Support First Things by turning your adblocker off or by making a  donation. Thanks!

In New York City, about one hundred and fifty churches and religious groups—many of them serving poor and immigrant congregations—are threatened with eviction from public buildings where other community groups are permitted to rent space because they engage in “worship” and not just religious “speech.” Every group is welcome but not churches, mosques, and synagogues.

Just thirty years ago, the Supreme Court protected the rights of groups like these to meet on public property. In 1981, in Widmar v. Vincent, an eight-to-one majority squarely rejected the notion that a public university can exclude religious groups when it allows other groups to meet, and a series of subsequent decisions extended that equal-access principle to other public settings. Indeed, even before Widmar, courts protected the right of religious groups to worship on public property—to hold vigils or sunrise services in parks, for example—on an equal basis with other groups, without questioning their right to choose their pastors, priests, or leaders.

Why has the freedom of law-abiding groups to meet on public property eroded in recent years? In an important new book, Liberty’s Refuge: The Forgotten Freedom of Assembly, legal scholar John D. Inazu puts forward a surprisingly simple and persuasive explanation: The right to meet has eroded because the Supreme Court has neglected the explicit textual source of that right, the freedom of assembly (often called freedom of association, but that term is not used in the Constitution), and has substituted for it weaker protections under principles of freedom of speech.

The legal shift, he argues, occurred in two stages. First, the distinctive elements of freedom of assembly—the rights to use public space and to organize associations around shared beliefs—were subsumed under freedom of speech. Second, because freedom of speech doctrine is focused on the message and not on where it is spoken or who says it, government can discriminate between groups so long as its distinctions pass a toothless “reasonableness” test and do not overtly target a group on the basis of its message.

Ironically, in their otherwise sparse deliberations over the freedoms of speech, press, assembly, and petition, the drafters of the First Amendment made one thing clear: These freedoms are separate and warrant individual enumeration and protection. In the past thirty years, without offering any reason and without considering this history, the Supreme Court has committed the one error the drafters most clearly tried to prevent.

On August 15, 1789, the House of Representatives began its deliberations over a proposal—at that time, the Fourth Amendment—that “the freedom of speech and of the press, and the right of the people peaceably to assemble and consult for their common good, and to apply to the Government for redress of grievances, shall not be infringed.” Immediately, Theodore Sedgwick of Massachusetts moved that the Assembly Clause be struck out. He argued that, once the Constitution had secured the freedom of speech, a separate provision for freedom of assembly would be redundant. “If people freely converse together,” he said, “they must assemble for that purpose.” It is a “self-evident, unalienable right” that “would never be called in question.” There was no need to include it in the list of constitutionally protected rights.

John Page of Virginia responded that people have “been prevented from assembling together on their lawful occasions,” referring obliquely to William Penn’s infamous prosecution in 1670 on the charge of unlawful assembly after he preached a Quaker sermon to a crowd assembled on Gracechurch Street in London: “Therefore, it is well to guard against such stretches of authority, by inserting the privilege in the declaration of rights.” An explicit Assembly Clause would ensure that the people would not be “deprived of the power of assembling under any pretext whatsoever.”

Page’s reference to Penn was doubly suggestive. Penn’s sermon occurred on a public street, without official permission. And its subject matter was religious rather than political and therefore beyond the proposed protection for assemblies gathered to discuss the common good. Page also seemed to realize that governments even in the new American republic might be prone to restrict the freedom on various pretexts—which is why a separate enumeration would be valuable.

With Page’s reminder of the Penn prosecution in mind, Congress rejected Sedgwick’s motion by a large majority and later omitted the words “and consult for the common good,” thereby extending the right of assembly to those assembling for nonpolitical matters such as religion and, presumably, for social, cultural, and other purposes. The sole limitation was that the assemblies be peaceable. The right of the people peaceably to assemble in public spaces escaped being merged into a more general freedom of speech.

