Abstract
Although understanding political obligations as content-independent is standard in the literature, this view cannot be defended. Central to this position is a single content-independent reason to obey virtually all laws because they are laws, as opposed to because of their particular contents. However, intuitively plausible and familiar this view may be, adequate foundational reasons are not to be had. My particular focus is “scope limitations” of content-independent moral reasons. Content-independent moral reasons commonly evoked can be seen to be unable to justify adequate ranges of laws. I demonstrate this problem in regard to what I view as the two strongest contemporary theories of political obligation, based on fair play and natural duties of justice. Accordingly, to preserve the content independence of political obligations, we must do so in a different way. The modified view I suggest combines different scope-limited content-independent reasons to justify an adequate range of laws.
As traditionally understood, political obligations are moral requirements to obey the law because it is the law. These requirements are independent of content. Although scholars debate the precise nature of content-independent (CI) reasons, their basic thrust is easily understood. The law may require that citizens pay progressive income tax or flat tax. If citizens have requirements to obey, these requirements have moral force because they are the law, rather than because of the specific contents of different laws. Moral requirements to obey the law remain constant across different laws.
Understanding political obligations in this way is standard in the literature. For instance, according to Leslie Green: “Political obligation is the doctrine that everyone has a moral reason to obey all the laws of his or her own state and that this reason binds independently of the content of the law” (Green, 1999: 309). Green views this feature as a “necessary one in any argument purporting to establish the existence of a political obligation.” It could not be abandoned “without abandoning part of any satisfactory analysis of political authority” (Green, 1988: 226, 239).
Many other theorists with similar views could be named (e.g. Raz, 1986: 35–37; Schauer, 2015: 52; Shapiro, 2002: 389), including critics of political obligations. For example, according to Robert Paul Wolff, in the work that pioneered philosophical anarchism: “Obedience is not a matter of doing what someone tells you to do. It is a matter of doing what he tells you to do because he tells you to do it” (Wolff, 1970: 9; his emphasis). Wolff (1970: 9) argues that this conception of obedience is incompatible with individual autonomy, and so the philosophical anarchist “will deny that he has a duty to obey the laws of the state simply because they are the laws” (his emphasis; see also Simmons, 2001: 106–107; Smith, 1973: 952).
In this article, I criticize this traditional view of CI political obligations. Central to what I regard as the traditional view is a single, expansive CI reason to obey virtually all laws because they are laws. This view is bound up with what Green calls the “self-image of the state” (Green, 1988: chap. 3). He claims that the state sees itself as a “duty imposer.” By passing a law or other edict, it changes people’s normative status, giving them reasons to obey the edict in question. CI political obligations are the reasons why people are required to obey. Although the state is not able to pass any law it pleases, central to the traditional view is that it has extremely wide latitude. Basically, it is able to pass any law that has rational basis and is not objectionable on moral or constitutional grounds. As long as these minimal requirements are satisfied, people subject to its jurisdiction are required to obey laws that are passed because they are laws. In opposition to this view, I advance two main claims, one critical and one constructive. First, I argue that this conception of CI political obligations cannot be justified. However, intuitively plausible and familiar it may be, adequate foundational reasons for it are not to be had. More than 40 years of critical scholarship has called into question all plausible bases of political obligations. 1 But in this article, I wish to avoid familiar debates about different principles of political obligation. I therefore sidestep these as much as possible to focus on my central concern, what I call “scope-limitations” of moral reasons. Accordingly, my main critical contention is that, because of scope limitations, CI moral reasons commonly evoked to justify political obligations are not able to uphold adequate ranges of laws. Exactly what I mean by this claim is discussed below. I support it in regard to what I view as the two strongest contemporary theories of political obligation, based on fair play and natural duties of justice. However, because of reasons of space and my desire not to rehearse familiar debates about political obligation, only the former is discussed in detail. Natural duties are only touched upon. There are two exceptions to the inability of traditional theories, theories of obligation based on consent and democracy which are not impeded by scope limitations. But these are weak theories, subject to immediate and fatal objections, and so supported by few contemporary scholars. Accordingly, my second claim, if we wish to preserve the content independence of political obligations, we must do so in a different way. The modified view I suggest combines different scope-limited CI reasons to justify an adequate range of laws.
Before proceeding, I should say something briefly about how the argument of this article fits into the current literature on political obligation. As I have noted, in the literature, scholars generally attempt to ground all relevant political obligations in a single moral principle. However, decades of critical scholarship have called these explanations into question, with the result that, at the present time, skepticism about the possibility of a workable theory of obligation may well be the dominant position in the literature (e.g. Buchanan, 2002: 696; Gur, 2013: 326; Morris, 1998: 214). In response to this situation, certain scholars, myself included, have attempted to develop what may be viewed as non-conventional or hybrid approaches, which draw on combinations of different moral principles (Klosko, 2004, 2018; Wellman, 2005: 33–34; Wolff, 2001). While scholars have discussed particular aspects of such approaches, especially how different principles can be combined, important implications of these approaches are relatively unexplored. In this article, I attempt to extend a hybrid model by examining its implications for one central aspect of moral reasoning, content independence. To properly assess a multiple principle approach, its wider implications must be considered. Moreover, the problems with conventional conceptions of content independence discussed below provide additional reasons why I believe it is necessary to move to a hybrid approach.
