EPIC logo
Internet "Indecency" Post-Trial Brief

UNITED STATES DISTRICT COURT
EASTERN DISTRICT OF PENNSYLVANIA


_________________________________________
                                         )
AMERICAN CIVIL LIBERTIES UNION, et al.,  )
                                         )
            Plaintiffs,                  )
                                         )
                  v.                     )   Civ. No. 96-963  
                                         )
JANET RENO, in her official capacity as  )
ATTORNEY GENERAL OF THE UNITED STATES,   )
                                         )
             Defendant.                  )
                                         )
_________________________________________)
                                         )
AMERICAN LIBRARY ASSOCIATION, et al.,    )
                                         )
            Plaintiffs,                  )
                                         )
                  v.                     )   Civ. No. 96-1458  
                                         )
UNITED STATES DEP'T OF JUSTICE, et al.,  )
                                         )
             Defendants.                 )
                                         )
_________________________________________)

ACLU PLAINTIFFS' POST-TRIAL BRIEF IN SUPPORT
OF THEIR MOTION FOR A PRELIMINARY INJUNCTION

INTRODUCTION

The 47 plaintiffs in these consolidated actions challenge the constitutionality of portions of the Communications Decency Act of 1996 (the CDA), which criminalize making "indecent" or "patently offensive" online communications available to any person under age 18.(1) The ACLU plaintiffs also assert, and the government concedes, that the CDA's amendment to 18 U.S.C. 1462(c), criminalizing online communications about abortion services, is unconstitutional.(2) Following five days of hearings, the ACLU plaintiffs now submit this post-hearing brief in support of their motion for preliminary relief enjoining the "indecency" and "patent offensiveness" provisions of 47 U.S.C. 223(a)(1)(B), (a)(2), and (d).

In their opening brief, the ACLU plaintiffs argued that the "indecency" and "patent offensiveness" provisions of the CDA were unconstitutionally vague and overbroad, and suppressed valuable, constitutionally protected expression without any showing by the government that its broad criminal ban was the least restrictive means available to achieve any compelling state interest.(3) The testimony at the five days of hearings amply supported these contentions. First, numerous witnesses, including the government's, did not understand or could not give coherent descriptions of what they thought 223(a) and (d) prohibited. Second, on its face and as applied to a broad range of online speech on sexual subjects or containing vulgar words, the CDA is plainly overbroad, sweeping within its terms communications on political, literary, artistic, science, health, and other subjects that are constitutionally protected and have serious value for older minors as well as adults. Indeed, the testimony demonstrated that because of the nature of online communications, a substantial number of content providers, from nonprofit organizations to unmoderated interactive forums, simply have no technologically or economically feasible way of screening out minors; the CDA thus becomes a total criminalization of constitutionally protected "indecent" or "patently offensive" speech.

The government was also unable to establish any compelling state interest that the CDA served -- in part, no doubt, because 223(a) and (d) sweep so broadly and give such unbridled discretion to prosecutors to decide what is indecent or patently offensive according to some undefined community standard. The government certainly did not meet its burden of showing that the statute served its asserted interest in protecting minors by methods that imposed the least restrictive possible burden on First Amendment rights.(4) Indeed, the evidence showed the contrary -- that the CDA, despite its draconian approach, will be ineffective in shielding minors from "indecent" or "patently offensive" material in the almost 50% of online communications that originate abroad, while parental screening mechanisms, by contrast, although admittedly not foolproof, have a far greater probability of actually blocking minors' access to sexually explicit material that may concern their parents or other adults.(5)

In the face of the CDA's overwhelming constitutional deficiencies, the government's response appears to be a proposed rewriting of 223(a) and (d) to cover only (or mostly) "pornography," and of 223(e), the section of the statute outlining "good faith" defenses to criminal liability, to permit or mandate self-labeling by online speakers or other content providers, combined with a presumed or hoped-for voluntary assumption of identity/age screening and blocking duties by the operators of Web browsers or other access providers somewhere down the communications pipeline. But, as we will show in I, infra, the operative terms in (a) and (d) cannot be saved from invalidation by the radical surgery that the government proposes; indeed, the government's own witnesses contradicted any claim that the CDA could be narrowly construed so that it did not include mere vulgar expressions or non-"pornographic" sexual subject matter. Moreover, the government's proposed definitional revisions would not cure the CDA's basic First and Fifth Amendment defects: the Act would still be unconstitutional because it effectively bans protected communications to adults as well as minors.

As to the government's tagging and blocking proposal, it cannot be a coherent defense to a criminal law that one is relying upon voluntary future action by entities not subject to the law -- in this case access providers and the manufacturers of browser software.(6) Moreover, self-labeling as proposed by the government would create new constitutional dilemmas by coercing speech under vague and subjective standards which would inevitably cause self-censorship of a substantial amount of protected and valuable expression. See infra, III.

FACTS

We will not recount in detail here the facts adduced at trial or in the numerous declarations and exhibits submitted by the parties, but will rely upon the Plaintiffs' Joint Proposed Findings of Fact (hereinafter "Prop. Fdgs."), and will cross-reference that document where appropriate. In very brief summary, however, these are the facts of the case:

The ACLU plaintiffs represent a wide variety of online speakers, including providers of information and ideas on World Wide Web sites, bulletin boards, mail exploders, and Usenet newsgroups; receivers of all these various forms of online speech; online access providers; users of e-mail; and participants in newsgroups and real-time chatrooms. The subject matter of their speech ranges from human rights, censorship, pornography, and lesbian and gay issues, to safer sex, rape in prison, and use of the "seven dirty words." None of the plaintiffs are pornographers, as that term is commonly understood,(7) but all of them provide or receive online speech that is sexually explicit or contains vulgar words, including detailed, graphic descriptions of rape and torture, explicit safer sex information and illustrations, discussions of masturbation and the meanings of the word "fuck," or art and literature about sexual relationships and sexual pleasure. Many of the plaintiffs believe that the information and ideas they convey online about these and other sexual issues are valuable, not harmful, for older minors -- a proposition not disputed by the government. Indeed, the record is undisputed that much of the sexually explicit speech that falls within the terms of the CDA has value for older minors, and that even exploitative or distasteful sexual material cannot be presumed to cause psychological harm.

Many of the ACLU plaintiffs are individuals or nonprofit organizations with limited budgets; some rely heavily on volunteer labor. For them, compliance with the CDA would be not only economically burdensome or prohibitive, but technologically unfeasible. Moreover, they do not know what the terms "indecent" and "patently offensive" mean, and so would be at a loss to determine which of their considerable volume of online communications must be shielded from minors under the Act. In addition, many of them believe that free and anonymous access to their speech is critical, so that requiring payment and/or even proof of identity and age will deter substantial numbers of their audience, both adult and minor, from accessing important and even life-saving communications.

The online world is vast, interactive, and quickly evolving. Communities form across local and even national boundaries. Indeed, more than 40% of online communications now originate outside the United States (where the CDA cannot be expected to reach); and that number is increasing.(8) Moreover, given the nature of the online medium, it would be costly and cumbersome if not impossible for many content providers to identify each of their communications that might be "patently offensive" or "indecent" and then screen out all those under 18 who may access those communications. By contrast, less burdensome alternatives such as Surfwatch, PICS, and a variety of other blocking mechanisms, enable parents and other adults much more effectively to shield minors from sexually explicit online material by blocking it at the receiving end.

The evidence in summary showed that the CDA, if not enjoined, will either force online speakers to purge all their communications of content that might be deemed "indecent" or "patently offensive," or radically restrict and restructure cyberspace by eliminating the many parts of the Internet that cannot readily accommodate adult ID checks for every visitor.

ARGUMENT

I. THE CDA BANS "INDECENT" OR "PATENTLY OFFENSIVE" SPEECH IN BROAD AND VAGUE TERMS, AND CANNOT BE REWRITTEN AS THE GOVERNMENT PROPOSES

The Justice Department's efforts to rewrite and thereby narrow the CDA to a statute that might be more constitutionally defensible should not distract the Court from the actual terms of the law, and Congress's purpose in enacting it. Congress's choice of the terms "indecent" and "patently offensive" was not accidental, and cannot simply be equated with "pornography," whatever the government may mean by that term.(9)

As the Supreme Court made clear in Virginia v. American Booksellers Association, 484 U.S. 383, 397 (1988), it is only if a statute is "'readily susceptible' to a narrowing construction" that such an interpretation will be applied to save an otherwise questionable law.(10) Thus, in Blount v. Rizzi, 400 U.S. 410, 419 (1971), the Supreme Court refused to salvage a federal obscenity law by adopting a narrowing construction because it was "for Congress, not this Court, to rewrite the statute." As the Third Circuit ruled recently in rejecting another Justice Department attempt to narrow a penal law, "the starting point is always the language of the statute itself. ... Courts presume that Congress expressed its legislative intent through the ordinary meaning of the words it chose to use." United States v. Knox, 32 F.3d at 744. Further, a court may not "rewrite a statutory scheme and 'create distinctions where none were intended.'" Consumer Party v. Davis, 778 F.2d 140, 147 (3d Cir. 1985) (quoting American Tobacco Co. v. Patterson, 456 U.S. 63, 71 n.6 (1982)).(11)

Congress's use of the terms "indecent"and "patently offensive" derive from Federal Communications Commission v. Pacifica Foundation, 438 U.S. 726 (1978), which upheld the FCC's authority to require time-channeling of constitutionally protected "indecent" radio broadcasts. The FCC derived its authority from 18 U.S.C. 1464, which prohibits both obscenity and "indecency" in radio communications.(12) The FCC at the time defined "indecency" to mean "language that describes, in terms patently offensive as measured by contemporary community standards for the broadcast medium, sexual or excretory activities and organs." See Pacifica, 438 U.S. at 732. This definition in turn derived from the second prong of the Supreme Court's three-part definition of obscenity in Miller v. California, 413 U.S. at 24.

It was understood at the time of Pacifica, and has been the case ever since, that "indecent" (or "patently offensive"), as opposed to "obscene" communications, may lack any of the prurient appeal commonly associated with pornography, and may contain serious literary, artistic, scientific, or other value.(13) Indeed, this was obvious from the "indecency" definition's elimination of the "serious value" and "prurient interest" prongs of the Miller obscenity test, and from the fact that the comic monologue at issue in Pacifica was not by any stretch of the imagination "pornography." (It was merely a numbing comic repetition of the famous "dirty words" in a satiric context.)(14) The FCC's enforcement of its "indecency" standard since Pacifica has, accordingly, encompassed sexually suggestive but nonpornographic comic routines like radio host Howard Stern's, as well as serious literary efforts such as the theatrical work, "Jerker," for the broadcast of which the agency imposed "indecency" sanctions in 1987. See Action for Children's Television v. FCC, 852 F.2d 1332, 1336 (D.C. Cir. 1988); see also John Crigler & William J. Byrnes, Decency Redux: The Curious History of the New FCC Broadcast Indecency Policy, 38 Cath.U.L.Rev. 329, 338-41 (1989) (describing indecision over broadcast of Ulysses); United States v. Evergreen Media Corp., 832 F.Supp. 1183 (N.D.Ill. 1993) (challenge to FCC sanction of radio station that broadcast talk shows containing sexual innuendo). As the FCC made clear in KSD-FM, Notice of Apparent Liability, 6 FCC Rcd 3689 (1990), the news value of a broadcast (in that case, about a highly publicized alleged rape) did not save the broadcast from "indecency"; "the merit of a work is 'simply one of many variables' that make up a work's context; it will not be singled out so as to support an approach 'that would hold that if a work has merit, it is per se not indecent'" (quoting Infinity Broadcasting, 3 FCC Rcd 930, 932 (1987)) (emphasis added).

It was this FCC/ Pacifica standard that Congress deliberately adopted for criminal liability under 223(a) and (d) -- a standard far broader than "pornography." Indeed, the Conference Committee Report explaining (d) explicitly states that Congress was adopting the Pacifica standard, and had considered and rejected the narrower "harmful to minors" test approved in Ginsberg v. New York, 390 U.S. 629 (1968), which exempts from criminal sanctions material that, taken as a whole, lacks prurient appeal, or contains serious value. H.R.Rep. No. 458, 104th Cong., 2d Sess. at 189 (1996). See also 142 Cong.Rec. H1166 (daily ed. Feb. 1, 1996) (statement of Rep. Berman) (conference committee at first accepted "harmful to minors" standard which would have "criminalized exposing children to online pornography ... without chilling entirely nonpornographic, but offensive, expression. However, the House conferees then approved by a 17-to-16 vote an oral amendment offered by Representative Goodlatte to replace the 'harmful to minors' standard ... with a then-unspecified indecency standard").(15) Moreover, the statute on its face applies to libraries and educational institutions; see 223(f)(a).

