Introduction

    The Internet has grown rapidly in the past decade.[1]  Likewise, the number of children using the Internet, both at home and at school, has increased significantly.[2]  A majority of the public schools and the public libraries in the United States are connected to the Internet.[3]  When using the Internet, children occasionally access Web pages containing inappropriate materials, usually after a series of mouse clicks.[4]  Arguably, some children may deliberately search for sexually explicit material.[5]  However, there is a lack of support showing that exposure of minors to inappropriate material “is a problem of ‘epidemic’ proportions.”[6]

    Both federal and state government have long had an interest in protecting children from harmful materials, including those available on the Internet.[7]  Additionally, obscenity and child pornography have never had constitutional protection.[8]  Moreover, both are illegal under various statutes.[9]  However, while the First Amendment[10] does not protect obscene speech, it does protect speech that is merely indecent.[11]

    Congress has passed criminal statutes and various other statutes to protect minors from receiving inappropriate material.[12]  Over the past few years, various plaintiffs have filed lawsuits challenging the constitutionality of these statutes.[13]  Several courts have held that the acts are unconstitutional, including the most recent Act.[14]

    On April 20, 2001, the Children’s Internet Protection Act (CIPA) took effect.[15]  CIPA requires schools and public libraries that receive “E-rate” subsidies to use filtering software to protect children from pornography.[16]  Under the E-rate program, eligible schools and libraries may apply for discounted eligible telecommunications, Internet access, and internal connection services (infrastructure to support networked computers).[17]  CIPA requires schools and libraries wishing to receive E-rate funding to certify that they enforce a policy of Internet safety, including using a “technology protection measure” that denies any user access to obscene pictures, child pornography, or images harmful to minors.[18]  Those schools and libraries failing to meet certification requirements lose their E-rate funding.[19] 

    In Reno v. American Civil Liberties Union,[20] the United States Supreme Court held that the Internet “is entitled to ‘the highest protection from governmental intrusion.’”[21]  However, as discussed in Mainstream Loudoun v. Board of Trustees of the Loudoun County Library,[22] a library installing filtering software may encounter constitutional problems under the First Amendment.[23]  The constitutional rights of minor students seeking information is “not automatically coextensive with the rights of adults” in similar situations.[24]  Additionally, while conditions of employment may place some restrictions on the teacher, most school districts have policies on academic freedom so long as the teacher implements the curriculum adopted by the district.[25]

    Education regulation has traditionally been a function of state and local government.[26]  Many states include computer instruction as part of the curriculum.[27]  Indeed, schools receive E-rate funding in order to help local school boards, especially those in poor areas, afford Internet access and networking capabilities.[28]  According to Senator Rick Santorum, a co-author of CIPA, “[i]t is vitally important that our students learn how to use the Internet and explore the many educational opportunities it provides, and the local community should play a critical part in ensuring that this is done in a safe and proper manner.”[29]

    CIPA leaves schools and libraries that receive funding under the Act with very little choice—they can either install filters on all computers or lose their funding.[30]   According to a recent Consumer Reports survey, only about two percent of all Web sites contain sexually explicit material.[31]  In total, there are approximately 2,469,940,685 Web pages.[32]  Some libraries and schools choose not to filter, as the amount of money received through CIPA is relatively small in comparison to their overall budget.[33]  Others choose not to filter because they feel their patrons should have access to all materials.[34]

    Part I of this Note examines the history of various congressional attempts to prevent children’s access to indecent material.  This includes background on the Communications Decency Act and the Child’s Online Protection Act, as well as a detailed discussion of the various provisions of CIPA.  Part II discusses the various types of computer blocking filters and their effectiveness.  Part III discusses the lawsuit challenging the constitutionality of CIPA and the district court’s opinion.  Finally, Part IV discusses the impact on students and teachers if the Supreme Court ultimately finds CIPA constitutional.  The Note concludes that CIPA, like its predecessors, will likely fail under a strict scrutiny standard of review and that there are less restrictive means of accomplishing Congress’s goal.

I. Acts Designed To Prevent Children’s Access To Inappropriate Internet Sites

    In the past several years, Congress has made several attempts to pass legislation restricting offensive material transmitted over computer networks.[35]  The United States Supreme Court held that the Communications Decency Act (CDA)[36] was an unconstitutional “governmental regulation of the content of speech [which] is more likely to interfere with the free exchange of ideas than to encourage it.”[37]  The Court subsequently considered the constitutionality of the CDA’s successor, the Child’s Online Protection Act (COPA).[38]  While holding that COPA was not unconstitutional per se, the Court remanded the case to the Third Circuit Court of Appeals to consider other possible constitutional violations.[39]  Most recently, Congress passed the Children’s Internet Protection Act (CIPA).[40]  As CIPA was about to take effect, the American Library Association and the American Civil Liberties Union filed separate lawsuits challenging its constitutionality.[41]  Brief reviews of the two prior Acts show that Congress tried to correct deficiencies in CIPA so it would pass constitutional muster.

A. The Communications Decency Act — A First Attempt

    Congress first attempted to prevent the communication and display of obscene, indecent, or patently offensive material to children in the Communications Decency Act (CDA).[42]  Two provisions of the CDA directly applied to children and the Internet.[43]  First, the CDA criminalized knowing transmission of obscene or indecent messages to anyone under eighteen years of age.[44]  Second, the CDA criminalized knowingly sending or displaying material which offensively depicts or describes, “as measured by contemporary community standards, sexual or excretory activities or organs,” to recipients under eighteen years of age.[45]  Additionally, the CDA provided affirmative defenses limiting the liability for some persons, including those who publish offensive material, if the publisher “has taken, in good faith, reasonable, effective, and appropriate actions under the circumstances to restrict or prevent access by minors to a communication specified in [subsection (a)(1)(B) or (d)] . . . .”[46]  Appropriate actions could include “any method which is feasible under available technology” or restrictive access by “requiring use of a verified credit card, debit account, adult access code, or adult personal identification number.”[47]

    The American Civil Liberties Union (ACLU) filed suit challenging the constitutionality of the two provisions intended to protect minors.[48]  According to the ACLU, these two provisions were too vague and violated the First Amendment’s freedom of speech clause.[49]  After an evidentiary hearing, the district court agreed and granted a preliminary injunction against enforcement of both provisions.[50] 

    The government appealed the decision, arguing that the CDA was “plainly constitutional” under prior Supreme Court decisions.[51]  Relying on Ginsberg v. New York,[52] the government contended that no constitutional right to distribute indecent material to children exists.[53]  The Court rejected this argument, noting that the statute upheld in Ginsberg was narrower in four important respects than the contested CDA provisions.[54]  First, the Court stated that in Ginsberg, the State could not prosecute a parent for purchasing indecent magazines for his child, while under the CDA, the parent would be guilty for providing Internet access to similar material.[55]  Second, the Court noted that the New York statute upheld in Ginsberg applied only to commercial transactions, whereas the CDA applied to anyone.[56]  Third, the Court stated that, while the Ginsberg statute included the requirement that material harmful to minors be “‘utterly without redeeming social importance for minors,’” the language of the CDA failed to “provide us with any definition of the term ‘indecent’ as used in § 223(a)(1) and, importantly, omits any requirement that the ‘patently offensive’ material covered by § 223(d) lack serious literary, artistic, political, or scientific value.”[57]  Finally, the Court noted that the New York statute applied to anyone under the age of seventeen while the CDA applied to anyone under the age of eighteen, “an additional year of those nearest majority.”[58]

    The government also relied on Federal Communications Commission v. Pacifica Foundation,[59] to argue that it may regulate indecent speech, even though the restriction is content-based, to prevent harm to minors when distribution of that speech to adults poses a substantial risk of exposure to children.[60]  In Pacifica Foundation, the Court, concluding that “the ease with which children may obtain access to broadcasts, ‘coupled with the concerns recognized in Ginsberg,’ justified special treatment of indecent broadcasting,” upheld a declaratory order holding that a radio station could be sanctioned for broadcasting an indecent monologue, even though broadcasting previously received the most limited First Amendment protection.[61] 

    Here, however, the Court rejected this argument by noting several differences between the order upheld in Pacifica Foundation and the CDA.[62]  First, the FCC’s order in Pacifica Foundation imposed sanctions for a specific broadcast, but no “agency familiar with the unique characteristics of the Internet” evaluated or limited the CDA’s broad prohibitions.[63]  Second, unlike the regulation in Pacifica Foundation, a violator of the CDA would be subject to criminal prosecution.[64]  Finally, the Court stated that the broadcast medium historically “‘received the most limited First Amendment protection’ in large part because warnings could not adequately protect the listener from unexpected program content,” but the Internet had no such comparable history.[65]  Additionally, the Internet is not as “invasive” as other broadcast media, most “sexually explicit images are preceded by warnings as to the content,” and the user must take several steps to access sexually explicit images rather than come across them accidentally.[66]

