Friday, April 05, 2013

Response to Bork and Landsburg

In a recent post, I argued that Robert Bork, in an old and famous law review article, and Steven Landsburg, in a recent and now infamous blog post, were making two sides of the same point, in both cases arguing that our legal rules (and, for many of us, our moral intuitions) were inconsistent in treating essentially similar cases differently. Various commenters on my post pointed in what I think is the right direction for explaining the puzzle and justifying the legal, and perhaps the moral, difference.

The first step is an old distinction in the economic analysis of law between property rules and liability rules, two different solutions to the problem of efficient allocation. Under a property rule, something belongs to someone and other people can use it only with his permission. The role of the legal system is to impose costs on those who use it without his permission large enough so that they won't. My control over my car is regulated by a property rule, as demonstrated by what happens to someone who steals it and gets caught.

Under a liability rule, in contrast, people are permitted to do things that impose costs on others, but have to pay for them. My use of my car to impose a risk of collision on other drivers is regulated by a liability rule. Unlike the case of a property rule, I don't need their permission to use their cars as possible unintended targets for mine. But if I do dent their cars and it is my fault, I am supposed to pay for the repairs. The role of the legal system is to estimate the cost and make me pay it.

Why do we use property rights to control who drives my car but liability rights to control how it is driven? The standard response is that, under property rights, things are moved to their highest valued use via market transactions—if ownership or use of my car is worth more to you than to me, you ought to be able to offer me a price to buy or rent it that I am willing to accept. Under liability rules, they are moved to their highest valued use via litigation, actual or potential. 

Controlling who drives a car by market transactions is reasonably simple and straightforward, as demonstrated every day by car dealers and rental agencies. Controlling how it is driven would either require me to buy permission from everyone else on the road to impose a cost of damage on their car, if they had a property right not to have such a risk imposed on them, or require them to buy a guarantee from me to drive carefully, if I had a property right to drive my car as I wished, whatever the risk to others. Neither is practical, since what I am imposing is a highly dispersed cost, a very small risk on each of a very large number of people.

Controlling who drives a car by a liability rule would mean that someone in a hurry to get to the airport would be entitled to hotwire my parked car, drive it to the airport, and leave it there—and it would be up to me to sue him for the cost and inconvenience imposed on me. That would create substantial costs, since I would have to identify the guilty party and sue him, and provide a very inaccurate and expensive control mechanism, since a court would have to somehow estimate the cost to me. Some of those problems exist for using a liability rule to control how I drive, but they arise only when I actually run into someone, which (fortunately) does not happen very often, and it is usually obvious at that point who was the driver. So it makes sense to use a property rule to control who drives the car but a liability rule to control how he drives it.

(For a more detailed account, see Chapter 5 of my Law's Order, under the subhead "Property or Liability Rules.")

The discussion so far leaves open one question of some importance: If we use a property rule, who owns the property? From the standpoint of economics, there are two different approaches to answering it.

One is that property should start out belonging to the person who created it, in order that people will have a suitable incentive to create things. The other is that property should belong to the person who can most easily control it, in order to minimize the cost of enforcing property rights.

We use the first rule for deciding who owns my car. It started out as the property of the Honda corporation, which built it, was sold by them to a dealer and by the dealer to me. We use the second approach for deciding who owns me.

One could hold—as Roman law did—that I belong to my parents (actually, under Roman law, my father—or his father if still alive), on the grounds that they created me. The problem with that approach, economically speaking, is that it is easy for me to control myself (at least as long as there are no bowls of potato chips within easy reach), hard for anyone else to control me. If I want my arm to move, I move it. For you to get my arm to move, supposing you are my owner, is a more difficult, indirect and costly process. That, economically speaking, is the argument against slavery.

Sometimes the two approaches are in conflict. Consider the case of copyright law. Giving me the right to control people who want to copy my book and charge them for doing so gives me a suitable incentive to write books. But controlling their copying is much easier for them to do than for me to do. How much easier depends on technology, which is why we have moved, over about a century, from a situation where copyright could be enforced without copyright law, as in the case of English authors collecting royalties for books published in America a century or so ago, to one where it is becoming increasingly difficult to enforce copyright in digital works even with copyright law.

How does all of this apply to the puzzle raised by Landsburg and Bork? Being more prudent than Landsburg, or perhaps less courageous, I start by replacing his example with a less emotionally loaded one that I think raises the same issues.

You happen to know that I am going to be out of town for a few months, perhaps because I have mentioned my plans on my blog. Coincidentally, you are going to be in town for the same few months, in need of a place to stay. Conveniently, one of your hobbies is picking locks. I leave, you move in. Shortly before I return you, being a considerate guest, tidy the place up, leaving it in exactly the same condition you found it in and so, arguably, imposing no cost on me. 

