Defending Liberty: Essays in Honor of David Gordon

CHAPTER 2

Nozick and the Natural Duty of Justice

Billy Christmas

Individuals have rights, and there are things no person or group my do to them (without violating their rights).

—Robert Nozick, Anarchy, State, and Utopia

The debate over the role of the state within libertarianism as a political philosophy has been done to death. However, I want to resurrect it because I think that an idea in Kant’s Doctrine of Right contains valuable resources that have not yet been brought to bear on it in its past life. My somewhat bold intention is to lay it to rest swiftly after having resurrected it; I do not intend to create a Kantian zombie.

Most libertarians are roughly Lockean inasmuch as they believe that the state could only have moral authority over us—that which it claims—if we consented to its rule. Of course, we do not consent, so it has no authority. Anarchist libertarians stop there and suggest various ways in which admittedly essential state functions could be provided without a state, and therefore without involving the concomitant injustices of the state.

Others proceed to believing that the state may be justified since we do not immediately have anything better on hand. Maintaining some degree of the status quo may avoid the greater injustice of what might happen if the state were to disappear, and therefore is all things considered justified, at least in certain areas of its operation. This is what John Simmons, the authority (excuse the pun) on Lockean consent theory, calls philosophical anarchism.1 On this view, under certain circumstances, states can justly monopolise coercion, but they do not have authority.2 Authority presupposes a general obligation of obedience, and therefore gives the state the moral power to create specific obligations. When a state lacks authority but justly monopolises coercion, it may only enforce natural duties, it cannot create new ones. Our obligation to such a state is merely to defer to it when it comes to the enforcement of rights, but not to defer to it in legislating the moral law.

Robert Nozick, typically regarded as a Lockean, offers an account of how a state can emerge through just processes without our consent such that so long as the state limited itself to the monopolisation of rights enforcement, and that monopolisation was de facto in origin, it would be justified.

Whilst Nozick’s account has not garnered widespread support even within libertarian political philosophy, I believe that it offers a plausible synthesis of the strongest aspects of two otherwise implausible rival accounts of state legitimacy: Lockean consent theory and Kantian natural duty theory.

The Lockean theory is that individuals come into the world with natural rights and that governments, if they exist, come into the world as a result of the choices of these right holders. Where right holders consensually transfer their natural power to enforce their own rights to a common third party, that third party is a state, and may legitimately coerce right-holders because they have already consented to its actions. The Kantian theory is that we come into the world with a need for rights, yet those rights are not real until and unless we create a state that coerces us in-line with these rights and correlative duties. Our most urgent duty is therefore to create a state so that our rights can become real.

What I will argue, is that Nozick’s argument for the legitimacy of coercing non-consenters in a state of nature into surrendering their right of self-enforcement and to paying for the dominant protective association, so long as that protective association protects its rights in every other way, succeeds. And I will use elements of Kant’s natural duty theory to show that this is so, without abandoning the Lockean commitment to the reality of natural rights in the state of nature. First, I will outline in greater detail the Lockean and Kantian respective theories of political legitimacy, with a view to highlighting at least the prima facie plausibility of Nozick’s approach.

LOCKEAN CONSENT THEORY

Locke regards everyone as moral equals insofar as no one has natural authority over anyone else. This gives everyone a right to acquire private property freely and unilaterally in the state of nature.3 Much of the divergence between Locke and Kant’s respective theories of legitimacy can be ascribed to subtle divergences in their views of the relationship between equality and property acquisition.

Illustrative of this natural equality is that no one comes into the world as an owner of the Earth or any part of it. In contrast to Robert Filmer who argued that the heirs of Adam are born into inheritance of the territory of their Kingdoms;4 Locke argues that no one enters the world with any property. Moral relations of equality also mean we do not need permission from anyone else before we go about engaging in world-interactive projects, which render parts of the world parts of us. The idea of mixing labour with hitherto unowned resources is how Locke understands unilateral acquisition. The purposive connection of persons with external resources through unilateral activity means that respect for persons requires respect for their property acquisitions.5 At least, so long as we do not use that property to subordinate others thereby violating natural equality.6

Unilateral acquisition of property is required for humans to be free in two ways. First, when humans freely act, their actions extend over the physical environment in a way which just makes it the case that for others to interfere with that space is for them to interfere with the former’s actions. Interference in another’s actions violates her freedom and defies the natural authority she has to do as she wishes without permission from any putative superior.7 Second, property is a sphere in which one can act without the permission of anyone else, so it only makes that freedom more robust if individuals get to construct that sphere of freedom freely.8 Without the power to particularise their freedom without permission from others, it is not clear that their rights are really their own, but rather something that can be granted or revoked by some superior power or the community as a whole. In Hohfeldian terms,9 unilateral acquisition means that we have an immunity with regard to our own property, which safeguards the claims, liberties, and powers therein. And a reflective disability with regard to the property of everyone else. These respective spheres of individual authority over ourselves and our own projects and affairs reflects our natural equality—no one naturally has any right over the life of another. We are equal inasmuch as we have authority over ourselves.10

All that has to happen, therefore, for a person to have rights and correlative duties she is bound to respect is for her to exist. And these rights and duties develop more refined and detailed content when we act. Unilateral action by individuals is sufficient for having full blown rights to person and property that no one else may revoke. Before there is a state, then, there is a moral law that all individuals must comply with.

According to Locke, individuals are free to enforce their rights against those who threaten them.11 Again, if one was not free to unilaterally protect one’s freedom from third party encroachment, then one was not really morally free in the first place. The moral law protects one’s freedom, but it cannot do so without also permitting you to protect yourself when you need to. Some have argued that it is internal to a claim right that its holder may enforce it,12 or that it must have counterfactual robustness of some kind.13 For example that if it is violated one has some recourse to seek restitution or punish the violator. Locke seemed to have believed something like this. The moral law, then, fully accounts for individual rights in a state of nature.