In an attempt to avoid Congress's explicit adoption of the FCC/ Pacifica standard, the government now pins its hopes on the "in context" qualification in 223(d), arguing that this additional vague term can and should be read to denote a narrowing of "patently offensive" to mean material without serious value. This argument, however, contradicts Congress's deliberate decision to adopt the FCC/ Pacifica standard, and reject the narrower "harmful to minors" standard that specifically exempted material with serious value. If Congress meant "indecent" and "patently offensive" not to include material with any serious value, it had ample opportunity to say so in the statute. Instead, the Conference Committee deliberately rejected such an exemption.(16) In any event, as the federal court explained in Community Television of Utah v. Wilkinson, "time, place and manner," or "context," as factors in determining whether communications are "indecent" (in that case, under a law regulating cable television) "neither provide the fact-finder with sufficient guidance nor adequately limit the scope of the Act to cure [its] constitutional infirmities." 611 F.Supp. 1099, 1117 (D.Utah 1985), aff'd, 800 F.2d 989 (10th Cir. 1986), aff'd, 480 U.S. 926 (1987).

The government next relies on the Conference Report's attempt to rescue (d) from unconstitutional vagueness and overbreadth by suggesting that "in context" implies an exemption for material with "serious value." H.R.Rep. No. 458, supra, at 189. But the Conference Report in this instance flatly contradicts Congress's explicit rejection of the "serious value" exemption under a "harmful to minors" standard or under Rep. Berman's proposal (see n.15, supra); and moreover, willfully misreads footnote 18 in Pacifica to imply that the plurality in that case thought "serious value" inconsistent with "indecency." In fact, footnote 18 of the Pacifica plurality opinion nowhere supports the idea that "indecency" contains a "serious value" exemption. Footnote 18 merely expressed the opinion of the three justices in the plurality that "[a] requirement that indecent language be avoided will have its primary effect on the form, rather than the content, of serious communication. There are few, if any, thoughts that cannot be expressed by the use of less offensive language." 438 U.S. at 743 n.18. This aside hardly amounts to the injection of a "serious value" defense into the FCC's indecency standard; indeed, it appears to acknowledge that expression with serious value may well contain offensive words.(17) Moreover, as noted, the FCC has consistently taken the position that "the merit of a work" does not absolve it of possible "indecency." KSD-FM, supra, 6 FCC Rcd 3689.

Finally, neither the government nor the Conference Report addresses the conspicuous absence of a "prurient interest" element in the CDA's "indecency" and "patent offensiveness" bans. "Prurient interest" is, after all, the defining characteristic of pornography, see, e.g., Roth v. United States, 354 U.S. 476, 487, 489 (1957), and if Congress intended to reach only pornography in the CDA it would therefore have said so by making some reference to the element of prurience.(18) Thus, if the government's theory is that Congress meant to criminalize "softcore pornography" which is not already subject to obscenity laws, i.e., material intended for sexual arousal but not depicting human genitals or ultimate sexual acts, such a theory simply does not square with an "indecency" or "patent offensiveness" standard that specifically includes material with serious value and without prurient appeal.

As the government's own witnesses demonstrated, "indecent" or "patently offensive" may easily be interpreted (and by people who are presumed to have some knowledge of the statute) to include material as wide-ranging as a partially nude image on the cover of a mainstream magazine such as Vanity Fair,(19) or a simple expletive such as "Fuck the CDA."(20) See also 142 Cong.Rec. H1174 (Feb. 1, 1996) (statement of Rep. Goodlatte) (if someone transmits certain passages in The Catcher in the Rye "out of context," they might violate CDA; indecency is "a matter of context and conduct ... [and] must be evaluated by prevailing community standards, not the views of just a few individuals"); 141 Cong. Rec. S8335 (June 14, 1995) (statement of Sen. Feingold) (Lady Chatterley's Lover, The Catcher in the Rye, and many novels by Kurt Vonnegut are potentially "indecent"); 141 Cong.Rec. S17962-63 (statements of Sens. Leahy & Feingold) (four-letter words and online communications about sexual abuse, rape, and AIDS are all potentially within CDA).

Thus, 223(a) and (d) are not "readily susceptible" to the narrowing construction that the government proposes. To the extent that online "pornography" is already subject to criminal prosecution under existing obscenity law,(21) Congress plainly designed its "indecency" prohibitions to range more broadly. To the extent that "indecency" and "patent offensiveness" specifically fail to include exceptions for material with serious value and without prurient appeal, the government's argument flies in the face of congressional intent. Ultimately, the government's argument is not only erroneous but circular, for what is patently offensive remains a subjective and indefinable judgment, whether or not it is patently offensive "in context." If "pornography" has any generally accepted meaning, it is precisely that subcategory of sexually oriented expression that is characterized by its prurient appeal. The CDA's inclusion within "indecency" of material that lacks prurience by itself disposes of the government's argument that the statute only (or mostly) reaches "pornography."

As for the government's proposed "narrowing construction" of the CDA to permit self-labeling as a defense, on the assumption that others may voluntarily undertake adult identification and blocking functions, this is also not a not a construction to which the statute is "readily susceptible." Congress knew how to write a mandatory or even "voluntary" labeling law when it wanted to. See, e.g., 551 of the Telecommunications Act of 1996 (mandating "V-chip" in new televisions and establishing a government ratings board if the industry does not set up a "voluntary" ratings system within a year). Section 502 of the CDA, in contrast to 551, does not require ratings but does establish criminal penalties for on any online speaker who does not assure that her communications are either not indecent, or else are not available to minors.(22) The ACLU plaintiffs are unaware of any precedent for the government's remarkable suggestion that those subject to a criminal law may avoid liability by taking measures well short of what the law requires and then relying for a defense on possible future action by entities not subject to the law. Indeed, as Dr. Olsen acknowledged, Congress "made a considered decision to impose no requirements on entities down the communications chain whatsoever," and no Web browsers currently available have the technology that he proposes.(23)

II. SECTIONS (a) and (d) ARE SUBSTANTIVELY UNCONSTITUTIONAL AS WELL AS IMPERMISSIBLY VAGUE AND SUBSTANTIALLY OVERBROAD

A. Sections (a) and (d) are Properly Subject to Facial Challenge

If, as we have shown, the CDA cannot be rewritten to avoid constitutional invalidation, the government next argues that the plaintiffs' facial challenge is inappropriate if the government can conceive of even one hypothetical situation in which the statute might be applied constitutionally -- to prosecute obscenity, for example, or to permit a (e) defense for commercial pornography providers. The government relies for this argument on New York State Club Association v. New York City, 487 U.S. 1, 11 (1988), and Rust v. Sullivan, 500 U.S. 173, 183 (1990) (a facial challenge may not succeed unless "no set of circumstances exists under which the Act would be valid"). But in making the argument, the government misunderstands New York State Club, Rust, and the entire purpose and nature of First Amendment facial challenges.

Facial challenges are critically important in the area of First Amendment rights precisely because, as the Court explained in New York State Club, statutes drawn too broadly "'create[] an impermissible risk of suppression of ideas.'" 487 U.S. at 11 (quoting City Council v. Taxpayers for Vincent, 466 U.S. 789, 798 n. 15 (1984)) (emphasis added). Facial invalidation is called for where "a statute imposes a direct restriction on protected First Amendment activity, and where the defect in the statute is that the means chosen to accomplish the State's objectives are too imprecise, so that in all its applications the statute creates an unnecessary risk of chilling free speech." Secretary of State of Maryland v. Munson, 467 U.S. 947, 967-68 (1984) (emphasis added) (citing Schaumburg v. Citizens for a Better Environment, 444 U.S. 620, 637 (1980)). It is in this sense that "no set of circumstances exists" under which a facially unconstitutional law "would be valid." Rust, 500 U.S. at 183.(24)

If the law were otherwise, then no statute could ever be facially unconstitutional, for some instance could always be conceived in which it could apply to constitutionally unprotected speech. If the government's argument were correct, for example, then the ordinance invalidated in Coates v. Cincinnati, 402 U.S. 611 (1971), which prohibited people from gathering on sidewalks and conducting themselves "in a manner annoying to persons passing by," would have been upheld because it could have been constitutionally applied "to prevent people from blocking sidewalks, obstructing traffic, littering streets, committing assaults, or engaging in countless other forms of antisocial conduct." Id. at 614. Similarly, the vague law in Smith v. Goguen, 415 U.S. 566 (1971) (criminalizing "contemptuous" treatment of the flag), would not have been invalidated because some "contemptuous"communications might have been constitutionally unprotected "fighting words." All of the cases invalidating laws that unconstitutionally repose too unbridled a discretion in government officers to decide whether and when to permit the exercise of free speech, from Shuttlesworth v. City of Birmingham, 394 U.S. 147 (1969), and Hynes v. Mayor of Oradell, 425 U.S. 610 (1976), to City of Lakewood v. Plain Dealer, 486 U.S. 750 (1988), and Forsyth County v. The Nationalist Movement, 505 U.S. 123 (1992), would have been wrongly decided because some instances could have been conceived in which those government officers might have exercised their discretion with evenhandedness and restraint.

Under the government's theory, even a patently unconstitutional viewpoint discriminatory law, prohibiting, for example, any speech in favor of capitalism, socialism, the Republican Party, or the repeal of the CDA, would not be facially unconstitutional because some speech falling within the statutory prohibitions (e.g., incitement, fighting words, obscenity, or defamation) could legitimately be punished. The Supreme Court in R.A.V. v. St. Paul, 505 U.S. 377 (1992), disposed of any such theory, noting in fact that even an obscenity law would be unconstitutional if on its face it criminalized "only that obscenity which includes offensive political messages." Id. at 387. In R.A.V. itself, the ordinance at issue was facially invalid precisely because it was viewpoint-based, even though it prohibited only constitutionally unprotected "fighting words."

Finally, the doctrine of substantial overbreadth would disappear entirely if the government's view were correct, for that doctrine's fundamental premise is that a law may be facially unconstitutional because of the risk it entails of chilling substantial amounts of protected speech even though by definition the law has constitutionally permissible applications. Thus, for example, in Houston v. Hill, 482 U.S. 451 (1987), an overbroad ordinance prohibiting interference with a police officer could have been applied constitutionally to disorderly conduct or "fighting words," see id. at 464-66; but that did not save the law from facial invalidation. The ordinance "criminalize[d] a substantial amount of constitutionally protected speech," and thus failed to provide the "'breathing space' that 'First Amendment freedoms need ... to survive.'" Id. at 466-67 (quoting NAACP v. Button, 371 U.S. 415, 433 (1963)). See also Gooding v. Wilson, 405 U.S. 518, 520 (1972) ("[i]t matters not that the words appellee used might have been constitutionally prohibited under a narrowly and precisely drawn statute"). If the government's gloss on facial challenges were correct, then the anti-nudity law in Erznoznik v. City of Jacksonville, 422 U.S. 205 (1975); the anti-live entertainment law in Schad v. Borough of Mt. Ephraim, 452 U.S. 61 (1981); and indeed, the anti-"indecency" law in Sable Communications v. FCC, 492 U.S. 115 (1989), would all have been immune from facial challenge because they could have permissibly been used to prosecute obscenity.(25)

B. Substantive Unconstitutionality (Sometimes Characterized as a Type of Overbreadth) Exists Where a Law, Like the CDA, on its Face Subjects Protected Speech to Criminal Penalties

There are two conceptually distinct types of First Amendment facial challenges -- in addition, of course, to vagueness challenges under due process and free speech principles; see IIE, infra. The first type, as noted, is "a challenge to a statute that in all its applications directly restricts protected First Amendment activity and does not employ means narrowly tailored to serve a compelling govermental interest." Maryland v. Munson, 467 U.S. at 965 n.13; see Taxpayers for Vincent, 466 U.S. at 797-98; Schaumburg, 444 U.S. at 637-39. Although this type of facial challenge is sometimes (confusingly) denoted a type of "overbreadth,"see Schaumburg, id. at 639; Munson, 467 U.S. at 965 n.13, it is actually a facial attack on the substantive constitutionality of a law that directly suppresses speech without any compelling justification. See Vincent , 466 U.S. at 797-98; Central Hudson Gas & Electric Corp. v. Public Service Commission, 447 U.S. 557, 565 n.8 (1980) ("[t]his analysis is not an application of the 'overbreadth' doctrine. . . . [T]he Commission's prohibition acts directly against the promotional activities of Central Hudson, and to the extent the limitations are unnecessary to serve the State's interest, they are invalid"). The second type of facial challenge, usually termed "substantial overbreadth," is proper where a statute even though it might be validly applied in some circumstances, "nevertheless is so broad that it 'may inhibit the constitutionally protected speech of third parties.'" Vincent, 466 U.S. at 798; see IID, infra.

