Concept
Identifying Content-Based Laws
Content-based regulation of protected expression
Identifying Content-Based Laws
A court cannot apply the presumption of invalidity until it decides which kind of law it faces, and that decision fixes the tier of review. A government “has no power to restrict expression because of its message, its ideas, its subject matter, or its content,” and content-based laws are presumptively unconstitutional; the government may justify one only by proving that it is narrowly tailored to serve a compelling state interest.1 Content-neutral laws, which restrict speech without regard to its message or subject matter, receive a more lenient balancing test: the government’s interest must be sufficiently substantial, and the restriction must not be substantially broader than necessary to protect it.2 Classification therefore decides whether a law faces strict scrutiny, with its presumption of invalidity, or a form of balancing.3 Sorting a law into one category or the other is categorization, which supplies the rules that govern the case; the alternative is balancing the government’s interests against speech case by case.4
Reed v. Town of Gilbert supplies the working definition. A regulation of speech is content based if it applies to particular speech because of the topic discussed or the idea or message expressed.5 Reed also recognized a second category, taken up under “Content-Neutral Justifications” below: a law that is neutral on its face is still content based if it cannot be justified without reference to the content of the speech, or if the government adopted it because of disagreement with the message.6 Either route leads to strict scrutiny.
Content-based laws come in two forms. A subject-matter restriction singles out a topic and reaches every speaker on it, whatever position the speaker takes. A viewpoint restriction singles out one position within a topic. Both trigger strict scrutiny, but courts treat viewpoint discrimination as the graver offense, so the subsections below take subject-matter restrictions first and viewpoint discrimination second. Two further questions complete the picture: when a law that is neutral on its face becomes content based because of its justification, and when a rule that draws a line by location stays content neutral.
Subject-Matter Restrictions
A court begins with the face of the law. Reed’s test requires it to consider whether the regulation “on its face” draws distinctions based on the message a speaker conveys.7 Some facial distinctions are obvious, because the law defines the regulated speech by its subject matter. Others are subtler, because the law defines the regulated speech by its function or purpose. Both draw lines based on message, and both receive strict scrutiny.8
Gilbert’s Sign Code shows the function-based form. It defined Temporary Directional Signs by whether a sign directed the public to a church or another “qualifying event,” Political Signs by whether a sign’s message was “designed to influence the outcome of an election,” and Ideological Signs by whether a sign communicated a message or ideas that fit neither category, and it subjected each category to different restrictions.9 No definition mentioned a viewpoint, yet each required an official to ask what a sign was for, and a sign’s purpose is part of what it communicates. The Town defended its categories with two interests, aesthetic appeal and traffic safety. Even assuming both interests were compelling, the distinctions failed as hopelessly underinclusive, because the Town could not claim that strict limits on temporary directional signs were necessary to beautify the Town while allowing unlimited numbers of other signs that create the same problem.10
A topic ban is content based even when it takes no side. In Federal Communications Commission v. League of Women Voters, the statute banned all editorials or endorsements, not merely those that objected to the government’s official views or to the views of incumbent office holders. The Court still found it censorial, because it prohibited discussion of an entire topic and so shaped and reduced the agenda for public debate.11 Boos v. Barry reached a similar classification for a ban on signs critical of foreign governments near embassies. The Court agreed that the display clause was not viewpoint based, because it fixed the acceptable viewpoint neutrally by looking to the policies of foreign governments, but it still held the clause content based.12
Picketing ordinances supply the oldest examples. The ordinance in Police Department of Chicago v. Mosley permitted peaceful picketing on the subject of a school’s labor-management dispute and prohibited all other peaceful picketing. The Court located the central problem there: the ordinance described permissible picketing in terms of its subject matter, so the operative distinction was the message on a picket sign.13 Carey v. Brown applied Mosley and found that the statute before it “accords preferential treatment to the expression of views on one particular subject; information about labor disputes may be freely disseminated, but discussion of all other issues is restricted.”14 Neither law told a picketer which side of a labor dispute to take. Each told picketers which topic they could discuss.
Frisby v. Schultz supplies the contrast. The Court upheld an ordinance that prohibited picketing “before or about” any residence, because the law was completely subject-matter neutral, even though the town adopted it in response to targeted picketing by anti-abortion protestors.15 The statute in Carey exempted one topic; the Frisby ordinance exempted none.
Discrimination against an entire subject matter can still skew debate by tilting it away from that topic, but it at least removes all perspectives on the topic, and in some contexts the government has legitimate interests in limiting the topics discussed.16 For that reason the government may define a “limited public forum” by the subject to be discussed.17 Forum analysis governs only when the government restricts speech on property it owns, such as streets and public buildings, or in a medium it owns, such as a public school newspaper or campus mail system; it does not apply to private property or privately owned channels of communication.18
Viewpoint Discrimination
Viewpoint discrimination goes beyond choosing a topic. It selects among the positions speakers take on a topic, and the Court examines it more closely than topic discrimination. As Justice Stevens put it, viewpoint-based restrictions on expression in general require greater scrutiny than subject-matter-based restrictions.21 The difference follows from the point made above: a topic rule removes every perspective on the topic, while a viewpoint rule removes only some and so lets the government decide which side the public hears.
