Libertarian Papers

Home | Mises Library | Causation and Responsibility: A New Direction

Causation and Responsibility: A New Direction

04/29/2009Matt Mortellaro

Matt Mortellaro*

In the past several years, a great deal of ink has been spilled on the issue of the proper view of causation (in its association with criminality),1 specifically as a reaction to the traditional Rothbardian position2 and the implications of this position, as presented by Walter Block.3 The critiques made by such libertarian luminaries as Hans Hoppe, Frank van Dun, Stephan Kinsella, and Patrick Tinsley have been in the direction of favoring an expansion of the Rothbardian framework to include actions outside the overt invasion of physical property, and thereby expanding the class of activities which are prohibited under libertarianism.4 This paper shall both attempt to defend the core of the Rothbardian view of criminal responsibility from the criticisms lodged against it, and then further seek to shrink the class of prohibited activities by applying the Rothbardian analysis in a more consistent and complete manner.

I. Hans-Hermann Hoppe

Of the three suggested expansions of the libertarian legal code based only on non-aggression, Hoppe’s, as presented in “Property, Causality, and Liability,” is the least objectionable from a strictly Rothbardian position. Indeed, it would seem that this is more akin to outlining a logical implication of the non-aggression principle than truly changing it in a fundamental manner. Hoppe’s view is influenced here by a specific view of liability (i.e., one informed by the related theories of Adolf Reinach) which differs from the Rothbardian view of this topic in several vital ways,5 but does not seem to be a radical adjustment from the traditional view.

In Section III of his paper, Hoppe explains the distinction between Reinach’s view of liability and the view held by Rothbard. Quoting Reinach, he explains:

In the case of a man’s death, it is not sufficient that the death resulted from the action of an accountable (sane) person; as an additional requirement of a punishable offense, intent and deliberation (premeditation) or intent without deliberation (negligence) or, as we can summarily say, fault must be present as well. Causation of success and fault are requirements of punishment.—Fault must always be found. (Hoppe 2004, p. 89)

This differs from Rothbard’s view in several important ways. First, it establishes fault as a criterion for establishing liability. This eliminates the problem of a liability being imposed without any action on the part of the “aggressor”—which intuitively seems necessary for any responsibility to be shouldered—in that it requires that true action in the Misesian sense takes place, not merely an uncontrolled reflex.6 Second, it subtly, but importantly, changes the traditional Rothbardian view of causation. The defender of the non-aggression principle needn’t ignore blatant acts of aggression simply because no overt physical invasion is made under Reinach’s view of causality, such as in the case of entrapment (which shall be discussed at length below). Third, the emphasis on fault allows us to distinguish between what Hoppe calls the “proper assumption of risk” and actual invasive actions.7 This gives the Rothbardian a framework which allows for a rigorous defense of the non-aggression principle against critics such as van Dun.8

Hoppe would seem to agree with this view of his revision, stating explicitly that, “Rothbard would have likely agreed [with this analysis]… However, this implies admitting that the narrow causality criterion is inadequate.”9 So far, it may therefore be argued that the Rothbardian has nothing to fear from Hoppe’s revision, that it is simply a more rigorous formulation of the traditional Rothbardian non-aggression principle.

The main thrust of Hoppe’s argument is revealed in Section IV and explicitly stated in Section V; he seeks to rid the non-aggression principle of the view that only “‘objective’ (external, observable)” invasions can be criminal, leaving out indirect actions and internal, subjective considerations.10 Several demonstrative examples seek to prove this point. One is particularly illustrative:

Consider another example. A, B’s employer, orders B to come directly to him, knowing that half-way there is a concealed trap. B walks into the trap and is injured. Reinach would find A liable. Rothbard would let him go, because there is no “overt physical invasion” initiated by A. A merely says something (which in itself is clearly a noninvasive act) to B; and then “nature” takes its course with no further interference on A’s part. That is, entrapment, as an indirectly and by in itself noninvasive means effected physical harm, would have to remain free of punishment. (Hoppe 2004, p. 92)

It seems obvious here that, indeed, the requirement of an “overt physical invasion” is too great and that, intuitively, we would have trouble denying the criminality of A’s actions. All the same, it would seem that this is no disagreement with the non-aggression principle per se, but rather with what we might call “Block’s Paradox.”11 In this construction of the non-aggression principle, any combination of acts which, by themselves, are non-invasive cannot create an act which is itself invasive (and therefore illicit). Yet, rejecting this particular construction does not seem to, by itself, threaten the non-aggression principle nor many of its traditionally-assumed conclusions.12

For instance, we can take a case where all would agree that an invasive action occurs, murder by shooting. And yet, it becomes hard to see what invasive actions make up this criminal act in all circumstances. What objective, overt physical actions must be taken for the murder to occur? First, the gun must be pointed at the victim. Here, we might have an immediate objection, for pointing a gun at someone is surely a threat. However, let us assume the victim in this case is blind and thus cannot see the pointed gun. Is this still a threat? Unless the victim has some way of knowing that the gun is pointed at him, certainly it is not.13 The second thing that must happen is the murderer must pull the trigger. Now this, in itself, is not objectionable at all—if the gun was pointed at a target in a shooting gallery, or at an attacking criminal, we would have no cause for saying that the murderer has committed a violation of the non-aggression principle. So, where did the invasive action occur? It appears that this is another example of two actions which, separately, are non-invasive combining to form an illicit action (namely, murder). With this in mind, we have reason to believe that rejecting Block’s Paradox is far from a rejection of the non-aggression principle, and indeed may be a welcome correction.14

There might be reason to think that it is unfair to Block’s position to use this example in this way—it might be objected that pulling the trigger whilst pointing it at an innocent are not two separate acts, as described above, but rather is simply one complex act. And, indeed, such an argument is not totally without merit, though such a system seems far from stable. For example, why not extend such analysis to the issue of blackmail? This would set us up with two opposing choices. On the one hand, we could take the position Block does and say that the demand for money and the threat of telling secrets are two separate acts, and since each is licit on its own, we may reject the notion that combined they become criminal.15 But, on the other hand, what if we were to construct the situation such that the demand for money and the threat of telling secrets was not two separate acts, but rather just one complex act? It would seem that we might have reason at that point to call the action criminal—after all, payment is being extracted by threat and some legal theorists even within libertarian circles oppose blackmail.16 The case of entrapment, too, could appeal to this “complex action” loophole, in effect making it as though the boss threw his worker into the trap. However, this entire process of deciding what actions count as separate or complex seems to drift dangerously into the realm of ad hoc, arbitrary judgments, and as such seems especially unattractive when there is a solid, rigorous alternative such as the one presented by Hoppe and Reinach.