What difference did this make? Because there were few “assembly” cases—and indeed no “freedom of speech” cases in the Supreme Court either—for over one hundred years after adoption of the First Amendment, we have no authoritative adjudication of the differences between these two related freedoms. Inazu’s discussion of the history and text of the Clause strongly suggests two ways in which the right of assembly supplements the freedom of speech, arguing that the freedom of assembly has to include the right to use public spaces and the right of associations to plan their assemblies.

First, freedom of assembly by its nature involves public spaces, like the public street in which the founder of Pennsylvania preached. If crowds of people are to assemble, they must assemble somewhere. In early America that somewhere was ordinarily the village green, the commons, the streets, the wharves, or the parks, not just private spaces like taverns, coffeehouses, churches, and courthouses. A right of assembly without a right of access to public spaces would be an empty right.

Today, the courts treat the right to speak in public places as a branch of free-speech law called the “public forum doctrine.” The government may regulate speech on government property that has traditionally been available for public expression only when necessary to serve a compelling public purpose. But this is not a natural proposition. Although, as the Supreme Court recently explained in the 2011 decision Brown v. Entertainment Merchants Association, the “most basic” principle of freedom of speech is that, with certain very limited exceptions, “government lacks the power to restrict expression because of its message, ideas, subject matter, or content,” this principle logically carries no entitlement to use government resources. The government may prevent speakers from occupying public parks, streets, or commons to protect alternative uses of the land. Freedom of speech is ordinarily a negative freedom, not a positive claim on public property.

As Inazu points out, however, the Court’s earliest public-forum cases involved claims of freedom of assembly in addition to freedom of speech. In a 1939 decision, Hague v. CIO, the Court declared that public streets and parks have been used “immemorially” and “time out of mind” for “purposes of assembly, communicating thoughts between citizens, and discussing public questions.” The Court cited no authority for this assertion. In fact, the Court was wrong.

In Britain, the people were not free to assemble in the streets and parks without official permission. Unauthorized groups of twelve or more could be charged and prosecuted, like Penn, for unlawful assembly. Colonial governors tried to suppress the Sons of Liberty on similar legal bases. America’s declaration of a freedom of assembly was a break from this history, not a continuation of rights that had existed “time out of mind.” Nothing in the history of freedom of speech suggests any change in rights of access to public property. The creation of the freedom of assembly embodied this legal change in the right of access.

Second, freedom of assembly was understood to protect not only the assembly itself but also the right to organize assemblies through more or less continual associations and for those associations to select their own members by their own criteria. The Sons of Liberty’s public meetings were not purely spontaneous gatherings; they were planned, plotted, and led by men who shared a certain vision and met over a period of time, often secretly, to organize them. In this respect, the freedom of assembly is preparatory to the freedom of speech. The freedom of speech presumably suffices to protect what is said at an assembly. Freedom of assembly or association is necessary to protect the seedbed of free speech: the group that plans and guides the speech.

This aspect of freedom of assembly was tested in the first decade of the republic. Political opponents of the Washington administration formed dozens of “Democratic-Republican societies” in every region of the nation. Typically they held public meetings every month and more frequently near elections or at times of political controversy, and in between the public meetings met among themselves to discuss ideas, principles, and strategies. Especially in the wake of the Whiskey Rebellion, President Washington and his Federalist allies came to view the Democratic-Republican societies as instigators of sedition, incipient American versions of the Jacobin clubs of revolutionary France. Washington called for legislative action against them.

The particular charge was that the societies were “self-created,” meaning that they were deliberately organized and limited to like-minded members rather than spontaneous and fully public. One Federalist critic contrasted the societies to a “regular town meeting,” warning of their “nocturnal meetings, of individuals, after they have dined, where they shut their doors, pass votes in secret, and admit no members into their societies, but those of their own choosing.”