Locke does note certain “inconveniences”14 of a state of affairs in which each individual is afforded the sole right to enforce the moral law on their own behalf. People will often interpret the moral law somewhat in their favour, even in good faith. They will regard harms done to them somewhat more serious than equal harms committed by them. They will want to extract a bit more compensation than they are really due, and feel they owe a little less than they really do. Whilst the moral law does provide a determinate moral standard for the just ordering of actions, the state of nature does not yet contain anything that is intersubjectively ascertainable that can be pointed to, to overcome the disagreements and conflicts that arise out of such divergences of judgement. All there is are rival interpretations, the holders of which respectively believe correspond to the genuine moral law. Given the potentially insurmountable disagreement over the conclusion of any given conflict, there will always be uncertainty over whether any given matter is in fact settled. To overcome this inconvenience, there must be something which binds all persons that is not only accessible through the individual agent’s moral reasoning and judgement, but rather something that is intersubjectively ascertainable—something public like a written document or an individual person’s spoken ruling, or the like. Something which, while it may imperfectly express the moral law, is something that all persons are bound to for the sake of overcoming the potential effects of divergent interpretations of how the moral law applies to their own case. Locke believes that interpretation and enforcement of rights by a common third-party arbitrator would solve the inconvenience. But how could this agent make final pronouncements on justice without the consent of all affected parties? An agent’s arrogating themselves with the normative ability to determine how other persons rights are to be enforced would involve a presumption of natural authority over those person’s rights; a natural authority Locke believes no one has. Therefore, such an agent could only be legitimate if everyone whose rights they administered consented to them.15

In other words, Locke believes that freedom involves no natural duty to form a state. We are entirely at liberty to voluntarily join or not to join any given commonwealth. Our rights are our own, and not the state’s or the community’s, and any duty we have to the state is strictly contingent and acquired, not natural.

KANTIAN NATURAL DUTY THEORY

Kant’s starting position is similar to Locke’s. He also believes that each person has an innate right to freedom, where freedom is understood as independence of the will of any other person.16 He also believes, like Locke, that property both particularises our individual freedom, and protects our exercise of our freedom. However, he has deep concerns about the status of property in a state of nature. Concerns which result in a very different theory of political legitimacy. Kant believes our rights not to have our bodies physically interfered with in a state of nature is clear enough, so we can at least be said to have a right to freedom from assault. Moreover, we may enforce this right ourselves—rights are essential authorisations to use force. But a fully-fledged right to external freedom is one that is merely provision until we have a state. There are a number of steps to this argument.

Firstly, a mere right against having our bodies interfered with is insufficient for full external freedom. External freedom means not only that we are respected in our physical integrity and possession of our bodies (and anything physically attached to our bodies, like chattels we can hold in our hands), but also respected in our intelligible possession of external objects. Our physical possession of our bodies and external objects we physically possess is protected by our basic rights not to be physically interfered with. However, this means that if one were to initiate some activity with regard to an external object, the moment the physical connection between one’s body and the object is severed, one’s possession of it is no longer protected by one’s bodily rights. If one started building a house, as soon as one lays a brick atop some mortar, sets down one’s trowel, and walks away, one relinquishes any protected claim that others are duty bound to respect with regard to the brick. Kant things this will not do for freedom, and yet it is the most we can have in a state of nature. In order for our sphere of freedom to extend beyond the physical body (and whatever is attached to it), intelligible possession must be protected. That means that we have a right to recover physical possession of external objects not currently in our physical possession at any given time. That means, a right to external objects that excludes others, regardless of our purposes with regard to it—property.17

Why can one not claim external objects as one’s property in a state of nature in the way Locke thinks, according to Kant? The problem for Kant is that the moral law is partially indeterminate as to the particular property rights of any given individual. Whilst the moral law is clear that everyone has a right to property, it is not clear which particular property rights any particular person has. Property is indeterminate in a state of nature. This problem is compounded when such rights need enforcing. When I enforce my interpretation of my rights against you—the moral law as I see it—I am subjecting you to my will by imposing my own interpretation of rights upon you. This violates your freedom inasmuch as you are no longer independent of my will. Interpretation and enforcement of rights in a state of nature is unilateral. Furthermore, a situation in which everyone imposes coercion upon one another in accordance with their own private judgements of justice, we are in a state of constant threat and vulnerability to one another. This vulnerability licences each of us to use unilateral coercion to pre-emptively protect ourselves from being subjected to the will of others.18 The lack of assurance of our rights generates a state of affairs in which no assurance can emerge.19

In a state of nature, we cannot particularise our freedom without rendering one another unfree. Where there is no property, there is no freedom, but when we try to create property, we simultaneously subordinate one another to private power. We’re damned if we do, and damned if we don’t. What is required, is for there to be an authority that can create and enforce some schema of rights in external objects without thereby subjecting us to any person’s private will—a public authority. Whilst the individual agents of the state necessarily act on their own will even when they execute the state’s laws, those subjected to their coercion are not subjected in a way that renders them unfree. Coercion that is executed in accordance with laws that apply to all equally with the function of creating external freedom for all, does not constitute an imposition of any individual’s unilateral will but rather the omnilateral will.