R.A.V. v. City of St. Paul, which a later section develops, shows a viewpoint rule hiding inside what looks like a topic rule. Under the St. Paul ordinance, a speaker could hold up a sign saying that all “anti-Catholic bigots” are misbegotten, but not that all “papists” are, because the second would insult and provoke violence “on the basis of religion.” The Court objected that St. Paul had no authority “to license one side of a debate to fight freestyle, while requiring the other to follow Marquis of Queensberry rules.”22 Even within a class of speech the government may proscribe, a distinction survives only when its basis is the very reason the class is proscribable. A State might prohibit only the obscenity that is most patently offensive in its prurience, but it may not prohibit only the obscenity that includes offensive political messages.23
Forum cases show the line between the two forms most clearly. The government may limit a forum to a topic, but viewpoint discrimination is prohibited in any forum, so a speaker challenging an exclusion need only show that the expression fits within the topic of whatever forum the government has created.24 In Rosenberger, a university excluded a Christian magazine because it “promote[d] or manifest[ed] a particular belie[f] in or about a deity or an ultimate reality.” The university argued that its standard excluded atheistic as well as theistic publications and so excluded the entire subject of religion. The majority answered that the policy disfavored two viewpoints, atheism and theism, while subsidizing publications reflecting many other perspectives, since the magazine wrote on the same cultural, moral, and political topics as the publications that received support.25 Evenhandedness between two disfavored views did not make the policy viewpoint neutral.
One regulation can raise both classifications. Justice Blackmun’s dissent in Rust v. Sullivan argued that the Title X counseling and referral provisions were plainly content-based regulation of speech, because grantees could provide counseling and referral on a wide range of family planning and other topics, save abortion.26 The dissent also rejected the majority’s conclusion that the government “has not discriminated on the basis of viewpoint; it has merely chosen to fund one activity to the exclusion of the other,” calling that claim hollow in light of the broad range of nonpreventive services the regulations authorize.27 The majority’s framing recalls the view that the greater power to deny a benefit includes the lesser power to impose conditions on its receipt. The doctrine of unconstitutional conditions reflects the triumph of the opposing view, that government may not do indirectly what it may not do directly.28
Governments also defend viewpoint rules by pointing to how audiences will react, and the trademark cases rejected that defense. A concurring opinion in Matal v. Tam explained that the government may not insulate a law from charges of viewpoint discrimination by tying censorship to the reaction of the speaker’s audience, and it added that viewpoint discrimination need not take the form of an intent to suppress a speaker’s beliefs in every instance.29 The danger of viewpoint discrimination is that the government removes certain ideas or perspectives from a broader debate, and the danger grows when an audience might find the ideas offensive at first hearing, because an initial reaction may prompt further reflection and a more tolerant position.30 A speech burden based on audience reactions is simply government hostility and intervention in a different guise: the government targets the speech because it disapproves of the speaker’s choice of message, and the government itself decides whether the audience would find the speech offensive.31 The Court’s cases have accordingly long prohibited the government from justifying a First Amendment burden by pointing to the offensiveness of the speech to be suppressed.32 The offense can run deep. A footnote in one of the Court’s opinions collects the intelligible remarks recorded at a rally, from racial slurs to slogans such as “Save America” and “Freedom for the whites.”33 Speech of that kind tests the principle, and the principle makes no exception for it.
