Comstock in the '90, or the Index of Prohibited Web Books

Testimony for McCain hearings

By Seth Finkelstein sethf@mit.edu

I'll start my testimony by describing myself briefly. I am an MIT graduate with degrees in Math and Physics, who has been working as a computer consultant for over a decade. I have maintained a web site at MIT about general free-speech issues (http://www.mit.edu/activities/safe/) for more than four years now, and have made available on it an extensive collection of resources called the Information about Labeling and Rating Systems page (http://www.mit.edu/activities/safe/labeling/summary.html), for at least two and a half years. Incidentally, this entire site was at one time blacklisted by a leading censorware program, CyberPatrol as evil under categories ranging from Full Nudity to Militant/Extreme to Satanic/Cult. I felt honored.

Let me begin here by commenting on some of the language I used above. I like to speak of "censorware" which "blacklists" or "bans" sites, not "filtering programs" which "filter" the Internet. As one who write software by trade, I am not at all intimidated by technology, but I have seen too many commentators who are. The word "filter" is much too kind to these programs, it conjures up inaccurate gee-whiz images of sophisticated, discerning, choice. When these products are examined in detail, they usually turn out to be the crudest of blacklists, long tables of hapless material which has run afoul of a stupid computer program or person, perhaps offended by the word "breast" (as in possibly "breast cancer") and so consigned to a modern-day electronic version of the "Index of Prohibited Books".

It is not my intention to declare that there is no material on these blacklists which would arguably be legally obscene. But then blacklists of the McCarthy period undoubtedly contained many real Communists. It is the unchecked and shameless power that such panics put in the hands of demagogues which is the critical issue.

Unlike many writers on this topic, who lack a historical perspective, I was not at all surprised when several censorware programs were exposed to have been categorizing everything from feminism to gay-rights to anti-censorship information to be, colloquially, "pornography". What do you expect when you take the most censor-minded people around, and have them make up a secret unaccountable blacklist, with the intent of suppressing material far and wide? Anthony Comstock, one of the US government's early censors, classed information about birth-control and abortion as obscenity, and prosecuted people for such crimes against his concept of virtue. His successors of today have not been quite as successful in victimizing their targets, but they cast as wide a net.

The above isn't simply airy rhetoric or an isolated instance. For example every censorware program with an extensive list, which has been examined in detail, has been caught placing feminist organizations on their blacklist. CyberPatrol, one of the most common, bans a USENET social-issues feminist discussion group ("soc.feminism"). NetNanny and CyberSitter blacklist the National Organization for Women. X-Stop was exposed to be banning the American Association of University Women, even in its special "Library Edition" which allegedly contained only material meeting a legal standard of banning (on this last, see https://www.spectacle.org/cs/xstop.html and https://www.spectacle.org/cs/seth.html)

Why did these companies get away with such outrageous conduct, and still do to a large extent? Because their blacklists are almost all secret, and even if they becomes available one way or another, almost no-one has the time to examine them. CyberPatrol claims to have over 50,000 entries. At a minute per item, that's more than ONE HUNDRED WORKDAYS for a person just to give the whole thing a cursory inspection. This realization, by the way, should immediately destroy the myth that a librarian or schoolteacher can look over a product and make more than insignificant adjustments for his or her own values. As one of the authors who participated in producing a report (https://www.spectacle.org/cwp/) containing one of the few detailed examinations of a censorware program ever to be found, I am in a rare position to testify as to the difficulty involved and the effort such evaluation requires.

I would like to join my college Michael Sims in repudiating David Burt's citation regarding our report. He writes about it: "After a group calling itself "The Censorware Project" decrypted CyberPatrol's list, they found about 60 bad blocks, or about .1% of the total blacklist." This is a mendacious distortion, on several fronts. The whole blacklist was not available to us, only a small portion, and even if it were, we didn't have the time or resources to go through it all. We never claimed to be doing an exhaustive search. On the other end, the phrasing "60 bad blocks" is extremely misleading. Some entries encompassed huge numbers of individual authors. Moreover, we didn't put everything we could have investigated into our released report. We didn't want some aspects to be distracted by, for example, arguments as to whether dating services deserved to be banned as referring to "Sexual Acts".

This gives just a small glimpse into the manifold technical, social, and last but should not be least, Constitutional, problems with the legislation under discussion. I urge you in the strongest terms to reject it completely.