You only make one small mistake—not noticing the video camera unobtrusively located on a high shelf in the living room, plugged in, running, and feeding a very large hard drive. 

Just as in Landburg's example, the act violates existing rights but imposes no material cost. As in that example, it might impose a psychic cost—I may be upset to find that someone else has been living in my house without my permission. And, as in that example, it is illegal. Should it be?

It should. The reason it should is that the use of houses is better allocated by a property rule than by a liability rule. The liability rule, under which you are not liable because you imposed no cost, or at least no cost of a sort reasonably measurable by a court, is difficult to enforce, given that not all empty houses have security cameras running and not all trespassers are easily identified, located and sued. Further, the cost to me is better measured by the price I am willing to accept for the use of the house than by a court's estimate, in part because that includes the subjective cost to me of knowing that my house has been occupied and any secrets it contained, perhaps my stash of porn, revealed, a cost hard to prove or measure in court.

A further argument, pointed out by commenters on my earlier post, is that a property rule makes it possible for me to charge you for your use of my house, which in turn affects my incentive to make the house available—my earlier point about allocating property to its creator. The ability to collect rent on the house when I am not using it reduces the cost to me of spending a quarter visiting at GMU and so makes me more willing to do so—and should, since the availability of the house for you to occupy is a real benefit produced by my absence.

I think I have now answered Steve Landsburg's puzzle. The difference between his example (or mine) of an action that imposes only subjective costs and his example of an activity such as reading pornography, or Bork's of using contraception, that imposes only subjective costs, is not the nature of the harm. The difference is that in the one case the cost is of a sort that can be measured, the action controlled, via a property rule. In the other, it is not. 

More precisely, the property rule under which I have a right to read porn and you can only stop me by offering to pay me not to do so produces its result by ignoring the cost my porn reading imposes on you, since, as with the case of risks imposed by careless driving, including that cost requires an unworkable contract between all of the prudes and all of the would-be consumers of porn. The property rule under which you have a right to forbid me, or anyone else, from reading porn, produces its result by ignoring the cost your ban imposes on me, for the same reason. Neither property rule gets the cost/benefit calculation correct, but the former rule is a great deal less expensive to enforce than the latter, which is an argument for it. 

What about a liability rule? That is the point at which the subjective nature of the harm comes in. It is true that, from the standpoint of economics, all harm is ultimately subjective—having my arm broken or my car dented would not be a cost under sufficiently bizarre assumptions about my preferences. But some subjective costs are a lot easier to measure externally than others. When I claim damages for my wrecked car, there are market prices out there for repairing or replacing it that provide a court with a reasonable basis for estimating the cost. When I announce that your reading of porn, or oil drilling in a wilderness I never plan to visit, inflicts large psychic harm on me, there is no such basis for checking my claim.

I think I have now answered Bork's argument as well as Landsburg's. Insofar as air pollution imposes measurable costs in ill health or increased laundry bills, it is possible to use a liability rule or something similar to measure those costs and charge the polluter for them. But if my use of contraception imposes only immeasurable psychic costs on you, that approach does not work. A property rule takes account of such costs, since they are reflected in the prices people are willing to pay or accept, but a property rule assigning the right to either the would-be user of contraception or the opponent of his use does not work to reflect those costs because of their dispersed nature. And, between the two alternative property rules, there is a strong presumption for the one that libertarians prefer, since it requires no expensive and intrusive apparatus to enforce it.

Bork, unfortunately, is no longer around to read my answer, but Landsburg is, and probably will.

In my earlier post, I commented that explaining why I thought Bork was wrong would require a different and longer post. I have now written it, and it did—about twice as long.


David Friedman said...

Several of my previous posts got very large numbers of comments. Max put up the only comment this had gotten after about 24 hours, and it wasn't really a comment.

Which is interesting. This post is, I think, the most important and (to me) interesting of the recent series. But it is also the hardest, sketching what I think is an important point in the economic theory of the law.

Max said...


That's what I was thinking; I sought to say the same thing with a single word, its superficial simplicity intended to act as a meta-commentary on what I perceive to be the shallow analysis of every post I've seen on the subject except for this one.

Sorry if that didn't come across as clearly as I'd hoped. =X

123 said...

David, this post is so perfect, comments can only spoil it, and for the last 24 hours I was happy there were no comments. I was disappointed a bit when I saw the comment count, but I'm happy again because all the comments are metacomments.

Anonymous said...

If you are wondering whether this post was uninteresting or too narrow in scope, I would respond in the negative. Your argument was well formed and accessible, but I had no prior views on the issue- and therefore I wasn't impelled to critique an argument that seemed credible anyway.

Max said...

