The second constitutional era of the right of association is the equality era, which began in the mid-1960s. It includes the transformation of the right of association into intimate and expressive components in Roberts v. United States Jaycees. As I suggested at the end of the previous chapter, this transformation in some ways took its cues from the foundations established during the national security era. But the equality era also introduced its own political, jurisprudential, and theoretical factors that influenced associational freedom. This chapter focuses on three of these factors, each of which further contributed to the decline of the protections for group autonomy.
The primary political factor affecting the right of association in the equality era involved ongoing efforts to attain meaningful civil rights for African Americans. As the Civil Rights Movement gained traction, civil rights activists shifted from protecting their own associational freedom (as represented in the NAACP cases chronicled in the previous chapter) to challenging segregationists’ arguments for group autonomy. Questions about the limits of exclusion became increasingly complex as these challenges extended to private groups.
The jurisprudential factor affecting the right of association in the equality era was the development of another constitutional right that appeared nowhere in the text of the Constitution: the right to privacy. Privacy and association had been linked in some of the Court’s earliest cases on the freedom of association, but new connections emerged in the 1965 decision of Griswold v. Connecticut. Griswold’s framework eventually led to the right of intimate association recognized in Roberts.
The theoretical factor dominating the equality era was the rise of Rawlsian liberalism. Rawlsian questions about the relationship between liberty and equality, the limits of public reason, and the contours of individual autonomy dominated scholarly discussions about associational freedom during the equality era. Rawlsian premises also permeated the legal academy during this time, and they are implicit in Roberts and the scholarly commentary that followed the Court’s decision.
Unlike the emergence of association in the national security era, the transformation of association in the equality era was already a few decades removed from the disappearance of the right of assembly in legal and political discourse. But retracing the transformation of association in this later era highlights how the new version of association further departed from the deference to group autonomy once encompassed by the right of assembly: (1) rejecting dissenting and destabilizing groups in the interests of consensus norms; (2) depoliticizing social practices that once counted as part of political life; and (3) construing groups once seen as forms of expression as merely means of expression. Like the previous chapter, this chapter highlights the historical context in which these changes unfolded. The next chapter explores the implications of these changes by developing a theory of assembly.
Civil Rights in Public and Private
The right of association that emerged in the national security era introduced crucial protections to the NAACP and its efforts to promote equality and civil rights for African Americans. During the equality era, freedom of association claims arose from a much different corner: segregationists who wished to curb the march of integration. The segregationist challenge raised complicated questions about the line between public and private groups. Herbert Wechsler had infamously argued the year after NAACP v. Alabama that the freedom of association was implicated in any effort at integration, arguing that “integration force[d] an association upon those for whom it [was] unpleasant or repugnant.”1
Although Wechsler had directed part of his critique against Brown v. Board of Education, his argument lacked plausibility in public settings like the schools at issue in Brown—it made little sense to argue that segregationists had a freedom to associate (or a right to exclude) in situations where the government provided a public good or service. Within a decade of Brown, forced integration in public education, public transportation, public buildings, and public recreational facilities had been widely accepted. Integration also extended to private entities doing business on public property. By 1961, integration applied “to virtually any private concern operating on public property,” and three years later, segregation in “most forms of public life” had come to an end. As Gerald Rosenberg suggests, “by the late 1960s and early 1970s there was not as large-scale or as deep-seated a social and cultural aversion to desegregation as there had been in the pre-1964 years.”2
The critical question for the right of association during the civil rights era was the extent to which it could justify private discrimination by whites against African Americans. Three important legal developments answered that question: (1) the Civil Rights Act of 1964; (2) the Court’s 1968 decision in Jones v. Alfred H. Mayer; and (3) the Court’s 1976 decision in Runyon v. McCrary.
Title II of the Civil Rights Act of 1964 prohibited racial discrimination in places of “public accommodation.” The legislation encompassed inns, restaurants, gas stations, and places of entertainment but exempted private clubs and other establishments “not in fact open to the public.” The act’s broad scope brought to the forefront an underlying tension between the clash of public and private interests. As Justice Goldberg recognized in his concurrence in Bell v. Maryland, a decision issued ten days prior to the enactment of the act, “a claim of equal access to public accommodations” against a restaurant “inevitably involves the liberties and freedoms both of the restaurant proprietor and of the Negro citizen.” But despite significant resistance from segregationists, the Civil Rights Act left no doubt in which direction that tension would be resolved. Five years later, the Court made clear in Daniel v. Paul that sham attempts to meet the private club exception would not prevail.3
The second important development defining the scope of the right of association in the civil rights era was the Court’s 1968 decision in Jones v. Alfred H. Mayer, which interpreted a Reconstruction era statute, the Civil Rights Act of 1866, to bar racial discrimination in the sale or lease of private property. The Court reasoned that the 1866 act reached even private discrimination because “the exclusion of Negroes from white communities” reflected “the badges and incidents of slavery” that the Thirteenth Amendment sought to remedy. It extended the reach of Jones to membership in a community park and playground in Sullivan v. Little Hunting Park and a private swimming pool in Tillman v. Wheaton-Haven Recreation Association. Jones, Sullivan, and Tillman all involved sales or leases related to real property covered under the Fair Housing Act of 1968. The Court’s reliance on a somewhat strained interpretation of the Civil Rights Act of 1866 rather than a straightforward application of the Fair Housing Act prompted Justice Harlan (joined by Justice White and Chief Justice Burger) to dissent in Sullivan, noting that the “vague and open-ended” construction of the 1866 act risked “grave constitutional issues should [that authority] be extended too far into some types of private discrimination.”4
These two developments—the Civil Rights Act of 1964 and the Court’s decision in Jones—both represented major steps toward ending segregation. Both also constrained group autonomy. But few people today object to these constraints along racial or any other lines—the idea that owners of businesses open to the public or sellers of private homes should have a constitutional right to discriminate finds few defenders. In other words, if the constraints on group autonomy ended with these applications, contemporary debates would be virtually nonexistent.
