Power Under Law: The Constitutional Idea

The King Who Had to Write It Out by Hand

Lesson 8.1 left you with a question and an assignment: read the instrument that grants your authority, and mark every place it takes something away. If you did it, you noticed something uncomfortable. The document that makes you powerful is the same document that makes you restrainable, and there is no way to accept the first half without the second. That is the constitutional idea, and now we ask what happens when the person holding the power is also the person the law is trying to bind.

Start where almost nobody starts, in a legal code from the ancient Near East. Deuteronomy 17 anticipates that Israel will one day want a king like the surrounding nations. It does not forbid the request. It regulates it, and the regulations are startling given the neighborhood. He shall not multiply horses for himself, which is a limit on military accumulation. He shall not multiply wives, which is a limit on dynastic alliance-building. He shall not greatly increase silver and gold for himself, which is a limit on personal enrichment from the office. Every one of those was standard operating procedure for a Bronze Age monarch. Then comes the strangest provision of all.

'Now it shall come about when he sits on the throne of his kingdom, he shall write for himself a copy of this law on a scroll in the presence of the Levitical priests. It shall be with him and he shall read it all the days of his life, that he may learn to fear the LORD his God, by carefully observing all the words of this law and these statutes, that his heart may not be lifted up above his countrymen and that he may not turn aside from the commandment, to the right or the left.'

Sit with what that text is doing. The king does not receive the law; he copies it, physically, by hand, in front of witnesses who are not his employees. He keeps it with him. He reads it for the rest of his life. And the stated purpose is not administrative competence — it is 'that his heart may not be lifted up above his countrymen.' The Hebrew phrase behind 'a copy of this law' is mishneh hattorah, mishneh carrying the sense of a second or a duplicate. When the Septuagint translators rendered that phrase into Greek they produced the word that eventually named the whole book: Deuteronomy, the second law. The name of the book is a fossil of the king's homework.

Understand why this is not a devotional detail. In the surrounding cultures the king was frequently the source of law, sometimes a god or the god's earthly stand-in. Here he is a reader of a law that predates him, subject to it in the same way as the farmer in Judah, and the mechanism of his subjection is a discipline he performs on himself in public. That is a constitution in everything but vocabulary: a written, prior, public standard binding on the person with the most power to ignore it.

The King's Own Copy

Call the framework The King's Own Copy, and give it four marks. Written. Public. Prior. Binding on the writer.

Written matters because memory is a servant of desire. An unwritten understanding drifts toward whatever the powerful party currently prefers, and both parties will sincerely remember it differently. Public matters because a rule only some people can see is not a rule, it is a lever — and the person who controls access to the text controls the outcome regardless of what the text says. Prior matters because a standard adopted after the facts are known is not a standard, it is a verdict wearing a costume. And binding on the writer is the mark that separates constitutionalism from every sophisticated imitation of it. Anyone can write rules for other people. The whole innovation is a rule the rule-maker cannot escape.

Israel's own history is the test case, and it does not flatter the monarchy. You already worked the sharpest instance in Lesson 5.3, where Nathan walks into David's court and lets the king pronounce sentence on himself before closing the frame with four words. We are not running that scene again. Take from it only the structural fact this lesson needs: a prophet with no army told a king with an army that the king had broken a law the king did not make, and the king's answer was confession rather than execution. Whatever else you think about ancient Israel, that exchange presupposes an operating assumption that the throne sits under the statute.

Then read 1 Kings 21, which shows you the same system failing. Ahab wants Naboth's vineyard, adjacent to the palace, and offers a fair trade or cash. Naboth refuses: 'The LORD forbid me that I should give you the inheritance of my fathers.' That refusal is not stubbornness or sentiment. Ancestral land in Israel was held under an inalienability rule; Leviticus 25:23 states the principle — the land is not to be sold permanently, because the land is God's. Naboth is not driving a hard bargain. He is obeying a law that binds the king as much as it binds him, and he tells the king so to his face.

Watch what Ahab does, because it is the most instructive thing in the chapter. He goes home and lies on his bed facing the wall and will not eat — 'sullen and vexed.' He does not seize the vineyard. He sulks, which means the constraint is real to him even though he is king. It is Jezebel who says the quiet part: 'Do you now reign over Israel?' In her framework, the question answers itself. A sovereign who cannot take a vineyard is not sovereign. And then she solves the problem in the most chilling way available. She does not send soldiers. She proclaims a fast, seats Naboth at the head of the people, produces two witnesses to testify that he cursed God and the king, and has him executed on their testimony. Two witnesses is the evidentiary standard Deuteronomy itself requires. Every procedural step is correct. The result is judicial murder.

