Law and the Entitlement to Coerce *
Law and the Entitlement to Coerce *
Abstract and Keywords
This chapter argues that the justification of the power to make law does not entail the existence of an entitlement to use or to threaten coercion. In a society of morally very good people, there can be a justified legal system in which no public agency or private party has any entitlement to enforce law coercively. In such a society, neither the need to address good-faith disagreement nor the need to provide a coercive assurance of people's rights would justify the risk to the innocent that coercive enforcement necessarily presents. In a society of flawed human beings, governments sometimes have an entitlement to coerce, but this entitlement may be more restrictive than is commonly supposed. A coercive response to free-riding is not always required to make laws morally binding. Depending on social circumstances, the need to address unjustified law-breaking may or may not warrant exposing the innocent to risk.
A long tradition in political and legal philosophy regards coercion as central to the very idea of law. Some historical figures, such as Hobbes, Locke, and Austin, took the position that there can be no law without a coercive sanction. Many philosophers of law, most famously H. L. A. Hart, have called this view into question.1 Nonetheless, many political and legal philosophers continue to believe that law is necessarily connected with coercion in a subtler way. Whenever government is entitled to make a law that imposes a direct requirement on conduct, it is entitled to use coercion to enforce this requirement. Some endorse this position explicitly.2 Others commit themselves to it when they argue against certain kinds of laws or legal arrangements by claiming that coercive enforcement of those laws would be objectionable. Such arguments presuppose that a government is justified in making law only if it has some sort of justification for enforcing law coercively. The view that the entitlement to make law necessarily comes with an entitlement to coerce is challenged rarely, if ever. Nonetheless, this view is mistaken.
I shall argue that the justification of the power to make law does not entail the existence of an entitlement to coerce. It is possible for there to be a legal system in (p.184) which making law is justified but neither government nor any private party has any entitlement to enforce law coercively. In societies like ours, governments may sometimes have an entitlement to coerce, but this entitlement does not follow from the bare fact that some citizens are inclined to violate legal restrictions their governments are entitled to impose. Whether a government is entitled to enforce any given legal restriction coercively depends on concrete empirical facts, not only on abstract truisms about human nature, and on moral features of the restriction in question. Empirically, it matters to what extent compliance with a particular restriction could be secured without coercion and what effects coercive enforcement would have. Morally, it matters whether the value of coercively addressing violations of a given law is great enough to justify establishing a coercive enforcement mechanism that inevitably risks injuring innocent people. Because the answers to these questions can vary, a government’s entitlement to coerce may not extend to all justified legal requirements on conduct, and it may expand or contract as social conditions change.3
The fact that an entitlement to coerce does not simply follow from an entitlement to make law has consequences for the way political discourse should be conducted. Both civil and economic libertarians often criticize legislation by arguing that the legislation in question cannot permissibly be coercively enforced.4 Such arguments must be either abandoned or refined. One cannot straightforwardly argue that a legal requirement should not be imposed by arguing that it would be wrong to enforce this requirement coercively. Recognizing that government can be entitled to make laws that it is not entitled to enforce coercively can also help us to make more sophisticated diagnoses of injustice. For example, arguments for economic egalitarianism often appeal to the injustice of using coercion to enforce an inegalitarian system of property.5 If there is a moral problem with an inegalitarian system of property, it is important to ask whether the problem is only with the coercive enforcement of the system or also with the government’s claim that people ought to comply.
To show that the entitlement to make law does not always come with an entitlement to coerce, I will first argue in section 2 that it is conceptually possible for there to be a legal system that neither uses nor threatens coercion. If it were impossible to make law without using or threatening coercion, an entitlement to make law would, of course, require an entitlement to coerce. Sections 3 and 4 address two ways of arguing that the entitlement to make law generally comes with an entitlement to coerce even if non-coercive law is possible. Section 3 considers and rejects arguments that the entitlement to make law necessarily comes with an entitlement to enforce law coercively. On this view, in any possible society, even a society of morally very good people who do not break the law without justification, (p.185) government has an entitlement to coercively enforce every law it is entitled to make. Section 4 addresses arguments that the entitlement to make law always comes with an entitlement to coerce in a society in which some people are inclined to break the law unjustifiably.
2. The Conceptual Possibility of Law without Coercion
Before arguing that no entitlement to coerce is required to have the power to make law or to be justified in exercising that power, I will explain why it is possible to make laws—not mere public guidelines or recommendations, but laws—that lack coercive enforcement. Indeed, it is at least conceptually possible for there to be an entire legal system that lacks coercive enforcement. Showing that there can be law without coercion is a necessary first step to showing that there can be an entitlement to make law without an associated entitlement to coerce. If making law necessarily involved using or threatening coercion, it would be senseless to try to argue that lawmaking bodies are entitled to make laws they are not entitled to enforce coercively.
I am not the first to defend the conceptual possibility of an entirely non-coercive legal system. Raz and Shapiro have both argued for this possibility.6 My argument for the conceptual possibility of an entirely non-coercive legal system will differ from Raz’s and Shapiro’s arguments in that it does not presuppose a particular account of the nature of law. In particular, it will not presuppose a position on legal positivism, the view that non-moral social facts are the only facts that determine the content of the law.7 Raz’s and Shapiro’s arguments for the possibility of non-coercive legal systems both presuppose particular, fully-developed accounts of the nature of law, which in both cases are positivist accounts. Perhaps because these arguments presuppose positivist accounts of law, the possibility of a non-coercive legal system has not been widely recognized among anti-positivists. Notably, according to Ronald Dworkin’s account of the concept of law, “The most abstract and fundamental point of legal practice is to guide and constrain the use of governmental power,” where “power” means coercive power.8 On this view, law would be either impossible or pointless in the absence of a coercive government whose power needs to be constrained.
I will show that both positivists and anti-positivists can accept the conceptual possibility of an entirely non-coercive legal system. To do this, I will offer a set of jointly sufficient conditions for the existence of a legal system. Since these (p.186) conditions are jointly sufficient, not necessary or constitutive, both positivists and anti-positivists should be able to accept them. I will then show that in a society of morally very good people, there could be a system of legislative and judicial bodies that satisfies these conditions and that does not use or threaten coercion.9
2.1. Sufficient Conditions for the Existence of a Legal System
One might be tempted to think that any characterization of law must mention law’s coercive character. Intuitively, laws seem to be binding in a way that other rules are not. The presence of public coercive enforcement is the obvious way to explain this difference between laws and other rules. So one might think that purported legal restrictions on conduct that lack coercive enforcement are not really laws; they are merely public guidelines, recommendations, or advisories. A system of cooperation that lacked coercive enforcement for its rules would not constitute a legal system because its rules would not be binding in the way that is characteristic of law. I suggest there is another way of seeing the distinction between laws and other, non-mandatory public rules. Even if laws are not binding in the sense that they are coercively enforced, laws can be morally binding in a way that guidelines, recommendations, or advisories are not.
When complying with guidelines is morally mandatory, it is mandatory for reasons that are independent of the guidelines’ existence. For instance, if complying with safety guidelines is morally mandatory, compliance is mandatory because of independent facts about what safety requires, not because a public body decided to establish the guidelines. By contrast, laws do not in general aim merely to report independently existing moral obligations. To be sure, some laws do not change people’s moral obligations and do not aim to do so. For instance, laws criminalizing wrongful violence aim to codify independent moral requirements.10 Most laws, however, aim to change what obligations people have, either by directly imposing obligations or by changing the ways in which obligations can be established, transferred, and eliminated.11 Laws regarding the ways in which property may be acquired and transferred, for instance, aim to affect what actions constitute morally wrongful theft.
