First Amendment and the Internet
By far the most important decision in the area of First Amendment law as applied to the Internet to date is Reno v. American Civil Liberties Union [cases/Renoaclu/] in which the Supreme Court essentially determined that the Internet was entitled to the highest degree of First Amendment protection and simultaneously demonstrated a substantial understanding of the nature, potential and problems posed by the Net. In Reno, the Court struck down one CDA provision and a portion of another on the grounds that they were not narrowly tailored enough to justify the unacceptably heavy burden they placed on speech. The provisions prohibited the knowing transmission of obscene or indecent messages to any recipient under 18 and prohibited the knowing sending or displaying of patently offensive communications in a manner that is available to a person under 18.
The Court was particularly troubled that the provisionsí vague definitions of prohibited communications would chill free speech on the Net about a variety of issues, such as birth control practices, homosexuality, and the consequences of prison rape, because Net users could not determine if such discussions would violate the CDA. The Court was also troubled by the breadth of the CDAís coverage which it found to be "wholly unprecedented," noting that the undefined terms "indecent" and "patently offensive" cover large amounts of non-pornographic material with serious educational or other value. The importance of the Reno decision can scarcely be exaggerated as it provides the definitive guide to Internet First Amendment issues for the foreseeable future.
In a second case interpreting the CDA, Apollomedia Corp. v. Reno, a three-judge panel of the United States District Court for the District of Northern California ruled that a provision which on its face criminalized obscene and indecent communications made with an intent to annoy was not impermissibly broad and vague in violation of the First Amendment because it interpreted the provision to regulate only obscene, and not indecent, speech. One judge dissented, arguing that, just like the provisions at issue Reno v. ACLU, [/cases/Renoaclu/] this provision should be struck down on First Amendment grounds, and expressing concern that it was unrealistic to expect the tens of millions of daily Net users to know that the words of the provision do not mean what they say.
Unlike in Reno v. ACLU, in Zeran v. America On-Line, Inc., the CDA emerged as a shield protecting interactive computer service providers like AOL by providing immunity to state law claims alleging that a provider was liable for defamatory messages placed on its service by a third party. The United States Court of Appeals for the Fourth Circuit held that the federal CDA preempted (i.e., forbade under these circumstances) the plaintiffís state law claims pursuant to the Supremacy Clause, which essentially makes federal law superior to state law where they conflict.
The United States Court of Appeals for the Fourth Circuit upheld the lower courtís dismissal of the case on the ground that the CDA provides ISPís with immunity with respect to information placed on their services by third parties. The court noted that Congress had two prime concerns with respect to this aspect of the CDA. First was the fear that, should ISPís be liable for messages posted to their services, they would likely respond in ways that would limit speech on the Internet, for example by restricting the typo or number of messages that could be posted. Second, Congress wanted to encourage ISPís to do content-screening and editing in the face of recent case law that had held that a provider which did such screening was liable for a defamatory statement that it missed, whereas a provider that did no screening would not be liable for the same statement.
The court premised its decision on the Commerce and Supremacy Clauses of the U.S. Constitution, holding that the former limited what states could do in areas of international concern like the Internet, and that, under the latter, the CDA preempted the state law action. (American Libraries Assn. v. Pataki, discussed below, also considers the limitation imposed by the Commerce Clause on what states can do with respect in Net regulation.)
The ability of the CDA to shield civil and criminal defendants is limited. For example, the immunity conferred by another CDA provision which generally protects providers and users of interactive computer services against liability arising out of their attempts to restrict access to Internet material containing sexual or violent content does not extend to public libraries which seek to block the availability of those materials to their patrons. In Mainstream Loudon v. Bd. Of Trustees of the Loudon County Library, the United States District Court for the Eastern District of Virginia considered a motion to dismiss, or grant the defendants summary judgment in, a lawsuit brought by patrons of a public library against the public library for its policy which required site-blocking software to be installed on all library computers so as to block child and hard core pornography and material deemed harmful to juveniles on the ground that the policy impermissibly blocked their access to protected speech in violation of the First Amendment.
The court ruled first that the CDAís immunity provision relied upon by the defendants was unavailing because its purpose was to minimize state regulation of Internet speech by encouraging private content providers to self-regulate against offensive material, and not to insulate government regulation of Internet speech from judicial review. As for the merits of the plaintiffsí First Amendment claim, the court held that a public library may not, without violating the First Amendment, enforce content-based restrictions on access to Internet speech absent a compelling state interest and means narrowly tailored toward that end. The court further ruled that the allegations that the filtering software denied access to such sites as the Quaker Home Page, that the criteria used to make blocking decisions was secret, and that only material fit for children to view was not blocked were sufficient to allege that unreasonable means had been employed such that the law suit could go forward.
