Why Classical Liberals Care about the Rule of Law (And Hardly Anyone Else Does)

Politics Has Captured the American Legal System


In 1776 John Adams declared that America was “a nation of laws, not men.” Politicians of all persuasions have used Adams’s phrase ever since to claim the moral high ground. Such rare agreement among the political classes, even if only rhetorical, is an indication of the power of the idea of the rule of law.

What does it mean to have the rule of law? Adams’s opposition of “a nation of laws” to one of “men” suggests part of the answer. A “nation of men” would be one governed by the desires and whims of the rulers, unconstrained in their impositions on those they ruled. Disputes would be decided according to the rulers’ preferences, not principle, leaving individuals dependent on having a protector among the rulers. A “nation of laws,” on the other hand, would substitute principle for preference.

Consider a simple question of whether John must repay a loan from Mary. In a “nation of men,” the fact of the loan would be irrelevant; what would matter was whether John’s patron outranked Mary’s. In a “nation of laws,” on the other hand, if Mary could establish that John had borrowed money from her and not repaid it, the principles of contract law would require John to repay the loan regardless of whether John had more powerful friends than Mary.

In other words, societies without the rule of law would be Hobbesian states of nature, with escape possible only through the intervention of the Leviathan. That escape would be purchased with the surrender of liberty to the Leviathan and the acceptance of his exactions as the price of preventing the greater losses of the war of all against all. In the society with the rule of law, on the other hand, individuals would not fear their neighbors, since disputes would be settled through the application of principles known in advance. These individuals would have no reason to submit to the local Leviathan. (What we know about a variety of primitive” societies suggests that the rule of law was far more common than was once believed. Bruce Benson’s The Enterprise of Law, for example, documents the widespread existence of the rule of law in a variety of pre-modern societies.)

In a society governed by the rule of law, we should expect to observe two key features. First, the principles by which disputes will be resolved are known in advance. John knows before he borrows the money from Mary that promises to repay loans are enforceable. Second, the result of the application of those principles to a dispute does not depend on who the parties are. Powerful people are governed by the same rules as the weak, the rich by the same rules as the poor.

These are necessary but not sufficient conditions for the rule of law.While we sometimes take these components of the rule of law for granted today, they have often been absent under tyrannical rulers. For example, the Roman Emperor Caligula had imperial decrees written in small letters and posted so high up on pillars that the decrees could not be read, ensuring that citizens could not know whether or not they had violated the law.

These conditions are enough to exclude arbitrary tyranny like Caligula’s, but not enough to get us to a free society. We need to constrain the content of laws for two reasons. First, for there to be the rule of law, both John and Mary must be secure enough in their property rights to have accumulated sufficient assets to make their transaction possible. Second,we need to be able to eliminate orderly tyrannies such as fascism and communism, which oppress people with laws adopted in accordance with “proper procedures,” as well as merely arbitrary ones such as Caligula’s.

The German Nazi regime, for example, had many well-known, well-understood, and equally applied rules. The infamous Nuremberg laws, for example, outlawed marriages between “citizens of German or some related blood” and Jews. The law was known and understood by the public, and applied equally. As in most dictatorships, of course, enforcement of the law was sometimes arbitrarily waived — but our primary objection to the Nazi regime as violating the rule of law cannot be that its hideous laws were imperfectly enforced. Some restraint on substance of what lawmakers can do seems necessary.

Of course, the Nuremberg laws were built on the distinction between those of “German or some related blood” and Jews, a distinction we today recognize as having no validity whatsoever. Germany is not unique in having introduced such distinctions: much turned on minute distinctions of the degree of “African blood” under pre-Civil War and segregationist American laws, and similar nonsense continues today, from the United States (over “native Hawaiian” and American Indian status) to New Zealand (over Maori status). It is also present in nonracial distinctions—employment-discrimination laws treat people differently depending on whether they are large or small employers, employers or employees, or disabled or able-bodied, to name but a few such distinctions. Indeed, one of the fundamental problems in today’s legal system is that our rights often depend on how the law categorizes us, an issue to which we will return below. The problem of treating individuals differently is thus not merely a relic of past tyrannies but a real problem in today’s legal system.

Constraining Content

Constraining the content of rules requires that we constrain the power of the bodies that create them. Any reasonable theory of the rule of law ought to be able to distinguish the Nazi legal system from the legal system of a free society. Indeed, this seems like a minimal requirement. There are lots of closer cases where we want our theory to do much harder work in drawing distinctions. We thus need to add some kind of substantive constraint to our initial set of conditions. This is where things get tricky.

