by Stephan Kinsella
I intend here to provide a short guide to the relevant literature followed by a limited response to Murphy’s and Callahan’s (hereinafer, MC) critique of Hans-Hermann Hoppe’s argumentation ethic.
Hoppe’s “argumentation ethic” defense of libertarian rights was first published, to my knowledge, in three articles in 1988: “On the Ultimate Justification of the Ethics of Private Property,” Liberty (September 1988); “The Justice of Economic Efficiency,” Austrian Economics Newsletter (Winter 1988); and a longer piece, “From the Economics of Laissez Faire to the Ethics of Libertarianism,” in: Walter Block & Llewellyn H. Rockwell, eds., Man, Economy, and Liberty: Essays in Honor of Murray N. Rothbard (Mises Institute, 1988). These were included as chapters 10, 9, and 8, respectively, of Hoppe’s book The Economics and Ethics of Private Property (Kluwer, 1993). The most definitive elaboration of Hoppe’s theory is found in “The Ethical Justification of Capitalism and Why Socialism Is Morally Indefensible,” chapter 7 of his monumental A Theory of Socialism and Capitalism (Kluwer 1989; more info; hereinafter referred to as TSC). This chapter is similar to the 1988 chapter in Man, Economy, and Liberty. These and other materials are available at Hoppe’s website.
Over the next year or three, Hoppe’s theory was intensely debated and commented on by several libertarians. Several replies and reviews, for example, were published in Liberty and elsewhere, by libertarians such as Murray Rothbard, David Gordon, Tibor Machan, David Friedman, Loren Lomasky, David Osterfeld, Sheldon Richman, Leland Yeager, David Ramsay Steele, Douglas Rasmussen, David Conway, and others. Hoppe responded to many of these pieces at length; these responses are reprinted in “Four Critical Replies,” Appendix to The Economics and Ethics of Private Property.
Several of the replies to Hoppe were unusually nasty and unfair. Some were shocked anyone would argue for “untrammeled anarchism” and others were turned off by the idea that libertarian rights could be rigorously proved. Others badly misconstrued Hoppe’s argument. Still others, like Rothbard, recognized that Hoppe’s theory was a revolutionary advance in libertarian theory, as have a growing number of adherents over the years. (One of the most recent papers addressing Hoppe’s view of rights is the fascinating new working paper Hopp(e)ing Onto New Ground: A Rothbardian Proposal for Thomistic Natural Law as the Basis for Hans-Hermann Hoppe’s Praxeological Defense of Private Property, by Jude Chua Soo Meng, which attempts to elaborate and extend Hoppe’s argumentation ethic and its relation to Rothbardian Austrian economics, praxeology, and libertarian theory.)
I have commented on Hoppe’s work in detail: first, in my review essay of The Economics and Ethics of Private Property, “The Undeniable Morality of Capitalism,” St. Mary’s Law Journal (1994); in my complementary “estoppel” theory of rights, e.g. “Punishment and Proportionality: The Estoppel Approach,” Journal of Libertarian Studies (Spring 1996); and in my survey article “New Rationalist Directions in Libertarian Rights Theory,” Journal of Libertarian Studies (Fall 1996). These and other papers can be found at my website.
To fully appreciate Hoppe’s argument and to fairly evaluate MC’s critique, I suggest reading at least the following, in this order: (1) chapters 1 and 2 of TSC (esp. pp. 5-6 & 8-18, discussing notions of scarcity, aggression, property, norms, and justification); (2) chapter 9 of TSC, “The Ethical Justification of Capitalism and Why Socialism Is Morally Indefensible” (esp. pp. 130-145); (3) “Four Critical Replies”; and (4) my survey article “New Rationalist Directions in Libertarian Rights Theory.” Then read MC’s piece, in view of my following comments.
The central question here is: does Hoppe’s theory establish that there are libertarian rights?
Scarce resources are those things over which there can be conflict; two or more individuals may want to use or control a given scarce resource at the same time, but only one of them can, because use by one excludes use by the other. Thus, as Hoppe explains, a theory of interpersonal ethics must be a theory of property rights, “a theory of the assignment of rights of exclusive control over scarce means” (TSC, p. 235 n.9; also p. 8). The purpose of rights is to specify which individual has the right to control a given scarce resource, so that conflicts may be avoided. The person who has the right to control a given scarce resource its owner is the person who is justified in using the resource, in excluding others, and in enforcing this exclusion against non-owners who would act in disregard of the owner’s property rights.