In order for our freedom to be real we must, as a matter of priority, establish and obey a state. Our respective innate right to freedom, then, cannot be acted on until there is a public authority that lays down a uniform codification of rights. Outside of a state we cannot wrong each other by invading one another’s external freedom, yet we “do wrong in the highest degree” by remaining in that state.20

Kant believes we have a natural duty to establish and obey a state.21 We entirely lack rights and correlative duties until we join a commonwealth. Our rights are merely provisional, waiting to be realised in the coercive authority of the state. Our duty toward that state is necessary and natural, rather than contingently acquired. Kantian freedom involves deep anxieties about the state of nature—it is devoid of justice, and a place in which we cannot be fully human. Where we cannot be assured of our rights whatsoever, we lack the full status of rights-bearing individuals. What Locke wrote off as mere inconveniences to our rights, Kant believes penetrates to our very status as rights-bearers whatsoever.

NOZICKIAN INVISIBLE HAND THEORY

Nozick affirms the Lockean commitment to the reality of rights in a state of nature, whilst denying that consent is necessary to a legitimate state. Nozick’s is a contingent derivation of a non-voluntary state, where Locke’s is a contingent derivation of a voluntary one, and Kant’s a necessary derivation of a non-voluntary one. Whilst many libertarians (and with them, almost all political philosophers22) reject Nozick’s derivation of the state, I think that at least as far as it rejects the aspects of the Lockean and Kantian respective accounts that are too strong, it has plausibility. I will reconstruct Nozick’s derivation of the state here and supplement it with some important Kantian insights to show how they support Nozick’s claims (and how we should soften Kant’s).

In his invisible hand argument for the state, Nozick is self-consciously positioning himself against the quasi-Lockean, individualist anarchism of Benjamin R. Tucker23 and Lysander Spooner24 that he was exposed to through his contemporaries Murray N. Rothbard,25 David Friedman,26 and Morris and Linda Tannehill.27 These anarchists, like Locke, believed that we have a right to freedom that we may self-enforce in the absence of any political institutions we have consented to. Indeed, they believed the more likely consensual scenario was not a single voluntary political association at all, but a plethora of protection associations operating simultaneously, protecting their respective clients with whom they individually contract. Nozick intended to show that a non-voluntary state could emerge from a system of market anarchy in a way that nonetheless did not violate anyone’s rights. He did not believe that we consented to any actually existing state, but he wanted to show that it was in principle possible for a certain kind of state to emerge non-consensually and yet without violating our rights.

The starting point from which Nozick departs in his speculative derivation of the state, then, is that of market anarchy—a system in which a given geographical territory inhabited by a society contains a number of protective associations competitively seeking contracts to enforce the rights of individuals. Nozick believes that agencies who disagree over whose client was in the wrong, or over how much restitution is owed, etc. will enter into “battle.”28 Since they cannot agree on right, they must resort to might. Nozick believes such battle can tend toward any one of three outcomes. (1) One agency tends to win more often than the other, and eventually grows in power due to clients of the vanquished agency transferring their subscription to the evidently stronger agency, in which case all but the most dominant agency remain, and everyone is either a member of that association, or they are independent—members of no association. (2) The victories of respective agencies tend to correlate by geographical region, meaning a similar effect as the above, except each smaller subregion has its own dominant agency. The competing agencies find comparative advantage closer to home, and the geographical area we started with is balkanised. (3) There is rough equality in the violent capacities of the respective agencies, and none distinguishes itself as dominant. To avoid further costly violent conflicts, these agencies recognise it to be in their own self-interest to agree to nominate a third-party arbitrator to rule where they cannot find agreement bilaterally. In each one of these cases, a dominant protective association emerges either monolithically over an area, or one dominates each subregion, or one is generated by an agreement among the competing associations. The reason why protective associations tend toward monopolisation is because the value of membership of the larger association grows with its size whilst the value of membership of the smaller shrinks with its size. The greater the number of clients a given association has, the higher the value of its product, since its product facilities protection from a larger volume of co-associates.29

In each case, the dominant association has not quite got the characteristics of a minimal state. A minimal state limits itself to protecting rights but does not permit any other agent to unilaterally engage in rights protection. Moreover, it reserves the right to extract payment from everyone within its jurisdiction, not just its voluntary subscribers. As yet, the dominant association does not have the right to tax, nor may it prohibit self-enforcement of rights by non-subscribers.

Nozick then goes on to say that the entrance of a competitor agency (notwithstanding its likely economic unviability) would introduce risk relative the status quo—risk of violence towards the members of the dominant association. As a rights-protector, then, the dominant association may prohibit entry into the rights-protection market simply as an extension of their duty to protect their members. At this point, the dominant association now takes the form of an “ultra-minimal state.” It monopolises legitimised violence in that it prohibits competitors, yet there would still be those who choose not to subscribe, and it would lack any power to coercively extract payment from them and does not extend protection to them. So, it is a monopoly, but lacks a power to tax and does not provide general protection.30

Nozick then considers how the presence of independents raises problems for members of the dominant agency.

An independent would be allowed to proceed to enforce his rights as he sees them and as he sees the facts of his situation; afterwards the members of the protective association would check to see whether he had acted wrongly or overreacted. If and only if he had done so, they would punish him or extract compensation from him.

But the victim of the independent’s wrongful and unjust retaliation may be not only damaged but seriously injured and perhaps even killed. Must one wait to act until afterwards? Surely there would be some probability of the independent’s misenforcing his rights, which is high enough (though less than unity) to justify the protective association in stopping him until it determines whether his rights indeed were violated by their clients? Wouldn’t people choose to do business only with agencies that offer their clients protection, by announcing they will punish anyone who punishes a client without first using some particular sort of procedure to establish his right to do this, independently of whether it turns out that he could have established this right?31

Nozick seems to be relaxed about the prospect of misenforcement of rights by members of the dominant association because they have already voluntarily transferred their right to enforce their own rights to that agency. Of course, they might enforce their rights themselves, even misenforce them, it is just that the agency is authorised to pre-emptively stop them from doing so and has presumably been successful in extinguishing the potential for self-enforcement among its members. Since the dominant agency is authorised to pre-emptively stop all self-enforcement, it a fortiori avoids all misenforcement. It is not authorised to do this with independents, it must wait and see if they misenforce before the use of force on their part is authorised.

