What Is Rule of Law? 

Bo LI

Perspectives, Vol. 1, No. 5

"We are free because we live under civil laws." --- Charles de Secondat Montesquieu

"Rule of law" is one of the much said but little understood concepts in popular press and daily conversations in China today. What is rule of law? What is its significance? Does rule of law mean that there is no "rule of person?" What are the institutional conditions and cultural content of rule of law? How do we achieve rule of law? I intend to tackle these questions in a series of essays. In this essay, I will focus on the meaning of the rule of law and its values. In the next essay of this series (to be published in the June issue of "Perspectives"), I will talk about the implementation.

As a starting point, I want to point out that when we say "rule of law" these days, we mean something different from the instrumentalist conception of "rule by law" of the legalist philosophers in ancient Chinese history. When we say "rule of law" today we intend to describe a key component of the social and political orders found in the United States and other liberal democratic states of our time. In other words, by "rule of law" we mean a western tradition that can be traced back to the Roman republics and was fully developed by the liberal constitutionalism. It is characterized, in the words of Max Weber, by "legal domination."

The difference between "rule by law" and "rule of law" is important. Under the rule "by" law, law is an instrument of the government, and the government is above the law. In contrast, under the rule "of" law, no one is above the law, not even the government. The core of "rule of law" is an autonomous legal order. Under rule of law, the authority of law does not depend so much on law's instrumental capabilities, but on its degree of autonomy, that is, the degree to which law is distinct and separate from other normative structures such as politics and religion. As an autonomous legal order, rule of law has at least three meanings. First, rule of law is a regulator of government power. Second, rule of law means equality before law. Third, rule of law means procedural and formal justice. We will take up these meanings of rule of law one by one.

First, as a power regulator, rule of law has two functions: it limits government arbitrariness and power abuse, and it makes the government more rational and its policies more intelligent.

The opposite of rule of law is rule of person. There are two kinds of rule of person. The first kind is "rule of the few persons," examples of which include tyranny and oligarchy. The second kind of rule of person is "rule of the many persons," an example of which is the ancient Greek democracies. The common feature of rule of person is the ethos that "what pleases the ruler(s) is law." That is, under rule of person, there is no limit to what the rulers (the government) can do and how they do things.

In contrast, a key aspect of rule of law is "limitation;" i.e., rule of law puts limits on the discretionary power of the government, including the power to changes laws. This is why the western juridical tradition is Roman, not Greek. One of the major problems of ancient Greek democracy is that its conception of law does not contain the idea of limitation. The Greek word "eleutheria," commonly translated as "freedom," connotes a freedom that extends into the principle that what pleases the people is law. In other words, there were no limits to the (democratic) governments of ancient Greece, and the popular will, be it short-term passion or long-term rationality, would always become law if the demos so wished. "As soon as law lost its sacred character, popular sovereignty was placed above the law, and, by that very act, government by laws was once again fused and confused with government by men" (Sartori, 1987, p. 307).

Unlike the Greek system, the Roman system of law limited the ability of the rulers to change laws, and it greatly influenced the Anglo-Saxon version of rule of law. At the core of the Anglo-Saxon conception of rule of law is the idea that the discretionary power of the government should be limited. "[W]henever there is discretion there is room for arbitrariness, and . . . in a republic no less than under a monarchy discretionary authority on the part of the government must mean insecurity for legal freedom on the part of its subjects" (Dicey, 1982, p. 110). The solution to this problem, say liberal democrats, is rule of law.

There are two common misconceptions in China. First, when some writers describe "rule of person," they mean only "rule of the few persons." These writers think that, as long as we have democracy ("rule of the many"), we have justice and rule of law. These writers forget that the popular will can rule with or without constitutional and legal limits. Without constitutional and legal limits, popular will can be as destructive as, or even more destructive than, the unfettered discretion of "the few." Examples include the injustices of classical Greek democracy, terrors of the French Revolution and the crimes against humanity committed during the Chinese Cultural Revolution. Second, some writers in China think that, as long as laws are passed through democratic procedures, they represent the general will (a la Rousseau) and therefore are just laws. These writers forget that "popular will" is not necessarily "general will" (as Rousseau himself warned us more than two hundred years ago). Unfettered popular will can not only deviate from the Rousseauian general will, it can run against and destroy the general will.

In more specific terms, how is government arbitrariness constrained? The answer lies in several important principles of rule of law. First, if we are to limit government caprice, rule of law requires the supremacy of law as opposed to the supremacy of the government or any political party. To the noted English jurist A. V. Dicey rule of law means, "in the first place, the absolute supremacy or predominance of regular law as opposed to the influence of arbitrary power, and excludes the existence of arbitrariness, of prerogative, or even wide discretionary authority on the part of the government" (Dicey, 1982, p. 120).