Everyone has at least an implicit view of rights. An aggressor, or at least one who would try to justify his aggression â€“ maintains that he is entitled to a given scarce resource “because” he is strong enough to take it. Others, such as socialists, believe that the state is entitled to the means of production “because” â€“ well, because they are the state, “because” capitalists “exploit” workers, and so on. Mainstream liberal-democratic types believe that, for example, the poor are entitled to property formerly owned by the not-poor, “because” the property is transferred from the latter to the former by means of a democratic process, which is “legitimate.” Everyone assigns each disputed scarce resource to some owner â€“ whether to a thief, the state, or a relatively-poor “needy” person â€“ for some reason.
The libertarian view of rights, as distinct from all others, says that we answer the question of who owns a scarce resource simply by asking who was its first user. In the case of bodies, the person himself is the owner of his body, as he was its first user. In the case of external scarce resources, the owner is the original appropriator (or someone he transferred it to). Thus, under libertarianism, an individual has (a) a right to the exclusive control of the scarce resource of his body, sometimes called “self-ownership”; and (b) a right to the exclusive control of other, previously-unowned scarce resources that are originally appropriated by the individual or by his ancestor-in-title.
So the question is, does Hoppe’s theory establish that the libertarian view of rights, as opposed to competing views, is the correct one?
Hoppe’s Theory: Let’s Try Again
I do not intend here to restate Hoppe’s entire argument, as I believe it has been adequately explicated and defended already by Hoppe, in the literature referenced above. And he has already replied to numerous criticisms, some of them similiar to MC’s. Instead, I will try to show, as simply as possible, why Hoppe succeeds. I’ll then address, in view of this, a few of MC’s concrete critiques, but it should be clear by this point why I think their criticism is off base.
Hoppe starts by noting that if any proposed theory of rights is going to be justified, it has to be justified in the course of an argument (discourse). I fail to see how MC can disagree with this without falling into contradiction. It follows that if any norms, ethics, facts, or rules of discourse are necessarily presupposed by participants in argumentation simply by virtue of arguing, then no theory that contradicts these presupposed facts or norms could ever be justified. By contrast, any proposed theory that is consistent with, indeed implied by, these presuppositions, would have to be seen as irrefutably justified. This type of reasoning is called the “apriori of communication and argumentation,” and was pioneered, to my knowledge, by German philosophers Jürgen Habermas (Hoppe’s PhD advisor) and Karl-Otto Apel (but was applied by them to reach non-libertarian results).
Again, I fail to see how MC can disagree with any of this, in general. Rather, the disagreement is over what norms are actually implicit in the activity of argumention that is, over what participants in discourse must presuppose to be true in order to participate in argument. Whatever these presuppositions are, they rule out of court any proposed norms inconsistent with them. And, any such normative presuppositions, or norms deduced from these presuppositions, would have to be considered to be ultimately and irrefutably justified, as their validity could never be coherently denied.
So let’s see what Hoppe contends. First, any norm proposed in argumentation is presumed to be universalizable. Writes Hoppe:
“Quite commonly it has been observed that argumentation implies that a proposition claims universal acceptability, or, should it be a norm proposal, that it is ‘universalizable.’ Applied to norm proposals, this is the idea, as formulated in the Golden Rule of ethics or in the Kantian Categorical Imperative, that only those norms can be justified that can be formulated as general principles which are valid for everyone without exception.” (TSC, p. 131).
In other words, any proposed norm that is, an attempted justification for a given action is not justified if it is not universalizable. This rule is presupposed by the very attempt to argumentatively justify something, because “argumentation implies that everyone who can understand an argument must in principle be able to be convinced of it simply because of its argumentative force.” Because the universalizability priniciple is an inherent feature of argumentation in general, “the universalization principle of ethics can now be understood and explained as grounded in the wider ‘apriori of communication and argumentation.'” (TSC, p. 131) I.e., no one can deny that only universalizable norms can be justified.