Nozick believes that it is important not merely that our rights be protected, and that when they are violated restitution is made.32 It is also important that we know that our rights are already secured ex ante as far as possible. It is not enough that it is possible and permissible to exact retribution or extract restitution after one’s rights are violated, it is far more important that we know they not be violated. Such ex-post remedies do not erase the rights-violation, they are simply the best one can do once the damage is done. The residue of wronging remains in two sense. The first is that restitution or retribution trivially do not undo the fact that one’s rights were violated. Restitution may be a fine, but is not a fee. The second is that the absence of knowledge that one’s rights will not be violated is itself a kind of harm.

Although it is not a rights violation (in the absence of a rights violator), there is some significant moral cost to us when we rightly believe ourselves to be vulnerable. Firstly, because we are not safe to actually make use of our rights; to get the full value of our freedom. Without the actual ability to safely act on our rights, it is almost as if we do not have them. Secondly, it is not clear that our moral status as the kinds of being who bear rights is reflected in the social reality in which we live. It is not a rights-violation for one’s rights to not be given a fully-fledged social recognition, but it is certainly a kind of downgraded wrongful status one is left in that we should be concerned about for the same reason we should be concerned with rights at all.

This is why Kant believed that in a state of nature we could take pre-emptive action against one another; we were all presumptive threats, and given that there are no social institutions in place to extinguish that collective and mutual threat, we are licenced to protect ourselves as a last ditch effort to protect our status as rights-bearers.33 Even though (indeed, perhaps in virtue of the fact that) we are all entitled to pre-emptively attack one another for the sake of our own security, this state is not a state where we really have rights. In failing to institute a social state of affairs in which we can act on our rights in the knowledge that others will not violate them, we do not wrong one another, but we nonetheless do wrong “in the highest degree.”34

Whatever the nature of the wrong, two things are clear. It is not a violation of anyone’s rights, but it is still supremely wrong due to the normative importance of rights. Nozick concludes that the dominant agency may impose itself upon the independents at this point and prohibit their unilateral enforcement of their own rights—they must defer to the agency. Given that they benefit from the relative security provided—now to all—by the agency, the agency may force them to pay for the protection received.

Most interpretations of Nozick say that he has compromised the notion of rights he starts out with; a notion of “side-constraints” rather than maximally valuable consideration to feed into cost-benefit analysis. Nozick rejects what he calls a “utilitarianism of rights.”35 That is, the idea that an action that violates rights can be deemed rightful when its consequences are the avoidance of a larger numerical volume of, or more serious kinds of, rights violations. Rights are not a master value to be fed into utility calculus, they are a side-constraint on our pursuit of all other values.

Nozick’s idea that membership of a protective association can be forced upon a group (in pro tanto violation of their rights) so long as the membership assures them of robust protection of their rights in all other regards, is normally read to mean that it is permissible to violate someone’s rights so long as you provide them a stream of benefits that more than make up for it. The result is a Pareto improvement because the imposition is Kaldor-Hicks optimal, it is just that the currency is the frequency of the non-violation of rights, rather than units of welfare.

I think that David Gordon’s interpretation36 is the right one: Nozick thinks that, in some sense, unsecured rights are no rights at all. Even if it is not what Nozick himself meant, I think it makes a plausible amendment to the theory, not only because it maintains faith to rights as side-constraints,37 but also because it makes sense of a problem that Nozick leaves vague. The problem of not knowing whether independents plan to violate your rights or not, and not knowing if they will unilaterally try to extract too much compensation from you if you violate theirs. It is not that rights turn out to be less deontic than we thought, and we can violate them in marginal cases for the sake of their protection in general cases. Rather, it is that rights are so weighty, that we ought not have a wait-and-see attitude about whether or not they will be violated in certain kinds of social contexts. Those social contexts are ones in which we have some reason for thinking rights are systematically at risk—where the assumption that individuals are not threats to one another unless and until they initiate some ostensive action constitutive of a threat directed at another agent, vanishes. It is not that the association provides a benefit and recoups payment for it. It is that this arrangement is necessary for one to have the status of a rights-bearer whatsoever. It is a deontic claim, not a utilitarian one.

Respect for rights is something that is too important for us to make it a baseline assumption, rather than something requiring the positive presence of particular kinds of institutions. This, I think, echoes Kant’s assurance problem. For Kant, where there are no state legal institutions, we are automatically threatening to one another. This is not purely because in lieu of enforcement we may (in the physical possibility sense of “may”) invade one another’s rights, potentially leaving behind irreparable harms. Rather, Kant believes that we need a state legal system even if we are all morally perfect; so, it cannot be a worry about mere physical security. Rather, it is that whatever physical security there may be, it is not security we can mutually have knowledge of the effectiveness of—we cannot trust in it. Where there are no legal institutions, Kant thinks, we have nothing other than hope that others will both know what justice consists in,38 and be motivated to act justly. Hope is not the same thing as trust. When we have to hope that our rights are known and acted on by others, this not only shapes the way we use our freedom, such as it is, but also undermines our status as rights-bearers. That is not to say that Kant is right that only a state-like entity could provide this assurance, only to say that where assurance is lacking, we lack full standing. When our abstract rights are not given fully-fledged social reality, embedded in patterns of action across the society through which we move, our status is damaged. Facing one another as rights-bearers is what gives reality to our status as persons of equal moral status—it gives reality to the abstract humanity which we share. Living in social circumstances where rights are systematically insecure and unrecognised, as they are wherever we lack knowledge that others recognise our rights, as a life that is not fully human. As humans, we have a duty to overcome these circumstances and create institutions that recognise our status as moral equals and mutually protect our rights to be free. The rights to be free come after the institutions are created. Therefore, whatever institutions best secure our rights, we have a prior duty to uphold them. This is why a dominant protective association might be justified in imposing itself upon independents, so long as it really is the best way to secure our rights—and that it protects the rights of those it imposes itself upon would be partial evidence of that.