Second, if the government is to be restricted in its exercise of discretion, the government has to follow legal procedures that are pre-fixed and pre-announced. As F. A. Hayek puts it, rule of law "means that a government in all its actions is bound by rules fixed and announced beforehand -- rules which make it possible to foresee with fair certainty how the authority will use its coercive powers in given circumstances, and to plan one's individual affairs on the basis of this knowledge" (Hayek, 1994, p.80). For example, in constitutional and criminal law, there is a prohibition on "ex post facto" laws, that is, no one should be punished for a crime not previously defined in law. In other words, the government cannot simply define a new crime and apply the new definition retrospectively. The rationale for this principle is that, first, the government should not be allowed to abuse its power by punishing individuals for political or other conveniences; second, it would be grossly unjust and oppressive for the government to punish someone for behavior that was not known to be criminal at the time of commission; third, to so punish individuals would result in so many uncertainties that it would create great inefficiencies.

Rule of law as a constraint on government power is well recognized, but its cognitive value in enhancing government's rationality is often less understood. Rule of law not only limits the arbitrariness of the government, it also makes the government more intelligent and articulate in its decision making. For one example, as Professor Stephen Holmes writes, "[o]nly a constitution that limits the capacity of political decision makers to silence their sharpest critics . . . can enhance the intelligence and legitimacy of decisions made" (Holmes, 1995, p. 8). For another example, the key reason why liberal democrats do not believe in the pure will theory of legality is that, without rule of law as a limit, popular will can easily be corrupted by passions, emotions and short-term irrationalities. As such, liberal democrats demand rule of law because it helps us to behave according to our long-term interest and reason.

One comment on the relationship between rule of law and liberalism is due here. Liberalism requires a limited government, and it naturally finds rule of law (as a regulator of government power) a necessary institutional arrangement. We can say that liberalism requires rule of law, and that rule of law is an institutional realization of liberal ideals. Historically, however, rule of law existed before liberalism. According to Dicey, rule or supremacy of law was firmly established in England before the end of the sixteenth century, when liberalism as a political and social philosophy had not been completely born yet -- John Locke was born in 1632 and his Second Treatise of Government was first published in 1690. However, rule of law as a constraint on government arbitrariness might have provided a strong institutional and cultural support for the birth of liberalism in England, whose core teachings included the doctrine of limited government and the right of revolution.

The second meaning of rule of law, according to Dicey, is equality before law. "[N]ot only that . . . no man is above the law, but (what is a different thing) that . . . every man, whatever be his rank or condition, is subject to the ordinary law of the realm and amenable to the jurisdiction of the ordinary tribunals. . . . [T]hough a soldier or a clergyman incurs from his position legal liabilities from which other men are exempt, he does not (speaking generally) escape thereby from the duties of an ordinary citizen" (Dicey, 1982, p. 114-115). To Dicey, even in 1915 this principle of rule of law was not universally true among the liberal democratic countries of Europe. In England the idea of legal equality had been "pushed to its utmost limit" by 1915, but in France the officials were "to some extent exempted from the ordinary law of the land, protected from the jurisdiction of the ordinary tribunals, and subject in certain respects only to official law administered by official bodies" (Dicey, 1982, p. 115). By now, however, equality before the law is a universally recognized principle in all liberal democratic countries, although different countries might, at the margins, have different interpretations of what that equality entails.

The third meaning of rule of law is formal or procedural justice. What is formal or procedural justice? Before we answer this question, we need to answer a more preliminary question: what is formalism? Max Weber categorizes legal systems into four kinds: formally irrational, substantively irrational, formally rational, and substantively rational. Rationality refers to the generality and universality of law. Formality refers to the characteristic that the criteria of lawmaking and lawfinding are intrinsic to the legal system itself; that is, all rules, procedures and decisions can be deduced from the legal system itself. In contrast, a legal system that emphasizes substantive qualities of lawmaking and lawfinding uses factors outside law, such as ethical, emotional, religious or political factors, to evaluate cases. To Weber, only a formally rational legal system can achieve "legal domination" (rule of law) through consistent application of general rules, because only a formally rational legal system can maintain a "consistent system of abstract rules" that is necessary for rule of law.

A formally rational legal system, according to the western legal tradition, also results in justice that we desire. This kind of justice is called formal or procedural justice, which "connotes the method of achieving justice by consistently applying rules and procedures that shape the institutional order of a legal system" (Shen, 2000, p. 31). More specifically, formal or procedural justice consists of several principles. First, the legal system must have a complete set of decisional and procedural rules that are fair. Second, the fair rules of decision and procedure must also be pre-fixed and pre-announced. Third, these decisional and procedural rules must be transparently applied. Fourth, these decisional and procedural rules must be consistently applied. When these four conditions are satisfied, western judges and lawyers will say that they have achieved a certain kind of justice, which is called formal or procedural justice. Note that this notion of justice is more concerned with process and procedure than with the end result. As Selznick puts it, "legality has to do mainly with how policies and rules are made and applied rather than with their contents" (Selznick, 1969, as cited in Shen, 2000, p. 30). In other words, as long as the process is fair, transparent and consistent, justice is obtained and legality is achieved.

