Concept
Internet Content Regulation
Content-based regulation of protected expression
Internet Content Regulation
Two decisions show how the First Amendment treats laws that try to shield minors from sexual material online. Reno v. American Civil Liberties Union struck down a federal statute that criminalized sending “indecent” or “patently offensive” messages to minors, because senders had no effective way to screen out minors and the statute therefore operated as a ban on speech that adults may receive.1 The later decision in Ashcroft v. ACLU held that an age-verification statute for sexually explicit websites cannot stand while less restrictive alternatives, such as filters that parents install, remain available, because the Government must show that its measure is the least restrictive effective alternative.2
Sexually explicit speech sits low in the First Amendment hierarchy, usually because it is said to contribute little or nothing to the exposition of ideas.3 Low rank still leaves protection in place. Sexual expression that is indecent but not obscene is protected, and the government’s legitimate interest in shielding children does not justify suppressing speech addressed to adults any more broadly than necessary.4 Both cases therefore turn on tailoring. Reno asks what a law does to adult speech when speakers cannot tell who is listening, and Ashcroft asks who must prove what when a narrower tool is available. Reno’s rejection of the Government’s “cyberzoning” label also bears on the secondary-effects material earlier in this chapter.
Reno v. ACLU
The Communications Decency Act (CDA) made it a crime, punishable by up to two years in prison for each violation, to send “indecent” messages to minors or to display “patently offensive” material to them online.5 The Court affirmed the District Court’s judgment against the statute.6 Its reasoning starts with what a sender can know about the audience.
One Government defense was that the statute’s knowledge requirement protected communication among adults, and the District Court’s findings made that premise untenable. With no viable age-verification process, a sender addressing a large audience must be charged with knowing that one or more minors will likely see the message. If a sender knows that a member of a 100-person chat group is a minor, sending the group an indecent message becomes a crime, and the adult members’ conversation bears the burden.7 The narrower “specific person” requirement fared no better. Most Internet forums, including chat rooms, newsgroups, mail exploders, and the Web, are open to all comers, so any opponent of indecent speech could log on and announce that his 17-year-old child was present, and the statute would hand him a “heckler’s veto.”8
Technology offered speakers no escape. At the time of trial, no effective method existed for a sender to keep minors from its communications on the Internet without also denying access to adults, and no effective way existed to determine the age of a user reaching material through e-mail, mail exploders, newsgroups, or chat rooms. Verifying the ages of Web-site users was also prohibitively expensive for noncommercial speakers and for some commercial ones. These limitations must inevitably curtail a significant amount of adult communication.9 Criminal penalties magnified the effect, because the severity of the sanctions may lead speakers to stay silent rather than communicate even arguably unlawful words, ideas, and images.10 A speaker’s choices were to withhold the message from everyone or to send it and risk prosecution.
Adults have a protected stake in what the statute suppresses. The Court has repeatedly recognized the government’s interest in protecting children from harmful materials, as in Ginsberg v. New York and Pacifica, but that interest does not justify an unnecessarily broad suppression of speech addressed to adults.11 Indecent but not obscene sexual expression is protected, and where obscenity is not involved, the fact that protected speech may offend some listeners does not justify suppressing it.12 Pacifica, the case about a broadcast of the Carlin monologue, put the point further: if the speaker’s opinion is what gives offense, that consequence is a reason to protect the speech, and protection might be required if the offensiveness could be traced to political content or to satire of contemporary attitudes about four-letter words.13 That echoes this chapter’s earlier treatment of giving offense as a viewpoint.
The Court classified the CDA as a content-based blanket restriction on speech, one that cannot be “properly analyzed as a form of time, place, and manner regulation.” The Government’s “cyberzoning” label failed on two grounds. The CDA applied to the entire universe of cyberspace, and its purpose was to protect children from the primary effects of “indecent” and “patently offensive” speech, not from any secondary effect. Regulations that focus on the direct impact of speech on its audience are not analyzed under Renton, and listeners’ reaction to speech is not a content-neutral basis for regulation.14
Nor did it help the Government to argue that adults remained free to speak elsewhere. The Court called that position “equivalent to arguing that a statute could ban leaflets on certain subjects as long as individuals are free to publish books.” It recalled that it had invalidated leafleting bans that applied regardless of content because “one is not to have the exercise of his liberty of expression in appropriate places abridged on the plea that it may be exercised in some other place.”15 That reasoning applies with at least equal force to a law that targets speech because of its content.
