Answers to Questions About Libertarian Punishment and Estoppel

An interchange with someone with questions about one of my articles, which sets out my “estoppel” theory of libertarian rights. For more background on these issues, see the links interspersed below, and those in the following endnote: ¹

From Mr. S:

I just read your article “A Libertarian Theory of Punishment and Rights”, which was very interesting, thank you.  It raised a number of questions in my mind which I wanted to raise with you in case you’ve dealt with them elsewhere and can point me to these sources.

(Note that I am ignoring for now that both aggressor and victim may have protection insurance policies and so the punishment scale might already have been agreed to; I’m more focused on what a libertarian judge should decide in the absence of preset penalties.)

1. I raised with [a certain libertarian philosopher] the problem of failed attempts.  You deal with this a little in the area of assault (p.640), but not sufficiently to answer my question.  If A shoots at B but misses, what is the punishment B can levy on A?   Is it just that B can shoot at A and miss, which seems pointless?   But if that wouldn’t instill in A the same amount of fear that B suffered, per your assault example, is B justified in actually hitting A with the bullet and potentially killing him, with A being estopped from complaining about this?  Should failed attempts be punishable at all?  According to Rothbard, neither deterrence nor rehabilitation are valid bases for punishment; rather, only restitution and retribution are appropriate.  But in my example, there is nothing to be restituted and retribution would imply shooting and missing.  Then, to make things more interesting, what if A shot at B and missed, and B didn’t even know A had shot at him, so never suffered any fear (but witnesses saw it and reported A)?   Again, what is the basis on which failed attempts should be punishable?  In a failed attempt there has been no physical invasion of body or property, and mental distress cannot be the basis for action since one cannot have property in one’s feelings (that would raise a host of conflict-creating problems).
2. It seems to me that the estoppel principle should be more narrowly stated.  In your examples, you say that if A murders B then  A is estopped from complaining about being murdered as punishment.  Yet shouldn’t this be restated to say that A is estopped from complaining about being murdered as punishment by B or his representative?  In other words, I don’t think you mean to imply that A is estopped against the whole world from complaining about being murdered; D (a complete stranger) cannot murder A and then claim that by A’s action in murdering B, A is estopped against anyone from complaining.  Yet that would be one reading of the estoppel principle as stated, since A has by his actions apparently indicated that he sees nothing wrong in murder, so it could be open season on A.
3. If, as stated on p.635, the goal of punishment is to equalize damage suffered, not just the actions that caused the damage, then that could work against the victim.  If nice person A beats up gang member B, since B is used to getting beaten in his daily life the damage suffered is probably not that great.  Thus he would have to reduce the punishment beating he exacts on A.  Perhaps the theory should be that the victim can exact the greater of (x) equalizing action and (y) equalizing damage.  However, saying “should” is somewhat normative, and I wonder what the positive theory behind such a “greater of” concept would be.
4. That raises a broader point: what is actually being estopped and therefore what forms the basis for outlining the bounds of punishment: (A) the actions of the aggressor [e.g., punching the victim], (B) the result caused [e.g., burst spleen] or (C) the damage suffered by the victim [e.g., inability to continue working as a laborer]?  Can an aggressor be estopped from any one of these that was not obvious at to him at the time?  If so, then does estoppel not really rest on what the aggressor has actually acknowledged by his actions, but rather what a “reasonable aggressor” should have realized what he was doing?
5. Moreover, why can the victim choose to exact a dollar remedy for a physical aggression (leaving aside the situation where the aggressor bargains for this with the victim to avoid physical retaliation)?  Under estoppel the aggressor has only acknowledged he does not believe hitting is wrong, but it doesn’t mean he has acknowledged that taking someone’s money is wrong.  It seems that the broader the range of remedies the victim is entitled to exact, the looser the connection to the aggressor’s actions which give rise to the estoppel.  We could end up effectively concluding that the aggressor indicated by his actions that he does not believe in the sanctity of private property at all, and thus any punishment is warranted.  Surely that’s not where we should end up?
6. What is the theory underlying why heirs can take action on behalf of a murder victim (assuming there is nothing in the victim’s will saying so, and there is no protection policy for which the heirs are the beneficiaries)?  If each person’s body is his own property, how can an heir claim to have been damaged when only the victim’s body was invaded?

Kinsella reply:

Stephan:I just read your article “A Libertarian Theory of Punishment and Rights”, which was very interesting, thank you.  It raised a number of questions in my mind which I wanted to raise with you in case you’ve dealt with them elsewhere and can point me to these sources.

