Two judges appointed by President Donald Trump allowed him to deploy the National Guard into Portland, Oregon, on Monday, granting the executive branch sweeping authority to suppress peaceful protests based on false claims of widespread violence. The 2–1 decision from the 9^th U.S. Circuit Court of Appeals rested on a credulous reading of the record: The majority largely accepted the administration’s portrayal of Portland as a “war zone,” dismissing the district court’s extensive findings to the contrary. As dissenting Judge Susan Graber pointed out, Trump’s “illegal deployment of troops” rests on “false pretenses,” “fabrication,” and “propaganda”—all “political theater” that no court should dignify.

As bad as Monday’s decision was, it could have been far worse. In a concurrence, Judge Ryan Nelson argued that the judiciary has no power at all to review (or halt) a president’s deployment of the National Guard to suppress alleged domestic unrest. In Nelson’s view, Trump enjoys absolute discretion to send troops into American cities for any reason he deems necessary, and no court may stand in his way. It may be tempting to dismiss this argument as Nelson’s audition for a Supreme Court seat. But it must be taken seriously, because it is precisely what the Trump administration is now asking SCOTUS to embrace. And it is wrong from top to bottom: As a matter of constitutional principle, statutory text, precedent, and history, there is no remotely plausible argument that courts cannot review a president’s domestic deployment of the National Guard. Indeed, the weight of evidence points in the opposite direction.

Before delving into why Nelson’s claim is wrong, consider its radical and disturbing consequences. Trump could mobilize a state’s National Guard over its governor’s objections for blatantly illegal purposes, and courts would be powerless to stop him. He could send in the Guard to punish voters, protesters, and lawmakers who oppose his agenda—as he is doing in Portland and Chicago—responding to political disagreements with a militarized crackdown. He could flood Democratic cities with troops during an election to obstruct people going to the polls. He could dispatch soldiers to surround critical media outlets, sending the message that dissent will be met with force. The opportunities for repression are boundless, which is one reason why the Framers were so hostile to the idea of a standing army responsible for domestic security.

In light of these dire consequences, one might expect Nelson to mount a robust defense of his belief that Trump’s mobilization of the Guard “may not be reviewed by the federal courts.” But he devoted a sum total of fewer than six pages to this radical claim and spent most of them mischaracterizing Supreme Court precedent. The key case, to his mind, is 1827’s Martin v. Mott, a dispute over President James Madison’s authority to call up the states’ militias (the predecessors to our National Guard) during the War of 1812. In his unanimous opinion, Chief Justice Joseph Story held that Madison’s mobilization was lawful because the British invasion constituted a genuine emergency. The ruling included a sentence that the Trump administration has repeated ad nauseam: “We are all of opinion,” Story wrote, “that the authority to decide whether the exigency has arisen, belongs exclusively to the President, and that his decision is conclusive upon all other persons.”

Plucked from context, these words seem to imply that the president alone may determine when there is an emergency that requires state militias, and courts may not second-guess his decision. That’s precisely what Nelson insisted on Monday. But as professor Steve Vladeck has explained, the sentence means something entirely different in the context of the opinion. Mott addressed a debate about whether the president could make the initial determination that an emergency exists, or whether officers lower in the military’s chain of command could do so instead. The Supreme Court, professors Joshua Braver and John Dehn have shown, was concerned about preserving the military chain of command, specifically during a foreign invasion; it wanted to ensure that the commander in chief’s military subordinates could not defy his decision in a crisis.

To that end, Mott announced that the president, rather than officers of inferior rank, could make the first call to deploy the militia in a national crisis. The ruling did not say, or remotely imply, that courts lack authority to review the president’s decision, as Nelson wrongly asserted. That mistake has been definitively debunked by Braver, Dehn, and a comprehensive account of the case by the Constitutional Accountability Center.

There is another reason why neither Mott nor any other 19^th-century case that Nelson cites could possibly stand for the proposition that courts may not review the president’s declaration of an emergency: They involved a different statute than the one Trump now relies upon. Those cases interpreted the Militia Acts of 1795 and 1807, which broadly empowered the president to “call forth” the militia to counter an insurrection or rebellion “as he shall think proper.” These statutes gave immense deference to the president’s judgment and imposed few limitations. But Congress later stripped out that deferential language and replaced it with explicit restrictions on federalization of the National Guard. The modern statute allows the president to send in the troops only in the case of insurrection, rebellion, or an inability “with the regular forces to execute the laws of the United States.”

Nelson ignored this significant revision of the statutory text. He wrote that the current statute “uses similar language” to its predecessors, disregarding Congress’ deletion of a clause that gave more deference to the president. But as Judge Graber pointed out in dissent, today’s statute includes careful limitations on the president’s discretion. And those limits create a “judicial responsibility” for courts to rebuff the president when he deploys the Guard “in a situation far divorced” from what Congress envisioned.

There is, Graber noted, a deeper principle that counsels judicial review here. The Constitution does not give the president control over the militia; it allows Congress to decide when the militia may be called up. Congress has decided when deployment is appropriate and instructed the president to send in troops in those circumstances alone. Nelson’s theory would nullify Congress’ constitutional authority over the National Guard by letting the president summon it whenever he wants. Without judicial review, after all, who could stop him? Congress might try to deny funding for his partisan deployments, but Trump has already asserted the right to redirect money appropriated for other purposes. Combined with his ability to unlawfully withhold federal funds—which SCOTUS recently greenlit—Trump now has a license to take money from programs he dislikes and repurpose it for his own military objectives.

The Supreme Court is currently considering these questions in a closely related case out of Chicago. The administration has asked the justices to freeze an order by a cross-ideological panel of the 7^th U.S. Circuit Court of Appeals that blocked Trump’s deployment of troops in the Windy City. SCOTUS’ decision may tell us whether a majority of justices share Nelson’s fringe view that the judiciary has no power to enforce Congress’ restrictions on presidential exploitation of the Guard. If a majority embraces his position, it will effectively approve Trump’s use of military force to combat not a foreign enemy, but American citizens exercising their First Amendment freedoms. There are words to describe regimes like that, but democracy is not one of them.