The idea that it can never be justified to impose institutions for the protection of rights upon parties without their consent is founded in the maxim of the presumption of innocence. Independents who have not actually initiated any overtly threatening actions towards members of the dominant association have not done any wrong; they have not initiated any rights-violating action that would render them liable to having their own rights defensively invaded by agents of the association. However, Kant and Nozick have a notion of threat that goes beyond individual actions, but that can include collective omission.

Of course, individuals can initiate threats that constitute rights violations, and these render them liable to defensive force where necessary and proportionate. However, certain kinds of social environments can themselves be threats to rights generally, apart from any identifiable agent’s actions. It is possible that moral agents are threatened as such simply by existing in a certain institutional setting, not because any particular person has engaged in a positive act that itself directs a threat toward any one in particular. Kant thinks that all states of affairs apart from a state represent these circumstances. Nozick thinks that a lack of shared rights-protection association does, since he thinks that the default for co-members is one of security and the default between members and independents is one of hostility. In failing to overcome a state of affairs in which the default assumption is rightful hostility, Kant describes us not as doing wrong against any particular person in the sense of constituting a rights violation, but of “doing wrong in the highest degree” by failing to take positive actions that would create a state of affairs in which we can recognise each other as rights-bearing equals.

I do not intend to argue that only a state can overcome that state of affairs, or that states necessarily can overcome such states of affairs. I intend to argue that it is justified to coercively deprive individuals of their rights of self-enforcement when and if that is necessary to overcome that state of affairs. And the provision of security to those agents is an implication of the justificatory fact that such institutions are the best for securing our status as rights-bearing equals. What I argue then is that we have a contingent duty of justice to establish and obey as far as necessary institutions that facilitate reciprocal trust between agents, such that they know that they are regarded as rights-bearers and will be treated accordingly as a default assumption. Deference to an enforcement and arbitration agency that is common to all agents may be the best way of securing such trust in some circumstances. The duty is not natural because the circumstances it evades are not necessarily natural. There should be no default assumption about what the default assumption is regarding the threat posed by other persons. It all depends upon the social context. However, if and where social institutions (or lack thereof) are such that individuals are not justified in holding a default assumption that other members of society will respect their rights, then all members of society acquire a duty to change those institutions and comply with ones that do provide such security. This is consistent with political anarchism, since stateless institutions could in principle do the work necessary to secure rights.39 Consent to the institutions that are best for securing justice is not itself a requirement of justice.

MARKET ANARCHISM

Market anarchists are unconvinced by the case Nozick makes.40 They argue that it is not necessarily the case that a single association would become the dominant one, and that it would nevertheless be unjust and possibly unnecessary to compel the obedience of independents. I will reconstruct these arguments below and argue that they cannot escape the potential need to impose compliance with a rights-protecting institution upon non-consenting agents, and whether or not that institution is a state or not is not particularly important.

The market anarchist denies the necessary truth of Nozick’s claim that competing associations that disagree would necessarily resort to physical battle. Indeed, they make heavy work of Nozick’s third suggestion that they may have the foresight to agree to a third-party arbitrator. They deny however that there must necessarily be a single third-party arbitrator to all pairs of disputing associations. It is a fallacy of composition to affirm that since all disputes needs third party arbitration, there must be one third party arbitrator to all disputes. What happens when two or more among this plurality of arbitrators have an incommensurable disagreement? Well, hopefully there are some social norms in place that act as constitutional constraints such that when appropriate procedures are not followed arbitrators are sanctioned. However, such norms could solve the problem at a more basic level; before we even need to resort to third party arbitration. There could simply be social norms that emerge that ensure that the associations themselves are sanctioned when they do not comply with explicit or implicit agreements over how to proceed in cases of disagreement. Such norms fill the juridical void that statists tend to think must be filled by a state. Instead of a state appointed judge and written constitution, you simply have effective and stable informal norms.41 The norms have de facto juridical monopoly rather than any one of the particular agencies as such.42

Where such norms effectively generate trust, Nozick’s worries about the effects of insecurity fall away. We could, in principle, secure our rights under anarchy, and have well-placed faith in that security.43

However, I think there is still the conceptual space for Nozick’s worries to re-emerge. What if an association emerged that defied the norms; would it be legitimate for the other associations to impose the norms upon them coercively? If the answer is yes, then it looks like these other associations are behaving like a state (even if they do not consolidate into a “single” agency). If the answer is no, then should the members of the norm-abiding associations not fear for their rights in the same way they did with the independents?

To this, the anarchist reply is that such problems are present within a state as well. State agents might break the law as much as independents break the norms in a stateless society. Perhaps Nozick is wrong to assume that members of the same association can trust one another, as well as the enforcement agents. The point however is surely that we would concede that if a state agent broke the law and misenforced someone’s rights we would think that the state had failed in doing something that was not merely permissible for it to do but was also duty bound to do—to ensure that misenforcement did not take place. The same should go for the independent who breaks the norms of procedure that avert the danger of misenforcement; coercive measures should and could have been employed to prevent it from happening.

