Internet Blocking Programs and Privatized Censorship

by Seth Finkelstein

In July 1996, a CyberWire Dispatch article by Brock Meeks and Declan McCullagh, entitled "Keys to the Kingdom", revealed many unsavory aspects about the heretofore secret contents of popular Internet blocking program (i.e., software packages which prevent viewing of "undesirable" material). Though commonly discussed as programs which block "porn" or "smut", the actual range of material suppressed was far wider. To be sure, the lists of offensive content contained much sexually-explicit material. But various programs turned out to also ban everything from feminist discussion to gay youth support newsgroups to animal-rights organizations (there's an extensive archive on this topic at the Information about Labeling and Rating Systems section of the MIT Student Association for Freedom of Expression WWW pages).

Although many people were surprised at this revelation, it was in fact completely predictable from a historical perspective. Too much discussion of the future of unfettered electronic communications takes place in a social vacuum, from an extremely simplistic viewpoint (I refer to this the "net.libertarian" mindset). Because of a perspective that might be rendered "government action bad, private action good" There's great unwillingness to think about complicated social systems, of private parties acting as as agents of censorship.

Before going further, it's important to specifically define the meaning of censorship here. The definition I'm using in this article is roughly "The suppression of information due to government power". I think this is the most restrictive of common definitions, what I call the "Libertarian" definition. But note the following is NOT an aspect of this definition:
a) The exercise of government power must be crude, blatant, and direct - any subtlety, sophistication, or indirection makes it NOT CENSORSHIP.
b) The information suppression must be total, complete, and perfect - if any marginalized markets exist anywhere, then it's NOT CENSORSHIP.

These items are much too frequently tagged onto the term in some discussions (implicitly or explicitly). It sounds silly when put so directly, but the using term censorship is so fraught with argument over the definition it's often critical to be absolutely clear as to the meaning.

There are various ways to implement censorship (i.e., government-backed suppression of information), from the crude to the sophisticated. The Communications Decency Act is very crude. It's a bona-fide government criminal law with penalties of up to two years in jail. But even this very blatant censorship legislation, when studied in detail, has a level of indirection that is worthy of note. It doesn't state that saying anything which "depicts or describes, in terms patently offensive ... sexual or excretory activities or organs" (a very broad and vague prohibition) is itself criminal. The key provision is that making such material available to a minor is the crime. The trick is that making available to a minor can be argued to encompass all but the most restricted public audience, certainly any wide-scale distribution. So a public ban has been achieved in effect, but without actually stating it as such.

This subtlety is usually well-understood by anti-censorship advocates, but it is worth emphasizing in order to highlight the slight sophistication present. It's not a direct criminalization, but a prohibition under conditions (who can breezily defend giving "patently offensive ... sexual or excretory" material to children?) which effectively create a near-absolute ban.

Although it can be annoying to slog through endless Protect The Children rhetoric, this particular indirection usually doesn't give much trouble in censorship discussion, because what's at issue is the purest form of government action. But such application of criminal law is not the only possible method for achieving widespread suppression of information. Various blocking programs or labeling schemes may perform this function as well.

I call these various types of systems "privatized censorship". The administration is placed in private hands - a software company, a trade group, a ratings board - but, critically, they still draw power and authority from government threat. There's yet one more layer between the application of the power of the State and the resulting information suppression, but it's still ultimately a government-backed creation. In the real world, not an abstraction, these systems are deeply tied into fear of government action, and even jail terms. This was quite clear in the Philadelphia Communications Decency Act trial hearing:

[From the 5-10-96 transcripts, on]

[Lawyer arguing against the CDA]