Hoppe’s case for entrapment being a form of aggression thus is not in contention with the non-aggression principle and should be unobjectionable to the Rothbardian. In many ways, the insights of Reinach and Hoppe up to this point are a significant improvement on the traditional understanding of the non-aggression principle. Despite the thrust of the present paper being towards a less prohibitive interpretation, it cannot be denied that the inclusion of entrapment as an aggressive act is an important and positive contribution to the current understanding of libertarian law. However, Hoppe’s other suggested inclusions, namely, failed attempts and incitement, are cause for concern.

Hoppe’s position that failed attempts at aggression should be cause for liability seems odd on its surface for a Rothbardian and somewhat inconsistent with his earlier positions. To quote his example directly:

A wants to kill his wife, B. He buys deadly poison from the pharmacist, and regularly adds it to   B’s tea. However, the pharmacist has made a mistake. He did not sell A poison but something entirely harmless. B dies in an unrelated car crash. The pharmacist discovers his error and the entire case unravels. Should A be held liable or go free (B’s heirs are suing A)?

Reinach would find A liable. There is intent (and hence fault) and there is (failed) causality. A performs a series of actions that he believes to be and which objectively are suited to bringing about the desired result. It is only because of an incidental (accidental) causal event (the pharmacist’s error) that the result does not occur as desired. (Hoppe 2004, pp. 93–94)

However, the difficulty arises initially when we attempt to consider what the proportional restitution and punishment for the criminal in this case should be. In most libertarian scholarship on the issue of restitution17 the focus is on making the victim whole again, or as close as possible, so what would the criminal be liable for? Since the criminal has done no harm, the just and proportional restitution would be zero (or, perhaps, zero times two,18 which, of course, remains zero). As to the issue of punishment, proportionality would again make it difficult to find any non-zero punishment19 for the criminal in this case—the most that could be done is frightening the criminal, assuming that we take Rothbard’s position on the issue of failed attempts,20 and this hardly seems to be much more than zero. Thus, if the liability of the criminal is zero (or at most, zero plus a frightening of the criminal), they will not be taken to any market court where costs are non-negative in prosecuting.21

We might not even be persuaded to think that A in this case is liable at all, even if the liability is zero, for it is intuitively hard to accept the premise that a failed action can create a liability. Hoppe’s defense of this position is with an analogy to homesteading:

We do not require that an act of original appropriation (homesteading) be successful in order to find that it has taken place and to determine ownership. For example, A clears the underbrush from a previously unowned piece of woodland in order to create a park. However, in doing so he accidentally burns down all trees. A’s action was unsuccessful. This is not the outcome he wanted. Is he nonetheless the owner of the burned forest? It seems so. However, if there are unsuccessful attempts of appropriation which count nonetheless as acts of appropriation, why should there not also be unsuccessful attempts of aggression which nonetheless count as aggression? (Hoppe 2004, p. 94)

In this case, the analogy between a “failed” attempt at homesteading on the one hand and a failed attempt at aggression on the other seems to not be warranted. It would seem that, in fact, A does not fail at homesteading the forest, that is, changing it significantly from its natural state and thereby mixing his labor with the land.22 Indeed, a true failed attempt at homesteading would not be homesteading at all. An example of this might be simply placing a flagpole in the ground and claiming an entire continent or dropped flags from a plane on an area, both would be failed attempts at homesteading and, thus, not homesteading.23 This is the more accurate analogy to a failed attempt at aggression. If no harm is done, how can we possibly consider this to be aggression? Even by Reinach’s criteria, this would not seem to qualify. While A believes what he is using will result in the death of B, this is objectively not the case due to the mistake of the pharmacist. It is analogous to the example of A praying for B’s death—A may believe this to be effective, but it objectively is not, and thus cannot be considered a legitimate causality.24

One final objection to Hoppe paper from a Rothbardian perspective would be his insinuation that “incitement” can be cause for liability.25 This stands in stark contrast to the position taken by Rothbard.26 However, very little argument is given by Hoppe in favor of this position; indeed, it is somewhat unclear as to what exactly Hoppe means by “incitement.” He says this of it:

Surely, if A tells B that he wished C were dead, and B kills C we would not hold A liable. But would we do the same if A paid B, or if A and B were members of an organized gang of which A               were the gang’s leader, and B killed C? (Hoppe 2004, p. 93)

This gives us some reason to believe that Hoppe is strictly speaking of incitement by monetary payment and incitement by coercion, both of which are considered illegitimate by Block.27 I will deal fully with these cases in the third section of this paper, but suffice to say that Hoppe would not be in disagreement with what is generally considered the Rothbardian view on the matter if he objects only to these two types of incitement.28

II. Stephan Kinsella & Patrick Tinsley

Like Hoppe’s paper, “Causation and Aggression” by Kinsella & Tinsley29 is relatively modest in its suggested expansion of the non-aggression principle. As mentioned above, the focus of this paper seems to be on the issue of incitement and its legality under a libertarian law code. The uniqueness of this article, however, is that it takes a decidedly different tact than both Hoppe and van Dun in critiquing the Rothbardian position, using a means-ends analysis and relying heavily on the concept of “human means.”30