The broadcasting order in Pacifica differed in kind from the CDA. It came from an agency that had regulated radio stations for decades, it targeted a single broadcast that departed dramatically from traditional program content, and it decided when, rather than whether, the program could air. The CDA’s broad categorical prohibitions were not limited to particular times and did not depend on any evaluation by an agency familiar with the Internet. The order was also not punitive, and the Court expressly refused to decide whether the broadcast would justify a criminal prosecution, whereas the CDA carried prison terms.16 The Court found no basis in its cases for qualifying the level of First Amendment scrutiny that applies to this medium.17
Vagueness aggravated the problem. Vagueness in a content-based regulation raises special First Amendment concerns because of its obvious chilling effect on speech, and the criminal penalties compounded that chill.18 The Government compared the CDA’s “patently offensive” standard with the second prong of the Miller obscenity test, but the comparison failed. Miller requires that the proscribed material be “specifically defined by the applicable state law,” which reduces the vagueness inherent in an open-ended term like “patently offensive,” and Miller reaches only “sexual conduct,” while the CDA also covers “excretory activities” and “organs” of both a sexual and excretory nature.19
The statute’s defenses and the Court’s remedial options did not cure these defects. The defense built on age verification was not economically feasible for most noncommercial speakers, and the Government offered no evidence that verification techniques actually keep minors from posing as adults. With the risk of criminal sanctions hovering over each content provider “like the proverbial sword of Damocles,” the District Court correctly refused to rely on unproven future technology, and the Government failed to prove that the defense would significantly reduce the burden on adult speech.20 The Court severed the words “or indecent” from the transmission provision, leaving the ban on obscene material standing, but no other textual surgery could save either provision.21 It also declined to redraw the statute itself. Drawing lines through an overly broad statute when Congress has sent inconsistent signals about where they belong “involves a far more serious invasion of the legislative domain,” and the Court will not rewrite a law to conform it to constitutional requirements.22
The holding rests on the requirement of less restrictive alternatives. To deny minors access to potentially harmful speech, the CDA effectively suppressed a large amount of speech that adults have a constitutional right to receive and to address to one another, and that burden is unacceptable if less restrictive alternatives would be at least as effective in achieving the statute’s legitimate purpose.23 The breadth of the restriction placed an especially heavy burden on the Government to explain why a less restrictive provision would not work as well, and it did not do so. The alternatives mentioned in argument included tagging indecent material to facilitate parental control, making exceptions for messages with artistic or educational value, tolerating parental choice, and treating commercial Web sites differently from chat rooms. Congress had made no detailed findings and held no hearings on the CDA’s special problems, so the Court concluded that the statute was not narrowly tailored “if that requirement has any meaning at all.”24 By contrast with sender-side screening, the District Court found that user-based software would soon give parents a reasonably effective way to keep children from sexually explicit material they consider inappropriate.25 That alternative works at the recipient’s end, where a household knows who is using the computer, while the CDA placed the duty on a sender who cannot know.
The Government’s last argument was that regulating content would encourage the Internet’s growth, and the Court found it singularly unpersuasive. The dramatic expansion of the new marketplace of ideas contradicted the factual basis of the contention. “As a matter of constitutional tradition, in the absence of evidence to the contrary, we presume that governmental regulation of the content of speech is more likely to interfere with the free exchange of ideas than to encourage it,” and the interest in encouraging freedom of expression outweighs any theoretical but unproven benefit of censorship.26
Ashcroft v. ACLU
Ashcroft v. American Civil Liberties Union held that a content-based restriction on Internet speech, here a statute requiring age verification for access to sexually explicit websites and criminalizing the failure to verify, cannot stand where less restrictive alternatives such as parent-installed filters are available, because the Government bears the burden of showing that its measure is the least restrictive effective alternative.33 The statute was the Child Online Protection Act (COPA), Congress’s second attempt to restrict and punish this speech after Reno.34 Congress had studied Reno in drafting COPA, and the Court said the Judiciary must proceed with caution before invalidating the Act. Respect for Congress, however, does not permit departing from well-established First Amendment principles, so the Court held the Government to its constitutional burden of proof.35
COPA aimed at commercial pornography on the Internet and required commercial providers to place pornographic material behind “screens” that adults could pass by producing age verification.36 Screening provided only an affirmative defense, so even full compliance could not guarantee freedom from prosecution. A speaker who dutifully placed content behind age screens might still land in court and have to prove the lawfulness of the speech on pain of criminal conviction.37 The District Court found that the statute was likely to burden some speech protected for adults, and the Government did not dispute that finding.38
Because COPA was a content-based restriction, the presumption of invalidity governed from the start. The Constitution demands that content-based restrictions on speech be presumed invalid and that the Government bear the burden of showing their constitutionality, even when Congress has twice tried to find a constitutional means to restrict and punish the speech.39 The allocation matters most when the record is thin, because a gap in the Government’s proof then counts against the statute.