Of course, not all rule-making or rule-applying bodies that can change people’s moral obligations are law-making or law-applying bodies. For instance, private voluntary associations may be able to make rules or decisions that change their members’ moral obligations. There are two ways in which legal systems’ normative powers are distinctive. First, legal systems typically claim, and often in fact have, a (p.187) moral power to regulate everyone in a geographically defined society. They do not limit their normative claims to a proper subset of the population, such as a list of volunteers. Second, legal systems characteristically regulate certain areas of life. Though there is probably no area of life that some legal system has not tried to regulate, there are certain areas of life all effective legal systems do regulate. For instance, every effective legal system has something to say about the allocation of physical resources within its territory.
I suggest the following as a set of jointly sufficient conditions for the existence of a legal system:
(1) There is a system of legislative and judicial bodies that makes rules and decisions concerning a subject matter that is characteristically governed by law—that is, a subject matter that is regulated by the legislative and judicial branches of government in every actual society that has an effective government.
(2) These rules and decisions are morally binding on everyone in a geographic territory.
(3) The rules and decisions are morally binding partly because these legislative and judicial bodies have made them. The bodies in question do not merely identify independent moral truths and report them; their pronouncements actually affect what moral obligations people have.
I emphasize that these conditions are jointly sufficient conditions for the existence of a legal system, not necessary conditions. I leave open whether there could be legal systems that do not meet one or more of these conditions (e.g. whether there could be a legal system with a legislature and no judiciary). I also emphasize that these conditions are conditions for the existence of a legal system. They do not claim that an individual rule must be morally binding to count as a law. Legal systems typically have some laws that do not alter people’s moral obligations, in some cases because these laws do not aim to change people’s moral obligations, in other cases because they try and fail. Such laws may count as laws not because they are morally binding or because they are binding in some other way, but because they are products of the same legislative process that in other cases produces morally binding rules with characteristically legal content and applicability.
It is clear that anti-positivists can accept these conditions for the existence of a legal system. Despite the conditions’ reference to morality, positivists should also be able to accept them, since these conditions are supposed to be sufficient conditions for the existence of a legal system, not necessary or constitutive conditions. Positivists can hold that any system of legislative and judicial bodies that satisfies (1)–(3) would have all of the non-moral, social features that it must have to count as a legal system. Most notably, the system should satisfy positivist requirements concerning the efficacy of law. For a system of legislative and judicial bodies to alter people’s moral obligations, the system and its rules and decisions (or some core subset of its rules and decisions) must be widely treated as authoritative. If no part of a would-be legislature’s output were widely respected, the would-be legislature (p.188) could not alter people’s moral obligations; it could only describe them. So positivists and anti-positivists alike can accept (1)–(3) as jointly sufficient conditions for the existence of a legal system. They will disagree about what facts make a legal system exist and about what features of a legal system distinguish it from other, less binding systems of rules. The anti-positivist will hold that the distinctive way in which legal systems can alter people’s moral obligations distinguishes them from other normative systems. The positivist will hold that legal systems are distinguished by the way in which their rules are socially regarded as binding.
2.2. Meeting the Conditions
Having outlined a set of jointly sufficient conditions for the existence of a legal system, I shall argue that in a society of morally very good people, an entirely non-coercive system can satisfy these conditions. An entirely non-coercive legal system is thus a conceptual possibility. Though there are various forms of law that a non-coercive system could make, I shall concentrate on property law. A society’s system of property provides rules for the allocation of its physical resources.12 To have a property interest in an object is to have an entitlement to use it in some way or set of ways, free from the interference of others. Property is a form of cooperation in which governments are characteristically involved. All actual effective governments make laws regulating the acquisition and transfer of property (e.g. about the transfer of property by inheritance) and specifying what rights come with property interests of different sorts (e.g. what rights landlords and tenants have). All actual effective governments also settle disputes about property, such as boundary disputes between neighbors. Moreover, governments are justified in making property law, since it serves morally important purposes. Whether every society needs a system in which private property is the dominant form of property ownership is debatable, but it is clear that every society needs some system for the allocation of physical resources.13 Having widely respected property rules is important for the functioning of a sophisticated economy. Having a system of property is also important for individual autonomy. If I have no property interests, including property interests in common or collective property, others are entitled to interfere with anything I may choose to do with any physical resource. All human activities require the use of physical resources. (Some activities, like thinking and dancing, do not directly involve the use of movable objects, but even these activities involve the use of space.) Individuals need property in order to have spheres (p.189) of possible action in which they can make choices without needing the permission of others.14
Under ideal social circumstances, it is possible for legislative bodies to make morally binding rules of property without using or threatening coercion and without authorizing private parties to coerce. To see this, imagine a society made up of morally very good people, people who are strongly committed to treating each other justly, competent at moral reasoning, and free of weakness of will. Suppose that this society has a deliberative body that has published a code of property rules and that periodically revises this code in response to objections and concerns. A large majority of people follow the code because they recognize the importance of having generally accepted property rules and they think the rules set out in this code are good. For the rule-making body to have the moral authority that legal systems typically claim to have, its code must be morally binding on everyone, whether or not they happen to approve of the code.15 The mere fact that a large majority of the population has embraced the code gives everyone a strong reason to follow it. It is important for there to be a single system of property rules that everyone follows, at least with respect to uses of important resources that exclude or interfere with others’ use of those resources. Once a system of property rules has obtained widespread acceptance, refusing to follow it is justified only if there is a compelling reason to reject it in favor of some other system.
People could have two sorts of reasons for preferring another system of rules to the one that is widely accepted. First, people could have purely self-interested reasons for preferring another code. A commitment to treating others fairly does not exclude reasonable self-interest. But the mere fact that the generally accepted system of property is less to one’s advantage than some other possible system does not justify refusing to follow the accepted system. The alternative system one prefers would disadvantage others if it were adopted. If people were entitled to disobey existing rules merely because some other system would be better for them, there could be no system of property rules that everyone is morally required to obey. That the current system is not to one’s advantage could justify disobedience only if the disadvantage one experiences is unfair.
Moral reasons, such as considerations of fairness, are the second sort of reasons people could have for preferring another code to the currently accepted one. If either the content of the code or the process by which it is created and amended is grossly unjust, people may be justified in refusing to follow it even if there is no significant chance that doing so may help bring about consensus behind a better code. Suppose that the code’s failings are not this great. It is morally better for everyone to follow the existing code than for some to follow it and for others to (p.190) follow another code or no code at all. Suppose further that most people accept the established legislative process as the only legitimate way of changing the rules of property. Disobedience will thus have no significant chance of helping to bring about consensus behind better rules. Under these circumstances, disobeying the generally accepted rules of property on moral grounds will be unjustified. So the legislature in this society makes rules of property that are morally binding on everyone, including people who disagree with them.
