In Bahrain, Iran, Saudi Arabia, the United Arab Emirates, and Yemen, ISPs--either under government orders or pressure--all block web sites on the basis of their content.⁽⁵²⁾ In the first four if not all of these countries, blocking extends to cultural and/or political content.

U.A.E. officials are perhaps more forthcoming than others that engage in blocking. They justify it as a means of limiting access to pornography. Persons in the Emirates who dial up a local service provider do not have unmediated access to the Internet. Rather, their requests are routed through a proxy server maintained by the government-controlled telecommunications company, Etisalat.⁽⁵³⁾ When a user requests a web site, the request will be refused if the filtering software on the proxy server detects "objectionable" material on the site or finds that the site is on a list of banned sites. The presence on the U.A.E.'s list of taboo sites of a political and cultural web site for Arab homosexuals refutes official claims that only pornography is banned (see below).

A proxy server such as the ones in place in the U.A.E. and Saudi Arabia can be used by authorities to track which computer terminals are accessing which web sites and for how long. U.A.E. authorities deny monitoring individual web use. In Saudi Arabia, however, users who request a site that is blocked get a message on their screens warning that all access attempts are logged.⁽⁵⁴⁾

State-controlled or state-influenced Internet Service Providers in Tunisia, Iran, and Bahrain block web sites containing political or human rights criticism of the government. Tunisia has acknowledged blocking only those sites that offend moral values, but apparently this includes the sites of various human rights organizations critical of the government (see below).

Bahrain has been less aggressive than neighboring Saudi Arabia and the U.A.E. in trying to block pornography centrally. Although there are reports that it is blocking some sexually explicit sites, Batelco, the state-controlled monopoly service provider, seems to rely more on encouraging subscribers to download filtering programs, should they desire it.

The ability of minors to access online pornography was one impetus for the enactment in 1996 by the U.S. Congress of the Communications Decency Act (CDA). The act criminalized online communication that is "obscene, lewd, lascivious, filthy, or indecent, with intent to annoy, abuse, threaten, or harass another person," or "obscene or indecent" if the recipient of the communication is under eighteen years of age.

The CDA was opposed by a coalition of human rights and free expression groups. As Human Rights Watch, a member group of the coalition, argued in an affidavit to the Supreme Court, the law's prohibition of "indecent" speech could be applied to its own human rights reporting that includes graphic accounts of rape and other forms of sexual abuse. In an important ruling for Internet liberties, the U.S. Supreme Court in 1997 voided key provisions of the act on the grounds that the Internet is entitled to the highest level of free speech protection and the act violated that principle.⁽⁵⁵⁾

While cautioning that all methods of controlling access to Internet content pose dangers to the right to impart and receive information, some international civil liberties organizations have argued that filtering software installed by the end-user raises fewer freedom-of-speech concerns than censorship imposed from above.⁽⁵⁶⁾

Filtering software works by examining a data stream and blocking any material that matches specified criteria, such as the presence of "stop words" in a web site's home page or URL, or the presence of the URL on a list of banned sites A more aggressive type of software, such as the KidDesk Internet Safe program, "white-lists" material by prohibiting access to all materials except those that are explicitly approved by the parent or supervisor.

Making the case that end-user filtering software is preferable to government censorship, the Center for Democracy and Technology, a Washington, D.C.-based organization, argues:

By placing control over content in the hands of individual parents, as opposed to bureaucrats and prosecutors, policy makers can assure full respect for our constitutional protection of freedom of expression and enable the Internet to grow free from unnecessary and ineffective regulatory interference....A great variety of blocking and filtering software exists which can serve the diversity of family values of American communities, providing choice to families online without infringing the constitutional First Amendment rights of Internet users.⁽⁵⁷⁾

Whether employed by end-users or government servers, filtering software is an imprecise tool that in practice almost always blocks materials beyond its stated purpose. This is true even of the more sophisticated products that are regularly updated by staff members who check web sites for "suitability."⁽⁵⁸⁾ According to popular anecdotes, software that hunted for specified "stop words" ended up blocking sites that mentioned the English county of Middlesex and recipes for chicken breasts. One popular program blocked a web site containing information on AIDS prevention. Even a pro-filtering web site was once blocked by Surf Watch, a leading filtering program, because it shared a commercial server with a pornographic site.⁽⁵⁹⁾

59. Matt Richtel, "Tables Turn on a Filtering Site As It Is Temporarily Blocked," New York Times, March 11, 1999.