More complicated questions arose from the Court’s line of cases addressing private school segregation that culminated in its 1976 decision in Runyon v. McCrary. Preliminary challenges to private school segregation focused on government financial support, and in the late 1960s, the Court affirmed a number of decisions enjoining state tuition grants to students attending racially discriminatory private schools. In 1973, the Court concluded in Norwood v. Harrison that state-funded textbook loans to students attending these schools were “not legally distinguishable” from tuition grants.5
Norwood was the Court’s first explicit consideration of the conflict between antidiscrimination norms and the right of association. Summarizing recent legislative and judicial developments, Chief Justice Burger reasoned: “Invidious private discrimination may be characterized as a form of exercising freedom of association protected by the First Amendment, but it has never been accorded affirmative constitutional protections. And even some private discrimination is subject to special remedial legislation in certain circumstances.” Burger also noted that “although the Constitution does not proscribe private bias, it places no value on discrimination,” and simply because “the Constitution may compel toleration of private discrimination in some circumstances does not mean that it requires state support for such discrimination.”6
Shortly after Norwood, the justices addressed the use of public recreational facilities by private segregated schools in Gilmore v. City of Montgomery. Justice Blackmun’s majority opinion noted that, in contrast to the relatively easy question about integrating public facilities and programs, “the problem of private group use is much more complex.” The dispositive question was whether the use of public facilities made the government “a joint participant in the challenged activities.” The Court concluded that municipal recreational facilities, including parks, playgrounds, athletic facilities, amphitheaters, museums, and zoos, were sufficiently akin to “generalized governmental services” like traditional state monopolies such as electricity, water, and police and fire protection. Accordingly, the government’s acquiescence in the use of these facilities by private groups that discriminated on the basis of race did not rise to the level of government endorsement of discriminatory practices. But Blackmun went even further, noting that the exclusion of a discriminatory group from public facilities would violate the freedom of association. He asserted: “The freedom to associate applies to the beliefs we share, and to those we consider reprehensible,” and “tends to produce the diversity of opinion that oils the machinery of democratic government and insures peaceful orderly change.” At the same time, he cautioned that “the very exercise of the freedom to associate by some may serve to infringe that freedom for others. Invidious discrimination takes its own toll on the freedom to associate, and it is not subject to affirmative constitutional protection when it involves state action.”7
Two years later, in Runyon, the Court retreated from both its defense of the right of association and its state action requirement when it construed another provision of the Civil Rights Act of 1866 to bar racial discrimination by “private, commercially operated, nonsectarian schools.” Rejecting the suggestion that the legislation “[did] not reach private acts of racial discrimination,” Justice Stewart wrote: “From [the principle of the freedom of association] it may be assumed that parents have a First Amendment right to send their children to educational institutions that promote the belief that racial segregation is desirable, and that the children have an equal right to attend such institutions. But it does not follow that the practice of excluding racial minorities from such institutions is also protected by the same principle.” Stewart buttressed his argument with a truncated quotation from Norwood. Burger had written in Norwood that “although the Constitution does not proscribe private bias, it places no value on discrimination.” Stewart’s quotation omitted Burger’s prefatory clause and asserted: “As the Court stated in [Norwood], ‘the Constitution . . . places no value on discrimination.’” The abbreviated language stood for a broader legal principle. Norwood had prevented government subsidization of a disfavored social practice. Runyon precluded the practice itself and marked the first time that the Court had denied the right of existence to a private group neither with ties to state action nor meeting the definition of a public accommodation.8
Runyon’s symbolic importance is beyond challenge. The decision made clear that the Court understood the Civil Rights Act of 1866 “to reach all intentional racial discrimination, public and private, that interfered with the right to contract,” and that it trumped the right of association. That core holding has been undisturbed—it was, in fact, codified in the Civil Rights Act of 1991. Few people today believe that private schools ought to have a constitutional right to exclude African Americans.9
Runyon’s doctrinal significance is less clear, and it is on this doctrinal level that the case maintains its greatest significance for contested questions of group autonomy today. Two moves in particular are open to question, both of which the Court adopted eight years later in the much different context of Roberts. The first is Stewart’s argument that forced inclusion of unwanted members would not change the core expression of a discriminatory group: “there is no showing that discontinuance of [the] discriminatory admission practices would inhibit in any way the teaching in these schools of any ideas or dogma.” Setting aside the political and moral context of Runyon, the claim is not persuasive.10
Stewart’s second questionable doctrinal move is his artificial distinction between the act of discrimination and the message of discrimination. In Stewart’s view, the right of association protected only the latter, and the exclusion of African Americans counted only as the former. In other words, the right of association extended only to the expression of ideas, and exclusion wasn’t expression. But that argument makes an arbitrary distinction between act and message that could be applied to any form of expression—burning a flag or a draft card, for example. It tells us nothing about the value or harm of the expression itself.11
Association and Privacy
The clash between integration and the right to exclude paralleled a line of cases that emphasized a wholly different aspect of associational freedom: privacy. Frankfurter and Douglas had linked association and privacy in cases during the national security era. And Harlan had referred to “the vital relationship between freedom to associate and privacy in one’s associations” in NAACP v. Alabama. But the connection deepened after the Court recognized a constitutional right to privacy in Griswold v. Connecticut. Because privacy, like association, appeared nowhere in the text of the Constitution, the Court’s earlier recognition of the right of association in NAACP v. Alabama became an important example of the kind of “penumbral” reasoning that justified the right of privacy in Griswold.12
There was, however, a definitional problem with the meaning of privacy in the context of association. Brandeis and Warren’s classic definition of the right “to be let alone” in their 1890 law review article conveyed a sense of individual autonomy. But references to privacy in the association cases during the national security era had more to do with protecting group autonomy than with endorsing individual autonomy. As Harlan had argued in NAACP v. Alabama, “Inviolability of privacy in group association may in many circumstances be indispensable to preservation of freedom of association, particularly where a group espouses dissident beliefs.” The kind of privacy envisioned by the Court in NAACP v. Alabama did not mean not public; to the contrary, groups like the NAACP and the Communist Party had actively sought public visibility and recognition. Before Griswold, privacy in the context of association existed largely to facilitate public and political actions rather than to protect secret or intimate actions.13
Griswold struck down a Connecticut law that prohibited the use of contraceptives and the giving of medical advice about their use, specifically the application of this law to the use of contraceptives by married persons. Warren assigned the opinion to Douglas. In a draft that he shared only with Brennan, Douglas made scant reference to a right of privacy and rested his argument almost entirely on the First Amendment freedom of association. Douglas argued that while marriage did “not fit precisely any of the categories of First Amendment rights,” it was “a form of association as vital in the life of a man or woman as any other, and perhaps more so.” He reasoned that “we would, indeed, have difficulty protecting the intimacies of one’s relations to [the] NAACP and not the intimacies of one’s marriage relation.”14
After reviewing the draft, Brennan urged Douglas to abandon his exclusive reliance on the right of association. Brennan argued that marriage did not fall within the kind of association that the Court had recognized for purposes of political advocacy. He suggested that Douglas instead analogize the Court’s recognition of the right of association to a similar broadening of privacy into a constitutional right. Because neither privacy nor association could be found in the text of the Constitution, if association could be recognized as a freestanding constitutional right, then so could privacy. Douglas followed Brennan’s suggestions and wrote that the “specific guarantees in the Bill of Rights have penumbras, formed by emanations from those guarantees that help give them life and substance.”15
In addition to its recognition of privacy, Douglas’s final opinion also contained some extended language about the constitutional source of the freedom of association. In locating one of the penumbras of privacy in the First Amendment, Douglas wrote:
In NAACP v. Alabama, we protected the “freedom to associate and privacy in one’s associations,” noting that freedom of association was a peripheral First Amendment right. Disclosure of membership lists of a constitutionally valid association, we held, was invalid “as entailing the likelihood of a substantial restraint upon the exercise by [the NAACP’s] members of their right to freedom of association.” In other words, the First Amendment has a penumbra where privacy is protected from governmental intrusion. In like context, we have protected forms of “association” that are not political in the customary sense but pertain to the social, legal, and economic benefit of the members [citing NAACP v. Button].