So, we have our first presupposition: that only universalizable ethics can be possible candidates for being justified. By the same token, so-called “particularizable” norms are not justifiable. However,
“the universalization principle only provides a purely formal criterion for morality. To be sure, checked against this criterion all proposals for valid norms which would specify different rules for different classes of people could be shown to have no legitimate claim of being universally acceptable as fair norms, unless the distinction between different classes of people were such that it implied no discrimination, but could instead be accepted as founded in the nature of things again by everyone. But while some norms might not pass the test of universalization, if enough attention were paid to their formulation, the most ridiculous norms, and what is of course even more relevant, even openly incompatible norms could easily and equally well pass it. For example, ‘everybody must get drunk on Sundays or be fined’ or ‘anyone who drinks alcohol will be punished’ are both rules that do not allow discrimination among groups of people and thus could both claim to satisfy the condition of universalization.” (TSC, p. 131-132; emphasis added)
But even though universalizability is merely a formal requirement, it does eliminate many proposed norms, such as those underlying most versions of socialism which amount to “I can hit you but you cannot hit me” particularizable rules.
“[T]he property theory implicit in socialism does not normally pass even the first decisive test (the necessary if not sufficient condition) required of rules of human conduct which claim to be morally justified or justifiable. This test, as formulated in the so-called golden rule or, similarly, in the Kantian categorical imperative, requires that in order to be just, a rule must be a general one applicable to every single person in the same way. The rule cannot specify different rights or obligations for different categories of people (one for the red-headed, and one for others, or one for women and a different one for men), as such a ‘particularistic’ rule, naturally, could never, not even in principle, be accepted as a fair rule by everyone. Particularistic rules, however, of the type ‘I can hit you, but you are not allowed to hit me,’ are at the very base of all practiced forms of socialism.” (TSC, p. 5.)
Thus universalizability acts as a first-level “filter” that weeds out all particularistic norms. This reduces the universe of possibly justified normative claims but does not finish the job since many incompatible and unethical norms could be reworded in universalizable ways.
It is for this reason that Hoppe next examines other, more substantive, presuppositions inherent in argument itself. These are then used in a second filtering process to reject additional proposed norms, those that are universalizable but incompatible with the other presuppositions of discourse. And, because some of these presuppositions turn out to be presupposed norms, Hoppe then shows that the libertarian conception of rights can be deduced from these presupposed norms and facts.
Substantive Facts and Norms Presupposed in Argumentation
The universalization principle filters out many possible norms, but many possible, mutually incompatible, and nonlibertarian candidates remain (“anyone who drinks alcohol will be punished”).
“However, there are other positive norms implied in argumentation aside from the universalization principle. In order to recognize them, it is only necessary to call three interrelated facts to attention. First, that argumentation is not only a cognitive but also a practical affair. Second, that argumentation, as a form of action, implies the use of the scarce resource of one’s body. And third, that argumentation is a conflict-free way of interacting.” (TSC, p. 132; emphasis added)
Participants in discourse cannot deny the existence of scarcity (discourse is a form of action, after all, and action implies scarce resources, in one’s body and in external objects) nor the possibility of conflict over these scarce resources. They also value the ability to participate in argument (they are engaging in it, after all) and thus its practical preconditions, namely the ability to actually use scarce resources in order to survive (for argumentation is not possible without survival). And because argumentation/discourse is cooperative, civilized, peaceful activity, and because “justifying means justifying without having to rely on coercion” (TSC, p. 133), participants in discourse necessarily value being able to use scarce resources in a conflict-free way. One adopting a civilized, peaceful stance and trying to justify a norm cannot coherently advocate non-peaceful norms. In fact, the very attempt to justify a resource allocation norm is an attempt to settle conflicts with regard to the use of that resource. Thus, a participant in discourse could never justify the proposition that there is no value to being able to use resources, or that conflict should not be avoided, or that cooperation and peacefulness are bad things. Valuing the avoidance of conflicts also presupposes the value of attempting to find rules that make conflict avoidance possible. I.e., property rules.
Accordingly, participants in discourse, in particular those seeking to justify proposed norms, implicitly recognize the value and legitimacy of assigning specified property owners to specified scarce resources for reasons that are universalizable and that make conflict-avoidance possible. However, property rights make conflict avoidance possible by establishing perceivable boundaries to property indicating the property’s borders and who the owner is, and by basing the assignment on universalizable rules that could be accepted as fair by all potential arguers. For this reason, the assignment of property rights has to be based on some objective link between the claimant and a particular resource.
