Published by The Lawfare Institute
in Cooperation With
Public discussion of the 25th Amendment—Section 4 of which provides for transferring an incapacitated president’s powers—has been simmering throughout the Trump administration. After every odd or unnerving public statement by Trump (most recently: crudely altering a weather map with a Sharpie, tweeting a baffling explanation of his use of the word Liddle’, and raising the specter of a “civil war” if he is removed from office), one can find his adversaries on social media questioning his mental fitness. They fume that only cravenness or corruption can explain Vice President Pence and the Cabinet’s failure to invoke the 25th Amendment. “If the 25th Amendment doesn’t apply to President Trump,” they ask, “who would it apply to? What’s it even for, then?"
But Section 4 has worked just as it was intended. The failure so far to invoke Section 4 is a reflection of the 25th Amendment’s design, not a perversion of it.
The original Constitution provided for the vice president to take over whenever the president suffered an “inability to discharge [his] powers and duties.” But the Constitution incorporated no process for determining inability, and it designated no decision maker. It also left unclear whether an incapacitated president would be able to retake power if he recovered. The result was that every time a president was incapacitated, the vice president failed to step up. Among numerous such episodes, the most troubling involved Presidents Garfield (who died 79 days after being shot) and Wilson (who was felled by a stroke and for months was unable to fulfill his duties).
President Eisenhower, who suffered multiple health crises, found it intolerable that there might be no one at the helm were a nuclear war to break out. Eisenhower executed an agreement with then-Vice President Richard Nixon to transfer power if Eisenhower became incapacitated. Eisenhower also had his administration propose a more formal and permanent solution. After President Kennedy’s assassination, the issue gained additional urgency. Congress approved the 25th Amendment in 1965, and state ratification was completed in 1967.
Section 3 of the amendment, which has been used three times, allows the president to declare himself “unable to discharge the powers and duties of his office.” When this happens, he hands over power to the vice president. Upon declaring that he has recovered, the president takes power back—unilaterally and immediately.
Section 4 covers situations in which the president is similarly unable but cannot or will not invoke Section 3. When the vice president and a majority of the Cabinet declare the president unable, power immediately transfers to the vice president. But in this case, the president cannot take power back unilaterally. If and when the president declares that “no inability exists,” the vice president and the Cabinet have four days to respond. In the absence of such a counter-declaration, the president retakes power at the end of the four days. If the Cabinet and the vice president reassert that the president is unable to discharge his duties, however, the dispute goes to Congress to resolve. Unless two-thirds majorities in both the House and the Senate vote within 21 days that the president is unable, he retakes power. If the president loses the vote in Congress, the vice president remains in charge, but the president can try repeatedly to return.
The congressional debates about Section 4 made some things clear about its creators’ intentions. More importantly, these intentions are reflected in the structure of Section 4 itself.
Again and again, congressional debaters cited the twin specters of Garfield and Wilson. Garfield lapsed in and out of consciousness, with executive work going undone and Vice President Chester Arthur and the Cabinet refusing to act. While Wilson lay unable to fulfill his duties, Vice President Thomas Marshall was unwilling to take over. This was the paradigm legislators had in mind: the unmanned helm.
In this context, the goal of Section 4 was continuity. The concern was not unfitness, screwiness or irresponsibility (i.e., an oaf or a knave holding the reins), it was inability (i.e., no one holding the reins). Sen. Birch Bayh, the prime mover behind the amendment, explained in 1965 that “when it is possible actually to destroy civilization in a matter of minutes,” it is imperative to “make absolutely certain that there will be a President of the United States at all times.”
Bayh and others understood that such a functional vacancy could stem from other things besides an incapacitating physical or mental illness. The president could be captured, go missing or any one of a number of other possibilities. The touchstone, again, was the unmanned helm. This was distinct from inability in the sense of unfitness or unpopularity—though one might say colloquially that a president plagued by these conditions “can’t do the job,” Section 4 was not supposed to cover this kind of inability. Also excluded from Section 4’s coverage was misuse of power; if a president trampled on the law or turned out to be corrupt, that was a matter for the impeachment process, not Section 4.
The contrast with impeachment is key. Even though Section 4 requires that the president be “unable,” its framers realized that any mechanism for stripping power would be tempting to use against any objectionable chief executive. But the United States does not have a parliamentary system in which leaders can be deposed instantly by a no-confidence vote. Presidents are elected for four-year terms with a powerful presumption that they will finish their time in office. The impeachment process provides a narrow exception to that presumption, and legislators did not want Section 4 to be used as a way to evade the limits of that exception.
