Freedom to Tinker

Today I’ll wrap up Vice Week here at Freedom to Tinker with an entry on porn labeling. On Monday I agreed with the conventional wisdom that online porn regulation is a mess. On Tuesday I wrote about what my wife and I do in our home to control underage access to inappropriate material. Today, I’ll suggest a public approach to online porn that might possibly do a little bit of good. And as Seth Finkelstein (a.k.a. Eeyore, a.k.a. The Voice of Experience) would probably say, a little bit of good is the best one can hope for on this issue. My approach is similar to one that Larry Lessig sketched in a recent piece in Wired.

My proposal is to implement a voluntary labeling scheme for Web content. It’s voluntary, because we can’t force overseas sites to comply, so we might as well just ask people politely to participate. Labeling schemes tend not to be adopted if the labels are complicated, or if the scheme requires all sites to be labeled. So I’ll propose the simplest possible labels, in a scheme where the vast majority of sites need no labels at all.

The idea is to create a label, which I’ll call “adultsonly” (Lessig calls it “porn” but I think that’s imprecise). Putting the adultsonly tag on a page indicates that the publisher requests that the page be shown only to adults. And that’s all it means. There’s no official rule about when material should be labeled, and no spectrum of labels. It’s just the publisher’s judgment as to whether the material should be shown to kids. You could label an entire page by adding to it an adultsonly meta-tag; or you could label a portion of a page by surrounding it with “adultsonly” and “/adultsonly” tags. This would be easy to implement, and it would be backward compatible since browsers ignore tags that they don’t understand. Browsers could include a kids-mode that would hide all adultsonly material.

But where, you ask, is the incentive for web site publishers to label their racy material as adultsonly? The answer is that we create that incentive by decreeing that although material published on the open Internet is normally deemed as having been made available to kids, any material labeled as adultsonly will be deemed as having been made available only to adults. So by labeling its content, a publisher can ensure that the content’s First Amendment status is determined by the standard obscenity-for-adults test, rather than the less permissive obscenity-for-kids test. (I’m assuming that such tests will exist and their nature will be determined by immovable politico-legal forces.)

This is a labeling scheme that even a strict libertarian might be able to love. It’s simple and strictly voluntary, and it doesn’t put the government in the business of establishing fancy taxonomies of harmful content (beyond the basic test for obscenity, which is in practice unchangeable anyway). It’s more permissive of speech than the current system, at least if that speech is labeled. This is, I think, the least objectionable content labeling system possible.

It’s day two of porn week here at Freedom to Tinker, and time to talk about the tools parents have to limit what their kids see. As a parent, I have not only an opinion, but also an actual household policy (set jointly with my wife, of course) on this topic.

Like most parents, we want to limit what our kid sees. The reason is not so much that there are things we want our kid never to see, but more that we don’t think our kid is ready, yet, to see and hear absolutely everything in the world. Even the Cookie Monster is scary to kids at a certain age. Good parents know what their kids can handle alone, and what their kids can handle with a trusted adult present. We want to expose our kid to certain things gradually. Some things should be seen for the first time with a parent present to talk about what is being depicted.

But how can we do this, in the real world? It’s not enough simply to say that we should supervise our kid. To watch a kid nonstop, 24/7, is not only impractical but creepy. We don’t want to turn our home into a surveillance state.

Instead, we rely on architecture. For example, we put the only kid-accessible computer and TV in the busiest room of the house so that we’re less likely to lose track of what’s happening. But even that isn’t foolproof – it doesn’t work in the early morning hours when kids tend to be up while parents sleep.

This is where filtering technology can help. We find the TV rating and filtering system quite useful, despite its obvious flaws. This system is often called the V-chip, but we don’t actually rely on the V-chip itself. Instead, we rely on our Tivo to allow restrict access to shows with certain ratings, unless a secret password has been entered. We know that the technology overblocks and underblocks. But overall, we prefer a policy of “watch any kid-rated show you want, but ask a parent if you want to watch anything else” to the alternatives of “watch anything you want” or “always ask a parent first”. (A welcome side-effect: by changing the rating threshold we can easily implement a “no TV today” policy.)

It’s worth noting that we don’t use the federally mandated V-chip, which is built into our TV. We simply use the ratings associated with shows, and the parental controls that Tivo included voluntarily in its product. For us, the federal V-chip regulation provided, at most, the benefit of speeding standardization of the rating system. We’re happy with a semi-accurate, voluntary system that saves us time but doesn’t try to override our own judgment.

Adam Thierer at Technology Liberation Front offers a long and interesting discussion of the online porn wars, in the form of a review of two articles by Jeffrey Rosen and Larry Lessig. I’ve been meaning to write about online porn regulation for a while, and Thierer’s post seems like a good excuse to address that topic now.

Recent years have seen a series of laws, such as the Communications Decency Act (CDA) and the Child Online Protection Act (COPA), aimed at restricting access to porn by minors, that have been the subject of several important court decisions. These cases have driven a blip in interest, and commentary, on online porn regulation.

The argument of Rosen’s article is captured in its title: “The End of Obscenity.”
Rosen argues that it’s only a matter of time before the very notion of obscenity – a word which here means “porn too icky to receive First Amendment protection” – is abandoned. Rosen makes a two-part argument for this proposition. First, he argues that the Miller test – the obscenity-detection rule decreed by the Supreme Court in the 1970’s – is no longer tenable. Second, he argues that porn is becoming socially acceptable. Neither claim is as strong as Rosen claims.

The Miller test says that material is obscene if it meets all three of these criteria: (1) the average person, applying contemporary community standards, would find it is designed to appeal to the prurient interest; (2) it depicts [icky sexual stuff]; and (3) taken as a whole, it lacks serious literary, artistic, scientific, or political value.

Rosen argues that the “community standards” language, which was originally intended to account for differences in standards between, say, Las Vegas and Provo, no longer makes sense now that the Internet makes the porn market international. How is an online porn purveyor to know whether he is violating community standards somewhere? The result, Rosen argues, must be that the most censorious community in the U.S. will impose its standards on everybody else.

The implication of Rosen’s argument is that, for the purposes of porn distribution, the whole Internet, or indeed the whole nation, is essentially a single community. Applying the standards of the national community would seem to solve this problem – and the rest of Rosen’s essay supports the notion that national standards are converging anyway.

The other problem with the Miller standard is that it’s hopelessly vague. This seems unavoidable with any standard that divides obscene from non-obscene material. As long as there is a legal and political consensus for drawing such a line, it will be drawn somewhere; so at best we might replace the Miller line with a slightly clearer one.

Which brings us to the second, and more provocative, part of Rosen’s essay, in which he argues that community standards are shifting to make porn acceptable, so that the very notion of obscenity is becoming a dinosaur. There is something to this argument – the market for online porn does seem to be growing – but I think Rosen goes too far. It’s one thing to say that Americans spend $10 billion annually on online porn, but it’s another thing entirely to say that a consensus is developing that all porn should be legal. For one thing, I would guess that the vast majority of that $10 billion is spent on material that is allowed under the Miller test, and the use of already-legal material does not in itself indicate a consensus for legalizing more material.

But the biggest flaw in Rosen’s argument is that the laws at issue in this debate, such as the CDA and COPA, are about restricting access to porn by children. And there’s just no way that the porn-tolerant consensus that Rosen predicts will extend to giving kids uncontrolled access to porn.

It looks like we’re stuck with more of less the current situation – limits on porn access by kids, implemented by ugly, messy law and/or technology – for the foreseeable future. What, if anything, can we do to mitigate this mess? I’ll address that question, and the Lessig essay, later in the week.