What's a Little Information Between Neighbors? -- The Policing of Offensive Content on the Internet

    Imagine that you operate an Internet site where users can offer personal items for sale to the entire Internet community. You operate your site from the United States, attracting millions of users from around the world buying and selling millions of items. Transactions are impossible to monitor due to the large number of users and items sold. The site is divided into numerous categories to make it easier for users to post items for sale and purchase those items. Users of your system buy and sell everything from cars to computer supplies to books to jewelry and everything in between. Sites such as yours are often thought of as online auctions. Your site is different from the traditional notion of an auctioneer, however. Once the site is in place, it is up to the users to do all the work. The users post their items for sale, select items to purchase, and are responsible for completing the transaction once purchased.

          Your business is operating smoothly, facilitating millions of transactions, when you surprisingly find that you have been charged with a crime in a foreign country. The government of that country claims that you have broken its laws, even though you have never even set foot there! Nobody from that foreign country had even used your site, although anybody with an Internet connection could have found materials for sale that are illegal to buy and sell in that country, even though completely legal in the U.S. The foreign government claims that by operating your Internet site, you are committing a crime. They give you an ultimatum: either monitor the contents of the goods on your system and filter out those offending items from users in that country, or face harsh fines.

          The Internet company Yahoo! faced just such a problem in 2000 when the French government charged it with violating a French law prohibiting the advertisement, exhibition, or sale of any objects likely to incite racial hatred. The charges came after Jewish and anti-racist groups in France expressed outrage that Nazi memorabilia were among the items available on Yahoo!'s auction pages. Although Yahoo!'s French language portal, Yahoo.fr, does not contain any of the offensive material, French users are able to access the offensive material through the Yahoo! English language site. There were no documented instances of any French citizens ever purchasing or making available the offensive material.[i]

          This example illustrates the problem that the developing uses of the Internet present when they encounter the numerous conflicting cultural and political differences that exist in the world. Traditionally, territorial borders determine the domain of governments and thus those governments' power to control the people within their borders. The laws those governments implement reflect the moral and religious beliefs of the people within their boundaries. Often laws prohibit certain activities that violate those beliefs. The Internet, in contrast, has no territorial boundaries, because the cost and speed of message transmission on the Net is almost entirely independent of physical location.[ii] Is it possible for a particular government to protect its people from material on the Internet that offends its moral and religious beliefs? Can one particular government legislate standards that are applicable the whole of the Internet? Does Yahoo! have an obligation to protect individuals from accessing material on its system that might offend a person in the Middle East?

          This paper will discuss these questions and what solutions exist when conflicting cultures meet the global entity known as the Internet. First, I will examine the various forms this conflict takes and how different countries have dealt with the problem. Second, I will examine the technological capabilities currently in place to allow governments and Internet providers to customize the content available to different countries. Third, I will examine possible solutions to the problem and offer a recommendation to maximize the potential of the Internet while respecting the Interests of all parties.

Cultural conflict and the Internet

          The Internet is a widespread series of interconnected computer networks, often referred to as a “network of networks.”[iii] The owners of the computers and networks that make up the Internet include government and public institutions, non-profit organizations, and corporations.[iv] One court described this new technology as: “The resulting whole is a decentralized, global medium of communications - or “cyberspace” - that links people, institutions, corporations, and governments around the world.”[v] One of the defining characteristics of the Internet is its durability. As envisioned by its original creators, the Internet can still function if portions of it do not work, by finding alternate routes for information to travel through the system. Another defining characteristic of the Internet is the absence of any geographic boundaries: once a user is connected to the Internet anywhere in the world, he or she is able to have the same access to information as any other user in the world.

          New issues in cultural conflicts are arising as the Internet continues to develop in new parts of the world. In 1995, over two-thirds of Internet users were located in the United States. By the year 2005, the United States is projected to have less than one-third of all Internet users. The fastest areas of growth will be in areas such as the Middle East, South and Central America, and Asia[vi]. These areas of the world have cultures dramatically different from that of the United States and Western Europe, who up to this point have dominated the Internet in terms of number of users. As Internet users from diverse cultures become more and more prevalent, solutions will be needed to deal with the inevitable growth in cultural conflicts that arise.