The CDA is plainly subject to a substantive First Amendment challenge. On its face, it criminalizes constitutionally protected speech. See Sable, 492 U.S. at 126; Cruz v. Ferre, 755 F.2d 1415 (11th Cir. 1985); Jones v. Wilkinson, 800 F.2d 989 (10th Cir. 1986), aff'd, 480 U.S. 926 (1987); ACLU Brief at 26-34. As the evidence has amply shown, given the nature of cyberspace, the 223(a) and (d) bans on transmitting or displaying "indecent" or "patently offensive" speech in a manner available to minors amounts to an a ban on protected speech for adults.(26) And even if identifying minors and blocking their access to "indecent" or "patently offensive" communications were possible for any but a small portion of online speakers, the CDA would still be unconstitutional because the government has shown neither that it has a compelling interest in barring all persons under 18 from this large and vague category of protected speech, nor that Congress has chosen the least restrictive available means to do so. See Turner Broadcasting System v. FCC, 114 S. Ct. 2445, 2459 (1994); Simon & Schuster, Inc. v. New York State Crime Victims Board, 502 U.S. 105, 118 (1991); Fabulous Associates, Inc. v. Pennsylvania Public Utility Commission, 896 F.2d 780, 784 (3d Cir. 1990).(27)

The Department of Justice itself acknowledged in a letter to Congress opposing the CDA that it would criminalize constitutionally protected speech, and urged that because the legislation "raises complex policy issues that merit close examination prior to Congressional action, ... [w]e recommend that a comprehensive review be undertaken of current laws and law enforcement resources for prosecuting online obscenity and child pornography, and the technological means to enable parents and users to control the commercial and noncommercial communications they receive over interactive telecommunications systems." See 141 Cong.Rec. S8342 (June 14, 1995) (letter to Sen. Leahy). Congress ignored this advice, and although it made minor changes in the CDA after June 1995, it persisted in its determination to criminalize a broad range of constitutionally protected speech without any inquiry regarding whether the CDA was necessary or narrowly tailored to achieve any compelling interest.

C. The Government Has Not Met Its Burden of Showing that the CDA Serves a Compelling State Interest by the Least Restrictive Means Available

1. No Compelling State Interest Has Been Shown in Criminalizing All "Indecent" or "Patently Offensive" Communications that May be Accessed by Minors

The government's claimed compelling interest in protecting minors from constitutionally protected speech on sexual subjects does not extend beyond the specific confines of the "harmful to minors" standard articulated in Ginsberg v. New York.(28) See Sable, 492 U.S. at 126 (citing Ginsberg); ACLU Brief at 43-47. As the Supreme Court explained in Erznoznik v. Jacksonville, 422 U.S. at 212-14:

[M]inors are entitled to a significant measure of First Amendment protection, ... and only in relatively narrow and well-defined circumstances may government bar public dissemination of protected materials to them. ... Speech that is neither obscene as to youths nor subject to some other legitimate proscription cannot be suppressed solely to protect the young from ideas or images that a legislative body thinks unsuitable for them.

There is simply no persuasive precedent for the broader proposition that a three-judge plurality opinion in Pacifica establishes governmental authority to criminalize any "indecent" or "patently offensive" communication in any medium at any time because it may be read or seen by a minor, without even the need to show any actual harm.

When First Amendment rights are at stake, courts cannot defer to a legislative judgment but must make an independent inquiry to assess whether the record establishes a compelling interest justifying governmental regulation. See ACLU Brief at 28-30 (citing Sable, 492 U.S. at 129; Landmark Communications, Inc. v. Virginia, 435 U.S. 829, 843 (1978); Turner Broadcasting, 114 S.Ct. at 2471). Congress considered no evidence, and the government presented no evidence at the hearing, to support a "compelling government interest" in protecting minors from "indecent" or "patently offensive" material in the online medium.(29) In contrast, plaintiffs presented substantial evidence that the categories of "indecent" and "patently offensive" speech include a vast quantity of material that is valuable to minors, especially older minors.

Plaintiffs Critical Path AIDS Project, Wildcat Press, and Stop Prisoner Rape, and Plaintiffs' witness Robert Croneberger of the Carnegie Library all testified about the value to minors of their online information on safer sex, gay and lesbian issues, prison rape, and other sexually explicit materials.(30) Dr. William Stayton, a respected therapist and an expert on human sexuality, with 25 years of clinical experience, testified that exposure to sexually explicit material does not harm minors of any age, and is valuable to older minors. The government offered no evidence to the contrary.(31) In fact, the government's only witness to address this point, Special Agent Howard Schmidt, testified that he believed his own son, by the time he is 17, "will have a sufficient basis of input from [me] and from his life experiences that he would not be harmed even if he were exposed to even exploitative sexually explicit material on line."(32)

U.S. census data reveal that about 123,000 people under age 18 are enrolled in college, 117,000 of them as freshmen.(33) According to the Alan Guttmacher Institute, in 1994 more than half of teenage females and almost 3/4 of teenage males had been sexually active before their 18th birthday.(34) These numbers were generally confirmed by Dr. Stayton.(35) Kiyoshi Kuromiya testified that, according to a March 1996 report of the President's Office of National AIDS policy, one quarter of all new HIV infections in the U.S. are estimated to occur in young people between the ages of 13 and 20.(36) In the face of these realities, the government would be hard-pressed to demonstrate that it has a compelling justification for barring older minors from access to explicit safer sex information or other communications that may help them deal with the onset of sexuality, and indeed, it did not attempt to do so. Thus, the evidence is undisputed that sexually explicit communications are more likely to be of value to older minors than to harm them.

The government has attempted to persuade the Court that it has a compelling interest in barring minors' access to the type of online material reflected in Agent Schmidt's numerous exhibits and testimony regarding his affirmative search for and downloading of many commercial pornographic visual images. But to the extent that this is hardcore material that could be prosecuted under existing obscenity laws, Agent Schmidt's demonstration does not advance the government's case. See ACLU Plaintiffs' Memorandum in Support of Motion in Limine. The obscenity standard articulated by the Supreme Court in Miller v. California, 413 U.S. at 28, 34, clearly covers "hardcore pornography"of the type reflected in Agent Schmidt's collection of lascivious depictions or descriptions of human genitalia or ultimate sexual acts. As the Justice Department made clear in its May 3, 1995 letter to Senator Leahy, there was no need for the CDA because the Department was successfully prosecuting online obscenity, child pornography, and child solicitation, and intended to continue zealously doing so. 141 Cong.Rec. S8342 (June 14, 1995).

In addition, "the unique physical limitations of the broadcast medium" that contributed to the Pacifica plurality's finding of a "compelling interest" in regulating "indecency" in that medium are notably absent in cyberspace. Cf Turner Broadcasting, 114 S. Ct. at 2456. See ACLU Brief at 33-34; Plaintiffs' Prop. Fdgs., III. Dr. Donna Hoffman testified that, unlike radio and television, in cyberspace "individuals do not passively receive information, nor does information suddenly appear."(37) Rather, online users "must seek out the information they want to consume."(38) The courtroom demonstrations of the online medium by Plaintiffs' witness Ann Duvall and Defendant's witness Howard Scmidt aptly illustrated that to get information in cyberspace, you have to go looking for it. In addition, the demonstrations showed that almost all information in cyberspace contains a headline, providing a "warning" to online users of the content they will view if they proceed to download the information.(39)

The government suggests that it has a "compelling interest" in protecting minors from inadvertently encountering "pornography" in cyberspace. Its only evidence, however, was that Agent Schmidt, a law enforcement officer highly trained specifically to hunt down sexually explicit images online, was able to find pornography when he affirmatively looked for it. Even his search for "Jasmine," a popular Disney character, was prompted by his knowledge that "Jasmine" would produce pornographic images.(40) In response to a direct question about the likelihood that an online user would "come across a sexually explicit site by accident," Agent Schmidt admitted that "the odds are slim."(41) Given the slim possibility of encountering "pornography" that is not already subject to prosecution under obscenity laws, the clear warnings that precede this material, and the existence of software programs like SurfWatch that parents can install to screen such content, the government's evidence falls far short of its heavy burden in establishing a compelling interest in the CDA -- which in any event bans a much larger category of speech than "pornography."

2. The CDA is the Most Restrictive Speech Ban in Any Medium

The CDA, on its face, imposes criminal liability on anyone -- whether a single individual or a large organization, whether commercial or non-commercial -- who displays "indecent" or "patently offensive" online communications in a manner available to minors. Indeed, the literal terms of 223(d) would criminalize the actions of parents who, in the course of assisting their children, "use[] an interactive computer service to display" to them online expression that a prosecutor decides is "patently offensive" according to "community standards." The evidence in this case proves that, because the overwhelming majority of speakers in cyberspace have no realistic way to determine the age of each member of their audience, the CDA also operates, on its face, as a ban on "indecent" speech among adults. Such an outright ban has never been upheld for any other medium, even under the guise of protecting children.

In the print medium, constitutionally permissible regulation of speech with sexual content is limited to obscenity, and materials that may fall within the definition of "indecency" or "patent offensiveness" are clearly available to both adults and minors in libraries and bookstores around the country. Even "offensive" material with serious value for a 17-year-old cannot be criminalized.(42) The CDA, by contrast, subjects to two-year prison terms and up to $250,000 in fines anyone who displays in a manner available to minors expressive materials that may be given to minors in written form.

With motion pictures, while the industry has developed a voluntary ratings system to assist parents in determining which films are appropriate for their children, the courts have uniformly rejected government requirements of, or reliance on, content-based labeling of sexually explicit films under penalty of law. See Interstate Circuit, Inc. v. Dallas, 390 U.S. 676 (1968); Swope v. Lubbers, 560 F.Supp. 1328, 1334 (W.D.Mich. 1983); National Association of Theater Owners v. Motion Picture Comm'n, 328 F. Supp. 6 (E.D. Wisc. 1971); Motion Picture Association of America v. Specter, 315 F. Supp. 824 (E.D.Pa 1970) (3-judge court); Engdahl v. Kenosha, 317 F. Supp. 1133 (E.D. Wisc. 1970); see also infra at III. Thus, any parent may choose to take her child to see a movie containing potentially "indecent" material, and any theater's decision to deny entrance to children is not mandated by law.

In television and radio broadcast, while the Supreme Court has held that some restrictions on indecency are constitutionally permissible, it has repeatedly emphasized that this exception is emphatically narrow, focusing on the limited, noncriminal nature of time-channeling and the unique invasiveness of the broadcast medium. Pacifica, 438 U.S. at 750; Sable, 492 U.S. at 128; Bolger v. Youngs Drug Products, 463 U.S. 60, 74 (1983). See also Turner Broadcasting, 114 S. Ct. at 2456. The Supreme Court has never expanded the three-judge plurality opinion in Pacifica to other media. Moreover, when the FCC issued a 24-hour ban on broadcast indecency, the D.C. Circuit rejected it as unconstitutional, Action for Children's Television v. FCC, 932 F.2d 1504 (D.C. Cir. 1991)("ACT II"), and the Supreme Court denied review, 503 U.S. 913 (1992). The D.C. Circuit has since upheld a midnight-to-6 a.m. time-channeling requirement, Action for Children's Television v. FCC, 58 F.3d 654 (D.C. Cir. 1995) ("Act III"); the Supreme Court declined certiorari, 116 S.Ct. 701 (1996), and thus has not revisited the question of the permissible extent of regulation of broadcast "indecency" since Pacifica.(43) In any event, the broadcasting "indecency" regulations, unlike the CDA, are not a total ban, do not impose criminal penalties, are narrowly justified by the special intrusiveness of radio and television, and have essentially been limited to the broadcast context.(44)

The Supreme Court is currently considering the constitutionality of "indecency" regulations in yet another medium, cable television. See Denver Area Educational Television Consortium v. FCC, 116 S.Ct. 471 (1995), granting certiorari in 56 F.3d 105 (D.C.Cir. 1995). The D.C. Circuit in that case did not reach petitioners' First Amendment challenge to provisions of the Cable Television Consumer Protection and Competition Act of 1992 that allowed cable operators to refuse to carry "indecent" or "sexually explicit programming" on cable access channels, because it held that Congress's delegation of such censorship authority to private companies was not state action. 56 F.3d at 123. The court noted that if cable operators' decisions were "treated as decisions of the government, the Commission and the United States would be hard put to defend the constitutionality" of those provisions. Id. at 113.(45) While the ultimate constitutionality of the cable "indecency" regulations has yet to be determined, they are clearly less harsh and restrictive than the CDA's criminal ban. Moreover, "indecency" on non-access cable channels remains completely unregulated, precisely because Congress recognized that a government-imposed ban would be unconstitutional.