In a dissenting opinion issued just a month prior to Griswold, Douglas had referred to a singular “right of assembly and association.” But now he argued that NAACP v. Alabama and Button “involved more than the ‘right of assembly.’” Instead: “The right of ‘association,’ like the right of belief, is more than the right to attend a meeting; it includes the right to express one’s attitudes or philosophies by membership in a group or by affiliation with it or by other lawful means. Association in that context is a form of expression of opinion; and while it is not expressly included in the First Amendment its existence is necessary in making the express guarantees fully meaningful.”16
Douglas’s conception of the right of assembly as no more than “the right to attend a meeting” in Griswold departed from his past descriptions of that right. Having thus confined assembly, Douglas suggested that the right of association was “necessary in making the express guarantees [of the First Amendment] fully meaningful.” But there was no reason that meaningful protections of assembly required a separate right of association. The Court had long since set forth the broad contours of the rights of speech and assembly in Thomas v. Collins: “If the exercise of the rights of free speech and free assembly cannot be made a crime, we do not think this can be accomplished by the device of requiring previous registration as a condition for exercising them and making such a condition the foundation for restraining in advance their exercise and for imposing a penalty for violating such a restraining order. So long as no more is involved than exercise of the rights of free speech and free assembly, it is immune to such a restriction.” Douglas, in fact, had quoted this language in his 1961 dissent in Communist Party v. Subversive Activities Control Board, adding that “the vices of registration [of an organization] may be not unlike those of licensing.” Yet despite his repeated arguments against this kind of prior restraint in the area of free speech, he failed to make the same connection with assembly.17
Douglas nevertheless maintained an important understanding of association in Griswold that would be lost a decade later in Stewart’s instrumental characterization in Runyon. He argued that the right of association “includes the right to express one’s attitudes or philosophies by membership in a group or by affiliation with it or by other lawful means.” In other words, as he had argued in a dissent four years earlier, “joining is one method of expression.” For Douglas, the act of association was itself an intrinsically valuable form of expression. For Stewart, it became merely an instrumental means of facilitating expression.18
Douglas’s reasoning in Griswold failed to convince all of his colleagues. Harlan “fully agree[d] with the judgment of reversal” but rejected the incorporation argument that he saw as implicit in Douglas’s insistence that “the Due Process Clause of the Fourteenth Amendment does not touch this Connecticut statute unless the enactment is found to violate some right assured by the letter or penumbra of the Bill of Rights.” Harlan based his objection on the now familiar liberty argument: “The proper constitutional inquiry in this case is whether this Connecticut statute infringes the Due Process Clause of the Fourteenth Amendment because the enactment violates basic values ‘implicit in the concept of ordered liberty.’ ” Black also disagreed with Douglas’s penumbral argument. His dissent lamented:
One of the most effective ways of diluting or expanding a constitutionally guaranteed right is to substitute for the crucial word or words of a constitutional guarantee another word or words, more or less flexible and more or less restricted in meaning. This fact is well illustrated by the use of the term “right of privacy” as a comprehensive substitute for the Fourth Amendment’s guarantee against “unreasonable searches and seizures.” “Privacy” is a broad, abstract and ambiguous concept which can easily be shrunken in meaning but which can also, on the other hand, easily be interpreted as a constitutional ban against many things other than searches and seizures. I have expressed the view many times that First Amendment freedoms, for example, have suffered from a failure of the courts to stick to the simple language of the First Amendment in construing it, instead of invoking multitudes of words substituted for those the Framers used.
Black’s words were odd in light of his repeated endorsement of the right of association, which had certainly been a failure “to stick to the simple language of the First Amendment in construing it.” Moreover, as the Court’s association cases in the national security era had shown, substituting a new right of association for the right of assembly had proven “one of the most effective ways of diluting or expanding” the constitutional protections for communists and civil rights activists, respectively.19
In 1972, the Court extended Griswold’s holding to unmarried persons desiring access to contraception. Brennan’s majority opinion in Eisenstadt v. Baird relied heavily on Griswold but not on Douglas’s reasoning. In Griswold, Douglas had maintained that part of the right to privacy rested on the “association” of marriage: “We deal with a right of privacy older than the Bill of Rights—older than our political parties, older than our school system. Marriage is a coming together for better or for worse, hopefully enduring, and intimate to the degree of being sacred. It is an association that promotes a way of life, not causes; a harmony in living, not political faiths; a bilateral loyalty, not commercial or social projects. Yet it is an association for as noble a purpose as any involved in our prior decisions.” This relational focus may have drawn an unlikely connection between a married couple and the NAACP, but it resisted the kind of individualism that equated associational privacy with “the privacy of private life.” In Eisenstadt, Brennan shifted the focus away from Douglas’s emphasis on the marriage relationship: “It is true that, in Griswold, the right of privacy in question inhered in the marital relationship. Yet the marital couple is not an independent entity, with a mind and heart of its own, but an association of two individuals, each with a separate intellectual and emotional makeup. If the right of privacy means anything, it is the right of the individual, married or single, to be free from unwarranted governmental intrusion into matters so fundamentally affecting a person as the decision whether to bear or beget a child.” Brennan’s language thus converted an understanding of associational freedom rooted in relationships between people to a right of individual autonomy. As H. Jefferson Powell argues, “Brennan’s reading of Griswold turned Douglas’s reasoning on its head” and signaled “the identification of a radically individualistic liberalism as the moral content of American constitutionalism.” Ironically, Brennan’s reasoning drew upon the liberty argument that Harlan had advanced in NAACP v. Alabama and other cases (including his Griswold concurrence). The right of privacy utterly detached from the right of association had no First Amendment basis; it came rather from the “liberty” of the Due Process Clause of the Fourteenth Amendment—exactly where Harlan had argued against Brennan that the right of association was itself located.20
The Rise of Rawlsian Liberalism
As the Court and commentators applied the new right of association in civil rights and privacy decisions, John Rawls introduced to political and legal discourse a theoretical resource with important implications for these developments. The appearance of Rawls’s Theory of Justice in 1971 breathed new life into the discipline of political theory, which had increasingly been exiled from political science by the behavioralism of postwar pluralists. But while Rawls came to be viewed as a kind of normative antidote to the ostensibly descriptive pluralist claims that had ruled political science in the 1950s and 1960s, his basic framework echoed many pluralist assumptions. As John Gunnell has written, “the new pluralism is, in many respects, not the same as the old pluralism . . . but it is, at bottom, the same theory.” The continuity is particularly evident with respect to questions about group autonomy. Pluralist political thought insisted on a consensus bounded by shared democratic values; Rawlsian liberalism presumed an “overlapping consensus” in which egalitarianism rooted in an individualist ontology trumped and thus bounded difference. Pluralists attributed harmony and balance to group interaction to explain the relative stability that they perceived; Rawls feared a loss of stability and made the preservation of peaceful interactions a cornerstone of his normative theory. Like the pluralist assumptions that preceded them, the Rawlsian premises of consensus and stability pervaded political discourse and influenced the ways in which the equality era reshaped the right of association.21