In defense of their activities, the clubs vocally asserted “the right of citizens to assemble.” Washington responded that “no one denies the right of the people to meet occasionally, to petition for, or to remonstrate against, any Act of the legislature” but that it was “pernicious to the peace of Society” for a “self-created permanent body” to organize and sustain political dissent. In a letter to James Monroe, James Madison described this speech as “perhaps the greatest error” of Washington’s political career. Although the Federalist-dominated Senate censured the Democratic-Republican societies as Washington proposed, the House conducted a serious debate over whether the clubs had “a constitutional right to exist.” Under Madison’s leadership, the House eventually drafted a reply to Washington that omitted any censure of the groups.

This was regarded as a great victory for freedom of assembly, and as setting a precedent that groups of citizens have the right not just to “meet occasionally” but to form groups of like-minded persons to pursue common political, religious, cultural, or social objectives. By the 1830s, French observer Alexis de Tocqueville regarded this freedom of association as the most distinctive feature of American public life. Inazu aptly calls it “liberty’s refuge,” because the freedom of assembly allows downtrodden groups in American life the space to meet, share ideas, and prepare a more public message if that is what they wish to do.

Inazu also stresses the nonpublic and nonexpressive functions of Tocquevillian groups. With their public meetings no longer confined to consultations over the common good, constitutionally protected associations in America could provide opportunities for individuals to gather together, talk among themselves, generate ideas or offer mutual support, worship, play, or enjoy each other’s company, whether or not they had a particular message to impart to the public at large. Much of this important human interaction, essential to a healthy society, is lost when the courts reduce the freedom of groups to a freedom of speech and protect groups only from regulations that would impair their public messages.

After setting forth the legal and historical foundations of freedom of assembly, Inazu provides a lively history of assemblies, conventions, and associations in the nineteenth and twentieth centuries. He pays perhaps insufficient attention to the legal obstacles faced by the abolitionist movement before the Civil War but describes in detail the meetings, marches, and parades of freedmen and women’s-rights advocates in the aftermath of the war.

He rightly stresses the symbolism of freed blacks marching and meeting in public space as if they were full and equal citizens, entitled to be there. Which they were. He similarly describes public meetings where women occupied the main floor and the stage, with their male supporters relegated to the galleries: a silent but eloquent demonstration of the capacity of women, no less than men, to lead and to govern. Inazu’s unstated but provocative point is that the very racial and sexual homogeneity of these groups was part of their message. Liberal integrationism would have blunted, if not destroyed, their witness.

Inazu’s story of the mid-twentieth century is dominated by three movements, each plagued by governmental opposition and each loudly invoking in self-defense the freedoms of assembly and association. The NAACP generally prevailed in the Supreme Court, giving birth to a formal recognition of the freedom of association as a fundamental constitutional right. The Ku Klux Klan and the Communist party generally lost. Unfortunately, there was no common thread to the Court’s reasoning in these cases. NAACP protests were protected even when they were not entirely peaceable (see the 1982 decision NAACP v. Claiborne Hardware Co.), and Communists were often unprotected even when they were (see the 1951 decision Dennis v. United States). The cases thus produced a jurisprudence of uncertainty rather than a refuge for liberty.

In the modern period, the distinctive elements of freedom of assembly—the right to speak in public places and the freedom of association—came together in cases involving student groups on public campuses. Here Inazu’s legal history is a bit thin. In a 1972 decision he does not even mention, Healy v. James, the Supreme Court overturned the efforts of a public university to exclude a local chapter of Students for a Democratic Society (SDS) from participation in its program of recognized student groups, and did so expressly on grounds of freedom of association rather than just freedom of speech.

This was significant for two reasons. First, the Court analogized public-university meeting space to the traditional public forums of streets, parks, and commons, rejecting the university’s claim that “recognition” imparted government approval, which the government was entitled to withhold from groups of which it disapproved. Second, it did so in the context of a student group that refused to forswear the use of violence. Evidently, the “peaceableness” limitation of the Assembly Clause would not be triggered without an outbreak of outright violence.