Rule of Law, Rule by Law

That is why the distinction between rule of law and rule by law is not academic hair-splitting, and if you take one transferable idea out of this lesson, take this one. Under rule by law, law is an instrument the ruler uses on other people; forms are observed, courts sit, statutes are cited, and the outcome is whatever the powerful party wanted before the proceeding began. Under rule of law, the ruler's will is itself subject to the standard. Both systems produce paperwork. Only one produces protection. Jezebel ran a lawful-looking proceeding to accomplish a murder, and every authoritarian government since has understood the technique perfectly.

Lon Fuller gave us the most useful tool for telling them apart. In The Morality of Law he argues that law has an internal morality — not a list of good policies, but eight conditions a system must meet before it can function as law at all. Rules must be general rather than aimed at individuals. They must be promulgated so the governed can know them. They must not be retroactive. They must be intelligible. They must not contradict each other. They must not demand the impossible. They must be stable enough to be relied on. And official action must actually correspond to the declared rule. Fuller's claim, argued against H.L.A. Hart in their 1958 exchange in the Harvard Law Review, is that a regime failing these badly enough has not made bad law; it has failed to make law.

Run Naboth's trial through Fuller's list. The charge was fabricated, so official action did not correspond to any declared rule. The proceeding was aimed at one man, so generality fails. The evidence was manufactured, so the congruence condition collapses entirely. The forms were immaculate and the internal morality was absent, which is precisely Fuller's point.

Joseph Raz sharpened the discussion in a way a serious leader should not skip, because it cuts against comfortable assumptions. Raz argued that the rule of law is a virtue of legal systems the way sharpness is a virtue of knives — it makes law effective at guiding behavior, and it is compatible with a good deal of injustice in the content of the rules. A regime can be procedurally impeccable and still be evil. That is an honest and unwelcome conclusion, and it means the rule of law is necessary rather than sufficient. Brian Tamanaha's survey of the concept's history documents how many incompatible things the phrase has been made to carry, and reading it will cure you of using the term as an applause line.

Here is the application to you, and it is uncomfortable on purpose. Every leader who has ever violated a rule while technically complying with it has run Jezebel's play at a smaller scale. The consultation held after the decision was made. The competitive process with a predetermined winner. The investigation assigned to someone who reports to the person being investigated. The consent obtained from people who could not safely refuse. If you have done any of those, you did not break a procedure. You used one. Name the instance before you read the next section, because a leader who cannot see this pattern in himself will not recognize it in a government.

Runnymede Is Not What You Were Told

Now the history, and we are going to correct a story you have heard wrong, because a candidate who repeats the romantic version will be corrected in public by someone who read the document.

In June 1215, English barons in armed rebellion forced King John to seal a charter at Runnymede. Most of its sixty-odd clauses concern feudal grievances of the baronial class — wardship, scutage, forest law, fish weirs on the Thames. It was not a declaration of universal rights. The famous clauses, numbered 39 and 40 in the 1215 text, promise that no free man shall be seized or imprisoned or dispossessed except by the lawful judgment of his equals or by the law of the land, and that right and justice will not be sold, denied, or delayed. Read the qualifier: liber homo, a free man. In thirteenth-century England the unfree were a very large share of the population, and the charter did not reach them.

Then the part almost nobody mentions. Magna Carta failed almost immediately. Pope Innocent III annulled it by papal bull on 24 August 1215, roughly ten weeks after the sealing, declaring it null and void; England fell back into civil war. What survived was not the 1215 document but its reissues — 1216, 1217, and the 1225 version, with Magna Carta finally taking statutory form in 1297. Only a handful of clauses remain on the statute book in England today. The charter's constitutional stature is substantially a later achievement, built by seventeenth-century lawyers reaching for a usable precedent against the Stuart kings, and then by American colonists reaching for it against Parliament.