A society of morally very good people could also have a judiciary that makes morally binding decisions about property disputes. Morally very good people can have good faith disputes about who is entitled to use what resources. They can disagree about relevant matters of fact, and they can disagree about the interpretation of the generally accepted rules of property. Since they are competent moral reasoners, their views will be plausible even if they are false.16 If two people who are competent moral reasoners have a good faith dispute about whether one of them may use a particular resource in a way that excludes the other’s use of it, it is unethical for the first party simply to start using the resource or for the second party simply to prevent this use. Morally, the parties must reach agreement, either by discussing the issue together or by submitting the issue to the decision of a third party. Suppose they cannot agree either on the substance of the issue or on a private party they both trust to settle the dispute fairly. If the generally accepted code of property identifies a court system with the power authoritatively to resolve disputes about the code’s application, the parties will be morally required to submit their dispute to this court system. Unless the courts issue a grossly unjust ruling—which they will not, assuming that the rules of property and the rules of civil procedure are not grossly unjust and that the courts perform their duties in good faith—both parties will be morally required to comply with the court’s decision even if they disagree.
The society just described has a judicial system that makes morally binding decisions in disputes about property. It also has a legislature that makes morally binding property rules. These rules and decisions concern a subject matter that is regulated by the legislative and judicial branches of government in every society that has an effective government. So condition (1) is satisfied. The rules and decisions are morally binding on everyone in the society, so condition (2) is satisfied. The rules and decisions are binding in part because the legislature and the courts made them; the legislature and the courts have the power to change what moral obligations people have. So condition (3) is satisfied. This society therefore has a legal system. Yet neither the legislature nor the courts ever uses or threatens coercion, nor do they authorize any other party to coerce on their behalf. Thus, it is conceptually possible to have a legal system that lacks coercive enforcement of law.
Having shown that there can be law without coercion, I now turn to my central claim, that a body can be justified in making law without having an entitlement to enforce law coercively. This section will argue that there is no strictly necessary link between the entitlement to make law and an entitlement to coerce. It is possible for there to be a society in which a legal system is entitled to make law, but no one has any entitlement to enforce law coercively. I will spend some time defending this claim because there are plausible reasons for thinking that there is an entitlement to enforce law coercively in any society, including societies of very good people. In particular, one might think that a law-making body is entitled to establish a coercive enforcement mechanism even if this mechanism would not address a need. One might also think that every society needs coercive enforcement to deal with lawbreaking that stems from good-faith disagreement. Finally, one might think that people need a coercive assurance that others will respect their rights whether or not violations are likely. Since these arguments are supposed to demonstrate the existence of an entitlement to coerce even in the best social circumstances, examining their application to a society of very good people will show why they are unsound.
3.1. An Entitlement to Coerce Needs Justification
One might think that a legal system has an entitlement to authorize coercive enforcement of law even if coercive enforcement is not practically necessary. Sometimes a political body is entitled to do something it has all-things-considered reason not to do. In a society of people who are strongly disinclined to break the law, there may be policy reasons not to establish a coercive enforcement mechanism. Building prisons and hiring police officers may be a poor use of public resources, and it may not be a good use of the legislature’s time to set penalties for violations of law. There could still be an important sense in which law-making bodies are entitled to enforce law coercively or to authorize other parties to enforce law on their behalf.
A legislature can also have an entitlement to coerce in the face of certain sorts of moral reasons against coercive enforcement. Law-making bodies could have a general entitlement to enforce law coercively even if there were weighty moral reasons not to enforce particular laws coercively. For instance, a government might have reason not to enforce a particular law coercively because coercive enforcement would deter innocent and desirable conduct that is in the neighborhood of prohibited conduct. A law-making body might have an abstract entitlement to coerce though none of the available forms of coercive enforcement are morally permissible. Imagine, for instance, that a society’s severely limited resources made corporal punishment, public humiliation, and confinement under inhumane conditions the only available sanctions. But a law-making body lacks an entitlement to coerce if there are unmet moral objections to that body’s use or threat of coercion (p.192) as such.17 It is incoherent to say that a body has an entitlement to coerce though there are unmet moral objections to the body’s use of coercion in any form and for any purpose.
There is in all societies a pro tanto objection to coercive enforcement of law—an objection that rarely gets its due. Most explanations of governments’ entitlement to coerce focus exclusively on showing either that prospective offenders have no legitimate objection to state coercion, or that the interests of the victims of legal wrongdoing outweigh or override the objections of prospective wrongdoers. An adequate explanation of government’s alleged entitlement to coerce must also address the effects of governmental coercion on those who do not violate the law and have no desire to do so. Any coercive enforcement mechanism, however well designed and however virtuous its agents, sometimes interferes with the entitlements of innocent persons. Interference with entitlements of the innocent is not merely a feature of particular forms of governmental coercion. It is a necessary feature of coercive enforcement. This does not show, of course, that coercive enforcement is impermissible. But it shows that coercion stands in need of justification. It is necessary to show that the problem coercion solves is more serious than the problem it introduces.
There are broadly two ways in which laws can be enforced coercively, and they interfere with the entitlements of the innocent in different ways. One way of enforcing a law coercively is to threaten a sanction for non-compliance. The threat of sanctions for non-compliance sometimes interferes with the freedom of the innocent because whoever decides when sanctions should be applied will inevitably sometimes apply a sanction to an innocent person. Such errors need not reflect malice or negligence on the part of law enforcers; they can result from good faith mistakes. If a coercive enforcement mechanism is meant to address negligence and intentional wrongdoing, as well as legal wrongs that result from good-faith, non-negligent error, we must also be concerned about negligence and malice on the part of those who make accusations of legally cognizable wrongdoing, those who make decisions about guilt or innocence, and those who decide what sanctions to impose. Legal systems can be designed to deter false accusations and to prevent people from being wrongly convicted or found liable. However carefully a legal system is designed, inevitably the evidence will sometimes support imposition of a legal sanction on a person who is in fact innocent. Assuming that legal sanctions imposed on innocent people deprive those people of things to which they are entitled, the imposition of coercive sanctions necessarily interferes with the entitlements of the innocent.
The other form of coercive enforcement is the direct application of force to compel compliance. Direct application of force includes application of force against someone’s person—for instance, dragging a trespasser off private property. It also (p.193) includes at least some other actions that would, outside of the context of law enforcement, violate a right of the actions’ target. For instance, police use force when they break into a private home to retrieve stolen property or to obtain evidence of crime. When direct application of force is used to enforce a law incorrectly, or when it is used to enforce an incorrect governmental order, the innocent target of enforcement suffers a double injury. If the police drag someone away from a place he was legally entitled to occupy, they do not only deprive him of his entitlement to be where he was; they also violate his entitlement not to be manhandled. If a court incorrectly finds that an object has been misappropriated, and it orders police to retrieve it from a private home, the residents do not only lose an object they were entitled to possess; they also suffer a violation of privacy. Now, in a society that lacks coercive enforcement of law, people can be subject to incorrect orders from courts, police, or other public officials. If these orders are made with morally legitimate authority, people may be morally required to give up things they are entitled to, and if they are morally upright, they will comply. But they will not suffer the additional injuries they would have suffered had the incorrect orders been enforced by direct compulsion. Because some governmental orders will inevitably be incorrect even if government officials always perform their duties in good faith, the direct application of force to compel compliance with orders will inevitably sometimes violate the entitlements of the innocent. Moreover, it will sometimes violate entitlements that would not be violated in a legal system that lacks coercive enforcement.