The mere fact that Reno v. ACLU held that Internet communications are entitled to the broadest First Amendment protections does not prevent the federal government or states from criminalizing certain types of Internet speech. In United States v. Kufrovich, the United States District Court for the District of Connecticut considered a First Amendment challenge to a prosecution for a violation of a federal statute which made illegal the use of the Internet to persuade a minor to engage in a sexual activity for which anyone may be prosecuted. The court rejected the defendantís argument that both charges against him were based on communications made over the Internet which, he claimed, Reno v. ACLU had rendered "presumptively protected speech." The court held that "otherwise criminal conduct is not protected by the First Amendment simply because it involves the use of speech" and ruled that the First Amendment did not prohibit the defendantís prosecution.
In other criminal contexts, the First Amendment has been found to afford Internet users themselves important protections against federal criminal prosecutions based on the content of e-mail messages that those users send. In United States v. Baker, the United States District Court for the Eastern District of Michigan was confronted with a criminal prosecution based on a statute outlawing threats transmitted in interstate or foreign commerce. At issue were a number of e-mail messages sent by a Michigan resident to an unknown Canadian, which threatened sexual violence against unspecified women and young girls.
The court held that the First Amendment limited what sort of content could be prosecuted regardless whether the medium was an e-mail, regular mail or a phone call. On the facts of the case, the court ruled that no prosecution could be brought with respect to any of the e-mail messages either because no specific potential victim was clearly identified or no statement of an intention to act was present.
In addition to the First Amendment, one court has also held that the federal Constitutionís Commerce Clause severely limits the statesí ability to prohibit certain speech on the Internet. The Commerce Clause, in part, restricts the individual statesí ability to interfere with the flow of interstate commerce directly or indirectly. In American Libraries Assn. v. Pataki, the United States District Court for the Southern District of New York enjoined enforcement of a state criminal statute analogous to the CDA on the basis of the Commerce Clause. (Because the Supreme Court, in Reno v. ACLU, was about to rule on the First Amendment issues raised by the New York Act and because the Commerce Clause analysis determined the outcome of the case, the court did not consider whether the Act ran afoul of the First Amendment.)
The court acknowledged that the Internet represents an innovative instrument of interstate commerce and held that the Act violated the Commerce Clause in three respects. First, the Act in practice unconstitutionally applied to transactions involving citizens of other states because Internet users outside of New York who did not intend to communicate with New Yorkers nevertheless could not prevent New Yorkers from visiting Web sites, joining list servers and so on, and therefore could be subject to prosecution for violations of the Act. Second, the Actís burdens on interstate commerce clearly exceeded any local benefit derived from it in terms of protecting children because the Act was of dubious efficacy; it had no effect on communications originating outside the United States by persons beyond the reach of the New York courts and the Act only governed pictorial messages, not written communications, from pedophiles to minors.
Third, the court held that the Internet is one of those areas of commerce which must be marked off as a national preserve to protect users from inconsistent legislation in the fifty states which, taken to its most extreme, could paralyze development of the Internet altogether. Pataki, while not controlling in any jurisdiction, is likely to be a persuasive opinion and indicates that state attempts at regulation of speech on the Internet must clear not one, but two, hurdles: the First Amendment and the Commerce Clause.
Although, as noted in Zeran, the CDA does not prohibit states statutes or court-made common law from impacting the Internet provided that their laws do not conflict with the CDA, Pataki [cases/pataki/]raises significant questions about what kinds of state laws affecting the Internet are possible in light of the limitations that the Commerce Clause imposes on the states.
Like New York, Georgia passed its own statute regulating content on the Internet and, like the New York statute, Georgiaís law was enjoined, albeit on First Amendment grounds. In brief, in A.C.L.U. v. Miller, the United States District Court for the Northern District of Georgia enjoined a statute which made it a crime for persons knowingly to falsely identify themselves on the Internet and to employ a trade name, registered trademark, logo, legal or official seal, or copyrighted symbol in Internet communications in a manner that would falsely indicate that the person was authorized to use it.