It is difficult to disagree with the idea that humans can be used as means; indeed, the authors correctly explain that, “there is no reason that other humans cannot also be one’s means. What else does it mean to ‘employ’ a worker, or to cooperate with others to produce wealth?”31 They further seek to buttress their point with regard to a criminal action:

A terrorist builds a letter-bomb and mails it to his intended victim via courier. The courier has no idea that the package he is delivering contains a lethal device. When the addressee dies in an explosion after he opens the package, whom should we hold responsible? The obvious answer is: the terrorist. Why not the courier? After all, the courier is causally connected to the killing. But because he did not know he was carrying a bomb, he did not have the intent to aggress against the victim. Instead, he was connected to the killing only as a means. When the bomb exploded, it was the terrorist’s action, not the courier’s, that was completed. The courier simply handed over a      letter. The terrorist, by contrast, intentionally used means—the bomb materials, but also the unwitting courier—to cause his victim’s death. (Kinsella & Tinsley 2004, p. 102)

In this case, it seems undeniable that Kinsella is correct in removing liability from the courier—there is no intent (fault) and thus he cannot be held liable.32 However, a significant difference between this situation and one of incitement can be quickly noticed—in this case, the courier has no idea that his actions are involved in this criminal action, rather, he is simply following his normal routine without being aware of the results of his actions.

Initially, this view may be assumed to be overly broad and implicate persons who are clearly not the cause of a criminal action. Kinsella saves it from this objection by offering a method of differentiating between the possible causers of a criminal action:

The law has long recognized that one accused of a crime or tort is not responsible if the damage was really caused by an “intervening act” that breaks the chain of causal connection” between the actions of the accused and the damage that occurred. The idea is that the intervening act is the true cause of the harm caused. (Kinsella & Tinsley 2004, p. 103)

Kinsella quickly qualifies this criterion, in order to stop his position being misconstrued as opposing the inclusion of incitement as a crime due to intervening acts of free will which “break” the causal chain.33 He explains,

Even the law recognizes that an intervening force only breaks the chain causal connection when it is unforeseeable. As the Restatement of Torts provides, “The intervention of a force which is a normal consequence of a situation created by the actor’s . . . conduct is not a superseding cause of harm which such conduct has been a substantial factor in bringing about.” (Kinsella & Tinsley 2004, p. 103)

So, the framework proposed now puts all actions which cause the physical invasion of person or property, without the intervention of unforeseen willful acts, into the category of prohibited actions under libertarian law. This, of course, is not limited to the use of innocent human means, but rather explicitly includes “criminal conspiracies.”34

To this point, I can offer no objections to Kinsella’s conclusions regarding liability in the two cases he has given (the unknowing courier and the criminal conspiracy). His focus on causation35 is also a welcome addition to the previously analyzed paper by Hoppe, which focused more heavily on fault, and together the two analyses make for a stronger and more rigorous application of the non-aggression principle. However, I believe Kinsella goes astray when he delves into issues beyond this basic core.

On the issue of incitement, it is not difficult to ascertain the view Rothbard took. In For a New Liberty he rather explicitly explained his position:

  What, for example, of “incitement to riot,” in which the speaker is held guilty of a crime for whipping up a mob, which then riots and commits various actions and crimes against person and property? In our view, “incitement” can only be considered a crime if we deny every man’s freedom of will and of choice, and assume that if A tells B and C: “You and him go ahead and riot!” that somehow B and C are then helplessly determined to proceed and commit the wrongful                act. But the libertarian, who believes in freedom of the will, must insist that while it might be      immoral or unfortunate for A to advocate a riot, that this is strictly in the realm of advocacy and should not be subject to legal penalty. (Rothbard 1978, pp. 93–94)

However, Kinsella believes that a revision of this view is in order.36 He seeks to provide a variety of examples which will prove the strength of his position and these shall be analyzed below.

The first and most in depth example given by Kinsella is the following:

A malcontent, A, purchases a remote-controlled tank. With the remote control he can steer the tank and fire its cannon. He directs the tank to blow down the walls of a neighbor’s house, destroying the house and killing the neighbor. No one would deny that A is the cause of the killing and is guilty of murder and trespass. However, after the rampage, a hatch opens in the tank, and an   evil midget jumps out. It turns out, you see, that the midget could see on a screen which buttons were pressed on the remote control, and he would operate the tank accordingly. We submit that A is equally liable in both cases. From his point of view, the tank was a “black box” that he used to attain his end, regardless of whether there was a human will somewhere in the chain of causation. (Of course, the evil midget is also liable.) (Kinsella & Tinsley 2004, p. 104)

In the first part of the example, where A controls the tank entirely, it is obvious that A is the criminal and liable for the damage of his rampage. But when we introduce the midget, it seems that there are three possibilities as far as liability is concerned. (1) In the first case, the midget is unaware of what he’s doing, when the buttons come up on the screen to tell him how to operate the tank and he does so without any knowledge of what the results of his actions will be. Here, we have a situation analogous to the courier who carries the terrorist’s letter bomb to the target. It is obvious that only A can be liable and the midget is without fault. (2) In the second case, the midget is aware of what he’s doing, but is unable to operate the tank without the help of A. Perhaps A serves as a spotter or the midget does not know what sequence of buttons must be pressed to do what he wishes with the tank. In this case, we have an analogy to the criminal conspiracy, where neither A nor the midget can achieve their goal without the other. I would, in this case, agree with Kinsella’s conclusion that both A and the midget are liable for the damages. (3) In this third case, the midget is both aware of what he is doing and can operate the tank without any assistance from A. It would seem that in this case, all A is doing by pushing the buttons is suggesting, hoping, pleading with the midget to take certain actions. However, mere hopes and wishes cannot be cause for liability.37

Do A’s actions in this case count as mere hopes and wishes in this case, or is he a part of a criminal conspiracy with the midget? Kinsella would argue that it is the latter—the speech acts of A can be interpreted as causation and intent on the part of A. On the other hand, it seems that A’s actions here are no different from that of any statist when it comes to the actions of the State. Statists, by definition, support some sort of aggression on the part of the State against its citizens, no matter if they are minarchists or Bolsheviks. To that extent, they can be compared with A in this scenario—they are not necessary for the actions of the State38 and the State-criminals are aware of their actions. So, unless we are prepared to indict the vast majority of the world’s population, it seems that A’s actions cannot be considered criminal simply because he has pleaded with the midget to attack certain targets.

