced appreciation of America as a democratic alternative to totalitarian regimes, the Court was increasingly confident that majoritarian repression of free expression struck at the very heart of a democratic society, and the number of free speech challenges brought against legislation in the period suggests that many other Americans found freedom of speech culturally significant.
From the 1950’s on, heightened Carolene Products review has functioned as a surrogate for the cultural stature of particular constitutional claims. Claims based on equal justice for racial minorities, or on freedom of speech and religion, or on the right to vote or to comment on matters of public concern, have had very high cultural standing and have been subjected to the highest levels of constitutional review. Claims based on gender have progressed from virtual invisibility to considerable resonance, and the heightening scrutiny of such claims has reflected that trend. Claims based on intimate sexual choices, for both heterosexuals and homosexuals, have made a comparable progression. In the latter two lines of cases the Carolene Products justification for heightened scrutiny which emphasizes prejudice against powerless minorities seems to have been particularly prominent, serving to transform what appeared to be natural and necessary classifications into ones based on uninformed stereotypes.
But, as the Burger Court recognized in Cleburne, the logic of strands two and three of Carolene Products review—keeping open the political channels that would produce change and redressing prejudice against minorities—points in the direction of an almost endless series of claims for equal justice by groups who feel disadvantaged by a legislative policy. It also points, as reflected in the Court’s recent decisions interpreting the First Amendment’s religion clauses, toward a heightened sensitivity toward majoritarian policies that allegedly disadvantage any form of religious worship or privilege traditional religious beliefs. At some point sorting out the differences between legislative classifications with functional significance and those based on stereotypes requires the Court to make ultra-fine distinctions or educated guesses, and at some point the concept of majority rule itself becomes threatened.
If America is assumed to be a culturally and politically diverse, rather than a homogeneous, society, the legislative forum takes on more of the character of an arena in which discrete minority groups compete with one another, as well as with the “majority,” for influence. Scrutiny of legislation disadvantaging a cultural or political minority thus becomes not simply a matter of ensuring that the minority is not unfairly marginalized by the majority, but rewarding that minority, at least provisionally, at the expense of other minorities competing for consideration and influence. Thus instead of heightened review simply functioning to prevent a homogeneous or prejudiced majority from excluding or disadvantaging those who do not share its characteristics or beliefs, it can be seen, in today’s America, as functioning to give some minorities a provisional comparative advantage in competition against other minorities. In sum, two forms of heightened Carolene Products review can be seen as performing the very task that the Carolene Products Court felt was appropriately performed by legislatures when they enacted policies affecting ordinary economic transactions.
Thus the Rehnquist Court may be anticipating the eventual abandonment of Carolene Products review for another model. Under that model, heightened scrutiny would be reserved for instances in which legislation was in direct conflict with a constitutional provision. This would not necessarily mean that the Court’s general scrutiny of legislation would be relaxed, because, as the Rehnquist Court’s racial discrimination cases suggest, the model permits judicial glossing of open-ended constitutional provisions such as the Equal Protection Clause, so individual Courts might be inclined toward broad interpretations of the constitutional text for the purposes of invoking heightened review. But under the model the Court would employ the same approach toward the shifting cultural and political dynamics of the legislative process that it employed toward the economic dynamics of policymaking. Legislative marginalization of minorities would either raise a facial conflict with the Constitution or receive relaxed scrutiny.
It is far too early to suggest that Carolene Products review has become obsolescent. But the history of American constitutional law in the 20th century does not merely suggest that changing postures of review have been the most important variable affecting the course of constitutional decisions. It also suggests that those postures, being indices of deeply shared assumptions about the role of Supreme Court justices as constitutional interpreters and about the current cultural stature of particular constitutional claims, are also fated to change with time. Just as it was no accident that the Court decided to relax its scrutiny of economic legislation in the wake of an economic depression and heighten its scrutiny of legislation restricting freedom of expression around the outbreak of World War II, it is no accident that the growing cultural diversity of America has complicated the meaning of legislative “prejudice.” As the Supreme Court responds to new cultural configurations and constitutional claims, we are likely to see political labels attached to that response. But the most significant clues to the Court’s response will be found in whether it continues to maintain a posture of selectively deferential and aggressive constitutional review, and how it seeks to justify the posture it adopts. In those justifications one will catch a glimpse of the values currently taken as undergirding American constitutional law.