Rawls’s theory was self-avowedly motivated by a concern for political stability that could avoid the kind of sectarian religious violence that followed the European Reformation in the sixteenth and seventeenth centuries. Rawls believed that this stability could be attained through the “well-ordered society,” that is, “a society effectively regulated by a public political conception of justice.” In this society, “everyone accepts, and knows that everyone else accepts, the very same principles of justice.” This agreement could be reached without “the oppressive use of state power.” Rawls initially asserted that citizens, in spite of their differences, could pursue a common understanding of justice from an “Archimedean point . . . by assuming certain general desires, such as the desire for primary social goods, and by taking as a basis the agreements that would be made in a suitably defined initial situation.” He later came to believe that liberal society could never overcome the interminable disagreement that flowed from what he called “conflicting and incommensurable doctrines.” But he insisted that we might nonetheless attain political stability that was more than a mere modus vivendi. Rawls believed that while “reasonable pluralism” permitted “a diversity of reasonable comprehensive doctrines,” we could discover an “overlapping consensus” about justice from among these comprehensive doctrines by constraining dialogue to “public reason.” He thought that the overlapping consensus of reasonable belief would produce agreement over the “basic structure” and the “primary social goods” of society, which include rights, liberties, opportunities, income and wealth, and self-respect.22
Rawls included the freedom of association among his “basic liberties.” The freedom of association was related to what he called “private society,” which “is not held together by a public conviction that its basic arrangements are just and good in themselves.” As a result, “there are many types of social union and from the perspective of political justice we are not to try to rank them in value.” In fact, “a well-ordered society, and indeed most societies, will presumably contain countless social unions of many different kinds.” Importantly, “government has no authority to render associations either legitimate or illegitimate any more than it has this authority in regard to art or science.”23
Yet at the same time, Rawls’s vision for stability depended on consensus, and consensus could only be reached by constraining certain modes of discourse through what Rawls called “public reason.” He maintained that the requirement of public reason enabled consensus because political views could be detached from comprehensive doctrines: “We always assume that citizens have two views, a comprehensive and a political view; and that their overall view can be divided into two parts, suitably related.” (Years later, in response to numerous critics, Rawls cryptically suggested that the requirement of public reason “still allows us to introduce into political discussion at any time our comprehensive doctrine, religious or nonreligious, provided that, in due course, we give properly public reasons to support the principles and policies our comprehensive doctrine is said to support.”)24
Rawls may not have been cited in the legal decisions that reshaped the freedom of association in the first part of the equality era, but his influence was close at hand. Legal academics eager to provide intellectual cover to the Warren Court’s decisions and its recognition of fundamental rights not found in the text of the Constitution embraced his framework. In 1969, Frank Michelman’s foreword in the Harvard Law Review adopted a Rawlsian framework for analyzing income and wealth inequality. Eight years later, Kenneth Karst’s foreword employed a Rawlsian approach to conclude that the “substantive core” of the Fourteenth Amendment was “a principle of equal citizenship, which presumptively guarantees to each individual the right to be treated by the organized society as a respected, responsible, and participating member.” The Rawlsian influence in the legal academy did not go unchallenged, and Laura Kalman suggests that “in the end, Rawls proved helpful only to legal scholars predisposed toward political liberalism who were looking for a way to justify its continuance.” But in the first part of the equality era, those scholars held significant sway in the law schools and on the courts.25
One of the most important legal scholars shaped by Rawlsian premises was Ronald Dworkin. Dworkin’s legal theory made explicit an important assumption underlying Rawls’s theory of justice: individual rights prevailed over majoritarian democracy. The “constitutional conception” of democracy held out “rights as trumps” that limited majoritarian preferences when they constrained fundamental values like “equal concern and respect.” This meant that “a society in which the majority shows contempt for the needs and prospects of some minority is illegitimate as well as unjust.” Dworkin’s theory also exposed (and replicated) a tension inherent in Rawls’s theory of justice: accepting the “counter-majoritarian difficulty” presupposed an agreement about liberal values. But once illiberal minorities laid claim to fundamental liberal rights, the conflict between competing liberal claims became unavoidable: members of a group engaging in illiberal practices could consistently claim the liberal right to group autonomy.26 Unlike Tocqueville and Madison, Rawls and Dworkin never really recognized the value of protecting antimajoritarian groups on nonideological grounds. For Madison and Tocqueville, group autonomy was a boundary marker that didn’t engage in a substantive weighing of values. For Rawls and Dworkin, group autonomy and freedom of association were conditioned by equality, self-respect, and other liberal values.
Roberts v. United States Jaycees
The influence of Rawlsian liberalism and the two strands of case law that emerged over the right to exclude and the right to privacy coalesced in Roberts v. United States Jaycees, the most important case on the freedom of association in the equality era. In a sweeping decision with significant consequences for associational freedom, the Court simultaneously endorsed the implicit connection between privacy and association and severely curtailed the right to exclude.
The background to Roberts began in 1974 and 1975, when the Minneapolis and St. Paul chapters of the Jaycees began admitting women as regular members, in violation of the national organization’s bylaws. According to the national organization, women could be “associate individual members” who were ineligible to vote, hold office, or receive certain national awards but could “otherwise participate fully in Jaycee activities.” After the national organization threatened to revoke their charters, the two Minnesota chapters filed sex discrimination charges with the Minnesota Department of Human Rights based on the Minnesota Human Rights Act, which declared that it was an unfair discriminatory practice “to deny any person the full and equal enjoyment of the goods, services, facilities, privileges, advantages, and accommodations of a place of public accommodation because of race, color, creed, religion, disability, national origin, or sex.” In response, members of the national organization filed suit, alleging that the act violated their rights of speech and association.27
The Supreme Court upheld the constitutionality of the act without a dissent. Justice Brennan’s majority opinion asserted that previous decisions had identified two separate constitutional sources for the right of association. One line of decisions protected “intimate association” as “a fundamental element of personal liberty.” Another set of decisions guarded “expressive association,” which was “a right to associate for the purpose of engaging in those activities protected by the First Amendment—speech, assembly, petition for the redress of grievances, and the exercise of religion.” Expressive association to pursue “a wide variety of political, social, economic, educational, religious, and cultural ends” was “implicit in the right to engage in activities protected by the First Amendment.”28
The constitutional hooks for Brennan’s categories of intimate and expressive association roughly tracked the liberty argument and the incorporation argument. But in an odd doctrinal twist, the intimate association corresponding to the liberty argument now commanded greater constitutional protection than the expressive association corresponding to the incorporation argument, a reversal of the positions debated on the Court during the national security era. Brennan contended that intimate and expressive association represented, respectively, the “intrinsic and instrumental features of constitutionally protected association.” These differences meant that “the nature and degree of constitutional protection afforded freedom of association may vary depending on the extent to which one or the other aspect of the constitutionally protected liberty is at stake in a given case.”29
Brennan began his analysis by considering whether the Jaycees was an intimate association and announced that “several features of the Jaycees clearly place the organization outside of the category of relationships worthy of this kind of constitutional protection.” In the second section of his opinion, Brennan concluded that the Jaycees was an expressive association. He appeared to recognize the significance of the consequences of the Minnesota law to the Jaycees: “There can be no clearer example of an intrusion into the internal structure or affairs of an association than a regulation that forces the group to accept members it does not desire. Such a regulation may impair the ability of the original members to express only those views that brought them together. Freedom of association therefore plainly presupposes a freedom not to associate.” And in a critical comment, Brennan noted that “according protection to collective effort on behalf of shared goals is especially important in preserving political and cultural diversity and in shielding dissident expression from suppression by the majority.” The sentiment could have come straight from Madison and Tocqueville (absent their pluralist gloss). It reflected the importance of dissenting groups that the freedom of assembly had once recognized.30