A further problem with Kinsella’s conclusion exists, namely, that it seems to be in tension with his view on punishment.39 Now, this is not to say that his estoppel approach in any way prevents incitement from being seen as a crime, far from it, but merely that it makes things rather difficult to carry out. As he explains in “Punishment and Proportionality”:

Just because aggressors can legitimately be punished does not necessarily mean that all concerns about proportionality may be dropped. At first blush, if we focus only on the initiation of force itself, it would seem that a victim could make a prima facie case that, since the aggressor initiated           force—no matter how trivial—the victim is entitled to use force against the aggressor, even including execution of the aggressor. Suppose A uninvitedly slaps B lightly on the cheek for a rude remark. Is B entitled to execute A in return? A, it is true, has initiated force, so how can he complain if force is to be used against him? But A is not estopped from objecting to being killed. A may perfectly consistently object to being killed, since he may maintain that it is wrong to kill. This in itself is not inconsistent with A’s implicit view that it is legitimate to lightly slap others. By sanctioning slapping, A does not necessarily claim that killing is proper, because usually (and in this example) there is nothing about slapping that rises to the level of killing. (Kinsella 1996, p. 64)

Given this, what could be done against the inciter, A? At most, it would seem that the victim, B, would be entitled to engage in incitement against A. What more has A done to B, except this? B would certainly be justified in doing more to those who were incited by A, those that actually did damage to his property and thus are estopped from objecting to B extracting compensation from them. However, it would seem that Kinsella’s position on punishment deeply undermines his view that incitement should be deemed criminal if all that could justly be done in response is incitement against the inciter.

At this point, I am obliged to provide an alternative standard for dealing with human means.40 Rather than inquiring as to whether the actions of the “intervening wills” are “unforeseen,” I propose that instead we consider two criteria: 1) are the human means aware of what they are doing? and 2) are the employers of the human means necessary for the actions which are taken? This standard, in this author’s earnest opinion, deals satisfactorily with the issue and its conclusions have intuitive appeal. In the case of the courier who carries the bomb to the target, the first criterion eliminates the responsibility from the unwitting postal worker and places it, correctly, on the terrorist. In the case of the bank heist, the second criterion implicates those who were not involved in the stealing of money qua cracking the safe and removing the loot, but still involved in the robbery itself. In the case of the soapbox orator “inciting” a mob to riot, this standard places the blame squarely on the shoulders of the rioters, as they were both aware of what they were doing and the inciter was entirely unnecessary for them to commit these actions.

Kinsella goes on in his paper to criticize the inconsistency of some of the defenders of the Rothbardian position for making ad hoc exceptions to the theories which they support.41 This will be discussed in greater detail below, but suffice to say for now that I am entirely in agreement with Kinsella on this point and can bring no substantive objection to his criticisms.

In sum, Kinsella’s paper has many beneficial aspects—most importantly, a focused look at causation which provides a more rigorous method of looking at liability, especially in cases where human means are employed. However, it is this author’s opinion that Kinsella’s particular method of analyzing situations, i.e., to consider willful actions only if they are “unforeseen,” should be revised.

III. Walter Block

Moving now from defending the Rothbardian position, this paper shall attempt to improve upon its current application. Arguably the most stringent and radical adherent to Rothbard’s legal philosophy is Walter Block, whose writings on a multitude of issues are often used interchangeably with Rothbard’s own writings. And yet, it is the goal of this paper to show that Rothbardians should be even more radical than the esteemed Block when it comes to issues of causality and responsibility. This section shall deal with Block’s two reservations regarding incitement (incitement-by-monetary-payment and incitement-by-extortion42) and why a truly consistent Rothbardian must reject the liability of the inciter in both cases.

For the first case, incitement-by-monetary-payment, we shall take the paradigm example of the hitman contract to analyze the actions of this type of inciter. This is, for good reason, precisely the situation Block examines. His argument is the following43:

It is [not my] view that he who hires a killer to murder an innocent person should not be found guilty; after all, the initiator of murder-for-hire did not himself pull the trigger. [In my view,] in subsidizing evil, one becomes a part of it. Paying for a murder-by-hire is not merely an act of free speech; it is part and parcel of gangster-like activity. (Block 2004a, p. 17)

At first glance, it seems rather odd that Block would take this view, given his position on incitement-by-words. This initial feeling is caused by the extremely deterministic outlook Block has on the hitman, an outlook that is most certainly not applied to rioters. The implicit assumption here seems to be that hitmen not only must follow through on any agreement to kill someone, but furthermore that they must accept any and all requests. But what if we were to apply the same to rioters? If rioters had no choice but to do as the soapbox orator said, certainly Block would recognize incitement in all cases as a crime. Indeed, both Block and Rothbard stress the importance of the rioters’ free will in their reasoning for opposing liability for inciters. In explaining why a rioter is different from a bullet, Block explains:

[T]he rioter is a human being, presumably with free will; no one could say the same of a piece of lead.

Third, there are many cases in which an inciter incites until his lungs give out, and no subsequent                riot takes place, further attesting to the distinction between free will and inanimate objects that mars Van Dun’s analogy. (Block 2004a, p. 16)

Rothbard deals similarly with this topic:

Since every man is free to adopt or not adopt any course of action he wishes, we cannot say that in some way Green determined the members of the mob to their criminal activities; we cannot make      him, because of his exhortation, at all responsible for their crimes. (Rothbard 1998, p. 81)

Given this, why should we assume that the hitman’s actions are determined? Why should we assume that by the mere act of offering money the inciter is able to take control of the hitman’s body and make him do the dirty work? The same argument which underpins the Blockian and Rothbardian support for the right to incite-by-words can be used to bolster the right to incite-by-monetary-payment.