I came back to this post with the intention of saying no more than this:

"Perhaps people aren't commenting because there's absolutely nothing that you've left unsaid. All who read these words agree with every one of them, and upon realizing that the problem has been solved, they lose the will to continue discussing a puzzle that once seemed very interesting, but now seems so obvious as to not merit further discussion."

Imagine my glee at finding that comment #4 pretty much said exactly what I'd been planning to in addition to mirroring other thoughts I'd had but did not plan to share.

(My decision to break the comment barrier was born of a fear that someone might sully this sacred beauty with something unworthy of its splendor; the purity of a single word with multiple embedded meanings buried beneath the simplest possible exterior was too much to pass up.)

Patrick said...


Like others, I refrained from posting mostly since your post closely represents my own views about this. Of course, my views on the matter have been largely formed from reading your prior writings (blog and printed), so this isn't exactly surprising for me. I was happy, though, that some of my comments in your earlier post seemed to have affected some of what you wrote in this one.

I tend to post to disagree with a commenter or answer an open question. Since your post had nothing I disagreed with, and no one voiced an opposition, I found I had little to say.

Unknown said...

I'll be the student unafraid to raise his hand and risk revealing his ignorance.

"When I announce that your reading of porn, or oil drilling in a wilderness I never plan to visit, inflicts large psychic harm on me, there is no such basis for checking my claim."

Suppose neuroscience and brain-scanning technology made it cheap and easy to objectively measure psychic harm. Would that make a liability rule more attractive? If so, would it be more attractive than a property rule, since "[n]either property rule gets the cost/benefit calculation correct"?

Unknown said...

One more thought, which occurred to me when reading Landsburg's OP, is that a private market in law solves the apparent contradictions on a practical level. In MoF, Ch. 31, you pointed out that people oppressed by "compulsory puritanism" "are willing to pay a much higher price to be left alone than anyone is willing to pay to push them around."

Thus anarcho-capitalism solves the problem. Prude probably won't be willing to pay what it costs to take away my porn. On the other hand, I'm willing to pay a relatively high price to deter someone from raping me.

In some ways that answers a different question, but in some ways it's the whole point, no?

Guest 1b said...

I'll be the second student to raise their hand.

What if the parameters around the porn example are changed? For example, what if the porn being viewed:
- Is in some way extreme, for example excessively violent.
- Was created using actors who were not participating voluntarily.
- Is animated and depicts acts that would otherwise be illegal (e.g. involving minors).

Secondly, some psychologists contend excessive viewing of porn is causing dramatic shifts in body image and perception of normal human sexuality in young adults. This affects their capacity to form healthy relationships. If large scale research proves this to be true, does it change the argument for allowing porn?

Patrick said...


My immediate reaction to your neurotech solution is that in such a world, the standard property solution works, too. After all, if we get machines that can read minds cheaply and accurately, wouldn't that solve bargaining problems?

I could imagine that the standard way of coming to a bargain would be for both sides to submit to neural scanning to determine the real range of acceptable bargains, and agree in advance to stick the price in the middle. Either side could opt out to try and get a larger share of the pie, but doing so would revert to costly bargaining and the possibility of a breakdown in negotiations. A guaranteed half of a large pie is worth more, I think, than an uncertain division of a smaller pie (even if you're such a good bargainer that you're sure you'll get the larger share).

Alternatively, if the same technology can be used to trick the machine, then we couldn't trust it to quantify the psychic harms for the purpose of a liability rule, so we're back to square one.

Patrick said...

Guest 1b,

In response to each of your points generally, if the porn is so horrible that it obviously shouldn't be viewed, then either people will voluntarily not view it or it will be worth it for others to pay them to stop. In response to your scenarios individually:

- One man's extreme is another's tame. We should no more prohibit extreme porn than we should prohibit extreme political opinions (German law notwithstanding).

- This kind of porn is illegal to produce, obviously. The criminal here would be the producer, distributor, etc. Basically, anyone who had control over the decision to use force rather than just pay actors to make an accurate simulacrum.

- Under that rule, your post should be illegal, too. After all, it depicts (represents in words) illegal porn. While there are at least tenuous reasons that can be given for criminalizing child porn, I can see no reason to support outlawing artistic depictions of the same. If anything, animated porn is a substitute good for real child porn, so its existence should be seen as a good thing if child porn is bad.

David Friedman said...

Noah: A good way of measuring such costs would make the liability rule more attractive than the property rule, assuming a sufficiently reliable court system to enforce it. Property rules come much closer to being self-enforceable—indeed, in some societies are mostly self-enforced—which is an additional argument for them.

Guest 1b: I don’t think any of your first set of assumptions change the situation, although if the actors were not voluntary the making of the porn should have been treated as a crime. I don’t see any good reason why a film that appears to show sex with minors but was made without actually using any minors—young looking actors or computer generated images—should be illegal.