Restrictions of sexually explicit commercial phone messages have had mixed success in the courts. In Fabulous Associates, 896 F.2d at 788, the Third Circuit struck down as facially invalid a Pennsylvania state statute that required adults to obtain an advance identification code in order to access sexually explicit phone messages (under the Ginsberg "harmful to minors" standard). In Sable, 492 U.S. at 129, the Supreme Court struck down a statute that banned "indecent" commercial telephone messages, holding that "the congressional record contains no legislative findings that would justify us in concluding that there is no constitutionally acceptable less restrictive means, short of a total ban, to achieve the Government's interest." Congress responded with 223(b) of the Communications Act of 1989, which imposes criminal penalties on commercial entities that knowingly make available "indecent communications for commercial purposes" to any person under 18. Two circuits have upheld the federal restrictions and the Supreme Court has not decided the question. See Information Providers' Coalition v. FCC, 928 F.2d 866 (9th Cir. 1991); Dial Information Services Corp. v. Thornburgh, 938 F.2d 1535 (2d Cir. 1991), cert. denied, 502 U.S. 1072 (1992). Unlike the CDA, 223(b) does not censor the communications of millions of non-commercial speakers, nor does it ban "indecent" or "patently offensive" communications among adults. There was no showing in either Information Providers or Dial that the statute reached constitutionally protected speech that had real value or lacked prurient appeal.

As the comparisons above illustrate, the CDA would impose a ban on sexually explicit speech in cyberspace that is unprecedented in any other medium. Dr. Donna Hoffman testified that online environments "provide a uniquely democratizing means of communication for all people, rich and poor, powerful and less powerful, urban and rural, American and foreign."(46) Howard Rheingold testified that cyberspace "transforms every desktop into a printing press and place of assembly, a component of community-building in a technological society."(47) Scott Bradner testified that the online medium has inspired -- in the absence of government regulation -- remarkable innovations in human communications.(48) Congress has, ironically and unjustifiably, chosen to impose a more restrictive and draconian censorship law than it has ever applied to any other medium, to a medium with vast potential to foster communication, promote democracy, and make the "marketplace of ideas" a reality in the 21st century.

3. The "Defenses" Under 223(e) Do Not Satisfy the "Least Restrictive Means" Test Because They Are Unavailable to the Vast Majority of Online Speakers and Because Existing Voluntary Methods Exist that Permit Parents and Other Adults to Shield Minors from Material they Believe Inappropriate

The government argues that the CDA satisfies the "least restrictive means" test because 223(e)(5) provides defenses for online users who restrict minors' access to "indecent" materials or who require use of a verified credit card or adult access code. This argument ignores the unavailability of these defenses for huge sections of the Internet and the great majority of online speakers, and the immense economic burden such defenses would impose on other users.

First, the evidence at the hearing established that the initial burden of compliance with the CDA would require online speakers to review large quantities of information currently in databases on the World Wide Web and elsewhere to determine whether it is "indecent" or "patently offensive." See Plaintiffs' Prop. Fdgs., IV. Many of the plaintiffs have large numbers of files on their web sites. Mr. Croneberger testified that it would take 180 additional employees to screen all the files at the Carnegie Library.(49) Critical Path, with only one full- and one part-time employee, would find it logistically impossible to review all of the material it makes available for potential "offensiveness" or "indecency."(50) The ACLU has only one full-time staff person responsible for the creation and maintenance of all of the content in the ACLU's online sites, and would have to devote tremendous staff resources to screen its 2000 existing files and the growing amount of new information it puts online.(51) Speakers using other online fora such as e-mail, newsgroups, mail exploders, and real-time chat rooms, would have to make such a judgment each and every time they "spoke" online.(52)

Once an online speaker makes a judgment that some of his communications may be "indecent" or "patently offensive" under the CDA, the speaker must then take "good faith, reasonable, effective, and appropriate actions under the circumstances to restrict or prevent access by minors." 223(e)(5)(A) (emphasis added). Although the statute does not specify what "good faith, reasonable, effective, and appropriate" actions would be, certainly ascertaining the age of each person with whom the speaker communicates online appears to be a necessary first step. Yet many popular web sites receive tens of thousands of visitors(53) each day.(54) As testimony at the hearing established, with these levels of volume at their online sites, the plaintiffs -- like the great majority of online users -- have no viable way to determine the ages of those who access their communications.(55)

The evidence showed that there is no existing technology that allows a speaker to screen for age on newsgroups, mail exploders, and chat rooms -- the "town halls" of cyberspace -- or when using electronic mail.(56) Dr. Olsen conceded this fact.(57) Regarding age screening on the World Wide Web, the only available methods appear to involve either online credit card verification or contracting with an "adult ID" company such as those used by commercial pornographers. In either case, the costs would be prohibitive for the plaintiffs and other nonprofit organizations and individuals who, with limited budgets and volunteer labor, operate large web sites with many visitors.

Plaintiffs presented unrebutted evidence regarding the procedures and cost of setting up credit card verification on a web site.(58) Credit card companies will not verify cards in the absence of a commercial transaction.(59) This option is therefore unavailable to those who wish to supply information online for free, even if they could afford to establish a verification system.

As to cost, Barry Steinhardt testified that it would cost the ACLU, in the first month of operation alone, at least $144,000 to implement credit card screening on its web site.(60) Given a limited yearly budget of $75,000 for the maintenance of all its online resources, the ACLU could not absorb these costs, and would be forced to shut down its site.(61) Patricia Nell Warren testified that credit card screening on the YouthArts web site would cost $9000 for two days. YouthArts is produced entirely by volunteer labor, and has no funding sources -- the cost would be devastating for such a non-profit publication.(62) Passing the cost on to the recipients is not an option for many online speakers, who believe it is important to their mission to continue to provide their information for free.(63) The evidence thus shows that the "credit card" defense, for all practical purposes, is unavailable for non-commercial speakers on the web. The government did not rebut this evidence: its two witnesses knew nothing about the actual operation or cost of adult ID systtems or whether credit card verification could be done without a financial transaction.(64)

In Erznoznik, the Court found an unconstitutional deterrent effect on free speech where, to avoid prosecution, theater owners were required either to "restrict their movie offerings or [to] construct adequate protective fencing which may be extremely expensive or even physically impracticable." 422 U.S. at 217. See also, e.g., Fabulous Associates, 896 F.2d at 785 (noting constitutionally unacceptable cost of adult access code requirement for sexually explicit phone messages). The burdensome deterrent effect of the CDA is much more harsh than that ruled unconstitutional in Erznoznik and Fabulous: the technical impossibility of screening for age on newsgroups, chat rooms, and mail exploders would force online users either to censor constitutionally protected speech or to abandon these online fora entirely. Similarly, the cost of screening for age on the World Wide Web would be prohibitive for non-commercial speakers and non-profit organizations and would exclude large numbers of minors and adults (those without credit cards) from accessing constitutionally protected material.

The government's only response to Plaintiffs' substantial evidence on the technological and economic infeasibility of the CDA defenses was the testimony of Dr. Dan Olsen. Dr. Olsen proposed a number of possible variations on a hypothethical "-L18" tagging and blocking scheme for preventing minors from having access to potentially "indecent" material.(65) Under Dr. Olsen's proposal, all online speakers would have to examine each of their communications to determine if any of them were subject to the CDA, then "tag" each such potentially "indecent" word, idea, image, or document. Age screening and blocking would then be done by speakers themselves or through contracts with "third parties." Alternatively, client browser software could be reconfigured to read tags and block access from the receiving end; it is not clear under this scenario who would do age screening or activate the blocking mechanism. Dr. Olsen devised this scheme two weeks before testifying, and did not submit it to any peer review.(66)

Dr. Olsen's scheme contains several flaws that render it useless in the government's attempt to establish the constitutionality of the CDA. First, Dr. Olsen's scheme does not solve the immense initial burden of requiring all online speakers to review their content to determine whether it is "indecent" or "patently offensive."(67) Mandatory labeling also raises serious First Amendment problems of its own, see III, infra. Moreover, to the extent that Dr. Olsen's scheme relies on online speakers themselves to screen for age, the proposal fails to alleviate the overwhelming logistical and economic burdens of that task. Dr. Olsen conceded that there is no way for speakers to verify age when communicating via newsgroups, mail exploders, chat rooms, and e-mail.(68) He also conceded that the only existing way to verify age on web sites was by credit card.(69) Dr. Olsen had no answer to plaintiffs' evidence that implementation of credit card screening is only available to commercial providers, costs a great deal of money and time to implement, and excludes large numbers of adults from accessing constitutionally protected speech.(70) Thus, it is only to the extent that Dr. Olsen relies on the hypothetical voluntary actions of third parties to verify age, and of browser companies to reconfigure their software, that the government can even make an argument that the CDA would not be unconstitutionally burdensome for at least some online speakers.

Second, other than relying on speakers themselves to screen and block, Dr. Olsen's plan would require the cooperation of many entities other than speakers themselves -- entities that are not subject to liability under the CDA.(71) He proposes that speakers could be relieved of the burden of screening for age if Netscape re-configured its browser to block tagged material at the receiving end. But there is nothing in the Act to require Netscape to change its browser, or to give speakers assurance that client-end blocking would even comply with the CDA.(72) Dr. Olsen also proposes that third-party verification systems could relieve the speakers' burden of verifying age. But the only such systems that currently exist on the web require a credit card or some other form of payment, and provide adult access codes explicitly for the purpose of obtaining commercial pornography(73) -- an option that is costly, and is certain to inhibit many users in search of nonpornographic communications such as free AIDS information or human rights reports.

Finally, Dr. Olsen's tagging scheme is only hypothetical and is not available to online speakers today.(74) In fact, Olsen estimated that the restructuring of the Internet to accommodate his tagging scheme would take at least one to two months.(75) It thus provides little comfort to the millions of online users who will immediately risk real-life criminal prosecution if the CDA is not enjoined.

In contrast to Dr. Olsen's burdensome scheme, the voluntary blocking mechanisms that are currently in place, such as Surfwatch, Net Nanny, and Cyberpatrol, or that soon will be, such as PICS, will address the government's concern that minors not access pornographic material in a far more effective and less burdensome manner than the CDA. See IIC4, below; discussion in ALA Plaintiffs' Post-Trial Brief. Thus, even if the Court thought that the government's interest in the CDA were compelling, the government has completely failed to meet its burden to overcome the presumption of unconstitutionality by establishing that the CDA is the least restrictive means to achieve its purpose.

4. The CDA is Far Less Effective in Achieving the Government's Goal Than are Less Burdensome Existing Alternatives

To pass the strict scrutiny test, the government bears the burden to "'show that its regulation is necessary to serve a compelling state interest and is narrowly drawn to achieve that end.'" Simon & Schuster, 502 U.S. at 117 (quoting Arkansas Writers' Project v. Ragland, 481 U.S. 221, 231 (1987)). This means that the government must show there are "no adequate alternatives," Carey v. Brown, 447 U.S. 455, 465 (1980), i.e., that its "restriction serves a compelling interest and that there are no less restrictive alternatives." Fabulous Associates, 896 F.2d at 787. (Indeed, even under the intermediate scrutiny applicable to content-neutral regulations and commercial speech, a law "may not be sustained if it provides only ineffective or remote support for the government's purpose," Central Hudson, 447 U.S. at 564; the government bears the burden of showing that its scheme will "in fact alleviate . . . the alleged harms in a direct and material way." Turner Broadcasting, 114 S.Ct. at 2470.)

Yet because of the global nature of the online medium, even the CDA's total ban will be ineffective at ridding online networks of "indecent" or "patently offensive" material. See ACLU Brief at 31-32. As Professor Hoffman testified, the number of foreign-based sites on the Internet is growing, and will soon reach 50%.(76) Even if the Court accepts the argument that the CDA covers only (or mostly) "pornography," the government admits that a large number of "pornographic" sites originate outside the United States. Even under Dr. Olsen's scheme, only tagged sites would be blocked from a minors' view, and all untagged material -- including pornography from foreign sites not subject to the CDA -- would reach minors.(77) Therefore, the government's proposal, even if accepted, would do nothing to prevent minors from accessing many "pornographic" sites.

By contrast, user-based technologies like Surfwatch and PICS can effectively block foreign sites. In fact, under the PICS proposal, a parent could block all unrated sites.(78) Moreover, some blocking mechanisms offer the advantage of enabling parents to block whatever content they feel is inappropriate, consistent with their values, whether it be sexual, violent, or religious in nature.(79)

The government has argued that surely the First Amendment does not render it powerless to prevent minors from accessing online pornography (Def. Brief in Opp. to ALA Motion for P.I. at 4). The existence of less burdensome and more effective alternatives to the heavy criminal artillery of the CDA is one answer to the government's concern. Another is the vigorous enforcement of existing obscenity laws. To the extent that the government is concerned that parents are not able to control their children's online activity, education, advocacy, and promotion of parental control tools are possible remedies, as Agent Schmidt testified.(80) Finally, Congress, after studying the issue to determine what, if any additional legislation might be needed (as the Justice Department urged in its May 1995 letter), could attempt to fashion a narrowly drawn law to address any clearly identified problem.