The argument of this article draws directly upon the critical efforts of philosophical anarchists. To a large extent, I believe these scholars have succeeded in criticizing the traditional theories of obligation, and so that, it is not possible to establish a single moral reason to obey all laws. As a result, the philosophical anarchists argue, to determine what one is morally required to do, one should take into account all relevant moral considerations, including moral principles that bear on one’s circumstances (Simmons, 1979: chap. 8). However, in a crucial respect, the philosophical anarchists have not succeeded. In keeping with hybrid approaches, I believe it is possible to circumvent their criticisms by cobbling together different moral principles. A hybrid theory developed in this way is able to accomplish the practical aim of theories of political obligation: to demonstrate the existence of moral requirements for virtually all citizens to obey virtually all laws. Accordingly, even if we accept the criticisms of the philosophical anarchists in regard to single moral principles, in practical term, these criticisms turn out to be “toothless” (Gans, 1992: 90). As discussed in the previous paragraph, the present article moves beyond largely familiar aspects of the debate with philosophical anarchists, by exploring how a hybrid theory would work.
Content Independence
Content-independent reasons were first clearly analyzed by H. L. A. Hart (1958, 1982). According to his account, central to such reasons is distinctively loose relationship between reasons for action and actions themselves. For example, in what we may regard as an ordinary case, there is “a connection of content” between reason and action (Hart, 1982: 255). Thus, my reason for closing the window is that the room is cold. With CI reasons, this connection is severed. Such reasons allow the substitution of different contents without otherwise affecting reasons for action. Many different kinds of reasons have been identified as CI, including orders, commands, threats, promises, and others as well (Markwick, 2000). For instance, if the sergeant orders a private to stand at attention, the latter is required to do so. But if the sergeant ordered him to stand at ease or to present arms, the private would be required to perform these actions, and he would be required to do so not because of the inherent value or other features of the actions but because he had been ordered to do so. To take another example, imagine that a father tells his young daughter to eat her peas. In a case, such as this, we are able to distinguish the reason why she should eat her peas, which is the father’s command, and the action commanded. If her father told her to go to bed, her reasons to comply would be similar, in spite of the different contents of the two commands. 2
While the exact nature and workings of CI reasons are subjects of scholarly debate (Markwick, 2000; Sciaraffa, 2009), helpful analysis is provided in a recent article, by N. P. Adams (2017). Adams’ main contribution is interpreting such reasons through the metaphor of a container and its contents, with the CI reason identified as the former and the actions to which it refers as the latter. Thus, in the case of the sergeant, his order is the container, while the content of the container changes with different orders. This account allows clear identification of different aspects of content independence. Following Adams, we can distinguish the existence of a CI reason, its force, and the action to which it refers, or its content. Adams argues that the existence and weight of a CI reason remain constant with different contents (Adams, 2017: 148). For instance, in the example of the father, the container is clearly identified as the father’s command. While the contents of the container change from eating peas to going to bed, in both cases, the reason the child should comply is because her father told her to, and the force of the requirement to eat her peas is akin to that to go to bed.
While this account is helpful as far as it goes, I believe we can develop it farther. An additional factor that has not received adequate attention is the foundation on which a CI reason rests, in other words, the basis for the authority of the party that issues it. 3 In many cases, this is clear and requires little comment—which is likely one reason it has not received the attention it merits. Thus, the private has a requirement to obey his sergeant’s orders because of the system of military ranks and discipline in which they participate. It is perhaps less clear exactly why the daughter is bound to obey her father. But we may set complicating factors aside. There are many plausible accounts of the authority of parents over their children; for present purposes, any of these will do. However, in spite of the intuitive clarity of many cases along these lines, we must realize that, in the absence of adequate foundations, a given CI reason will not bind. If a stranger or some other person with whom the daughter has no relationship tells her to eat her peas or to go to bed, she is not required to comply (over and above whatever reasons stem from the independent desirability of these actions). Similarly, if a civilian orders the private to stand to, this order will have no force.
In this article, we are especially concerned with two additional aspects of foundational CI reasons. First, we must recognize that such reasons contain scope limitations. Not any contents may be swapped with their containers, but only those that are successfully grounded by given foundational reasons. Let us say that the foundation for a particular CI reason is legitimate if it is able successfully to ground whatever requirement is in question. In the above examples, both the sergeant and the father have legitimate authority, in that they are able to present binding commands. However, the nature of a particular CI reason imposes limits on the reason-giver’s authority. While the foundation of the sergeant’s authority allows him to require privates to stand at attention, march, face right, and many other such things, it does not include requiring a given private to vote for a particular candidate for public office, vacation in Florida, or root for the New York Yankees. Something similar is true of the father. If we assume ordinary circumstances, his authority does not give the child reasons to jump off a tall building or to cut off her hand. Similarly, when the former child is an adult, his commands to eat her peas or go to bed retain little if anything of their former force.