The notion that the rule of law requires a constraint on the reach of lawmaking powers is far from universally accepted. For example, Hans Kelsen, one of the twentieth century’s major legal theorists, does not accept it. Kelsen (who was forced to flee Germany when Hitler came to power) thought the Nazi racial laws met the definition of law precisely because he defined law in terms of the ability of a state to back a command with the threat of force and did not impose substantive constraints on the content of the law. His theories are still taught throughout the world, particularly in civil-law countries, and he numbers among the fans of his jurisprudential theory Judge Richard Posner, one of the most influential figures in American law today. (Fortunately, given his position, Posner’s own jurisprudence is considerably better than Kelsen’s.)

We also have the problem of where to find the constraints we will impose. In his book A Brief History of Time, Stephen Hawking relates an anecdote of a scientist who, after delivering a lecture on the structure of the solar system and galaxy, was approached by a little old lady. She told him that his talk was rubbish because everyone knows that the earth is a flat plate balanced on the back of a giant turtle. The scientist responded by asking her what the turtle stands on. She replied, “You’re very clever, young man, very clever. But it’s turtles all the way down.” Unfortunately many attempts to locate substantive constraints on the legal system devolve into resting our premises on ever-increasing numbers of turtles.

As F. A. Hayek perceptively noted in 1973, the problem is the legal system’s refusal “to recognize as binding any rules of conduct whose justification had not been rationally demonstrated or ‘made clear and demonstrative to every individual,’ ” a problem that has been an “ever recurring theme” since the nineteenth century (Law, Legislation and Liberty, vol. 1, p. 25; subsequent page numbers are from this book). Just as central economic planning intruded on the spontaneous order of the marketplace throughout the twentieth century, central legal planning displaced ever more of the spontaneous order of the common law. As a result it is turtles all the way down for most modern legal theorists. As they seek to impose their ever-more complex, planned, rational legal orders on society, they stack turtle on turtle in trying to locate a plausible source for the legislation they write.

External Sources of Constraints

One possible source for substantive constraints is to look outside the legal system for a set of principles to guide the law. There are an infinite variety of possible sources: feminism, Marxism, fascism, divine law, or whatever ideology is the current favorite. All of these suffer from the turtle problem, of course, since we’ve simply substituted the problem of justifying feminism, Marxism, and so on for the problem of justifying the constraints on the legal system. Adherents to these various worldviews recognize the fundamental truth of their particular views, but convincing the rest of us has generally come down to using state power to coerce compliance. Of course, the next turtle down is the problem of how to resolve conflicts over which theory to use. For those who identify law only as the command of the sovereign, this is not a problem. The ism with the most guns gets to tell the rest of us what to do, and we should just be thankful if it doesn’t turn out to be the Khmer Rouge or the Taliban running the show. This is not to suggest that my religious or philosophical beliefs shouldn’t play a major role in shaping my conduct, just as yours do in shaping your conduct. The issue is whether my beliefs get to play a role in shaping yours and vice versa.

The records of societies that looked outside the legal order to religious, political, or other moral codes as the source of legal constraints are neither libertarian nor particularly happy ones. The Soviet Union, Puritan Massachusetts, and the Taliban regime in Afghanistan are just a few of the failed societies that relied on an external set of beliefs as the source of rules in the legal system. Unfortunately, they are also societies more famous for witch trials (political as well as religious) or mass killings than for flourishing economies or freedom. The lesson is that if we are to have a society in which people are free to hold diverse opinions about religious and ethical issues, we cannot rely on such sources to constrain power without igniting conflict over which religion or ethical system on which to rely.

Another possible source of constraints is to limit the subjects about which laws can be written. Disputes over the morality of consensual behavior between adults, for example, can be left to the pulpit and water cooler if the state is not authorized to legislate about such matters. The U.S. Constitution takes this approach by carefully enumerating the subjects over which the national government has authority. Through such limits, the Framers hoped to constrain the national government to a relatively narrow sphere.

Unfortunately, however, this strategy proved insufficient to permanently prevent the expansion of government power. Once the federal government was in place, it became a permanent lobby for enhancing its own powers and over time, bit by bit, many of the constraints imposed by the Constitution were worn away. For example, the Supreme Court allowed a steady expansion of congressional power under the Constitution’s Commerce Clause (Article I, section 8, clause 3), ultimately holding in a 1942 opinion (Wickard v. Filburn) that Congress could rely on its interstate commerce power to punish a wheat farmer for growing wheat on his own land for his own consumption. (Sadly, the Court again endorsed this approach this year in Gonzales v. Raich,
upholding federal rules barring the medical use of marijuana.)

Although more recently the Court has attempted to revive the notion of enumerated powers as a meaningful constraint on the federal government, the extent of its powers today would surely shock even the most ardent proponents of a strong national government among the Founders. While imposing as many constraints on state power as possible through a written constitution is worthwhile, what must be recognized is that such constraints are vulnerable to erosion from the ever-present pressures from governments and interest groups.

