H.L.A. Hart’s legal positivism, set out in his 1961 book The Concept of Law, describes a legal system as a structured union of two kinds of rules held together by a shared standard that officials use to decide what counts as law. Primary rules tell people what to do or not do. Secondary rules govern how the primary rules are made, changed, and applied. Sitting above both is the “rule of recognition,” a criterion that judges, legislators, and administrators accept as the authoritative test for legal validity. Hart’s positivism holds that a rule’s status as law depends on whether it was produced through these recognized institutional channels, not on whether it is morally good.
What Legal Positivism Means in Hart’s Framework
Hart’s central commitment is the “separation thesis”: whether a rule is law and whether it ought to be law are two different questions. A statute can be validly enacted and morally repugnant at the same time. Legal validity turns on institutional pedigree, not moral merit.
This puts Hart against the natural law tradition, which holds that a sufficiently unjust rule fails to qualify as law at all. Hart saw that position as harmful, not because he was indifferent to justice, but because collapsing the two questions robs citizens of a clear vocabulary. If an immoral law “isn’t really law,” you lose the ability to say plainly, “this law exists and it is wrong.” Keeping validity and morality separate lets people criticize the legal system honestly without pretending the rules they oppose have disappeared.
Hart did not claim law and morality never overlap. Moral ideas regularly enter legislation and judicial reasoning. His narrower point was that morality is not a necessary condition for legal validity. A rule that comes through the right institutional channels is law, whatever anyone thinks of its content.
Why Austin’s Command Theory Was Not Enough
Before Hart, the dominant positivist account came from the nineteenth-century philosopher John Austin, who defined law as the commands of a sovereign backed by threats of punishment. Hart dismantled this model with a thought experiment that has become a fixture of legal philosophy. Imagine a gunman who points a weapon at you and demands your money. You hand it over. You were obliged to comply, in the sense that the threat gave you a powerful reason. But no one would say you had a legal obligation to hand over your wallet.
Hart’s point: Austin’s theory cannot distinguish the gunman from the tax collector. Both issue demands backed by consequences. The difference is that the tax collector operates within a system of rules that people accept as legitimate. Austin treats all law as the gunman situation scaled up, missing exactly the feature that makes a legal system different from organized coercion. The rest of Hart’s book is an attempt to explain that missing feature.
Primary and Secondary Rules
Hart’s alternative starts with a distinction between two kinds of rules. Primary rules tell people what to do or refrain from doing: don’t steal, pay your taxes, drive below the speed limit. Every functioning society has them.
A society with only primary rules would run into three chronic problems. The rules would be uncertain, because there would be no authoritative way to resolve disagreements about what they require. The system would be static, unable to adapt when circumstances change. And disputes about whether someone had broken a rule would drag on, because no one would have official authority to settle them.
Secondary rules solve each of these problems by operating on the primary rules themselves. Rules of recognition identify which primary rules count as legally valid. Rules of change authorize specific people or institutions to create, amend, or repeal primary rules; Article V of the U.S. Constitution, which requires a two-thirds vote in both chambers of Congress to propose an amendment and ratification by three-fourths of the states, is a rule of change operating at the highest level of the American legal system.1Constitution Annotated. Article V Rules of adjudication empower officials such as judges to determine authoritatively whether a primary rule has been broken and what consequences follow.
The framework explains how a legal system can be both stable and flexible. Primary rules give citizens the predictability they need to plan their lives. Secondary rules give the system the mechanisms to evolve, resolve disputes, and maintain its own integrity over time.
The Rule of Recognition
Among the secondary rules, the rule of recognition holds a special place. It is the foundational standard that identifies which other rules count as valid law within a given system. Unlike a statute you can look up in a code book, the rule of recognition is not written down in any single document. It exists in the shared practices of legal officials who operate the legal machinery day after day.2University of Pennsylvania Law Review. What Is the Rule of Recognition in the United States?
When a judge strikes down a local ordinance because it conflicts with the U.S. Constitution, that judge is showing the rule of recognition in action. The judge is applying a shared understanding that the Constitution sits at the top of the legal hierarchy and that any rule conflicting with it is invalid. Everyone within the system agrees on this hierarchy, even when they disagree fiercely about what the Constitution means in particular cases.
In the United States, the rule of recognition is not a single, simple formula. Legal scholars have described it as a hierarchical bundle of criteria: the Constitution overrides federal statutes, federal statutes override state law, and Supreme Court precedent determines how those documents are interpreted, unless overridden by constitutional amendment.2University of Pennsylvania Law Review. What Is the Rule of Recognition in the United States? The rule of recognition provides the ultimate test for legal validity. Without it, there would be no principled way to resolve disagreements about which rules are enforceable and which are not.