In addition to having limited scope, CI reasons justify moral requirements in a particular way. Justification they provide may be viewed as on two levels. Return to the father. He tells the daughter she should eat her peas. Why should she obey? On the first justificatory level, she should do what he tells her to do because he told her to. It is not unlikely that, if she protested, his response—his initial response—would be: “because I told you to.” But this of course is not a satisfactory explanation. Why does the fact that he told her bind her? Responding to this request for foundational reasons moves us to a second level, at which we encounter what we may view as genuine justificatory reasons. Whether or not a given justification is satisfactory depends on whether the reasons in question are able legitimately to establish requirement in regard to the relevant action. For the sake of argument, assume that the Ten Commandments are binding. The second-level reason, then, would be because of the commandment to “Honor thy father and thy mother.” But circumstances with other CI reasons, including those that ground political obligations, are more complex.
Political Obligations
Turning to political obligation, we encounter analogous moral reasoning, in regard to both scope limitations and two-level justification. Let us begin with the latter. As noted above, the centerpiece of the traditional view of CI political obligations is a single overall reason to obey the law. Imagine that a citizen, Abel, questions a law requiring him to pay a tax pegged at 30% of his income. On the first justificatory level, the state’s traditional response is that Abel should obey the law in question, because it is the law. This reason is of course CI. If the tax had been set at 20% or 40% the state’s response would be the same, as would Abel’s requirements to obey. However, in itself, this reason is not adequate. Abel may want to know why he is required to obey the law. In response, the state must invoke a second-level moral reason, but because of disagreement in the literature about exactly which moral reason this should be, this level of justification should admit different ones. In this respect, justification on this level is open-ended:
Second-level response: Abel’s requirement to obey the law because it is the law rests on “some moral reason that is able to ground CI political obligations.”
For instance, if Abel had consented to obey the law, this response would likely support his requirement to obey the law. Different principles could be substituted for consent, and, if they survived scrutiny, they too could support obedience.
Turning to scope limitations, we encounter difficulties with moral reasons invoked at the second level. Once again, the traditional view of CI political obligations appeals to a single expansive reason to obey the law. Although theorists disagree about preferable principles, traditional discussion of single moral reasons is a central feature of the literature on political obligation. 4 For example, Beran defends consent; Arneson, Dagger, and I defend fair play; Wellman argues according to Samaritanism; Waldron, Stilz, and Ripstein defend a natural duty of justice; Horton appeals to association; Hare to consequentialism; Walker to gratitude, and many others could be named. 5 Critics of traditional theories of political obligation also adhere to single principles, as they generally criticize the different theories in turn, on the belief that to point out difficulties in all of them is to demonstrate the impossibility of political obligations. 6
It is here that scope limitations kick in. If we assume that the second-level response appeals to a single moral principle, it is not clear it can succeed. I agree with Green that content independence is a central feature of our understanding of political obligations, and so that, it must be preserved (see also Valentini, 2018). But to be blunt, the content independence of political obligations cannot be sustained on the basis of a single moral principle. Even if a principle is able to overcome the objections in the literature, because of scope limitations, it will still not be able to uphold a traditional CI reason to obey the law. 7
Construing CI political obligations as based on a single moral principle confronts two difficulties. First, as noted above, to a large extent, I believe the philosophical anarchists and other likeminded theorists have been successful. Their scholarship has raised severe doubts about theories of obligation based on all familiar principles, and so the prevalence in the literature of skepticism about traditional theories. Accordingly, we must confront the possibility that the content independence of political obligations cannot be justified, because there are no defensible moral reasons able to justify it. Although this is a fundamental difficulty, because I wish to avoid familiar debates, I will largely set it aside in favor of scope limitations, which is the second difficulty.
The problem here is that, even if we were able successfully to defend a particular principle from the philosophical anarchists and other critics, it must be of adequate scope, able to ground adherence to what may call an acceptable range of laws. For our purposes, it is not necessary (and likely not possible) to lay out precise parameters in the abstract. We may assume that an acceptable range encompasses laws in regard to central functions of the state, for example, providing law and order, defense, public works, and additional tasks that are performed by most if not all existing liberal democracies. Other laws are also included and will be discussed below. Whether or not a particular principle is subject to this difficulty is seen by testing it against different possible laws. Inability of a given principle to create obligations to obey particular kinds of laws that are included in the acceptable range implies that it does not ground an adequate theory of obligation. Obviously, the severity of this problem depends on the particular laws affected. Particular cases will be examined below. But to take a simple example, I believe that a principle of political obligation that cannot establish obligations to contribute to national defense is disqualified on this basis.
These two problems beset all current theories of political obligation, but they affect different principles in different ways. Although my main concern in this article is the second problem, the first is necessary in two cases. Once again, I believe that theories of political obligation based on consent and democracy are able to establish CI obligations to obey an acceptable range of laws, and so that to overcome the second problem. 8
Consider consent. According to Locke’s theory, if some body of people consents to government, they agree “to be concluded by the majority”, in regard to the areas of authority subject to the state (Locke, 1988: sec. 96). These obligations are pre-eminently CI, as, within broad parameters, those who have consented will have requirements to accede to the state’s determinations. It is also intuitively clear that the laws in question will constitute an acceptable range. A political theory based on democracy is less familiar, but I assume that it also passes this test. 9 However, both of these are notably weak theories, held by few if any current theorists. Consent’s difficulties are so well known as not to require discussion in this article (Simmons, 1979: chaps. 3–4; cf. Beran, 1987). Problems with democracy are similarly disqualifying. But because these are less familiar, I will discuss them briefly, in the third section.
