Law and the Rule of Law

A regime passes a statute stripping a group of citizens of their property and their standing in court. The bill is introduced by a body constitutionally empowered to introduce it, debated according to the standing orders, passed by the required majority, signed by the head of state, published in the official gazette, and applied by judges who follow it accurately and impartially.

Was that a law?

Two answers have been given for something close to two thousand years, and almost everything else in the philosophy of law follows from which one is taken.

The thing in question

The definition is deliberately non-committal, and where it is non-committal is the point. It says the rules purport to bind, and does not say they succeed.

The first serious modern attempt to say what makes something law was John Austin's, in 1832. On his account a law is a command issued by a sovereign, backed by the threat of a sanction, where the sovereign is a person or body habitually obeyed by the population and habitually obeying no one else. The account has the great merit of being clear, testable, and free of anything mysterious.

H. L. A. Hart dismantled it in The Concept of Law in 1961, and the objections are worth having because they show what a legal system actually contains.

Not all laws are commands. A statute setting out how to make a valid will commands nobody to make one; it confers a power, and someone who ignores it has not disobeyed anything, only failed to make a will. Laws of this kind cannot be recast as commands without distorting them beyond recognition.

Laws bind those who make them. Austin's sovereign is by definition above the law, since it commands and is not commanded. Real legal systems routinely bind legislators, and the legislature that enacted a statute is subject to it.

Laws persist. When a legislature is dissolved, its statutes remain in force, and Austin's account has trouble explaining why the commands of a person no longer there are still law.

And most importantly, the account cannot distinguish law from the gunman. A robber who demands your money under threat issues a command backed by a sanction. Austin's theory says that a sovereign is a robber who is habitually obeyed and obeys nobody, which is not a difference in kind. Hart's diagnosis was that what Austin had left out was the internal point of view: participants in a legal system do not merely predict that they will be punished, they treat the rules as standards for criticising conduct, including their own. His replacement account made a legal system a union of rules imposing duties and rules about how those rules are made, changed, and identified, with the second kind being what a gunman lacks.

The old division

Both positions are regularly misstated and the misstatements are worth clearing away first.

Positivism is not the view that law and morality have nothing to do with each other. Positivists observe that legislators are usually trying to enact good law, that legal systems frequently incorporate moral standards by reference, and that moral criticism of law is entirely legitimate. What they deny is that a rule's moral quality is part of what makes it legally valid. Nor is positivism a doctrine that laws should be obeyed. The clearest positivists have been reformers, and the separation of what the law is from what it ought to be was defended precisely so that bad law could be identified as law and criticised.

Natural law theory is not the view that morality and law are identical, and no serious natural lawyer has ever held that an unjust rule has no effects. Aquinas gave the classic formulation in the Summa Theologiae, holding that a rule contrary to the human good is an act of violence rather than a law, quoting Augustine to the effect that what is not just does not seem to be a law, and concluding that such rules do not bind in conscience. The claim is about the obligation to obey and about what deserves the name, not about whether the courts will enforce it.

So the answer to the question at the top depends on which position is taken. On the positivist view the statute was law, and appallingly unjust law, and saying so is the clearest possible way to condemn it. On the natural law view something has gone wrong at the level of validity, and what the officials applied was not law but force wearing its clothes.

Hart and Lon Fuller conducted the argument in its modern form in the Harvard Law Review in 1958, in two papers published back to back. Hart's position was that describing wicked statutes as valid law is not an endorsement of them, that it keeps the two questions apart where they belong, and that a person facing a monstrous statute is helped rather than hindered by being able to say that it is law and ought not to be obeyed. Fuller's was that this concedes too much, and that a system of rules has to satisfy certain conditions before it counts as a legal system at all.

Governing by law rather than by will

A. V. Dicey gave the phrase its currency in English in his study of the law of the constitution in 1885, and the core of his account was that everyone, whatever their rank, is subject to the ordinary law administered by the ordinary courts.

The ideal is not the same as having laws. Every regime has laws. What the rule of law requires is that power be exercised through them rather than around them: that officials act on published rules rather than case-by-case decision, that the rules mean what they say, and that a person can find out in advance where they stand.

The reason this is a demand of political morality rather than mere administrative tidiness is that it is what makes planning a life possible. A person governed by known rules can arrange their affairs in reliance on them. A person governed by whatever an official decides on the day cannot arrange anything, and is dependent on the official's goodwill in a way that is a form of subjection whether or not the goodwill is forthcoming.

Constitutionalism goes beyond the rule of law and the two are often conflated. The rule of law asks that power be exercised through general rules; it is satisfied by a regime with unlimited powers that exercises them through clear published statutes. Constitutionalism asks that some powers not exist at all, and its characteristic devices are entrenchment, separation of powers, judicial review, and the placing of certain matters beyond the reach of ordinary legislation.

The obvious question is what makes it binding, since a constitution is itself a legal document and provides no independent force. The answers on offer include the practice of officials who treat it as binding, the difficulty of amendment, the interests of the parties who would lose from ignoring it, and the underlying moral claims that the entrenched provisions protect. None of them makes a constitution self-enforcing, and every constitution that has failed did so because enough people with power stopped treating it as binding.

The tension with democratic self-government is real and should not be smoothed over. Entrenching a provision means that a present majority cannot change it, which is either a protection of individuals against majorities or the dead hand of an earlier generation, depending on the case and on your view. Later tiers take this up directly.

What this tier has done

Six articles have laid the ground rather than settling anything on it.

The state's claim to command was distinguished from its capacity to compel, and the two came apart. The state itself was defined and its purposes examined against the reply that they could be served otherwise. Three demands that get made of political arrangements were separated and shown to conflict. The bases on which shares might be allocated were sorted, and the criterion ordinary thought reaches for first turned out to be the hardest to defend. The vocabulary of rights was analysed into components that ordinary usage runs together. And law was distinguished from organised force, with the two available accounts of how that distinction works set out.

Nothing here has told you what to think. What it has done is make it possible to state the positions in the intermediate tier without confusion, which is where the arguments actually get made: the social contract tradition and what consent could mean, the modern theories of justice and their critics, the traditions that organise political life, and the case for and against democratic rule.