The second objection levelled by market anarchists is that we are not permitted to initiate coercion against those who might violate our rights; we are only permitted to use force to counter the actual initiation of force. No one has any duty to defer to an agency they have not consented to purely because other people might be violating one another’s rights. So protective agencies do wrong when they deny the right to self-enforcement to independents because it is they who initiate the use of force. Moreover, why should I get my hands dirty in obeying and supporting an agency that uses coercion against those who are only seeking to enforce their own rights—as is their right? Well, I think there are two things that can be said in response to this.

The first is that the normative power that threats to one’s rights have in authorising the use of defensive force does not need to come from the positive actions of anyone in particular. Social states of affairs can be hostile without any identifiable individual initiating a threatening action against any particular person. Hostility can be an emergent feature of collective action or of collective omission. Security of rights is not necessarily natural, but in some contexts require the positive presence of institutions. However, even if this is denied there is another problem. It is true that we have a duty not to aggress against others or violate their freedom. And it is true that one can abstain from violating rights without joining, obeying, or supporting any particular set of institutions that positively protect the rights of others. Inaction on one’s own part does guarantee that one does not violate rights. However, it does not guarantee that rights are not violated. Hence the duty to defer to a common agent to secure rights is not one that directly correlates with the rights of others, but it is one that has to be implied by the existence of rights at all.

An account of rights that says it is more important to avoid getting our hands dirty than contributing to a social order in which rights are realised is not a kind of anarchism that can have a stable relationship with liberalism. Liberalism seeks freedom on Earth among sinners; and that means adjusting our behaviour in light of that sin, even if we ourselves are good. Keeping your hands clean may get you into heaven but not into a free society. If seeking the Kingdom of God is to be reconciled with social life, it cannot require that we not get our hands dirty, as is often believed to be the correct implication of Matthew 6:33; that one first seek the Kingdom of God.

A SYNTHESIS OF LOCKEANISM AND KANTIANISM

Locke believed that legitimacy was something states acquired only when they were consented to. Kant believed that legitimacy was something states necessarily had because of the kind of thing they are. What seems more plausible is that states are legitimate if and when they generate the assurance of rights better than any alternative.

If there is an argument for the state, it is that deferring to it enforcement of rights is the best way of enjoying our rights among others. But it is not because there is anything conceptually to do with the state being “one” agency or a monopoly as such; legitimacy does not attach to those properties per se. It must be that through those things it is able to generate a situation of trust among rights-bearers. A state is legitimate, then, if that legitimacy functions to protect our rights. The use of coercion in a stateless society, however, is also legitimate if it functions to protect our rights. The distinction between anarchy and statehood is not particularly morally interesting, since legitimacy in both is simply achieved through providing better assurance of rights than there would be sans their coercion. Nozick’s position—or my reconstruction of it—is superior to the Kantian one because it does not say we have no rights of justice outside of institutions that give them a social reality. Rather, it says that we do have such rights, and they are so important that it is legitimate to non-consensually deprive some of the right to self-enforcement for the sake of rights generally. It does not deny that independents have a right outside of their incorporation into the dominant agency; it just denies that merely because they do not initiate any positive action that constitutes a threat, that their non-incorporation does not constitute a threat. One can play a role in generating an unsafe environment through one’s omissions, and not only through one’s actions. Kant’s notion that abstaining from entering a civil condition undermines our status as rights-bearers, and that such a situation is so inhumane that we have a duty to depart from it is a plausible one that I think Nozick shared. But Kant is wrong to think that a state is necessarily the only way to do this. The market anarchists are right to say that in principle, non-monopolistic organisations could provide the kind of security needed for our rights to have social reality, but they are wrong to think that this thwarts any need to non-consensually impose any institutions whatsoever upon non-consenting parties. Refusal to do the latter may very often leave our rights at systematic risk.

Whilst our rights do in fact exist in a state of nature, and we are all compelled by the moral law of justice not to invade one another’s rights, we also have a duty to take whatever positive action might be required for the sake of the protection of rights generally; this might mean giving up our prima facie right of self-enforcement if abstaining from doing so when generalised presents such a threat to others that their status as rights-bearers is undermined.

It is important, I think, to note that on this view, the problem (when there is a problem) with statelessness is not that the use of force to protect rights is unilateral as Kant was concerned. It is that we cannot trust that force will be used justly.44 When unilateral force is used justly enough such that we can all trust that we face one another as rights-bearing equals, or as much so as is possible, that force is legitimate. Perhaps at the point of development of the dominant agency, the state, or the nexus of overlapping competing agencies, where rights are secured and there is substantively just rule of law, the use of force can be said to be omnilateral. But the whole point is that it will have to start out as unilateral. Justice cannot be secured suddenly and all in one go, by an institution that has legitimacy prior to its actual securement of rights. The functionality of rights protective institutions is retro-actively legitimising.

1An early version of this essay was presented to the Colloquium on Market Institutions and Economic Processes at NYU, and the Political Theory research group in the department of Political Economy at King’s College London, in 2020. I am very grateful to both audiences for their helpful feedback and the discussions that ensued. Special thanks are due to Carmen Pavel for written comments.

1A. John Simmons, Moral Principles and Political Obligations (Princeton: Princeton University Press, 1981); A. John Simmons, On the Edge of Anarchy: Locke, Consent, and the Limits of Society (Princeton: Princeton University Press, 1993); A. John Simmons, Justification and Legitimacy: Essays on Rights and Obligations (Cambridge: Cambridge University Press, 2000).

2Ralf M. Bader, ‘De Facto Monopolies and the Justification of the State’, in The Routledge Handbook of Anarchy and Anarchist Thought, ed. Gary Chartier and Chad Van Shoelandt (New York: Routledge, 2021).