For reasons of space, to defend my claims in regard to problems of scope, I must be selective, and so I focus on fair play and natural duties of justice, the two theories of obligation that are the most defensible and most strongly supported in the literature at the current time, although reasons of space entail that discussion of natural duties must be somewhat cursory. For the sake of argument, in examining these theories, I set aside questions of overall defensibility and focus only on scope limitations. Because these theories will be seen to be unable to support acceptable ranges of laws, I conclude that no theory of obligation currently available is able to uphold the traditional view of CI political obligations.
If these negative conclusions are correct, where do they leave us? In the fourth section, I argue that a successful theory of obligation must be CI in a somewhat different sense. Rather than advancing a general moral reason to obey all laws because they are laws, it defends particular laws or classes of laws on the basis of reasons peculiar to them. For reasons of space, I cannot present an overall or complete defense of this modified view. I can only point the direction in which content independence can be rescued and respond to some important objections. But if traditional theories of obligation cannot ground an overall reason to obey the law because it is the law, a modified view, such as the one developed here is the only alternative.
Democracy
In this section, I briefly examine a theory of obligation based on democracy—on the assumption that consent need not be discussed. Theorists who argue in favor of a democratic theory focus on the need that laws are made in ways that protect equality. For example, Thomas Christiano (2010: 2) appeals to “public equality, which requires not only that people be treated as equals by “society’s institutions”, but that they be seen to be treated as equals.” Democratic political processes are arguably best suited to meet these standards.
It is clear that democratic provenance is able to generate laws that are CI. If the law says that people should pay one third of their income as tax and democratic provenance is able to generate obligations to comply, then people have this moral requirement. Similar requirements exist as the content of the law changes, as long as democratic provenance is able to establish obligations. Thus, democracy is well suited to conform to the “self-image” of the state, and democratic theorists describe the obligations generated by their theories as CI (Viehoff, 2014: 367–368).
The problems with a democratic theory of obligation are, however, fundamental. Democracy itself is not able to generate obligations to comply with its decisions. If a democratic decision is to apply to a specific set of people, moral principles other than democracy are necessary to bind them (Simmons, 2016: chap. 3) Consider an example. The neighborhood birdwatching association decides that everyone in the neighborhood should contribute US$50 to preservation of the house sparrow. Does this decision bind me? The answer is obviously “no,” unless there is some reason that it applies to me. For instance, I might have joined the club and agreed to comply with its decisions. But in the absence of such a reason, I am obviously not bound. The fact that the association made the decision democratically would not change things in regard to me. Assume that all members of the association had equal rights to participate in the decision and that votes were free and fair, counted accurately, and so on. Still, in the absence of a reason connecting me to the decision, I am not bound by it.
There are various ways in which I could be bound to a given association. Different principles of political obligation could accomplish this. However, in the absence of such a binding reason, democracy itself does not create political obligations; that task would be accomplished by whatever binding reason was in effect. And while democracy itself is not scope-limited, whatever reason generated the requirement to comply would be. As a rule, rather than themselves grounding obligations, democratic decision procedures are used to determine the content of given binding reasons. For instance, if we have consented to pay income tax, the fairest way to determine the form this should assume would be democratic procedures. Like requirements that laws not be grossly immoral or significantly violate people’s rights, democracy guarantees requirements of procedural fairness, that laws be made fairly. But because democracy itself does not establish moral requirements to obey the law and must rely on other moral principles, it is unable to overcome the scope limitations that affect these other principles.
Fair Play and Natural Duties
If we rule out consent and democracy, we must consider other theories of obligation. For practical reasons, discussion of these alternatives must be less than complete. I will discuss fair play and natural duties of justice, although, once again, the latter only briefly.
(Iva) Fair Play
The principle of fair play was first clearly formulated by Hart (1955) . Its moral basis is mutuality of restrictions. Under specified conditions, the sacrifices made by members of a cooperative scheme to produce benefits also benefit non-cooperators, who do not make similar sacrifices. According to the principle, this situation is unfair. It is intended to justify the obligations of non-cooperators, to achieve “just distribution of benefits and burdens” (Lyons, 1965: 164).
For the sake of argument, I will assume that the principle is able to establish obligations to cooperate in the supply of benefits that are not “accepted” in the usual sense, and so that applies to the production of public goods. 10 But in the literature, this condition is satisfied most easily in regard to supply of public goods that are indispensable or necessary for people’s lives (Klosko, 1992: chap. 2). Included in this class are public goods bearing on physical security, most notably national defense, law and order, and protection from environmental hazards and severe threats to public health. Because provision of such goods requires that people’s activities be coordinated by the state, theorists contend that the principle grounds moral requirements to obey the state in regard to the relevant cooperative schemes (Klosko, 1992: chap. 2).