What all this means is that anyone ever attempting to (argumentatively) justify any norm is already presupposing a host of norms and argumentative rules. The substantive presupposed norms rule out many proposed norms, even if they are universalizable. For example, a rule such as “no one should ever be able to use any scarce resource” could never be justified. It is incompatible with the speaker’s evident value for the ability to use scarce resources, because he has to use (and be able to use) the scarce resource of his body in order to engage in any activity, including argumentation. And he, or someone, had to be able to use other scarce resources such as food, shelter, etc., so that the arguers are alive and able to argue (remember, discourse is a practical affair, and requires the speakers to be alive, to have control of their bodies and their standing room, etc.).
In addition, a rule specifying that all resources, or even some resources, should have no owner at all, simply does not allocate ownership in the scarce resources at issue, i.e. it does not fulfill its function of conflict-avoidance. Unless property rights are allocated to someone, conflict over each scarce resource is possible; that is the nature of scarcity. (As a practical matter, most such rules also imply that, if a given resource should not be “owned,” then some person or agency is authorized to prevent others from using the thing. In which case the rule is in reality assigning ownership to the agency with control, and would need to be justified. For example, the public forests are said to be “unowned” but the federal government prevents homesteaders from moving in. Clearly here the federal government is asserting ownership. The necessity of justifying this cannot be avoided by the fiction that the property is not owned.)
There is no way any norm can be justified that does not seek to assign ownership of every scarce resource to particular owners, based on an objective link between the owner and the owned resource. No rule could ever be justified if it refrains from deciding who owns a particular resource, or if it specifies that no one owns a resource. And any justification offered has to be universalizable. The reasons for all these requirements should be clear by now. Universalizability has been discussed. Particular owners must be assigned to each and every scarce resource this is what any theory of property any ethic has to do. There must be an objective link between the owner and the resource, so that conflicts can be avoided, and also to comply with universalizability. “Every” scarce resource must be owned by someone, for conflict-avoidance and other reasons given above.
To this point the case is fairly general, and only establishes the framework for examining various competing norms. The libertarian insistence on objective links between resources and owners, and its particular view of what constitutes such objective links, is what completes the case.
Objective Links: First Use, Verbal Claims, and the Prior-Later Distinction
So now we come to libertarianism. It turns out that libertarianism is the only theory of rights that satisfies the presuppositions of discourse, because only it advocates assigning ownership by means of objective links between the owner and the property. This link, of course, is first use, or original appropriation. Only the norm assigning ownership in a thing to its first user, or his transferee in title, could fulfill this requirement, or the other presuppositions of argumentation.
There is clearly an objective link between the person who first begins to use something, and emborders it, and all others in the world. Everyone can see this. No goods are ever subject to conflict unless they are first acquired by someone. The first user and possessor of a good is either its owner or he is not. If he is not, then who is? The person who takes it from him by force? If forcefully taking possession from a prior owner entitles the new possessor to the thing, then there is no such thing as ownership, but only mere possession. But such a rule that a later user may acquire something by taking it from the previous owner does not avoid conflicts, it rather authorizes them. It is nothing more than mights-makes-right writ large. This is not what peaceful, cooperative, conflict-free argumentative justification is about.
What about the person who verbally declares that he owns the good that another has appropriated? Again, this rule is not justifiable because it does not avoid conflicts because everyone in the world can simultaneously decree that they own any thing. With multiple claimants for a piece of property, each having an “equally good” verbal decree, there is no way to avoid conflict by allocating ownership to a particular person. No way, other than an objective link, that is, which again shows why there must be an objective link between the claimant and the resource.
As Hoppe states:
“Hence, the right to acquire such goods must be assumed to exist. Now, if this is so, and if one does not have the right to acquire such rights of exclusive control over unused, nature-given things through one’s own work, i.e., by doing something with things with which no one else had ever done anything before, and if other people had the right to disregard one’s ownership claim with respect to such things which they had not worked on or put to some particular use before, then this would only be possible if one could acquire property titles not through labor, i.e., by establishing some objective, intersubjectively controllable link between a particular person and a particular scarce resource, but simply by verbal declaration; by decree.  The separation is based on the observation that some particular scarce resource had in fact for everyone to see and verify, as objective indicators for this would exist been made an expression or materialization of one’s own will, or, as the case may be, of someone else’s will.” (TSC, pp. 135-136; see also pp. 142-144)
As Hoppe notes, assigning ownership based on verbal decree would be incompatible with the “nonaggression principle regarding bodies,” which is presupposed due to the cooperative, peaceful, conflict-free nature of argumentative justification. Moreover, it would not addess the problem of conflict avoidance, as explained above.
