Stark Revelations

Last year the U.S. Department of Justice hired Philip B. Stark, a statistician at the University of California at Berkeley, to scour the Web for porn. The government wanted hard numbers to bolster its defense of the Child Online Protection Act, a 1998 law requiring commercial websites to verify users’ ages before allowing access to anything deemed “harmful to minors.” The American Civil Liberties Union (ACLU) had challenged the law on constitutional grounds, and the Supreme Court had sent the case back to the U.S. District Court for the Eastern District of Pennsylvania.

When the trial began last October, the Department of Justice was ready with some shocking revelations from Stark’s report, which cost upward of $1 million, required a subpoena to pry private data from Google, and took months to complete. Among them: “The number of sexually explicit websites is huge”; “Search results often include sexually explicit material”; and 1.1 percent of the websites indexed by the Google and MSN search engines are sexually explicit.

The government ostensibly wanted the numbers to underscore the need for censorship. But it’s hard to see how this percentage, or any other, is significant, since Internet space is not for practical purposes limited. It says even less about the average kid’s interaction with sleaze, since the Web is rarely experienced as a random sequence of sites. As the ACLU’s Chris Hansen put it in the trial’s closing statement, “Some people find [it] a scary number. Some people find it a reassuring number. What it mostly is, is an irrelevant number.” And an expensive one.