In this chapter and the next, I present a novel account of what the rule of law demands and why we should care. The account brings together pretheoretical evaluations of rule of law institutions in real states, functional generalizations of those institutions, and an account of their moral worth.
Together, Chapters 1 and 2 defend two key theses. First, the rule of law is morally valuable because it is required for the state to treat subjects of law as equals (“the equality thesis”). Specifically, the rule of law fosters vertical equality between officials and nonofficials and horizontal equality among nonofficials.
Second, states comply with the rule of law to the extent that they satisfy the following three conditions (“the three principles”):
Regularity: Officials are reliably constrained to use the state’s coercive power only when authorized by good faith and reasonable interpretations of preexisting, reasonably specific, legal rules.
Publicity: The rules on which officials rely to authorize coercion are available for subjects of law to learn; officials give an explanation, on reasonable demand, of their application of the rules to authorize coercion in individual cases; and officials offer those who are the objects of state coercion the opportunity to make arguments about the application of legal rules to their circumstances; the public at large may observe these reasons and the arguments about them.
Generality: Neither the rules under which officials exercise coercion nor officials’ use of discretion under those rules make irrelevant distinctions between subjects of law; a distinction is irrelevant if it is not justifiable by public reasons to all concerned.
Each condition presupposes the satisfaction of those before it. (It is possible, however, to combine the partial satisfaction of a later principle with only a partial satisfaction of an earlier principle – for example, to have a state that is partly general, but excludes some discrete class of individuals from the protection of the laws, and hence is also only partly regular and public.) Regularity and publicity together lead to vertical equality. A state that has achieved them has achieved a weak version of the rule of law: its officials cannot easily abuse their power, and individuals can be fairly secure in their legal rights against the state. Generality leads to horizontal equality. A state that has achieved it has achieved a strong version of the rule of law: in it, individuals are genuinely equal under the law.
When states achieve vertical equality, their legal institutions guard against hubris, officials’ use of their powers to claim superior status. They also guard against terror, the use of the state’s power to cow individuals into submissiveness.
When states achieve horizontal equality, their legal institutions prevent legal caste, the state’s support of hierarchies among individuals, particularly along ascriptive group lines. They serve the obligation of reciprocity that individuals have to one another to share alike the cost to produce the public good of law and order. Finally, they ensure that everyone is counted – the interests of no one in the community are treated by the law with complete disregard.
The heart of this conception of the rule of law is responsiveness to reasons. The weak version of the rule of law treats people with respect, as minimally capable of responding to reasons given them by preexisting rules that govern their behavior, while also restricting those officials who wield coercive power to acting in accordance with those reasons, rather than simply their own wills. Moreover, they directly recruit the capacity of ordinary people to reason about reasons by making the use of state force against them conditional on their having an opportunity to publicly contest and deploy the reasons given by law.
The reasons at play in the weak version are artificial (or artifactual) in that they are creations of an act of rule making: by saying that they are “reasons,” I simply mean that they take the form of reasons: externally specified normative standards for behavior that can form the basis for better or worse rational arguments, and that can be deployed effectively to criticize people for failing to comply with them. The form of a reason is the opposite of arbitrariness, which is disconnection from reasons: the arbitrary official, when asked why she or he coerced some citizen, can just say “Because,” whereas the official who is bound to legal reasons at least has to be able to say something that is comprehensible to the person over whom power is exercised. In doing so, that official treats the other like an adult, and an equal.
The reasons at play in the strong version are real reasons, in that someone whom the state orders about on the authority of general laws will able to understand the laws as actually meaningful to her (at least in a constructive sense if not a subjective one). The law will recognize her as a stakeholder in the society, and recognize that she ought not to be ordered about at gunpoint unless the orders can be understood to capture something that she has reason to do independent of the unadulterated will of some lawmaker.
This chapter explicates the weak version of the rule of law. Chapter 2 will cover the strong version.
I Opening technicalities
Mindful of the fact that the audience for this book includes not only philosophers and lawyers, but also political scientists, economists, and development experts, most of the technical philosophical groundwork has been deferred to the end of this chapter; some has been published elsewhere.1 In particular, many clarifications, fine points, and answers to objections appear in an article entitled “The Rule of Law and Equality”; the reader who thinks I have missed something obvious is advised to double-check there.2 However, three points are important enough to highlight at the beginning of the main argument.
First, the rule of law has institutional (or what I have previously called “descriptive”) and evaluative components. To see what I mean, compare an idea like democracy to an idea like justice. Justice covers a multitude of normative principles and concrete social practices; there are innumerable uses of the term that bear at most a family resemblance to one another. One’s favored conception of justice might be instantiated in anything from a tax-and-transfer system of redistribution to a society ruled by Platonic guardians. By contrast, while there are numerous theories of democracy, they all occupy pretty much the same territory: all have something to do with will or opinion aggregation, hearing minority views, removing disobedient officials, and so forth. The rule of law is more like democracy than justice: it has a relatively concrete set of practical extensions. This has important methodological implications: it suggests a coherentist account of the rule of law that draws together normative ideas about the value it serves and observations about the social practices that we ordinarily associate with it.3 It also suggests that the best account of the rule of law will have the property that I have elsewhere called “verisimilitude” – it will describe actual societies in the real world better than alternative, equally coherent, conceptions.4
Second, the rule of law regulates states when they exercise their power over individuals. It does not regulate the private use of coercion or violence.5 Particularly, it does not give us a reason to object to anarchy, nor does it oblige nonofficials to obey the law. However, as will be seen in Chapters 6 and 8, the rule of law does require nonofficials to be sufficiently committed to its preservation to be willing to collectively defend it against officials who might abuse their power. That is, in the first instance, a practical rather than a conceptual requirement, although, in view of the goal of this book to state a conception of the rule of law that bridges philosophical and social scientific approaches to the topic, it will soon become clear that such strong practical constraints feature in the concept of the rule of law as well.