D. Sections 223(a) and (d) are Substantially Overbroad

The government's main argument in response to the claim that (a) and (d) are unconstitutionally overbroad is to say that none of the plaintiffs has standing to assert the claim. According to the government, substantial overbreadth is really only a standing doctrine; a party whose speech is constitutionally protected may not bring a First Amendment overbreadth challenge, but is confined to contesting the constitutionality of the law as applied to her or him. This theory ignores the importance of entertaining facial challenges where overbroad statutes have an unacceptably deterrent effect on free expression. Taxpayers for Vincent, 466 U.S. at 798; New York State Club, 487 U.S. at 11. It also ignores the Supreme Court's explicit rejection of the government's theory in Board of Trustees v. Fox, 492 U.S. 469, 484 (1989).

1. The ACLU and ALA Plaintiffs Have Standing to Bring a Substantial Overbreadth Challenge

As the Supreme Court in New York State Club affirmed, a statute may be facially challenged on First Amendment grounds either if it "'could never be applied in a valid manner'" because "every application ... create[s] an impermissible risk of suppression of ideas," id. at 11 (quoting Taxpayers for Vincent, 466 U.S. at 798)(81), or if it is "substantially overbroad," i.e., it might be validly applied to some expression but it still risks an unacceptable chilling effect, Munson, 467 U.S. at 466; Vincent, 466 U.S. at 798. It is in these "substantial overbreadth" cases that the doctrine of third party standing arose, to permit facial challenges to laws that are not so overbroad as to risk suppression of speech in all their attempted applications, but that nevertheless may inhibit free expression "almost as easily by the potential or threatened use of power as by the actual exercise of that power." Id.; see also Broadrick, 413 U.S. at 611-13; Brockett v. Spokane Arcades, 472 U.S. 491, 504 (1985).

The government turns this third party standing doctrine on its head to argue that because all of the plaintiffs in this case are asserting their own First Amendment rights (in addition to the rights of their members and readers), none of them has standing to bring a substantial overbreadth challenge. For this proposition, the government mistakenly relies on Brockett, 472 U.S. at 504, and Kreimer v. Bureau of Police, 958 F.2d 1242, 1265 (3d Cir. 1992).

Brockett involved a statute that defined "prurient" to include "lasciviousness or lust," 472 U.S. at 494; the unconstitutionally overbroad term "lust" could easily be excised, and the law thus legitimately narrowed and only partially invalidated. Id. at 504.(82) Similarly, Kreimer held that an ordinance relating to library patronage simply did not restrict a substantial amount of protected speech, i.e., it was not "substantially overbroad." 958 F.2d at 1265. In contrast to the broad sweep of 223(a) and (d), in Kreimer there was no "'realistic danger that the statute itself will significantly compromise recognized First Amendment protections of parties not before the Court ...'" Id. (quoting Taxpayers for Vincent, 466 U.S. at 801).

The Supreme Court's overbreadth cases reveal that the third party standing doctrine is an expansion of, not a limitation on, a party's right to bring a facial challenge. See Board of Trustees v. Fox, 492 U.S. at 484. In Schaumberg, for example, which struck down a charitable solicitation ordinance on substantial overbreadth grounds, the plaintiff Citizens for a Better Environment (CBE) was almost surely engaged in constitutionally protected solicitation. The Court noted that it did not matter whether or not this was so, 444 U.S. at 634; there was no suggestion that CBE's challenge would be precluded in the event its conduct was constitutionally protected. The point of substantial overbreadth/third party standing doctrine was that it did not matter.(83)

Similarly, in Erznoznik, 422 U.S. at 212-13, an ordinance prohibiting drive-in movie theaters from exhibiting any films containing nudity was facially overbroad even though a narrower statute, i.e., limited to material deemed obscene as to minors within the meaning of Ginsberg, 390 U.S. 629, might have passed muster. The plaintiff's speech was constitutionally protected, even as to minors. See 422 U.S. at 213-14.(84) Yet there was no suggestion that he lacked standing to bring a facial overbreadth challenge. Indeed, such a result would be counterintuitive and counterproductive: as the Erznoznik Court said, the unconstitutional "deterrent effect" of the ordinance was "both real and substantial," id. at 217, and this was true regardless of whether the individual challenging the ordinance had engaged in protected or unprotected speech.(85) See also, e.g., Houston v. Hill, 482 U.S. at 458-67 (no suggestion that plaintiff gay rights activist had engaged in unprotected activity); Airport Commissioners v. Jews for Jesus, 482 U.S. 569, 574-76 (1987) (overbroad ban on "First Amendment activities" at airport subject to challenge whether or not plaintiff's own speech might properly be prohibited; case-by-case adjudication to bring the law within constitutional bounds would be "intolerable" because of the "chilling effect ... on protected speech in the meantime"); Schad v. Borough of Mt. Ephraim, 452 U.S. at 66 (unnecessary to decide what free speech protection should be extended to nude dancing offered by appellants because their claims were "rooted in the First Amendment, and they are entitled to rely on the impact of the ordinance on the expressive activities of others as well as their own"). As the Court explained in Board of Trustees v. Fox, 492 U.S. at 484, "while the overbreadth doctrine was born as an expansion of the law of standing, it would produce absurd results to limit its application strictly to that context."

2. Sections (a) and (d) are Substantially Overbroad

For the reasons set forth in the opening briefs of the ACLU and the ALA plaintiffs, the CDA's "indecency" prohibitions are classic examples of substantially overbroad laws that are not written with anything like the precision required of statutes regulating speech. See ACLU Brief at 43-51; Cruz v. Ferre, 571 F.Supp. 125, 130-31 (S.D.Fla. 1983), aff'd, 755 F.2d 1415 (11th Cir. 1985) ("indecency" ban is unconstitutionally overbroad); Community Television v. Wilkinson, 611 F.Supp. at 1106-09 (same). This is particularly the case where, as with the CDA, the law regulating speech is criminal in nature. See, e.g., Houston v. Hill, 482 U.S. at 459 (courts must examine statutes that criminalize speech with particular care).

The government cannot, despite its attempt, succeed in "narrowing" the coverage of the CDA to avoid its substantial overbreadth. First, as we have already discussed, substituting "pornography" for "indecent" or "patently offensive" speech contradicts the CDA's clear purpose and legislative history. See I, supra. Second, substituting the undefined term "pornography" does not by itself necessarily narrow (a) and (d), particularly where the government's explanation of what constitutes pornography is circular: that is, sexually explicit material that, "in context," is considered "patently offensive" according to some (undefined) "contemporary community standard." Indeed, the government's own witnesses testified that they understood "indecent" or "patently offensive" to include such politically inspired vulgarisms as "Fuck the CDA";(86) and such relatively tame partial nudes as the Demi Moore photo on the cover of Vanity Fair, or leather-clad women posing for Playboy.(87)

It is no defense to a First Amendment overbreadth challenge that the government can conceive of some circumstance in which speech encompassed by 223(a) and (d) could constitutionally be punished, and that the government promises or suggests that its officers will exercise their discretion responsibly to prosecute only such speech. See IIE, infra. A formidable array of First Amendment precedent establishes that it is simply not constitutionally acceptable to leave such unfettered discretion to public prosecutors under facially overbroad laws. E.g., Hill, 485 U.S. at 465 and cases cited at n.15; Broadrick, 413 U.S. at 612-13; Shuttlesworth v. Birmingham, 394 U.S. at 157; Thornhill v. Alabama, 310 U.S. at 97-98. .

E. Sections (a) and (d) are Unconstitutionally Vague

For the reasons set forth in our opening brief, see ACLU Brief at 34-43, the CDA is unconstitutionally vague. And nothing in the government's attempted "narrowing" of the statute cures the problem -- even assuming such a judicial rewriting were permissible; see I, supra. For the Justice Department neither defines (other than in circular fashion) what it means by "pornography," nor explains whether and to what extent such "pornography" is not already subject to prosecution under existing obscenity laws.(88) In essence, the government is asking the plaintiffs and the Court to "trust us"; even though the statutory terms are vague, we will only prosecute purveyors of "pornography." This is not a permissible defense to a First Amendment challenge; on the contrary, a major vice of vague criminal laws is precisely that they give unfettered discretion to prosecutors. E.g., Hynes v. Mayor or Oradell, 425 U.S. at 622. As the Court said in Baggett v. Bullitt, 377 U.S. 360, 373 (1964), "well-intentioned prosecutors and judicial safeguards do not neutralize the vice of a vague law."(89)

The government's case, if anything, has highlighted rather than obscured the intolerable vagueness of the CDA. The responses offered by government witnesses Schmidt and Olsen to the Court's questions illustrated just how freewheeling the subjective, discretionary judgments of police and prosecutors under the CDA would be. Schmidt, for example, opined that an extremely explicit safer sex illustration of how to put a condom on an erect penis would not be subject to prosecution because its "context" was educational, not "purely for pleasure purposes."(90) He seemed to have the same view about depictions of "couples copulating" in highly explicit erotic Indian sculptures, so long as they were presented in an "cultural" or "educational" context rather than on one of the commercial pornography sites that he surveyed.(91) By contrast, a much less explicit partial nude on the cover of Vanity Fair magazine would be covered, according to this experienced federal law enforcement officer, because the "context" was not primarily educational, but was "for fun, basically ... more for fun than anything else."(92)

Similarly, Dr. Olsen opined that any of the "seven dirty words" made famous by the Pacifica decision, or their synonyms, could be subject to (a) and (d), and should therefore be "tagged," as should nudes even if displayed on a museum web site.(93) Whatever the government may now say in extenuation of these remarkably broad and varied interpretations of the CDA, they surely illustrate that its operative terms are hopelessly vague and subjective. If an experienced federal law enforcement officer and a computer specialist, both prepared by the government and presumably knowledgeable about its interpretation of the law, could so radically differ from the Justice Department's lawyers as to what the law really means, its vagueness is manifest. "'No one may be required at peril of life, liberty or property to speculate as to the meaning of penal statutes,' and this is particularly true of laws 'having a potentially inhibiting effect on speech ...'" Hynes v. Mayor, 425 U.S. at 620 (quoting Lanzetta v. New Jersey, 306 U.S. 451, 453 (1939), and Smith v. California, 361 U.S. 147, 151 (1959)).

Finally, according to Special Agent Schmidt, "community standards," and therefore criminal liability under the CDA, would vary depending on whether the material was uploaded in Minnesota, New York, or another venue -- despite the fact, undoubtedly known to Schmidt, that online communications are accessible virtually anywhere regardless of their "community" of origin.(94) The government has not yet indicated whether it believes, as its witness evidently does, that "community standards" under the CDA are local, or whether they are national or even global. This uncertainty highlights yet another way in which the statute is intolerably vague, for a local standard is logistically impossible for most of cyberspace, as the great majority of online speakers do not and cannot know the location of every person who accesses their speech, and certainly cannot block access depending upon location.(95) As for imposing a national or international standard of "patent offensiveness," communities nationally and worldwide are simply too diverse, cultures too disparate, and variations far too great, for "community standards" to supply any measure of specificity, no less the precision that the First Amendment requires. See Miller v. California, 413 U.S. at 30 (recognizing that national standard of "offensiveness" cannot be articulated).(96)

F. The CDA Violates the Constitutional Right to Receive Information Privately and Anonymously

For the reasons already set forth in the ACLU's opening brief, at 52-56, the CDA's requirement that online speakers whose communciations might be "indecent" or "patently offensive" identify and determine the age of anyone seeking access to those communications violates the well-recognized First Amendment and privacy rights to receive information anonymously in certain circumstances. Those circumstances are amply present here with respect to safer sex information, gay and lesbian online resources, rape, human rights and domestic abuse issues, and a host of other communications dealing with sexual subjects. See Plaintiffs' Prop. Fdgs., IV. The CDA's requirement that every online user provide identification and proof of age before accessing material that may be offensive or indecent directly infringes the right to anonymity, with potentially disastrous effects in some cases, for example, for those who need access to safer sex information.

Mr. Kuromiya testified confidential access to AIDS information is crucial for many users of Critical Path AIDS Project.(97) Similarly, because of the trauma of prisoner rape and the stigma of incarceration, some of those on the Stop Prisoner Rape mailing list have asked to remain anonymous.(98) It is important that access to the Queer Resouces Directory remain anonymous if the user so desires because many visitors to the site are fearful about revealing their sexual orientation.(99) As Dr. Stayton testified, the feelings of adults and minors about sexual subjects, particularly sexual orientation or traumas such as rape, often require assurances of confidentiality.(100)

The decision in favor of anonymity may be motivated by fear of economic or official retaliation, by concern about social ostracism, or merely by a desire to preserve as much of one's privacy as possible.

[ . . .]

Anonymity is a shield from the tyranny of the majority. It thus exemplifies the purpose behiind the Bill of Rights, and of the First Amendment in particular: to protect unpopular individuals from retaliation -- and their ideas from suppression -- at the hands of an intolerant society.