(Note that I am ignoring for now that both aggressor and victim may have protection insurance policies and so the punishment scale might already have been agreed to; I’m more focused on what a libertarian judge should decide in the absence of preset penalties.)

1. I raised with [philosopher X] the problem of failed attempts.  You deal with this a little in the area of assault (p.640), but not sufficiently to answer my question.  If A shoots at B but misses, what is the punishment B can levy on A?   Is it just that B can shoot at A and miss, which seems pointless?

In the meantime I’ve had various blogposts and other articles dealing with various issues. Some are collected here https://stephankinsella.com/llw.

On this particular issue I’d refer you to:

• The Limits of Armchair Theorizing: The case of Threats
• Fraud, Restitution, and Retaliation: The Libertarian Approach
• The Problem with “Fraud”: Fraud, Threat, and Contract Breach as Types of Aggression

So in light of those, I’d say–first, not sure how much we can deduce ahead of time (as armchair theorists) about such issues. Still, my inclination is as outlined in the article you noted plus the ensuing posts. My view is that attempt, sometimes called assault, and threats, are treated as species of aggression. My view is that aggression in its pure form is a combination of intention plus invasion, or, rather, intentional invasion. In the case of a threat, or assault, or attempt, the invasion does not quite happen, but the intent is there, so if you consider it as sort of a product (intent times damage) then the latter factor is lesser in the case of an unsuccessful attempt, but the intent is still there. So it’s not as serious as succesful trespass, but still sanctionable. Likewise, negligence is the other way around: intent x damage is there product, but in this case the damage is there, but the intent is partial. It is not mere unintentional behavior, it is negligent so it is “partly” intentional. So again, the product is less than fully intentional crime. In the case of assault or attempts, the intent part is 100%, the damage part is less; in the case of negligence, the intent part is partial, but the damage part is complete. So my view is that both negligence, and attempted crime, are both sanctionable but the damages are less than would be the case than for full intentional (premeditated) successful invasion/trespass. More than this, I do not think we can say, from our armchairs–though I do think consulting the way the common law and great legal systems tend to handle practical cases like this, is a good rough guideline. If we were in the situation where someone needed me, Kinsella, or someone else, to give an answer, to outline a libertarian legal code, I’d probably start by a sophisticated and realistic legal analysis of the legal systems of the world–i.e., I’d just ask legal experts, or consult their many legal treatises. I’ve done this to some respect and others have and can too. I just myself go into only so much detail, simply because this endeavor seems premature to me at this point. I’ve toyed with the idea of writing the ideal libertarian law code, or constitution, or civil law/code, and I have an idea how I would do it (I would start with one of the common law restatements of the law, and/or a good Roman civil code, and try to emend it where it obviously strays, to conform with libertarian principles), but I am not sure what would be the point of this endeavor at the current time (or ever).

But if that wouldn’t instill in A the same amount of fear that B suffered, per your assault example, is B justified in actually hitting A with the bullet and potentially killing him, with A being estopped from complaining about this?  Should failed attempts be punishable at all?  According to Rothbard, neither deterrence nor rehabilitation are valid bases for punishment; rather, only restitution and retribution are appropriate.

I think this view is a bit confused, as may be clear from my estoppel argument. I think deterrence or rehabilitation (etc.) are reasons one might want to engage in punishment, but are not justifications for it. The justification for it is that the aggressor has no reason to complain.

But in my example, there is nothing to be restituted and retribution would imply shooting and missing.  Then, to make things more interesting, what if A shot at B and missed, and B didn’t even know A had shot at him, so never suffered any fear (but witnesses saw it and reported A)?

The civil law defines assault as either the attempt to commit battery, or putting someone in fear of receiving a battery. I believe both can be classified as species of aggression. The former because if A tries to harm B, and thus B is enttled to “try” to harm A too; but probabilities enter into it. The latter because the oly way to make the defendant fear receiving a battery from B is if B really has the legal right to try to harm A.

Again, what is the basis on which failed attempts should be punishable?  In a failed attempt there has been no physical invasion of body or property, and mental distress cannot be the basis for action since one cannot have property in one’s feelings (that would raise a host of conflict-creating problems).

I think in general causing mental distress is not enough for liability, as the law recognizes. You have to have it caused by some physical event, like the result of negligence. My view is once you commit invasion–which is an intentional use of someone’s property without their permission, and which by extension can include both negligence (lowered but not zero intent) and attempt (lowered but not zero harm [in the paradigmatic case]),– then the harm occasioned can be taken into account. The law has various rules that recognize this, e.g. the “eggshell skull doctrine”.

