Everyone knows where to find the most powerful person in Washington. He lives and works in an old neoclassical mansion known as the White House. Every day, swarms of tourists, cranks, and clowns gather at the fence beyond his front door, tacitly acknowledging his preeminence.

Where can one find Washington’s second most powerful person? Is he at the main executive building? Is she in an office next to the Capitol? Perhaps he’s at the Supreme Court? Plausible suggestions all. But no one in these places can get much done without the support of other—sometimes many other—equally mighty and ambitious people.

There’s a case to be made that the second most powerful person in the federal government sits a block from the West Wing, in a drab concrete structure the sightseers invariably pass without a glance. The building houses the Consumer Financial Protection Bureau, the director of which implements more than a dozen major statutes—and answers to no one. Not to voters. Not to other bureaucrats. Not even to the chief executive. Although a CFPB director can be fired for misconduct, she cannot be removed due to her policies. The protocols she imposes, the priorities she sets, and the tactics she uses are in her hands to make. Her choices are her own. The president is stuck with them.

What are we to make of this? What do the traditions of our nation and our law have to say about this regulatory lord and her fiefdom?

*

“Will no one rid me of this turbulent priest?” Four knights immortalized this howl of protest, indignantly emitted by Henry II while his blood was up, by taking it seriously. They took their leave, rode to the Norman coast, and crossed the Channel. They found the Archbishop of Canterbury, Thomas Becket, in his cathedral, and they hacked him to pieces.

Becket had once been the king’s friend and counsellor. When Henry had him appointed archbishop in 1162, however, he promptly went native. The Church of that day was a power apart—a body with its own lands and privileges, its own laws and courts—and Becket became its champion. He denied the Crown’s authority, undermined royal policy at every turn, and excommunicated clerics loyal to the king. Henry could do little about any of it except shout and sputter. When at last, in 1170, the knights mistook one of Henry’s many impotent tirades as a command and removed Becket by cutting him down, the scandal shook the kingdom to the core. Becket was hailed a martyr and a saint. Henry spent years atoning for the sacrilege he had set on foot.

Just as the King of England could not always remove a man, neither could he always retain one. Edward II’s favorite, Piers Gaveston, gave the high and mighty of the realm nicknames such as “burst belly” and “the cuckold’s bird.” The barons took him to the woods and executed him in 1312.

Parliament impeached Charles I’s first minister, the Earl of Strafford, in 1640. When Strafford proceeded to defend himself a little too ably at his trial, the Puritans dropped the impeachment, attainted him, forced the king to sign a death warrant, and had him beheaded before 100,000 spectators. Charles never forgave himself. When he in his turn stepped onto a scaffold in 1649, he declared that God was punishing him for the “unjust sentence” that he had “suffered to take effect.”

But these dramatic episodes are aberrations. They are exceptional. For centuries England’s high officials—its chancellors, its judges, and, once Henry VIII had his way, even its bishops—served at the king’s pleasure. A minister’s powers were the king’s powers. They were, as Maitland explained, “royal prerogatives” that “the king might lawfully exercise himself were he capable of discharging personally the vast business of government.” An untrammeled power of appointment and removal was itself such a prerogative.

Some of these prerogatives were stripped away just as the American colonies were becoming a going concern. After the Revolution of 1688, the Crown lost finally and for all time the power to suspend a law. No sovereign has vetoed a bill from the throne since Queen Anne did so in 1708. When Anne died in 1714, the monarchy’s power to remove a judge “during good behavior” died with her. Royal authority withered further under George I and George II, Hanoverians who cared little about England or its affairs. Although George III took some interest in governing, by his day the king no longer attended cabinet meetings.

Yet the king remained formidable, at least in theory. He retained his say in foreign affairs. And his power to appoint and remove officers was untouched. The king, Blackstone wrote in 1765, was still “the fountain of honour, of office, and of privilege.” It was for him alone, therefore, to decide “in what capacities, with what privileges, and under what distinctions his people [we]re best qualified to serve and to act under him.” He could even create new offices, albeit only with what money parliament might supply.

The Declaration of Independence accuses George III of committing “every act which may define a Tyrant.” At times the document seems to rail against a despot whose writ ran no farther than Thomas Jefferson’s imagination. Accurate or no, however, Jefferson’s view of the king played on the minds of the men who assembled for the Constitutional Convention in 1787. The charter they crafted prises many royal prerogatives from the grasp of our chief executive. In line with the British practice by that time, he may not remove judges. But he also may not declare war or create offices; those powers belong to Congress. And he may not make treaties or appoint senior officers by himself; he needs the Senate’s approval.

But the British monarchy was not the Framers’ only point of reference. The Articles of Confederation had created a meagre executive power and assigned it to Congress. Forget George III; the absence of a separate executive was, Jefferson exclaimed, “the source of more evil than we have ever experienced from any other cause.” “Nothing is so embarrassing nor so mischievous in a great assembly as the details of execution,” he cried. Hamilton agreed. The lack of a “proper executive” led, he wrote, to a “want of method and energy.” “Responsibility” was too “diffused.” The Framers wanted to fix this problem.

So it is important to note how the Framers went about trimming the executive bough. First, they invested a chief executive with his panoply: they roundly vested “the executive Power” in a single “President.” Then, when they wanted to revoke some prerogative or other, they did so openly and in plain words. The president may not “provide and maintain a Navy”; he may not grant anyone a “Title of Nobility”; and so on. The Framers wanted to reduce the president’s authority to a point, and no further. “All the powers properly belonging to the executive department of the government are given,” as Fisher Ames put it to the First Congress, “and such only taken away as are expressly excepted.”