    The government also relied on Renton v. Playtime Theatres,[67] where the Court upheld a zoning ordinance preventing adult movie theaters from opening in residential neighborhoods.[68]  The government contended that the CDA acts as “cyberzoning” on the Internet, “channeling indecent material to other ‘neighborhoods’ in cyberspace where only adults may communicate.”[69]  The Court disagreed, stating that the CDA was “a content-based blanket restriction on speech” because it applied to the entire Internet, and, as such, it cannot be “properly analyzed as a form of time, place, and manner regulation.”[70]  Thus, the Court stated that these prior decisions did not require it to uphold the CDA and were “fully consistent with the application of the most stringent review of its provisions.”[71]

    The Court held that the terms “indecent” and “patently offensive” in the CPA were so vague as to violate the First Amendment because of their “chilling effect on free speech” and because of the criminal penalties imposed for each violation.[72]  The Court noted that the standard it announced in Miller v. California[73] was more narrow and specific than the CDA.[74]  The Court was concerned that the requirement that the proscribed material be “specifically defined by applicable state law” was missing from the CDA, as were two of the Miller prongs.[75]  The stricter requirement was needed to reduce the inherent vagueness of the term “patently offensive.”[76]  According to the Court, the two missing prongs, (1) that the material taken as a whole appeals to the “prurient” interests and (2) that it “lac[ks] serious literary, artistic, political, or scientific value,” acted to limit the obscenity definition.[77]  Particularly important, the Court stated that the missing third prong was not judged by contemporary community standards and thus allowed appellate courts to impose some limitations on the definition of obscenity “by setting, as a matter of law, a national floor for socially redeeming value.”[78]  Because the CDA used a “community standards” criterion, the standards of the community most likely to be offended would serve as the criterion in judging communication available on the Internet, thus potentially suppressing speech adults have a constitutional right to receive.[79]  The Court noted that technology did not exist to prevent minors from receiving explicit material without also denying access to adults.[80]  The Court also noted that “it would be prohibitively expensive for noncommercial—as well as some commercial—speakers who have Web sites to verify” the user’s age, thus chilling speech on the Internet.[81]  Additionally, less restrictive alternatives were available, such as requiring indecent material to be “tagged” so it could be filtered, making exceptions for speech with “artistic or educational value,” and regulating commercial Web sites differently from non-commercial ones.[82]  Because the CDA placed an “unacceptably heavy burden on protected speech” and was not narrowly tailored, the Court declared it unconstitutional.[83]

B. The Child Online Protection Act — An Attempt to Correct the CDA’s Deficiencies

    In response to the Supreme Court’s ruling in Reno II, Congress attempted to correct the CDA’s deficiencies by enacting the Child Online Protection Act (COPA).[84]  Passed as part of the Omnibus Appropriations Act in October 1998,[85] Congress included changes from the CDA to conform to Ginsberg.[86]  First, the provisions of COPA did not prohibit parents from purchasing prohibited materials for their children.[87]  Second, COPA applied only to those individuals or entities involved in “communication for commercial purposes.”[88]  Further, Congress restricted the scope of commercial purposes to those on the World Wide Web rather than to the entire Internet.[89]  Third, Congress changed the term “indecent” to “harmful to minors,” adopting language similar to that of the three-prong Miller test.[90]  Finally, Congress adopted the definition of a minor to mean “any person under [seventeen] years of age.”[91]  Additionally, Congress included an affirmative defense to prosecution if the commercial operator restricted minors’ access to harmful material “(A) by requiring use of a credit card, debit account, adult access code, or adult personal identification number; (B) by accepting a digital certificate that verifies age; or (C) by any other reasonable measures that are feasible under available technology.”[92]

    The ACLU again filed suit, contending that COPA was unconstitutional by “burdening speech that is constitutionally protected for adults,” by violating the First Amendment rights of minors, and by being unconstitutionally vague.[93]  The District Court for the Eastern District of Pennsylvania found that COPA constituted a content-based restriction on speech that required review under a strict scrutiny standard.[94]  The court held that COPA placed too great a burden on protected expression because (1) Web publishers would incur high economic costs implementing age verification systems, (2) Web publishers would censor more material than necessary to avoid a violation, and (3) adults would be deterred from accessing Web sites requiring age verification, thereby affecting Web publishers’ abilities to provide the communications in the future.[95]  Additionally, the court determined that COPA was not narrowly tailored to meet the compelling interest of protecting minors from harmful material.[96]  While COPA required Web publishers of harmful material to make it less readily available to children rather than banning it, the court noted (1) that minors would still have access to similar material on international Web sites, (2) that some minors legitimately possess credit cards, defeating COPA’s affirmative defense, (3) that COPA prohibits an entire category of content rather than limiting its coverage to pictures, images, and graphic file images, and (4) that parental blocking or filtering software may be as effective with fewer burdens on speech.[97]  The court held that COPA would have a chilling effect on speech, resulting in the censoring of constitutionally protected speech.[98]  Because of this irreparable harm, the district court issued a preliminary injunction, enjoining the government from enforcing or prosecuting cases under COPA.[99]

    The government appealed the decision, and the Third Circuit Court of Appeals affirmed.[100]  The court agreed that COPA was a content-based restriction on speech subject to strict scrutiny.[101]  The court of appeals noted the district court’s finding that COPA was not narrowly drawn and, therefore, could not survive strict scrutiny.[102]  However, the Third Circuit based its determination of unconstitutionality on the overbreadth of the definition of “harmful to minors” in “COPA’s reliance on ‘contemporary community standards’ in the context of the electronic medium of the Web to identify material that is harmful to minors.”[103]  Because the Web is not geographically contained, any information published on a Web site is immediately accessible to all users.[104]  Additionally, since no technology existed to limit access to a Web site based on the geographic location of the user, Web site publishers would have to censor their publications or implement age verification systems to “‘comply with the regulation imposed by the State with the most stringent standard’ . . . .”[105]  Because Web publishers would have to meet the more stringent standards of “less liberal” communities, thereby depriving adults of their constitutional right to view certain material, the court held that COPA imposed an overreaching burden on constitutionally protected speech.[106] 

    The government appealed the decision, and the Supreme Court granted certiorari on “the narrow question [of] whether [COPA’s] use of ‘community standards’ to identify ‘material that is harmful to minors’ violates the First Amendment.”[107]  The Court first reiterated that courts should apply the three-prong Miller standard to test whether material is obscene.[108]  The Court then pointed out that the Miller test uses community standards “to be certain that . . . [material] will be judged by its impact on an average person, rather than a particularly susceptible or sensitive person—or indeed a totally insensitive one.”[109]  However, because “‘Web publishers are currently without the ability to control the geographic scope of the recipients of their communications,’” the Court had to determine whether such “technological limitation render[ed] COPA’s reliance on community standards constitutionally infirm.”[110]

    The Court stated that courts or jurors would use community standards that “need not be defined by reference to a precise geographic area” but “will inevitably draw upon personal ‘knowledge of the community or vicinage from which [they] come[].’”[111]  Therefore, even if courts instruct juries to use “the standards of the adult population as a whole,” juries in various parts of the country could still “reach inconsistent conclusions as to whether a particular work is ‘harmful to minors’” because of varying standards throughout the country.[112]

    In holding the CDA unconstitutional, the Court expressed concern over using a standard that would evaluate “any communication available to a nationwide audience . . . by the standards of the community most likely to be offended by the message.”[113]  The CDA was overbroad, restricting not only those communications that were patently offensive, but also those “with serious literary, artistic, political, or scientific value,” without limiting the restrictions to those communications that “appeal to the prurient interest.”[114]  Consequently, the CDA restricted much material that would be protected in some communities, but not in others.[115]

    In contrast, as the Court noted, COPA applied to much less material and defined materials harmful to minors as those which are patently offensive, designed to appeal to the prurient interests, and “‘taken as a whole, lac[k] serious literary, artistic, political, or scientific value for minors,’” thereby incorporating language similar to the three-pronged Miller test.[116]  Therefore, COPA more narrowly defined the material it covered, including material that “is in some sense erotic,” without restricting access to material that “‘a reasonable person would find . . . value in . . . taken as a whole.’”[117] 