Your second issue seems to me to be analogous to the question of whether people ought to be allowed to overeat or to use drugs, both of which might have negative effects on them. From both a moral and consequentialist standpoint, it seems to me that we are better off leaving those decisions to the people in question rather than making them for them. I agree that, as with Noah’s hypothetical, the closer an effect comes to being objectively measurable from the outside the more practical it becomes to use liability or something similar to control it. But you still have the obvious problem of A making a decision for B, where A not only lacks relevant information that B has but lacks B’s incentive to decide on the basis of B’s welfare.

Guest 1b said...

I posed the the three scenarios because all are already illegal where I live.

Porn is classified (R&X for films, R1&R2 for literature). Films and literature depicting extreme violence are refused classification (RC) and basically illegal to possess.

All porn involving and depicting minors is illegal. Porn where actors have not consented includes 'sexting' by minors, since minors by law cannot give consent. I deliberately used the term minors because depictions of individuals who may or may not be just below the age of consent is illegal.

At some point, our elected representatives have created these laws. With the exception of 'sexting', there appears to be no political movement, or public demand, to change any of these laws.

I take your point about A making decisions about B. Two questions then. At what point does society's collective knowledge or majority view outweigh B's personal information (which may not be fully informed) and incentive? What happens if B's behavior rises to the level of addiction (to drugs, porn or food), and B is no longer able to make rational, informed decisions?

David Friedman said...

For Guest 1b:

Bork's argument re Griswold was that the elected representatives of the people of Connecticut had made a law against contraception, and that the Supreme Court did not have a good argument for finding it unconstitutional. My claim is that they did--more precisely, that Bork's argument for why they didn't is wrong (there could be other good arguments for the conclusion).

With regard to A and B, part of the problem is that it is "society," i.e. the political and legal system, that is deciding when its expertise outweigh's B's. For the same reasons I don't trust it to make the decision for B, I also don't trust it to decide whether this is one of the special cases where it can be trusted to make the decision because B's decision for himself will be even worse.

I agree that such special cases exist, but not that the political system can be trusted to identify them.

A similar problem arises with regard to protecting children. I see no moral argument against taking children away from a sufficiently bad parent, but I see a strong practical argument against giving the state the power to do so--because I don't trust the state to decide when it will do a better job than the parent. The most dramatic recent example I know of that is the Texas FLDS case, which I blogged about at length some years ago, where the Texas child protection people took some three hundred children away from their parents because they disapproved of the parents' religion, lied about the facts of the case (to a generally credulous media) for months, and only surrendered the children back to their parents after both the state appeals court and the state supreme court had ruled unanimously that they had no grounds to take them. I think treatment of First Nation children in Canada some time back provides a still clearer case, and one in which the legal system never forced the state to stop what it was doing.

Anonymous said...


I know there has been a lot of praise for your considerable post, so please indulge an old student of yours some observations/questions.

1) By admittedly being less courageous than Landsburg, changing the subject from rape to real estate, you’ve removed much of what is both interesting and problematic about the question in the first place. There is always a market for house rental, but by definition, there is no market for rape, and there never can be.

2) Landsburg’s original is a very confusing and poorly written post. It’s hard to figure out exactly what he meant. He wrote: “As long as I’m safely unconscious and therefore shielded from the costs of an assault, why shouldn’t the rest of the world (or more specifically my attackers) be allowed to reap the benefits?” I assumed this meant that the victim was always shielded from the costs of assault and NEVER found out, via any means, including the internet. If my assumption is correct, you need to remove the camera evidence from your analysis. The owners of the home never find out. Is it still a problem that someone entered? This is the more interesting question anyway. If your assumption is correct, and you think that Landsburg thinks that all the torment that girl is receiving, including the death threats, is outweighed by the “benefits” the attackers gained, then you may want to rethink the utilitarian calculus and consider finding some new friends.

3) Finally, by drawing an analogy between a woman’s body and a house, you are basically ascribing full property rights to the human body. Despite the appeal to the notion that we own ourselves, the full implications of it are not all that easy to embrace. Can I sell a kidney? Can I sell a kidney, not to save someone, but to an artist who wants to make art out of body parts? Can I then sell a second kidney, even though it will kill me? Can I sell myself into slavery to you in order to send my kid to college, and then can you in turn sell me to somebody else? Can I put an ad in the paper seeking someone willing to be killed and eaten (actual case: see “Cannibal of Rotenburg” – By the time he was arrested, our cannibal had consumed over forty pounds of his consenting victim, cooking some of him in olive oil and garlic)? If the libertarian claim is right, banning consensual cannibalism is unjust, a violation of the right to liberty. Is a human body really equivalent to a house?


Max said...


1) You don't know much about human trafficking, huh? There's a sense in which that's fortunate, I suppose. Suffice it to say that there is indeed a market for rape, with supply and demand playing their customary roles in determining price. You can find out more about this market if you'd like, but I suggest that you don't.