This view also is in tension with the Austrian theory of value, namely, that it is entirely subjective.44 That being the case, we have no reason to condemn money payments while turning a blind eye to psychic value.45 Money payments are not necessarily more valuable than the multitude of psychic goods that could be used to incite someone into action, be it the desire to please a popular orator or even just the wish to be a part of a group. Neither of these is considered criminal motivators by Block, yet money payments are. Kinsella rightly pointed out:

But paying someone is simply one means of inducing them to do something to obtain money that they subjectively value. They could be induced or persuaded by giving them other things they value, such as gratitude. (Kinsella & Tinsley 2004, p. 106n)

The standard for judging liability with regards to human means which was proposed above in Section II can be used in the case of the hitman contract as well, meeting this challenge. With regards to awareness, we can easily see that the hitman is aware of what he is doing. Unless fooled into thinking the target was a criminal who deserves death, there can be no denying that the hitman would be liable. With regard to the necessity of the inciter, it would seem that the hitman has the ability and means to engage in the crime without the help of the inciter. Indeed, unless the inciter plays some other role—if he helps hide the hitman from the authorities, drives the getaway car, picks the lock on the target’s door, or something actually involved in the crime itself, then and only then would he have been necessary for the hitman to carry out the crime.

Let us assume this was not the case, that one could be part of a criminal conspiracy without having any part in the actual commission of the crime (including avoiding detection by the police). What would be the results of such a stance? It would seem that, beyond merely implicating those who pay hitmen, any and all persons who engaged in “hate speech” would become criminals. For instance, take the case of a book which calls for the elimination of all redheads. If someone reads that, then goes out and kills a redhead, are we to blame the author of the book for the killing? What if the author of the book wrote it specifically in the hopes that people would indeed start killing redheads? By the logic presented for making the inciter-by-payment a member of a criminal conspiracy with the hitman, it would seem that the author would become a criminal. But, we can be quite certain that Block opposes such a conclusion,46 and is therefore left with the position that the hitman, fully aware of the criminality of his actions, and acting entirely independently in his commission of the crime, retains full and total liability.

Moving to the second case, we have that of incitement-by-extortion, an example would be a criminal demanding that you steal something for him with the threat that if you refuse, he will kill you. The situation here is much trickier, but, in this author’s opinion, it does not differ substantially from the previous cases. It is somewhat dubious taking Block’s comments on incitement-by-extortion by themselves as to what he actually means. He writes:

According to Van Dun’s interpretation of my viewpoint, they [Hitler, Stalin, et al] would therefore be “guilty” of no more than exercising their free speech rights, and should be considered innocent of all wrongdoing.

However, Van Dun reckons in the absence of threats. To reiterate, the libertarian legal code proscribes not only invasive acts, but also intimidation. Hitler, Stalin, et al. were not merely engaging in their free speech rights. Rather, they were issuing orders to their subordinates to maim and kill innocent people. Implicit in these commands was the threat that if they were not obeyed, those who failed to carry out these orders would be summarily dealt with. (Block 2004a, p. 15)

When Block says that dictators are not “innocent of all wrongdoing” does he mean that they are liable for the crimes of their subordinates or that they are liable for the crime of threatening their subordinates? A somewhat clearer position is taken in his second round with van Dun, but it could still be interpreted both ways:

“[O]rdering one’s followers to commit murder.” This case alone on the above list, as I have been at great pains to elaborate, would constitute guilt under the libertarian order as I understand it…

A gang leader does not merely incite his followers to criminal behavior, he orders them to do it, or threatens that if they do not, they will be visited with physical sanctions. Under the libertarian legal code, he would be guilty. (Block 2004b, p. 67)

However, given the context in which Block is making these arguments, it would seem that he means to say that incitement-by-extortion does indeed make the inciter liable for the actions of the incited.

Intuitively, this thesis seems very reasonable, after all, the inciter, A, is “forcing” the incited, B, to do his bidding. Yet, this is not entirely accurate. The situation is not that A is grabbing B’s hand and physically forcing him to pull the trigger on an innocent victim. This would clearly be a case where A would be liable. Rather, A is asking B to trade, namely, to trade a rights violation against himself (e.g., A shooting B) for a rights violation against an innocent third party (e.g., B shooting C). Certainly this is no voluntary trade, of course, but nonetheless, it is a trade of one rights violation for another. How can we possibly deem B to be justified in doing this? A’s contribution to this situation is notable only to B, those that B aggresses against have no recourse with A for B’s rights violations. The only person whose rights are violated by A in this scenario is B, and this can be shown if we consider the criteria proposed in Section II. First, is B aware of what he is doing, that is, does he realize that his actions are violating the rights of C? As with the hitman, the answer seems to clearly be affirmative, unless B has been fooled into thinking that C is a criminal deserving of death. Now, is A necessary for the commission of the crime? Here, as with the hitman, the answer seems to be “no.” Unless A is engaged in more than simply inciting B to kill C, he would fail this test. If A had kicked in C’s door, or provided B shelter, a “safe house” so to speak, or anything of that nature, then we could put A down as necessary and therefore liable. But, barring those circumstances, we have no cause for placing liability for B’s actions onto A, at least as far as C is concerned.

The argument against this position might run essentially as follows: because A has threatened B into action he would not have taken on his own, he becomes necessary to the commission of the crime that B carries out, thus making A part of a criminal conspiracy with B to violate C’s rights. But this is a gross stretching of the necessary-criterion that was previously proposed, which was in no way an attempt to suggest that anyone who influenced the criminal to commit the crime would be considered guilty as part of a criminal conspiracy (such an interpretation might lead to parents and schoolteachers being locked up for the crimes of their children, who they undoubtedly influenced). Rather, it was to include those who played necessary roles as a part of the crime, such as the one who drives the get-away car, yet does not directly engage in rights-violations. The role of A in this case, at least as far as C is concerned, is merely as an influence on the criminal actions of B, but because A is not in any sense an actual part of the crime itself, he cannot be held liable the criminal actions against C (though, he would of course be liable for his aggression against B).