McIntyre v. Ohio Elections Commission, 115 S.Ct. 1511, 1516, 1524 (1995) (citation omitted).

III. THE GOVERNMENT'S PROPOSED "NARROWING" CONSTRUCTION OF (e) DOES NOT SALVAGE THE CDA AND IN FACT CREATES ADDITIONAL FIRST AMENDMENT PROBLEMS

Evidence at the hearing suggested that, in an effort to avoid the constitutional infirmities of the CDA, the government is attempting to rewrite the defenses in 223(e) to transform the Act into a mandatory labeling or rating statute. For the reasons discussed in I, supra, there is no basis for substituting content labeling either as a requirement of or as a viable defense under the Act. Even if the Court were to accept the government's radical rewriting of 223(e), however, the proposal has serious constitutional defects.

First, the government acknowledges that self-rating or "tagging" does not in itself constitute compliance with the CDA; someone still has to do the screening and blocking, and under the government's scenario, it would probably be the content providers themselves, either through credit card verification or under contract with "third parties."(101) The government's proposal thus hardly resolves the problems of technical and economic infeasibility that plague the CDA. See IIB3, supra. Moreover, self-labeling is neither simple nor inexpensive, for as the government's expert Dr. Olsen acknowledged, someone must review every document to determine if it is potentially offensive or indecent. This includes not only content providers on Web pages, bulletin board archives, and FTP and gopher sites, but occasional contributors to newsgroups, mail exploders, chatrooms, MUDs, and MUSEs.

Second, the proposal creates its own constitutional problem, since mandatory self-labeling amounts to unconstitutional forced speech. As the Court recently affirmed in Turner Broadcasting, 114 S.Ct. at 2458, the First Amendment prohibits "Government action that . . . requires the utterance of a particular message favored by the Government." It is thus well-established that the First Amendment protects "both the right to speak freely and the right to refrain from speaking at all." Wooley v. Maynard, 430 U.S. 705, 714 (1977); see West Virginia State Board of Education v. Barnette, 319 U.S. 624, 645 (1943); Miami Herald Publishing Co. v. Tornillo, 418 U.S. 241, 254-58 (1974); Pacific Gas & Electric Co. v. Public Utilities Commission, 475 U.S. 1, 9 (1986) (First Amendment prohibits government from forcing people to "alter their speech to conform with an agenda they do not set").(102)

Critical Path AIDS Project, the American Civil Liberties Union, Wildcat Press, the American Library Association, and other plaintiffs believe that their sexually explicit online materials are valuable to both adults and minors. Their organizational beliefs affirmatively oppose censorship of any kind, including forced labeling. As Barry Steinhardt testified, because of the ACLU's strong belief in free speech, "we would object to being forced to rate ourselves."(103) A labeling law would force the plaintiffs to censor their own speech in direct violation of their viewpoints, and to present their information and advocacy to the public in a stigmatizing and misleading manner by lumping their socially valuable speech into a vague and overbroad category that includes pornography.(104)

Forced labeling also violates the First Amendment because it inhibits both the speaker's right to express and the listener's right to access constitutionally protected material. In order to avoid the extensive burden of categorizing and then labeling all potentially "indecent" or "patently offensive" speech, see discussion supra at IIC3, many online speakers will simply refrain from addressing sexually explicit subjects entirely; or else, rather than be forced to label and thus risk being "foreclosed . . . from a significant portion" of their audience, would choose to say "nothing but the innocuous." Interstate Circuit, 390 U.S. at 684. See also Bella Lewitzky Dance Foundation v. Frohnmayer, 754 F. Supp. 774, 783 (C.D. Cal. 1991) (recognizing chilling effect of requirement that NEA grant recipients certify that funds would not be used to produce obscene material). The stigma of an "indecency" label would also inhibit many online users from accessing a wide range of socially valuable speech on subjects like AIDS education and gay and lesbian issues. (See Plaintiffs' Prop. Fdgs., IV.) Yet the First Amendment clearly protects against "inhibition as well as prohibition" of free speech. Lamont v. Postmaster General, 381 U.S. 301, 309 (1965) (Brennan, J., concurring). It prohibits both direct government censorship and "more subtle governmental interference." Bates v. City of Little Rock, 361 U.S. 516, 523 (1960).

Self-labeling or certification requirements, like mandatory oaths, are particularly troublesome where the underlying standards are inherently vague. "The vice of vagueness is particularly pronounced where expression is sought to be subjected to licensing." Interstate Circuit, 390 U.S. at 683. The potential for vague laws to chill protected speech is not "rendered less objectionable because the regulation of expression is one of classification rather than direct suppression." id. at 688-89,(105) and the vagueness problem is exacerbated, not cured, when the speaker himself must become the censor. See Baggett v. Bullitt, 377 U.S. at 372 (vague statute will cause forced oathtaker to "'steer far wider of the unlawful zone' than if boundaries of the forbidden areas were clearly marked"; conscientious oathtakers can only avoid threatened sanction "by restricting their conduct to that which is unquestionably safe") (quoting Speiser v. Randall, 357 U.S. 513, 526 (1958)).(106) Since many pornographers can move offshore or post materials through anonymous remailers, self-labeling of online content is most likely to chill other kinds of speakers, "whose conscientious scruples [a]re the most sensitive." Cramp v. Board of Public Instruction, 368 U.S. 278, 286 (1961).(107)

Mandatory labeling also amounts to a prior restraint to the extent that, as Dr. Olsen described it, content providers would either have to silence themselves entirely until the self-labeling process was complete,(108) or initially tag entire files or sites as "indecent" during the "month or two" it might take to review all files for potentially indecent or offensive content.(109) The government's witness thus contemplates an extended period of self-censorship of communications clearly not covered by the Act, as online speakers engage in the laborious process of reviewing material that they would have to tag initially just to be on the safe side. This amounts to a prior restraint on protected speech which, as the Supreme Court has repeatedly said, "bear[s] a heavy presumption against its constitutional validity." Bantam Books v. Sullivan, 372 U.S. 58, 70 (1963).(110) Dr. Olsen's proposal betrays an insensitivity to free speech values that is simply staggering.

Government attempts to impose government-mandated rating systems have been struck down repeatedly by the courts. In Interstate Circuit, 390 U.S. at 678-80, the Supreme Court invalidated a city ordinance that imposed a misdemeanor penalty for showing movies deemed by a classification board "not suitable for minors." In Soundgarden v. Eikenberry, 123 Wash.2d 750, 871 P.2d 1050, cert. denied, 115 S. Ct. 663 (1994), the Washington Supreme Court struck down a law requiring "erotic" music recordings to be labeled "adult only." See also National Association of Theater Owners v. Motion Picture Comm'n, 328 F. Supp. 6 (E.D. Wisc. 1971) (striking ordinance that empowered commission to classify movies "adult only" and to revoke theater license if exhibitor failed to submit movie for classification); Motion Picture Ass'n v. Specter, 315 F. Supp. 824 (E.D. Pa. 1970) (striking statute that imposed criminal sanctions for showing films "not suitable for children").(111) Thus, even if the Court were to accept the government's proposal to rewrite the CDA as a labeling statute, it would still be unconstitutional.

IV. PRELIMINARY RELIEF IS WARRANTED BECAUSE PLAINTIFFS WILL SUFFER IRREPARABLE HARM IF THE CDA IS NOT ENJOINED

Plaintiffs fully satisfy the requirements for preliminary injunctive relief. The evidence presented during the five-day hearing, and through declarations and exhibits, establishes that a broad range of valuable speech is potentially subject to the CDA and will be chilled if 223(a) and (d) are not enjoined. Given that even temporary suspension of First Amendment rights is irreparable harm, and as the government has not presented any evidence of a problem that cannot be better addressed by existing obscenity laws and parental control mechanisms than by the CDA, preliminary relief will also safeguard rather than harm the public interest. See ACLU Brief at 24-26, 58-59.(112)

CONCLUSION

For the foregoing reasons, the Plaintiffs' motion for preliminary injunctive relief should be granted.

Respectfully submitted,

                      /s/ Marjorie Heins                  
                      Christopher A. Hansen 
                      Marjorie Heins 
                      Ann Beeson 
                      American Civil Liberties Union 
                      132 West 43 St. 
                      New York, NY 10036 
                      212-944-9800
                      Laura K. Abel 
                      Catherine Weiss 
                      Reproductive Freedom Project 
                      American Civil Liberties Union Fdn. 
                      132 West 43 St. 
                      New York, NY 10036 
                      212-944-9800
                      Stefan Presser 
                      ACLU of Pennsylvania 
                      125 South Ninth St. #701 
                      Philadelphia, PA 19107 
                      215-923-4357
                      David L. Sobel 
                      Marc Rotenberg 
                      Electronic Privacy Info. Center 
                      666 Pennsylvania Ave. SE  #301 
                      Washington, DC 20003 
                      202-544-9240
                      Michael Godwin 
                      Electronic Frontier Foundation 
                      1550 Bryant St.  #725 
                      San Francisco, CA 94103 
                      415-436-9333
                      Roger Evans
                      Legal Action for Reproductive Rights
                      Planned Parenthood Federation Of America 
                      810 Seventh Avenue
                      New York, New York 10019
                      212-261-4708

April 26, 1996


NOTES

1. 47 U.S.C. §§223(a)(1)(B), (a)(2), and (d), as amended or added by Title V, §502, of the Telecommunications Act of 1996, Pub. L.No. 104-104, 110 Stat. 56 (1996).

2. Judge Buckwalter denied a TRO as to 18 U.S.C. 1462(c), noting that "at this early stage of the litigation, it seems clear that no irreparable harm will befall plaintiffs." ACLU v. Reno, No. 96-963, slip op. at 2 n.1 (E.D.Pa. Feb. 15, 1996). Plaintiffs agree that given the Justice Department's statement that it will not enforce 1462(c), they are not in imminent danger of irreparable harm, and therefore they do not seek a preliminary injunction against the enforcement of 1462(c). Plaintiffs reserve the right, however, to seek permanent relief against the enforcement of this provision, for it is clear that the government's nonbinding statement of policy could change at any time -- certainly, with a change in Administration -- see, e.g., United States v. Knox, 32 F.3d 733, 739 (3d Cir. 1994), cert. denied, 130 L.Ed.2d 782 (1995).

3. The ACLU plaintiffs do not intend to repeat here the arguments set out in our opening brief. This Introduction merely summarizes those arguments. In this brief, we will concentrate on how those arguments are informed by the evidence adduced at the hearing, and on new issues raised by the evidence or by the government's briefs. Where appropriate, we will cross-reference portions of our opening brief (Plaintiffs' Brief in Support of Motion for Temporary Restraining Order and Preliminary Injunction, filed Feb. 8, 1996, or "ACLU Brief").

4. As noted in the Plaintiffs' Joint Response to the Court's Questions Regarding the Burden of Proof, it is well-settled law that the government bears the burden of proof on both the existence of a compelling state interest in suppressing speech, and the absence of less restrictive alternative means of accomplishing that interest.

5. The ACLU plaintiffs also argued in their opening brief that the CDA infringed free speech and privacy rights connected with the anonymous accessing of online communications; see ACLU Brief at 52-53; infra at IIF.

6. To the extent the government's proposed alternatives for complying with (e) rely on content providers' undertaking the laborious tasks of age identification and/or blocking, they do nothing to ease the untenable burdens imposed by the Act. See IIB & C, infra.

7. "Pornography" is not a legal term, although the Supreme Court in Miller v. California, 413 U.S. 15, 27-28 (1973), appeared to equate depiction of "hard core" sexual conduct with "'hard core' pornography," and ruled that only such hard-core pornography could be prosecuted under obscenity laws. Plaintiffs use the term "pornography" here in its usual sense, to describe sexually explicit material whose primary purpose is sexual arousal and whose claims to serious literary, artistic, or other value are generally meager or nonexistent.

8. See Parties' Stipulations in Preparation for the Preliminary Injunction Hearing 3 ("[t]oday . . . 9,400,000 host computers . . . are linked to the Internet"); Direct Testimony Declaration of Dr. Donna Hoffman ("Hoffman Test. Decl.") 107-112; Testimony of Dr. Donna Hoffman ("Hoffman Testimony"), Tr. Vol. II, at 68-69. Dr. Hoffman testified that the number of servers on the World Wide Web is doubling approximately every 2 months; the number of host computers has been doubling annually since 1981 or '82. Id. at 59:5-10. As of 1995, "[t]he Internet connect[ed] more than 45,000 separate networks and 25 to 30 million users in more than 100 countries." Fred H. Cate, The First Amendment and the National Information Infrastructure, 30 Wake Forest L.Rev. 1, 34 (1995).