2) I disagree that Landsburg's post is necessarily confusing or poorly written. I think it IS confusing to many people who are not especially good at reading dispassionately.

You assume that the psychic distress of the victim upon finding out later obviously counts as a "cost" of the assault. Do you also assume that the psychic distress of Farnsworth McCrankypants counts as a "cost" of allowing people to look at pornography? Do you assume that the psychic distress of would-be rapists at not being allowed to rape counts as a "cost" of banning rape? If your answer is no - what distinguishes the first situation from the second and third?

If the owners of the home in David's example never find out, then no, it is not a problem that someone entered. Same goes when we use Landsburg's original example of a harmless rape that no one finds out about. You think otherwise because you are not actually imagining a trespass/rape where no one finds out and no harm is done - you are imagining an instance that you know about and then heavily weighing the disgust you feel into your moral calculation of whether a wrong has occurred. But we stipulated up front that no one found out and no harm was done. You are simply unable or unwilling to imagine such a scenario because the word "rape" drives you crazy.

You do not understand what Landsburg thinks. He does not think that the torment the Steubenville girl received/is receiving is outweighed by the benefits (why do you put that word in quotes?) the attackers gained. You are just completely mistaken about this, because you have allowed passion to cloud your reasoning.

3) The full implications of self-ownership are much easier for libertarians to embrace than non-libertarians because we experience much lower levels of digust than you do. Organ markets are widely supported by libertarians and widely condemned by non-libertarians. A few libertarians DO support the right of people to sell themselves into slavery; the majority who don't generally cite either the innate inalienability of one's property right in one's body or economic efficiency as the reason why. And libertarians are much more willing to consider legalizing consensual cannibalism than are non-libertarians. The reason for this is that we are much more willing/able to see human bodies as the equivalent of houses than other people are. We do not sacralize things to the same extent that you do (well - except for human freedom). This is why we think things and hold values that you find repugnant.

Jonathan Haidt has written and said a lot about this topic if you're interested in learning more:

I'm sure there are a bunch of other links I could share, but I doubt you'll even check all of these, so I'll stop now.

Rebecca Friedman said...


Actually, I think Anonymous makes a good point that the camera does reintroduce the psychic cost of finding out, and I agree that it is a more interesting question without it. I thought that post was thoughtful and sensible, certainly not a reflection of someone being "driven crazy" and unable to think rationally.

"Us versus them" is, in my experience, a fairly natural, I believe innate human instinct. It is very hard not to identify with one's own group, one's "us", and view "people like me" as generally better than everyone else.

That is no reason to indulge it.

Max said...


Of course the camera (re?)introduces the psychic "cost" of finding out. The question is whether that "cost" should be counted when crafting policy. If your answer is "yes," as mine (and Landsburg's) is, then there are a few other questions I'd like to know your answer to. ("Do you also assume that the psychic distress of Farnsworth McCrankypants counts as a 'cost' of allowing people to look at pornography? Do you assume that the psychic distress of would-be rapists at not being allowed to rape counts as a 'cost' of banning rape? If your answer is no - what distinguishes the first situation from the second and third?")

Without the camera, the question becomes supremely uninteresting/easy in my mind for reasons that I thought were laid out clearly or at least implied in my response to Anonymous - namely, that the question is defined in such a way that the only possible answer is no, unless you do not actually comprehend/consider the question as posed: If an event transpires that causes no harm or detriment to any conscious entity and confers a benefit on at least one conscious entity, is it wrong? If you think this is an interesting question, then I'd appreciate an explanation of why.

If you think the right question is not "is it wrong" but "should it be illegal," then realize that embedded in the act of making something illegal is the assumption that someone other than the perpetrator has found out - thus it is necessarily the case that all acts which none but the perpetrator knows occurred are de facto legal, because logically speaking the law cannot possibly operate upon them.

If we're talking about a situation where the trespass/rape victim never finds out, but third parties do, and whether or not THAT should be illegal, then we're simply back to Granola McMustardseed and oil drilling in Alaska and psychic distress in general.

Max said...

Ironically, I'm fairly certain that it is your own myside bias that's caused you to perceive my response to Anonymous as being full of myside bias. When crafting it, I was very cognizant of the possibility that this particular form of human weakness might color my words and was careful to be as neutral as possible. However, because you are (I assume) psychologically predisposed towards libertarianism yourself, you read objective statements about the psychology of libertarians as being favorable when in fact people who do not share our peculiar personality type and worldview find those same statements to be extremely unfavorable. As evidence, I offer the following video. It's only two minutes; please watch the whole thing:

When you hear Haidt speak, does it sound to you like he's saying good things about libertarians? Clearly the ones in the room with him think so - they're laughing cordially, and the one at the microphone is enthusiastically agreeing with both Haidt's assessment of libertarians and his perception that most libertarians proudly endorse that assessment.