Brennan quickly downplayed these concerns in light of “Minnesota’s compelling interest in eradicating discrimination against its female citizens.” He reasoned that Minnesota furthered its compelling interest by assuring women equal access to the leadership skills, business contacts, and employment promotions offered by the Jaycees. Because the Jaycees’s willingness to admit women as associate individual members presumably already afforded them most of these opportunities—the associate status precluded voting, holding office, and eligibility for national awards—it is unclear how forced admission of women as full members helped to eradicate gender discrimination in Minnesota. But even more troubling than Brennan’s failure to link remedy and harm was his claim that the forced integration of women would have no effect on the expressive interests of the Jaycees. There was, according to Brennan, “no basis in the record for concluding that admission of women as full voting members [would] impede the organization’s ability to engage in . . . protected activities or to disseminate its preferred views.”31
Justice O’Connor’s oft-cited concurrence is sometimes viewed more favorably than Brennan’s majority opinion. Contrary to Brennan, O’Connor viewed expressive association as more than instrumentally valuable. She asserted that “protection of the association’s right to define its membership derives from the recognition that the formation of an expressive association is the creation of a voice, and the selection of members is the definition of that voice.” If the Jaycees was in fact an expressive association, O’Connor believed it would be entitled to protection from intrusion by the state’s antidiscrimination legislation.32
Rather than distinguishing between expressive and nonexpressive associations, O’Connor instead proposed drawing a line between predominantly expressive and predominantly commercial organizations. She acknowledged that while the Jaycees was not a political organization, “the advocacy of political and public causes, selected by the membership, is a not insubstantial part of what it does.” Nevertheless, she reasoned that the Jaycees’s attention to and success in membership drives meant that it was, “first and foremost, an organization that, at both the national and local levels, promotes and practices the art of solicitation and management.” Accordingly, “the State of Minnesota ha[d] a legitimate interest in ensuring nondiscriminatory access to the commercial opportunity presented by membership in the Jaycees.” For these reasons, the Jaycees presented for O’Connor a “relatively easy case for application of the expressive-commercial dichotomy.” Elaborating upon her reasoning, O’Connor explained that “an association should be characterized as commercial, and therefore subject to rationally related state regulation of its membership and other associational activities, when, and only when, the association’s activities are not predominantly of the type protected by the First Amendment. It is only when the association is predominantly engaged in protected expression that state regulation of its membership will necessarily affect, change, dilute, or silence one collective voice that would otherwise be heard.”33
O’Connor’s focus on the commercial nature of groups as a boundary line for associational protections holds some promise, but her reasoning is problematic on three counts. First, O’Connor posits a false dichotomy between commercial and expression associations: some commercial associations are expressive and (at least under the Court’s understanding of the expressive and nonexpressive divide) some noncommercial organizations are not expressive. Second, her requirement that an association be “predominantly engaged” in protected expression to avoid being classified as commercial leaves vulnerable to regulation some groups that because of their size or unpopularity must devote a substantial portion of their activities to fundraising or other commercial activities. Finally, O’Connor leaves unclear which activities are “of the type protected by the First Amendment.”34
The Artificiality of Intimate and Expressive Association
The most serious doctrinal shortcoming in Roberts was Brennan’s creation of the categories of intimate and expressive association. Brennan’s arguments implied two corollaries: (1) some associations were “non-intimate,” and (2) some associations were “non-expressive.” His reasoning thus suggested four possible categories of associations: (1) intimate expressive associations, (2) intimate nonexpressive associations, (3) nonintimate expressive associations, and (4) nonintimate nonexpressive associations. Since Roberts, it has become clear that there is no constitutionally significant distinction between these first two categories. Intimate associations receive the highest level of constitutional protection regardless of whether they are also expressive.
The same is not true for the distinctions between the other categories. Brennan’s parsing of intrinsic and instrumental value and his reference to the varying “nature and degree of constitutional protection” for intimate and expressive associations signaled a clear privileging of the former over the latter. And the category of expressive association drew a line that left nonintimate nonexpressive associations without any meaningful constitutional protections. The Roberts framework thus created the following hierarchically ordered categories of associations:
A) intimate associations
B) nonintimate expressive associations
C) nonintimate nonexpressive associations
It turns out that the groups in B sometimes lose, and the groups in C always lose. Yet despite these consequences, neither intimate nor expressive association represents a constitutionally defensible drawing of lines.35
The category of intimate association likely originated in a 1980 article by Kenneth Karst in the Yale Law Journal, “The Freedom of Intimate Association.” Karst’s article sought to recover the relational emphasis in Griswold that Brennan had abandoned in Eisenstadt. He began by noting that Douglas had focused specifically on the association of marriage. Karst contended that this language had established a freedom of “intimate association,” which he suggested was “a close and familiar personal relationship with another that is in some significant way comparable to a marriage or family relationship.” The problem with Karst’s argument was its implicit corollary that some groups were “non-intimate associations”—and that a constitutionally significant line could be drawn between intimate and nonintimate ones. The argument fails for the simple reason that all of the values, benefits, and attributes that it assigns to intimate associations are equally applicable to many if not most nonintimate associations.36
Karst at times recognized the broader applicability of his claims. He noted that “an intimate association, like any group, is more than the sum of its members; it is a new being, a collective identity with a life of its own.” And near the end of his article, he wrote that “one of the points of any freedom of association must be to let people make their own definitions of community.” Yet despite these occasional concessions, Karst repeatedly placed special value on the relationships that form in intimate associations. For example, he emphasized the importance of “close friendship” in intimate association. For Karst, it was “plain that the values of intimate association may be realized in friendships involving neither sexual intimacy nor family ties,” and that “any view of intimate association focused on associational values must therefore include friendship.” He also tied intimate association to the kinds of bonds that form through personal interaction: the “chief value in intimate association is the opportunity to satisfy” the “need to love and be loved”; “the opportunity to be cared for by another in an intimate association is normally complemented by the opportunity for caring” that requires a “personal commitment”; “caring for an intimate requires taking the trouble to know him and deal with him as a whole person, not just as the occupant of a role,” which “limits the number of intimate associations any one person can have at any one time, or even in a lifetime.”37
Karst’s attention to friendship and personal bonds is eminently reasonable. But the potential for and existence of such close friendships can be found in many kinds of associations. While it may well be that attributes of friendship and personal bonds distinguish small or local groups from large and impersonal groups like behemoth mailing list organizations, surely many small associations that fall outside the bounds of intimacy are capable of producing “close friendships” of the kind that Karst describes. To be sure, some relationships between members of these groups will undoubtedly be superficial and “casual.” But this is also true of the relationships that constitute many intimate associations. Karst recognized that protecting the values he saw as inherent in intimate association required offering “some protection to casual associations as well as lasting ones.” In fact, “one reason for extending constitutional protection to casual intimate associations is that they may ripen into durable intimate associations.” Karst argued that “a doctrinal system extending the freedom of intimate association only to cases of enduring commitment would require intolerable inquiries into subjects that should be kept private, including states of mind.” It is hard to understand why these principles wouldn’t apply equally to many larger, nonintimate associations.38
Karst’s other attempts to mark the bounds of intimate association were similarly unavailing:
- “Transient or enduring, chosen or not, our intimate associations profoundly affect our personalities and our senses of self. When they are chosen, they take on expressive dimensions as statements defining ourselves.”
