Landsburg v Bork: What Counts as Injury?

My friend Steve Landsburg appears to have entered a competition with my friend John Lott over who can make the most politically incorrect argument. John's old entry was an article arguing that the fact that rich criminals were less likely to be convicted than poor criminals was evidence for, not against, the efficiency of the legal system (my discussion of that article is in chapter 15 of my Law's Order, under the subhead "Should the Rich Pay Higher Fines"). Steve's new one, which has gotten a good deal more attention, is a blog post asking why rape of an unconscious victim who suffers no injury in the ordinary sense should be illegal.

Reading attacks on Steve's piece, it occurred to me that I had seen the essential point before—from the other side, in an old article by Robert Bork which explained, among other things, why he was not a libertarian. It is a long time since I read it, and when I recently reread it I discovered that it was not mainly about the part I remembered, but that particular argument was what impressed me, since I am both a libertarian and an economist and the argument appeared to show that the two were inconsistent with each other.

Bork's argument, in my words not his, goes as follows:

When I pollute the air I am injuring other people, so it is legitimate for the legal system to respond by penalizing me. What makes it an injury is not the fact that I affect the air but that the effect does harm; one could imagine an effect, such as a change in the ratio of two stable isotopes of trace gases in the atmosphere, that would not matter to anyone and so would not be seen as an injury or a proper subject for legal action.

Harm, however, is ultimately subjective, since it depends on the preferences of the harmed individual.  When I smell the roasting ribs from my neighbor's barbecue, that isn't harm because I enjoy the smell. When the smoke from the barbecue makes me cough that is harm, because I don't like coughing—and would be harm even if the smoke had no adverse effect on my health. From the point of view of economics, "harm X" simply means "lower X's utility."

Suppose that, instead of polluting my neighbor's air, I engage in behavior that he disapproves of—read pornography, use contraceptives, work on Sunday. That too causes him disutility. Since the defining characteristic of harming someone is lowering his utility, I am harming him. Since I am harming him, my activity is just as much a legitimate target for legal action as my polluting his air would be. Hence the libertarian principle that I have a right to engage in what Mill referred to as self-regarding actions, actions that only affect me, is either false or empty. Either I don't have a right to read porn if doing so offends others without affecting them, or their offense counts as an effect of what I am doing so my reading porn isn't really a self-regarding action and there is no reason in principle why it shouldn't be banned.

When I first read it, it struck me as an interesting and persuasive argument for a conclusion I disagreed with—explaining why I thought it was ultimately wrong would take a different, and longer, post than this. 

Steve Landsburg's piece, responding in part to the Steubenville rape case, makes the same argument from the other side. We—at least Steve (and I)—don't feel that the argument for banning pornography or contraception is a legitimate one. Our reason is that the "harm" in those cases is purely subjective—I haven't actually done anything to you, so your unhappiness at my self-regarding behavior is your problem, not mine, and you have no right to use the legal system to make me conform to your wishes. And even if you argue that I have done something to you—acted in a way that resulted in your knowing what I was doing, knowledge that pained you—that doesn't count, because "knowledge that pains you" isn't injury in the same sense as causing you to get cancer is.

Which gets us to the part of Steve's post that gives lots of people reason, or excuse, to attack him. Suppose an unconscious woman is raped in a way that results in no injury—in the Steubenville case, "rape" actually consisted of digital penetration. She only finds out it happened several days later, at which point the harm is purely subjective, consists of her being offended at the knowledge that it happened. Why is this different from the subjective harm suffered by the person offended at someone else reading pornography? It feels different—to me and obviously, from his post, to Steve. But is it different, and if so why?

That, it seems to me, is an interesting question, one relevant to both law and  morality. It is ultimately the same question raised by Bork, although from the other side. Bork was arguing that the harm caused by the use of contraception and the harm caused by air pollution were ultimately of the same sort, that it was legitimate to ban pollution hence legitimate to ban contraception—his article was in part an attack on Griswold v. Connecticut, the Supreme Court case that legalized contraception, a fact I had forgotten when I started writing this post. Landsburg is arguing that rape that does only subjective harm is of the same sort as reading pornography that does only subjective harm (unlike Bork, it isn't clear that he is thinks his argument is right, only that he thinks it interesting), that it is not legitimate to ban the reading of pornography hence not legitimate to ban that particular sort of rape. 

I agree with both Bork and Landsburg that there is a real puzzle in our response to the legal (and moral) issues they raise. Hence I disagree with the various commenters whose response to the Landsburg piece was that it showed he was crazy, evil, or both.