The obligations established by the principle of fair play are undoubtedly CI. Consider defense. In modern countries, people will have different opinions about how defense should be provided. Some people might want to bolster conventional military, for example, with larger armies and larger fleets of bombers and ships. Other people might prefer greater reliance on special forces and high tech weaponry, for example, cruise missiles and modern cyber warfare. Disagreements along these lines can be settled fairly only through democratic processes (Christiano, 2010). Democratic processes function as containers, providing solutions to questions concerning the production of benefits that bind all who receive them. Since we will assume that the cooperative systems in question are coordinated by law, citizens will have CI reasons to obey the law.
However, even if we grant these points, fair play is beset by scope limitations. To begin with, even if the principle is able to establish obligations to contribute to benefits that are indispensable, governments provide countless other benefits that are less valuable. Are people obligated to contribute to these? These include building roads, providing education and cultural goods, such as museums, and recreational facilities, such as national parks, and much more. I assume that some package of such goods falls in the acceptable range, and so that, an adequate theory must be able to support obligations in regard to at least some of them. It could be argued that, even if many benefits provided by government are not themselves essential to acceptable lives, they are necessary to the provision of benefits that are. For instance, while an interstate highway system is not itself essential, it is part of an overall infrastructure that is necessary for provision of defense, law and order, and the like (Klosko, 1992: chap. 4). But even if we grant this, it is clear that this argument will not require obedience to many important laws in the acceptable range. Such laws include education and cultural, as well as economic regulation, which, though generally beneficial to society, is not necessary for provision of indispensable public goods. Moreover, in all developed countries, governments play major roles in seeing to the well-being of their citizens. Modern welfare states provide income support during periods of unemployment and retirement and healthcare to most or all inhabitants. Because the principle of fair play is a “self-benefit” principle—justifying obligations on the basis of benefits that obligees themselves receive (Arneson, 1982)—it is not able to account for moral requirements to support efforts, such as these that benefit other people.
Accordingly, although the obligations generated by fair play are CI, they do not overcome central scope limitations. But still, because of the range of CI obligations fair play is able to establish, its importance should be recognized. Its contribution to a modified view of CI political obligations will be discussed in the following section.
However, before proceeding, we must confront another possibility, an interpretation of fair play that does allow it to ground CI obligations in the traditional sense. On this account, the rule of law is an overall cooperative scheme, the benefits of which depend on cooperative efforts of all or almost all members of society (Dagger, 2018: esp. chaps. 2, 4). Because all members benefit from rule of law, all incur moral requirements to cooperate in the activity that makes rule of law possible. They too must obey the law. Obligations established in this way are obviously CI. Thus, if convincing, this line of argument would support moral requirements to obey an acceptable range of laws, and both traditional content independence and the self-image of the state.
Although this argument is attractive in many ways, I do not believe it succeeds. Construing the entire legal system as a single cooperative scheme requires that we understand the law as a whole as a “seamless web”,” with each component of the overall system closely integrated in a common structure (Finnis, 1984: 120). Although there are respects in which all components of the law are parts of a common system, all are not integrated sufficiently tightly to function as a cooperative system in the relevant sense. The legal system is so large, addressing so many aspects of society, that its breadth approximates that of society as a whole. The key factor telling against construing the legal system as a whole as a single cooperative scheme is great variance in the force of obligations to obey different laws. As discussed in the first section, the force of CI reasons is constant across different contents of the CI container. This is not the case in regard to the overall container of the legal system.
Consider the coordination required to provide national defense. This system encompasses a significant portion of society, coordinated by government. People could abrogate law-governed responsibilities under defense in different ways, for example, by dodging the draft or war-profiteering. Intuitively, it seems that these actions would be subject to condemnation that was both strong and reasonably similar in strength. I believe that this is not true of the legal system as a whole. Clearly, violations of many laws would receive strong condemnation, for example, laws requiring us to pay taxes or to comply with various environmental restrictions. But violation of many other laws would receive little or no condemnation. Consider our reaction to jaywalking, or having 31 a in a room limited by the fire code to 30. 11 The fact that people respond so differently to violations of different laws tells against the idea that law is a seamless web.
In response to these claims, one might contend that the consequences of the violations just noted are much less serious than in regard to failure to pay taxes or generating environmental pollution. But this response will not work, because in many cases in which people would be strongly condemned for violating specific laws, direct consequences of the violations would be undetectable. If the numbers involved are sufficiently large, failure to cooperate by a given individual or by a small number of individuals would ordinarily not affect the goal of the law in question. For example, imagine that Beth fails to pay her income taxes. At the present time, the annual US federal budget is more than 5 trillion dollars, and the annual deficit in the area of 2 trillion dollars. Because these figures are so large, Beth’s failure to pay her 10 or 15,000 dollars would not be noticed. Even if it were noticed, its effect on the deficit would be infinitesimal. Paradoxically, in cases of this sort, society could actually benefit from Beth’s non-compliance, as she could use her extra money to produce tangible benefits. 12 Still, if Beth is known not to have paid her taxes, she would likely be strongly condemned. Thus, I conclude that the law is not a seamless web and, because it is not, political obligations grounded in the existence of the law as a single system should not be viewed as CI in the traditional sense.