9. In its opening briefs, the government was carefully equivocal about what exactly it meant by narrowly construing (a) and (d) to reach only (or primarily) "pornography," and nowhere actually stated what it meant by "pornography" or whether it viewed non-"pornographic" sexually explicit speech or vulgar words as entirely outside the coverage of (a) and (d). The government's opposition to the ACLU Plaintiffs'Motion for a TRO and Preliminary Injunction, at 2, asserted that "patently offensive" was equivalent to "pornographic," while its response to the ALA Plaintiffs' Preliminary Injunction Motion, at 2, was more equivocal: the formulation there was "sexually explicit subject matter and language that, while 'patently offensive,' does not rise to the level of being legally obscene." As we will show infra, IIE, both formulations are essentially circular because they rely on subjective and indefinite notions of what may be "patently offensive."

10. The Court articulated this principle as long ago as United States v. Reese, 92 U.S. 214, 221 (1875), when it said: "It would certainly be dangerous if the Legislature could set a net large enough to catch all possible offenders and leave it to the courts to step inside and say who could be rightfully detained and who should be set at large. This would, to some extent, substitute the Judicial for the Legislative Department of the Government."

11. Although a federal court has somewhat more leeway when it is construing a federal as opposed to a state law, it still may not substitute its judgment for Congress's and impose different terms on a statute than those that Congress specifically chose. Compare Blount v. Rizzi, 400 U.S. at 419, with United States v. 37 Photographs, 402 U.S. 363, 369-70 (1971); see Eubanks v. Wilkinson, 937 F.2d 1118, 1124-25 (6th Cir. 1991) (37 Photographs is "a very narrow exception to the general rule" that courts do not "add words of limitation to statutes because they are aware of the dangers of intruding on the legislative function").

12. The Pacifica plurality rejected the argument that, to avoid constitutional difficulties, "indecent" should be interpreted as synonymous with "obscene," as the Court had done in earlier cases, e.g., Hamling v.United States, 418 U.S. 87, 105 (1974); United States v. 12 200-Foot Reels, 413 U.S. 123, 130 n.7 (1973).

13. See Pacifica, 438 U.S. at 740 ("[p]rurient appeal is an element of the obscene, but the normal definition of 'indecent' merely refers to nonconformance with accepted standards of morality"); id. at 741 n.16 (nudity, or scenes from Lady Chatterley's Lover, although constitutionally protected, would raise serious "indecency" concerns if broadcast); Cruz v. Ferre, 755 F.2d 1415, 1418 (11th Cir. 1985) (cable TV ordinance with "indecency" definition almost identical to FCC's is unconstitutional in part because it "does not require that the challenged materials, 'taken as a whole, appeal to the prurient interest in sex,'" and "does not inquire whether the materials, 'taken as a whole, do not have serious literary, artistic, political, or scientific value'") (quoting Miller v. California, 413 U.S. at 24).

14. Indeed, in view of the fact that Pacifica was not about pictures at all, but about vulgar words, it is difficult to understand how the government can assert that Congress's deliberate and explicit adoption of the Pacifica test encompassed pornographic pictures but excluded vulgar words.

15. Rep. Berman went on to explain that a further amendment was proposed to define the indecency standard to include the third prong of the Miller-Ginsberg "harmful to minors" test -- that is, "to include statutory language clarifying that the indecency standard included only material that 'taken as a whole, lack[s] serious literary, artistic, political or scientific value for minors.' I and others supported this proposal in an effort to avoid criminalizing display of valuable material that might nevertheless be considered 'patently offensive' according to the standards of some local communities. However, the proposal was rejected ..." (emphasis added).

16. "Patent offensiveness" simply does not equal "lack of serious value," as any consideration of the history of censorship will attest. Many now-acknowledged masterpieces of art and literature, from Joyce's Ulysses to Manet's "Dejeuner Sur L'Herbe" and "Olympia" were considered highly offensive in their time. See also ACLU Brief at 29 n.63 (offensiveness is not a permissible standard for censoring speech); Papish v. University of Missouri, 410 U.S. 667, 670 (1973) ("dissemination of ideas -- no matter how offensive to good taste -- ... may not be shut off in the name alone of 'conventions of decency'").

17. In addition, the notion that the First Amendment permits the government to dictate polite "form" so long as the "content" of speech is not censored has never had the endorsement of more than three justices, and is probably wrong as a matter of First Amendment law. See, e.g., Cohen v. California, 403 U.S. 15 (1971); Gooding v. Wilson, 405 U.S. 518 (1972); Papish v. University of Missouri, 410 U.S. 667 (1973); Hustler Magazine v. Falwell, 485 U.S. 46 (1988); Texas v. Johnson, 491 U.S. 397 (1989).

18. Nor does the government explain why it thinks the "indecency" ban in 223(a) -- as opposed to the "patent offensiveness" ban in 223(d) -- contains a secret "serious value" exemption. "Indecency" in (a) is wholly undefined, and its meaning is not addressed in the Conference Report.

19. Schmidt testimony, Tr. Vol. IV, at 138-39.

20. Olsen testimony, Tr. Vol. V, at 53-54.

21. See 18 U.S.C. 1462 and 1465 and 47 U.S.C. 223(a).

22. The Conference Committee rejected a proposed amendment that would have included content-labeling as a defense. See Rep. Rick White, Leg. Counsel draft, Nov. 17, 1995 (suggesting defense based on "appropriate measures to restrict [access by] minors, including content selection standards ..."); see 142 Cong.Rec. 1145, 1165-66 (Feb. 1, 1996).

23. Olsen Testimony, Tr. Vol. V, at 43; Vol. IV, at 214.

24. In Munson, the Supreme Court held facially invalid a statute that prohibited charitable organizations, in connection with any fundraising activity, from paying expenses of more than 25% of the amount raised. Despite the government's argument that the statute conceivably could be applied constitutionally to some charities, the Court struck down the statute because "in all its applications [it] falls short of constitutional demands. ... [A] percentage limitation on fundraising unnecessarily restricts protected First Amendment activity." 467 U.S. at 965 n. 13.

25. Indeed, the whole doctrine of enhanced standing in substantial overbreadth cases presupposes that some plaintiffs' speech could permissibly be regulated under a properly drawn law, but not under an overbroad law that is facially unconstitutional because of its chilling effect. See IID, infra.

The ACLU Plaintiffs, moreover, agree with the ALA Plaintiffs that this Court could, if it deemed it necessary for purposes of the preliminary injunction motions, also consider their as-applied challenge to the CDA.

26. See Plaintiffs' Prop. Fdgs, III, IVB-C; Hoffman Test. Decl., 67; Rheingold Test. Decl., 14-15; Bradner Test. Decl. 29, 49-50, 71.

27. The government makes a half-hearted suggestion that something less than the compelling state interest/strict scrutiny standard should apply to the CDA's explicitly content-based criminalization of protected speech. There is no basis, however, for for this Court to apply anything less than strict scrutiny. The Pacifica plurality's intimations that "indecency" could be considered "low value" speech had the support of only three justices; and Pacifica in any event has repeatedly been limited to its facts. See Sable, 492 U.S. at 126, 129; ACLU Brief at 41; Turner Broadcasting, 114 S. Ct. at 2459 (content-based laws require "the most exacting scrutiny"). Even in the special context of broadcasting, strict scrutiny applies to "indecency" regulation. See Action for Childlrens Television v. FCC, 58 F.3d 654, 656, 659-64 (D.C.Cir. 1995) ("ACT III"), cert. denied, 116 S.Ct. 701 (1996).

Nor do the Supreme Court's cases allowing cities to regulate the location of adult bookstores and theatres through zoning have any application to the CDA, as zoning regulations do not (and constitutionally cannot) constitute a total ban; they apply only to commercial businesses; they must be justified by a showing of "negative secondary effects" from such businesses in the surrounding community; and they impose only civil rather than criminal penalties for violation. See, e.g., Renton v. Playtime Theatres, 475 U.S. 41 (1986); Young v. American Mini Theatres, Inc., 427 U.S. 50, 75 (1976) (Powell, J., concurring).

28. Courts since Miller v. California have modified the Ginsberg standard to be consistent with the Miller test, rather than the earlier "Roth/Memoirs" obscenity test. See cases cited in ACLU Brief at 45-46 n.85; Erznoznik, 422 U.S. at 213 n.10 (Court has "not had occasion to decide what effect Miller will have on the Ginsberg formulation").

29. Congress ignored suggestions that it should hold hearings to determine whether additional criminal prohibitions beyond the laws already governing obscenity, child pornography, harassment, and solicitation were necessary. See Justice Department Letter to Sen. Leahy, 141 Cong.Rec. S8342 (June 14, 1995); 142 Cong.Rec. H1165 (statement of Rep. Berman) (Feb. 1, 1996); 142 Cong.Rec. S694 (statement of Sen. Leahy) (Feb. 1, 1996) (no witnesses at Senate Judiciary hearing on earlier version of CDA addressed constitutionality of indecency standard). Congress also rejected an amendment proposed by Senator Leahy that would have required the Justice Department to study whether the CDA was in fact necessary. See 141 Cong. Rec. S8327 (June 14, 1995); Jonathan Wallace and Mark Mangan, Sex, Laws, and Cyberspace 178 (1996).

30. See Plaintiffs' Prop. Fdgs. IIIE; Kuromiya Test. Decl. 16, 21, 26; Kuromiya Testimony, Tr. Vol. I, at 181:12-182:12; Warren Test. Decl. 11-24, 45-46; Donaldson Test. Decl. 19, 26-27.

31. While the government relied heavily on an excerpt from the 1986 Meese Commission Report on Pornography (Def. Ex. 80) to attempt to discredit Dr. Stayton's testimony through cross-examination, Dr. Stayton testified on re-direct that the study was almost universally discredited within the field, and that testimony was not rebutted by the government. Tr. Vol. I, at 214:12-215:8. Moreover, the assertions contained in the Meese Commission Report are plainly inadmissible as both hearsay and non-expert opinion; and are contradicted by several other exhibits (Def. Exs. 81-90), including scholarly articles finding no harm to minors from exposure to sexual subject matter, and the 1970 Report of President Nixon's Commission on Obscenity and Pornography (Def. Ex. 83), which came to the same conclusion. The government initially proferred and then, after the close of the testimony, attempted to withdraw these exhibits.

32. Schmidt Testimony, Tr. Vol. IV, at 130:24-131:7. Schmidt was admittedly not an expert in any aspect of family relations, including the asserted "inability" of parents to supervise their children online; see Tr. Vol. IV, at 74:23-76:11. He acknowledged that adults, not children, are the ones who buy computers and establish passwords/accounts for their use. Id. at 130:12-14.

33. U.S. Dep't of Commerce, Bureau of the Census, School Enrollment - Social and Economic Characters of Students: October 1992, table 1 (Oct. 1993).

34. Alan Guttmacher Institute, Sex and America's Teenagers 19 (1994).

35. Stayton Testimony, Tr. Vol. I, at 217-19; Stayton Test. Decl. 9, 10.

36. Kuromiya Test. Decl. 22.

37. Hoffman Test. Decl. 66-69; Hoffman Testimony, Tr. Vol. II, at 26:7-11.

38. Hoffman Test. Decl. 66.

39. See Schmidt Testimony, Tr. Vol. IV, at 49:17-18, 92:12-93:22.

40. Schmidt Testimony, Tr. Vol. IV, at 105, 135 (reading from the deposition: "Because one of my searches on line, I came up with the WEB site name Jasmine that related to some exotic, sensual aids, and sex things . . . .").

41. Id. at 103.

42. While many states have enacted "harmful to minors" laws which apply to the print medium, these laws use the variable "obscene as to minors" standard articulated by the Supreme Court in Ginsberg, presumably as modified by Miller. See note 28 supra.

43. Because some stations were allowed to broadcast "indecency" after 10 p.m., a distinction the D.C. Circuit considered unjustified, the actual ruling upheld only a 10 p.m. to 6 a.m. "safe harbor."

44. The Pacifica plurality recognized that a criminal penalty for broadcast "indecency" would pose greater First Amendment concerns. 438 U.S. at 739 n.13, 726 n. 25 ("[e]ven the strongest civil penalty at the Commission's command does not include criminal prosecution"), and 750 ("[w]e have not decided . . . that this broadcast would justify a criminal prosecution").

45. The D.C. Circuit upheld one provision that was obviously state action as well; it required cable operators who did not ban "indecent" leased access programming to segregate and scramble it until subscribers requested unscrambling. Although burdensome, this form of segregation and blocking is considerably less restrictive than a criminal ban.

46. Hoffman Test. Decl. 69.

47. Rheingold Test. Decl. 3.

48. Bradner Testimony, Tr. Vol. I, at 166:25-167:19.

49. Croneberger Test. Decl. 32.

50. Kuromiya Test. Decl. 36.

51. Steinhardt Test. Decl. 4, 11, 21; see also Godwin Supp. Decl. 3 (Electronic Frontier Foundation web site contains 13,939 files); Troyer Supp. Decl. (Safer Sex Web Page contains 967 files).