In this, we may make an analogy to trademark infringement and its associated liabilities. As Kinsella explains,47 the only ones who can legitimately sue trademark infringers are those who are actually defrauded by them, i.e., the customers. Those whose trademarks are infringed upon have no recourse, for their rights were not violated by the infringers. So it is with those who have their property damaged by B, they may only bring suit against him, not against A, who has nothing to do with them. B, on the other hand, does have legal recourse against A, for threatening him.48 In this way, we might say that incitement-by-extortion is indeed illicit, but that does not imply that third parties (i.e., persons other than the target of the threat and his agents) are justified in extracting compensation from or punishing the inciter for his actions, in other words it is merely the extortion portion of the act which creates liability, not the incitement.49

Taking this all into account, it would seem to this author that the traditional Rothbardian position, as defended by Walter Block, must be revised to include all types of incitement, be they by words, money, or threats, from creating a liability on the inciter. In order to maintain its appealing consistency, these revisions, no matter how counterintuitive they may initially seem, must be made.

IV. Conclusion

Though this paper has relied entirely upon the non-aggression principle in vacuo, it is not the opinion of this author that any society, libertarian or otherwise, would depend entirely on this principle. The non-aggression principle is best seen not as a guide to a complete legal system, but rather as a basis for judging legal systems, to see if they meet the requirements to be deemed “libertarian” or “moral.”

But, in a world of private property, it is far from required that any legal system limit itself to purely applications of the non-aggression principle. Indeed, Rothbard would be quick to maintain, his was only a political philosophy, which dealt exclusively with the just use of force.50 However, just because one has the right to do something, does not mean they have the right to do it anywhere.51 Property owners have full discretion to forbid whatever sorts of activities they wish, be it drugs, prostitution, or, if they choose, incitement. Any action taken on someone else’s property is subject to whatever rules the relevant property owner has devised. Thus, when libertarians speak of the right to do something, it is only in the context of taking this action with the permission of the relevant property owner.

Would a free society react with such strong sanction against inciters that they voluntarily choose to compensate persons who are harmed by those who follow them?52 Perhaps, and this could be said for a great number of activities ibertarians deem to be non-criminal, such as libel, blackmail, or racial discrimination. We cannot possibly know the actions of individuals in the nonexistent free society. Therefore, again, it must be stressed that such issues where the non-aggression principle is invoked in vacuo, the issue is only one of the justified use of force, and only in the context of the actor having permission from the relevant property owner. Such investigations cannot (and should not) make any guarantees regarding the shape of a free society.