While there are many regulative principles for both private and state violence, the unique significance of state violence generates a unique principle, the rule of law, to guard against its abuse. States are distinguished from all other entities by their expressive and practical significance. By “expressive significance,” I refer to the facts identified by Weber and legal philosopher Joseph Raz: states ordinarily claim a monopoly over the legitimate use of violence in their territories, and ordinarily claim that those in the territory are obliged to obey their commands.6 With that, they typically forbid individuals from resisting the force that they wield; often they also claim to be acting in the name of the citizenry as a whole or some constitutive social, national, or ethnic group.7
By “practical significance,” I refer to the fact identified by Thomas Hobbes: states have overwhelming force within their territories. In ordinary political life, in a modern state, it’s impossible or staggeringly costly for individuals to resist its power; moreover, there’s ordinarily no external power to which individuals can turn to protect them from it.
These features (which I will henceforth call the “Weberian and Hobbesian properties”) make state power different in kind from the private use of force. Its expressive significance confers upon it more morally relevant dimensions than private power: a mugger who assaults you on the street doesn’t claim you’re obliged to quietly submit to the violence, or that it’s done in your name. The state does. Its practical significance makes it more fearsome and influential: you might have a chance to fight back against the mugger, or at least call upon the state to defend you. There’s no one to defend you from the state.
However, in some societies, there can be forms of nonstate power that sufficiently resemble state power, particularly in its practical significance, so that their control too becomes a matter for the rule of law. For example, in Classical Athens, as will be seen in Chapters 5 and 6, the rich and powerful had an interest in undermining the state, and were capable of overwhelming ordinary citizens on a one-to-one basis. There, the rule of law required their nonstate power be regulated, too, just because it threatened to take on the Hobbesian and Weberian properties. At the end of Chapter 3, I explain more precisely where the rule of law requires private power be regulated.
Finally, I claim that the egalitarian account of the rule of law and its moral value is factually and normatively robust. By “factually robust,” I mean that the arguments offered do not depend on strong assumptions about facts about social arrangements, human motivations, or the like that differ from society to society. This is a weaker criterion than necessary truth: the arguments might not be true in every possible social world, but they are true for a substantial range of reasonably common social worlds. By “normatively robust,” I mean that these arguments are, as far as possible, nonsectarian. They are meant to be acceptable without taking on overly controversial normative commitments. This argument for the moral value of the rule of law relies on the ecumenical core of the ideal of equality: if nothing else, those who value equality object to hierarchies of status and esteem, and demand that the state treat individuals with equal respect and take each of their interests into account.
The robustness criteria respond to concerns with the scientific and the political usefulness of a conception of the rule of law. First, a conception of the rule of law ought to be compatible with a social scientific explanation of its appearance in societies characterized by different political institutions and ethical ideals. Thus, we ought not to say that the rule of law responds to institutions and motivations that have appeared in few rule of law societies, or to values their citizens have rejected. In Chapters 5 and 7, I show that a variety of societies have accepted egalitarian values as the basis of their rule of law, or at least have been influenced by them; in Chapter 9, I put the robustness of this conception of the rule of law to immediate use, by developing a proof of concept scale to aid social scientists in measuring the rule of law.
Second, an account of the rule of law ought to be able to motivate citizens and officials to act in accordance with it in the real world. Explaining the moral value of the rule of law is a political task as well as a justificatory one. Not everyone supports the rule of law, and political leaders may think they have good reason to ignore its constraints in the pursuit of their vision of the public good. Those who think that the rule of law generates normative obligations should be able to offer their arguments to those leaders. For those arguments to be persuasive, they should not appeal to sectarian values that may not be shared by those addressed, and they should be applicable across the broad range of human societies in which we may wish to offer them.8
Normative robustness is important for a second reason. As Chapter 6 will show, the rule of law depends (practically) on a commitment to its preservation on the part of those whom the law protects. A conception of the rule of law that is normatively robust is more likely to be able to sustain that commitment – those who would encourage their fellows to defend it will have moral as well as pragmatic arguments to offer for it. In Chapter 8, I will argue that the egalitarian value of the rule of law helps us understand how rule of law states actually hold on to it; this explanatory insight depends on the normative robustness of the egalitarian account. In Chapter 9, I explain how the egalitarian values underlying the rule of law can actually be used by development specialists and others to promote it across the world.
Finally, throughout this book, I speak of “subjects of law,” or, more informally, “individuals” or “citizens” as those whose equal standing the rule of law protects. Sometimes, I say “nonofficials” to highlight the distinction between ordinary individuals and those who wield official power. By all of these terms, I mean those whom the state claims the authority to command. This category is not limited to those who count as members of the political community (“citizens” in the conventional sense), and includes, for example, aliens stopped at the border, transients incidentally in the territory, and those whom the state has disenfranchised or enslaved. I cannot here reach any conclusions about the scope of the protections of the rule of law as applied to aliens found outside the territory (e.g., in military conflicts). Also, by the use of the term “individuals,” I do not mean to deny that groups, corporate entities, and the like can claim the protection of the rule of law; this question, too, is beyond the scope of the present work.
Having gotten the technical hurdles out of this way, I now turn to defend the principles of regularity and publicity, and the conception of equality that they serve. They respond to our experience, throughout history, of societies in which the rule of law catastrophically fails – societies such as Haiti under the Duvaliers, or the Soviet Union, in which individuals live in constant fear of officials and officials behave arrogantly toward nonofficials. First, I will elaborate publicity and regularity, their scope, and how they are derived from the specific practices of rule of law states. Then I will show that publicity and regularity are necessary to free states from hubris and terror, and sufficient to at least greatly circumscribe them.
II The weak version of the rule of law in two principles
We begin with the core concept: the rule of law imposes the twin demands on the state to control the use of its monopoly over violence with rules and to make those rules accessible to those over whom that monopoly is used.
A Regularity
If the rule of law means anything, it must mean that those who control the power of the state may not use it whenever and however they want, bound only by their untrammeled whims – their power must be bound by law in some meaningful sense. I express this fundamental idea in the principle of regularity, the minimum condition for a state to have even a rudimentary version of the rule of law. A state satisfies it if officials are reliably constrained to use the state’s coercive power only when authorized by good faith and reasonable interpretations of preexisting, reasonably specific rules.9
Regularity defines the line between states that control official violence and those that are run at the will of executive officials, or in which soldiers and police use violence willy-nilly against individuals who have something they want or who anger them. Because this is the most fundamental function of the rule of law, most conceptions contain something like it.
