Further response to WelcometotheUnknown: punishment theories

Here is WelcometotheUnknown's response to my blog post. My response follows.

Thank you for your detailed reply. Before I reply, first note that I used the word “retribution” in my posts accidentally multiple times. I meant to use the word “restitution” instead. Interestingly, they have similar meanings, so my mistake led you to bring up a good point.

First to clear up a few things: I certainly do not consider myself a pacifist. I think there is a difference between uses of force in defense and retaliatory uses of force. Perhaps the Non-Aggression Principle does not forbid either kind of force, but there is still a difference. For example, if you tried stealing from me I would consider it just to use force to prevent you from stealing from me. If you managed to steal something from me, I would also feel justified using force against you to get it back (“restitution”). Would I consider it just to demand you give me back something extra in return for my troubles or for punishment of some kind against you for the act of theft? Perhaps, but I think we soon run into trouble saying that such uses of retaliatory force for retribution are justified.

I think there may be a gray area between your description of what I call anarchy and what you call anarchy. The gray area is centered on what instances we say people can legitimately use physical force and how far they go with that use of force. The gray area may simply be a spectrum from “restitution” to “retribution.” These terms may be ambiguous in and of themselves seeing as it is very difficult if not impossible to come up with a clear objective definition of what counts as “restitution” and what counts as “retribution.” For example, if an item of subjective value is stolen (e.g. a painting, rather than an amount of money as in the DD vs. Bill example), then what counts as restitution? Surely what one person may say counts as restitution another will say is retribution due to the fact that the value of the property stolen is completely subjective, rather than something like a quantity of money.

If someone were to steal a painting from me and then destroy it, I wouldn’t be able to get it back, no matter how much coercion I used. Could I thus demand some amount of money from the thief and legitimately use force against the thief to get the money from him in the name of restitution? Even if I’m someone who believes that obtaining only restitution rather than retribution is just, there is still a problem in defining what counts as restitution. What if I value the painting at a billion dollars? If I draw a stick figure on a piece of paper (my painting) and then Bill Gates takes out a match and burns it, am I really justified in using physical force against Gates to take a billion dollars from him? Perhaps the Non-Aggression Principle doesn’t forbid my actions because my actions are indeed retaliatory force rather in reaction to an act of aggressive force, but I think you and I would agree that there is something unjust about this. But, what is unjust about it? Can we define it?

And what about crimes other than theft? If someone commits rape or murder, what counts as “restitution” and what counts as “retribution”? I can’t think of an objective definition for either term. Thus, I think that unless I’m prepared to say that no amount of retaliatory force is justified (beyond re-obtaining something that was stolen), even in the case of murder, then I must accept the subjective spectrum of retaliatory force in subjectively defined degrees as legitimate.

The only time that I would confidently say retaliatory force is justified is when it is used to re-obtain something that was stolen. As soon as retaliatory force is used to get something extra (retribution) or even something “equivalent” but in a different form (restitution in the form of money, for example, for a stolen/destroyed piece of property without a market value (i.e. something that is not replaceable), e.g. a one-of-a-kind painting) then we run into the problem of calling it just to coercively seize a billion dollars from Bill Gates for destroying my stick figure painting.

So am I prepared to say that such retaliatory force to demand restitution and retribution from murderers, rapists, and thieves (thieves who have destroyed what they have stolen or done something else to what they have stolen so that the stolen property can no longer be returned, even using force to return it) is unjust and unnecessary? I certainly would consider some acts of retaliatory force unjust, but others seem very reasonable, such as demanding that a murderer pay the victim’s family/next of kin at least some small fee as “restitution” for his crime. Due to the subjective nature of them and our inability, at least as far as I now see, to objectively define the degrees of retaliatory force, I think that if I want to accept the use of retaliatory force to obtain a small fee of “restitution” for the murder victim’s family as legitimate, then I have to agree with you and accept the whole spectrum of retaliatory forces in all degrees as legitimate. While I may personally disagree with some uses of retaliatory force, such as using retaliatory force to obtain a billion dollars from Gates for the stick figure drawing he burned, the difference is subjective as far as I see.