The Internal Point of View
Hart drew an important distinction between two ways of looking at a legal system. An outside observer who doesn’t accept the rules but watches how people behave has the external point of view. This person might notice that drivers stop at red lights and conclude they do so to avoid fines. From this angle, rules are just predictions about what happens if you step out of line.
The internal point of view belongs to someone who accepts the rules as standards for their own conduct and uses them to evaluate others. A person with this perspective doesn’t stop at the red light merely to dodge a fine. They stop because running it would be wrong, and they would criticize someone else for running it too. The rules function as genuine reasons for action, not just warnings about consequences.
This distinction matters enormously for the rule of recognition. Hart argued that for a legal system to function, officials must adopt the internal point of view toward the rule of recognition. Judges, legislators, and administrators must genuinely accept the system’s criteria of legal validity as a shared standard, not merely go through the motions out of fear or habit. Without that internal commitment, what looks like a legal system from outside lacks the normative glue holding it together.
Two Minimum Conditions for a Legal System
Hart distilled these ideas into two conditions any legal system must meet to be said to exist. First, the primary rules that are valid according to the rule of recognition must be generally obeyed by the population at large. Ordinary citizens don’t need the internal point of view; they can obey for any reason, including fear of punishment. Second, the officials who run the system must take the internal point of view toward the secondary rules, especially the rule of recognition. They must accept those rules as common standards for official behavior.
The asymmetry is one of Hart’s sharper insights. A legal system can survive widespread cynicism among ordinary citizens as long as they broadly comply. What it cannot survive is cynicism among its officials. If judges and administrators stop treating the rule of recognition as a legitimate standard and start treating their positions as nothing more than power, the system collapses from within.
Open Texture and Judicial Discretion
Legal language cannot anticipate every situation. Hart called this the “open texture” of law and illustrated it with a rule that has become a staple of law school classrooms: “no vehicles in the park.” The rule clearly applies to cars and trucks. Nobody disputes that. These sit in what Hart called the “core” of settled meaning.
But what about a bicycle? A motorized wheelchair? A remote-controlled toy car? A military memorial with a decommissioned tank mounted on a pedestal? These cases fall into the “penumbra,” a zone of uncertainty where the language of the rule runs out and no amount of staring at the word “vehicle” will produce an answer. Hart introduced this example during his 1958 exchange with Lon Fuller, partly to push back against legal realists who he thought exaggerated the amount of uncertainty in law. Most cases are easy. But the hard ones are genuinely hard, and pretending otherwise is dishonest.
In penumbral cases, Hart argued, judges must exercise discretion. They function as something like limited legislators, making a choice that the rule’s drafters never explicitly made. This discretion is not unlimited. It is constrained by the purpose behind the rule, by precedent, and by the broader framework of the legal system. But it is real discretion, not just mechanical application of pre-existing answers. Recognizing this keeps legal theory honest about what judges actually do when they hit a novel situation.
The Minimum Content of Natural Law
Despite his commitment to separating law from morality, Hart was not blind to the fact that every surviving legal system shares certain basic features. He identified five “truisms” about human nature and the physical world that any legal system must account for if the society it governs is going to survive.3Marquette Law Scholarly Commons. Law and Morality in HLA Harts Legal Philosophy
People are vulnerable and can be injured or killed, so any legal system needs rules restricting violence. People are approximately equal in strength, so even the strong can be overpowered by alliances, which makes peaceful compromise necessary. People have limited altruism; they care about others to some degree, but not reliably enough to sustain cooperation without rules backed by sanctions. Resources are limited, so property rules and some system for adjudicating competing claims are inevitable. And people have limited understanding and strength of will, meaning even those who accept the rules are tempted to break them, so enforcement is needed to protect the compliant majority from the minority who would exploit them.
Hart called this the “minimum content of natural law,” and it represents his most significant concession to the natural law tradition. He was willing to say that certain rules are so closely tied to human survival that any viable legal system will contain them. But he framed this as a factual observation about what human nature requires, not a moral claim about what law must be. The overlap between law and morality here is real but contingent, a product of the kind of creatures we happen to be.
The Nazi Informer and the Test of the Separation Thesis
The separation thesis faced its hardest test in the aftermath of World War II. In a famous case discussed during the 1958 Hart-Fuller debate, a German woman had reported her husband to Nazi authorities in 1944 for privately criticizing Hitler. She did so under statutes that criminalized remarks undermining public confidence in the regime’s leadership. The husband was sentenced to death, though the sentence was commuted to service on the Eastern Front. After the war, a German court prosecuted the wife for using the Nazi legal apparatus to destroy her husband.