But try taking off your "libertarian" hat and watching that same video - or reading my comment - with your "Christian conservative" hat on. Or your "left liberal" hat. You might have some trouble; after all, libertarians score extremely low on empathy. Did you forget that part? Or did you just not focus in on it too much because, well, that's a bad thing, and he said so many other GOOD things about us . . .

Try reading Leon Kass's "The Wisdom of Repugnance" first (, and then read my comment again while doing your best to adopt the mindset of the man who said, "Shallow are the souls that have forgotten how to shudder." Takes on a whole new light, doesn't it? All of a sudden, I don't appear to be speaking about how rational and level-headed libertarians are. Instead, I'm saying that we're amoral, cold, heartless, and ultimately about as closely aligned with the god-damned devil as it's possible for human beings to be.

Now listen to Sam Seder for three minutes (, and then read my comment again.

"Holy fucking shit! This retarded asshole is ENDORSING SLAVERY, saying that poor people in the Third World should be forced to sell their organs to eat/survive, and claiming that there's nothing wrong with FUCKING CANNIBALISM. Jesus Christ! No wonder libertarians are part of the Right; that's just where evil tends to congregrate. These people are MONSTERS who are trying, as Chomsky once said of Rothbard, to bring about 'hell on earth.' Thank God there are sane liberals like me here to stop them."

Now try putting on your "objective observer" hat. (You've got one, haven't you? After all, you're a libertarian!) Is anything I said about libertarians untrue? Does any of it necessarily speak for or against us? I submit that it does not.

And if it seems like I've completely misunderstood your point, and if it seems like nothing I'm saying meaningfully addresses or responds to what you said . . . well, that's no great surprise, is it? After all, I'm a libertarian. We suck at empathy, which means we suck at obtaining insight into the minds of others. =P

Max said...

Presented without comment:

Jonathan Campbell said...

seems pretty unsatisfying -- if it became much much easier to prevent someone else from reading books of his choice, would you support transferring the property right to the would-be prohibitor?

David Friedman said...

"If an event transpires that causes no harm or detriment to any conscious entity and confers a benefit on at least one conscious entity, is it wrong?"

It's worth looking a little more closely at the slightly different but closely related question: "Should the legal system treat it as wrong?" or even "Is it a good thing for our morality to treat it as wrong?"

The answer isn't so clear. Legal systems, after all, make mistakes. Treating harmless rights violations as permissable might result in misclassifying violations that are not harmless as harmless and permitting them. So the closest to an optimal system we could have might well involve punishing some harmless offenses.

For an analogous case I have thought a good deal more about, consider the question of whether, from the standpoint of economic efficiency, a legal system ought to deter all and only offenses where the harm done to the victim is less than the benefit to the offender. The obvious economic answer is that it should —hence the simple idea that expected punishment should equal damage done, in order to deter all and only inefficient offenses.

For reasons I have worked out in print, among other places in Law's Order, that conclusion is wrong because the legal system cannot costless deter offenses. The most efficient option might involve expected punishment greater than damage done, because if we deter offenses we don't have to catch and punish the offenders, and it might involve expected punishment less than damage done, because deterring an offense might cost more than the net damage it does.

I don't think your response that the legal system can't deal with it if nobody finds out is adequate. For one thing, the perpetrator always finds out, so if we switch from law to morality I can convert the argument into one about why a perpetrator ought not to commit an act he believes harmless.

In Landsburg's hypothetical, I think many, perhaps most, people would be deterred by the feeling that what they were doing was wrong, even if they believed it was harmless. And, on the whole, I would be less comfortable interacting with someone if I believed he wouldn't be deterred in that situation.

Going back to the legal system, imagine that an agent of the legal system finds out but the victim and those who care about the victim don't. Further imagine the agent has no emotional feelings--perhaps we have an sf legal system run by a computer. Do we want to program the computer not to punish harmless rights violations? The answer isn't clear, since the computer, like human enforcers, can make mistakes.

Anonymous said...

First, Rebecca, thank you for your kind post.


I am the anonymous from above, not one of those other imposter anons that have made comments elsewhere. Thanks for the links, I skimmed the three articles, although the second and third really reference the study itself. I’ll have to save the youtube links for another time. If it turns out I’m bad at understanding the libertarian position, you can rest assured you have done your duty and you can blame my old professor instead ;-).