“It may help if you can clarify what degree of retaliatory coercion you see as justifiable.”

It’s just to use retaliatory force to retrieve something that was stolen.

It may or may not be just, in my own subjective opinion, to use retaliatory force to obtain “restitution” or “retribution” for property that is stolen that is not returnable or replaceable or to use retaliatory force or obtain “restitution” or “retribution” for acts of rape, murder, etc.

It may be just to use retaliatory force to lock up a murderer, in the name of defensive force. I haven’t thought about this very much, but I think a case can be made to say that locking up someone who has committed a murder in the name of defending others from other possible acts of murder may be justified. I’m not very confident about saying this is just, however, because I think it could also be argued that this idea could be extended to locking up other violent criminals for the same reason. What if someone starts a fist fight? Can he be locked up for life as well for the same reason? If force can be used to lock up the murderer in the name of defending people from other possible future attacks, then can’t the same be said of the person who started the fist fight? The differences don’t seem too clear to me.

Out of curiosity, what degree of retaliatory coercion do you think is justifiable? I don’t think the terms “restitution” and “retribution” define the degrees well enough. They have some subjective meaning, but objectively I think they fail to define what acts of retaliatory forces are justified and which are not.

Lastly, while I called your description of a free society a system of “competing governments” in my post on YouTube, my realization in this post that the degree of retaliatory force that is justified does not seem to be objectively definable and the fact that I think it would be very reasonable to use retaliatory force against a murder to make him pay at least a small sum of money to the victim’s next of kin in the name of “restitution” has made me change my mind: I now consider what you described in the video as a system of law without government, law in a stateless/free/anarchic society.

Furthermore, I have realized that the distinction between what you describe and government is not the lack of monopoly, but rather, the lack of compulsory monopoly. Even if Dawn Defense were to obtain a monopoly in a geographic region and be the sole provider of such law services in a geographic region, it still would not qualify as a government/state, for the very reason that people would choose to fund it voluntarily. If people in the region where Dawn Defense had gained a monopoly choose to stop funding it Dawn Defense would allow this. If Dawn Defense did not allow it, then the funding would be renamed from “voluntary” to “compulsory taxation” and at this point Dawn Defense would be a government/state.

If everyone in the world subscribed to Dawn Defense’s services and nobody else in the world offered any law services, Dawn Defense would have a monopoly on law services, but still would not be a government. Thus, I believe the defining distinction between private law organizations and governments is not the presence of competing firms vs. a monopoly, but rather voluntary funding vs. compulsory taxation. Having said this, I have no doubt that when we transition from compulsory taxation to a free society of voluntarily funded law organizations, competition will thrive. So while law services in a free society may be marked by competition, all I mean to say is that the competition is not the defining aspect of what makes the society free.

Glad I helped you understand it better.  You bring up a lot of good points here.

The nature of the beast is that, whenever political philosophies are applied in the real world, there are always going to be “continuum issues” or “boundary issues”. Libertarianism is no exception to this. From the armchair, we can explain principles, and we can construct hypotheticals to try to demonstrate how to apply these principles, but we cannot escape continuum issues.

For example, what exactly constitutes homesteading of a plot of land? A court can be fully libertarian, but it still must ask, did what Joe did to the plot of land qualify as homesteading? Perhaps he fenced off the land, and put a sign up. Is that enough? Or does he need to do some other physical thing to the land to make it his, and if so, what?

Of course, the answer is that we cannot answer this. It has to be determined by custom; by whatever works. The courts will draw a line somewhere and say ‘to homestead a field, the following requirements must be filled…’ We can only be vague about what these requirements might be.