As such, Section 4 provides a weaker remedy for those looking to remove a president—Section 4 displaces the president only temporarily, while removal through the impeachment process is permanent. Moreover, transferring power away from the president using Section 4 is much harder to pull off. Impeachment and removal require a simple majority in the House and a two-thirds vote in the Senate. By contrast, displacing an unwilling president via Section 4 requires a two-thirds vote in both chambers, as well as initial action from the vice president and the Cabinet. In other words, if a president is behaving abusively, impeachment is the way to handle it. If there is insufficient support for impeaching and removing a president, there surely will not be sufficient support for using Section 4 as an alternative.
There are two categories of Section 4 cases. First are the easy cases—a coma, for instance—in which the vice president and the Cabinet will have no reason to hesitate, and power will transfer swiftly and smoothly.
Then there are hard cases: those in which the president is conscious, present and does not want to give up power. Here, Section 4 stacks the deck in the president’s favor. Not only are congressional supermajorities required to displace him, but he also has the power to preempt the matter in the first place by firing any Cabinet members he suspects of plotting against him.
The point is not to ensure that the president wins every one of these struggles. Rather, the point is to dissuade the vice president and the Cabinet from going down this path in all but the most serious cases.
So which cases rise to that level? Section 4 answers this question not by providing a detailed definition of “unable” but, instead, by designating the decision makers. In one typical passage in the debates, Rep. Richard Poff, one of Section 4’s key proponents in the House, justified the vice president’s participation precisely because he would be “a man of the same political party [and] a man originally chosen by the president.” As for the Cabinet, Poff touted the fact that its “members were appointed by the president” and that they would be “most loyal to him politically.” In other words, Section 4 requires that the president be incapacitated not just according to his adversaries (who, presumably, would already want him gone) but to the satisfaction of his senior leadership team as well. The reluctance of the vice president and the Cabinet to act is a feature of Section 4, not a defect.
While Section 4 could have required that a team of doctors examine the president and make the call, the amendment’s framers specifically rejected this approach. (Section 4 does empower Congress to replace the Cabinet by designating an alternative group through legislation, but Congress has never done so and the president would presumably veto any such attempt in the middle of an actual struggle.) To be sure, medical evidence may be relevant to assessing the president’s condition, and Section 4’s framers expected that the vice president, the Cabinet and Congress would consult with doctors as needed in making their decisions. But medical advice would play less of a role in Section 4 deliberations than one might think. It does not require a medical degree to conclude that a comatose president is “unable,” for example. By contrast, if the president believes himself to be sufficiently able but people disagree about his capabilities, the question of displacing him will be as much a matter of public policy as of medicine. The vice president and the Cabinet (and, in case of a dispute, Congress) were chosen as the politically legitimate engineers of such a policy decision.
The result of all this is that there are only two apparent categories of cases in which a Section 4 action would succeed even against a conscious, resistant president. The first is a president whose impairment is severe enough that the helm is, effectively, unmanned, even if he is still somehow able to claim that he is able to discharge his powers and duties. This could arise for a number of reasons, such as a severe stroke, a psychotic break or moderate dementia. But the bar is high; for the Cabinet and the vice president to invoke Section 4, only for the president to beat back the effort, could make matters worse rather than better. The vice president and the Cabinet would need to be confident of winning the requisite two-thirds majorities in Congress.
In a case like this, it might not matter that the bar is higher for Section 4 than for impeachment. Section 4 might be an easier sell for representatives of the president’s base: Politically, agreeing with the vice president and the Cabinet regarding the president’s incapacitation could be less fraught than agreeing with the opposition party that the president is a crook worthy of impeachment. But the reason it would be more palatable politically is rooted in the understanding that the vice president and the Cabinet would act only if the president was, in fact, thoroughly and observably incapacitated. Again, the bar would be high.
Then there are cases in which the president might be not quite so incapacitated but, nevertheless, impaired to the point of teeing up a disaster. Consider, for example, an unhinged president who orders a capricious nuclear strike against another country—the problem here is not that the president is “unable” so much as all too able to wipe out millions of lives. But this is dangerous territory. Perhaps watering down the “unable” standard would be morally justified to save so many lives, but that would hold only if the Section 4 effort actually worked and if the president did not return to power with a vengeance. And even if it were morally justified, it would present a potential slippery slope, opening the door for future actions against not-actually-impaired presidents posing less dire threats.
All of this is no doubt unsatisfying to those who want President Trump gone and who see Section 4 as an obvious way to make that happen. But that’s exactly the point: At present, Section 4 really is not a way to make that happen. Section 4 is designed principally to displace presidents who cannot fight back, but Trump can. When presidents can fight back, Section 4 is set up to protect them in all but the most extreme cases. Section 4 is what it is—a tool that is much narrower and more limited than it may look at first glance.