Because both coercive threats and direct application of force necessarily risk violating the entitlements of the innocent, no government can have an entitlement to coerce unless establishing a coercive enforcement mechanism would solve a problem that matters more than the problems it would create. Now, the most obvious problem a coercive enforcement mechanism could address is the problem of lawbreaking that arises from human moral failings. There are two other problems coercive enforcement could address, however. These are problems that could arise even in a society of morally very good people. One is the problem of moral disagreement. Competent moral reasoners can make errors of moral judgment. In particular, they can mistakenly believe that a particular law or governmental order is not morally binding, or worse, that they have a moral obligation to disobey. One might think that coercive enforcement of law is needed to address violations of law that result from such errors of moral judgment. Second, one might think that public coercion provides an important assurance even if citizens would not in fact violate the law in the absence of public coercion. Coercive enforcement provides each member of society an assurance that others will not violate their legal rights without providing compensation. Individuals need this assurance even if there would be no rights violations without it because they should not have to depend on others’ good will. The next two subsections will show that in a society of morally very good people, neither of these arguments succeeds in showing that there is an entitlement to enforce law coercively.
Morally very good people, as I have characterized them, are not immune from moral error. They reliably reason competently about moral matters, but because some moral questions are hard, competent moral reasoning can lead to plausible but incorrect judgments. Competent moral reasoners can reasonably disagree about various moral questions, including the justice of an existing property system. Even if the currently accepted system of property rules is not unjust in any way that justifies disobedience, a competent moral reasoner might reasonably think otherwise. Likewise, a competent moral reasoner might erroneously but reasonably think that a court’s decision is so unjust that disobeying the court is morally permissible. One might think that coercive enforcement of law is needed, even in a society of morally very good people, to deal with well-intentioned but wrongful violations of the law.
The first point to make about this line of argument is that there is only a limited range of cases in which fully competent moral reasoners will disagree with each other not only about what the law should be, but also about whether one should obey the law even if the law is not what it should be. Though competent moral reasoners can make errors of moral judgment, they do not make gross errors. It would be a gross error to think that it is unimportant for a society to have a generally respected system of rules for the allocation of physical resources. It would also be a gross error not to recognize that it is often morally better to follow the generally accepted rules of property than to follow the rules one thinks people should be following. There are many possible systems for the allocation of physical resources. Some of them are morally intolerable. All competent moral reasoners recognize, however, that a variety of systems of property rules are morally tolerable in the following sense: it is morally better for one of these systems to have everyone’s compliance than for some to follow that system while others follow the rules of another system they regard as morally better.
There are some cases, however, in which competent moral reasoners can disagree not only about what rule they should follow, but also about whether following the same rule as everyone else is more important than following the best rule or the right rule. Imagine, for instance, that there is disagreement about whether it should be possible to own animals of a certain species. The animals of this species have a high cognitive capacity, though they do not have the same mental powers as human beings. Most people believe that it is acceptable to keep these animals in captivity under good conditions, and that it would be wrong to release a captive animal without the putative owner’s consent. Some animal rights activists believe that animals of this species are close enough to humans in cognitive capacity that keeping them in confinement is tantamount to kidnapping. Releasing captive animals of this species without the alleged owner’s consent is not only morally permissible; it is required. The activists disagree with the majority not only about what the rules of property should be, but also about whether in this case it is important to follow the same rules as everyone else if the rules are bad. Suppose that (p.195) the animal rights activists are mistaken in these moral judgments. Though the animal rights activists are in error on this matter, their error does not reflect moral incompetence or lack of moral seriousness. It is the sort of error that competent moral reasoners can make.18 Many of the animal rights activists will disobey their society’s property law, though they are not morally justified in doing so. Governmental coercion would help to protect animal owners from the well-meaning but misguided animal rights activists.
Though moral disagreement of this sort can lead to morally wrongful violations of law, even in a society of very good people, it does not justify establishing a coercive enforcement mechanism. When private individuals disagree with the public both about what the law should be and about whether individuals should obey the law even if it is the wrong rule, sometimes the dissenters will be wrong. Sometimes, however, the dissenters will be right. Perhaps well-designed public bodies are more likely than morally competent private individuals to be right about whether disobeying a given law would be morally justified. Even well-designed public bodies will be far from infallible about this matter, however. When dissenters correctly believe both that a law is bad and that it should be disobeyed, even though it is the law, it would be wrong to attempt to coerce the dissenters to comply. An attempt to coerce compliance would be an attempt to coerce people to do something they are entitled not to do. It may even be an attempt to coerce people to do something wrong.
Thus, if controversial laws are enforced coercively, sometimes the enforcing body will prevent wrongs or address wrongs that have occurred, but sometimes the enforcing body will commit wrongs itself. If controversial laws are not enforced coercively, there will be wrongs by private parties that go unaddressed, but no public body and no private party authorized to act on behalf of the public will be guilty of wrongful coercion. Perhaps there is reason to believe that the private wrongs coercive enforcement would address would be somewhat greater in number than the public wrongs coercive enforcement would involve. There is no reason to suppose, however, that in a society of competent moral reasoners, government would necessarily be right a great deal more frequently than dissenters. There is also no reason to suppose that the private wrongs coercive enforcement would address are more serious wrongs than the public wrongs coercive enforcement would involve. There are two reasons to think the public wrongs should matter more from the perspective of the legislature. First, there is at least arguably a general moral distinction between doing and allowing: it is worse to commit wrongs than to allow wrongs of comparable severity. Second, when public officials commit wrongs while acting in their official capacity, they purport to act on behalf of all (p.196) citizens. Because their wrongful acts implicate people other than themselves, they have more gravity than wrongs committed by private citizens not acting under color of law. The somewhat greater frequency of private wrongdoing does not clearly make up for the greater importance of wrongs committed by public agents. So although moral disagreement would lead to a limited amount of non-compliance with law in a society of morally very good people, that fact alone does not entail that there would be an entitlement to enforce law coercively.
3.3. Rights and Assurance
There is another reason one might have for thinking that a society of morally very good people would need coercive enforcement of law. One might argue that such a society needs a coercive enforcement mechanism to provide individuals assurance that others will not violate their rights without providing compensation. Arthur Ripstein offers an argument, on Kant’s behalf, that it would be important to have such a coercive assurance even under social conditions in which violations of law do not occur. Ripstein argues that any society needs public coercive enforcement of legal entitlements to property, as well as certain other legal rights of individuals, among them the right to bodily integrity.19 In order for us to enjoy external freedom, we each need to be able to set and to pursue ends of our own choosing. Our choice of what ends to pursue must not be “subject to the choice of another person.”20 I can set and pursue ends of my own choosing only if my use of my means, including my person and my property, is not subject to the choice of others. For instance, if you can decide to lock me in a room, my use of my person is subject to your choice. If you can decide to drive my car without my permission, my use of my car is subject to your choice. My use of my means is subject to your choice even if it is empirically unlikely that you will interfere. A slave does not enjoy external freedom if he has a benevolent master who allows him to make his own decisions and gives him access to resources with which to act. The slave is subject to his master’s will even if as an empirical matter he can expect his master never to forbid him from doing anything he wants to do or to require him to do something he does not want to do.21
For us to enjoy external freedom, then, our freedom needs to be subject to reciprocal limits. You must not be free to interfere with my use of my means, and I must not be free to interfere with your use of your means. Interferences with individuals’ use of their means must therefore be stopped. Violations of rights may be coercively stopped before they can be completed; for instance, a theft in progress may be forcibly interrupted. When a violation is completed, the offender must be forced to restore the victim’s means. For instance, a thief must be compelled to return stolen property or, if this is no longer possible, to pay monetary compensation. Now, limits on our freedom are not reciprocal if rights are enforced privately, since some people are stronger, smarter, or more willing to confront (p.197) others and thus more able to enforce their rights effectively. So for our freedom to be subject to reciprocal limits, the government must enforce the rules that specify the limits on our freedom.22 In particular, it must provide a coercive system by which those whose rights are violated can have their means restored to them. It would not be enough to have a non-coercive court system that hears disputes about alleged rights-violations and makes judgments about whether compensation is owed, and in what amount. There needs to be a coercive mechanism for ensuring that court-ordered damages are paid. On Ripstein’s view, such a system is needed even in a society in which people are seldom tempted to wrong each other, whether because they have little to gain from wrongdoing or because they have strong moral commitments.
