We are accustomed to treat rulings by the U.S. Supreme Court as authoritative guides in deciding questions of constitutional law. But what happens if a ruling depends on a plain misconception of what someone said? Why continue to rely on judicial error? This abstract question takes contemporary form with a decision by the U.S. Court of Appeals for the D.C. Circuit in May in Zivotofsky v. Clinton. The Supreme Court has yet to decide whether it will hear the case, but if it does, the justices would have an opportunity to acknowledge a judicial mistake.

The appeals court in Zivotofsky ruled that a 2002 congressional statute “impermissibly intrudes” on the president’s power to recognize foreign governments. The statute required the secretary of state to record “Israel” as the place of birth on the passport of a U.S. citizen born in Jerusalem if the parent or guardian so requested. The court acknowledged that neither the text of the Constitution nor the Framers’ intent “provides much help” in deciding the scope of the president’s recognition power. In weighing the president’s implied recognition power against Congress’ implied authority to set passport policy, the court decided in favor of executive power.

In doing so, the court admitted that it relied on dicta by Justice George Suther­land in United States v. Curtiss-Wright Corp., a ruling issued in 1936 that the Supreme Court has cited many times. Although the court has not actually held that the president exclusively possesses the power of recognition, the D.C. Circuit said it was appropriate for an inferior court to treat as authoritative “carefully considered language of the Supreme Court, even if technically dictum.” That conclusion is “especially” strong, the D.C. Circuit said, if the court frequently repeats the dictum.

Four times the D.C. Circuit sought guidance from Sutherland’s dicta. Was his analysis “carefully considered”? Here the record demonstrates clear judicial error. In advocating independent presidential power in foreign affairs, Sutherland relied on a speech given by John Marshall in 1800 when he served in the House of Representatives. Marshall referred to the president as the “sole organ of the nation in its external relations.” Sutherland took that language to mean a “plenary and exclusive power” of the president in international relations. Marshall never at any time advanced that argument, a point that is obvious once we understand why he gave the speech. In 1800, Thomas Jefferson ran for president against John Adams. Jefferson’s supporters in the House threatened to impeach or censure Adams for turning over to Great Britain someone charged with murder. Because the case was pending in an American court, critics accused Adams of encroaching on the judiciary and violating separation of powers. However, Secretary of State Timothy Pickering concluded that the person was a native Irishman, as did the district judge who was asked to turn the prisoner over to the British.

In his floor statement, Marshall explained that a section of the Jay Treaty contained an extradition provision, directing each country to deliver up to the other all persons charged with murder or forgery.

Adams was therefore not the “sole organ” in making foreign policy. He was the sole organ in implementing it. He did what the Constitution commands him to do: to take care that the laws be faithfully executed. At no time in this speech or in his lengthy career as chief justice did Marshall ever endorse independent, plenary and exclusive authority of the president in external affairs. In describing Marshall’s speech, Sutherland committed judicial error. Although clearly a misrepresentation, his dicta is routinely cited by the judiciary, the Justice Department and scholars to inflate presidential power.

A judicial error should not be treated as an authoritative source when deciding constitutional law. No matter how often the court repeats an error, it remains an error. It does not, by repetition, emerge as truth.

Perpetuating Sutherland’s error injures the reputation of the judiciary. Decisions are respected when grounded on a correct understanding, not on misleading secondary sources. As with other human institutions, the Supreme Court makes mistakes. Over its history, it has known how to correct them. In a dissent in 1932, Justice Louis Brandeis remarked: “The Court bows to the lessons of experience and the force of better reasoning, recognizing that the process of trial and error, so fruitful in the physical sciences, is appropriate also in the judicial function.” It may seem politically awkward and publicly uncomfortable for the Supreme Court to announce that what it said in 1936 was judicial error and has improperly guided courts for nearly eight decades. We know the judicial process is slow and does not easily correct itself. The erroneous dicta in Curtiss-Wright should not be treated as a legitimate precedent. Justices have admitted mistakes in the past. They need to do so again.

Louis Fisher is scholar in residence at the Constitution Project, visiting professor at College of William and Mary Marshall-Wythe School of Law, and author of “The Law of the Executive Branch: Presidential Power,” to be published by Oxford University Press in 2014.