    Further, the Court pointed out that “[w]hen the scope of an obscenity statute’s coverage is sufficiently narrowed by a ‘serious value’ prong and a ‘prurient interest’ prong, we have held that requiring a speaker disseminating material to a national audience to observe varying community standards does not violate the First Amendment.”[118]  In prior cases, the Court upheld a federal statute prohibiting the mailing of obscene materials[119] and a federal statute prohibiting the use of the telephone to “make ‘obscene or indecent’ . . . communications ‘for commercial purposes.’”[120]  Additionally, the Court noted that while in those cases the speaker could control where they communicated their material in the community, that ability was not significant in its legal analysis.[121] 

    Further, the Court did not find a need to adopt a different approach for the unique characteristics of the Internet.[122]  If a publisher does not want his material judged by the standards of a particular community, he needs to use a medium that allows him to publish material into a target community, such as using a credit card verification service.[123]  The Court also noted that there was “no reason to believe that the practical effect of varying community standards under COPA . . . is significantly greater than the practical effect of varying community standards under federal obscenity statutes.”[124]

    Finally, the Court stated that the plaintiffs failed to show that COPA was unconstitutionally overbroad since they did not demonstrate how the statute’s reach was both real and substantial.[125]  The Court held that “any variance caused by the statute’s reliance on community standards [was] not substantial enough to violate the First Amendment.”[126]  However, the Court noted that its decision was narrowly tailored, applying only to the identification of material harmful to minors using community standards.[127]  Consequently, the Court remanded the case to the Third Circuit Court of Appeals to determine “whether COPA suffers from substantial overbreadth for other reasons, whether the statute is unconstitutionally vague, or whether the District Court correctly concluded that the statute likely will not survive strict scrutiny analysis . . . .”[128]

    Interestingly, none of the nine Supreme Court justices upheld COPA’s constitutionality outright and Justice Stevens  indicated in his dissent that he would have affirmed the circuit court in declaring the Act unconstitutional.[129]  In her concurring opinion, Justice O’Connor agreed that the plaintiffs did not show that COPA was overbroad on the basis of differing community standards.[130]  However, Justice O’Connor also noted  the possibility “that the use of local community standards will cause problems for regulation of obscenity on the Internet, for adults as well as children, in future cases.”[131]  Additionally, fearing that Web producers would suppress their speech rather than bearing the burden of controlling who receives it, Justice O’Connor proposed Congress should adopt a national standard for regulating Internet obscenity.[132]  She pointed out that precedent did not prevent the Court from adopting a national standard.[133]  Rather, Justice O’Connor argued that precedent “left open the possibility that jurors would apply any number of standards, including a national standard, in evaluating material’s obscenity.”[134]  While obtaining a national standard may be difficult, Justice O’Connor pointed out that juries in large states like California already consider the standards of very diverse communities—the approach the Court approved in Miller itself.[135]  Therefore, it is not unreasonable to assume that the nation as a whole could make similar assessments, especially since jurors are “more aware of the views of adults in other parts of the United States” because of the Internet.[136]

    Justice Breyer, in his concurring opinion, stated that the statute did not specifically define the word “community.”[137]  Considering COPA’s legislative history, Justice Breyer stated that “Congress made clear that it did not intend this ambiguous statutory phrase to refer to separate standards that might differ significantly among different communities.”[138]  Therefore, courts and juries should use a community standard that is both “national and adult.”[139]  Additionally, using a national, adult community standard avoids other First Amendment problems.[140]  According to Justice Breyer, “[t]o read the statute as adopting the community standards of every locality in the United States would provide the most puritan of communities with a heckler’s Internet veto affecting the rest of the Nation.”[141]

    In his concurring opinion, Justice Kennedy stated it is likely that COPA is overbroad and, therefore, unconstitutional.[142]  However, because Congress was aware of the Court’s decision rendering the CDA unconstitutional when it promulgated COPA, Justice Kennedy pointed out that courts must carefully analyze whether the statute is truly overbroad before invalidating it.[143] 

    In this case, the court of appeals based its decision strictly on the use of the contemporary community standard while ignoring the rest of the statute.[144]  Justice Kennedy stated that whether the variations in the community standard render COPA unconstitutional “depends on the breadth of COPA’s coverage and on what community standards are being invoked.”[145]  The court of appeals needed to look at all speech that COPA would burden, including “the amount of commercial communication, the number of commercial speakers, or the character of commercial speech covered by the Act.”[146]  Justice Kennedy noted that venue was an issue that neither the parties nor the circuit court addressed, but would “be bound up with the issue of varying community standards” since the court must use the standard of that venue.[147]  COPA does not address whether venue is proper in a jurisdiction where the speaker published the material, where it passed through, or where the viewer ultimately saw it.[148]  “The more venues the Government has to choose from, the more speech will be chilled by variation across communities.”[149]

    Justice Kennedy also noted that COPA offers no guidance on judging material “as a whole” in regard to the World Wide Web.[150]  The statute does not indicate if “as a whole” means “a single image on a Web page, a whole Web page, an entire multipage Web site, or an interlocking set of Web sites.”[151]  Because the circuit court only considered varying community standards and not the speech or speakers covered under COPA, Justice Kennedy concurred that the case should be remanded for further consideration of the statue’s other aspects.[152]  However, he noted “if the District Court correctly construed the statute across its other dimensions, then the variation in community standards might well justify enjoining enforcement of the Act.”[153]

    In his dissent, Justice Stevens stated that although “the community standard [once] provided a shield for communications . . . offensive only to the least tolerant members of society . . . [i]n the context of the Internet, . . . community standards become a sword . . . .”[154]  If in the most puritan town the communication is offensive, the publisher would violate the statute and be subject to civil and criminal sanctions.[155]  Therefore, while acknowledging that COPA was an improvement over the CDA, Justice Stevens agreed with the circuit court that COPA’s community standard rendered the statute unconstitutional.[156] 

    Justice Stevens noted that, unlike previous cases where statutes prohibited selling or mailing indecent materials to minors, COPA limits protected speech not specifically targeted at children but posted on the Internet available for adults and children to view.[157]  Protecting children from harmful material is a compelling governmental interest, but the Court has “repeatedly rejected the position that the free speech rights of adults can be limited to what is acceptable for children.”[158]

    Justice Stevens noted that the Court overlooked COPA’s comparison to the CDA with its similar community standard provision, which the Court invalidated because of its overbreadth.[159]  While COPA has other provisions requiring material to “‘appeal[] to the prurient interest of minors,’ [to be] . . . ‘patently offensive with respect to minors,’ [and to] lack ‘serious literary, artistic, political, or scientific value for minors,’” these provisions do not substantially narrow the amount of speech that will be limited.[160]  The jury will judge whether the material is appropriate for minors using community standards, but the material that is limited may be appropriate for adults.[161]  Thus, “within the subset of [material] deemed to have no serious value for minors, the decision whether minors and adults throughout the country will have access to that speech will still be made by the most restrictive community,” even though many other communities would not consider the material harmful to minors.[162]  Consequently, Justice Stevens would hold that COPA is substantially overbroad and therefore unconstitutional.[163]

C. The Children’s Internet Protection Act — A Further Attempt

    In January 1999, Senator John McCain introduced the first of several versions of the Children’s Internet Protection Act in order to protect children from accessing inappropriate Internet material.[164]  The Act would require that an

elementary or secondary school having computers with Internet access may not receive services at discount rates . . . unless . . . [it] (i) has selected a technology for its computers with Internet access in order to filter or block Internet access through such computers to (I) material that is obscene; and (II) child pornography; and (ii) is enforcing a policy to ensure the operation of the technology during any use of such computers by minors.[165]

However, fearing insufficient support, Congress took no action on this form of the bill.[166]

    Representative Bob Franks introduced several versions of a similar bill, which included language restricting “matter deemed to be inappropriate for minors.”[167]  His version, passed by the House in June, 1999, but never enacted, would have required schools receiving E-rate subsidies to use filtering software and to certify that they had selected, installed, and were using “technology . . . to filter or block (i) child pornographic material . . . (ii) obscene materials . . . and (iii) during use by minors, materials deemed to be harmful to minors.”[168]  Additionally, Representative Ernest Istook introduced a version that would have required schools and libraries using federal funds to purchase computers to use filtering software.[169]

    In the fall of 2000, Congress created an omnibus budget bill to resolve all outstanding budgetary issues, including CIPA.[170]  President Clinton signed this budget into law on December 21, 2000, with CIPA taking effect on April 20, 2001.[171]  As passed, CIPA consisted of two different acts, the Children’s Internet Protection Act and the Neighborhood Children’s Internet Protection Act.[172]  Consequently, while having the same goal of protecting minors from inappropriate and obscene materials, each separate act regulates Internet activities based on the type of federal funds the institution receives.[173]