- “An intimate association may influence a person’s self-definition not only by what it says to him but also by what it says (or what he thinks it says) to others.”
- “When two people [voluntarily enter into an intimate association],they express themselves more eloquently, tell us more about who they are and who they hope to be, than they ever could do by wearing armbands or carrying red flags.”
- “First Amendment doctrine cautions us to be sensitive to the need to protect intimate associations that are unconventional or that may offend a majority of the community.”
Each of these claims applies with equal force if we remove the adjective “intimate.” Some associations and associative acts will lack significance for some people, but the extent to which expression, self-definition, and unconventional norms unfold in a group’s practices is not contingent upon whether the group is an intimate association.39
Some of the conceptual problems with Karst’s approach to intimate association likely arose because he wasn’t explicitly attempting to distinguish intimate from nonintimate associations. His focus appears to have been on trying to develop a category of intimate association as an alternative to the then-nascent right of privacy and in using the right of intimate association to advance legal protections for homosexual relationships. Today, these particular goals are unlikely to be advanced by the right of intimate association, as evidenced by Lawrence v. Texas, the Court’s overruling of its decision in Bowers v. Hardwick. Bowers had drawn two dissents, one from Justice Stevens that emphasized Griswold’s liberty arguments, and one from Justice Blackmun that drew upon Griswold’s connections between privacy and intimate association (and included two citations to Karst’s article). Lawrence relied on Stevens’s dissent and never mentioned the right of intimate association.40
The Court’s avoidance of intimate association in Lawrence suggests that the doctrinal value of the category may be minimal. But what about the doctrinal harm? The dangers of privileging intimate association become apparent in Brennan’s Roberts opinion. Brennan began by noting, “Certain kinds of personal bonds have played a critical role in the culture and traditions of the Nation by cultivating and transmitting shared ideals and beliefs; they thereby foster diversity and act as critical buffers between the individual and the power of the State.” This language attempts to draw the reader into a kind of Tocquevillean ethos in which intimate associations at once facilitate support for “the Nation” and resistance to “the State.”41
But Brennan’s argument lacks coherence and specificity. What is the difference between nation and state? What are the national culture (singular) and national traditions (plural) brought about by “shared ideals and beliefs”? How do personal bonds “foster diversity” and act as “critical buffers” from state power? More to the point, why are these functions unique to intimate associations? If Brennan’s argument is that intimate associations sustain some kind of shared culture (“cultivating and transmitting shared ideals and beliefs”), then why can’t nonintimate associations also serve as “schools of democracy”?42 Conversely, if Brennan means to position intimate associations as “mediating structures” between individuals and the state (“foster[ing] diversity and act[ing] as critical buffers”), then don’t some of the largest (and least “intimate”) groups have the greatest capacity to resist the state?43 The passage also belies a more troubling vagueness. It contains an irresolvable tension that doesn’t let the reader know whether Brennan is ultimately prioritizing the state, the nonstate group, or the individual, and the answer to that question matters a great deal. From the rest of his opinion and his broader jurisprudence, we might infer that Brennan wants to privilege the individual, then the state, and lastly, the group. But if that is where his argument rests, then some language—“critical buffers,” “traditions,” “shared ideals”—becomes much harder for him to employ in an unqualified sense.
Brennan next enlisted notions of liberty and autonomy in his defense of intimate association, embracing the individualistic gloss that his Eisenstadt opinion cast on Griswold: “The constitutional shelter afforded [intimate associations] reflects the realization that individuals draw much of their emotional enrichment from close ties with others. Protecting these relationships from unwarranted state interference therefore safe-guards the ability independently to define one’s identity that is central to any concept of liberty.” These phrases—emotional enrichment, defining one’s identity, and the concept of liberty—again call to mind lofty ideals, but their meaning is imprecise. As before, Brennan fails to explain why his reasoning extends only to intimate associations. People form close ties with others through all kinds of associations. Some lifelong friendships emerge from within nonintimate associations; some intimate associations collapse in a matter of months. Self-definition also comes from myriad forms of associations—one’s decision to join the ACLU or make a financial contribution to Greenpeace can speak volumes about one’s identity.44
Like Karst, Brennan fails to offer a convincing rationale for privileging intimate associations over nonintimate ones. His theoretical anchor is the residue of Eisenstadt that supplants the inherently relational aspects of association with an individualistic notion of privacy. Intimate association is reduced to intimate individualism.45
The second category that Brennan announced in Roberts was expressive association. He characterized expressive association as “for the purpose of engaging in those activities protected by the First Amendment— speech, assembly, petition for the redress of grievances, and the exercise of religion.” The Court had “long understood as implicit in the right to engage in activities protected by the First Amendment a corresponding right to associate with others in pursuit of a wide variety of political, social, economic, educational, religious, and cultural ends.”46
Despite his instrumental characterization of expressive association, Brennan proposed an ostensibly protective legal test: “Infringements on [the right of expressive association] may be justified by regulations adopted to serve compelling state interests, unrelated to the suppression of ideas, that cannot be achieved through means significantly less restrictive of associational freedoms.” The language of “compelling state interests, unrelated to the suppression of ideas,” calls to mind the strict scrutiny standard established in other areas of the Court’s First Amendment law. But notice how the reference to “means significantly less restrictive” differs from the usual strict scrutiny language of “least restrictive means.” On closer examination, what resembles a strict scrutiny test might actually invert the presumption favoring the protected First Amendment activity to one that favors the government. Brennan’s phrasing suggests that a government regulation “to a large extent but not significantly more restrictive of associational freedoms than a less onerous regulation” would survive the test. Although Brennan elsewhere intimated that he was applying strict scrutiny, his only formulation of the legal test proposed a different standard. Unsurprisingly, some courts have construed Roberts as intending something less than strict scrutiny.47
Brennan’s category of expressive association implied that some associations were “nonexpressive.” The problems with this line-drawing are not merely doctrinal—they are philosophical. The purported distinction between expressive and nonexpressive association fails to recognize: (1) that all associations have expressive potential; (2) that meaning is dynamic; and (3) that meaning is subject to more than one interpretation. These three claims rely on hermeneutical arguments whose full consideration exceeds the scope of this book, but I will address them briefly in the next chapter.48
Justice Brennan’s reasoning in Roberts has been roundly criticized. Nancy Rosenblum observed: “The Jaycees’ ‘voice’ was undeniably altered once it was forced to admit young women as full members along with young men.” Aviam Soifer contested: “Surely the Jaycees . . . will be a different organization. Surely that difference will be felt throughout an intricate web of relationships and different voices in immeasurable but nonetheless significant ways.” George Kateb suggested: “Brennan’s claim that young women may, after their compulsory admission, contribute to the allowable purpose of ‘promoting the interests of young men’ is absurd.” With an eye toward the legal effects of Roberts, Andrew Koppelman has noted that “the Court’s obliviousness to the obvious burden the antidiscrimination law imposed [on the Jaycees] is not reassuring about future applications of the Roberts rule,” and Jason Mazzone has cautioned that “the modern notion of ‘expression’ is a dubious peg on which to hang a constitutional right of free association.”49