(IVb) Natural Duty of Justice
Serious attention to a theory of political obligation based on a natural duty of justice began with John Rawls’s Theory of Justice. According to Rawls, natural duties are moral principles that hold in regard to all people, unlike obligations, which have to be grounded in specific transactions or relationships between people. For instance, if Abel makes a promise to Beth, the obligation binds only Abel and is owed only to Beth. Other people, not involved in the transaction through which the obligation is generated neither owe nor are owed what has been promised. With a natural duty, in contrast, all individuals are bound by the requirement in question, which is also owed to all people. Several of Rawls’s natural duties are familiar, intuitively clear moral principles. These include the duty not to harm or injure others (Rawls, 1999: 98), the duty to show others the respect due to them as moral beings (Rawls, 1999: 297), and mutual aid, “the duty of helping another when he is in need or jeopardy, provided that one can do so without excessive risk or loss to oneself.” (Rawls, 1999: 98).
To properly assess a natural duty theory of political obligation, it is advisable to remove it from the context of Rawls’s theory of justice. While Rawls attempts to justify the natural duties by demonstrating that they would be chosen by the representative individuals in the original position, I do not believe this is necessary. Several of the natural duties are intuitively clear moral principles, which we may assume are binding. This category includes duties, such as those just mentioned. Especially important is mutual aid. That this duty holds is intuitively clear. And so I assume the existence of a requirement to come to the aid of other people who need help, including unfortunate members of society, for example, orphans, the mentally ill, the blind, and others who cannot care for themselves.
Theorists have attempted two main variants of natural duty theories of political obligation, which may be described as negative and positive. 13 I begin with the former. This approach follows the position developed by Immanuel Kant in The Metaphysics of Morals. Imagine how people would fare in a situation without law, a state of nature. Rather than focusing on the threats to people’s survival that dominate other social contract arguments, notably those of Hobbes and Locke, Kant concentrates on the danger of inflicting injustice on other people. In the absence of a code of law to govern his interaction with other people, Abel would be in danger of dealing with them unjustly. For instance, in regard to property, unless the rules concerning what belongs to whom were clearly spelled out, Abel could unwittingly violate the rights of other people. This danger could be alleviated through introduction of clearly formulated laws of property. Accordingly, Kant argues that, to avoid the risk of inflicting injustice on others, people have a duty to leave the state of nature for civil society and its code of laws, to achieve a “rightful state.” 14
For the sake of argument, I will grant these contentions and that the obligations that are established are CI. However, the obligations so established are beset by scope limitations. People are to submit to law to avoid violating the rights of other people. This argument grounds only obligations to laws that concern areas in which such violations are possible. It is not an accident that proponents of this argument frequently focus on the need for a property system, since this is one area in which, in the absence of clear laws, people may well violate other people’s rights. But the same is not true in multitudinous other areas of the law. Consider once again welfare state activities, which I assume are necessary components of an acceptable range. Note once again, that the natural duty in question here is negative, a requirement to avoid inflicting injustice on other people. For a citizen’s non-compliance to be unjust to the needy, her natural duty must be positive—to come to the aid of others—rather than to avoid harming them. We will consider this kind of natural duty directly, but the point here is that principles other than the negative natural duty must be invoked to require compliance to an acceptable range of laws.
Let us turn now to a theory of obligation based on a positive natural duty. This is a requirement to advance other people’s interests, rather than to avoid harming them. The main thrust of this argument is a requirement to aid other people, especially if they are in danger, along the lines of Rawls’s duty of mutual aid. Once again, I believe such a duty is intuitively plausible, and so that, this approach cannot be faulted on those grounds.
A serious problem with this position concerns the strength of this natural duty. Rawls and other proponents of natural duties generally limit the costliness of actions that can be required. Thus, as noted above, Rawls (1999: 98) says that the duty of mutual aid is to help others when they are in need, “provided that one can do so without excessive risk or loss to oneself.” Our general intuitions concerning the natural duties support the claim that they are of limited force. While Abel would be severely condemned if he allowed a child to drown to avoid inconveniencing himself by performing a rescue, it is not clear that he would be similarly condemned if rescuing the child required significant risk, for example, having to rush into a burning building that could collapse at any time. If such costly actions are not required, then we can see that a natural duty of justice cannot ground obligations of acceptable range. Requirements to pay burdensome taxes or to obey other costly laws could not be supported. 15
Alternative View of Content Independence
If we grant that no viable theory of political obligation is able to ground a single reason to obey the law because it is the law and we wish to preserve content independence, we must approach it in a different way. We must consider the possibility of a theory of obligation that establishes CI political obligations, although not in the traditional sense.