So what does it actually establish? Something narrower than the myth and more durable than the myth deserves: a king conceded in writing that his will was subject to law, and the concession was written down where later generations could find it. That is the whole seed. Everything constitutional in the English tradition afterward — the courts' slow accumulation of precedent, the Petition of Right, the settlement of 1689 — grows in the space that concession opened. And it is worth saying plainly that England reached constitutional government without ever writing a single constitutional document, which should keep anyone from believing that the text alone does the work.

What England did instead was accumulate the thing in pieces, and the pieces are worth naming because they are the working parts of every constitutional order since. Sir Edward Coke, as chief justice and then as a parliamentarian, pressed the claim that the common law bound the Crown and drafted the Petition of Right in 1628 against imprisonment without cause and taxation without parliamentary consent. The Bill of Rights of 1689 settled the substance after a king had been removed: no suspending or dispensing with laws by royal authority without consent of Parliament, no levying money without a parliamentary grant, free elections, and freedom of speech in Parliament. The Act of Settlement in 1701 added the piece most people never think about and every serious executive should — judges would hold office during good behavior rather than at the Crown's pleasure, removable only on an address of both Houses. Independent tenure for the people who tell you no is not a decorative feature. It is the load-bearing wall.

Philadelphia in 1787 tried the opposite experiment: put the whole thing in one document, ratify it by convention rather than by legislature, and make the amendment process deliberately difficult. Madison's argument in Federalist 51, which is the subject of the next lesson, explains the machinery. What matters here is the design commitment. A written constitution is the political version of the king copying out the law by hand — a standard set down before the officeholder arrives, published where the governed can read it, and difficult enough to change that the person it restrains cannot quietly amend it in his own favor.

Your Own Copy: The New York City Charter

The New York City Charter is a mayor's own copy of the law, and the discipline Deuteronomy commands of a king is available to you in the most literal form: read it, keep it with you, know it better than the people who will quote it at you.

Start with what it grants, because a trustee who is timid about real authority is failing too. Charter section 3 makes the mayor the chief executive officer of the city. Section 8 grants the mayor the powers vested in the city 'except as otherwise provided by law,' and assigns him responsibility for the effectiveness and integrity of city government operations. Section 6 gives the appointment and removal power over commissioners and department heads — the mayor may remove any public officer holding office by appointment from a mayor whenever in his judgment the public interest requires it. Section 249 puts the executive budget in his hands. That is a strong executive by American municipal standards, and pretending otherwise is false modesty.

Now the limits, in the same document. Section 21 vests the legislative power of the city in the Council. Section 28 gives the Council power to adopt local laws for the government of the city, and the mayor cannot repeal a local law by any act of his own. Section 37 sets the veto: the mayor may disapprove a local law and return it with written objections, and two-thirds of all the council members can repass it over him. On the budget, section 254 lets the Council increase, decrease, add, or omit units of appropriation; section 255 lets the mayor disapprove increases within five days and lets the Council override by two-thirds. Section 29 gives the Council investigatory power over matters within its jurisdiction. Section 31 conditions certain appointments on the advice and consent of the Council after a public hearing. Every one of those is the Charter binding the person who administers it.

And there is a copy above your copy. New York City is a creature of the State. Its home rule powers come from Article IX of the New York Constitution and the Municipal Home Rule Law, the Legislature can preempt the field, and the Charter itself can be amended by local law or, for certain changes, only by the voters at referendum. A mayor who does not internalize this will spend four years frustrated that Albany will not stop existing.

Which brings us to the instrument every new executive overestimates. An executive order is real and it is limited. It directs agencies under the mayor's jurisdiction; section 8 lets him organize the executive office and delegate functions. It cannot repeal a local law, override state law, bind the Council or the courts, or reach independently elected officials and state authorities. It does not by itself create a legally enforceable rule of general applicability — that requires rulemaking under the City Administrative Procedure Act, Charter section 1043, with publication of the proposed rule in the City Record, a comment period and hearing, and codification in the Rules of the City of New York. And an executive order lasts exactly as long as the next mayor allows. Mayor Bloomberg's Executive Order 41, signed in September 2003, established the city's confidentiality policy on immigration status by executive action; when the Council wanted the city's detainer practice to survive an administration, it legislated, in Local Laws 58 and 59 of 2014. Learn the difference now. An executive order is how you move on Monday. A local law is how you are still standing in ten years. Lesson 8.3 asks the question underneath all of it: why should anyone design a system on the assumption that the person in the chair cannot be trusted?