The Constitution requires the president to rely on Congress to erect and fund offices, and on the Senate to approve principal officers. Meanwhile, however, it commands the president, and the president alone, to “take Care that the Laws be faithfully executed.” And although it says that judges “shall hold their offices during good behavior,” it extends no like protection to executive officials. A balance has plainly been struck. The president answers to Congress, but the government answers to the president. An officer must, in Washington’s words, “assist the supreme Magistrate in discharging the duties of his trust.”

When Jefferson became president, he circulated among his heads of departments a letter setting Washington’s administration as his standard. Washington had required his officers to keep him “always in accurate possession of all facts and proceedings.” He had “formed a central point for the different [executive] branches, preserved a unity of object and action among them,” and “met himself the due responsibility for whatever was done.” Jefferson contrasted this approach with “Mr. Adams’s administration,” in which the president, during “his long and habitual absences,” let the government be “parceled out” among “four independent heads, drawing sometimes in opposite directions.” “That the former is preferable to the latter course,” declared Jefferson, “cannot be doubted.” Washington—and Jefferson—clearly believed that the president may guide, command, and, when necessary, remove government officials. Indeed, although no law granted the president a removal power, Washington, Adams, and Jefferson each dismissed many officers. Jefferson fired 124 of them.

The early presidents’ conduct was not challenged. To the contrary, the First Congress endorsed the notion that the president enjoys an unfettered removal power. When Madison moved to establish a Department of Foreign Affairs, “the head of which” was to be an officer “removable by the President,” a debate erupted about the nature of the removal power. A few representatives argued that removal required an impeachment trial in the Senate. Others argued that the Senate’s approval, at least, was necessary. Still others believed that, although the president should be allowed to remove people at will, his power to do so came not from the Constitution but from Congress. Madison, for his part, contended that “the lowest officers, the middle grade, and the highest” all “depend, as they ought, on the president.” And because he in turn depends on the “community,” the “chain of dependence” terminates in “the people.” An unqualified removal power ensures, in other words, that voters may hold the president to account for his officers’ actions.

In what is now known as the Decision of 1789, Congress passed several bills that contained no removal clause, but that discussed who would manage the papers of a removed officer. The traditional view holds that Congress thereby affirmed that the Constitution empowers the president to remove officers at will. Legislators on both sides of the debate placed that gloss on the affair in their private letters. The votes turned, one senator wrote, on “whether the President had a constitutional right to remove; not on the expediency of it.” Madison told Jefferson that his colleagues had taken the position “most consonant” to “the text of the Constitution” and “the requisite responsibility and harmony in the Executive Department.”

It was not until the twentieth century that the Supreme Court turned its gaze squarely on removal. When at last it did so, it threw its weight behind what was by then the custom and understanding of more than 130 years. Although the president holds all “executive power,” the court said in Myers v. United States (1926), he “alone and unaided could not execute the laws.” He must “execute them by the assistance of subordinates,” and, to do so effectively, he must be able to remove “those for whom he cannot continue to be responsible.”

*

This has been a story about the American government as it was designed and built. The government described in grade-school textbooks. The government in which one branch makes the law, another administers it, and a third applies it to cases and controversies. The plan never worked perfectly. That was never a possibility. The lines between the powers are at times too obscure, the humans tasked with finding them often too fallible and corrupt. But the model endured. Its imprint was real.

That model is gone. Not long after Myers, the Supreme Court declared, in Humphrey’s Executor v. United States (1935), that some officers can be subject only to for-cause removal. The court permitted the creation of a new, fourth branch of government. The fourth branch exercises not only executive powers, but also “quasi-judicial” and “quasi-legislative” ones. This is the branch of the commissions and boards—the branch of the Federal Trade Commission, the Securities and Exchange Commission, and the National Labor Relations Board, to name but a few. These independent agencies have their own rules, their own judges, and their own domains. They are, like Becket’s Church, a power apart.

The court has approved, in particular, of agencies governed by panels. Spreading control among a number of board members or commissioners, we are told, makes the novel bodies’ autonomy and clout more tolerable. We are to rest easy because, in a committee of five, only a confederacy of at least three may ignore, foil, and defy the president and the public.

In recent years, however, in what seems almost like a deliberate effort to wreak constitutional havoc, Congress has taken to creating departments governed by individual directors. The most powerful of these is the head of the Consumer Financial Protection Bureau. She is tasked with administering a raft of consumer-protection laws. She decides what rules her agency will issue, against whom they will be enforced, and what the penalties for breaking them will be. She draws her budget—more than $600 million—from another independent body, which shields her appropriations from the threat of presidential veto. She serves a five-year term, which means she often will proceed under a president who did not even choose her. And still, she can be removed only for “inefficiency, neglect of duty, or malfeasance in office.”

Earlier this month the Supreme Court agreed to decide, in Seila Law LLC v. Consumer Financial Protection Bureau, whether the CFPB’s structure is constitutional. The case will be contentious. Many issues will be hotly debated. The question of when and how the court may overturn its own rulings will likely attract much attention. The supposed virtues of politically insulated experts, and the palpable vices of the current chief executive, will surely worm their way into the discussion. There can be no doubt, however, about what the constitutional text and history teach us. The president, they say, can remove officers at will.

Also published by Forbes.com on WLF’s contributor page.