The difficulty with this argument is that it is unclear what sort of reciprocal limits on freedom we need if we are not to be subject to each others’ wills. The limits we need could be moral limits: I enjoy external freedom only if you normatively may not interfere with my use of my means and I am entitled to compensation if you do interfere. Alternatively, the limits we need could be empirical limits: I enjoy external freedom only if it is empirically not possible for you to interfere with my use of my means without providing compensation. If the limits we need to enjoy external freedom are purely normative limits, then at least in ideal social circumstances, external freedom does not require coercive enforcement. No coercion is needed for people to have moral entitlements to use resources without others’ interference. No coercion is needed for people to be morally entitled to compensation if others do interfere with their means: if one person in a society of very good people takes another’s coat by mistake, the accidental thief is morally required to return it when she learns of her mistake. Moreover, no coercion is needed for these moral entitlements to be socially and politically recognized.23 Providing coercive remedies for wrongdoing is one way of publicly recognizing an entitlement, but it is not the only way. In a society of very good people, entitlements under the property system are publicly recognized in the rules of the code and in documents such as wills and deeds that the system recognizes as having legal force. So if coercive enforcement of law is needed even in ideal social circumstances, as Ripstein claims, this must be so because the limits on freedom that need to be established are at least partly empirical limits, not merely normative limits.
If external freedom requires that it be empirically impossible for anyone to interfere with other people’s use of their means without being forced to restore them, then external freedom is unattainable. Even under the best legal systems, (p.198) people can sometimes get away with violating others’ rights without providing compensation because there is insufficient evidence to support a judgment against the wrongdoer. Moreover, if it is a contingent matter whether private citizens will conform to their legal duties, it is also a contingent matter whether rights enforcers will perform theirs. Judges can be corrupt, incompetent, prejudiced, or negligent. Juries can find for a sympathetic defendant even if the evidence supports a finding for the plaintiff. For both of these reasons, there is no way to eliminate the possibility that one person could interfere with another’s use of her means. In a society of morally flawed people, a good legal system can make unjustifiable violations of rights less common, and it can make it more likely that people whose rights are violated will receive compensation, but it cannot ensure that every unjustifiable violation of a right will be compensated. In a society of morally very good people, uncompensated violations of rights are always possible, even if they never actually occur. Coercive enforcement cannot remove this possibility.
So in a society of morally very good people, a coercive enforcement mechanism would not give people anything that they did not already have. Even in the absence of coercive enforcement, people could have socially and politically recognized rights that others are morally required to respect. Because of their strong moral commitments, violations of these rights would not occur, except in cases in which the obligation to respect these rights is legitimately controversial—and in these cases, coercive protection of rights is unjustified. It remains possible that others could violate their rights without providing compensation, but no coercive enforcement mechanism can eliminate this possibility. Thus, in a society of morally very good people, a coercive enforcement mechanism would serve no good purpose. It is necessary neither to provide citizens with a coercive assurance mechanism nor to address the problem of reasonable moral disagreement. In a society of morally very good people, then, there is no way to overcome the pro tanto objection to the establishment of a coercive enforcement mechanism—that it necessarily places burdens on the freedom of the innocent. Therefore, the legislature in such a society would have an entitlement to make law, but it would have no entitlement to enforce law coercively or to authorize others to enforce law coercively.
One might be tempted to characterize the powers of the legislature in a society of very good people differently: the legislature has an entitlement to coerce (or to authorize coercion), but under current social circumstances, it should not exercise that entitlement. This position does not make sense. The fact that coercive enforcement mechanisms interfere with the freedom of the innocent is no more contingent than the fact that coercive enforcement mechanisms interfere with wrongdoing. The interests of potential victims of government abuse, error, and intrusiveness are no less important than the interests of potential victims of private wrongs. To show that a legal system has an entitlement to coerce, it is necessary to show that its benefits are more important than any moral objections to its fundamental features. In a society of very good people, this cannot be done. Perhaps if social circumstances were to change, law-making bodies would acquire an entitlement to coerce. In the happy circumstances people find themselves in, neither the legislature nor any other party is in any sense entitled to enforce law coercively.
Actual human societies, of course, are not made up entirely of morally very good people. I do not wish to deny that in a large society made up of morally flawed human beings, government sometimes has an entitlement to enforce law coercively, and that it sometimes has good reason to make use of this entitlement. But there is no justification for a general presumption that any government in a non-ideal society is entitled to coercively enforce any law it is entitled to make. That government has an entitlement to coercively enforce a legal requirement on conduct does not follow from the fact that it has the power and the entitlement to impose this requirement together with abstract truisms about human nature (e.g. that human altruism is limited and that humans are vulnerable to weakness of will).24 To show that government is entitled to coercively enforce a requirement on conduct, it is necessary to do an empirical investigation that goes beyond truisms and examines the extent and character of citizens’ moral failings in an actual society. It is also necessary to ask the moral question whether the need to address violations of this particular legal requirement justifies establishing a coercive enforcement mechanism that inevitably puts the innocent at risk of being injured by the government or its agents. Though there may be no actual human societies in which government lacks the entitlement to coerce altogether, there are imaginable social circumstances in which government’s justified power to impose legal requirements on conduct is considerably more extensive than its entitlement to enforce these requirements on conduct coercively.
There are two reasons one might have for believing, to the contrary, that in any society not made up entirely of superhuman angels or saints, government will be entitled to enforce whatever requirements on conduct it is entitled to impose. First, one might think that in a society not made entirely of morally very good people, it would not be possible to establish laws that are morally binding. The existence of unpunished free-riders would undermine others’ obligation to obey the law. Second, one might think that the presence of people who are inclined to break the law unjustifiably will necessarily give the government an entitlement to address lawbreaking coercively. In any imaginable human society, the threat of unjustifiable lawbreaking will present a problem that justifies burdening the innocent with the risk of being mistakenly or maliciously subjected to legal sanctions or governmental force. I will address these arguments in turn.