To attempt to answer the one actual question I think you asked me, I put ‘benefit’ in quotes because there are at least 2 and a half schools of moral philosophy that question the assumption that the rapist benefits (even if nobody ever finds out). The virtue philosophers claim the rapist is actually harmed, the rights philosophers claim that submission to such impulses is just another form of slavery to urges not freely chosen, and some of the utilitarian’s tried to rescue utilitarianism from becoming simply a crude calculus of pleasure and pain by adding character into the equation. If any of these schools are right, then the assumption that the rapist benefits is a wrong one. As I think David wrote elsewhere, nobody can say with certainty which school of moral philosophy is the right one, so the jury, so to speak, remains out (and no, I don’t think the seminal thinkers of these schools simply put their morals in up front only to find them emerging from the back). Does the rapist really benefit? It’s a question worth pondering.

Anonymous said...


“Going back to the legal system, imagine that an agent of the legal system finds out but the victim and those who care about the victim don't. Further imagine the agent has no emotional feelings--perhaps we have an sf legal system run by a computer. Do we want to program the computer not to punish harmless rights violations? The answer isn't clear, since the computer, like human enforcers, can make mistakes.”

Interesting post. The legal system can make mistakes, but on the flip side, we live in a world where it actually might be possible to sometimes, but not always, make a victim unaware of a violation. The current topic that would fit this category is slipping roofies to a victim prior to rape. Sometimes the victim never knows, and sometimes the victim figures out what happened (I’d suggest this is mostly the case, but there is no way to actually prove this).

In this world, our world, even if the victim and anyone who might care are completely unaware, and only a computer sees the crime (other than the perp), I’d suggest most of us would want the computer to punish the violation for the deterrent effect -- because the next victim may remember.

windwheel said...

'Going back to the legal system, imagine that an agent of the legal system finds out but the victim and those who care about the victim don't. Further imagine the agent has no emotional feelings--perhaps we have an sf legal system run by a computer. Do we want to program the computer not to punish harmless rights violations? The answer isn't clear, since the computer, like human enforcers, can make mistakes.'
In the rape case, the State may have guardianship of an incapacitated person and thus is obliged to act if only on the 'broken window' theory.
Depneding on the jurisdiction, and the property/liability regime, the State may always have this obligation even for torts. In practice something which starts off as liability type may turn into property and vice versa. Advowson, it could be argued, did so. Initially the vicar having the advowson could levy tithes on parishioners by compelling attendance and charging pew-rent. The law then was of a liability sort but then this was transmuted into a charge upon the tenants of the parish. Since property rights are enforced thru liability law and since some liability claims are both survivable and assignable, the distinction doesn't seem that useful.
One problem with property rights are they are more information intensive and therefore contestable. Furthermore, assuming agents are identical, a rent accrues to those better able at managing Property or better able to get people to trust them to manage their property. So you still have allocative inefficiency from local monopoly, agent principal hazard, dynamic inefficiency by reason of strategic or speculative behavior and so on.
I suppose if what you are doing is mechanism design for an ideal Tiebout model for fellow Libertarians- the exercise is still useful. I'd be glad for any suggestions for further reading, if that is the case.

Alan said...

I tried asking Landsburg this question but got no response so I'll try it here, reframing it in the language you use in your post.

Given that I accept your rationale and discussion and that I believe you have (as you say) come up with a good solution to the conundrum using a regime of property rights... I am still motivated to ask why property rights are the correct framework to use here? Why are property rights considered more important than, say, personal rights such as rights of self-determination or personal autonomy.

This framework would provide a distinction between the cases Landsburg used, the controversial one of which involved a severe violation of a person's ability to determine her own future and autonomy - though it would not be as applicable to your transformed case of a home invasion. Is this not a valid framework to use?

David Friedman said...


To answer your question, I think I would need a clearer idea of your alternative framework--in particular, how it generates a consistent set of results.

Does my right of self-determination and/or autonomy include a right to have other people provide me the resources needed to do what I want to do with my life? To what extent do my rights of that sort depend on facts invisible to an outside observer, such as how important one accomplishment or another is?

Part of the attraction of a system of property rights, as sketched in my post, is the degree to which they avoid the latter problem.

Alan said...

David, thanks for the quick response. I'm not sure I can generate a consistent enough set of results here. Let me explain: we have this notion of a person as an "adult" in society and law, which concept we use in a number of ways. But we apply it in very inconsistent ways. You can be "adult" enough to be drafted and given a gun to shoot other people, but not "adult" enough to buy a beer. We have differences in age for "adult" enough to marry, and sometimes we have special cases such as "emancipated minors" who are given many rights of adults under the law and conversely we have "tried as an adult" in which we apply adult-level sanctions against someone whom we might not otherwise think was an adult.

I have the sense that my concept of personal autonomy is analogous to this (admittedly fuzzy) legal/social notion of "adult." Maybe that's too fuzzy - and certainly we can point to outcomes in law that seem ridiculous that are based on this notion. But I do think it captures a vital distinction between Landsburg's case of personal assault and your case of home invasion.