The need for this approach is seen clearly, if we return to two-level justification. Beth asks why she should pay her taxes. The state responds on the first level that she should pay because that is the law. But as we have seen, on the second level, the state is unable to defend this claim with a convincing traditional reason. Accordingly, I believe that, to overcome this difficulty, it must invoke a narrower reason that is able to undergird the specific law in question. Although this moral reason will be unable to support requirements to obey all laws, it can succeed with this one. On a larger scale, CI moral requirements to obey all laws can be established by combining different scope-limited, CI reasons to obey different laws. My expectation is that there is a great deal of overlap among these reasons and that, in combination, they will establish obligations to obey an acceptable range of laws. It is likely that this combination of arguments will leave us with a situation that is messy and in that way less than satisfying. In this respect, traditional theories are clearly superior, as they possess a desirable elegance. But as we have seen, this elegance carries prohibitive costs, as it cannot be defended adequately.
On the modified approach, justification is in regard to specific laws, coming into play when requirements to obey those laws are called into question. Because of the messy, particular nature of this approach I cannot provide a full defense here. Nor can I do so on an abstract level. Rather, arguments must be worked out piecemeal fashion, in response to whatever laws are challenged. Consider laws bearing on national defense, arguably government’s most important function. If Abel asks why he should submit to the draft, the first-level response is of course that it is the law. But on the second level, adequate reasons are forthcoming. To uphold this requirement, the state could appeal to the principle of fair play, or some other principle that is able to establish the relevant moral requirement, thereby successfully addressing the immediate challenge. According to this conception, justification is practical. When Abel is confronted by a particular law claiming his obedience, he should determine whether it lies in a particular area of the law obedience to which is supported by defensible moral reasons. For instance, if we assume that the principle of fair play is able to establish obligations to contribute to cooperative schemes that provide indispensable public goods, then people can be seen to be bound to numerous cooperative schemes. These include schemes that supply defense, but also others that provide clean air, protection from threats to public health, and others as well. If reasons are required for why laws in these different areas should be obeyed, fair play can be invoked in each case, while in each of these cases, the moral requirements established by fair play will be both CI and scope-limited. Moral requirements in regard to each specific law will depend on the particular cooperative scheme that provides the benefits that ground the obligation. People will be required to support the clean air scheme because it provides them with clean air and the disease protection scheme because it provides that benefit. Something similar is true of moral requirements stemming from other principles of obligation. In each case, the requirements are limited by the nature of the principle on which they rest.
To get a clearer idea of how this modified view works, we may consider a few examples. Even if we grant that fair play establishes a wide range of obligations important laws fall beyond its reach, and other principles must be appealed to. Laws of familiar social welfare programs are in this category. Because fair play obligations depend on benefits that obligees themselves receive, it cannot support such laws. 16 But, a natural duty of justice can and so it should be appealed to. There is no incompatibility between fair play and a natural duty, even though this approach breaks from the tradition of single principle approaches. Consider food stamps. If Abel asks why he should be required to pay taxes to support such a law, the second-level response is because of his duty to help prevent unnecessary suffering, in accordance with the duty of mutual aid. In cases along these lines, it is necessary to combine moral principles, in keeping with hybrid approaches discussed above. In this particular case, his natural duty should be supplemented with a different version of fair play, according to which all citizens are required to do their fair part in preventing unnecessary suffering (Wellman, 2005: chap. 2). Thus, I believe laws of this kind can be accommodated by the modified approach.
A more difficult case concerns providing benefits to other people that are not intended to alleviate suffering, and so that, do not fall under a duty of mutual aid. Consider laws intended to provide education and cultural benefits to other people. While I do not believe such laws can be justified through simple appeal to a natural duty of justice, I believe they can be supported by other considerations, especially the need for democracy. While an argument in support of this position must be abbreviated here, it runs something like this. Because Beth is an inhabitant of the country in question, she will be required to obey its laws, if they can be justified. But unless the laws are made democratically, citizens will not be treated equally. Requiring some people to obey laws they have had no say in making would fall short of public equality, as Christiano argues. Therefore, laws must be made democratically, with all citizens given means to participate. But if citizens are to have effective rights to participate in democratic processes, these rights cannot be solely formal. For example, if citizens lack basic education or have no time or resources to become informed, politically active citizens, whatever political rights they have cannot be effectively exercised. Thus, the need for equal participation requires that citizens have a certain level of material resources. They require what we may call an adequate level of income—adequate to allow them leisure to become informed about political affairs—along with sufficient education and cultural opportunities to make it possible for them to do this and to participate actively. 17
An implication of this modified approach is that it will likely be more difficult to justify obedience to the full range of laws than it is under a traditional theory—on the assumption that the traditional theory is in fact able to justify all obligations. Because of the particular nature of foundational reasons, it is possible that various laws will fall through the cracks. In order for the state to support requirements to obey different laws, on the second level of justification, it must provide convincing reasons why whatever law is in question should be obeyed. These reasons can of course be based on different moral considerations. However, if adequate reasons cannot be presented to defend a particular law, citizens will not have moral requirements to obey it. Once again, in this respect the modified view falls short of the ideal situation in which, on the basis of a single foundational reason, the state is able to justify obedience to all laws with reasonable basis. But, there are a great number of moral principles that can be invoked to support obedience to particular laws. On the whole, these are not incompatible, and it is highly likely that, by combining different principles, we will be able to justify obedience to an acceptable range of laws—or possibly all reasonable laws.