In the case of U.S. v. Thomas[vii], the obscenity-based convictions in Tennessee of a California couple, who operated a "for fee" Bulletin Board Service that allowed members to download pornographic materials, were upheld after the court rejected the defendants claim that under the First Amendment, the "community standards" by which the "obscene" nature of the materials should have been measured was that of the "cyberspace community" and not that of Memphis, Tennessee. The Thomases were indicted by a grand jury in the Western District of Tennessee and were convicted under, among other federal statutes, 18 U.S.C. § 1465 for knowingly using a facility and means of interstate commerce -- in this case, the combined computer/telepphone system – for the purpose of transporting obscene, computer-generated materials in interstate commerce. The United States District Court for the Western District of Tennessee rejected the defendants’ First Amendment arguments. One prong of  First Amendment analysis is whether the average person, applying contemporary community standards, would find that the work in question, taken as a whole, appeals to the prurient interest. It was argued on behalf of the defendants that the relevant community standards were not those of Memphis, Tennessee, where the defendants had been prosecuted, but rather a new definition of community was needed, i.e., one that was based on cyberspace, rather than the geographic, connections among people. The court rejected this argument, first holding that "obscenity is determined by the standards of the  community where the trial takes place" and that "it is not unconstitutional to subject interstate distributors of obscenity to varying community standards." The court acknowledged the concern expressed by the defendants that such a ruling might lead to an impermissible chill on protected speech because BBS operators cannot select who gets the materials they make available on their bulletin boards and would be forced to censor their materials so as not to run afoul of the standards of the community in America with the most restrictive standards. However, the court believed that the First Amendment was not implicated by this prosecution because, unlike the hypothetical bulletin board operator who has no knowledge or control over the jurisdictions where materials are being distributed for downloading or printing, access to the defendants’ BBS could only be obtained by revealing one’s geographic location. Because on the facts of this case the defendants "had in place methods to limit user access in jurisdictions where the risk of a finding of obscenity was greater than in California" and could avoid liability in jurisdictions with less tolerant obscenity standards by refusing membership to persons in those communities, the Court held that it did not need to adopt a new definition of "community" for use in obscenity prosecutions involving BBSs and left for another day the First Amendment questions implied, but not directly presented, by this case.

          The First Amendment of the United States Constitution states that “Congress shall make now law … abridging the freedom of speech.”[viii] However, the Constitution does not protect all forms of speech. Obscene material does not receive any First Amendment protection. Indecent speech, on the other hand, is constitutionally protected and subject to strict scrutiny. This means that in order for a regulation to survive strict scrutiny “the State must show that its regulation is necessary to serve a compelling state interest and is narrowly drawn to achieve that end.” The strict scrutiny standard is applied to any government legislation that limits the content of protected speech. The Court, however, treats forms of communicative technology differently when applying First Amendment standards because of the different characteristics of each medium.[ix]

In the landmark Reno v. ACLU[x] the Court decided to give the Internet the same level of protection as that received by the print media. In Reno, the ACLU claimed that the Communications Decency Act of 1996 (CDA) provisions concerning the Internet infringed upon protected First and Fifth Amendment rights.[xi] The U.S. District Court for the Eastern District of Pennsylvania conducted an extensive fact-finding including everything from the history of the Internet to the specific issue of sexually explicit material.[xii] In a unanimous decision, the three-judge panel granted an injunction of the enforcement of the CDA and ruled the CDA unconstitutional. The case was then directly appealed to the Supreme Court, where the Court affirmed the decision.[xiii]

The Court analyzed the statute under strict scrutiny. The Court first attacked the statute as being overly broad, and then noted that alternatives less restrictive of First Amendment rights were available for achieving the same purpose. Therefore, the majority concluded that the CDA was not narrowly tailored to meet its objective of protecting minors because less restrictive means existed to achieve the stated purpose.[xiv]

The Internet is a new area of media technology that the Court chose not to stifle by allowing the passage of a statute that was overly broad in scope. The Internet is now closer, in terms of First Amendment protection, to print media than it is to broadcast media or common carriers. The Court’s reasoning in Reno seems to place the Internet in the realm of protection offered by Miami Herald Publ’g Co. v. Tornillo,[xv] and suggests that a vague statute implicating First Amendment freedoms will not pass the strict scrutiny standard used by the Court. Reno is indicative of the strong protection the United States uses for expression, in contrast to many other countries.

          Interestingly, the United States is not immune to the threat of offensive content that other countries and cultures could find acceptable, or at least allowable within their legal frameworks. The FBI recently began an investigation of an Internet site, Bonsaikitten.com, run by MIT students, after receiving complaints from people offended by the site's depiction of animals being treated cruelly.[xvi] Although the site is clearly a spoof, claiming to apply the ancient bonsai techniques of Japan to kittens, FBI officials apparently were not amused. The probable crime involved is a federal law passed in December 1999 prohibiting the transfer across state lines a depiction of animal cruelty, which includes the Internet. Although this law has not been challenged on Constitutional grounds as of yet, it is possibly unconstitutional. Regardless of its constitutionality, it illustrates how the United States, with its tradition of allowing a free flow of information, may have its limits that can be violated by the borderless Internet.