    To comply with CIPA, schools and libraries receiving E-rate funding for Internet access, service, and internal connections face the greatest number of requirements.[174]  First, the institution must adopt an Internet safety policy that addresses minors’ access to “inappropriate matter,”[175] minors’ safety and security when using forms of direct electronic communication such as e-mail and chat rooms, minors’ unauthorized access such as hacking and other unlawful online activities, unauthorized disclosure of minors’ personal identification information, and measures designed to restrict minors’ access to harmful materials.[176]  Second, the institution must provide notice of and hold at least one public hearing on, the proposed Internet safety policy where the public may discuss concerns and revise the policy before adoption.[177]  Third, the institution must certify to the Federal Communications Commission (FCC) that it has adopted and implemented an Internet safety policy that includes a “technology protection measure”[178] that blocks pictures that are obscene, that are child pornography, or that are harmful to minors (if the computer is in use by a minor) and that it enforces the operation of the technology protection measure during the computer’s use.[179]

    For schools that do not receive E-rate funding but do receive ESEA Title III funds, the institution need only comply with a certification requirement stating that it has adopted and is enforcing an Internet safety policy.[180]  Likewise, a library that does not receive E-rate funding, but receives funds under the Museum and Library Services Act,[181] must certify that it has, and is enforcing, an Internet safety policy.[182]

    Under CIPA, all of an institution’s computers with Internet access must have a technology protection measure installed, regardless of the age of the intended user.[183]  However, an institution may disable the filter when an adult is using the computer for “bona fide research or other lawful purpose.”[184]  Yet, CIPA contains no definition of these terms or how the institutions should apply them.[185]

    Failure to comply with CIPA, regardless of the type of funding received, results in ineligibility for the funding program.[186]  When a school or library is not in compliance, the FCC must notify the institution and suspend its funding until it has complied.[187]  However, if an institution receives E-rate funds and is non-compliant, it can be liable for reimbursing the FCC for funds already received.[188]

    CIPA allows for expedited court review of any civil action challenging the constitutionality of the Act.[189]  If there is a challenge, it “shall be heard by a district court of [three] judges convened pursuant to the provisions of section 2284 of title 28, United States Code.”[190]  If the district court finds the Act unconstitutional, the government can appeal directly to the United States Supreme Court within twenty days of the decision.[191]

II. Computer Filtering Software — Not a Guarantee of Safety

A. What is a Filter?

    A filter is defined as “a device or material for suppressing or minimizing waves or oscillations of certain frequencies (as of electricity, light, or sound).”[192]  A computer filter or blocking software is a program intended to

block access to Internet sites listed in an internal database of the product; block access to Internet sites listed in a database maintained external to the product itself; block access to Internet sites which carry certain ratings assigned to those sites by a third party, or which are unrated under such a system; scan the contents of Internet sites which a user seeks to view and block access based on the occurrence of certain words or phrases on those sites.[193]

There are presently around 150 filtering tools available, many of which are free or of minimal cost.[194]  These programs work in several different ways including “limit[ing] access . . . based on verbal identification of that material; . . . based on URLs—Internet addresses . . . ; and . . . utiliz[ing] algorithms that analyze pictoral [sic] blends of color and texture in order to identify skin tone or the presence of other objectionable materials.”[195]  The user can install the software in a variety of ways.[196]  For example, with a “client-based” product, the user installs and runs the software on each individual machine.[197]  With a “server-based” product, the user installs the software on the file server that intercepts the information from the Internet and blocks it before it reaches the workstation.[198]  Finally, an Internet Service Provider (ISP) can install the software on its server, letting the recipient choose whether to receive pre-filtered service.[199]

B. How Does a Filter Work?

    1. Filters That Block Access to Internet Sites Listed in an Internal Database

    One type of filter blocks sites listed in an internal database of the product.[200]  With this type of filter, the operator can preview a particular Web site and if she determines it appropriate, add it to the database of available sites pre-set in the program.[201]  This allows a teacher or librarian to determine the content that users would be able to view, a process similar to book acquisition for the classroom or library.[202]

    2. Filters That Block Access to Internet Sites Listed in an External Database Updated by the Product Manufacturer

    A second type of filter is one that blocks sites listed in an external database.[203]  With this type of filter, the software manufacturer and its team of reviewers classify Web sites and material in categories such as “obscenity,” “nudity,” or “violence.”[204]  A disadvantage with these filters is that the company, not the user, determines which web sites are blocked and often the user is unable to add to or to customize the database.[205]  Additionally, the filtering companies do not release the names of the Web sites blocked and only provide limited information on their criteria for blocking.[206]

    3. Filters That Block Access to Internet Sites Which Carry Ratings Assigned by a Third Party or Which Are Unrated

    A third type of filter blocks content based on a rating voluntarily attached to the Web site by its administrator.[207]  The rating becomes part of the Web site’s programmed code, and the user may select which rating he wants the software to block.[208]  If someone attempts to access a site within the range selected, the site is blocked.[209]  However, there are only a small number of rated sites on the Internet.[210]  Additionally, because many of the sites are self-rated, the publisher may not rate his site accurately or strictly enough to prevent minors’access to inappropriate sites.[211]

    Because there are so few rated sites, a person may have adverse results using this type of blocking software.[212]  For example, some rating-based filtering software denies access to any unrated site, preventing the user from being able to view appropriate sites only because they are unrated.[213]  On the other hand, other similar types of software allow the user to access all unrated sites, including those that are inappropriate for the user.[214]

    4. Filters That Block Access to Internet Sites Based on the Occurrence of Certain Words or Phrases on Those Sites

    A fourth type of filter blocks access based on the presence of certain words or phrases within a Web site.[215]  As with external database filters, it is the software company, not the user, that determines the words and phrases that trigger the block.[216]  Accordingly, the filter may prevent the user from viewing sites that, while appropriate, contain one of the keywords.[217]  Further, software manufacturers frequently do not release the list of keywords or their other blocking criteria.[218]  If a publisher does not identify a photograph on a Web site as having sexual or inappropriate content using one of the keywords, the filter will not block that photograph and the child would receive it.[219]

    5. Filters That Analyze Pictorial Blends of Color and Texture in Order to Identify Skin Tones or Objectionable Graphics

    A fifth type of filter uses algorithms to analyze pictorial blends of color and texture.[220]  This filter “looks” at a picture and blocks it when the program identifies it as having a large amount of skin tones, thereby assuming it is objectionable material, such as a nude photograph.[221]  A disadvantage to this type of filter is that it does not actually “see” the photograph; thus “it cannot discriminate between obscenity and protected speech, or child pornography and protected speech.”[222]

III. Will CIPA Pass Constitutional Muster?  American Library Association v. United States[223]

A. The Lawsuit Challenging CIPA’s Constitutionality

    Believing that “mandatory filtering is misguided and unworkable in the context of a public institution,” the American Library Association (ALA), the ACLU, various library associations, patrons, and Internet authors and publishers filed suit challenging CIPA’s constitutionality seeking declaratory and injunctive relief.[224]  On July 26, 2001, a three-judge panel of the federal district court denied the government’s motion to dismiss the case, concluding “that plaintiffs’ complaints in these actions contain enough factual allegations to withstand dismissal.”[225]  The court then consolidated the cases.[226]

    The plaintiffs challenged CIPA’s constitutionality on several grounds.[227]  First, they contended that spending programs, like the one under CIPA, are subject to First Amendment scrutiny.[228]  Second, the plaintiffs argued that the First Amendment applies to public libraries.[229]  Third, the suit alleged that CIPA violates the First Amendment by conditioning funding on the mandatory use of filtering software that blocks access to a substantial amount of constitutionally protected speech.[230]  Finally, the plaintiffs contended that CIPA’s disabling provision will have a “chilling effect” on speech and will allow the library officials unfettered, standardless discretion.[231]

    1. Spending Programs Must Survive First Amendment Scrutiny

    The plaintiffs first argued that this case involved First Amendment issues.[232]  The defendant, however, contended the case involved the Spending Clause, and relying on South Dakota v. Dole,[233] argued that the court must address whether Congress constitutionally exercised its spending power.[234]  As the plaintiffs noted, CIPA fails under the Dole test because the Constitution prohibits Congress from exercising its spending power in a way that violates another constitutional provision, such as the suppression of speech under the First Amendment.[235]

