technical filtering process implemented by ISPs resulted in large-scale overblocking
that further violated free speech rights. Ultimately, the Federal District Court in Philadelphia ruled that the law violated the First Amendment. This ruling details the process, both legal and technical, that led to the blocking of 1.5 million legitimate Web
sites while trying to block access to approximately 400 Web sites suspected of containing child abuse images. The decision by Judge Jan E. duBois stated:
There is little evidence that the Act has reduced the production of child pornography or the child
sexual abuse associated with its creation. On the other hand, there is an abundance of evidence
that implementation of the Act has resulted in massive suppression of speech protected by the
First Amendment. 32
The decision explicitly confirmed what had been implicitly noted by NGOs throughout their involvement in international agreements and conferences on child pornography: state-mandated national filtering was technically ineffective and posed a threat to
the right of freedom of expression. While the case of the United States may be exceptional given the strong preference for the First Amendment guaranteeing the right to
freedom of speech, the events in Pennsylvania did not go unnoticed. A report published in 2004 by UNESCO after the World Summit on the Information Society in
2003 noted that some countries were beginning to follow the legislative approach of
the United States by introducing their own legislative efforts to ‘‘require the implementation of filtering at ISPs and gateways.’’ 33 However, these attempts were plagued by
the same issues faced in the United States, and widespread adoption did not occur until
filtering was reconstituted as a legitimate practice.
The Legitimization of Filtering
The legitimization of filtering centers on three interlocking developments: a model of
implementation in which the role of the state is reduced, the delinking of filtering and
free speech concerns through technological developments, and a reframing of the effectiveness of filtering.
In 2004, British Telecom (BT), the largest ISP in the United Kingdom, and the Internet Watch Foundation (IWF), a ‘‘tip line’’ for reporting illegal online content, developed
a partnership in consultation with the government. 34 British Telecom agreed to block
access to a list of Web sites compiled by the IWF. This model of private partnership
removes the need for government to implement legislation requiring ISPs to filter—it
is a voluntary private initiative. This development changed the perception of filtering
from one of state imposition to one of private initiative. While technically a private arrangement, the state’s involvement in the process leading to the implementation of filtering allowed it to retain some influence. 35 British Telecom’s filtering system, known
as Cleanfeed, was designed to be extremely precise and cost effective. 36 Cleanfeed elegantly avoids the pitfall of overblocking, the key objection that was consistently raised
62
Nart Villeneuve
that further violated free speech rights. Ultimately, the Federal District Court in Philadelphia ruled that the law violated the First Amendment. This ruling details the process, both legal and technical, that led to the blocking of 1.5 million legitimate Web
sites while trying to block access to approximately 400 Web sites suspected of containing child abuse images. The decision by Judge Jan E. duBois stated:
There is little evidence that the Act has reduced the production of child pornography or the child
sexual abuse associated with its creation. On the other hand, there is an abundance of evidence
that implementation of the Act has resulted in massive suppression of speech protected by the
First Amendment. 32
The decision explicitly confirmed what had been implicitly noted by NGOs throughout their involvement in international agreements and conferences on child pornography: state-mandated national filtering was technically ineffective and posed a threat to
the right of freedom of expression. While the case of the United States may be exceptional given the strong preference for the First Amendment guaranteeing the right to
freedom of speech, the events in Pennsylvania did not go unnoticed. A report published in 2004 by UNESCO after the World Summit on the Information Society in
2003 noted that some countries were beginning to follow the legislative approach of
the United States by introducing their own legislative efforts to ‘‘require the implementation of filtering at ISPs and gateways.’’ 33 However, these attempts were plagued by
the same issues faced in the United States, and widespread adoption did not occur until
filtering was reconstituted as a legitimate practice.
The Legitimization of Filtering
The legitimization of filtering centers on three interlocking developments: a model of
implementation in which the role of the state is reduced, the delinking of filtering and
free speech concerns through technological developments, and a reframing of the effectiveness of filtering.
In 2004, British Telecom (BT), the largest ISP in the United Kingdom, and the Internet Watch Foundation (IWF), a ‘‘tip line’’ for reporting illegal online content, developed
a partnership in consultation with the government. 34 British Telecom agreed to block
access to a list of Web sites compiled by the IWF. This model of private partnership
removes the need for government to implement legislation requiring ISPs to filter—it
is a voluntary private initiative. This development changed the perception of filtering
from one of state imposition to one of private initiative. While technically a private arrangement, the state’s involvement in the process leading to the implementation of filtering allowed it to retain some influence. 35 British Telecom’s filtering system, known
as Cleanfeed, was designed to be extremely precise and cost effective. 36 Cleanfeed elegantly avoids the pitfall of overblocking, the key objection that was consistently raised
62
Nart Villeneuve