4.1. Free-riding and the Moral Force of Law
The concern about free-riding begins with the thought that people have a moral obligation to do their part in a cooperative activity only if a cooperative activity is actually taking place. For a cooperative activity to take place, enough people need to (p.200) be doing their part in that activity. So for people to have a moral obligation to do their part in a particular system of property rules, for instance, there need to be enough people who actually follow the system’s rules. If there are not enough people following the rules of a particular system of property, then nobody has an obligation to follow that particular system, though everyone has a duty to try to bring about the existence of a system that is widely followed. Now, perhaps a system of property that lacked coercive enforcement might be widely obeyed on an island occupied only by very good people. In a society of flawed human beings, however, many people are less than fully committed to treating each other fairly. Thus, a non-coercive system of property could not retain sufficient obedience for it to count as the established system. For rules of property to remain morally binding over time in a society of flawed human beings, there must be a coercive assurance that people will generally continue to comply. Or so one might argue.25
The force of this argument depends on how severe people’s moral failings actually are. If these moral failings are mild enough that a non-coercive system of property could secure widespread compliance, though not universal compliance, that system could be morally binding. Imagine a society of people who fall short of sainthood but are, as a group, fairly good. A small number of them do not understand the moral reasons for respecting a system of property. The rest are committed in principle to respecting each others’ private property and to respecting limits on the use of public property, but on rare occasions they suffer from weakness of will. Unjustifiable intentional violations of the rules of property will occur in this society, but they will not be common. For such people, moderately flawed though they are, a morally binding non-coercive system of property would be possible. That people occasionally fail to do their part in a cooperative activity and that a few people persistently free-ride does not entail that no cooperative activity is taking place, nor does it undermine people’s obligation to do their part in that activity. To determine whether morally binding rules of property can be made non-coercively, it is necessary to have empirical evidence about how much compliance could be achieved without the use or threat of coercion. The abstract truism that human beings are sometimes inclined to break the law does not entail that law cannot be morally binding without coercive enforcement.
One may object to this point with a further argument. There are some forms of cooperation for which the existence of even a few free-riders undermines everyone else’s obligation to comply. Imagine, for instance, that businesses in a highly competitive market are told to pay a tax in support of important public goods. Suppose that the tax is not coercively enforced. If a small number of business owners, caring little for the public good, opt not to pay the tax, these businesses will have a competitive advantage. The owners of other businesses in this market may (p.201) very much want to pay the tax, but paying the tax may put their businesses at risk, depending on how burdensome the tax is and on the economic circumstances these businesses face. To borrow Hobbes’s words, these businesses fear that by paying the tax they will make themselves “prey to others.”26 Suppose that economic circumstances are such that businesses will, in fact, be putting themselves at risk if they pay the tax and some of their competitors do not. Then it would be unreasonable to demand that a business pay the tax unless business owners can be reasonably confident that other businesses are paying. In a society of saintly people, business owners could perhaps be confident that their competitors voluntarily comply with their tax requirements. In a society of flawed human beings, we cannot be so confident that competitors are voluntarily complying with the rules. So in order for businesses to have a morally binding obligation to pay this tax, there needs to be an enforcement mechanism to ensure that payment is compatible with a business’s ability to compete on fair terms.
There are two problems with this argument. First, coercive enforcement is not the only way of providing businesses with an assurance that competitors are complying with a tax. The government could call for businesses’ taxes to be paid in regular installments and publish information about businesses’ tax payments. This would provide businesses with information about which of their competitors pay their taxes and which, if any, evade it. The second problem with the argument is that it only applies to free-riding by businesses in a competitive market. By way of contrast, suppose that a tax paid by individual consumers is not coercively enforced, and that 90 percent of individuals make a good faith effort to pay the amount they owe. They would not thereby make themselves prey to the non-compliant 10 percent. Even in lean times, most consumers do not compete with each other for survival. Having more money is good, but it does not make the difference between survival and death. That 10 percent of consumers shirk their obligation to pay a tax does not undermine other taxpayers’ moral obligation to pay.27 Many of the rules of property are more like taxes on individual consumers than they are like taxes on businesses in a competitive market. Though very widespread non-compliance with a rule would perhaps undermine the obligation to obey it, a moderate amount of non-compliance would not. If everyone regarded themselves as perfectly free to ride any bicycle found on the street (even if doing so involved picking a lock), perhaps nobody would be under moral obligation not to do so. Bicycles would, in effect, be common property. If only a small but substantial minority has this attitude toward (p.202) bicycles, however, the moral obligation to treat bicycles as private property is not undermined. That some people steal bicycles with impunity does not give me justification for engaging in theft myself, nor does it undermine any of my other obligations with respect to the property system.28
In order to answer the question whether non-coercive rules of property can be morally binding, it will be necessary to examine different parts of the property system separately. Some rules or sets of related rules will be morally binding if they are widely followed; the obligation to follow them is not undermined if the rules are not universally followed or if there is no effective enforcement mechanism. Other rules or sets of related rules will be morally binding only if they are universally followed or if they are enforced (coercively or otherwise) in such a way that compliance is appropriate even in the face of non-compliance by competitors. In order to determine which category a rule or a set of related rules falls in, it is necessary to address both the empirical question what competitive pressures agents face as market participants and the moral question what risks a market participant can be reasonably asked to assume. Having determined what level of compliance is required for a rule or set of related rules to be morally binding, there will be a further empirical question whether this level of compliance can be achieved without coercive enforcement.
I cannot pretend to have enough knowledge of psychology or sociology to answer definitively the question how much obedience non-coercive rules could secure, but I will mention some motives that could lead people to comply with law voluntarily. First, though there are often powerful incentives of self-interest to violate laws, sometimes violating a law is not in one’s self-interest even if legal and social sanctions are not taken into account. For example, it is typically not in a radio broadcaster’s interest to transmit on frequencies that others in the area have an exclusive right to use. If a broadcaster transmits on a frequency that someone else is legally entitled to use, and that person is making use of her entitlement, neither party’s use of the frequency is likely to be a success. Second, when violations of a rule are visible to the public, concern for one’s reputation can provide an incentive to follow the rule, and this incentive is not always a coercive incentive. No doubt some informal social responses to perceived wrongdoing are coercive, but it is not clear that all of them are. (If prospective customers avoid businesses they regard as dishonest, not to punish those businesses but to avoid being cheated, is the resulting pressure coercive?) Furthermore, concern for one’s reputation is not only concern about social sanctions or rewards that may result from perceived good or bad behavior; people care about other’s good opinion for its own sake.29 Finally, it is important not to underestimate the significance of motives other than narrow self-interest. Even people who are not of saintly character often obey the law either out of the belief that there is (p.203) a moral obligation to obey or because they believe that there are independent moral reasons to act as the law requires.30
In light of these considerations, I think it is likely that a large human society could have a system of property rules in which all the rules are morally binding and at least some of the rules are not coercively enforced. The more important point, though, is that a substantial amount of empirical evidence is needed to settle the question whether and when non-coercive rule making could be morally binding in a large human society. Abstract truisms about human nature do not settle the question.
4.2. Responding to Unjustified Lawbreaking
Given that government can make a morally binding law without coercively enforcing it, there is a further question about whether the government has an entitlement to enforce it coercively. One might think that that the presence of citizens who are disposed to break the law unjustifiably will always present a problem that provides a justification for establishing a coercive enforcement mechanism. To show that government has an entitlement to coerce, however, it is not enough to show that coercive enforcement would solve a problem. It is necessary to show that coercive enforcement would solve a problem that matters more than the burdens a coercive enforcement mechanism necessarily places on the innocent.