One example will help illustrate the concept of procedural or formal justice in contrast to substantive justice. If, in truth, a person has killed another person, substantive justice requires that the killer be punished according to law. However, if the killer is illegally tortured by the police to confess to his crime and, as a result of the confession, the police find conclusive evidence (i.e., evidence proving guilt beyond reasonable doubt), such as the weapon, the body of the victim, etc., for the court to convict the killer (which results in substantive justice), there is no procedural justice because the process of finding guilt has violated the basic rights of the killer who, before the conviction, is a citizen entitled to the full protection of the Bill of Rights.

In this case, based on the well-established law of criminal procedure, an American judge will not allow the record of confession (obtained through torture) and anything found as a direct result of the confession (such as the weapon and the body) to go into the court as evidence. As such, the jury will never see these items as evidence, and if the prosecutors have no other good evidence, the killer is likely to be acquitted, even though substantive justice requires that the killer be punished (because, for example, the weapon and the body might prove the guilt beyond reasonable doubt due to the fact that the killer knows where the weapon and the body are, and the weapon contains the killerí»s fingerprints.) In this way, in the United States, procedural justice triumphs over substantive justice in this particular case. In the end, the American judge will claim that justice is done simply because the pre-determined procedural rule (e.g., illegally obtained evidence is not admitted in court) is consistently and transparently applied.

One prominent Chinese diplomat in the United States once complained to some American friends that the United States should not blame China for human rights violations. The United States itself, the diplomat observed, is not a just society. The diplomat then went on to use the case of O. J. Simpson as an example of the lack of justice in American society. With all the evidence, said the diplomat, it was obvious that Mr. Simpson was guilty, and an overwhelming majority of American people also thought he was guilty, but the court set him free in the criminal trial. "Is this justice?" The diplomat asked in disbelief. Obviously, the diplomat did not understand the concept of procedural justice. In the case of O. J. Simpson, the judge could confidently conclude that justice was achieved because the trial was conducted according to pre-fixed and pre-announced procedures in a fair, transparent and consistent way. If the government could not prove Mr. Simpson's guilt beyond reasonable doubt, Mr. Simpson should be set free. This procedural rule ("prove guilt beyond reasonable doubt in an ordinary court") was a well-established legal rule known long before the trial started, and the rule was fairly, transparently and consistently applied to Mr. Simpson's trial.

You might ask: does it make sense to emphasize procedural justice? The general answer is yes. In a system that sacrifices procedural justice for the sake of substantive justice, the danger of arbitrary government power and the threat to individual liberty will be too great. Eventually, that system will lead to substantive injustice as well. In contrast, in a system that emphasizes procedural justice, arbitrary government power will be checked, liberty will be protected, and substantive justice will be preserved in the long term (if we believe that truth is best obtained through contest and debate between equals).

More specifically, formal or procedural justice has at least three values. First, without fair and just procedure, there is no guarantee that the end result will be just (that is, substantive justice cannot be guaranteed). As such, procedural justice is seen as a necessary condition for substantive justice. This is why the western legal tradition places a much higher value on formal or procedural justice than its East Asian counterpart, which puts more emphasis on substantive justice. In fact, some western legal scholars regard procedural justice as the only workable method for reaching substantive justice, and to these scholars procedural justice should be the only concern of the players within the formally rational legal system.

Second, formal or procedural justice is a condition for constraining government arbitrariness and protecting individual rights. When the government is required to follow pre-fixed, transparent and fair procedures before it can deprive a person's life, liberty or property, the danger of government arbitrariness is substantially reduced and the prospect for wrongful deprivations of individual rights is also significantly diminished.

Third, as Max Weber points out, procedural justice results in consistency, predictability and calculability that are desirable aspects of economic and social life. This second value of procedural justice is independent of any value we place on substantive justice and strengthens the argument for the western tradition of emphasizing procedural justice.

A comment on formalism is due here. Marxist writers often criticize the capitalist style rule of law as a sham. In these writers' vocabulary, "formal" is often synonymous with "superficial" and "unreal," and the distinction between formal and substantive justice is turned into a distinction between apparent and real justice. This characterization of rule of law is a gross mis-characterization. In western law, being formal has nothing to do with being superficial or unreal. "At the heart of the word 'formalism' . . . lies the concept of decisionmaking according to rule" (Schauer, 1988, p. 510). Furthermore, being formal is probably the only way through which a juridical order can achieve certain level of rationality. Professor Giovanni Sartori puts this point in more stark terms. "When we speak of 'juridical form' we are singling out the very requisite of a legal order. The form of law and the formal nature of law constitute . . . the characteristics by virtue of which a law is a law. . . . Formal is the method, not the result."