It seems to me that the estoppel principle should be more narrowly stated.  In your examples, you say that if A murders B then  A is estopped from complaining about being murdered as punishment.  Yet shouldn’t this be restated to say that A is estopped from complaining about being murdered as punishment by B or his representative?

Yes.

In other words, I don’t think you mean to imply that A is estopped against the whole world from complaining about being murdered; D (a complete stranger) cannot murder A and then claim that by A’s action in murdering B, A is estopped against anyone from complaining.

Actually I am not quite sure about that part. I’ve refrained from being emphatic about that, because of that uncertainty. I’ve hinted at it in a few places, eg. where I am skeptical of the ability of A (or A’s agent) to “forgive” B. Even if A forgives B, B is still an aggressor, and … so does it really violate B’s rights if A “changes his mind” and punishes B anyway? Why is it worse to punish B for harming A, after “forgiving” B,  than to just punish B without forgiving him? Maybe the added psychological torment B feels in the former case is justified. And even if A and his line forgive B, isn’t B a “standing threat” to third parties C, D, et al.? So this issue to me is not a completely clear one (not that I don’t have opinions about it).

Yet that would be one reading of the estoppel principle as stated, since A has by his actions apparently indicated that he sees nothing wrong in murder, so it could be open season on A.

That is one argument that I don’t think is dismissible. My inclination is that unless the malefactor has done something so heinous or a repeated series of crimes as to show himself to be irredeemable in some sense, and thus a standing threat, then the community would tend to defer to the wishes of the victim or his agents, in the interest of justice and avoiding feuds and the like. And even in the case of an irredeemable outlaw, in most cases it is hard to imagine an institutionalized system of execution or even incarceration, but probably more likely some kind of outlawry due to social ostracism and the like.

If, as stated on p.635, the goal of punishment is to equalize damage suffered, not just the actions that caused the damage, then that could work against the victim.  If nice person A beats up gang member B, since B is used to getting beaten in his daily life the damage suffered is probably not that great.

Correct. I think I mention in my article on estoppel that the victim should have “options”–for just this reason. A rape victim might not want to have to have her rapist raped; and anyway, he might not mind it as much as she did. That’s why I always favor the vitctim, in case of any doubt. And I think any just legal system would. Thus the burden of arguemnt and proof would be on the aggressor, to show why any proposed retribution by the victim is not just.

Thus he would have to reduce the punishment beating he exacts on A.  Perhaps the theory should be that the victim can exact the greater of (x) equalizing action and (y) equalizing damage.

I actually do say basically this, by giving the victim the choice. see sec. B on p. 65 of this JLS version of the article: http://www.mises.org/journals/jls/12_1/12_1_3.pdf . (Remember also that I wrote this 20 years ago, and would probably try to improve my formulations now, though I find that I stand by most of what I wrote back then.)

However, saying “should” is somewhat normative, and I wonder what the positive theory behind such a “greater of” concept would be.

That raises a broader point: what is actually being estopped and therefore what forms the basis for outlining the bounds of punishment: (A) the actions of the aggressor [e.g., punching the victim], (B) the result caused [e.g., burst spleen] or (C) the damage suffered by the victim [e.g., inability to continue working as a laborer]?

I think here we might reach the limits of a quick email response, and would be happy to talk to you about this sometime. But in short, I think of it like this. What is estopped is a claim. The claim of the aggressor, that the victim doesn’t have the right to do XYZ  to the aggressor, in response to the aggression complained of. It is the claim that is estopped, that is not heard. the reason is that the claim is based on some normative proposition that is incompatible or inconsistent with the maxim of the earlier act of aggression. In other words, the estoppel analysis is aimed at satisfying the victim (or his agents, or the judge/the community) that it is justified to treat the aggressor in a certain way despite his manifested objection. Basically, it is a way of showing that the aggressor “does” consent (or “has consented to”) the violent response to his aggression. That is one reason I argue that the conventional idea of promissory estoppel for contract enforcement is in fact circular (see note 14 of the previous linked piece), BUT the use of aggression as I think of it, is not circular. Because the action of aggression has factually happened and cannot be undone. (Unlike a promise, which does not harm anyone, and which can be retracted. This is one reason I disagree with Blockean-type “inalienability” arguments.)

Can an aggressor be estopped from any one of these that was not obvious at to him at the time?  If so, then does estoppel not really rest on what the aggressor has actually acknowledged by his actions, but rather what a “reasonable aggressor” should have realized what he was doing?