3John Locke, Two Treatises of Civil Government, ed. Thomas Hollis (London: A. Millar, 1764), bk. 2 chaps 1–8.

4Robert Filmer, Patriarcha: Of the Natural Power of Kings (London: Richard Chiswell, 1680).

5Locke, Two Treatises of Civil Government, bk. II chap. 5. Whether the acquisitive action is really “labour” or not is irrelevant. Labour should be regarded as a placeholder for any activity that involves the subsumption of external resources into one’s ongoing projects. Judith Jarvis Thomson, ‘Property Acquisition’, Journal of Philosophy 73 (1976): 665; John T. Sanders, ‘Projects and Property’, in Robert Nozick, ed. David Schmidtz (Cambridge: Cambridge University Press, 2002); Billy Christmas, ‘Ambidextrous Lockeanism’, Economics and Philosophy 36 (2020): 193–215.

6Bas van der Vossen, ‘As Good as “Enough and as Good”,’ The Philosophical Quarterly Online first (2020).

7Bas van der Vossen, ‘What Counts as Original Appropriation?’, Politics, Philosophy and Economics 8 (2009): 355–73; Bas van der Vossen, ‘Imposing Duties and Original Appropriation’, The Journal of Political Philosophy 23 (2015): 64–85; van der Vossen, ‘As Good as “Enough and as Good”‘; Billy Christmas, ‘Hugo Grotius and Private Property’, Raisons Politiques 73 (2019): 21–38; Christmas, ‘Ambidextrous Lockeanism’; Billy Christmas, Property and Justice: A Liberal Theory of Natural Rights (London: Routledge, 2021).

8van der Vossen, ‘Imposing Duties and Original Appropriation’; van der Vossen, ‘As Good as “Enough and as Good”‘.

9Wesley Newcomb Hohfeld, ‘Some Fundamental Legal Conceptions as Applied in Judicial Reasoning’, The Yale Law Review 23 (1913): 16–59.

10Roderick T. Long, ‘Equality: The Unknown Ideal’, Mises Daily (blog), 2001, https://mises.org/library/equality-unknown-ideal; Roderick T. Long, ‘Liberty: The Other Equality’, Foundation for Economic Education (blog), 2005, http://fee.org/articles/liberty-the-otherequality/; Billy Christmas, ‘Social Equality and Liberty’, in The Dialectics of Liberty: Exploring the Context of Human Freedom, ed. Roger E. Bissell, Chris Matthew Sciabarra, and Edward W. Younkins (Lanham: Rowman and Littlefield, 2019).

11Locke, Two Treatises of Civil Government, bk. II chap. 2.

12H. L. A Hart, Essays on Bentham (Oxford: Oxford University Press, 1982), chaps 7–8; H. L. A Hart, The Concept of Law, 2nd ed. (Oxford: Clarendon Press, 1994); Joel Feinberg, ‘The Nature and Value of Rights’, in Rights, Justice and the Bounds of Liberty: Essays in Social Philosophy (Princeton: Princeton University Press, 1980); Carl Wellman, Real Rights (Oxford: Oxford University Press, 1995); Wellman; Carl Wellman, ‘Review of A Debate Over Rights by Matthew H. Kramer, Nigel E. Simmonds, and Hillel Steiner’, Mind 106, no. 436 (2000): 954–56; Nigel Simmonds, ‘Rights at the Cutting Edge’, in A Debate Over Rights: Philosophical Enquiries, ed. Matthew H. Kramer, Nigel E. Simmonds, and Hillel Steiner (Oxford: Oxford University Press, 1998); Hillel Steiner, ‘Working Rights’, in A Debate Over Rights: Philosophical Enquiries, ed. Matthew H. Kramer, Nigel E. Simmonds, and Hillel Steiner (Oxford: Oxford University Press, 1998); David Frydrych, ‘The Theories of Rights Debate’, Jurisprudence 9, no. 3 (2018): 566–88; David Frydrych, ‘What Is the Will Theory of Rights?’, Ratio Juris, Forthcoming.

13Daniel Layman, ‘Republican Liberty as Liberty within Rights’, in Oxford Studies in Political Philosophy, ed. Peter Vallentyne, Steven Wall, and David Sobel, vol. 8, 2022.

14Locke, Two Treatises of Civil Government, sec. II.13.

15Locke tells an unconvincing story about why he thinks, actually, we kind of have consented to the state under which we live. (Locke, bk. II chap. 8.) This is a very useful foil for teaching students to think seriously about consent, but we need not get into that here, since nobody is actually convinced by it. One of the few things political philosophers agree on is that John Simmons is right about consent theory, even if other aspects of his philosophical anarchism are open to challenge. (Fabian Wendt, ‘Against Philosophical Anarchism’, Law and Philosophy, Forthcoming.)

16Immanuel Kant, The Metaphysics of Morals, ed. Mary Gregor (Cambridge: Cambridge University Press, 1991), sec. 6.238.

17Kant, sec. 6.245–6.255.

18It is often argued that these three problems are independent of one another. I have argued elsewhere that they are all parasitical on indeterminacy (Billy Christmas, ‘Against Kantian Statism’, The Journal of Politics, 2021.)

19Kant, The Metaphysics of Morals, sec. 6.256–6.257; Arthur Ripstein, Force and Freedom: Kant’s Legal and Political Philosophy (Cambridge: Harvard University Press, 2009), chap. 6; Louis-Philippe Hodgson, ‘Kant on Property Rights and the State’, Kantian Review 15, no. 1 (2010): 57–87.

20Kant, The Metaphysics of Morals, sec. 6.307.