JUDGE SLOVITER: Okay, you were going to your third, the warning. MR. ENNIS: The third is give warning. The Government suggested there are basically two ways of coming within this safe harbor defense and that is to tag your speech, to self label your speech in some way as inappropriate for minors or to register your speech with the Internet Yellow Pages or with one or ten or fifty of the 200 and so additional directories of Internet listings.
[Lawyer for the CDA]
JUDGE BUCKWALTER: The defenses are easy to comply with? MR. BARON: That the -- that the defenses, if the content providers who are putting a material that is sexually explicit within the definition of the CDA out into cyberspace in various forms, if they take just simple measures to ensure that the world knows that they're there, the Department of Justice is saying that that's substantial evidence of an affirmative defense. JUDGE BUCKWALTER: That's not what the act says though. My point is that's not what the act says, that's what the Department of Justice says. MR. BARON: Well, we have keyed our defense of this act to the broadly worded provisions of E(5)(A) as well as the conference report that talks about a content selection standards. And I asked Dr. Olsen whether his scheme was a content selection standard, it is. The self-labeling scheme that's embodied as part of the PICs methodology is a type of content selection scheme for content providers. Congress set it out, we're not departing or going further afield than what's in the conference report.
So in the real world, I think almost any business-owner who's presented with a statement from anyone in the Department of Justice that if they follow a labeling scheme or deal with some blocking program, they won't be risking years in prison, is going to immediately agree to use that system. That's hardly "voluntary". Note very little has been definitely established as an actual defense, so the various bandwagon-jumpers are not even getting a guarantee, they're just desperately hoping something will protect them from prosecution.

This connects to the politics of blocking and labeling, and how they got labeled themselves as being "against government" when that's actually utter nonsense. The private proponents say "Use this system, otherwise the government will pass laws to put you in jail" (this is how they are using the term "alternative" to government censorship). The government proponents say "We've already passed laws, so use this system as a way to avoid going to jail" (i.e., a "safe harbor defense"). And people have seriously argued to me that all this has nothing to do with government threat.

I've had tremendous trouble getting this point across in many discussions. It sometimes seems that if I leave off for even a paragraph saying "it's about government force, government threat, government coercion, the men with guns ...", someone is sure to say "But blocking programs and ratings are private action, and that can't be censorship.".

In the sort of net.libertarian viewpoint, people are conditioned to fear and loath government, but love and cherish business. But the image of "government" is the proverbial "jack-booted thugs", marauding SWAT teams, etc. If it's a case where bureaucrats in suits are dealing with managers in suits over issues of markets and liability, that tends to fall into the "can't ever criticize business as taking action against freedom" mental slot, and since the fear/love usually works as an exclusive distinction, this leads to the conclusion the government isn't there. It's a little like the infant-psychology experiments where if an object is out of sight, it doesn't exist. Except here it's worse, it doesn't exist even if the angle is different from the familiar one.

Hopefully, the above long argument establishes various systems which have private elements as at least worthy of thought even under the strictest view of the meaning of censorship. When the administration of censorship is privatized, it may lose a measure of effectiveness (though how much is unclear), but it's very evident that the reach is broadened. To return to the specifics of the blocking programs, finding out about their true contents, not just the public-relations hype, pointed up a nasty bait-and-switch.

Throughout the Communications Decency Act debates, its proponents have insisted that material of serious value would not be targeted. But the so-called "alternative" solutions, blocking software, often touted as making that law unnecessary, turn out to reach content of unquestioned merit and legality. Even a Memphis prosecutor couldn't bring charges against some of the targets of these programs. And here is the bait-and-switch: Talk about "porn" to get people to widely use whatever system, then once it is in place, in the hands of censors subject to no legal challenge, load on all sorts of prejudices and political agendas. Who would have thought CyberPatrol (one company) would designate the League for Programming Freedom, an organization that opposes software patents and user interface copyrights, as a "Militant/Extremist" group? Or that CyberSitter would decide "Sexual orientation [is about sex] by virtue of the fact that it has sex in the name.".

But in fact, this sort of swindle is standard. My favorite historical example is from the 1954 Comics Code. Implemented ostensibly against horror in comics, in an atmosphere of somewhat similar social hysteria (a story in itself), it contained provisions such as

3) Policemen, judges, government officials, and respected institutions shall never be presented in such a way as to create disrespect for established authority.
That's a horror of a different kind. So it was utterly predictable that generations later, censors of a similar mindset would be using the same old corruption-of-the-youth excuses to suppress feminist information, gay and lesbian resources, or just plain old "disrespect for established authority".

People concerned with freedom of electronic information should be extremely wary of elevating any unaccountable organization as a savior against government action. The government can be sued, can be held accountable, can even on occasion be restrained. When censorship is implemented by government threat in the background, but run by private parties, legal action is nearly impossible, accountability difficult, and this makes the reach of the system extensive. Blocking software is not a censorship solution, but another, different, censorship problem.