Similarly, we need the courts to draw a line to distinguish coercion from voluntary exchange, because there is a continuum; there are ambiguous cases, which we cannot resolve from the armchair. There is no ‘libertarian place’ to draw the line. Libertarianism only outlines the principle itself; we need the courts to apply it, and in a free market setting, those courts will be guided by the customs of the society, and the opinions of the consumers.

All the questions you bring up in your post are matters of application, not of principle. Clearly all matters of application are subjective. This is regardless of whether the principles themselves are considered subjective or objective.

My point is that it is no vice to recognize continuum issues, or to admit that something is subjective. This is unavoidable. I don’t know how much compensation you will get from Bill Gates for stealing your precious stick-man painting. The only honest answer is that old libertarian cliché “let the market decide”. I can speculate that you probably won’t get a billion dollars from him, because any court that made such a preposterous decision would quickly lose customers.


With my descriptive hat on, I can predict that courts would enforce some kind of punishment or some kind of compensation that must be paid by aggressors. A “market for punishments” would develop. For example, in many evolved legal systems, it was customary for aggressors to have to pay the victim X times the price of what was stolen. X may be 2 or 3 or 5. I have not read of any system where X was 1. In other words, aggressors are usually made to pay to their victim more than what they stole. It really doesn’t matter how this was justified. Likewise, in a setting of competing private courts, it won’t matter how they try to ‘justify’ their policies. The merits of “punishment theories” don’t really matter when they are being actively tested out on the market.


Now I’ll put on my normative hat. What do libertarians have to say about the limits of retaliatory coercion?

There are retributionists and there are restitutionists. Both positions are nuanced. The difference between them is clear in theory, but not in practice. Given any court decision, it is futile to try to assign some portion of it as “restitution” and some other portion as “retribution”.

For example, take a punishment of 3X. Perhaps we call 1X the “restitution” part. But restitutionists do not stop here. They would rightly say that the victim is not “fully restored” just by giving back her property - the victim has also been “scared” and “inconvenienced”. They say that the aggressor has to pay more than 1X to account for this. They might say 2X covers these two things. So we cannot say that a punishment of 3X comprises 1X restitution plus 2X retribution. It may be all restitution.

The restitutionists are focussed solely on the victim. Their goal is to “make the victim whole again,” or to make it as far as possible as if they crime never happened.

The retributionists abandon the idea of “restoring the victim”. Their goal is to “make the aggressor suffer like he made the victim suffer”.

Both of these approaches are obviously beset by continuum issues, but we must not let such issues cloud the principles being discussed. In principle, does it make more sense for a court to focus on making the aggressor suffer, or on restoring the victim? In other words, should punishments be thought of primarily as retributive, or restitutive?

Traditionally, restitution has been the libertarian position, and people like Walter Block take this view. Stephan Kinsella convinced me that retribution should be the primary consideration, using his “estoppel” argument. There are just too many problems with restitutionism, even in principle.

Having said that, I believe that in a competitive market, while the courts may consider retribution primary when making their decisions, victims will probably “make deals” with their aggressors such that most people will think of the system as primarily restitutive. This adds another layer of complexity.

Personally however, I am inclined to think that there really isn’t a ‘correct’ libertarian theory of punishment. It’s a whole different subject. Libertarianism forbids aggression, but it doesn’t imply – without bringing in extra principles – anything about the limits of retaliatory coercion. It is a matter of taste whether a libertarian believes restitution or retribution ought to be primary in determining punishments.


Your last two paragraphs are spot on. The one thing I would point out is that some people (notably Rothbard) used the term monopoly to mean a compulsory monopoly. A firm that happened to gain 100% market share in a free market is not a ‘non-compulsory monopoly’: it is not a monopoly at all, but just a highly successful firm. This explains why I did not use the term ‘compulsory’ as a qualifier for ‘monopoly’ in my post, and also why the dichotomy that defines anarchy/government is free market vs. monopoly, and this is equivalent to voluntary vs. compulsory. Compulsory is implied by the term monopoly, to Rothbard and me.