21Jeremy Waldron, ‘Special Ties and Natural Duties’, Philosophy and Public Affairs 22, no. 1 (1993): 3–30; Jeremy Waldron, ‘Kant’s Legal Positivism’, Harvard Law Review 109, no. 7 (1996): 1535–66.

22For example, Samuel Freeman takes it to be constitutively illiberal to ground political institutions in private contracts. See Samuel Freeman, ‘Illiberal Libertarians: Why Libertarianism Is Not a Liberal View’, Philosophy and Public Affairs 30, no. 2 (2001): 105–51.

23Benjamin R. Tucker, Instead of A Book, By A Man Too Busy to Write One (Boston: Benajmin R. Tucker, 1888).

24Lysander Spooner, No Treason No. VI: The Constitution of No Authority (Boston: Self-Published, 1870); Lysander Spooner, A Letter to Grover Cleveland (Boston: Benjamin R. Tucker, 1886).

25Murray N. Rothbard, For A New Liberty: The Libertarian Manifesto, 2nd ed. (Auburn: The Ludwig von Mises Institute, 2006).

26David Friedman, The Machinery of Freedom: A Guide to Radical Capitalism, 2nd ed. (LaSalle: Open Court Publishing Company, 1989).

27Morris Tannehill and Linda Tannehill, The Market for Liberty, 2nd ed. (Auburn: The Ludwig von Mises Institute, 2007).

28Robert Nozick, Anarchy, State, and Utopia (New York: Basic Books, 1974), 16.

29Nozick, 17. My understanding of the subtleties of Nozick’s argument here have been greatly assisted by Gerald F. Gaus, “Explanation, Justification, and Emergent Properties: An Essay on Nozickian Metatheory,” in The Cambridge Companion to Nozick’s Anarchy, State, and Utopia, ed. Ralf M. Bader and John Meadowcroft (Cambridge: Cambridge University Press, 2011).

30Nozick, Anarchy, State, and Utopia, 26.

31Nozick, 55–56.

32Nozick, 57–63.

33One’s duty to one’s “rightful honour” is an internal duty because it is something no one else can enforce upon you. Kant, The Metaphysics of Morals, sec. 6.237; Ripstein, Force and Freedom: Kant’s Legal and Political Philosophy, 37–38.

34Kant’s claim here, I think, is that we do not commit any kind of directed wrongdoing such as a rights violation, since there are as yet no rights to violate. Nonetheless, that condition in which there are rights that we can speak of as respected or violated is itself a morally valuable condition (independently of the value of rights being respected) because it enables moral ordering. So, it is a form of wrongdoing, it is just that it is not a way of wronging a particular person or persons.

35For example, see Eric Mack, ‘Nozickian Arguments for the More-Than-Minimal State’, in The Cambridge Companion to Robert Nozick’s Anarchy, State, and Utopia, ed. Ralf M. Bader and John Meadowcroft (Cambridge: Cambridge University Press, 2011).

36David Gordon, Nozick’s Argument for the Minimal State (The Ludwig von Mises Institute, 2006), https://mises.org/library/nozicks-argument-minimal-state; David Gordon, Robert Nozick’s ‘Anarchy, State, and Utopia’ (The Ludwig von Mises Institute, 2018), https://www.youtube.com/watch?v=31pTBdm5qLk.

37As well has Nozick’s remarks regarding duties of fair play Nozick, Anarchy, State, and Utopia, 90–95.

38Kant also thinks that even well motivated agents will act on different interpretations of justice, each of which is equally reasonable, and that this offers further grounds for the need for a unified legal system. There needs to be one version of justice we are all compelled toward, rather than acting on a variety of interpretations as if our own is the legally binding one. See note 39 below for why this does not push in favour of the state versus stateless legal institutions.

39It requires the adoption of “methodological anarchism,” of suspending belief that the state or any other particular institution is necessary in lieu of empirical evidence. Jason Lee Byas and Billy Christmas, ‘Methodological Anarchism’, in The Routledge Handbook of Anarchy and Anarchist Thought, ed. Gary Chartier and Chad Van Shoelandt (New York: Routledge, Forthcoming).

40Michael Huemer, The Problem of Political Authority: An Examination of the Right to Coerce and the Duty to Obey (London: Palgrave Macmillan, 2013), pt. 2; Roderick T. Long, ‘Market Anarchism as Constitutionalism’, in Anarchism/Minarchism: Is a Government Part of a Free Country?, ed. Roderick T. Long and Tibor R. Machan (Aldershot: Ashgate Publishing, 2008); Aeon J. Skoble, Deleting the State: An Argument About Government (LaSalle: Open Court Publishing Company, 2008).

41Harrison P. Frye, ‘Freedom without Law’, Politics, Philosophy and Economics 17, no. 3 (2018): 298–316.

42This is why the additional Kantian concern with justice being made univocal by law does not add any further grounds for requiring a state versus statelessness. What is required is a single set of procedural norms, and there need not be a legal monopoly for that, it is just that law cannot be consensual (cannot be consensual? Why?—because some will not consent to just norms, and others will consent to unjust norms. Justice requires that their consent not matter in such cases). See Christmas, ‘Against Kantian Statism’, for my argument against the claim that concerns with the indeterminacy of rights offer a priori grounds for there needing to be a state.

43Indeed it is trust that is implicit in Nozick’s claim that openly aggressive outlaw associations will be unable to attract clients. Nozick, Anarchy, State, and Utopia, 19.

44Billy Christmas, ‘The Indispensability of Unilateral Coercion’, 200-Proof Liberals (blog), 2020, https://200proofliberals.blogspot.com/2020/08/the-indispensability-of-unilateral.html.