Showing that the value of coercive law enforcement justifies the burdens it imposes on the innocent is not straightforward either normatively or empirically. It is not straightforward normatively because the question must not be addressed as a utilitarian calculation. One cannot simply add up the harms a coercive enforcement mechanism could be expected to inflict, according to some measure of total harm, and compare this with the harm coercive enforcement can be expected to prevent.31 Even if such a comparison were possible, it would be inappropriate for three reasons. First, it is necessary to consider the way in which the burdens of coercive enforcement would be distributed. If innocent people of a particular class or social group would be disproportionately likely to be punished or subjected to compulsive force, this must be taken into account. Second, the injuries a coercive enforcement mechanism inflicts on the innocent, unlike most injuries perpetrated by lawbreaking individuals, come with the stigma of the government’s judgment that one is guilty or likely guilty of an offense.32 The stigma associated with a criminal conviction is most serious, but there is also a negative governmental judgment associated with other coercive sanctions, such as fines and punitive damages, and with direct applications of governmental force (e.g. being dragged away by the police). Finally, it is at least arguably more difficult in general to justify (p.204) doing injury than it is to justify merely allowing harm or injury to occur. Some defenders of a robust coercive criminal justice system question the relevance of the doing-allowing distinction to the unintentional punishment of the innocent. Instituting a practice of punishment knowing that some innocent people will be mistakenly punished is not morally comparable to intentionally punishing particular people known to be innocent.33 That said, there may well be a moral difference between doing something that risks injuring innocents without intending injury to anyone (e.g. driving a vehicle) and intentionally injuring people, believing each of these people to be guilty but knowing some of them will in fact be innocent.
Showing that coercive law enforcement presents a greater threat to the innocent than would its absence is also a complex matter empirically. The relative magnitudes of these two threats depend on precisely what moral failings people in a given society have. Imagine a decidedly non-angelic society deeply divided on ethnic lines. Because of the prejudices that infect this society, many people in the majority group believe that members of minority groups break the law far more often than they actually do. As a result, police officers of the majority group are far too quick to use force against members of minority groups. Criminal or civil juries dominated by the majority group find against members of minority groups on evidence that an unprejudiced juror would consider flimsy at best. Juries that are not dominated by members of one ethnic group tend to hang even when the evidence clearly points toward the defendant’s guilt, since most people are inclined to view others of their own ethnicity charitably. In this society, the coercive legal apparatus would inflict extensive wrongs on innocent people, and it would be largely ineffective at preventing, compensating, or justly punishing wrongdoing by private parties. So it would be a mistake to think that in any society that includes morally flawed people, a well-designed coercive executive would present less of a threat than would individual lawbreaking unchecked by a coercive executive. Which threat is greater depends on which moral flaws people have. If people’s flaws make misuse of public coercive power a greater threat than a given form of wrongdoing by private parties, government will lack an entitlement to address that form of wrongdoing coercively.
Even in a society of morally flawed people, then, the entitlement to make law does not straightforwardly entail an entitlement to enforce. No doubt actual governments sometimes do have an entitlement to enforce law coercively, but this entitlement may or may not extend to all of the laws governments are entitled to make. How extensive this entitlement is depends on many empirical facts, among them what non-coercive incentives there are to comply with various laws, what pressures people feel that could justify non-compliance in the absence of coercive enforcement, what temptations people feel most acutely, what temptations people are able to resist, and what prejudices and vices afflict people acting as agents of the state. The extent of the entitlement to enforce law coercively may wax and wane as social conditions change.
That the entitlement to make law does not entail an entitlement to enforce law coercively has two important consequences for the way political argument should be conducted. First, this conclusion shows that a common form of argument in both political philosophy and public political discourse is dubious at best. Both economic libertarians and civil libertarians often criticize proposed legal requirements by arguing that coercive enforcement of these requirements would be objectionable. Such arguments presuppose that the government is entitled to impose these legal requirements only if the government is entitled to enforce them coercively. This assumption is not true in general. Sophisticated libertarians need to consider whether they should advocate limits on the reach of law or whether they should instead advocate more non-coercive exercises of government’s law-making power. Note that the libertarian argument does not turn on the claim that every direct legal requirement on conduct must have a coercive remedy for non-compliance. It depends instead on the weaker claim that government is entitled to use coercion to enforce every requirement on conduct it is entitled to impose—and, conversely, that government is not entitled to impose any requirement that it is not entitled to enforce coercively. The conclusion that the entitlement to make law does not entail an entitlement to coerce is thus more significant for political discourse than the weaker conclusion that laws lacking coercive enforcement are possible.
Awareness that the entitlement to make law does not always come with an entitlement to coerce can also help us to make more refined diagnoses of injustice. As an example, many economic egalitarians argue that it is unjust to have a coercively enforced property system unless that system satisfies certain requirements of distributive justice.34 If such systems are indeed unjust, we can think more clearly about the nature of this injustice if we recognize that the justification of property law does not entail the justification of coercive enforcement of property law. We may find that the injustice of a coercively enforced property system lies solely in the wrongful use of governmental force. We may find instead that the injustice of an inegalitarian property system lies not only in what government forces people to do, but also in what it asks of them. A government that claims to issue morally binding laws and orders has a duty to ensure that those laws and orders reflect equal concern for all citizens, whether or not the laws are backed by the threat of force.
Alexander, L. (1983). “Retributivism and Inadvertent Punishment of the Innocent,” Law and Philosophy, 2: 233–46.
Aristotle (2002). Nicomachean Ethics. Translated by S. Broadie and C. Rowe. Oxford: Oxford University Press.
Dworkin, R. (1985). “Principle, Policy, Procedure,” in A Matter of Principle. Oxford: Oxford University Press, 72–103.
Dworkin, R. (1986). Law’s Empire. Cambridge, MA: Harvard University Press.
Dworkin, R. (2002). Sovereign Virtue: The Theory and Practice of Equality. Cambridge, MA: Harvard University Press.
Dworkin, R. (2006). Justice in Robes. Cambridge, MA: Harvard University Press.
Greenberg, M. (2004). “How Facts Make Law,” Legal Theory, 10: 157–98.
Hart, H. L. A. (1994). The Concept of Law. Second edition. Oxford: Oxford University Press.
Hobbes, T. (1994). Leviathan. Edited by E. Curley. Indianapolis: Hackett Publishing Company.
Kant, I. (1996). The Metaphysics of Morals, reprinted in M. Gregor (trans.), Practical Philosophy. Cambridge: Cambridge University Press.
Lamond, G. (2001). “Coercion and the Nature of Law,” Legal Theory, 7: 35–57.
Lippke, R. (2010). “Punishing the Guilty, Not Punishing the Innocent,” Journal of Moral Philosophy, 7: 462–88.
Nagel, T. (1975). “Libertarianism without Foundations,” Yale Law Journal, 86: 136–49.
Nozick, R. (1974). Anarchy, State, and Utopia. New York: Basic Books.
Rawls, J. (1971). A Theory of Justice. Cambridge, MA: Harvard University Press.
Raz, J. (1999). Practical Reason and Norms. Second edition. Oxford: Oxford University Press.
Ripstein, A. (2004). “Authority and Coercion,” Philosophy & Public Affairs, 32: 2–35.
Ripstein, A. (2006). “Beyond the Harm Principle,” Philosophy & Public Affairs, 34: 216–46.