Singapore

          As one might expect, the proliferation of cyberporn has motivated some nations to regulate Internet content in different ways. In policing the Internet, some nations have prohibited content based on broad terminology, such as being against public interest or public morality. As an example, in 1996, Singapore enacted an elaborate administrative law framework for Internet content reflecting its tight control of the media. In developing its Internet content regulations, Singapore had to resolve the obvious tension between its aggressive information technology growth strategies that allowed colossal amounts of uncensored information into the country via the Internet and the government’s traditional restrictions on media.

          Singapore has policies in place to actively encourage Internet development.[xvii] Singapore forbids Internet Service Providers and Internet Content Providers from providing “material that is objectionable on the grounds of “public interest, public morality, public order, public security, national harmony, or that is otherwise prohibited by applicable Singapore laws.”[xviii] To enforce the Code, ISPs and ICPs must use their “best efforts” to comply with the Code and must act to ensure that nothing is included in any broadcasting service (which includes Internet content) that is against what the government considers to be in the public’s interest and in good taste or decency.[xix] Although content providers within Singapore are prohibited from providing offensive materials, ISPs are not required to monitor the Internet or its users. They are, however, required to eliminate access to 100 “high impact” pornographic sites, as identified by the Singapore government, as a “statement of societal values.”[xx] In contrast to the rather lenient policies adopted by Singapore to offensive content, the actual enforcement of these policies is perhaps more stringent.[xxi]

          Despite Singapore’s high degree of technological advancement and the relatively small number of users and content it must regulate, censoring the Internet has proved virtually impossible. Singapore has realized that it is unfeasible to censor the Internet in the same manner as other types of media. Although the government has received assistance from ISPs (which are either owners by or connected to the government) in censoring the Internet, the head of the Singapore Broadcasting Authority, the agency in charge of Internet regulation, notes that “there is a limit to what domestic legislation can achieve in the face of a global and borderless medium like the Internet that it was impossible to fully regulate the Internet.”[xxii]

Australia

          Another example of a country that has taken a different approach to the regulation of Internet content is Australia. Due to fears that the relative ease of availability of any form of content via the Internet would shake the complex system of controls and regulations Australia places on telecommunications, publishing, and broadcasting, conservative elements pushed through the adoption of the Broadcasting Services Amendment Act 1999.[xxiii] This legislation, which attempts to protect Australia’s children from threats such as pornography, neo-Nazis, pedophiles, and bomb-making recipes, has faced mixed reactions from the public. The Act establishes a complaint-based system that gives the Australian Broadcasting Authoring (ABA), analogous to the Federal Communications Commission of the United States, the power to require illegal or offensive Internet sites to be taken down or for access to those sites to be prevented.

          Australia’s regulatory framework attempts to work on the principle that what is illegal in traditional communications should also be illegal online. The statute treats differently content depending on whether it is hosted within Australia or overseas. If a complaint is filed about material hosted in Australia, the host is ordered to cease carrying the offending material. Alternatively, if a complaint is filed about material hosted outside Australia, ISPs are required to take reasonable steps to prevent end-users accessing the prohibited material. Obviously, if no complaint is made, content is not restricted. Also, due to the shear number of possible offending sites, it would be impossible to regulate them all. There have as of yet been no legal challenges to the Australian legislation, and since Australia has no free speech protection in its constitution, it is not clear that there would even be any grounds compelling a court to reject the regulations.

          After a French court ordered Yahoo! to filter its system to make offensive memorabilia unavailable to people in France, Yahoo! removed the offending items from its auctions. Yahoo! did not comply with the court order, instead filing a countersuit in California arguing that: 1) it is technologically impossible for Yahoo! to comply with the french court's filtering order, and 2) the French court has no jurisdiction of U.S.-based Yahoo! Although the offending items were removed from its Webside, Yahoo! claims the action was due to a change in its own policies, and not to comply with the French court.[xxiv] Yahoo! likely does not wish to pursue a possibly lengthy and complex court battle. While French activists seem pleased with Yahoo!'s actions, the is far from settled, and they have expressed their desire to police offensive material