Two years after Tam, Iancu v. Brunetti struck down another provision of the Lanham Act, which prohibited registration of “scandalous” or “immoral” trademarks. A clothing manufacturer wished to register “F.U.C.T.” and was refused under that provision.34 All the Justices in Tam had agreed that a registration bar is doomed if it is deemed viewpoint based, so the question in Brunetti was whether the bar fit that description.35 Justice Alito’s concurrence observed that the decision does not prevent Congress from adopting a more carefully focused statute that precludes registration of marks containing vulgar terms that play no real part in the expression of ideas, and that the “F.U.C.T.” mark could be denied under such a statute.36
Content-Neutral Justifications
Reed’s second category exists because a purely textual test is easy to evade. A law neutral on its face is nonetheless content based if it cannot be justified without reference to the content of the regulated speech or if the government adopted it because of disagreement with the message.43 A facially content-based law, meanwhile, receives strict scrutiny whatever the government’s benign motive, content-neutral justification, or lack of animus toward the ideas it regulates, because innocent motives do not remove the danger that future officials will use the statute to suppress disfavored speech: “The vice of content-based legislation . . . is not that it is always used for invidious, thought-control purposes, but that it lends itself to use for those purposes.”44 Gilbert shows both halves. The Town offered only aesthetics and traffic safety, interests that sound unrelated to any message, yet the code drew strict scrutiny and failed.45
Aesthetic justifications illustrate why courts look past stated interests. The asserted interest in aesthetics may be only a facade for content-based suppression, because the real objective might simply be to eliminate the messages that signs typically carry, and a city might easily mask that objective by declaring that signs constitute visual clutter.46 A total ban justified solely in terms of aesthetics also frustrates the means inquiry, which deliberate or unintended government manipulation may impede.47
Justification matters because it helps decide which track a law travels: courts apply strict scrutiny, with a presumption of invalidity, in some cases and a form of balancing in others.48 United States v. O’Brien supplies the lenient standard for laws that regulate conduct with expressive elements, and the Court has clarified that this standard is little if any different from the one applied to time, place, and manner restrictions, which require that a law leave open ample alternative channels of communication.49 Texas v. Johnson, which tested a restriction on flag burning, applied the most exacting scrutiny after finding the restriction content based.
Permit denials show the same sorting by reason. When the Ku Klux Klan sought a permit to march in Louisville on Derby Day, the city could refuse on content-neutral time, place, and manner grounds: tens of thousands of people converge on the city for the Kentucky Derby, and city police are diverted to Churchill Downs for crowd control, so congestion and strain on police resources justified denying the march that day.50 The city could not have prohibited the Klan from marching on all days, or because it disliked the Klan’s message.51 The reasons that justified the denial exist whatever the marchers say, and a reason that depended on what they say would have condemned it.
The second direction of justification analysis begins with City of Renton v. Playtime Theatres, Inc., which involved a zoning ordinance that prohibited adult motion picture theaters from locating within 1,000 feet of any residential zone, single- or multiple-family dwelling, church, park, or school.52 The ordinance treated theaters that specialize in adult films differently from other theaters, so at first glance it fit neatly into neither the content-based nor the content-neutral category. The Court concluded that it was aimed not at the content of the films but at the secondary effects of such theaters on the surrounding community, and the district court had found that the City Council’s predominant concerns were those effects, not the content of adult films.53 Justice Powell’s observation in American Mini Theatres explains why that justification was credible: a city concerned with restricting the message would have tried to close such theaters or restrict their number, not circumscribe their choice of location.54
Secondary effects are harms that accompany a kind of business whether or not anyone objects to its message. In Renton they were crime, property values, and the protection of residential neighborhoods, effects almost unique to theaters featuring sexually explicit films, and the ordinance did not aim at the suppression of free expression.55 Renton could rely heavily on the experience of, and studies produced by, Seattle, because the Court of Appeals had imposed an unnecessarily rigid burden of proof.56 The Court has also reasoned, in discussing a similar thousand-foot restriction, that the restriction does not in itself create an impermissible restraint on protected communication, because the city’s interest in planning and regulating the use of property for commercial purposes is clearly adequate to support that kind of restriction applicable to all theaters within the city limits.57 Concurring in City of Los Angeles v. Alameda Books, Inc., Justice Kennedy argued that such ordinances “are content based, and we should call them so,” but that the Renton analysis should control anyway. He added a limit: a court must evaluate “how speech will fare” under the law, and an ordinance that reduces secondary effects by reducing speech, rather than merely dispersing it, is impermissible.58
Boos draws the boundary of the secondary-effects rationale. The respondents there did not point to congestion, interference with ingress or egress, visual clutter, or the security of embassies, any of which would have been a secondary effect of picket signs. They relied instead on the need to protect the dignity of foreign diplomatic personnel by shielding them from speech critical of their governments. That justification focuses only on the content of the speech and its direct impact on listeners, and the emotive impact of speech on its audience is not a secondary effect. Because the display clause regulates speech due to its potential primary impact, the Court held it content based.59 R.A.V. applied the same line, noting that even if a total proscription could ever be directed only at secondary effects, the St. Paul ordinance was not, because “Listeners’ reactions to speech are not the type of ‘secondary effects’ we referred to in Renton.”60