Because of the messy, particular nature of the modified view, it is subject to criticism. An important objection is that, even if it is able to deal effectively with scope limitations, it is too complicated to function as an adequate theory of political obligation. Adams presents a series of criticisms along these lines. One objection concerns the demands placed on citizens. In the absence of an overall reason to obey the law, in his words, citizens “must constantly reevaluate whether they have reason to do as the law says, resulting in significant inefficiencies”, a situation that is exacerbated by the pluralism of modern society and disagreements among its members. In addition, to quote Adams once again, the moral basis of people’s obligations “will often be obscure to ordinary citizens, who do not have the time or expertise to evaluate the grounds of the huge volume of complex laws in modern society” (Adams, 2017: 161). On the picture Adams presents, a given citizen must run through the moral reasons for and against requirements to comply every time she is confronted by a law. In a now famous, hypothetical case, Beth confronts a red light on an empty road in the middle of the desert, in the middle of the night. According to Adams, on a view, such as the modified one defended here, Beth must calculate the myriad factors involved before she can know whether or not she is required to stop. This daunting complexity is in comparison to the traditional view, according to which, by simply referring to her overall requirement to obey all laws because they are laws, Beth will understand her moral situation and not be overly taxed.
I agree that these are serious difficulties, but there are responses. First, we should be clear about the problems with Adams’ own view. Although a traditional reason to obey all laws because they are laws is of course convenient, it cannot be defended. To use it as a basis to criticize the modified view is not legitimate. Furthermore, we should be clear on the distinction between questions concerning people’s abilities to assess their moral requirements and questions as to whether they in fact have the requirements. Even if Beth is unable fully to understand the moral reasons bearing on her case, she still may have moral requirements to obey given laws. Contrariwise, even if she believes that she is required to obey given laws, this may in fact not be true. Whether or not she has moral requirements to obey particular laws depends on the moral facts, rather than on her ability to assess those facts. Once again, even though the modified theory is messy and complex, I believe it is still able to ground an acceptable range of political obligations, whatever difficulties people have in understanding this.
To a certain extent, people’s difficulties in understanding their moral situations can be circumvented by appeal to the generally reliable cognitive shortcut of assuming that they do have obligations. Once again, I believe this is true in most cases, and so that, this shortcut is just that, a shortcut. It is not a fiction. In this light, it may be useful to consider a related point in regard to the state’s point of view. In an important article, Noam Gur argues that it is in the state’s interest to tell people that they have overall obligations to obey all laws, whether or not they actually have them. Because it is necessary that people be law-abiding citizens, he believes it is important that the state cultivate law-abiding dispositions in its citizens (Gur, 2013: 342). One consideration supporting this position is cognitive errors to which people are liable. For example, “self-enhancement bias” leads people to exaggerate their own capabilities and to underestimate their deficiencies, such as incompetence and poor judgment. Similarly, people are prone to “hyperbolic discounting,” a tendency to overvalue immediate benefits in comparison to others that are more remote. These tendencies and numerous others have been confirmed by empirical research. Accordingly, as Gur has it, even if people are well-meaning and wish to promote society’s good, if they act on their own conceptions of this good, they will frequently err, to society’s detriment. Gur (2013) argues that belief that we should obey the law because it is the law is therefore useful and would promote desirable effects for society and its inhabitants, even if obligations to obey certain laws cannot be justified.
While I support a conclusion similar to Gur’s, I do so on different grounds. As we have seen, he argues that the state should promote belief in a general requirement to obey the law even though it may not be able to support this. As with Gur, I believe the state is well advised to promote this belief to help people avoid complex moral reasoning. However, unlike Gur, I believe that a requirement along these lines can in fact be justified. Because the modified view acknowledges scope limitations and works around them, it is able to establish moral requirements to obey all or almost all laws. 18 But because of factors, such as those Gur invokes, it is frequently advantageous for people simply to believe that they are required to obey whatever laws are in question.
Accordingly, in spite of its unorthodox features, the modified view is able to salvage a position more or less extensionally equivalent to that of the traditional view. But unlike the traditional view, it has the considerable advantage of being defensible. However, appealing it may be to believe that we should obey the law simply because it is the law, this position cannot be defended. Messy and complex as the modified view is, it is able to uphold these requirements in a way that is defensible.
In conclusion, I should note once again, that the modified view of political obligation proposed in this article is less desirable in many ways than traditional views. In particular, it lacks traditional views’ elegance and simplicity. But as we have seen, elegance and simplicity come at a cost. If it is not the case that the state is able to justify its self-image by providing an adequate foundational reason for obeying all laws, then, if we wish to preserve the content independence of political obligations, the modified view presents an attractive alternative. 19
Footnotes
Acknowledgements
The author wishes to thank Colin Bird, Matthias Brinkmann, Harrison Frye, Ross Mittiga, Justin Tosi, Jiafeng Zhu, and the editors and reviewers for Political Studies for valuable comments on previous versions of this paper.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