Thus, Hoppe is correct, when he writes:
“Hence, one is forced to conclude that the socialist ethic is a complete failure. In all of its practical versions, it is no better than a rule such as ‘I can hit you, but you cannot hit me,’ which even fails to pass the universalization test. And if it did adopt universalizable rules, which would basically amount to saying ‘everybody can hit everybody else,’ such rulings could not conceivably be said to be universally acceptable on account of their very material specification. Simply to say and argue so must presuppose a person’s property right over his own body. Thus, only the first-come-first-own ethic of capitalism can be defended effectively as it is implied in argumentation. And no other ethic could be so justified, as justifying something in the course of argumentation implies presupposing the validity of precisely this ethic of the natural theory of property.” (144)
Murphy’s & Callahan’s Critique
I am really at a loss as to where MC would part company with this theory. Do they deny, for example, that there is scarcity in the world, or that conflicts are possible? I doubt it. Do they deny that universalizability is not a requirement for justified norms? I doubt it, unless they are also ethical skeptics, in which case I wonder why they consider themselves libertarians. Do they deny that rights have to be justified, and that justification has to occur during argument? Such a denial would be a neat trick, as it would itself be an argument. Do they maintain that participants in discourse do not presuppose any truths? or do they just say that none of these are normative? Or do they think that argumentation is not a conflict-free way of interacting? in which case they would seem to think bashing someone over the head or stealing their wallet is also a form of peaceful, cooperative discourse.
Or, do they think it is coherent for a participant in the peaceful, cooperative activity of discourse, while searching with the other for a universalizable, conflict-avoiding property allocation rule, to advocate socialism, or any other non-libertarian approach? If they are libertarians surely there must be some advantage to libertarian rights, that would factor in to such a generalized argumentative justification context. Or, would MC seriously maintain that a norm could be argumentatively justified, if the norm, if followed, would render human life, and thus argumenative justification itself, impossible?
MC do not do attempt to debunk argumentation ethics in general, or, alternatively, to show just what ethics are implied in argumentation (and why these are not the ones that Hoppe proposes). Do they believe any norms are implied in argumentation? If not, they would seem to reject the entire edifice of work in this regard, including work by Jurgen Habermas, Karl-Otto Apel, Frank Van Dun, G.B. Madison, Alan Gewirth, Roger Pilon, Tibor Machan, and others discussed in my survey article “New Rationalist Directions in Libertarian Rights Theory.”
On the other hand, if they accept that argumentation implies some norms, which are they? Do these norms support libertarianism? Socialism? Or are they only non-rights-related interpersonal norms, like “be nice” or “don’t lie”? Are these argumentatively-presupposed norms at least consistent with libertarianism? MC write: “Whenever people engage in argumentation, they implicitly agree to a set of norms. For example, each participant implicitly agrees to try to persuade the other(s) through peaceful methods.” I am not clear whether they are only paraphrasing or whether they accept this is true. If they do accept it as true, are there no implications to be drawn from this? Does it place no constraints whatsoever on the legitimacy of norms propositionally advanced in the course of (peaceful!) argument?
It seems to me that if MC accept any form of argumentation ethics as valid that is, if there are some norms implied in discourse then, as libertarians who believe libertarian norms are (somehow) justified, they would have to believe that the argumentative norms are at least compatible with, if not the grounding for, libertarian rights.
What about universalizability? I am not sure if MC really reject the universalizability requirement but if they do, I fail to see how they can themselves adhere to any notion of rights; rejecting universalizability means that any norm whatsoever can be proposed, by simply making up a particularistic reason for it. Without the universalizability principle, literally “anything goes,” which of course leads to ethical relativism and/or skepticism. I will assume that MC are not ethical relativists or skeptics and thus do not reject universalizability. But I am not sure they fully appreciate this principle.