After the SDS decision, the Supreme Court handed down a series of rulings expanding the scope of the right of student groups to meet on the property of public schools: to religious groups on college campuses, to religious and other groups on elementary and high school campuses, and to religious groups wishing to share in the benefits of student-activity funds. Most of the lower-court cases involved two types of student group: homosexual (or homosexual rights) groups and Christian groups. Interestingly, the decisions in favor of Christian groups were cited to support the rights of homosexual groups, and the decisions in favor of homosexual groups were cited to support the rights of Christian groups. That is the way First Amendments rights should be: Rights transcend the identities and perspectives of the people involved.

The final chapters of Liberty’s Refuge recount the demise of the freedoms of assembly and association in the Supreme Court’s jurisprudence over the past thirty years, as public-accommodation laws originally directed only at certain businesses like railroads, inns, or theaters were applied to voluntary private groups. Increasingly, the Court forgets or neglects the independent textual roots of the freedom to assemble and depicts the freedom of association as merely “implicit” in the freedom of speech. And the Court is increasingly leery of groups that do not share the egalitarian ideology of the civil-rights revolution.

In Roberts v. United States Jaycees in 1984, the Court allowed a state to require the young men’s social and service organization to admit women as members, on the theory that this would have no effect on the group’s public advocacy. It might be true that the group’s public advocacy would be unaffected, Inazu argues (though even that is doubtful), but what about all the other attributes of voluntary group association, such as friendship, service, leadership, or development of role models? By focusing only on public advocacy—the “expressive” nature of the association—the Court essentially eliminated all constitutional protection for the group itself.

The Boy Scouts narrowly escaped a similar fate, with even more troubling constitutional implications. The organization understood its own teaching as being inconsistent with open homosexuality and accordingly declined to allow an openly gay young man to serve as assistant scoutmaster. In Boy Scouts of America v. Dale, decided in 2000, the five-justice majority concluded that the freedom of association protects the right of private noncommercial expressive groups to choose leaders whom the group thinks will exemplify their moral views. (I wrote the brief for the Boy Scouts.)

There were, however, four dissenters, Justices Stevens, Souter, Ginsberg, and Breyer. They did not precisely disagree with the majority’s legal framework, but they declared it “as plain as the light of day” that homosexuality did not offend the Boy Scouts’ creed. That is an assertion that should give any civil libertarian pause. Where do justices get the authority to say that a private group’s moral teaching is really the opposite of what it thinks? Yet much of the legal academy sided with the dissenters, and many cities across the nation responded by denying Boy Scout groups access to public facilities, such as marinas, parks, or campgrounds.

Thus did the right of groups to be “self-created”—the right to organize associations of like-minded citizens for mutual support as well as public witness—lose its legal force. If freedom of association is merely “implicit” in freedom of speech, groups lose control over their identity—their membership and leadership—if they cannot prove to a court’s satisfaction that their public message would be changed by the admission of outsiders. This removes all constitutional protection for the nonexpressive aspects of a group’s activities and invites courts to second-guess the meaning of their beliefs.

Inazu does not discuss the parallel reduction in protection for the right of assemblies to meet in public spaces. As already noted, in the SDS case and others, the Court initially extended the idea of “public forum” beyond the traditional open spaces of parks, streets, and sidewalks to public property that has been opened for use by the public for speech—places such as empty classrooms on a college campus, high school auditoriums over the weekend, airports, municipal buses, city-owned theaters, and even “metaphorical” forums such as the Combined Federal Campaign, student-activity funds, community-access television channels, and the like. The government may no more discriminate among speakers on public property that has been opened for free-speech purposes than it could in the parks and streets.

But this relatively simple idea was complicated by the fact that sometimes the government opens property not to the general public for speech on any subject but to specific uses. A public university might open classrooms for use by university student groups but not to community groups. A city council might institute an open-mic period for public comments on matters pertaining to city business without allowing blowhards to hold forth on irrelevant subjects. A public school might limit its poetry magazine to poetry or a science magazine to science. The courts properly approved these limits so long as the limits reasonably served the purpose for which the forum was created and did not discriminate among speakers on the basis of viewpoint.