    The plaintiffs further contended that even a spending program must survive First Amendment scrutiny if it restricts speech.[236]  When Congress restricts the use of government funds, it is the government, not a private party, that must act as the speaker.[237]  If the government appropriates funds “to convey a governmental message, it may take legitimate and appropriate steps to ensure that its message is neither garbled nor distorted by the grantee.”[238]  However, when Congress “‘expends funds to encourage a diversity of views from private speakers,’”[239] its “antecedent funding decision cannot be aimed at the suppression of ideas thought inimical to the Government’s own interest.”[240]

    One test the courts use to determine if there are impermissible restrictions on private speech is to ascertain whether “the Government seeks to use an existing medium of expression and to control it . . . in ways which distort its usual functioning.”[241]  In that situation, the court will determine if the restriction on the medium’s accepted usage “is necessary for the program’s purposes and limitations” because “[t]he First Amendment for[bids] the Government from using the forum in an unconventional way to suppress speech inherent in the nature of the medium.”[242]  Thus, the plaintiffs argued that CIPA involved private speech, that the government sought to use an existing medium (the Internet), and that mandating the use of blocking software in that medium distorts its usual function.[243]

    As “a unique and wholly new medium of worldwide human communication,” the Internet facilitates communication among any people with a connection using different kinds of services, such as e-mail, chat, newsgroups, and the World Wide Web, with that communication occurring among thousands of people simultaneously.[244]  There are currently approximately 400 million people using the Internet throughout the world.[245]  Internet users can access  2,469,940,685 web pages.[246]  Consequently, the range of content available on the Internet “is as diverse as human thought.”[247]  Therefore, the plaintiffs contended that there was no way to describe the majority of the speech communicated on the Internet “as conveying a government-sponsored message.”[248]

    The defendants, relying on Rust v. Sullivan,[249] argued that CIPA involves E-rate and LSTA funding programs for the specific and limited purposes of providing Internet access in public libraries for the public’s educational use.[250]  That, however, distorts the mission of libraries, which is “to provide free access to books, ideas, resources, and information for education, employment, enjoyment and self-government.”[251]  Additionally, the library is a community center that serves people of all ages and economic backgrounds.[252]  Public schools have the “inculcative function” of preparing students for participation as adults in society.[253]  While public school libraries supplement that role, public libraries were “designed by freewheeling inquiry.”[254]

    The plaintiffs also contended that CIPA suppresses speech beyond that funded by the statute.[255]  According to the statute, a library receiving E-rate or LSTA funds must certify that blocking software is installed and operational on “any of its computers with Internet access” during “any use of such computers.”[256]  The result is that every computer with an Internet connection, including those funded with non-federal money, blocks protected speech.[257] 

    Further, the library would not be able to install an additional facility offering uncensored Internet access as the defendant contended.[258]  First, the library is required to install blocking software on any computer with Internet access regardless of location.[259]  Second, the government did not include any provision in CIPA that allowed the housing of unfiltered computers in a separate location.[260]  Third, CIPA does not define “separate” nor offer guidance on how the library could accomplish providing a separate location.[261]

    2. The First Amendment Applies to Public Libraries

    The First Amendment states, “Congress shall make no law

. . . abridging the freedom of speech . . . .”[262]  This provision also encompasses the right to listen.[263]  In Reno v. ACLU, the Court held that the right to receive information also applies to expression on the Internet and is entitled to “‘the highest protection from governmental intrusion.’”[264]  Likewise, in Sund v. City of Wichita Falls,[265] the District Court for the Northern District of Texas expressly stated that “[t]he right to receive information is vigorously enforced in the context of a public library as ‘the quintessential locus of the receipt of information.’”[266]

    The state creates a forum when it opens public property “for use by the public as a place for expressive activity.” [267]  Because the library is a limited public forum, it may not impose content-based restrictions on rights to receive information through the Internet.[268]    While the government is not required to create such a designated public forum, once it does, it “is bound by the same standards as apply in a traditional public forum.”[269]  The state may only impose content-based restrictions on speech in that forum if the restrictions are “narrowly drawn to effectuate a compelling state interest.”[270]  

    Libraries are a limited public forum open “to the public for expressive activity, namely the communication of the written word.”[271]  Thus, the nature of the Internet is compatible with this purpose.[272]  Consequently, if the government restricts the speech of someone who is a member of the limited public forum, the court will review its action under strict scrutiny.[273] 

    The plaintiffs also contended that CIPA is similar to librarians removing books from the shelves after the library purchased them, because the library must actively restrict Internet access to a publication after making it available.[274]  This is more like a censorship decision than an acquisition decision.[275]  In Board of Education v. Pico,[276] a group of parents claimed that several books in the school library were offensive.[277]  As a result, the Board of Education removed nine books.[278]  The Supreme Court held that government officials could not remove the materials “simply because they dislike the ideas contained in those books and seek by their removal to [promote their personal views].”[279]  However, this principle is not limited to school libraries.[280]  Consequently, CIPA, through its imposition of mandatory blocking software, restricts access to substantially more material than the Board of Education attempted to suppress in Pico.[281]

    3. Filtering Software Blocks Access to a Substantial Amount of Constitutionally Protected Speech

    The plaintiffs next argued that the mandatory use of filtering software under CIPA violates the First Amendment because it blocks a substantial amount of constitutionally protected Internet speech based on the content of that expression.[282]  Additionally, because the software is required to be in use at all times on all computers with Internet access, CIPA violates adult patrons’ First Amendment right to receive constitutionally protected speech.[283]

    Presently, filtering software is unavailable for reliably blocking the type of speech prohibited under CIPA without also blocking other constitutionally protected speech.[284]  Most filtering technology either uses a database of Web sites and blocks those in the list, or searches for certain keywords and blocks those sites containing one or more of the words.[285]  Unfortunately, the software also blocks Web sites with appropriate, non-obscene material if they contain one of those words.[286]  Additionally, the filter may not detect sites containing obscene pictures that have no text associated with them, allowing the user to view the photograph and, thus, circumvent CIPA.[287]

    Additionally, the software manufacturer decides the criteria used to trigger a block on a particular Web site and often does not disclose that criteria to the user.[288]  Thus, the library using the software, or even judicial authorities, cannot review that decision.[289]  This forces the user to accept the manufacturers’ definition of obscenity, even if it is not the same as that defined in CIPA.[290]  The First Amendment protects speech that is sexually explicit, but not obscene.[291]

    Additionally, the definition of obscenity relies on the community standards test.[292]  Because the Internet is not limited to a single geographic location, anything published on a Web page would have to “abide by the most restrictive community’s standards.”[293]

    4. CIPA’s Disabling Provision Chills Speech and Allows Library Officials Unfettered, Standardless Discretion

CIPA contains a disabling provision, allowing the library official to “disable a technology protection measure . . . to enable access for bona fide research or other lawful purposes.”[294]  However, CIPA does not provide any guidelines to assist the library official.[295]  In fact, the plain language of the provision states that the official “may disable the technology protection measure,” but does not require it.[296]  In City of Lakewood v. Plain Dealer Publishing Co.,[297] the Court disapproved of vesting similar discretion in a government official.[298]  There, a newspaper publisher challenged a city ordinance that gave the mayor the authority to grant or deny applications for coin operated newspaper vending machines.[299]  According to the ordinance, the mayor could grant the permit for, among other reasons, “such other terms and conditions deemed necessary and reasonable by the Mayor.”[300]  The Court held that the “ordinance giving the mayor unfettered discretion to deny a permit application and unbounded authority to condition the permit on any additional terms he deemed ‘necessary and reasonable,’ was unconstitutional.”[301]

    Similarly, CIPA, by not providing any guidelines or definitions of “bona fide research or other lawful purpose,” allows the library official unfettered discretion to determine whether or not to disable the filter.[302]  The Federal Communications Commission (FCC), the agency overseeing the certification of schools and libraries under CIPA, expressly stated that it would not “promulgate rules mandating how entities should implement these provisions.  Federally-imposed rules directing school and library staff when to disable technology protection measures would likely be overbroad and imprecise, potentially chilling speech, or otherwise confusing schools and libraries about the requirements of the statute.”[303]  According to the FCC, this determination is best left to local community officials who are “most knowledgeable about the varying circumstances of schools or libraries within those communities.”[304]  However, the FCC did not compel local communities to adopt any methods or criteria for determining when to allow a library official to disable the filter.[305]  Consequently, because there are no standards nor a requirement for local communities to adopt any, librarians within the same community or even the same library, may or may not disable the filter for the same exact research purpose based on their own perception of what constitutes “bona fide research or other lawful purposes.”[306]