Consider this comment by MC: “To simply declare that ownership rights must be ‘universalizable’ is no help, either; after all, communists could cite the same principle to ‘prove’ that everyone should have equal shares to all property.” MC write here as if they are totally unaware that Hoppe has explicitly stated that “the universalization principle only provides a purely formal criterion for morality” (TSC, p. 131). Of course, even if socialism’s principles were reformulated in a completetly universalizable way, it will still be inconsistent with other norms presupposed in argumentation, as noted above.
And regarding universalizability, MC also state:
“Finally, we wish to note that, even if the above problems are overlooked, it’s still the case that Hoppe has only proven self-ownership for the individuals in the debate. This is because, even on Hoppe’s own grounds, someone denying the libertarian ethic would only be engaging in contradiction if he tried to justify his preferred doctrine to its ‘victims.’
“For example, so long as Aristotle only argued with other Greeks about the inferiority of barbarians and their natural status as slaves, then he would not be engaging in a performative contradiction. He could quite consistently grant self-ownership to his Greek debating opponent, while denying it to those whom he deems naturally inferior.  Aristotle need only contend [that] barbarians  are not as rational as Greeks.”
Do MC think that merely “deeming” or “contending” something to be so is automatically compatible with universalizability? I believe they are simply misapplying the universalizability principle here (or, rather, failing to apply it). For Aristotle to grant rights to himself and Greeks, but not to other individuals, would simply be particularistic. He would have to show that there is some reason, objectively grounded in the nature of things, that justifies rights in Greeks but not in other people identical to Greeks in all respects except for their Greekness. Again, either the universalizability requirement is taken seriously, or it is not. If not, the door to ethical skepticism is opened wide.
MC introduce supposed “counterexamples” of God and slavery. Take the slavery case. They recognize that “Hoppe and Rothbardian libertarians in general do not believe in universal self-ownership. In particular, they believe that criminals may be rightfully enslaved to pay off their debts to victims (or their heirs).” Well, of course! Hoppe is a libertarian. To advocate self-ownership means that a person has the right to control his body, as a default or prima facie matter. Buf if someone commits aggression, of course the victim now is a partial “owner” of the aggressor’s body, because he has a right to use force against it. So consider a victim who now “owns” an aggressor who, say, murdered the victim’s wife. Of course, the owner could engage in debate with the slave, but only by granting the slave the right to use his body for purposes of argument. But how does this change the fact that no one can argumentively deny the normative presuppositions that imply libertarianism? Let’s assume the owner is libertarian. He believes in the need for property rules and conflict-avoidance. He believes any norms have to be universalizable. If he advocated socialism, his argument would be incompatible with necessary argumentative presuppositions of peace, prosperity, and conflict-avoiding prosperity because socialist rules are either not universalizable or are not based on objective links between owner and resource. But his claim that he has a right to wield force against the slave is perfectly justified. It is universalizable, because the different treatment of the slave-aggressor and the master-victim is not arbitrary but is grounded in the objective fact of the act of aggression. It is compatible with objectively assigning property rights, because it is a way of enforcing objectively assigned property rights that are violated.
As for God you can’t just posit that God owns everyone and “therefore” we are not self-owners. Moroever, even if God does own us, it could be that we are still self-owners vis-a-vis each other. In any event, this in no way refutes the conclusion that only the libertarian norms can be argumentatively justified in discourse.
MC try to make much of their notion that propositions advanced “during” argument are not subject to the presuppositions of argument, if the rule is designed to be applied in a non-argumentative context. But propositions can only be justified during argumentation. A participant in discourse cannot deny that conflict-avoidance is good. When he seeks to justify something, it is always some action he seeks to justify. The justification takes place at one time; the action to be justified, at another. So what? Are MC saying that no action can ever be justified, other than argument itself? Consider an act of theft, or property acquisition, or rape: all non-argumentative actions. Obviously, these actions are not justifying-actions, because they are not arguments. The only time they could possibly be justified is at another time, during argument. In any event, this critique seems to miss the point. If two people seek to agree upon a fair, universalizable rule for assigning property rights in scarce resources to individuals in a way that would allow conflict to be avoided and the resources to be used of course the rule they are considering will be applicable to future property disputes. I am baffled at how they could think otherwise.
September 19, 2002