Despite its critical reception in academic circles, Roberts opened a large hole in the already attenuated freedom of association, and the Court endorsed its reasoning in two subsequent cases involving private organizations that refused membership to women. In 1987, the Court held in Board of Directors of Rotary International v. Rotary Club of Duarte that the Rotary Club had no First Amendment right to exclude women. The following year, in New York State Club Association v. City of New York, the Court upheld antidiscrimination laws applied to a consortium of New York City social clubs. Justice White’s opinion narrowed the scope of expressive association by announcing that a group must demonstrate that it was “organized for specific expressive purposes” and that “it will not be able to advocate its desired viewpoints nearly as effectively if it cannot confine its membership” to certain classes of people. White emphasized that the right to associate was by no means absolute: it did not mean “that in every setting in which individuals exercise some discrimination in choosing associates, their selective process of inclusion and exclusion is protected by the Constitution.”50
In 1995, the Court reviewed a challenge from the Irish-American Gay, Lesbian and Bisexual Group of Boston (GLIB) over its exclusion from a Boston parade jointly commemorating St. Patrick’s Day and Evacuation Day. Since 1947, the parade had been organized by the South Boston Allied War Veterans Council, a private organization. GLIB challenged its exclusion from the parade under Massachusetts’s public accommodations law. Justice Souter’s opinion for the Court rejecting GLIB’s claim relied on free speech rather than free association. Souter first classified the parade as a form of expression. Because the organizers were private speakers, they were free to select the content of their message. Therefore, they could properly reject GLIB’s request to march in the parade. In fact, “whatever the reason” the parade organizers had for excluding GLIB, their decision “boils down to the choice of a speaker not to propound a particular point of view, and that choice is presumed to lie beyond the government’s power to control.” The free speech analysis seemed fairly straightforward, but it was difficult to reconcile with the Court’s approach to association in Roberts, Duarte, and New York State Club Association.51
The Court’s most significant case on the right of association after Roberts was its 2000 decision in Boy Scouts of America v. Dale. A 5–4 majority upheld the right of the Boy Scouts to exclude from their membership a homosexual scoutmaster against a challenge brought under a state anti-discrimination law. Chief Justice Rehnquist’s opinion for the Court began by placing the case within the framework of expressive association: “To determine whether a group is protected by the First Amendment’s expressive associational right, we must determine whether the group engages in ‘expressive association.’ The First Amendment’s protection of expressive association is not reserved for advocacy groups. But to come within its ambit, a group must engage in some form of expression, whether it be public or private.” Rehnquist distanced himself from some of the Court’s earlier views on expressive association. Although New York State Club Association appeared to have narrowed the right of expressive association to groups that were organized “for specific expressive purposes,” Rehnquist argued: “Associations do not have to associate for the ‘purpose’ of disseminating a certain message in order to be entitled to the protections of the First Amendment. An association must merely engage in expressive activity that could be impaired in order to be entitled to protection.” For Rehnquist, the proper inquiry was “whether the forced inclusion of Dale as an assistant scoutmaster would significantly affect the Boy Scouts’ ability to advocate public or private viewpoints.” And this inquiry required that the Court defer to an organization’s purported views: “It is not the role of the courts to reject a group’s expressed values because they disagree with those values or find them internally inconsistent. As is true of all expressions of First Amendment freedoms, the courts may not interfere on the ground that they view a particular expression as unwise or irrational.”52
These strong words hardly seemed credible after Runyon and Roberts. Justice Stevens’s dissent highlighted the doctrinal tension that Dale created: “Until today, we have never once found a claimed right to associate in the selection of members to prevail in the face of a State’s antidiscrimination law. To the contrary, we have squarely held that a State’s antidiscrimination law does not violate a group’s right to associate simply because the law conflicts with that group’s exclusionary membership policy.” Moreover, while Dale’s holding favored the Boy Scouts, the Court reaffirmed the fundamental division between intimate and expressive association in Roberts.53
The Continued Costs of Intimate and Expressive Association
Since Dale, two decisions have potently illustrated the dangers of the framework of intimate and expressive association. The first is the Second Circuit’s decision in Chi Iota v. City University of New York. Alpha Epsilon Pi (“AEPi”) is a national social fraternity founded in 1913 “to provide opportunities for the Jewish college man seeking the best possible college and fraternity experience.” According to its Supreme Constitution, AEPi seeks “to promote and encourage among its members: Personal perfection, a reverence for God and an honorable life devoted to the ideal of service to all mankind; lasting friendships and the attainment of nobility of action and better understanding among all faiths.”54
In 2002, the Chi Iota Colony (“Chi Iota”) of AEPi formed at the College of Staten Island, a primarily commuter campus of just more than eleven thousand undergraduates. Since its inception, Chi Iota never had more than twenty members. Its past president described the purpose of the fraternity as fostering a “lifelong interpersonal bond termed brotherhood,” which “results in deep attachments and commitments to the other members of the Fraternity among whom is shared a community of thoughts, experiences, beliefs and distinctly personal aspects of their lives.” In furtherance of those goals, the fraternity limited its membership to males. Chi Iota applied to be chartered and officially recognized by the College of Staten Island in March 2004. The director of the Office of Student Life denied the application on the basis that the fraternity’s exclusion of women violated the college’s nondiscrimination policy. The denial of official recognition precluded Chi Iota from using the college’s facilities, resources, and funding, from using the college’s name in conjunction with the organization’s name, and from posting events to the college’s calendars.55
In 2005, the members of Chi Iota filed suit in the United States District Court for the Eastern District of New York, arguing violations of their rights to intimate and expressive association and equal protection. The district court granted the fraternity’s motion for a preliminary injunction against the college on its intimate association claim but concluded that Chi Iota had not shown a clear or substantial likelihood of success on its expressive association claim. On appeal, the United States Court of Appeals for the Second Circuit reversed the district court’s grant of a preliminary injunction and remanded the case, noting that the fraternity’s “interests in intimate association are relatively weak.” Although the district court would still have had Chi Iota’s intimate and expressive association claims before it on remand, neither looked to have a reasonable chance of success given the posture of the litigation. As the Second Circuit was considering the case, the Chi Iota Colony of the Alpha Epsilon Pi Fraternity at the College of Staten Island disbanded.56
Chi Iota is not the most sympathetic plaintiff to bring a freedom of association claim. Although its Jewish roots suggest religious freedom interests, most of its members were nonpracticing Jews. It was a social group, but some of its social activities were coarse and banal, including visits to strip clubs. It may well be that the brothers of Chi Iota were a self-focused, hedonistic group of boys who brought a collective drain on whatever community existed at the mostly commuter campus at the College of Staten Island. But all of this is beside the point. A group’s protections shouldn’t turn on whether its purposes or activities are sincere or wholesome to an outsider’s perspective. Chi Iota’s practices and activities meant something to its members. They meant enough for the brothers to pursue membership through an application and rush process, to participate in the group’s activities, and to bring a federal lawsuit in an attempt to preserve their associational bonds.57
