Reno v. American Civil Liberties Union

The CDA regulated speech because of its content, so it required the most demanding First Amendment review. The Government had a legitimate and important interest in protecting children from harmful material, but that interest could not justify unnecessarily suppressing speech that adults have a constitutional right to receive and exchange.

The Government’s principal precedents did not control. Ginsberg v. New York upheld a carefully limited restriction on commercial sales of material harmful to minors; unlike the CDA, it preserved parental choice and used a definition tied to material lacking value for minors. FCC v. Pacifica Foundation involved a noncriminal, context-specific sanction for a radio broadcast in a historically regulated and uniquely pervasive medium. Renton v. Playtime Theatres concerned zoning aimed at secondary effects, whereas the CDA directly targeted the content and primary effects of speech.

The Internet did not share broadcasting’s characteristics of spectrum scarcity, pervasive intrusion, or a history of extensive governmental supervision. Internet users generally must take affirmative and deliberate steps to access particular content, and they seldom encounter sexually explicit material by accident. The Court therefore found no basis for giving online speech less than the ordinary, full measure of First Amendment protection.

The statutory terms “indecent” and “patently offensive” were undefined and uncertain in scope. They omitted critical limits present in the Miller obscenity test, including a requirement that the regulated conduct be specifically defined by law, that the work appeal to prurient interest, and that it lack serious literary, artistic, political, or scientific value. The resulting uncertainty was especially troubling because the CDA imposed criminal penalties, creating a powerful incentive for speakers to censor protected expression rather than risk prosecution.

The CDA effectively forced speakers to limit their communications to what was fit for children. Its broad terms could reach serious discussions of birth control, homosexuality, prison rape, safe-sex practices, artistic nudity, and other nonobscene material. Its application to Internet communications available nationwide also meant that speakers could be judged by the standards of the community most likely to be offended.

The statute burdened adult speech because, absent a feasible way to identify and exclude minors, a speaker communicating in open chat rooms, newsgroups, e-mail lists, or websites had to assume minors might receive or view the material. Age-verification mechanisms were not practically available for many speakers, particularly nonprofits and individuals, and could deny access to adults who lacked credit cards or did not wish to disclose identifying information.

The CDA’s affirmative defenses did not cure this overbreadth. The defense for reasonable, effective measures could not safely be invoked through content “tagging” because no proven universal tagging and filtering system existed, and a speaker could not know whether every minor’s guardian used blocking software. Credit-card or adult-code verification was economically infeasible for many speakers and was not shown to prevent minors from posing as adults.

Less restrictive approaches were available or plausibly available, including user-based filtering tools, content tagging, exceptions for material with serious value, recognition of parental choice, and different treatment for distinct Internet services. Congress had not developed a record showing that its sweeping criminal prohibitions were necessary or narrowly tailored. The Government therefore failed to justify the CDA’s substantial burden on protected adult speech.

The Court recognized that the CDA’s imprecise language raised serious vagueness concerns. But it treated that uncertainty as relevant to the First Amendment analysis: vague content-based criminal laws chill protected speech because speakers cannot confidently determine what they may say.

Because the CDA’s burden on protected expression was already sufficient to establish its facial invalidity under the First Amendment, the Court found it unnecessary to resolve whether the statute independently violated Fifth Amendment due process requirements.

Obscene speech receives no First Amendment protection, and § 223(a) separately prohibited communications that were “obscene or indecent.” Because “obscene” was textually distinct from “indecent,” the Court severed the invalid indecency term while leaving the obscenity prohibition available for enforcement.

The Court declined to recast the broader statute as applying only to commercial pornographers, material without serious value, or some other narrower category. A court may adopt a limiting construction only when the statute is readily susceptible to one, and the CDA’s open-ended language supplied no clear, congressionally chosen line to enforce.

The case reached the Court as a facial challenge under a special statutory review provision. Given the broad range of plaintiffs and expressive activity, as well as the statute’s uncertainty, the Court could not practically convert the case into a collection of as-applied rulings. Rewriting the law to select permissible applications would improperly invade Congress’s legislative role.