Another way to express the ambition of the principle of regularity is that it requires the state’s coercive power be exercised impersonally. Regularity is violated when officials are permitted to treat the power with which they’re entrusted as part of their personal endowments, suitable for use in their private relations with members of the community; it’s respected when they are constrained to treat their power as the instrument of an agency relationship between themselves and the state, usable only for the purposes and under the conditions given by the terms of their legal authority.10 This is an ideal of role separation.
The heart of the idea of role separation is captured in a 1916 short story by Munshi Premchand, “Panch Parmeshwar.”11 The two main characters, Jumman Sheikh and Alagu Chaudhuri, begin as close friends, but Jumman mistreats his aunt, and when she convenes the local council (panchayat) to resolve the ensuing dispute, Alagu is nominated as the chair. Thanks to their friendship, Jumman expects an easy victory. However, moved by his official responsibilities, Alagu chooses justice over friendship, and rules in favor of the aunt.
After the judgment, Jumman, feeling betrayed, becomes Alagu’s enemy, and casts about looking for revenge. An opportunity comes a few months later: Alagu convenes the council to get payment on a disputed debt, and this time Jumman is named chair. Alagu is dejected, sure that he has lost his money. Here’s what happens next:
As soon as he accepted the headship of the panchayat, a new awareness of his duties dawned on him. He said to himself: “I am now the arbiter of justice and dharma. What I say will be accepted as a divine fiat, and I cannot allow my private prejudices to influence the sanctity of the divine word. I must not depart by a hair’s-breadth from the truth.”
After hearing the case, Jumman surprises everyone by ruling in Alagu’s favor. Explaining himself later, he says: “Today I have discovered a great truth – I have seen that when you become one of the panchayat, you are no one’s friend and no one’s enemy. You are only there to dispense justice. Today I have realized that God speaks through the panchayat.” Moved by the realization, the men again become friends.
This story captures the central idea of role separation through its most familiar exemplar, the impartial judge. Upon assuming the judicial role, each of the protagonists not only surrenders his personal attachments of friendship and enmity, but also surrenders his identity altogether, becoming the voice of God. This, in ideal form, is the principle of regularity. When an official (not just a judge) puts on her public role, she becomes the voice of the law. While the rules may leave her with some discretion, that discretion isn’t exercised as her; it’s exercised as the voice of the law, for the law’s ends.
I will argue that regularity and publicity together protect individuals from being subjected to official terror – from the specter of officials with open-ended threats who can use their power to make individuals live in fear and behave submissively. (By “open-ended threats” or “open threats,” I just mean a capacity to do harm to a citizen that an official can use substantially at a whim.) Just as Alagu feared the panchayat under his enemy, subjects will fear the power wielded by officials in irregular states. Unless official coercion is rule-bound, officials will be able to use their power to avenge themselves against their enemies, to expropriate property, and to extort deferential treatment from the population at large – to behave like the Tonton Macoutes or the KGB.
Numerous standard practices of rule of law societies serve the principle of regularity.12 Rule of law societies tend to forbid vague laws. They tend to use tools like the independent judiciary, appellate review, and the jury trial to impose checks on officials’ actually conforming their behavior to rules. They tend to require that the law be prospective and forbid bills of attainder. Each of these practices helps to protect individuals from living in fear of open-ended threats from unconstrained officials. For example, regularity forbids vague laws because officials can manipulate them to punish individuals whenever they want. Similarly, regularity forbids retroactive laws and bills of attainder because they can be enacted to retaliate against individuals who cross officials.13
The reader may understandably hesitate at the fairly vague requirement that the rules be “reasonably specific.” Unfortunately, this is a feature of the normative terrain. More specific rules leave officials less discretion in applying them, but there is no perfect specificity: all rules must leave officials some discretion, because no text can perfectly specify all situations to which it will apply.14 For want of a plausible formal way to specify how much that discretion must be constrained, I turn to context-dependent and pragmatic judgments to pick out the rules that are too open-ended.
We can give some content to reasonable specificity by appealing to the goals of the principle of regularity. A given power may pose more or less of a risk of generating open-ended threats; we can often determine how serious this risk is, and thus how much control is required for a particular power, with intuition and common sense. For example, the scrutiny of independent judges over the power of eminent domain in the United States is arguably sufficient to render it consistent with the rule of law despite its only being constrained by vague standards like public purposes and just compensation. By contrast, a state whose police arrested individuals under the similarly vague standard “whenever it is just” would confer open-ended threats on those police to an unacceptable degree. The power to arrest is much easier to abuse: it’s easy for an individual officer to deploy, and causes a serious short-term harm to the one arrested. By contrast, eminent domain is typically carried out by cumbersome elected or administrative bodies, and requires a further lengthy bureaucratic process before anyone is actually removed from the condemned property. The greater immediate harm the arrest power can cause gives officials who wield it more potential for open-ended threats, and thus gives us good reason to keep it on a tighter leash.
Further difficulty arises from the fact that all rules are open to different interpretations. It would be too demanding to insist that officials only ever use coercive power pursuant to accurate interpretations of the rules, for officials can make reasonable mistakes in applying them, and we do not ordinarily think that a state in which officials make mistakes offends the rule of law on those grounds. At the same time, it is too undemanding to adopt a fully subjective standard, which would permit unreasonable interpretations of law. Separately, it seems too demanding to say that officials be only motivated by the rules: certain kinds of reasons can fairly guide officials’ choices within the domain permitted by preexisting rules; a police officer, for example, may decide that drunk driving is a particularly dangerous crime and spend more of his efforts catching drunk drivers and less catching speeders without offending the rule of law.