Ripstein, A. (2009). Force and Freedom: Kant’s Legal and Political Philosophy. Cambridge, MA: Harvard University Press.
Sager, L. (1978). “Fair Measure: The Legal Status of Underenforced Constitutional Norms,” Harvard Law Review, 91: 1212–64.
Shapiro, S. (2011). Legality. Cambridge, MA: Harvard University Press.
Simmons, A. J. (1987). “The Anarchist Position: A Reply to Klosko and Senor,” Philosophy & Public Affairs, 16: 269–79.
Tyler, T. (1990). Why People Obey the Law. Princeton: Princeton University Press.
Waldron, J. (1988). The Right to Private Property. Oxford: Oxford University Press.
Wertheimer, A. (1977). “Punishing the Innocent—Unintentionally,” Inquiry, 20: 45–65.
* I am grateful to Jorah Dannenberg, Joshua Dienstag, Sonny Elizondo, Barbara Herman, Louis-Philippe Hodgson, Yannig Luthra, Eliot Michaelson, Letitia Meynell, Calvin Normore, Arthur Ripstein, Stefan Sciaraffa, Seana Shiffrin, Luca Struble, Wil Waluchow, Alan Wertheimer, and audiences at Northwestern and UCLA for their comments on this chapter and on earlier papers from which this chapter is derived.
(1) Hart notes, for example, that there is no coercive sanction directly attached to a law requiring someone who wants to make a will to get two witnesses’ signatures (1994: 28). Lawrence Sager provides another counterexample. The United States Supreme Court is bound by law in deciding cases, but its members are not subject to sanction if their opinions fail to follow the law. Congress can remove Supreme Court Justices from office by impeachment and trial, but presumably it is not in virtue of this rarely applied sanction that Justices are bound by law in making decisions (1978: 1222). Joseph Raz and Scott Shapiro have argued that is conceptually possible for there to be an entire legal system that lacks coercive enforcement. Raz (1999: 157–61); Shapiro (2011: 169–70).
(2) Arthur Ripstein does so in his defense of Kant’s political theory (2009; 2004). Kant expresses this view in the Metaphysics of Morals at Ak. 6:232 (1996: 388). Grant Lamond defends the related position that law necessarily claims the right to exercise coercion. He criticizes the view that law is necessarily coercive (2001).
(3) Requirements on conduct contrast with requirements that must be met to acquire, to transfer, or to exercise a legal power, such as the requirement that one have two signatures to make a valid will.
(4) Robert Nozick’s libertarian arguments take this form (1974: ix).
(8) Dworkin (1986: 93). Recently, Dworkin has taken a subtler view. He distinguishes the “sociological concept of law” from the “doctrinal concept of law.” The sociological concept of law, which is the concept relevant to the question whether a social practice counts as a legal system, “is not sufficiently precise to yield philosophically interesting ‘essential features’” (2006: 228). So there is no point in asking whether there could be a legal system, properly so-called, that lacks enforcement institutions (2006: 3).
(9) My argument resembles Raz’s in that it uses a thought experiment involving a society of morally very good people. Unlike Raz, I do not assume that such people would be disposed to comply with law; this is to be shown.
(10) The only people whose obligations these laws affect are the people who operate the criminal legal system. I do not claim that a society of morally very good people would have any use for laws criminalizing wrongful violence.
(12) Waldron characterizes property in a similar way (1988: 31–7). He notes that some systems of property allow people to have property interests in non-material objects, such as inventions and reputations. He argues that interests in material objects are more central to the concept of property because all human societies have felt the need to allocate scarce material resources. By contrast, not all societies have felt a need for a system of intellectual property or for property in other incorporeal objects (1988: 33–4).
(13) An example of a property interest that is not a private property interest is an entitlement to use the public roads.
(14) This view of the importance of property is inspired by Ripstein’s view. Ripstein claims that interfering with someone’s use of her property interferes with her ability to set and pursue her own ends (2006: 240–3). I agree with this, but I also take the stronger position that individuals need to have property in order to be able to set and pursue their own ends.
(15) The code may be only pro tanto morally binding; special circumstances may justify disobedience.
(16) I assume here that competence at moral reasoning includes not only an ability to draw plausible moral conclusions from non-moral facts, but also some degree of competence at assessing the non-moral facts relevant to moral questions.
(17) Likewise, private parties have no entitlement to enforce law coercively if there is an unmet moral objection to coercive law enforcement by private parties. The pro tanto argument against governmental coercion applies also to private coercive enforcement of law, except perhaps in the narrow case of self-defense against an imminent threat.
(18) This assumes that the animal rights activists are committed to limiting their use of violence. It is not plausible to think that one is justified in killing or injuring human beings in order to free captive animals. In general, when a great many competent moral reasoners think that doing X is impermissible, it is a gross error to believe that one is justified in killing or injuring others in order to facilitate doing X. That many competent moral reasoners believe that X should not be done ought to give pause to anyone who is tempted to use serious forms of violence to promote an opposing view. (I owe this point to Thi Nguyen.)
(22) Though Ripstein’s main argument is for the coercive enforcement of individual rights, it can be extended to an argument for coercive enforcement of other laws insofar as these laws are required for the establishment of a rightful condition in which limits on freedom are reciprocally enforced. For example, the state must redistribute wealth by means of coercively enforced taxation in order to prevent people in extreme need from becoming dependent on the will of other private individuals. (2009: 267–86); (2004: 33–5).
(23) Social and political recognition of rights matters; it is what the slave with the benevolent master lacks. Though this slave has a moral entitlement to decide for himself what to do, and though he can be confident that his master will never actually interfere, the slave’s entitlement to make his own decisions is not socially recognized.
(24) Hart identifies five such truisms: “human vulnerability,” “approximate equality,” “limited altruism,” “limited resources,” and “limited understanding and strength of will” (1994: 193–200).
(25) Rawls argues that even if everybody had a shared sense of justice and wanted to adhere to the existing system of cooperation, the legal system would have to threaten coercive sanctions (possibly mild sanctions) in order to secure stable cooperation over time. Without sanctions, people would be tempted not to fulfill their obligations because they would suspect that others are not doing their part (1971: 240).
(27) Another possible objection to a voluntary tax system is that it would put excessive demands on the will. Nagel argues that it is reasonable to impose a compulsory tax for the support of the poor but unreasonable to demand that people make voluntary financial contributions in the same amount. “The latter is an excessively demanding moral position,” he writes, “because it requires voluntary decisions that are quite difficult to make” (1975: 145). The problem Nagel describes would be a real problem in a non-coercive tax system, but it could be mitigated by structuring tax collection to make tax payment psychologically easier. For instance, it would probably be helpful to make use of automatic tax withholding from paychecks. I suspect that for most people it is psychologically harder to write a check than it is to refrain from demanding a refund.
(28) Furthermore, it is not unreasonable to demand that people refrain from stealing bicycles. It does not place excessive demands on the will to ask people not to steal even when they could get away with it.
(30) For evidence of this, see Tom Tyler’s study of several hundred Chicago residents’ compliance with commonly broken laws (1990).
(31) I do not mean to suggest here that retributive justifications of punishment should be off the table. Larry Alexander discusses retributivist approaches to tradeoffs between punishing the innocent and letting the guilty go unpunished (1983).