The court ruled that the plaintiffs were likely to prove that the statute imposes content-based restrictions which were not narrowly tailored to achieve the stateís purported compelling interest. Although the stateís interest in fraud prevention was compelling, the statute was not narrowly tailored because the statuteís criminal prohibition applied regardless whether a speaker intended to deceive or any deception had occurred. The court also ruled that the statute was overbroad because it criminalized protected speech such as "the use of false identification to avoid social ostracism, to prevent discrimination and harassment, and to protect privacy." Finally, the court ruled that the statute was void for vagueness because a number of crucial terms were undefined. The result was that Internet users did not know what conduct was prohibited and would self-censor legitimate speech to avoid any possibility of prosecution, and prosecutors and police officers would be empowered by the uncertainty inherent in the statute to single out those with minority views for prosecution.
As in Pataki,in Miller, the dangers of state-by-state regulation of the Internet are inherent in the facts if not explicit in the courtís discussion. While the court disposed of the statute on First Amendment grounds, it could have as easily invoked the Pataki courtís Commerce Clause analysis.
Yet another case which, like Reno v. ACLU and Pataki, considered which communityís standards should be employed to determine obscenity is United States v. Thomas. The case involved the prosecution of a California couple who ran a bulletin board system ("BBS") out of their home that permitted paying members to download pornography which the couple had scanned onto their BBS. The defendants were charged and tried in Memphis, Tennessee where a postal inspector who had bought a membership to the BBS had downloaded photos depicting sadomasochism, bestiality and the like. They were convicted of violating a federal statute outlawing the knowing use of means of interstate commerce (i.e., their computer and phone line) for the purpose of transporting obscene materials in interstate commerce.
The United States Court of Appeals for the Sixth Circuit rejected the defendantsí First Amendment argument that the relevant community standard under which the alleged obscenity of the materials must be determined was not that of Memphis, but rather that of the cyberspace community. The court rejected this argument on the ground that the law is settled that obscenity in the context of interstate distribution is determined by the standards of the community where the trial takes place even if that subjects the defendant to varying community standards. The court did not believe that this rule would lead to an impermissible chill on protected speech because the defendants asked for and received the postal inspectorís address and could have avoided subjecting themselves to the community standards of Memphis by not sending any materials there, unlike the users in Reno v. ACLU and Pataki who had no practical way to determine who might be accessing the information they put on the Net.
Nevertheless, Thomas seems at odds with Patakiís concern about the danger of inconsistent state or local standards hampering an important instrument of national commerce and the need for single national standards governing the Internet. On another point, read together, Thomas and Reno v. ACLU suggest that the provisions of the CDA struck down by the latter could be revived by the advent of new technology, for example, effective tagging.
Stern v. Delphi Internet Services Corp. represents a rare foray into First Amendment law by a state trial court. Stern is noteworthy because the court held that an Internet service provider is a news disseminator and therefore entitled to the same First Amendment protections as a library, a news vendor, a newspaper or a television network in the context of this case. More specifically, the court held that such a provider was entitled to use the photograph of a celebrity running for governor without his permission to advertise its on-line bulletin board service set up to debate his candidacy because of the "incidental use" exception to a state law which otherwise forbade such use. That exception essentially provides that a news disseminator is entitled to display the name and photograph of a person who is the subject of the services it provides for the purpose of attracting and selling those services.
In so ruling, the court observed that the First Amendment protects news distributors of publications, and similarly "[a]ffording [First Amendment] protection to on-line computer services when they are engaged in traditional news dissemination, such as in this case, is the desirable and required result." Like Reno v. ACLU, then, Stern affirms the notion that Internet service providers are entitled to full First Amendment protection.
Religious Tech. Center v. Netcom On-Line Communications Services, Inc., and two related cases demonstrate that the First Amendment provides some limited protection against a seizure of computer equipment from a defendant accused of copyright infringement. Although two other courts permitted the defendantsís actual computer hardware to be seized at least initially, in Netcom, the United States District Court for the Northern District of California refused to follow suit, even though the plaintiff asserted that the seizure was necessary to prevent continued posting to the Internet of its allegedly protected works. The court held that, while an appropriately tailored injunction in a copyright case did not offend the First Amendment, "attempting to shut down a criticís speech activities, including those that do not implicate the copyright laws in the least, would constitute an unwarranted prior restraint on speech." These cases together illustrate the risks run, and potential protections available to, alleged copyright violators who use the Internet.
Finally, Bernstein v. U.S. Dept. of State
is noteworthy because the United States District Court for the Northern
District of California held that both a programmerís source code and the
computer translated object code is speech protected by the First Amendment.
Although Bernstein involved a source
code written to function as an encryption and decryption system, the courtís
broad analysis of computer language would seem to apply to any code in
any language regardless of its function or the nature of its expression.