I propose to resolve these difficulties by saying that officials must follow the rules in good faith. By this, I mean that they must act as if they take the rules as generating reasons to act in compliance with them and forbidding their violation (the rule of law does not propose to examine the psychological motivations of officials, just their behavior). This forbids uses of power that officials ought to know that the rules do not permit (e.g., it would be bad faith for the officer of the previous paragraph to decide that driving in the rain is too dangerous and to arrest people for that) while permitting officials to use reasons fairly implied by the law (like drunk driving being more dangerous than speeding) to guide their application of the rules within the domain of discretion the law gives them.15 It also forbids unreasonable interpretations of the rules yet accommodates (the inevitable) disagreement.
In practice, this is perhaps a worryingly ambiguous criterion. However, the principle of publicity, to be addressed next, will help draw some boundaries around the idea of good faith. It will be seen in the course of discussing that principle that officials must be able to explain how their uses of power are permitted by the rules, and those explanations must be able to survive exposure to counterarguments offered by those over whom power is to be exercised. In a context in which officials listen to those counterarguments and take them seriously, as required by the principle of publicity, the process of external scrutiny will set an upper bound on the extent to which they can sustain exercises of power that are premised on unreasonable interpretations of the rules.
One might have the opposite worry, that regularity is too rigid. Some scholars, most notably Dworkin, deny that law is primarily a matter of specific rules.16 Instead, according to Dworkin, much of our legal practice involves the application of “principles” – normative standards that are to be weighed against one another in reaching decisions, and that require the extensive use of case-by-case judgment.
Regular legal systems may contain principles. What matters for regularity is that officials be constrained, not how they are constrained. Officials might be constrained by strict de jure rules, where their failure to do so subjects them to legal, social, or political sanction, or they might be constrained by looser de jure rules – open-ended principles leaving them a substantial amount of discretion – where that discretion is itself constrained by unwritten standards that fill out the content of the rules, by social norms that sanction officials for abusing their discretion, by political competition, by checks and balances from other officials, or by something else. Where the written rules constrain less, other tools must take up the slack to constrain more.17
B Publicity
The principle of publicity requires that the rules under which officials use power be accessible to nonofficials. It presupposes that there are such (effective) rules – that is, that regularity is satisfied to a significant extent. Specifically, publicity requires that (a) the laws that authorize official coercion be available for nonofficials to learn (i.e., not secret or unreasonably obscure); (b) officials publicly explain, on demand, their application of the law to authorize coercion in an individual case, where that law itself must meet the principle of regularity; and (c) officials offer those whom they coerce some opportunity to participate in the application of legal rules to their circumstances (i.e., by having an opportunity to make arguments for a particular interpretation of those rules). (The third requirement depends on the second: if officials do not say how the rules authorize their behavior, individuals will find it much more difficult to dispute official decisions.)
The principle of publicity is essentially a reason-giving requirement. Officials must be prepared to give the reasons for their uses of coercion over individual citizens, and those reasons must be statements of how the law, correctly interpreted and applied, permits their actions. Ordinarily this giving of reasons must happen in public: the community at large must be able to observe that officials are following the law and come to independent judgments about the extent of official faithfulness.
A state can run afoul of publicity, but not regularity, if officials’ power is actually constrained by preexisting rules, but nonofficials have no access to those rules or influence over what befalls them under their auspices. In such societies, the law is the exclusive domain of an elite class of officials, and nonofficials must rely on those elites to protect them.18
If regularity is the official-centered side of the rule of law, publicity is the subject-centered side. It responds to the concern not only that officials’ use of the state’s coercive power actually be constrained, but that subjects of law be able to know and to subjectively rely on the constraints. Thus, many of the same practices that serve regularity also serve publicity by involving individuals in the mechanisms to control official power. However, some standard practices of rule of law states serve publicity in particular: these include the prohibition against secret law, the requirement that subjects of law have notice and an opportunity to be heard before being coerced, the right to be represented by counsel, the right to be confronted by the evidence against oneself, and similar practices that allow subjects to observe that officials are constrained by rules and participate in the application of those constraints.19
Publicity allows nonofficials to verify for themselves that the state satisfies regularity. From a nonofficial’s perspective, a state that satisfies regularity but not publicity might not look very different from a state that satisfies neither.
Because publicity allows subjects to figure out whether officials are obeying the law, it also allows them to participate in its enforcement. Institutions (whether courts or otherwise) that force officials to give reasons for their uses of coercion in public permit the population in general to evaluate those reasons. Those that give subjects a forum to claim that officials have ignored the law also give the public a tool to come to consensus evaluations of officials’ actions, and thereby to collectively hold them to the law. This reveals how publicity and regularity come together in the weak version of the rule of law: although it might be possible in principle for officials to be constrained to obey the law absent coordinated enforcement by those who benefit (e.g., by some kind of game-theoretic equilibrium among officials with diverse interests who are motivated to hold one another to the law in order to maintain a political compromise), this is practically unlikely to be stable. In most realistic states, we should expect that dispersed power to observe and sanction official rule violations, whether held by relatively elite or relatively nonelite subjects, will be instrumental in the enterprise of controlling official power. (There will be much more on coordinated enforcement in subsequent chapters.)
If we look more closely at the matter, publicity and regularity begin to appear to dissolve into one another. For we may recast the notions of discretion and specificity, so central to the concept of regularity, into epistemic ideas.20 A minimally specific law, one that grants maximal discretion to the official implementing it, is one in which the meaning (in a nontechnical sense of “practical implications”) of its words is known only to that official. For example, if King Rex writes a law in a private language (Rexish), he has maximum discretion in applying it, simply because nobody else can tell him he is wrong. Put differently, the law may have a specific meaning in Rexish, but Rex cannot be constrained to apply the law only consistent with that meaning, for there is nobody with the capacity to constrain him – all the independent judges and well-armed nobles and engaged populaces in the world might have the power to force Rex to obey a law they can understand, but if they cannot observe when Rex has broken the law, then he is totally unconstrained.