    Therefore, the library officials act as a speech gatekeeper, unconstrained by any defined standards reviewable by a court.[307]  Previously, courts have held such prior restraints unconstitutional.[308]  In Mainstream Loudoun v. Board of Trustees of the Loudoun County Library,[309] a group of library patrons claimed that the library’s policy of using mandatory blocking software on all its computers blocked access to the Internet.[310]  The library used a filtering program that blocked sites based on criteria chosen by the manufacturer and kept unknown to purchasers.[311]  If the computer blocked a patron from a site, she could file with the librarian a written request, including the reason for unblocking the site.[312]  If the librarian determined that the site was acceptable under the library’s policy, he would manually override the filtering software.[313]  The policy contained no time limit or procedure for notifying the patron of the librarian’s decision.[314]  The District Court for the Eastern District of Virginia held the policy to be an unconstitutional prior restraint because it allowed the library board discretion to censor “essentially unbounded.”[315]  

    Without any objective and specific standards, CIPA’s disabling provision acts as a prior restraint because there is no way for the library official to consistently apply what constitutes “bona fide research or other lawful purposes,” nor is the ordinary citizen aware of what is permitted or forbidden.[316]  The Supreme Court has repeatedly held such prior restraints unconstitutional stating,

[a] free society prefers to punish the few who abuse rights of speech after they break the law than to throttle them and all others beforehand.  It is always difficult to know in advance what an individual will say, and the line between legitimate and illegitimate speech is often so finely drawn that the risks of freewheeling censorship are formidable.[317]

    Another problem with the provision is its chilling effect on speech because patrons must request that the filter be disabled.[318]  In Denver Area Education Telecommunications Consortium, Inc. v. FCC,[319] citizens challenged the constitutionality of three provisions of an act governing indecent and obscene programming.[320]  In order to receive patently offensive sex-related material, the cable subscriber had to submit a written request asking the cable operator to unblock or descramble the station.[321]  The Court held this provision unconstitutional because it “restrict[ed] viewing by subscribers who fear for their reputations should the operator, advertently or inadvertently, disclose the list of those who wish to watch the ‘patently offensive’ channel.”[322]  A similar result is possible and probable if a library patron is required to request that the filtering software be disabled.[323]  As in Denver Area Education Telecommunications Consortium, the material desired, though constitutionally protected, may be offensive to some.[324] 

    The library may implement more narrowly tailored options than mandated filtering software.[325]  For example, CIPA provides that the library must create and implement an acceptable use policy.[326]  Additionally, those libraries receiving E-rate funds must hold at least one public hearing where the citizens and patrons can offer input into and concern over the policy.[327]  Presently, many libraries employ a shoulder tap method, allowing the library official to tap the Internet user on the shoulder if the user is viewing inappropriate material.[328]  Other less restrictive means would include using privacy screens around the Internet computers, using filters that an adult can turn off, having computers for children’s use located in the children’s section, having the computer monitors recessed into the desktop, educating patrons on appropriate Internet use, utilizing time limits on Internet use, and enforcing criminal laws when violations occur.[329] 

B.  The District Court’s Opinion

    The District Court for the Eastern District of Pennsylvania conducted the eight-day trial, during which it heard various witnesses and received much evidence, primarily on the efficiency of the currently available filtering software.[330]  The court held that CIPA facially violated the First Amendment and permanently enjoined the government from withholding federal funds from those libraries who failed to comply with CIPA.[331]

    The court first considered the plaintiffs’ argument that  CIPA fails under the Dole test because it “induces” public libraries to violate the First Amendment rights of Internet content-providers and public library patrons.[332]  The government contended that the plaintiffs must demonstrate that “under no circumstances is it possible for a public library to comply with CIPA’s conditions without violating the First Amendment.”[333]  The court noted that in Dole, the Supreme Court did not explain “what it means for Congress to use the spending power to  ‘induce [recipients] to engage in activities that would themselves be unconstitutional’” and that it was “unclear what exactly a litigant must establish to facially invalidate an exercise of Congress’s spending power on this ground.”[334]   While a court will only sustain a facial challenge when the plaintiff shows that the statute has no constitutional application, the First Amendment’s overbreadth doctrine allows a court to sustain a facial challenge “of a statute that burdens a substantial amount of protected speech, even if the statute may be constitutionally applied in particular circumstances.”[335]

     The court did not decide whether the plaintiffs needed to show that it was impossible for public libraries to comply with CIPA without violating the First Amendment or whether CIPA will prevent library patrons from accessing a substantial amount of protected speech in violation of the First Amendment.[336]  Instead, the court stated that the plaintiffs needed to show that any public library complying with CIPA would violate the First Amendment.[337]

    Because presently available filtering technology would block a substantial amount of constitutionally protected material, the court looked to the whether the First Amendment would allow libraries to use filtering software that CIPA mandates.[338]  In order to determine if CIPA would violate the constitutional rights of the patrons, the court had to determine the appropriate level of scrutiny to use in reviewing the public library’s content-based restriction on Internet access, which, in turn, depended on whether a public library is a traditional public forum, a designated public forum, or a nonpublic forum.[339]  The plaintiffs contended that a public library is a designated public forum and that its restrictions on speech are subject to strict scrutiny, where the government must show that the restriction serves a compelling state interest, and that it is narrowly tailored to achieve that interest.[340]  Conversely, the government contended that a public library is a nonpublic forum and its restrictions on speech only need to be reasonable, as long as the library is not trying to suppress the speech because it disagrees with the speaker’s view.[341]

    The court had to determine whether the forum was the library’s entire collection, including print and electronic material, or merely the library’s provision of Internet access.[342]  The Supreme Court previously held that the relevant forum is determined by what material the speaker is seeking.[343]  The plaintiffs asserted their right to access the Internet or to provide information, through the Internet, to library patrons.[344]  Thus, the court concluded that “the relevant forum . . . [was the] specific [one] created when the library provides its patrons with Internet access.”[345]

    The court then analyzed content-based restrictions in a designated public forum, first by noting:

[T]he purpose of a public library in general, and the provision of Internet access within a public library in particular, is “for use by the public . . . for expressive activity,” . . . namely, the dissemination and receipt by the public of a wide range of information.  We are satisfied that when the government provides Internet access in a public library, it has created a designated public forum.[346] 

The court differentiated between several different instances where the government creates a designated public forum.[347]  First, when the government decides to fund a particular message, content-based restrictions on the speech the government chooses to fund is subject to a rational basis review.[348]  However, “where the government creates a designated public forum to facilitate private speech representing a diverse range of viewpoints, the government’s decision selectively to single out particular viewpoints for exclusion is subject to strict scrutiny.”[349]  Additionally, if the government creates a designated public forum on a specific topic and allows the public to speak on that topic, yet it selectively excludes some speakers based on the content of their speech, the court will review that restriction using a strict scrutiny standard.[350]  Finally, when the government opens the forum for “virtually unrestricted use by the general public for speech on a virtually unrestricted range of topics, while selectively excluding particular speech whose content it disfavors,” the court will review that restriction using a strict scrutiny standard.[351] 

    The court held that CIPA’s content-based restrictions were subject to strict scrutiny because it selectively excluded some speakers, while allowing others, in what is analogous to a traditional public forum.[352]  In finding that the restrictions selectively exclude some speakers, the court distinguished a library restricting patrons’ access to some Internet sites with filters from a library deciding how to spend funds to acquire books for its collection.[353]  Acquisition decisions, which a court would usually subject only to a rational basis review, are made after librarians, applying professional standards to develop their collection, have reviewed the content of the speech and decided that it meets the criteria of the library’s collection development policy.[354]  However, “in providing patrons with even filtered Internet access, a public library invites patrons to access speech whose content has never been reviewed and recommended as particularly valuable by either a librarian or a third party to whom the library has delegated collection development decisions.”[355]  Thus, in providing Internet access, libraries “intentionally open their doors to vast amounts of speech that clearly lacks sufficient quality to ever be considered for the library’s print collection.”[356]

    The court stated that even though the First Amendment allows public libraries to exercise editorial discretion to selectively acquire books based on content, it “believe[d] that where the state provides access to a ‘vast democratic forum[],’ . . . open to any member of the public to speak on subjects ‘as diverse as human thought,’ . . . and then selectively excludes from the forum certain speech on the basis of its content, such exclusions are subject to strict scrutiny.”[357]  The public library’s content-based restrictions on a patrons’ Internet access were similar to other content-based restrictions on speech that the Supreme Court had subjected to strict scrutiny.[358]