In Reno v. American Civil Liberties Union, the Government defended the Communications Decency Act as a sort of “cyberzoning” on the Internet. The Court answered that the statute applies broadly to the entire universe of cyberspace and aims to protect children from the primary effects of “indecent” and “patently offensive” speech, not from any secondary effect, so it is a content-based blanket restriction that cannot be analyzed as a time, place, and manner regulation.61 The later litigation over COPA carried the same concern forward. One opinion there observed that the statute’s adult-verification mechanisms provide only an affirmative defense, so even full compliance cannot guarantee freedom from prosecution, and speakers who place their content behind age screens may still find themselves in court, forced to prove the lawfulness of their speech on pain of criminal conviction.62 The same opinion adhered to the view that the Government may not penalize speakers for making available to the general World Wide Web audience what the least tolerant communities in America deem unfit for their children’s consumption.63
The two directions of justification analysis rest on different premises, and the Court has not fully reconciled them. Reed gives a facially content-based law strict scrutiny whatever its benign motive, yet Renton accepted a neutral justification as the reason to treat an ordinance that distinguished theaters by the content of their films as content neutral.64 Justice Brennan’s dissent in Renton pressed the concern about pretext. Other motion picture theaters and other forms of adult entertainment, such as bars, massage parlors, and adult bookstores, faced no comparable restrictions, and that selective treatment strongly suggested that Renton was interested in discriminating against adult theaters based on the content of their films, not in controlling secondary effects.65 The majority’s approach, the dissent added, lets a city conceal illicit motives by relying on similar restrictions adopted elsewhere, since a municipality can always find other ordinances to justify special zoning of adult theaters after the fact. The dissent would have required the city to show, as with any content-based restriction, a precisely drawn means of serving a compelling governmental interest.66
The working boundary is narrow. Reed said nothing about secondary effects, and the Court has never applied the doctrine to any context other than adult-oriented businesses. Perhaps the Reed majority ignored the matter because the ordinance could not pass any level of scrutiny, including, as Justice Kagan put it, “the laugh test.”67 A facially content-based law outside adult-business zoning is therefore tested under Reed, and Boos, R.A.V., and Reno confine Renton further by excluding any harm that consists of how an audience reacts to the message.
Location-Based Distinctions
Reed’s facial test has a boundary, and City of Austin v. Reagan National Advertising of Austin, LLC marks it. The decision limited Reed: a rule that distinguishes on-premises signs from off-premises signs is not content based merely because an official must read a sign to apply it.71 The reading serves only to draw a neutral, location-based line. Reed’s test asks whether a regulation draws distinctions based on the message a speaker conveys, and a line between a sign’s own site and some other site singles out no topic, idea, or message.72
Gilbert’s code differed because each of its categories depended on what a sign said. An official had to decide whether a sign concerned an election, a church event, or an idea, and the Court then tested the resulting distinctions against the Town’s aesthetic and safety interests and found them hopelessly underinclusive.73 A rule that draws only a location line never reaches that stage, because it faces the lenient test for content-neutral laws instead.
The Justices divided over where Austin’s rule falls. Justice Thomas’s dissent and Justice Breyer’s concurrence reflect a continuing debate about the breadth of the Reed rule.74 It is perhaps too early to tell whether the limitation on Reed will prove to be the beginning of a retreat from Reed or only a minor limitation.75
The picketing cases show the same line on the ground. Chicago’s ordinance in Mosley described permissible picketing in terms of its subject matter, and the Court faulted it for that. The Court has upheld ordinances that prohibited focused picketing at persons’ homes when the laws are completely subject-matter neutral, as with the ordinance in Frisby that barred picketing “before or about” any residence.76 One law drew its line by subject; the other drew it by place.
McCullen v. Coakley tested a statute whose trigger was location alone. The Massachusetts law made it a crime to knowingly stand on a public way or sidewalk within 35 feet of an entrance or driveway to any place, other than a hospital, where abortions are performed.77 Protestors who approached women outside such facilities to talk them out of abortions challenged the law. It exempted four classes of people, including employees or agents of the facility acting within the scope of their employment, and the legislature kept a separate provision that already proscribed the knowing obstruction of access to a facility.78 Whether a person violated the statute depended on where the person stood, and a person could violate it without saying a word, so the statute restricted speech without regard to its message. The Court treated it as content neutral on its face, yet it struck the law down because the buffer zone burdened substantially more speech than necessary.
Hill v. Colorado presents a harder pairing of place and message. The Colorado statute made it unlawful, within 100 feet of a health care facility’s entrance, to “knowingly approach” within 8 feet of another person, without that person’s consent, to pass out a leaflet or handbill, display a sign, or engage in oral protest, education, or counseling; the legislature passed it in part as a response to problems of access caused by abortion protestors.79 Unlike the Massachusetts statute, its reach depended on a kind of communication as well as a place. The Court nonetheless held it content neutral, because it applied to all protest, education, and counseling whatever the subject or viewpoint.
Holding a rule content neutral on its face does not end the case. Evidence that an impermissible purpose or justification underlies a facially neutral rule can still make it content based, because Reed’s second category reaches laws that cannot be justified without reference to content or that the government adopted because of disagreement with the message.80 A rule that remains content neutral must still be narrowly tailored to serve a significant governmental interest: the interest must be sufficiently substantial, and the restriction must not be substantially broader than necessary.81