Scaling that idea up, a law is more specific, in the sense that it grants less discretion to an individual official, if it is written in a language that only officials know (Officialish). If the police officer on the street badly enforces a law written in Officialish, she is constrained by other officials, but not the public. Practically speaking, this means that, if all the other officials do not care about the law being obeyed, it will not constrain the individual police officer. Equivalently, we may say that the law written in Officialish has the power to constrain an individual police officer relative to all officials, but does not have the power to constrain the class of all officials relative to everyone else.21
Now consider a law written in a language everyone speaks. Individual words in that language may be more or less penetrable to the public, but are necessarily penetrable to those with final decision-making authority about them. For example, the law may say “No one may drive at an unreasonable speed.” This word “unreasonable” may be quite unclear to ordinary people, but it crystalizes into clarity in particular cases at the moment a judge applies the rule to decide whether someone is guilty or not guilty of speeding, because the judge has the legal authority to finalize the case-by-case meaning of the word. Should she do so without giving any explanation, she is unconstrained (except by any appellate court), and the police are constrained only by the judge, for the judge is the only one with certain knowledge about what the word (legally speaking) means. It begins to appear like an instance of Judgish.
But the procedural demands of the principle of publicity expand that constraint. By allowing the one on trial to make arguments about the word “unreasonable,” the judge’s interpretation of it is opened to the influence from the linguistic and legal understandings of ordinary subjects. Moreover, the process of giving reasons to explain her ruling allows the judge to expand her epistemic power of making the ultimate judgment to the public at large: by making it possible for the public more generally to argue with judges as well as with police about the word, and to use political power to constrain them, she translates “unreasonable” from Judgish to Publicish.22
What this line of argument has suggested is that regularity and publicity refer to the same broad idea. Without publicity, regularity is also lacking, because the notion of being constrained by rules depends on the broad accessibility of both the meaning of the rules doing the constraining and the practical tools for constraint. Separating them is useful for heuristic purposes, to track the historical differences between societies that have not bound the powerful with rules at all and those that have ostensibly bound them with secret or obscure rules. For that reason, I shall continue to do so, but we must keep in mind that the distinction is artificial. There is a difference between the chaos of Caligula and the secrecy of the bureaucratized standards for getting on the Transportation Security Administration no-fly list, but we appeal to the same underlying idea in describing both as offenses to the rule of law.
The requirement that officials explain themselves to those whom they coerce has received less attention in the academic literature than the other elements of publicity, but is quite important in the legal culture of rule of law states. This is particularly visible in the requirements we impose on judges, whom we expect not only to have but to utter reasons for their decisions – a judge who fails to offer written opinions on serious controversies, or who issues significant rulings from the bench without any explanation, has seriously violated legal norms.23 Such a judge will be seen as high-handed, dismissive of the interests of the parties and of the fact that it might matter to them that they understand what is being done to them and have the opportunity to respond to the reasons given them with their own reasons. As I will argue shortly, such a judge is indulging in an act of antiegalitarian hubris: by declining to explain herself, she is expressing the idea that she doesn’t have to explain herself – that she is of sufficiently higher status than those appearing before her that she can give imperious commands and those coming before her should just shut up and do what they’re told.24
I now proceed to the details of that very point.
III Vertical equality
There are two major vices of a state in which publicity and regularity fail: first, officials treat nonofficials with hubris: they behave as if they are a superior class in a status hierarchy. Second, officials inflict terror on nonofficials: they force nonofficials to fear their power and make it rational for individuals to behave submissively in the face of it.
A Respect and hubris
When officials use coercive power without offering reasons, drawn from preexisting law (legal reasons), to the objects of that power, they deliver the message that the one using the power is superior to the one over whom the power is used, and actually constitute a relationship of subordination between themselves and subjects.25 I call this idea “hubris” to acknowledge its derivation from the classical Athenian hubris law, which forbade the striking of fellow citizens (and even slaves) because such striking expressed a disrespectful attitude of superiority toward its victims – it was a figurative as well as a literal slap in the face (see Chapter 5). By contrast, when a state complies with regularity and publicity, officials express respect toward nonofficials and the political community as a whole, and actually constitute a relationship of equality between them.26
By offering reasons for their use of power at all, officials express three distinct forms of respect toward those over whom power is to be used. First, they express the recognition that they actually have to have reasons, and hence that subjects of law are immune from the casual use of official power.27 Superiors do not need reasons to use their power over inferiors: masters need not have any particular reason to beat their slaves; bosses need not have any particular reason to fire their employees. Second, they express the idea that they are accountable to the particular individuals over whom power is used. Equals are accountable to one another; superiors are not accountable to inferiors: even if the master or boss has some coherent reason for his behavior, he need not explain it to his slave or employee. Third, they express respect for individuals’ powers of reasoning – they express that nonofficials are capable of understanding why they are to be coerced, and that it matters that they be given the opportunity to so understand. To be given reasons is to be treated like an adult.
So far, this doesn’t require very much respect. When the defendant asks why she’s going to jail, the judge might just say “Because I don’t like you.” That’s a reason, to be sure, and it’s perhaps a little better than “Shut up or I’ll double your sentence,” but it still clearly expresses an attitude of superiority.
It is slightly more respectful to offer a reason drawn from something other than the official’s personal will. Rather than saying “Because I don’t like you,” or “Because I felt like it,” she might say “Because your conduct posed a danger to the community.” This suggests that she doesn’t get to use her power just because she wants to – it implies that if the individual’s conduct hadn’t posed a danger to the community she wouldn’t have been entitled to punish him. However, this still falls short of the respect an official ought to offer a subject of law. For offering “Because your conduct posed a danger to the community,” standing alone, suggests that the official is the sole judge of what reasons suffice to use her power. She could have just as well said “Because you’re a jerk” or “Because the moon is in Virgo.”