    The court also found that the “public libraries’ provision of Internet access promote[d] First Amendment values in an analogous manner to traditional public fora, such as sidewalks and parks, in which content-based restrictions on speech are always subject to strict scrutiny.”[359]  The court noted that speech in streets, sidewalks, and parks is regulated because of the “speech-facilitating character” of those areas, such as their openness.[360]  Similarly, the public may enter a public library “to receive the speech that these fora facilitate,” and, like other public areas, the public pays for the library through taxes and may use it at no additional cost.[361]  Additionally, sidewalks and parks allow a speaker to communicate with a large number of people at low cost.[362]  The Internet also allows patrons to communicate with the public at a relatively low cost or, in the case of public libraries and schools, for free.[363]  Yet unlike traditional fora, the Internet allows a person to communicate with literally the entire world of people with an Internet connection without the “constraints of geography.”[364]

    Next, the court analyzed CIPA’s restriction on speech to determine whether it promoted a governmental interest, whether it was narrowly tailored to meet that interest, and whether there were less restrictive alternatives that would serve the same purpose.[365]  The government asserted three state interests promoted by a public library’s use of filtering software.[366]  First, CIPA’s use of filters is intended to prevent patrons from accessing obscene material, child pornography, or material harmful to minors on public library computers.[367]  Second, CIPA’s use of filtering software prevents patrons other than those using the computer from being unwillingly exposed to sexually explicit material.[368]  Third, CIPA’s use of filtering software reduces the occurrences of patrons engaging in unlawful or inappropriate conduct.[369]

    The court stated that “[t]he government’s interest in preventing the dissemination of [kind of material that CIPA proscribes] is well established.”[370]  Because the First Amendment does not protect speech that meets the Miller definition of obscene, the government has a compelling interest in preventing its distribution as well as preventing “the possession and distribution of child pornography.”[371]  Therefore, if the library’s use of filters is narrowly tailored to further this interest, CIPA’s restriction would survive strict scrutiny.[372]

    Next, the court examined whether the libraries’ use of filters to protect the unwilling viewer was a legitimate state interest.[373]  The court noted that the Supreme Court had previously been reluctant to recognize the protection of the unwilling viewer as a legitimate governmental interest.[374]  Rather, the Court has held that “[w]here the designed benefit of a content-based speech restriction is to shield the sensibilities of listeners, the general rule is that the right of expression prevails . . . . We are expected to protect our own sensibilities simply by averting our eyes.”[375]  Otherwise, those who were offended by the speech would be able to suppress it, “effectively empower[ing] a majority to silence dissidents simply as a matter of personal predilections.”[376]  Additionally, the court observed that “[t]he state’s interest in protecting unwilling viewers from exposure to patently offensive material” is part of the Miller test.[377]  However, the Supreme Court, in limited circumstances, has held that a state does have a legitimate interest in protecting the public from unwilling exposure to speech that is not obscene, such as when it intrudes into the home or when there is a captive audience.[378]  Because case law did not “categorically foreclose[]” the possibility of the state’s interest in protecting unwilling viewers, the court stated that the government “might have a compelling interest in protecting library patrons and staff from unwilling exposure to sexually explicit speech that, although not obscene, is patently offensive.”[379]

    The court examined the government’s third asserted legitimate interest of preventing unlawful or inappropriate conduct such as harassment, physical assault, and public masturbation.[380]  However, the court noted that the government did not offer any evidence to show how this conduct had changed after a library installed filters on the computers or how the use of unfiltered computers incited this conduct.[381]  Also, the library could deter this behavior in other ways, such as prohibiting the patron from further use of the computers.[382]

    Although the government does have a legitimate interest in preventing dissemination of material CIPA proscribes, the court held that any public library’s use of filtering software failed to be narrowly tailored to meet that end.[383]  Therefore, the court declared that CIPA was facially unconstitutional.[384]  In reaching this conclusion, the court noted that the currently available filters block thousands of Web pages that do not contain material harmful to minors, obscene material, or child pornography, and that there were substantially more Web sites that would be blocked than the expert witness discovered in testing only four filtering programs.[385]  Further, the court stated that no existing software could distinguish between sites containing proscribed visual depictions and those that did not.[386]

    Additionally, filters had other flaws.[387]  In order to identify Web sites containing proscribed material out of the ever-increasing number of Web sites, filtering companies use techniques that result in overblocking.[388]  Further, if the filtering companies did not use these techniques, the software would be ineffective or would underblock access to sites with proscribed material.[389]  As the court stated, “any filter that blocks enough speech to protect against access to visual depictions that are obscene, child pornography, and harmful to minors, will necessarily overblock substantial amounts of speech that does not fall within these categories.”[390] 

    The government has the burden to show that software exists to meet CIPA’s requirements without overblocking a substantial amount of constitutionally protected speech, but the government failed to meet that burden.[391]  Consequently, “any public library’s use of a software filter required by CIPA will fail to be narrowly tailored to the government’s compelling interest in preventing the dissemination . . . of visual depictions that are obscene, child pornography, or harmful to minors.”[392]  The court further stated that when strict scrutiny applies, “the government may not justify restrictions on constitutionally protected speech on the ground that such restrictions are necessary in order for the government effectively to suppress the dissemination of constitutionally unprotected speech.”[393]

    The court then determined that the public libraries had less restrictive alternatives that would serve the government’s interest in protecting the library patrons.[394]  For example, libraries can adopt Internet acceptable use policies that specify the material, such as child pornography, that patrons may not access and require that patrons sign the policy before using the computer.[395]  If a patron violates the policy, the library can “issue the patron a warning, revoke the patron’s Internet privileges, or notify law enforcement.”[396]  Further, the libraries may use the shoulder tap method, require minors to use specific computers, require minors to use blocking software in the absence of their parent, instruct patrons on the correct Internet search techniques to avoid inadvertently finding offensive material, or use privacy screens and recessed monitors.[397] 

    The government contended that even if the filtering software failed strict scrutiny analysis, CIPA’s provision allowing the librarian to disable the software cured the “lack of narrow tailoring inherent in filtering technology.”[398]  However, the court disagreed, stating that CIPA was unclear as to when a librarian could disable the program.[399]  Further, the court noted that the requirement that the patron wanting access to a blocked site has to “ask a state actor’s permission to access disfavored content violates the First Amendment.”[400]  This requirement would have a chilling effect, as the patron would not want to request that a librarian disable the filter when sensitive material is involved.[401]  Moreover, the court stated that the disabling provision failed to address the overblocking problem as the request was not acted upon immediately, imposing a significant burden on the patrons.[402] 

    Therefore, the court held that CIPA was facially invalid because it violated the First Amendment.[403]  A final issue the court analyzed was whether CIPA was severable from the other provisions of LSTA and E-rate funding.[404]  The court concluded that it was possible, as those other provisions would still be fully operational and there was no indication that Congress intended to repeal the LTSA and E-rate funding programs were CIPA to be found unconstitutional.[405]

    It is worth noting that the court did not analyze the plaintiffs’ alternate theories that CIPA was an invalid prior restraint or that it was unconstitutionally vague.[406]  However, while not deciding the issue, the court, in a lengthy footnote, analyzed whether CIPA is unconstitutional “because it requires public libraries, as a condition on the receipt of federal funds, to relinquish their own First Amendment rights to provide the public with unfiltered Internet access.”[407]  Approximately six weeks before this court decided American Library Association the Supreme Court decided Ashcroft v. ACLU and remanded the case to the Third Circuit Court of Appeals to resolve several issues that the circuit court had failed to consider.[408]  To potentially avoid a similar problem here, the district court found it necessary “to frame the disputed legal issues and to develop a full factual record for the certain appeal to the Supreme Court.”[409]  While the court did not decide the unconstitutional conditions theory, the court stated that its findings of fact and the “framing of the legal issue here, would allow the Supreme Court to decide the issue if it deems it necessary to resolve this case.”[410] However, the court also noted that “the plaintiffs have a good argument that CIPA’s requirement that public libraries use filtering software . . . in such a way that it constitutes an unconstitutional condition on the receipt of funds.”[411]

    On June 20, 2002, the government filed a direct appeal of the district court’s decision, as authorized under CIPA.[412] The Supreme Court noted probable jurisdiction and heard oral arguments on March 5, 2003.[413]