  • *. *Matt Mortellaro can be reached at
  • 1. See Frank van Dun, “Against Libertarian Legalism: A Comment on Kinsella and Block,” Journal of Libertarian Studies 17, no. 3 (Summer 2003); van Dun, “Natural Law and the Jurisprudence of Freedom,” Journal of Libertarian Studies 18, no. 2 (Spring 2004); Hans-Hermann Hoppe, “Property, Causality, and Liability,” Quarterly Journal of Austrian Economics 7, no. 4 (Winter 2004) [cited in the text parenthetically as “Hoppe 2004”; and Stephan Kinsella & Patrick Tinsley, “Causation and Aggression,” Quarterly Journal of Austrian Economics 7, no. 4 (Winter 2004) [cited in the text parenthetically as “Kinsella & Tinsley 2004”].
  • 2. See Murray N. Rothbard, The Ethics of Liberty, New York: New York University Press 1998, pp. 51–52, 77-82; and Rothbard, For a New Liberty: The Libertarian Manifesto, New York: Collier Books 1978, pp. 22, 93-94 [cited in the text parenthetically as “Rothbard 1978”].
  • 3. See Walter Block, “Toward a Libertarian Theory of Blackmail,” Journal of Libertarian Studies 15, no. 2 (Spring 2001). In addition to this, Block has written several replies to the criticisms lodged by van Dun, see Block, “Reply to ‘Against Libertarian Legalism’ by Frank van Dun,”Journal of Libertarian Studies 18, no. 2 (Spring 2004) [cited in the text parenthetically as “Block 2004a”]; and Block, “Reply to Frank van Dun’s ‘Natural Law and the Jurisprudence of Freedom’,” Journal of Libertarian Studies 18, no. 2 (Spring 2004) [cited in the text parenthetically as “Block 2004b”].
  • 4. Hoppe’s critique moves in both directions, both expanding in some areas and contracting in others, but the main thrust of his critique seems to be towards an expansion.
  • 5. Hoppe, “Property, Causality, and Liability,” p. 89.
  • 6. For a discussion of this concept of action vs. mere reflex, see Ludwig von Mises, Human Action: A Treatise on Economics, San Francisco: Fox & Wilkes 1996, pp. 11–17.
  • 7. Ibid, pp. 89–90.
  • 8. Hoppe’s position here have direct relevance to the arguments of other libertarian legal theorists. One such example is that of Frank van Dun who argues that liability may be imposed without any action (or negligence) on the part of the “criminal,” using the case of a dog destroying a flowerbed as an example of this (van Dun 2004, p. 41n). Hoppe’s position is clearly articulated in opposition to such a notion: “No one is liable for ‘accidents’ involving his person and property. Instead, the risk of accidents and the insurance against them must be assumed individually (by each person and property owner for himself). People can be held liable only for their actions, whether intentional or negligent (but not for accidents involving them). Actions, however, involve both “objective” (external) and “subjective” (internal) elements. Hence, the exclusive inspection of physical events can never be considered sufficient in determining liability (there must be fault, too, and one can only speak of fault if an event is caused by an action)” (Hoppe 2004, p. 90).
  • 9. Ibid, p. 90.
  • 10. Ibid, pp. 91–94.
  • 11. See Block, “Toward a Libertarian Theory of Blackmail,” pp. 56–57. Though the contention could be made for many different names (e.g., “Fletcher’s Paradox”), Block is the one who maintains that it is indeed logically inconsistent for two acts which, separately, are non-invasive to be illicit and is the commentator most relevant to the current discussion. Thus, it seems sensible to use the proposed name as shorthand in the context of this paper.
  • 12. For examples of these conclusions, see Block, Defending the Undefendable: The Pimp, Prostitute, Scab, Slumlord, Libeler, Moneylender, and Other Scapegoats in the Rogue’s Gallery of American Society, San Francisco: Fox & Wilkes 1991; Rothbard, The Ethics of Liberty, pp. 51–153; and Rothbard, For a New Liberty, pp. 78–302.
  • 13. Rothbard explains the necessary conditions for a threat to be considered aggression in The Ethics of Liberty, where he said, “It is important to insist, however, that the threat of aggression be palpable, immediate, and direct; in short, that it be embodied in the initiation of an overt act” (p. 78). It cannot be argued that a threat unknown to the person threatened is either palpable or overt, and thus we must deny its status as an aggressive action.
  • 14. Making more in-depth and complete criticisms of Block’s Paradox as an implication of the non-aggression principle is outside of the scope of this paper and was simply used as an illustration of the point that Hoppe’s critique is not necessarily in tension with the core of the Rothbardian position.
  • 15. Block, “Toward a Libertarian Theory of Blackmail,” pp. 55–56.
  • 16. See van Dun, “Against Libertarian Legalism,” pp. 72–73; and van Dun, “Natural Law and the Jurisprudence of Freedom,” pp. 34–36.
  • 17. An illustrative (and far from exhaustive) set of examples: Assessing the Criminal: Restitution, Retribution, and the Legal Process, Randy E. Barnett and John Hagel III, eds., Cambridge: Ballinger Publishing Co. 1977; Peter J. Ferrara, “Retribution and Restitution: A Synthesis,” Journal of Libertarian Studies 6, no. 2 (Spring 1982); Robert Nozick, Anarchy, State, and Utopia, New York: Basic Books, 1974; and Rothbard, The Ethics of Liberty.
  • 18. See Kinsella, “Punishment and Proportionality: The Estoppel Approach,” Journal of Libertarian Studies 12, no. 1 (Spring 1996), pp. 64–65; and Rothbard, The Ethics of Liberty, p. 88. Both offer strong reasons for supporting the principle of doubling the harm to the victim when assessing the liability of the criminal.
  • 19. It has been suggested by one reviewer that perhaps proportionality would allow B (or his agents) to attempt to kill A, putting A in the same situation that he placed B in—his fate dependent on the interference of a third party. However, it seems that this would be a grossly excessive reaction. To make an analogy, if a thief had tried to steal a painting worth one million dollars, but accidentally mistook a ten dollar copy of the painting as the real thing and stole that instead, the owner would not be justified in extracting one million dollars from the thief. Surely such a punishment by the owner would be seen as disproportionate to the crime which in actual fact occurred, the theft of a ten dollar item, regardless of what the thief had hoped to do.
  • 20. “[E]ven if the attempted crime created no invasion of property per se, if the attempted battery or murder became known to the victim, the resulting creation of fear in the victim would be prosecutable as an assault.” From Rothbard, “Law, Property Rights, and Air Pollution,” as reprinted by the Ludwig von Mises Institute, 2002. It might be noted that although Hoppe seems to suggest in his paper that this is backtracking on the part of Rothbard, such a statement is not inconsistent with Rothbard’s other views regarding liability. In The Ethics of Liberty, he says the following: “invasion may include [in addition to] actual physical aggression: intimidation, or a direct threat of physical violence… and this is equivalent to the invasion itself” (pp. 77–78). Further, Rothbard qualifies his statement regarding failed attempts at aggression in a footnote, citing Barnett in arguing that no liability exists in the case of the failed attempt not in itself being a violation of rights and the attempt being unknown to the victim. The example which Hoppe gives seems to fit both criteria, and thus Rothbard would indeed oppose the would-be criminal’s prosecution.
  • 21. Unless there is some value the victim places on seeing the criminal declared guilty in court which outweighs the cost of prosecution, there would be no reason for the case to be brought at all. Even in the case of prosecuting fees being shouldered by the guilty party, the plaintiff would still be unlikely to bring the case to court, because the intangible costs such as time would still exist, again, unless the goal is simply to spite the criminal with the costs of the trial (though, it seems likely the criminal would simply plead guilty and avoid this in that case). On this point, see Mises, Human Action: A Treatise on Economics, p. 13.
