When I blogged up what I thought was a fairly harmless and charming little remark from Steve Landsburg earlier this week, at least one of ErosBlog’s readers got at least a little bit perturbed. How dare I, or perhaps more accurately how dare Professor Landsburg, suggest that it’s okay for a minor to go surfing freely around the Internet, wherein she might encounter not just sexually explicit images, but {shudder} evil nasty porn?

A fair enough question, and one worthy of another one of my Sunday sermons.

Whenever someone gets up to defend the application of J.S. Mill’s Harm Principle — the view that the coercion of others can only be justified by self-defense and that we’re not allowed to force others to do or refrain from doing things for their own good — someone else will invariably hop up and ask, in plangent tones, “won’t anyone please think about the children?” (“Who will save the wee bairns?”)

What they mean by this is that if we allow adults full liberty to enjoy certain things like erotic materials, then Poor Innocent Children will also get exposed to them, and that this will be some sort of Very Bad Thing. So maybe we need to restrict the liberty of adults, at least to the extent of building high walls (real or virtual) around the Very Bad (if Poor Innocent Children will see it) Thing.

Hmm. Sounds like the putative well-being of children provides a very handy pretext for action by those who are not so much concerned to protect the well-being of children, as to extinguish the liberty of adults. And I don’t doubt that it is used exactly as such a pretext, much of the time. But let’s grant that those who would defend children — the Knights of the Wee Bairns, shall we call them? — are, in this instance, acting in good faith. Is there a harm here that merits our attention? Is Professor Landsburg’s daughter in some terrible danger from the Internet?

The Knights of the Wee Bairns, at the very least, want paywalls and adult filtering around “bad” content; some of the more maximalist among them want this content to disappear entirely, of course, and not just from the Internet.

People who fret about the danger that Internet porn supposedly represents to children most likely fear that free access to it will endanger their own ability to transmit their values and worldview to their children. This possibility is the “harm” that they fear. Perhaps their fears aren’t entirely unfounded. Maybe something children see on the Internet will affect their values or worldview in ways their parents won’t like. Too bad. In a free society, children are not robots to be programmed by their parents. You’re not entitled to demand that anyone else build a wall around anything just because of the worldview you want to transmit to your children.

I’ll sharpen this claim with an example, pointed right at myself. I am a religious nonbeliever. I am robust in my non-belief, and apologize for it to no one. It would be grounds for substantial disappointment, no, actually it would be grounds for considerable soul-searching and condign self-reproach, if either of my own daughters were to end up as some sort of evangelical Christian, because for my money that means they ended up believing something that is almost certainly false and probably pernicious as well. But I do not, not for one minute, think that this means that someone who runs a website devoted to Christian apologetics should be required to put up an age-determining wall, or require a valid credit card before visiting his site, or put some sort of objectionable content flag in his HTML code so that Atheist Net Nanny can filter out his content, with its explicit promotion of a worldview and values I’d rather my own children not end up subscribing to.

If we must censor content based on its potential to change some child’s values or worldview to something her parents won’t like, then there will be precious little liberty for anyone.

“Oh, but that’s not what we mean,” say the Knights of the Wee Bairns. “We’re only against porn, which is bad. We don’t mean to stifle vigorous debate in a free society. We don’t want to extinguish the liberty of adults. We think that children will be harmed if they see certain images or read certain stories. We insist only that there be some walls across the Internet, so the kids can’t get access to this sort of material.”

I would begin by noting that there really is no such thing a harmless wall across the Internet. The wall will never be truly voluntary. It will invariably be enforced by legal and social sanctions, which means that some content will disappear, simply because it isn’t worth the trouble or the risk of the provider to deal with the compliance burden. Other content, which should be available to all, will end up behind such a wall. To site just one prominent example, it’s been clear for years that Net Nanny and the censors keep material having to do with sexual health away from the teenagers who could really benefit from it.

So there’s a real cost to building walls across the Internet. Is there a corresponding benefit, in the form of harms avoided to minors who happen to view porn?


There are many reasons why I think there is a strong prima facie case why there is no such benefit, no avoided harm to minors. Among the two strongest are the relative resilience of minors and the fact that things on the Internet don’t really change the incentives associated with real-world behavior.

Does viewing stuff — even nasty stuff — make children into nasty adults? Where is the evidence? Let’s look at some history. About two generations ago there was, you might recall, a huge conflict called the Second World War. Children living in the United States experienced this conflict largely as a series of deprivations. But tens of millions of children living the vast zones between the North Sea and the Volga, between Hokkaido and Java, experienced the war as terror — bombings and shellings and occupations and persecutions. Nothing anywhere on the Internet is as obscene as what happens in the real world when a war sweeps through. And what became of this generation of children? Some were psychically scarred for life, sadly. But most of those tens of millions grew up in the postwar world to have pretty normal lives, no more prone to criminality or dysfunction than human beings generally. Children can be, and are, pretty resilient. And keep in mind that Internet surfing, unlike having your country bombed or invaded, is voluntary. Anyone, children included, who encounters a distressing image can always just surf away.

Meanwhile, the real world continues to impose its set of costs and benefits, and these will be more powerful shapers of human behavior than what children or teenagers see on the internet. Out in meatspace, aggressive sexual conduct toward non-consenting others can lose you your job, ruin your reputation, and even land you in jail. Poorly-timed pregnancies can derail your life. HIV infection can kill you. None of these hard facts change, no matter what you’ve seen on the Internet, and people, including underage people, know this. Costs are real. People’s behavior is highly sensitive to cost. Homo economicus might be a crude approximation of actual human beings, but he’s a hell of a lot better than the “monkey see, monkey do” psychology that seemingly can be attributed to many fear mongers about porn.

“But you still might be wrong!” cry the Knights of the Wee Bairns.

Sure. Anything might be wrong. But if the mere theoretical possibility of harm is enough to forbid someone from doing something, then there really will be no liberty for anyone. That much should be too obvious even to require exposition.

All the same, I have no desire to be a dogmatist or an armchair theorist. I’ll respect anyone who civilly disagrees with me. And I’ll go one better than that. I’ll even agree to change my mind, provided that someone can honestly meet the following challenge:

Begin by defining a Bad Life Outcome as something that is uncontroversially bad to happen to someone. I mean bad in a thin sense. It has to be an outcome that pretty much everyone would agree would be bad. In defining bad you don’t get to cheat and load into the concept of “bad” something distinctive about your own worldview. I know that if you’re a die-hard Democrat you might think turning into a Republican or if you’re a Christian you might think becoming an atheist is a bad life outcome, but these don’t count: these outcomes are only bad relative to your specific worldview. Spending your life in prison, or dying of some terrible disease at 25, are Bad Life Outcomes as defined here.

Now suppose further a hypothetical experiment. Take 200,000 nine year-olds. Assign 100,000 of them at random to a Control Group and 100,000 to a Test Group. The Control Group has to spend their time on the Internet until age 18 with Net Nanny filtering out most (surely not all) of the bad old porn on the Internet (along with quite a lot of other stuff, probably). The Test Group gets uncensored access to the Internet until the age 18.

If you can tell me what the number of Bad Life Outcomes will be in the Control Group and the Test Group, and give me a convincing explanation as to why a skeptical and competent social scientist — someone like, oh, Steve Landsburg, say — should credit your numbers; oh, and furthermore, if the rate of Bad Life Outcomes in the Test Group really is materially larger than in the Control Group, then I will cheerfully change my mind.