If an official offers legal reasons for her use of coercive power, matters are different. Even if the law just authorizes the defendant’s imprisonment based on the same reasons that our judge would otherwise have offered on her own (e.g., “Anyone whose conduct poses a danger to the community is liable to imprisonment”), legal reasons are embedded in a network of other people’s judgments; depending on the precise details of the legal system in question, a legislature or prior judges will have decided that there should be a law about the conduct in question, appellate judges will review the trial judge’s decision to ensure that it’s actually authorized by the law, prior cases will have filled out the legal rule and given the one over whom power is to be used some idea of what sort of evidence he might offer to defend himself, and so forth. An official who offers legal reasons treats individuals respectfully by showing that her use of power isn’t just a matter of her own judgment, but responds to the judgment of all those other people, too. When an official listens to an individual’s arguments about how the law is to be applied in a particular case, she also expresses respect for that individual’s judgment.28
A judge who offers legal reasons for her use of power also treats the political community as a whole respectfully in two ways. First, she acknowledges that it’s not ultimately her judgment about the rightful use of that power that matters, but the judgment of the political community. Her judgment is involved in applying the legal rules, but only within bounds specified by the collective judgment.29 Second, she acknowledges the agency relationship between herself and the state. The judge who uses her official power without appealing to legal reasons is like an employee who disrespectfully uses her employer’s property as her own, commingling them and not distinguishing between her purposes and her employer’s purposes.
Officials avoid hubris by maintaining a separation of role and personal identity.30 When an official acts in an official role, she is bound by rules and to the practice of explaining her acts in terms of those rules; the rules and the practice of reason-giving constitute the role. By contrast, an official who is not so constrained communicates that her right to exercise coercive power over another individual is a personal property, rather than a property of her role. The explanation “Because the rules say so” attributes authority and status to the law, whereas the explanation “Because I say so,” like no explanation at all, attributes authority and status to the official. Even an official who exercises discretion, when he acts in good faith by doing so on the basis of legal reasons, again attributes authority and status to the rules; by incorporating reasons drawn from his personal preferences or beliefs, he attributes authority and status to himself.
B Terror
Nonofficials in states that do not comport with regularity and publicity, whether or not they are actually targeted by official violence, have good reason to fear officials. For they know that their well-being is at the whim of a class of people who can wield overwhelming force at will, and those over whose heads that threat is held have no means of defense.
Those who are subject to such terror are rendered unequal twice. First, they are subjected to the experience of relative powerlessness and fear.31 Second, they are forced to act out their own subordination by behaving submissively toward powerful officials. When an ordinary citizen passed by a member of the Tonton Macoutes or the KGB, he must have felt a pang of alarm, an urge to cringe away and avoid the attention of the wielder of fearsome powers. He must have been obsequiously polite if he was forced to interact with them, and would have been inclined to submit to any “request” the official made.32 Note that this is also true in a society that has achieved regularity, but not publicity. Even if the KGB officer is constrained by rules, if an ordinary individual doesn’t know what those rules are, or will have no say in their application, he still has reason to fear the officer’s power. The individual doesn’t know the circumstances under which the officer will be able to do him harm, and will not be able to participate in his own defense if he does come into conflict with the officer, instead having to trust other officials to protect his legal rights. He is likely to feel powerless and fearful even if he believes that there really are background rules regulating the officer’s behavior. By contrast, a nonofficial who can help herself to the power of rules that constrain the power of officials need not bow and scrape, because she can rely on those rules to keep the officer from retaliating against her for failing to do so.
The asymmetry confronted by an ordinary person facing the might of the state is an essential feature of inegalitarian terror. In a Hobbesian state of nature, we may have unconstrained power over one another, but it doesn’t make us unequal. Defenselessness in the face of overwhelming power creates the pervasive fear characteristic of systems of state terror. This is one reason that the rule of law is a regulative principle for state violence, not private violence.
Before moving on, consider the following objection. The law might actually authorize officials to terrorize; for example, it may permit judges to issue a torture warrant.33 Under such circumstances, this objection goes, the rule of law would not prevent terror: officials with the power to torture are terrorizing regardless of whether their power to do so is regularized by the procedural apparatus of a rule of law legal system.
However, regularized torture is different in kind from the terror that is inflicted in the sort of states where one is always subject to the knock on the door in the middle of the night from some KGB officer. We already have regularized torture in the contemporary Western liberal democracies: think of the horror of a US prison, pervaded by rampant violence, the punitive use of solitary confinement, extraordinarily negligent medical care, and too many other forms of torture to list. Yet the prospect of being put in one of those prisons does not ordinarily cast a pall over day-to-day life in the United States because, at least in those privileged (e.g., white, upper-income) communities where officials comport with the rule of law, ordinary people know that they aren’t likely to be put in prison unless they commit an actual crime, that they’ll have a chance to defend themselves beforehand, and so forth.
If officials wished to adopt regularized procedures to create full-fledged terror, they doubtless could do so. They could, for example, create a system of secret ex parte torture warrants, and thus replicate the knock on the door in the night under the aegis of procedural propriety. But this would manifestly violate the principle of publicity, as would any system in which officials were authorized to inflict brutal treatment on subjects without notice and an opportunity to defend themselves. By guaranteeing a minimum of warning, by guaranteeing that those subject to brutal treatment at least have an opportunity to put up a defense in a forum where their objections will be listened to and taken seriously, and by making it possible for people to minimize their risk of being subjected to official brutality by complying with public rules, the rule of law puts a strong upper bound on the extent to which any legal system can inflict terror.