Moreover, why can the victim choose to exact a dollar remedy for a physical aggression (leaving aside the situation where the aggressor bargains for this with the victim to avoid physical retaliation)?  Under estoppel the aggressor has only acknowledged he does not believe hitting is wrong, but it doesn’t mean he has acknowledged that taking someone’s money is wrong.

Because the victim should have a wide arrange of remedies or options, and also because in most legal systems we can imagine actual institutionalized punishment is not available but as mentioned in various posts (see above and below) the level of monetary restitution/rectification awarded would be based to some degree on the severity of the crime and thus on the severity of punishment the victim ideally could inflict, and thus use to bargain for a higher degree of damages– see e.g. my discussion of the “millionaire problem” in some of these posts: The Libertarian Approach to Negligence, Tort, and Strict Liability: Wergeld and Partial Wergeld; Fraud, Restitution, and Retaliation: The Libertarian Approach; The Problem with “Fraud”: Fraud, Threat, and Contract Breach as Types of Aggression.

It seems that the broader the range of remedies the victim is entitled to exact, the looser the connection to the aggressor’s actions which give rise to the estoppel.

Well, maybe. I think  I would word it the other way round: the estoppel gives rise to a broad range of remedies the victim can choose from.

We could end up effectively concluding that the aggressor indicated by his actions that he does not believe in the sanctity of private property at all, and thus any punishment is warranted.  Surely that’s not where we should end up?

As I state in the longer estoppel argument, I actually think that is the “default” position, the victim is entitled to argue, thus shifting the burden of proof onto the assailant to argue that (say) capital punishment for a relatively minor trespass is not warranted. see e.g. Section IV.C on p. 70 of the JLS piece linked above. I can’t claim to have fully solved this, but I actually did explicitly try to address this very issue, there. In practice we could imagine defaults and pleadings and presumptions to arise, and rules of thumb, and so on, taking proportionality into account, so that if you just kill someone for stealing an apple, people view you as a murderer. So over time the job of aggressors becomes easier b/c of the work done before to refine the legal principles, but in principle, the victim stands in a superior position to the aggressor even with respect to standards and burdens of proof.

(Of course all this ignores epistemic issues–it is possible the victim or legal system is mistaken in claiming B was an aggressor–and this is of course one reason I think that the cost of a punitive institutionalized system is too high and thus it would tend to not be feasible. [Fraud, Restitution, and Retaliation: The Libertarian Approach])

What is the theory underlying why heirs can take action on behalf of a murder victim (assuming there is nothing in the victim’s will saying so, and there is no protection policy for which the heirs are the beneficiaries)?  If each person’s body is his own property, how can an heir claim to have been damaged when only the victim’s body was invaded?

I think we have two main issues here. The first, I think, is minor. The issue [one of my friends] has quibbled with me about, about the efficacy of a will. His argument is that at the moment of death you no longer own your body so why would your will count as to who owns your body now. I don’t see that as a big problem at all. I think this is philosopher’s quibbling that would make literally no difference to people in the real world, who would of course take the decedent’s last will and testament as the final word that is relevant as to what to do with or how to dispose with his estate, body, rights, etc. (and besides, lawyers could come up with workarounds for this, with trusts and so on, if this became a real problem in Pettifoggery Land).

Second, then I would view the right-to-retaliate (and related/derivative rights) as owned by the victim and thus as inherited by whomever he transferred the rights to, according to the will/testament or, failing an express will, according to the laws of intestacy. The latter goes by reasonable default presumptions, e.g. that A prefers his wife, or his children, or his parents, and then his siblings, cousins, neighbors, and so forth, to be his agent-after-death. So it’s not that the heirs are damaged by the invasion due to the dad’s death. It’s taht they inherit his right to seek restitution, because of the damage done to him when he was alive.

My best, Stephan

Mr. S:

Thanks Stephan, I really appreciate your prompt and complete reply.  I’m going to take some time to digest all of this and to hunt down some of the other pieces.

1. • Argumentation Ethics and Liberty: A Concise Guide
   • “Libertarian Legal Theory: Property, Conflict, and Society,” Mises Academy course, lecture 1 (episode KOL018)
   • The Social Theory of Hoppe,” Mises Academy course, lecture 3 (episode KOL153)
   • New Rationalist Directions in Libertarian Rights Theory
   • Defending Argumentation Ethics: Reply to Murphy & Callahan
   • A Libertarian Theory of Punishment and Rights
   • How We Come To Own Ourselves
   • Causation and Aggression
   • Hoppe: First significant thinker to get libertarianism totally right
   • other material linked at Selected Supplementary Material for Law in a Libertarian World: Legal Foundations of a Free Society . [↩]