IV. CIPA’s Effect on Students and Teachers[414]

    When CIPA became effective on April 20, 2001, school districts around the country had to prepare themselves to meet the mandated guidelines.[415]  According to the provisions, schools receiving E-rate funds had to submit certification within 120 days of the beginning of the first funding year following CIPA’s implementation.[416]  Schools receiving funds for Internet access only under Title III of the Elementary and Secondary Education Act must submit certification as part of the application process for the next program year.[417]  For most schools, this deadline was October 28, 2001, but institutions receiving discounts beginning on July 1, 2001 had to certify by the earlier date.[418]  Regardless of the deadline, school districts had much work to do.[419]

    The students and teachers using the Internet as an educational tool ultimately feel CIPA’s impact.[420]  As computer use has become common in the classroom, the teacher’s role and teaching styles have changed.[421]  Indeed, states are recognizing the importance of including the computer in the curriculum as part of the child’s education.[422]  Additionally, some cities are beginning to use schools as community centers similar to public libraries.[423]

    Traditionally, education regulation has been a function of state and local governments.[424]  In San Antonio Independent School District v. Rodriguez,[425] a group of parents challenged the constitutionality of the Texas system of financing public education, contending it violated their rights under the Fourteenth Amendment.[426]  In holding that the finance system did not violate the Fourteenth Amendment, the Supreme Court stated that education was one of the most important services performed by the state, but was not a fundamental right under the United States Constitution.[427]

    Therefore, state constitutions provide for the students’ education.[428]  One of the goals of public education is to prepare students to become productive adult citizens who are “self-reliant and self-sufficient participants in society.”[429]  Most state courts have recognized both civic participation and preparation for employment as the basic purposes of a student’s education.[430]  Additionally, the state decides on the content of its educational curriculum and frequently includes technology instruction.[431]

    Students do not “shed their constitutional rights to freedom of speech or expression at the schoolhouse gate.”[432]  However, those rights “are not automatically coextensive with the rights of adults in other settings.”[433]  Thus, in the school setting and as a part of legitimate pedagogical concerns, school officials may censor speech that substantially and materially interferes with school work,[434] that is school-sponsored and is lewd or indecent,[435] or that appears to have the school’s approval.[436]

    Teachers, likewise, do not lose their First Amendment rights upon crossing the school’s threshold.[437]  However, the teachers’ rights, too, are narrowed to that appropriate for the school setting.[438]  In order to restrict speech of the teacher, school authorities must show that a disruption in the workplace would arise.[439]

    CIPA poses special problems for teachers in public schools.  First, the use of filters in the school instills a false sense of security that inappropriate material will not reach the child’s computer.[440]  Thus, uninformed teachers and administrators may not take other reasonable precautions.[441]  Second, teachers may refrain from assigning research projects with themes such as “breast cancer” or “AIDS” because the students would not be able to access a broad range of, if any, material on the schools computer system.[442]  Third, students may not choose projects on sensitive subjects, thereby having a chilling effect on students’ speech.[443]  Fourth, the teacher is unable to preview or evaluate Web sites at school that she may wish to use for instruction.[444]

    States also recognize the importance of using the computer as a tool needed for future employment.[445]  As the New Hampshire Supreme Court stated in Claremont School District v. Governor,[446]

Mere competence in the basics—reading, writing, and arithmetic—is insufficient . . . to insure that . . . public school students are fully integrated into the world around them.  A broad exposure to the social, economic, scientific, technological, and political realities of today’s society is essential for our students to compete, contribute, and flourish in the twenty-first century.[447]

    Like public libraries, schools have other alternatives to mandatory filters on Internet computers.[448]  For example, the teacher can instruct the child how to avoid inappropriate sites by selecting search criteria.[449]  Schools can create an Internet Safety Plan or an Acceptable Use Policy, such as required by CIPA, which could include policies for student violations.[450]  The teacher could create a list of recommended Web sites to assist students in researching a particular project.[451]  Finally, the possibility of a student accessing an inappropriate site is reduced when the teacher monitors and assists the students while online.[452]

Conclusion

    The government has long had an interest in protecting children from obscene and indecent materials whether in print, motion pictures, or on the Internet.[453]  Even though the Constitution protects speech that is indecent, obscenity and child pornography have never enjoyed the same protection.[454]  However, the Constitution protects speech offensive to some people, while allowing others to view it.[455]

    The use of the Internet as an educational tool has grown significantly in the past few years.[456]  Congress has struggled with a means to prevent children from viewing indecent and harmful material.[457]  Congress first passed the CDA and citizens immediately challenged its constitutionality.[458]  The Supreme Court held that the CDA violated the First Amendment as it placed an “unacceptably heavy burden on protected speech” and was not narrowly tailored to achieve the government’s compelling interest.[459]

    Taking the suggestions of the Court, Congress attempted to rectify its errors by drafting COPA.[460]  In COPA, Congress included changes to conform to the tests the Court announced in Ginsberg v. New York and Miller v. California.[461]  COPA did not prevent parents from purchasing prohibited materials for their children, applied only to communication for commercial purposes, adopted the “harmful to minors” standard, and defined a minor as someone under the age of seventeen.[462]  Citizens immediately challenged the COPA’s constitutionality, and, as with the CDA, the court found that COPA violated the First Amendment.[463]  In its decision, the court held that COPA’s contemporary community standard was too limiting, as any speech on the Internet would have to “comply with the regulation imposed by the State with the most stringent standard . . .”[464]  While the Supreme Court held that the community standard did not render COPA unconstitutional per se, it remanded the case to the circuit court to consider other possible constitutional violations.[465]

    As a third attempt, Congress passed the Children’s Internet Protection Act.[466]  One major difference with CIPA is that it affects the receiver, not the publisher.[467]  In order to receive certain types of federal funding, libraries and schools must create an Internet Safety Policy, hold a public meeting to discuss that policy, and submit certification that it has installed blocking software on all computers in order to prevent minors from viewing visual depictions of obscenity, child pornography, or material harmful to minors.[468] 

    Again, plaintiffs challenged CIPA’s constitutionality on several grounds.[469]  The court held that public libraries were a limited public forum and any regulations imposed must survive First Amendment scrutiny by being narrowly drawn to achieve a compelling governmental interest.[470]  Further, the court stated that by opening its doors to the public, and then blocking a substantial amount of constitutionally protected material through the use of filters that cannot block the visual depictions the Act specifies, CIPA selectively excluded some speech.[471]  Additionally, software manufacturers cannot produce a filter that does not overblock or underblock.[472]  The court noted that the government does have a legitimate interest in protecting children from receiving obscenity, child pornography, and material harmful to minors.[473]   However legitimate that interest is, the Court held that CIPA’s use of filters was not narrowly tailored to meet that end.[474]  Additionally, a public library has alternatives that would be as effective as blocking software in meeting the governments’ compelling interest without preventing patrons from receiving constitutionally protected speech.[475]  Finally, the court held that CIPA’s disabling provision would chill speech by giving the library official unfettered blocking discretion.[476]  Adults, fearing the stigma in the community, would avoid asking a librarian to unblock or disable the filter.[477]  Additionally, because CIPA contains no guidelines, the library official has total discretion whether to disable the filter based on his own personal choice.[478]

    The Supreme Court will soon decide if CIPA is constitutional as the government has appealed the district court’s decision.[479]  However, comparing CDA, CIPA, and COPA, similarities emerge in how the courts determined the Acts’ constitutionality.  For example, CIPA requires the use of a virtually non-existent technology to block visual depictions.[480]  The Court found this a troublesome issue in holding the CDA constitutional.[481]  Likewise, the district court made a similar observation in holding that COPA violated the First Amendment.[482]  Additionally, the district court, in declaring CIPA unconstitutional, noted that both CIPA and the CDA “effectively prohibited the distribution to adults of material that was constitutionally protected with respect to adults.”[483]

    Regardless of whether the Court finds CIPA unconstitutional, teachers still have a duty to prepare students for their adult lives, and today Internet in the classroom can lead to a significant advances in teaching.[484]  Although students do not have the same rights as adults, filtered computers still block information that a child has a constitutional right to receive.[485]  Arguably, a child should not download obscene or sexually explicit materials.  However, there are other methods to deal with the few violators, such as school discipline or loss of Internet privileges.[486]  Teachers should instruct students on appropriate methods of searching Internet Web sites in preparation for the workplace, as most businesses do not use filters on their computers.[487]  Teaching self-control and restraint at an early age will help prevent problems in the future.[488]

    CIPA forces schools and libraries to make a choice.  Many institutions will elect to forgo funding.[489]  However, some poorer communities may not be able to afford that choice.[490]  Unfortunately for students, many schools will elect to filter the school’s computers, resulting in a potential loss of, literally, a world of information.

Michael J. Brown