Note the further important point that this entails: the weak version of the rule of law not only does not require liberal democracy, but can even be morally valuable in states other than liberal democracies. The case of the torture warrant shows that the weak version of the rule of law is morally valuable in a state that does not respect basic human rights. For another example, the weak version of the rule of law can be morally valuable in a state that does not respect political freedoms, and punishes dissidents, in virtue of the fact that it at least does not allow dissidents to be terrorized: at least they will get trials before they are punished, and the punishment will not come as a terrifying surprise. It follows that nonliberal states and nondemocracies are at least potentially blameworthy for not complying with the rule of law, and praiseworthy for complying with it, independent of their blameworthiness for not being liberal or democratic.34
C Normative robustness
I have said that an account of the evaluative side of the rule of law should be factually robust, in that it is likely to be true of a broad range of human societies, and normatively robust, in that it avoids controversial normative claims as much as possible. The case for factual robustness is primarily made over the following chapters, where the egalitarian conception of the rule of law is shown to match both history and the strategic structure of rule of law states.35
As for normative robustness, although equality is often a highly controversial ideal, some egalitarian ideas are uncontroversial: the claim that the state ought not to create a group of citizens (officials) who can engage in arrogant hubris over others or terrorize them into submission is unlikely to draw objections. The avoidance of hubris and terror is compatible with a very broad range of ways of thinking about equality and overall normative standpoints. Those, for example, who value treating the subjects of law with equal dignity can recognize that hubris and terror are wrong because of the way they create a status hierarchy between officials and ordinary people.36 Egalitarian democrats, who are concerned primarily with the distribution of political power, may note that the failure of the rule of law is inconsistent with the participation of ordinary citizens as equals in the political process, because officials could use terrorizing power to prop up their own rule against citizens’ wills.37 Those who are concerned with the egalitarian distribution of economic resources can note that terrorizing power enables rent seeking and exploitation. Welfarist egalitarians can note that the lives of those subjected to hubris and terror go dramatically less well than the lives of those who inflict it. Egalitarians concerned with capabilities can note that terror drastically reduces one’s functional opportunities.38 Even those who do not think of themselves as egalitarians, such as libertarians, can agree that the state ought not actually create hierarchies between individuals. Moreover, since the weak version of the rule of law does not require liberal democracy and its moral value is independent of liberal democracy, it does not require one to endorse liberalism or democracy in order to endorse its demands.
So much for the contemporary reason-giving power of hubris and terror, but one might object that the same cannot be said of the past. I advanced the normative robustness desideratum on the basis, in part, that it is necessary to make the philosophical/legal conception of the rule of law compatible with historical and social scientific explanations that take into account the actual motivations of those in rule of law states. However, the rule of law has been around, in various forms, much longer than the general consensus that the state should treat the subjects of law as equals. Pseudo-Xenophon, for example, criticized Athens for protecting slaves from hubris and terror.39
However, for determining whether a conception of the rule of law is normatively robust, the relevant motivations are those of officials and ordinary people who take the internal point of view in societies that already have the rule of law, as well as those who are fighting for the rule of law in societies without it. It is not relevant that Pseudo-Xenophon saw fit to criticize elements of the rule of law in the pursuit of his own oligarchic interests, interests naturally leading him to be opposed to the egalitarian institutions of democratic Athens, or that feudal states without the rule of law have been built on an ideology of natural inequality. (Chapter 5 will show that Athenian supporters of the rule of law indeed saw it as valuable for egalitarian reasons.)
Actually, Pseudo-Xenophon himself recognized the way that the rule of law promoted equality (although for him, being an aristocrat, this was a bug, not a feature). His account of the role of law in preventing hubris and terror provided much of the inspiration for mine:
Now amongst the slaves and metics at Athens there is the greatest uncontrolled wantonness; you can’t hit them there, and a slave will not stand aside for you. I shall point out why this is their native practice: if it were customary for a slave (or metic or freedman) to be struck by one who is free, you would often hit an Athenian citizen by mistake on the assumption that he was a slave. For the people there are no better dressed than the slaves and metics, nor are they any more handsome. If anyone is also startled by the fact that they let the slaves live luxuriously there and some of them sumptuously, it would be clear that even this they do for a reason. For where there is naval power, it is necessary from financial considerations to be slaves to the slaves in order to take a portion of their earnings, and it is then necessary to let them go free. And where there are rich slaves, it is no longer profitable in such a place for my slave to fear you. In Sparta my slave would fear you; but if your slave fears me, there will be the chance that he will give over his money so as not to have to worry anymore. For this reason we have set up equality between slaves and free men, and between metics and citizens.40
Pseudo-Xenophon invites us to compare two sorts of political community, represented by Athens and Sparta. In Sparta, citizens are more beautiful and richer than slaves, and express their superiority by striking their inferiors at will. Accordingly, slaves fear citizens and stand aside for them. By contrast, in Athens, slaves are immune from casual violence and, consequently, have no fear of citizens. Thus, slaves in Athens feel no need to stand aside. All of this entails that slaves and citizens are equal in Athens and unequal in Sparta.
The parallel between Pseudo-Xenophon’s insights and my argument is clear. First, Pseudo-Xenophon recognizes that unconstrained violence can be an expression of social status – consistent with the Athenian law against hubris, which recognized the insulting power of violence, to be discussed further in Chapter 5. Second, he recognizes the power of unconstrained violence to inflict terror on its victims and lead them to “stand aside” – to fear and behave submissively toward those who wield it. Regardless of whether any individual citizen actually strikes any individual slave, all slaves fear all citizens in Sparta, just by virtue of the fact that any citizen has the power to strike any slave. By contrast, in Athens, citizens and slaves, in Pseudo-Xenophon’s hyperbole, are equals, at least in respect of their immunity from casual violence.41 Pseudo-Xenophon shows us that the egalitarian significance of controlling arbitrary violence has been known for quite a long time.
IV Closing technicalities
Thus far I have assumed, but not defended, a variety of ideas about the sort of normative principle the rule of law is. Here, I list them. Most of these propositions either are uncontroversial, if often implicit, positions in the existing literature or are closely integrated into the conception as a whole such that they can be unproblematically accepted given the argument I’ve already made. However, before we move from the weak to the strong version, we need to have a clearer fix on the properties of the idea.
As I argued, the rule of law in the first instance governs states (or other discrete political communities), not individuals; we say that a state, not an individual official, comports with or violates the rule of law. However, to see whether the state comports with the rule of law, we must inspect officials’ use of the state’s coercive power. By “officials,” I simply mean those who wield the violent resources ordinarily associated with the state; this includes not only duly appointed officials of ordinarily legitimate states but usurpers, warlords in failed states, and the like. And by “coercive power,” I mean violence and commands backed up by the threat of violence.42
An official uses the state’s coercive power when she applies it to a specific person or known group of people. Ordinarily, officials who wield the state’s coercive power will be those exercising executive or judicial functions. The rule of law regulates the behavior of legislators indirectly: it commands that officials use the state’s coercive power only in accordance with laws holding certain properties; legislators can help bring it about that the state does or does not comport with the rule of law by enacting laws that do or do not hold those properties. Legislators may directly apply coercion to citizens in special cases, such as when passing bills of attainder.