52. Kuromiya Test. Decl. 26-31, 38 ("the CDA defenses are simply not available to Critical Path and its users when they communicate through" e-mail, mailing lists, Usenet newsgroups, and chat rooms).

53. While it is impossible to measure with accuracy the number of visitors to a given web site, web sites can ordinarily measure both the number of "hits" and the number of "unique host computers" from which users access the site. A "hit" is a measure of the number of times users have accessed a given file. "Hits" provide an upper bound figure of the number of people who visit a site. Hoffman Testimony, Tr. Vol. II, at 30:1-31:2, 33:16-21. A "unique host computer" is the address of the computer accessing the web site. A single computer often represents a number of online users, making the "unique host computer" figure a lower bound estimate of the number of visitors to a web site. Hoffman Testimony, Tr. Vol. II, at 60:6-61:21, 63:13-64:14, 67:12-16.

54. See, e.g., Godwin Supp. Decl. 3 (Electronic Frontier Foundation web site logs over 50,000 unique host computers each day); Steinhardt Test. Decl. 5 (ACLU web site logged 67,458 unique host computers during February 1996); Kuromiya Test. Decl. 37 (10,000 online users access the Critical Path AIDS Project each day); Donaldson Test. Decl. 6 (in one week, Stop Prisoner Rape web site logged 1,932 unique host computers from 39 countries); Warren Test. Decl. 38 (YouthArts web page received 9000 hits in two days).

55. See Hoffman Testimony, Tr. Vol. II, at 71:108; Steinhardt Test. Decl. 21; Donaldson Test. Decl 13; Kuromiya Test. Decl. 19; Warren Test. Decl. 25; Plaintiffs' Prop. Fdgs., IV.

56. Kuromiya Test. Decl. 26-31, 38; Steinhardt Test. Decl. 23; Donaldson Test. Decl. 13, 22; Warren Test. Decl. 25; Bradner Test. Decl. 29, 49-50..

57. Olsen Testimony, Tr. Vol. IV, at 219:23-220:7, 231:14-23, 233:18-234:25, 256:8-257:2.

58. Kuromiya Test. Decl. 36; Steinhardt Test. Decl. 25-33; Tr. Vol. III, at 183-186; Warren Test. Decl. 31-33, 36, 38-39; Bradner Test. Decl. 71; Plaintiffs' Prop. Fdgs., IV.

59. Steinhardt Test. Decl. 26; Anker Test. Decl. 21; Croneberger Test. Decl. 27; Olsen Testimony, Tr. Vol IV, at 197:19-23, 198:6-9.

60. Steinhardt Test. Decl. 29; Steinhardt Testimony, Tr. Vol. III, at 173:7-21, 185:7-17.

61. Steinhardt Test. Decl. 30, 35; Steinhardt Testimony, Tr. Vol. III, at 184:608.

62. Warren Test. Decl. 37. Ms. Warren testified that electronic magazines for young people have proliferated precisely because the cost of production is much lower, and the distribution much simpler, than for print media. Warren Test. Decl. 36; Warren Testimony, Tr. Vol. I, at 189:1-16.

63. See Kuromiya Test. Decl. 24 ("[c]ost should not be a barrier between . . . users and life-saving AIDS prevention and treatment information.").

64. Schmidt Testimony, Tr. Vol. IV, at 72; Olsen Testimony, Tr. Vol. V, at 197:5-201:2, 220:21-23.

65. Olsen Test. Decl 43-111.

66. Olsen Testimony, Tr. Vol. IV, at 207:11-14.

67. Olsen Testimony, Tr. Vol. IV, at 238:21-242:2 ("it could be a big task").

68. Id. at 219:23-220:7, 231:14-23, 233:18-234:25, 256:8-257:2. Dr. Olsen also failed to consider whether his scheme would work for large online content systems unconnected to the Internet, such as America Online, Compuserve, and Prodigy. (Id. at 215:25-216:5.)

69. Olsen Testimony, Tr. Vol. IV, at 217:6-12, 218:2-9.

70. Dr. Olsen proposed that, in order to obviate the delay of providing a credit card every time a visitor accessed a web site, each web site could issue an adult access code for future use. (Olsen Test. Decl. 80-86.) This also fails to solve the burden problem, as an adult online user would have to obtain a different adult access code every time he attempted to access or link to a new web site. It also adds an additional burden for speakers -- issuance and maintenance of a database of registered users. (Steinhardt Test. Decl. 28; Kuromiya Test. Decl. 37.)

71. Id. at 209:25-210:6, 211:8-11, 212:6-9.

72. Olsen Test. Decl 103-104; Olsen Testimony, Tr. Vol. IV, at 213:23-214:13.

73. Olsen Testimony, Tr. Vol. IV, at 251:13-255:8.

74. Olsen Testimony, Tr. Vol. IV, at 214:7-215:7.

75. Id. at 254:13-17, 256:1.

76. Hoffman Testimony, Tr. Vol. II, at 68-69.

77. Olsen Testimony, Tr. Vol. IV, at 226:4-18.

78. Olsen Testimony, Tr. Vol. IV, at 222:10-24.

79. Olsen Testimony, Tr. Vol. V, at 64:22-66:11.

80. Schmidt Testimony, Tr. Vol. IV, at 75:21-76:11; see also discussion in ALA Plaintiffs' Post-Trial Brief..

81. As noted in IIA, supra, 223(a) and (d)'s proscriptions on "indecent" or "patently offensive" speech amply satisfy this standard.

82. See also Lind v. Grimmer, 30 F.3d 1115, 1122 (9th Cir. 1994), cert. denied, 130 L.Ed.2d 786 (1995) (distinguishing Brockett, where statute's unconstitutional application was limited, and permitting substantial overbreadth challenge by party to whom a law was unconstitutionally applied because the law had "numerous other potential applications that are unconstitutional").

83. This was also the point in the case that originated the overbreadth doctrine, Thornhill v. Alabama, 310 U.S. 88 (1940): it did not matter whether the speech of the defendant was constitutionally protected. Id. at 96. Thornhill was a criminal prosecution in which the Court could simply have ruled that the law was unconstitutional as applied to the defendant. Instead, as the Court explained, facial invalidation was called for because "[i]t is not merely the sporadic abuse of power by the censor but the pervasive threat inherent in its very existence that constitutes the danger to freedom of discussion. ...A like threat is inherent in a penal statute, like that in question here, which does not aim specifically at evils within the allowable area of State control but, on the contrary, sweeps within its ambit other activities that in ordinary circumstances constitute an exercise of freedom of speech or of the press. The existence of such a statute, which readily lends itself to harsh and discriminatory enforcement by local prosecuting officials, against particular groups deemed to merit their displeasure, results in a continuous and pervasive restraint on all freedom of discussion that might reasonably be regarded as within its purview." Id. at 97-98.

84. Erznoznik was prosecuted for showing the R-rated film, "Class of '74"; see 422 U.S. at 206 n.1.

85. Even if a plaintiff in a substantial overbreadth case were required to show that his speech was unprotected, it is certainly possible that at least one of the plaintiff organizations or its members have displayed or will display in a manner available to minors some communication somewhere in cyberspace that could be constitutionally barred from access by at least younger minors, e.g., some of the nastier portions of the Marty Rimm study, which is available online from libraries and other sources. See Pl. Exs. 49, 203, 204.

86. Olsen Testimony, Tr. Vol. V, at 53-55.

87. Schmidt Testimony, Tr. Vol. IV, at 97-100, and Exs. 42 & 43.

88. Any "hardcore" pornography online is potentially subject to obscenity prosecution; see n. 7, supra; United States v. Thomas, 74 F.3d 701 (6th Cir. 1996). Thus, the exhibits submitted to the Court in connection with Special Agent Schmidt's testimony, which depict or describe sexual acts or lasciviously exhibit human genitalia do not advance the government's argument that there is a clearly discernible category of "pornography" targeted by the CDA that is not already covered by obscenity law.

89. See also, e.g., Keyishian v. Board of Regents, 385 U.S. 589, 599 (1967) ("[i]t is no answer" to a vague law for the government "to say that the statute would not be applied in such a case").

90. Tr. Vol. IV, at 137:5-138:19.

91. Id. at 140:20-141:10.

92. Id. at 138:23-139:14.

93. Olsen Testimony, Tr. Vol. V, at 53-55, 75-76.

94. Schmidt Testimony, Tr. Vol. IV, at 137-38.

95. A relatively limited exception may exist for commercial bulletin board operators like the defendants in U.S. v. Thomas, who track the location of every member who signs up for their services. See 74 F.3d at 711 (because the defendants knew where members were downloading their images, court found no constitutional problem with applying local Tennessee community standard for obscenity to online material that originated in California).

96. As noted in our opening brief, ACLU Brief at 40-41, the Court in Pacifica never considered whether the FCC's indecency standard was unconstitutionally vague.

97. Kuromiya Test. Decl. 21, 23, 35.

98. Donaldson Test. Decl. 14.

99. Casti Supp. Decl. 5, 10, 12; see also Steinhardt Test. Decl. 17; Pokempner Supp. Decl. 11.

100. Stayton Test. Decl. 14.

101. The only alternative the government proposed that relieves content providers of the laborious if not impossible burden of age-identifying and blocking was the suggestion that third parties who themselves have no obligations under the Act may develop screening and blocking as a result of market forces. As we have already demonstrated, this is not a coherent interpretation of the statute, because it depends on speculation about future voluntary actions by entities not subject to the Act. See I, IIB, supra.

102. See also Harper & Row Publishing, Inc. v. Nation Enterprises, 471 U.S. 539, 559 (1985) (citing Wooley on importance of "the right to refrain from speaking at all").

103. Steinhardt Testimony, Tr. Vol. III, at 192:3-4.

104. Olsen Testimony, Tr. Vol. IV, at 251:13-255:8.

105. See also Southeastern Promotions, Ltd. v. Conrad, 420 U.S. 546, 556 n.8 (1975) (quoting and approving this proposition in Interstate Circuit).

106. As Dr. Olsen recognized, content labeling decisions under an "indecency" standard would be sufficiently difficult in many instances to require consultation with an attorney. Olsen Testimony, Tr. Vol. IV, at 246:1-5.

107. Testimony at the hearing illustrated the problems of self-rating under vague standards. Albert Vezza testified that PICS rejected even a voluntary ratings system that relied solely on self-rating, because such a scheme would be arbitrary. Vezza Test. Decl. 16. A forced labeling scheme using the vague and overbroad "indecency" standard would transform the millions of online speakers into censors whose "individual impressions [would] become the yardstick of action, . . . result[ing] in regulation in accordance with the beliefs of the individual . . . rather than regulation by law." Interstate Circuit, 390 U.S. at 685 (quoting Kingsley Int'l Pictures Corp. v. Regents, 360 U.S. 684, 701 (1959)).

108. Olsen Testimony, Tr. Vol. IV, at 242:16-21.

109. See Olsen Testimony, Vol. IV, at 242:4-14, 254:13-17, 256:1, Vol. V, at 75-76. Dr. Olsen referred to this approach as "triage." Id. at Tr. Vol. IV, at 84:8-10.

110. Delays in the dissemination and receipt of communications are particularly devastating in the online medium, where high value is placed on the ability to post and access information quickly. The First Amendment does not permit government to delay the distribution of constitutionally protected speech: "A delay of even a day or two may be or crucial importance in some instances." A Quantity of Books v. Kansas, 378 U.S. 205, 224 (1964) (Harlan, J., dissenting).

111. Meese v. Keene , 481 U.S. 465 (1987), is easily distinguished from the labeling proposed by the government in the present case. In Meese, the Supreme Court reversed an injunction prohibiting the Attorney General from applying a "political propaganda" label to three films the plaintiff wished to distribute. The Court did not consider the constitutionality of the underlying labeling statute. Id. at 467. In addition, unlike the CDA, the label did not place the materials "beyond the pale of legitimate discourse," id. at 480, nor did failure to comply with the labeling and reporting scheme result in criminal sanctions.

112. The government appeared to argue in its opposition to the ACLU's Motion for a TRO, at 32 n.19, that preliminary relief is not warranted in advance of the FCC's issuance of recommendations under 223(e)(6). That argument is specious. If not preliminarily enjoined, the CDA will be enforceable immediately with no further guidance from the FCC, threatening millions of online speakers with criminal sanctions and chilling vast quantities of constitutionally protected speech. At best, the CDA grants authority to the FCC only to issue recommendations regarding specific measures that could be considered evidence of a "good faith" defense under 223(e). It does nothing to prevent prosecutions, and indeed, since the CDA explicitly denies any enforcement power to the FCC, prosecutors are not bound by any FCC recommendation.


Return to the EPIC Censorship Lawsuit Page