In recent years, however, some courts have interpreted this doctrine as allowing the government to limit any nontraditional forum—meaning any property other than streets, sidewalks, and parks—in any way, so long as the limits are reasonable and not overtly discriminatory against particular viewpoints.

For example, two circuit courts, the Second (in New York) and the Ninth (in California), have upheld policies opening government property to all community uses except religious “worship,” not on the ground that such use would violate the separation between church and state, a theory that was rejected thirty years ago, but on the ground that “worship” is a category of speech and not a viewpoint, which therefore may be excluded from a “limited public forum.” On this reasoning, the city of New York has decided to evict more than a hundred churches from public spaces they are now renting. The courts have rejected the churches’ free-speech claims, though their claim to the free exercise of their religion—their claim that government may not single out religious practices for unfavorable treatment—remains undecided.

Inazu calls particular attention to a recent decision, Christian Legal Society v. Martinez. (I represented the losing party, the Christian Legal Society, in the Supreme Court.) The Hastings College of the Law, a public-university law school in San Francisco, denied a small Christian student group its right to meet on campus after the school learned that the group believed that sexual relations are immoral outside of traditional marriage. The school’s excuse was a newly minted “all-comers” policy, which purportedly requires all student groups to allow anyone to vote and seek leadership positions, even if they do not share or indeed reject the group’s beliefs and purposes. The Christian Legal Society happens to be the only group this rule has ever been enforced against. Two years ago, by a five-to-four vote, the Supreme Court upheld Hastings’ policy against the Society’s First Amendment challenge.

Both of the recent changes in the law of freedom of assembly played a part in the Court’s affirming the school’s effective discrimination against the religious group. The Court first decided to “merge” the group’s association claim into its speech claim—stating that it “makes little sense to treat [CLS’s speech and association] claims as discrete” and that “the same ground rules must govern both speech and association challenges” in the context of exclusion from public property. In other words, there is no separate protection for assembly or association, beyond the protection for speech.

Then it rejected CLS’s free-speech claim on the ground that the “all-comers” rule was not overtly directed at any particular viewpoint. Addressing CLS’s argument that the rule in practice discriminated against small and unpopular groups, the Court responded that the policy is directed at CLS’s “conduct” and not its beliefs—its “conduct” being the exercise of its freedom-of-association right to set its own criteria for leaders and voting members.

Theodore Sedgwick would be horrified. He thought that freedom of speech was broad enough to protect the right of groups to organize and meet. It turns out, though, that according to the Supreme Court, freedom of speech protects only the message itself and not the process of organizing the message through the association of like-minded individuals. John Page and the First Congress were prescient in seeing that separate protection for assembly (as well as religion, press, and petition) would be necessary to prevent the government from using various “pretexts” to suppress assemblies that are contrary to the views of those in power. In Washington’s day, the unpopular meetings were of the Democratic-Republican clubs. In Alabama in 1958, they were of the NAACP. In San Francisco and New York today, the unpopular meetings are of religious citizens.

The title Liberty’s Refuge: The Forgotten Freedom of Assembly is therefore sad but all too true: Freedom of assembly has been forgotten. And unfortunately Inazu is right about another thing as well: It matters. America has long been distinguished by a vibrant and independent civil society, one possible only when voluntary associations can meet freely in public spaces and public institutions and when they can limit their membership and leadership to persons who share their beliefs. This means that groups will exist that we like and groups will exist that we do not like.

Under the Court’s current weak doctrine, governments can effectively pick and choose which groups are permitted to use public property, using pretexts like the “all-comers” rule or the policy against “worship.” The framers of the First Amendment thought they had guaranteed all associations the right to meet, with the sole limitation that they behave peaceably. That freedom has slipped away.

Michael W. McConnell is the Richard and Frances Mallery Professor and Director of the Constitutional Law Center at Stanford Law School.