  • 22. See John Locke, “An Essay Concerning the True Original Extent and End of Civil Government,” Two Treatises on Government, London: Awnsham and Churchill 1689, Chapter V “Of Property,” § 26.
  • 23. As Rothbard explains in The Ethics of Liberty, “Suppose Crusoe had landed not on a small island, but on a new and virgin continent, and that, standing on the shore, he had claimed ‘ownership’ of the entire new continent by virtue of his prior discovery. This assertion would be sheer empty vainglory, so long as no one else came upon the continent. For the natural fact is that his true property—his actual control over material goods—would extend only so far as his actual labor brought them into production” (p. 34).
  • 24. Hoppe, “Property, Causality, and Liability,” p. 92.
  • 25. Ibid, pp. 93–94.
  • 26. See Rothbard, The Ethics of Liberty, p. 81; and Rothbard, For a New Liberty, pp. 93–94.
  • 27. See Block, “Reply to ‘Against Libertarian Legalism’ by Frank van Dun,” pp. 15 and 17.
  • 28. Though, as will be explained below, picking out two forms of incitement as illegitimate while not opposing the rest seems rather ad hoc, and, this paper will argue, flawed.
  • 29. Henceforth, for the sake of brevity, I will refer only to “Kinsella” in the text, rather than “Kinsella & Tinsley.”
  • 30. Kinsella & Tinsley, “Causation and Aggression,” pp. 101–105.
  • 31. Ibid, p. 102.
  • 32. See Hoppe, “Property, Causality, and Liability.” A possible exception to this would be if in the courier’s contract with the victim there existed a clause requiring all mail to be checked for safety, then this might rightfully be called negligence and implicate the courier.
  • 33. Kinsella laments, “Using ostensibly similar reasoning, some libertarians would maintain that in the case above, the intermediate person, since he has free will, performs ‘intervening acts’ that ‘break’ the chain of causal connection between the terrorist and the acts committed by the intermediate person…But this premise is untenable…the notion that the use of another human to achieve one’s goals absolves one of responsibility for those results [is clearly absurd]” (Kinsella & Tinsley 2004, p. 103).
  • 34. Kinsella & Tinsley, “Causation and Aggression,” pp. 103–104.
  • 35. Ibid, p. 104. This theme is present throughout his paper and he succinctly notes, “The key is causation.”
  • 36. Ibid, p. 107. “What we maintain is that the inciter is not off the hook just because the rioters had free will. The question to be answered is: was the mob the means of the inciter? Was the inciter a cause of the mob rioting, or of their ensuing havoc?”
  • 37. See Hoppe, “Property, Causality, and Liability,” p. 91. Hoppe specifically excludes mere hopes and wishes from his construction of liability through fault, with an example of an employer (A) who sends his employee (B) into the woods, hoping that lightning will strike B dead. Hoppe then explains the solution, Has A caused B’s death or injury? Should A be liable? With regard to causation, Reinach would answer yes: without A’s authorized order to B, B would not have been killed. However, Reinach would deny that A is liable, not because there is no causality, but because there is no intent or negligence on A’s part (there is just hope). Thus, it makes some sense that Kinsella, with his focus on causation, may have the opposite conclusion. However, it is important not to forget fault as a necessary prerequisite for any action to create a liability. Strangely, later in his paper (pp. 110–11), Kinsella notes this exact example favorably, though it seems to contradict his earlier position. He interprets it rather differently, however, which may be the reason that it appears to serve both sides of this discussion.
  • 38. Evidence of this lies in the fact of non-State criminal gangs, who are supported only by their members, manage to operate regardless of the opinions of the community they terrorize and even when battled by community-supported State forces.
  • 39. See Kinsella, “Punishment and Proportionality: The Estoppel Approach,” pp. 64–65.
  • 40. Ludwig von Mises wisely pointed out, “An ‘anti-something’ movement displays a purely negative attitude. It has no chance whatever to succeed. Its passionate diatribes virtually advertise the program that they attack. People must fight for something that they want to achieve, not simply reject an evil, however bad it may be.” Mises, The Anti-Capitalistic Mentality, Grove City: Libertarian Press, Inc 1994, p. 88.
  • 41. Kinsella & Tinsley, “Causation and Aggression,” pp. 105–106.
  • 42. See Block, “Reply to ‘Against Libertarian Legalism’ by Frank van Dun,” pp. 15, 17.
  • 43. The wording has been slightly adjusted due to the original comments being directed at van Dun, but the meaning remains unchanged.
  • 44. It is important not to confuse explanatory value-subjectivism and normative value subjectivism, however. For more on this, see Roderick T. Long, “Economics and Its Ethical Assumptions,” Mises Daily Article 5/20/2006,
  • 45. This is precisely the mistake Block rightly criticized in his tremendous critique of the Coasian view of property rights. See Block, “Coase and Demsetz on Private Property Rights,” Journal of Libertarian Studies 1, no. 2 (1977).
  • 46. See Block, “Reply to ‘Against Libertarian Legalism’ by Frank van Dun,” p. 14. Here he vigorously denies the validity of any “hate speech” restrictions.
  • 47. See Kinsella, “Against Intellectual Property,” Journal of Libertarian Studies 15, no. 2 (Spring 2001), pp. 43–44; Kinsella, “A Libertarian Theory of Contract: Title Transfer, Binding Promises, and Inalienability,” Journal of Libertarian Studies 17, no. 2 (Spring 2003); and Kinsella, “Reply to Van Dun: Non-Aggression and Title Transfer,” Journal of Libertarian Studies 18, no. 2 (Spring 2004).
  • 48. An interesting possibility for a libertarian legal system would be to award B sufficient compensation from A to pay off those whose property he was threatened into violating, plus additional damages for the trauma of being threatened into committing these violations. This would give us the solution which is intuitively desirable, without threatening the sanctity of libertarian law.
  • 49. To extend it a bit further we might say that the complex act of extortion is itself made up of one licit and one illicit act, the former being whatever demand the extorter makes, and the latter being the threat of force. On this point, see Block, “Toward a Libertarian Theory of Blackmail,” pp. 56–57.
  • 50. As Rothbard notes in The Ethics of Liberty, “The intention is to set forth a social ethic of liberty… that deals with the proper sphere of ‘politics,’ i.e., with violence and non-violence as modes of interpersonal relations” (p. 25).
  • 51. Hoppe explains this concept wonderfully in his introduction to The Ethics of Liberty with reference to the question of abortion: The right to have an abortion does not imply that one may have an abortion anywhere. In fact, there is nothing impermissible about private owners and associations discriminating against and punishing abortionists by every means other than physical punishment. Every household and property owner is free to prohibit an abortion on his own territory and may enter into a restrictive covenant with other owners for the same purpose. Moreover, every owner and every association of owners is free to fire or not to hire and to refuse to engage in any transaction whatsoever with an abortionist. It may indeed be the case that no civilized place can be found anywhere and that one must retire to the infamous ‘back alley’ to have an abortion. [p. xli]
  • 52. On ways in which a free society might enforce rules without the use of force, see Robert P. Murphy, Chaos Theory: Two Essays on Market Anarchism, New York: RJ Communications LLC 2002.


Libertarian Papers, Vol. 1 (2009), Article No. 24

Cite This Article

Mortellaro, Matt. “Causation and Responsibility: A New Direction. Libertarian Papers 1, no. 24 (2009): 1–21.

Shield icon interview