Hayek’s Solution

Where, then, can we find a source of limits that will constrain state power and enable us to live under the rule of law rather than men? Hayek wrote extensively about law, applying the insights of Austrian economics to the study of legal institutions in The Constitution of Liberty (1960) and a three-volume work Law, Legislation and Liberty (1973, 1976, and 1979, respectively). Although there are certainly ambiguities and inconsistencies in his analysis, Hayek provided at least a partial solution to the problem of securing the rule of law, neatly addressing how to permanently constrain the domain of law-making and locating a mutually acceptable source of the constraint.

Central to Hayek’s theory of law is the distinction between law and legislation: law is a spontaneous order that came largely from custom; legislation is a planned order created by human institutions such as legislatures. The distinction seems at odds with common usage. We often talk of “laws” passed by Congress and state legislatures. Few modern statutes would meet Hayek’s definition of law, however. (Hayek did recognize the need for organizational statutes to structure the state, and transaction-cost-reducing measures that offered focal points such as clear rules on the formalities necessary to conclude a binding contract.) By introducing this distinction, Hayek pointed us toward a solution to the problem of constraining legislative power. Because it comes from custom and is the result of a decentralized process of dispute resolution, Hayekian law is not vulnerable to the interest-group pressures that bias the legislative process. (Hayek did not entirely exclude the legislature from contributing to the production of law. He allowed for intervention to save the legal system from conceptual “dead ends” (p. 100), although he never fully specified how one can distinguish a solution to a “dead end” from special-interest legislation.)

In a Hayekian legal order the judge “serves, or tries to maintain and improve, a going order which nobody has designed, an order that has formed itself without the knowledge and often against the will of authority, that extends beyond the range of deliberate organization on the part of anybody, and that is not based on the individuals doing anybody’s will, but on their expectations becoming mutually adjusted.” (pp. 18–19) Crucially, “[t]he question for the judge here can never be whether the action in fact taken was expedient from some higher point of view, or served a particular result desired by authority, but only whether the conduct under dispute conformed to recognized rules” (p. 87). As a result of limiting the production of law to the outcome of dispute resolution, a Hayekian legal order’s rules focus on making “it possible at each moment to ascertain the boundary of the protected domain of each. . . .” (p. 107)

Legal rules, Hayek argued, must satisfy a principle of generality; that is, they must be general, non-arbitrary, and applied equally to all. Legal rules produced by common law courts and ultimately rooted in custom meet these criteria; most of the product of legislatures does not. Hayek’s generality principle solves the problem of constraint without resting on a turtle. We avoid the problem of treating the Nazis’ Nuremberg laws as valid because a law that distinguishes between those of “German or related blood” and Jews violates the requirement of generality by introducing a distinction. We avoid the problem of special-interest legislation in the same fashion: special interests can’t be “special” if they cannot distinguish themselves from everyone else. As a result, they cannot confiscate our assets for their benefit.

What kinds of laws pass Hayek’s test of generality? Rules that allow individuals to make and keep private agreements, resolve conflicts, and structure their dealings meet the test. The essentials of property, tort, and contract pass unscathed.

What kinds of laws fail? Not much of the modern administrative welfare state would survive. Among the casualties:

• Environmental laws that prescribe differential treatment for different interests (such as the distinction between point and nonpoint sources under the Clean Water Act, which leads those who emit the exact same amount of the exact same pollutant from a field treated with fertilizers and from a factory to be treated radically differently).

• Employment laws that treat different groups of employees and employers differently.

• Regulatory laws that limit individuals’ freedom to enter occupations.

Indeed, all the many statutes that take the property rights of one person and indirectly award them to another (so-called regulatory takings) would fail Hayek’s test. (Hayek introduced some ambiguity on this point in volume three of Law, Legislation and Liberty, by suggesting there might be exceptions to the principles set out in the book for employment, environmental protection, and a surprising number of other areas. This is one of the rare occasions when I feel comfortable saying he was simply wrong.)

Why We Need the Rule of Law (and Not Much Else)

This would not leave us helpless, however. As Richard Epstein has eloquently written, simple rules turn out to be remarkably robust for resolving the problems of a complex world (Simple Rules for a Complex World, 1997). The principles of tort, property, and contract are enough to enable market forces to produce solutions to problems that bedevil regulators trying to write comprehensive regulations, while avoiding the special-interest problems inherent in government action. Indeed, this relatively small set of laws is all we need for the same reason that we do not need state assistance in making markets “work.” In fact, Hayek’s legal theory is rooted in his economics, particularly his understanding of the critical role dispersed knowledge plays.