The governing test comes from Reno. A statute that “effectively suppresses a large amount of speech that adults have a constitutional right to receive and to address to one another ... is unacceptable if less restrictive alternatives would be at least as effective in achieving the legitimate purpose that the statute was enacted to serve.” When plaintiffs challenge a content-based restriction, the burden is on the Government to prove that the proposed alternatives will not be as effective as the challenged statute.40 The court assumes that some protected speech may be regulated and asks which available, effective alternative restricts speech the least. It does not begin with the status quo of existing regulations and ask whether the restriction adds some ability to achieve Congress’s goal, because any restriction could be justified under that analysis.41
The District Court concentrated on plausible, less restrictive alternatives to COPA,42 and the Supreme Court found filtering software less restrictive. Filters impose selective restrictions on speech at the receiving end, not universal restrictions at the source. Under a filtering regime, adults without children may gain access to speech they have a right to see without identifying themselves or providing credit card information. Promoting filters also does not condemn as criminal any category of speech, so the potential chilling effect is eliminated or much diminished, however broadly or narrowly COPA’s definitions are construed.43 A filter puts the choice with the receiving household. COPA puts the burden on the speaker, who must prove the lawfulness of the speech after being charged.
The Government objected that filters depend on parental cooperation, and the Court rejected that objection. The need for parental cooperation does not automatically disqualify a proposed alternative: “A court should not assume a plausible, less restrictive alternative would be ineffective; and a court should not presume parents, given full information, will fail to act.” COPA presumes that parents lack the ability, not the will, to monitor what their children see, and Congress could give them that ability by promoting filtering software without subjecting protected speech to severe penalties. Congress itself had said its goal was to keep the Internet’s availability from letting minors reach material in a way that can frustrate parental supervision or control.44
The gap in the record decided the case. The evidence had a serious gap on the effectiveness of filtering software, and assuming without proof that filters are less effective than COPA would usurp the District Court’s factfinding role. By letting the injunction stand and remanding for trial, the Court required the Government to shoulder its full constitutional burden on the less-restrictive-alternative argument instead of excusing it.45 Because the Government bears the burden on the ultimate question, the respondents were likely to prevail unless it showed that their proposed alternatives were less effective than COPA.46
The case reached the Court on a preliminary injunction, so the decision is narrower than a ruling that COPA is unconstitutional. In the earlier round, the Court held only that COPA’s community-standards language did not, standing alone, make the statute unconstitutionally overbroad, and it remanded for the Court of Appeals to reconsider the preliminary injunction.47 This time the Court affirmed the injunction for the reasons the District Court gave and declined to consider the other arguments the Court of Appeals had relied on.48 Practical considerations supported that result. Where prosecution is likely and only an affirmative defense is available, speakers may self-censor rather than risk the perils of trial. Leaving the injunction in place cost little, because no prosecutions had begun and the Government could enforce the obscenity laws already on the books in the meantime.49
Three separate opinions mark the edges of the majority’s position. Justice Stevens, concurring, would have rested the decision on a different ground. He continued to believe 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, and he considered that principle a sufficient basis for deciding the case.50 The principle fits the older rule that the fact that society may find speech offensive is not a sufficient reason for suppressing it.51 Two dissents attack the majority from opposite directions. Justice Breyer read COPA narrowly: in his view the Act’s definitions confine it to legally obscene material, which has no First Amendment protection, “and very little more.”52 The majority’s reasoning did not depend on that dispute, because the points favoring filters hold however broadly or narrowly the definitions are construed.53 Justice Scalia would have gone further and refused to apply strict scrutiny at all. He wrote that commercial entities engaged in “the sordid business of pandering” by deliberately stressing the sexually provocative aspects of their nonobscene products engage in constitutionally unprotected behavior, and that nothing in the First Amendment entitles material of that kind to exacting review.54 His position builds on the low place the Court has long assigned to sexually explicit speech in the First Amendment hierarchy.55 Against it stands the Court’s insistence that First Amendment protection does not depend on an ad hoc balancing of a category’s costs and benefits, and that the Constitution “forecloses any attempt to revise that judgment simply on the basis that some speech is not worth it.”56
Sexually explicit programming on cable had already divided the Court once. In an earlier decision, without a majority opinion, the Court upheld a provision allowing cable systems to refuse to carry sexually explicit programming and invalidated two others, including one allowing systems to prohibit sexually oriented material on public, educational, or governmental channels.57 United States v. Playboy Entertainment Group, Inc. later supplied the rule that Ashcroft applied. The Court held there that, absent a showing that the proposed less restrictive alternative would not be as effective, the more restrictive option Congress preferred could not survive strict scrutiny, and it reversed because the record was silent on the comparative effectiveness of the two alternatives.58 The statute in that case already contained its own narrower tool, since one provision let viewers who wished to do so block sexually explicit programming.59 In Ashcroft the Court found the same gap and gave the same answer: at this stage the Government had not shown that the proposed less restrictive alternative would be less effective.60
Read together, Reno and Ashcroft show two flaws that rest on one tailoring requirement. Reno’s statute left a sender with no effective way to screen out minors without shutting out adults, so it operated as a ban on speech that adults are entitled to receive.61 Ashcroft’s statute reached only commercial providers and offered age screens, so the burden on adults was smaller, and the flaw lay in the Government’s failure to prove that filters would protect children less well.62 A statute drafted to cure the first flaw therefore remains exposed to the second.