I don't think this concept extends directly to the provision of resources, but it definitely relates. For example, we have the notion (again both in law and in society) of "neglect" when one person is supposed to be in care of another person such as a child. The person in-care-of has a duty to provide but there is no comparable right of the adult care-giver to be provided for by society in general.

It's sometimes said that our American set of rights is noticeably lacking in a generic right "to be left alone" and maybe that's what differentiates someone upset by a home invasion from someone upset by the idea that other people are looking at porn.

Sorry this is so vague...

windwheel said...

I would be grateful for any pointers to further reading on this liability vs. property rights approach.
This is because I'm worried by a type of 'principled' (a la Bork) Judicial activism which goes on creating more and more 'toxic rights' which, for convenience, create entitlements enforceable not against the State but property owners. To give an example, the Judiciary may decide that promoting some type of Equality, or Environmental or Health & Safety issue, or other Public Policy goal, is a fundamental legal principle which militates for more and more 'activist' judgments raising the bar for Businesses or other Property owners to comply with the Court's interpretation of what that principle requires, under threat of sanctions.
It may be that Friedman's approach gives us a reason to say to the Court 'you can't go on changing the rules of the game w.r.t the duties of the property owner on the basis that Judicial decisions have to be 'principled'.' This is because the very notion of Property rights is constitutionally based on a contradictory principle viz. that property rights are recognized by the Law because of some large and lasting efficiency type gain to the commonweal arising from their unimpeded and unmodified operation. One such efficiency gain may arise precisely from keeping both the Law and the State out of essentially bilateral voluntary transactions.'
To take an example, suppose I own and operate a School for my own profit. We may regard some restrictions on how I can operate it as reasonable and arising out of the operation of the market- e.g. the education I provide should be of merchantable quality- and, for convenience, either statute or Case law which restricts itself to codifying or otherwise crystallizing the penalties for transgression may be excusable. But what about a Court, or a Legislature, deciding on some 'principled'grounds that 'Education' means something different to what we all previously agreed it meant. Suppose the Judges say 'true' Education must be Egalitarian and non-discriminatory. A School isn't a proper School unless at least x percentage of its students can't afford to attend it.'
In this context, I would be grateful, as I said before, for any pointers to recent work either by lawyers or economists, especially such as incorporate Game theory of a Ken Binmore folk theorem type.
The fact is it is out of control Judges, mounted upon very high horses, who- at least in India and Europe- pose a greater danger than sleazy Legislators with their pork barrel politics.

Anonymous said...

I find this topic interesting (especially reading the comments), though viscerally disturbing now and when first reading the article at the Chronicle in which a commenter to said article linked to this post.

One of my visceral responses to being disturbed is to attempt to disturb in kind. So this question is directed to the Professors in the bunch:

If a person has accumulated debt and significant time served to pursuing a degree that a school allows the person to pursue, and which the person's professors/advisers say is achievable, and should the professors/advisers stretch time-to-degree out 'infinitely', or after significant time 'kick' the person out of the program, is the person justified in murdering at least one of said professors/advisers in retaliation for the lifespan they have lost? (This assumes that the time otherwise remaining in the professors/advisers lifespan is approximately equal to the time spent by the student in the degree program.)

Please, please, please!! Understand that this is a hypothetical, not a threat! I don't believe in destroying dreams or lives whoever said dreams or lives belong to. I consider life (especially sentient life) itself to be superior to, and outside of, any value system. And thus consider the taking of a sentient life to be only ever justifiable to immediately protect a life.

Anonymous said...

"But we stipulated up front that no one found out and no harm was done."

The action would somehow alter the actor (the rapist) (I know the difference between a thought and an action, and how the two differentially effect the thinker/actor). How can we determine what effects this alteration will have in the future? To both the rapist/rapee relationship, and the societal orientations of the rapist, and the rapists relation to others in non-societal contexts?

Potential for harm (I guess this would be the liability issue), and hell, potential for change (given that change in one person requires active change on the part of others - and consequent opportunity costs and gains), must always be considered, even if it can't be weighed. Out of such considerations come the less serious versions of "contributing to the delinquency of a minor". And out of such considerations come our 'grandfather' laws and payment for the exercise of eminent domain.

Anonymous said...

Apologies for not using apostrophes where appropriate. It's early.

Anonymous said...

"For one thing, the perpetrator always finds out, so if we switch from law to morality I can convert the argument into one about why a perpetrator ought not to commit an act he believes harmless."
"And, on the whole, I would be less comfortable interacting with someone if I believed he wouldn't be deterred in that situation."

Yes, thank you, this is the point-of-view that needs to be said before any artificial economic system point-of-view is referred to.

Get down to visceral roots / first principles - the things which drive and motivate us fundamentally. Because when you deal with such high level abstractions as economics you just avoid the foundational issues, and risk arguing in circles.