Technology

          As individuals, governments, and content providers grapple with the problems of offensive content, many see the problem that was created by technology as best solved by more technology. Content filtering software is becoming increasingly available in the marketplace. Filters come in different forms. While some are implemented "upstream," at the level of proxy servers that control access for whole schools, libraries, or busiensses, others are implemented "downstream" at individual workstations or PCs. Conceptually, they all work similarly: they start with "control lists" of addresses of unacceptable sites, then add automatic keyword filters to block additional sites that contain certain words and phrases. The user of the filter can specify the categories of sites to block, for example, "hate speech" or "sex acts." Technology also exists to monitor the sites accessed by users.[xxv]

          There are inherent inadequacies in the present filtering systems. To block objectionable material, the filter must be over-broad, blocking sites that aren't objectionable at all. For example, in a recent study, 1,000 randomly chosen addresses in the dot-com domain were submitted to the SurfWatch filter. Of the sites it blocked as "sexually explicity," more than four out of five were misclassified--for example, the sites of an antiques dealer in Wales, a Maryland limo service, and a storage company in California.[xxvi] The reason for this overblocking is because it is impossible to single out porn sites simply by the words they use. Other types of objectionable content are even harder to filter because they do not give themselves away so easily with genre-specific keywords such as "XXX". Another problem is that as the size of the Web continues to grow, the vast amount of objectionable material will likewise grow.

Solutions

          As the Internet continues its growth and an increasing amount of content is included within its realm, every conceivable type of information will become available, every conceivable viewpoint will be expressed in one way or another, and countless types of human creations will be displayed for all to see. This wealth of information, viewpoints, and creations will inevitably cross borders, sometimes to places where such offerings are not welcome. Such will be the cost of participation in the global network in which each society will be a part. This does not mean, however, that we will live in a homogenized world, where the only standard will be the lowest common denominator. Each of our cultures will form the basis for our participation in the global environment, the way in which we view the world, and the way in which the world views us. What will change is that underlying each culture's involvement in the global environment will be a heightened awareness, and tolerance of, the societal standards of others. Countries and cultures who more quickly and easily embrace the free flow of information will have a clear advantage over those who try to hold onto rigid standards of information control. And yet even the most libertarian cultures will face limits on their tolerance and need to somehow limit some types of Internet content in some situations.

          There will be three keys to solving this balancing problem: 1) technology; 2) education; 3) global cooperation. First, there will always be technological solutions to technological problems. Problems created by technology (the proliferation of offensive content via the Internet) can be solved by technology (filters). These solutions, however, will only be partial, and will have their own drawbacks. This is a fact of nature in the technological environment. New technology always creates problems that must be dealt with -  the technology is always one step ahead of the solution. Further, the pace of technological advancement is far quicker than policy advancement, so that legal standards in place to limit Internet content based on technological capabilities will always be at least a step behind. Governments will use technology to set standards on Internet content available to their citizens, but these technological solutions will not be fully adequate to solve the problem.

          Second, individuals, governments, and service providers will become more educated about the usefulness of the Internet and the opportunities it provides. This will lead to a better understanding of how to deal with the problem. A sort of maturity will develop in which people and institutions will learn to make available for themselves the content they want and ignore that which is offensive. For example, governments can provide recommended or required sites to be used as access points to the Internet, which could direct people toward content that meets their societal standards. Also, as individuals become more savvy, they can select content providers that will cater to their particular content requirements. A sophisticated user with access to the Internet will always be able to find his way into those dark alleys that contain offensive or obscene content, but the majority of people will stick to the well-lit highways that provide the content they need in the easiest manner possible.

          Third, global cooperation is required to strike the balance between a flourishing Internet and respect for cultures. Eventually, all countries and most people will likely have a presence on the Internet. People will associate through language, not location. Regardless of how the Yahoo! case, and those like it come out, it will be entirely impractical and ridiculous for one jurisdiction to attempt to censor a content provider over whom it has no power. The mere existence on the Internet of a piece of information, and therefor the availability of that information to anyone in the world with Internet access, does not give any one country the power to judge that information and determine its fate. The burden will be on those who attempt to restrict the free flow of information, not on those who provide information, no matter how offensive it might be. It should be the common goal to allow the free flow of all information to all nations to the extent each deems acceptable.

Necessary to achieve this goal will be international cooperation. An international Internet oversight body should be developed, at least part of which is focused on striking the balance that each country must find between the free flow of information on the Internet and societal standards that require content restriction. This body should have as its purpose the development and distribution of essential technologies to allow nations to adopt their own content standards to the extent possible. Also, this body should encourage the education of institutions and people about the possibilities and hazards of the vast array of content available online. The one overarching principle that this body must realize and embrace is that to maximize the effectiveness of this unprecedented resource for all people, undue burdens cannot be placed on the free flow of information.