The most recent case to illustrate the dangers of the weakened framework for associational freedom claims is the Supreme Court’s decision in Christian Legal Society v. Martinez. The Christian Legal Society (CLS) is a “nationwide association of lawyers, law students, law professors, and judges who profess faith in Jesus Christ.” Founded in 1961, its purposes include “providing a means of society, fellowship, and nurture among Christian lawyers; encouraging, discipling, and aiding Christian law students; promoting justice, religious liberty, and biblical conflict resolution; and encouraging lawyers to furnish legal services to the poor.” The society maintains student chapters at many law schools around the country. These student chapters invite anyone to participate in their events but require members (including officers) to sign a “statement of faith” consistent with the Evangelical Protestant and Catholic traditions. Part of this statement of faith affirms that sexual conduct should be confined to heterosexual marriage. Accordingly, CLS student chapters do not accept as members anyone who engages in or affirms the morality of sex outside heterosexual marriage.58
In 2004, the CLS chapter at Hastings College of the Law in San Francisco inquired about becoming a recognized student organization. Hastings officials withheld recognition on the basis that CLS’s statement of faith violated the religion and sexual orientation provisions of the school’s Nondiscrimination Policy. As a result, the school denied CLS travel funds and funding from student activity fees. It also denied the society the use of the school’s logo, use of a Hastings e-mail address, the opportunity to send mass e-mails to the student body, participation in the annual student organizations fair, and reserved meeting spaces on campus. Hastings subsequently asserted that its denial of recognition stemmed from an “accept-all-comers” policy that required all student organizations to accept any student who desired to be a member of the organization.59
The society filed suit in federal district court, asserting violations of expressive association, free speech, free exercise of religion, and equal protection. The district court granted summary judgment against CLS on all of its claims. With respect to CLS’s expressive association claim, the district court concluded that Roberts and Dale were inapplicable because “CLS is not being forced, as a private entity, to include certain members or officers” and “the conditioned exclusion of [an] organization from a particular forum [does] not rise to the level of compulsive membership.” The district court also asserted that “Hastings has denied CLS official recognition based on CLS’s conduct—its refusal to comply with Hastings’ Nondiscrimination Policy—not because of CLS’s philosophies or beliefs.”60
Despite resting its holding on the inapplicability of Roberts and Dale, the court held in the alternative that CLS’s claim failed under those authorities as well. It assumed that CLS qualified as an expressive association because Hastings did not dispute that characterization. But the court determined that “CLS has not demonstrated that its ability to express its views would be significantly impaired by complying with [the school’s nondiscrimination] requirement.” The court concluded: “Unlike the Boy Scouts in Dale, CLS has not submitted any evidence demonstrating that teaching certain values to other students is part of the organization’s mission or purpose, or that it seeks to do so by example, such that the mere presence of someone who does not fully comply with the prescribed code of conduct would force CLS to send a message contrary to its mission.” In fact, there was “no evidence” that “a non-orthodox Christian, gay, lesbian, or bisexual student” who became a member or officer of CLS “by their presence alone, would impair CLS’s ability to convey its beliefs.”61
The society appealed the district court’s decision to the United States Court of Appeals for the Ninth Circuit. The appellate court affirmed the district court with a terse two-sentence opinion: “The parties stipulate that Hastings imposes an open membership rule on all student groups—all groups must accept all comers as voting members even if those individuals disagree with the mission of the group. The conditions on recognition are therefore viewpoint neutral and reasonable.” CLS petitioned for a writ of certiorari to the United States Supreme Court, arguing, among other things, that the Ninth Circuit’s decision created a circuit split with a Seventh Circuit case invalidating the denial of official recognition to a CLS student chapter at the Southern Illinois University School of Law.62
A divided Supreme Court rejected CLS’s challenge. Justice Ginsburg’s majority opinion concluded that Hastings’s all-comers policy was “a reasonable, viewpoint-neutral condition on access to the student-organization forum.” The majority’s free speech analysis is not entirely persuasive—its reasoning obscures a tension between the viewpoint neutrality of the all-comers policy (under a public forum analysis) and Hastings’s nonneutral policy preferences expressed through its own speech and subsidies (under something akin to a government speech analysis). But in the context of this book, the majority’s failure to take seriously CLS’s freedom of association claim is even more disturbing.63
From the premise that it “makes little sense to treat CLS’s speech and association claims as discrete,” Ginsburg concluded that the Court’s “limited-public-forum precedents supply the appropriate framework for assessing both CLS’s speech and association rights.” The problem with this doctrinal move is twofold. First, it essentially elects rational basis scrutiny over strict scrutiny—and therefore all but preordains the outcome. Second, it casts aside the competing constitutional values underlying associational freedom. CLS’s associational claim highlights the underlying values conflict in this case: the clash between group autonomy and equality—the same tension at issue in Runyon and Roberts. Taking this values clash seriously means refusing to make an artificial distinction between expression and conduct and recognizing that, in some cases, they are one and the same. Contrary to Ginsburg’s insistence that “CLS’s conduct—not its Christian perspective—is, from Hastings’ vantage point, what stands between the group and RSO status,” CLS’s “conduct” is inseparable from its message.64
Ginsburg’s opinion misses this connection. Quoting from CLS’s brief, she writes that “expressive association in this case is ‘the functional equivalent of speech itself’” to set up the idea that expressive association is entitled to no more constitutional protection than speech. But CLS had asserted: “Where one of the central purposes of a noncommercial expressive association is the communication of a moral teaching, its choice of who will formulate and articulate that message is treated as the functional equivalent of speech itself.” CLS wasn’t arguing that association is nothing more than speech but that association is itself a form of expression—who it selects as its members and leaders communicates a message. The society underscored this point elsewhere in its brief, arguing that “because a group’s leaders define and shape the group’s message, the right to select leaders is an essential element of its right to speak.” Ginsburg interpreted this assertion to mean that “CLS suggests that its expressive-association claim plays a part auxiliary to speech’s starring role.” That interpretation may be consistent with the Roberts understanding of expressive association, but as I have argued throughout this book, it misses the more fundamental connection between a group’s message and its composition.65
Ginsburg distinguished the Court’s associational cases like Dale and Roberts because those cases “involved regulations that compelled a group to include unwanted members, with no choice to opt out.” But this is really a matter of perspective. Sometimes a group must choose between receiving benefits and adhering to its policies at the cost of those benefits. But withholding some benefits (like access to meeting space or e-mail lists or the opportunity to be part of a public forum) can be akin to stamping out a group’s existence. After Martinez, the Hastings Christian Group That Accepts All Comers can exist, and the Christian Legal Society for Hastings Law Students That Can Sometimes Meet on Campus as a Matter of University Discretion if Space Is Available but Can’t Recruit Members at the Student Activities Fair can exist. But the Hastings Christian Legal Society—whose views and purposes are in no way sanctioned by and able to be explicitly disavowed by Hastings—cannot.66
At the close of the equality era, the right of association bore little resemblance to the right of assembly that had existed for almost two hundred years of our nation’s history. The confluence of a growing Civil Rights Movement and the dominance of Rawlsian liberalism meant that when principles of equality collided with group autonomy, equality won. An already attenuated right of association established during the national security era now gave way to even more incursions into group autonomy in the equality era. These developments were facilitated by an odd connection between association and privacy that produced a right of intimate association and, in doing so, jettisoned all other groups to a “nonintimate” status. By the end of the equality era, a right of association now quite removed from the right of assembly had fundamentally altered the protections for group autonomy. The next chapter explores these changes and their consequences by sketching a theory of assembly.