We have discussed three core meanings of the western conception of rule of law. Several caveats are in order. First, we have so far ignored one fundamental question. Those laws that constitute constraints on government arbitrariness and establish formal justice cannot be just any laws. They also have to be "civil" (a la Montesquieu) or just laws in some substantive sense. We have so far left the question, "which laws are civil or just laws?" unanswered. In other words, we have been focusing exclusively on the form of law, not the content of law. This focus is not accidental, because it has been the focus of western jurisprudence in the last one hundred years. In the western system of liberal democracy, the substantive justice (justice in content), or "iustum" in Latin, of law has been guaranteed by a constitutional state. "A large number of constitutional devices are, in effect, intended to create the conditions of a lawmaking process in which ius [the Latin word for law] will remain tied to iustum [what is just], in which law will remain the right law. For this reason legislation is entrusted to elected bodies that must periodically answer to the electorate. And for the same reason we do not give those who are elected to office carte blanche, but we consider them power holders curbed by and bound to a representative role" (Sartori, 1987, pp. 322-323). As such, western jurisprudence could afford to focus exclusively on the form of law.

This exclusive focus on the form of law is, however, also worrisome. Because of the existence of the constitutional guarantee on the substantive justice of law, western jurisprudence has come to have a purely formal definition of law. "Unfortunately . . . the formalistic school of jurisprudence completely overlooks . . . the fact that the formal definition of law presupposes the constitutional state. Therefore, the high level of systematic and technical refinement achieved by this approach cannot save it from the charge of having erected an unsafe juridical edifice open to easy conquest" (Sartori, 1987, p. 323). As such, when we study the western conception of rule of law, we should not forget that western legal scholars have been living in a constitutional state for so long that they are used to forgetting the importance of liberal constitutionalism in guaranteeing the just content of law. We Chinese do not have such luxury -- we do not have a constitutional state yet. When we talk about legality in China, we should not focus only on the form of law; we should also pay particular attention to the content of law. In fact, as we will see in the next essay of this series, without a constitutional state, we probably cannot guarantee anything: neither the content nor the form of law can be guaranteed to be just. In other words, without a constitutional state, neither substantive justice nor procedural justice, either in lawmaking or in the application of law, can be guaranteed. In the next essay, we will therefore spend a good amount of time on liberal constitutionalism.

Second, promoting rule of law does not mean that we should, or can, eliminate rule of person. Literal rule of law has its own costs (such as rigidity) and in some cases it can conflict with our sense of justice. In addition, it is probably impossible to eliminate rule of person completely. After all, laws are not given; they have to be made by certain people. Laws' applications are not automatic; they have to be applied by certain people. Even in the most advanced liberal democratic countries of our time -- the countries that are regarded as having the most developed systems of rule of law -- human factors play important roles in shaping traditions, customs and institutional cultures that are integral parts of the liberal democratic machinery. The real question is not whether we should eliminate or keep rule of person. The real question is about how to strike a balance between rule of law and rule of person so we can achieve liberty, equality, and justice. In this regard, liberal constitutionalism has been the only successful system. "Liberal constitutionalism is the technique of retaining the advantages of [rule of law as well as rule of person] while lessening their respective shortcomings" (Sartori, 1987, p. 308). We will discuss how this is so in the next essay.

We have discussed the meaning and the ideals of rule of law in this essay. In the next essay of this series, we will turn to the implementation of rule of law, particularly liberal constitutionalism.

(The author is an attorney with the New York law firm of Davis Polk & Wardwell.)

References: 

1. Dicey, A. V. Introduction to the Study of the Law of the Constitution. Indianapolis: Liberty Fund, 1982. 

2. Hayek, Friedrich. The Road to Serfdom. Chicago: University of Chicago Press, 1994. 

3. Holmes, Stephen. Passions and Constraint: On the Theory of Liberal Democracy. Chicago: University of Chicago Press, 1995. 

4. Sartori, Giovanni. The Theory of Democracy Revisited. Chatham, New Jersey: Chatham House, 1987. 

5. Schauer, Frederick. "Formalism." Yale Law Journal 97 (1988): 509-548. 

6. Selznick, P. Law, Society and Industrial Justice. New York: Russell Sage Foundation, 1969. 

7. Shen, Yuanyuan. "Conceptions and Receptions of Legality: Understanding the Complexity of the Law Reform in China." In The Limits of the Rule of Law in China, ed. Karen G. Turner, James V. Feinerman and R. Kent Guy. Seattle: University of Washington Press, 2000. 

8. Weber, Max. Economy and Society, ed. G. Roth and R. Wittich. Berkeley: University of California Press, 1978.