The legal philosopher Gustav Radbruch, along with many postwar German courts, took the position that the Nazi statutes were so evil they never counted as law. Hart disagreed. He argued that the Nazi regime, however monstrous, operated something that “still deserved the name of law.” Denying that fact, in his view, let everyone off too easily. If the Nazi statutes were never really law, then the wife had not used the legal system as a weapon, and the moral horror of what happened gets papered over with a convenient fiction.
Hart proposed a different solution. The postwar government should have enacted a retroactive criminal statute to punish informers. That approach would have been honest about what it was doing, openly choosing justice over the normal rule against retroactive punishment, rather than pretending the old laws had never existed. The point was not to defend Nazi law but to insist that acknowledging evil law as law is the first step toward holding a legal system accountable for the evil it enables.
The Fuller and Dworkin Critiques
Two critics forced Hart to sharpen and eventually revise parts of his theory.
Fuller and the Inner Morality of Law
Lon Fuller, Hart’s sparring partner in the 1958 Harvard Law Review exchange, argued that law has an “inner morality” that positivism ignores. In his 1964 book The Morality of Law, Fuller listed eight principles that any system of rules must satisfy to count as law at all: rules must be general, publicly announced, prospective rather than retroactive, understandable, non-contradictory, possible to obey, relatively stable over time, and enforced consistently with how they are written.
Fuller’s challenge was pointed. If a regime issues secret laws, retroactive punishments, and contradictory commands, is it really operating a legal system? Fuller argued that the Nazis failed precisely these tests and that their regime amounted to a “perversion and violation of law” rather than a legal system with bad content. Where Hart saw morally evil law that was still law, Fuller saw a breakdown so severe that the label “law” no longer applied.
Hart responded that Fuller’s eight principles are requirements of efficacy, not morality. A poisoner needs to follow certain procedures to poison effectively, but nobody would call those procedures a “morality of poisoning.” A dictator who wants to control a population through law must make the laws clear, public, and prospective, but that is a matter of craft, not ethics.
Dworkin and the Problem of Principles
Ronald Dworkin launched his critique in 1967 with “The Model of Rules,” later expanded in Taking Rights Seriously. Dworkin argued that Hart’s framework, built entirely on rules, missed a category of legal standards that pervade actual judicial reasoning: principles. The distinction, as Dworkin drew it, is logical. A rule applies in an all-or-nothing fashion. Either a will must be signed by three witnesses or it need not be. If the rule is valid and the facts trigger it, the answer follows automatically.
Principles work differently. A principle like “no person should profit from their own wrongdoing” states a reason that pulls in one direction without dictating a result. In the classic case of Riggs v. Palmer, a court invoked that principle to prevent a murderer from inheriting under his victim’s will. But in countless other cases, people do profit from wrongdoing without legal consequence, such as when someone takes a better job after breaching a contract with a former employer. The principle carries weight without being decisive.
Dworkin argued that because principles are not identifiable by any simple pedigree test asking whether the legislature enacted them, Hart’s entire framework breaks down. If judges rely on principles that no institutional source enacted, then the rule of recognition cannot account for everything that counts as law.
Hart’s Postscript and Inclusive Positivism
Hart worked on a response to Dworkin for years but never published it in his lifetime. It appeared as a Postscript in the second edition of The Concept of Law, published in 1994, two years after Hart’s death.
Hart made two important moves. He conceded ground on principles, acknowledging that it was “a serious mistake on my part not to have stressed their non-conclusive force.” But he denied that his use of the word “rule” had ever been limited to all-or-nothing standards. Principles, in his view, could fit within the framework without destroying it.
More consequentially, Hart endorsed what is now called “inclusive” or “soft” legal positivism. He accepted that a society’s rule of recognition could incorporate moral criteria as part of its test for legal validity. A constitution that requires laws to respect “due process” or “equal protection” effectively builds moral evaluation into the rule of recognition. Hart argued that this possibility was always consistent with positivism, because whether a particular legal system includes moral tests is an empirical question about that system’s practices, not a necessary truth about law in general.4Yale Law School. The Hart-Dworkin Debate The concession split the positivist camp, with “exclusive” positivists like Joseph Raz arguing that moral criteria can never be part of the rule of recognition and that Hart had given away too much. That debate continues, and Hart’s framework remains the starting point for it.