The rule of law is observed or violated only by general patterns of behavior in a political community. If a single judge misuses his power in a way that is inconsistent with the three principles of the rule of law (by, for example, putting people in prison “just because I say so”), we don’t say the rule of law has failed; we say that particular judge is violating the law.43 If judges regularly do so, the rule of law has failed.
The rule of law’s constraints must be met reliably. It does not require any particular method of enactment or enforcement. For example, the United Kingdom has no written constitution, and adheres to the doctrine of parliamentary supremacy such that Parliament could, in principle, enact laws authorizing officials to violate the strictures of the rule of law at will. It nonetheless comports with the weak version of the rule of law to the extent that its constraints are a stable legal norm with which officials reliably comply, and to the extent that British officials who fail to do so can ordinarily anticipate social and political sanctions. (I discuss the extent to which these claims can and do correspond to British reality in Chapter 7.) However, a state will not count as adhering to the rule of law if its officials exercise their power in ways consistent with its demands out of mere benevolence, with no social, cultural, legal, political, or strategic constraints keeping them from violating it whenever they want.
The last two propositions – that the rule of law is about general patterns of behavior and norms, and that these patterns must be reliable – can be summed up in the claim that the rule of law is observed or violated only by a state’s institutions. I hesitate to say this, because the term “institution” pervades the academic literature and seems to have a different meaning every time.44 Here, I use the term “institution” to mean just the object of those last two propositions – reliable general patterns of behavior and norms. “Institutions” differ from the specific “practices” of a state, which are the sorts of things that are closer to a pretheoretic understanding of the word “institution,” such as the trial by jury, or the independent judiciary, or the writ of habeas corpus.45 Practices are nonexclusive ways of instantiating the three (institutional) principles; the independent judiciary, for example, is one way to bring it about that officials more reliably use coercion only pursuant to preexisting law. But the independent judiciary is not necessary for the rule of law – we can have the rule of law without it (for more, see Chapter 8). However, when people use the word “institutions,” they usually mean what I call “practices”; for ease of comprehension, I adopt this terminology toward the end of the book, particularly in the last few chapters, where I argue that no particular institutions (read: “practices”) are necessary components of the rule of law, though there must be some “institutions” that serve a given set of functions in each rule of law society. We only need the linguistic distinction between institutions and practices for a moment in order to get at the underlying conceptual idea.
The rule of law is (relatively) formal. By this, I do not mean that the rule of law does not regulate the substantive content of law – obviously it does, and when we reach the principle of generality it does so fairly pervasively. Instead, I mean to reject thicker conceptions of the rule of law, such as Dworkin’s “rights conception,”46 in which the rule of law basically requires liberal democracy, or the familiar notion that the rule of law requires an extensive system of private property rights. On the contrary, the rule of law does not require citizens to have any specific rights at all (other than the procedural rights to have access to the law and be heard in their own cases, as given by the principle of publicity), except insofar as those rights are necessary to constrain officials to use their power only consistent with the three principles. It is possible to have the rule of law without any private property, as in a state in which all property is held collectively but officials do not abuse their power.47 I will not defend this thin conception of the rule of law here (except to note that we already have perfectly good normative arguments for liberal democracy and private property rights).48
The rule of law is a continuum, not a binary: states can satisfy it to a greater or lesser extent.49 It is a continuum along three dimensions. First, a state can satisfy some of the principles but not others. Second, a state can satisfy a principle to a greater or lesser extent. Third, a state can satisfy its principles with respect to some citizens but not others – it could, for example, comport with publicity, regularity, and generality with respect to the elites but not the masses.
Finally, I will not say that the rule of law is necessary for a state to have law in the first place. There can be legal systems (Ancien Régime France, the Soviet Union) that radically fail to meet the standards of the rule of law, and still count as having law nonetheless. This is a controversial position.50 An alternative approach would follow Simmonds in understanding the concept of law to refer both to an ideal and to real-world practices. On Simmonds’s view, there is an “archetype” of law, which is “intrinsically moral,” and which tracks ideals like those typically captured under the notion of the rule of law; we understand real-world laws as such in virtue of their partaking, in very much a Platonic sense, of the ideal concept (however imperfectly).51
However, nothing important is at stake in this dispute. Both sides admit of the possibility of saying that even tyrannical states have law, even if the Simmonds position would call it law that is extremely flawed in virtue of its failure to conform to the Form of Law. We might then describe a continuum from Law0 to Law1, where the latter is Simmonds’s archetype, and the former is (say) the legalistic Soviet or Nazi bureaucracies. Similarly, all agree that there is a point along that continuum where we can find morally valuable properties. These points of agreement between those who would attach the rule of law to the concept of law and those who do not are enough for the argument of this book.
Likewise, all agree (following Fuller) that the morally valuable properties of rule of law systems have some connection to the technology of law itself.52 A core function of law is to give authoritative commands, and commands must at least be epistemically available to those who are supposed to obey them, and must be reliably backed by enforcement to get that obedience.53 This entails that the law must be minimally public, and officials must be minimally constrained.54
But here, too, we seem to be working with a continuum where moral value appears in the middle, not at the beginning. RuleofLaw0, a minimally regular and minimally public law that serves only to make it cheaper for an otherwise unconstrained top-level official to hold lower-level officials to her program and give commands to the rest, is unlikely to have moral value. Caligula could enact a (minimally public and regular) law when he wanted obedience, and just order people executed when he wanted amusement.55 Somewhere on the way to RuleofLaw1, the fully general and egalitarian state that will be described in the next two chapters, the rule of law begins to acquire moral value – when, for example, it begins to forbid Caligula’s amusement-executions.
The weak version of the rule of law is something like RuleofLaw0.5; I now turn to RuleofLaw1, and the principle of generality. The next two chapters are devoted to expressing the ideal of the rule of law in its most demanding form.
Open access