Just as Hayek showed in his 1945 essay,“The Use of Knowledge in Society,” that attempts to “fix” specific problems in the marketplace by bureaucratically altering prices damage the market’s ability to reconcile the diverse needs and resources of the millions of individuals, so too does he explain how attempts to “fix” legal problems through special-interest legislation damage the legal system’s ability to provide law. Once courts move beyond enforcing the expectations of the parties in an attempt to produce specific distributional outcomes, the legal system’s ability to reconcile the actions of individuals pursuing their diverse aims is damaged. No longer able to rely on the enforcement of their voluntary arrangements, people turn to Leviathan for protection, competing for recognition of their status as “deserving” and the special treatment such status brings. The result is a never-ending cycle of special-interest lobbying, bringing ever-increasing numbers of laws doling out favored status to clients of the powerful.

We can see this dynamic illustrated in the ever-increasing list of protected classes in employment law. The list of prohibited bases for employment decisions has expanded in some American jurisdictions to include sexual orientation, appearance, weight, and use of tobacco. With such an extensive list of prohibited bases for discrimination, of course, employers can no longer simply hire and fire employees. To avoid lawsuits, employers must invest in expensive compliance programs and vet their advertisements to avoid the appearance of impropriety.

Why No One Else Cares

Are classical liberals alone in caring about the rule of law? Increasingly it appears so. A nominally conservative administration in Washington seems more intent on outcomes than principle, abandoning federalism whenever it sees an opportunity to impose a desired outcome nationally. Some special-interest legislation is to be expected from any politician as the price of politics (the 2002 steel tariffs were a particularly clear example of such a measure), but the widespread abandonment of principle in areas as diverse as environmental regulation and tort “reform” makes clear the lack of commitment to the rule of law. At the same time, a nominally liberal opposition articulates its challenges to judicial nominees not in terms of their qualifications but based on whether they are “in the mainstream” of political discourse —in other words, whether the nominees will take the “correct” political position while on the bench.

Supreme Court Justice Antonin Scalia recently noted that the bitter fights over judicial confirmations are a sign that the courts are no longer about interpreting legal texts but about political decision-making.

If we are selecting lawyers, if we are selecting people to read a text and give it the fair meaning it had when it was adopted, yes, the most important thing to do is to get a good lawyer. If on the other hand,we’re picking people to draw out of their own conscience and experience a new constitution with all sorts of new values to govern our society, then we should not look principally for good lawyers.We should look principally principally for people who agree with us, the majority. . . . And that is why you hear in the discourse on this subject, people talking about moderate, we want moderate judges. What is a moderate interpretation of the text? Halfway between what it really means and what you’d like it to mean? There is no such thing as a moderate interpretation of the text. Would you ask a lawyer, “Draw me a moderate contract?” The only way the word has any meaning is if you are looking for someone to write a law, to write a constitution, rather than to interpret one.

Politics has captured the American legal system, an inevitable consequence of the judicial activism that flowed from the New Deal-era capitulation of the courts to legislative and executive power. As a result, interest groups from business lobbies to “public interest” groups see the courts as simply one more battlefield on which to seek special treatment. Many profit from this state of affairs — lawyers, lobbyists, regulators, and the interest groups themselves. They are not interested in ceding this profitable arena by acquiescing in the reinstitution of the rule of law.

All is not lost, however. Old constraints have revived. Richard Epstein’s book Takings (1989) launched a revival of the federal takings doctrine that promises to yield some limits to special-interest legislation. (The current Supreme Court’s determination not to follow Epstein’s analysis through to its logical conclusions is limiting the impact Takings ought to have. Unfortunately the decision in Kelo v. City of New London is only the most recent example of this.) Last year the Michigan Supreme Court reversed its pernicious Poletown decision and reinvigorated its state constitutional takings jurisprudence (County of Wayne v. Hathcock). Restoring the rule of law will be a long hard road, but articulating the benefits can help create the atmosphere in which it is possible. Ideas matter—and on this issue classical liberals have the better idea.

Download File


November 2005



Andrew P. Morriss is the D. Paul Jones, Jr. & Charlene A. Jones Chairholder in Law and Professor of Business at the University of Alabama. He is coeditor (with Roger E. Meiners and Pierre Desrochers) of Silent Spring at 50: The False Crises of Rachel Carson, forthcoming from the Cato Institute.

comments powered by Disqus


* indicates required


October 2014

Heavily-armed police and their supporters will tell you they need all those armored trucks and heavy guns. It's a dangerous job, not least because Americans have so many guns. But the numbers just don't support these claims: Policing is safer than ever--and it's safer than a lot of common jobs by comparison. Daniel Bier has the analysis. Plus, Iain Murray and Wendy McElroy look at how the Feds are recruiting more and more Americans to do their policework for them.
Download Free PDF