Could Donald Trump Pardon Himself?

Image credit: Photo by Jon Tyson on Unsplash

Five years ago, with unfinished investigations and the threat of indictments swirling around him, Donald Trump claimed that he had the power as president to pardon himself. Now the former president faces numerous felony charges in two separate federal cases, with one—the classified documents case—having a good chance of ending in a conviction before the 2024 presidential election. Should he nonetheless be reelected, a possibility we must take seriously given President Biden’s sagging approval ratings and Trump’s continuing hold on Republican primary voters, Trump could again be in a position to claim he can pardon himself. 

Can he? 

At first glance, and according to some conservative contextualist scholars, the Constitution appears to support Trump’s claim. Under Article II, Section 2, Clause 1, the president is given the “Power to grant Reprieves and Pardons for Offenses against the United States, except in cases of Impeachment.” With the exception of his pardon authority reaching only federal crimes and not covering individuals being or having been impeached, the president’s prerogative here appears plenary. If there are any limits, it would appear to be ones tied to the president’s oath of office and his duty to “take Care that the Laws be faithfully executed”—a largely self-enforcing limit recent presidential pardon practices have regularly blown past. 

In short, since the text of the Constitution does not explicitly forbid self-pardons, a president arguably has the power. 

Constitutional textualists put a great deal of weight on the fact that there are only the two listed exceptions to the pardon power: not to be exercised in cases of impeachment, and only for federal crimes. There is much to be said for looking at the text initially for guidance. As the late Justice Antonin Scalia wrote, “The text is the law, and it is the text that must be observed.”

Yet, remarkably, when citing the text as evidence that presidential self-pardons are not precluded, the meaning of the key word in that text—“pardons”—is regularly glossed over. 

The Latin root of “pardon” is “donare”—to give, as in the word “donate.” One doesn’t donate to oneself: to give is to give something to another. The Constitution’s architects did not consider self-pardons for the simple reason that it would have made no sense to them. 

Even today, when you say “pardon me,” you are asking another person to give you something—for example, either to repeat something that you didn’t quite hear, or perhaps to be excused for a gaffe such as accidently bumping into someone on the sidewalk. 

It’s true that contemporary psychologists have introduced the idea of “forgiving oneself” into our everyday lexicon—but this understanding was utterly foreign to those who drafted the Constitution. To interpret “pardons” as containing self-pardons is, ironically, to give the text a “living Constitution” patina which most adherents to textualism would normally avoid like the plague.  

We believe that, to the degree that there was no definitive statement against self-pardons, it was because self-pardons, in their original understanding of the term “pardons,” were not something either the Constitution’s defenders or its critics thought possible.

The core of the argument that originalist, textual proponents of the power of presidential self-pardon make is tied at the end of the day to the “silence” in the text, and to a similar silence in the constitutional history precluding it.  

Admittedly, there is little history to work with here. Nevertheless, the little we have is worth noting. 

In the Constitutional Convention, the most substantive discussion of the pardon power occurred after the Committee of Detail added the pardon power to the list of authorities given to the president in early August of 1787. (“He shall have power to grant reprieves and pardons, but his pardon shall not be pleadable in bar of an impeachment.”) More than a month later, as the Convention members were running through the Constitution’s provisions one last time, Edmund Randolph, the Virginia governor, argued that the provision as then written (“The President  … shall have Power to grant Reprieves and Pardons for Offenses against the United States, except in Cases of impeachment”) was “too great a trust,” especially when it came to “cases of treason.” Randolph raised the problem of when “the President may himself be guilty” and  when “[t]he Traytors may be his own instruments.” In rebuttal, James Wilson—a delegate from Pennsylvania and a key architect of the presidency during the Convention—argued against stripping the president’s pardon power in the case of treason, on the grounds that a timely pardon in such instances might be a necessary measure of statecraft. Moreover, Wilson argued, if the president was “a party” to such treasonous plots, he could “be impeached and prosecuted.” 

There are two things to bear in mind here. The first is that neither Randolph nor anyone else in the Convention raised the problem of a president pardoning himself. If the worry here was a president possibly abusing the pardon power, and self-pardons were even a possibility, one would have expected it to have come up in the debate. But it didn’t. 

Second, consider that Wilson says a president can be both impeached and prosecuted. What would be the point of bringing up the deterrent of being prosecuted if a president could simply pardon himself? While a Senate impeachment conviction would remove a president from office and prevent him from running again, a president with the power to pardon himself would have eliminated the most serious punishments associated with crimes like treason—such as imprisonment or death.

It’s also noteworthy that not a single Antifederalist raised the issue of presidential self-pardons during the Constitution’s ratification process and debates. Since the Antifederalists were looking for arguments to help defeat the new constitution’s adoption, and many of them were already suspicious of what they took to be the quasi-monarchic features of the office, it would be quite an omission on their part not to debate the matter if they believed presidential self-pardons were an option. 

It’s possible, as some have argued, to read Wilson’s statement that a president could be “impeached and prosecuted” as referring not to the separate process of impeachment and criminal prosecution, but, instead, a president being impeached by the House of Representatives and being convicted by the Senate. The problem, with this reading of Wilson’s words is that it isn’t consonant with his argument in the Pennsylvania ratifying convention. There Wilson denies that “the first magistrate” retains “a single privilege or security that does not extend to every person throughout the United States.” “Far from being above the law,” Wilson says, the president “is amenable to them in his private character as a citizen, and in his public character by impeachment.” Put more directly by co-Federalist Tench Coxe—writing as “An American Citizen”—a president “may be proceeded against like any other man in the ordinary course of law.” (Emphasis in the original.) 

The standard rebuttal to the originalist, textualist interpretation of Article II, Section 2, typically reaches beyond the text to a broader principle of the Anglo-Saxon legal tradition. As the Justice Department’s Office of Legal Counsel put it in the days just before Richard Nixon’s resignation in August 1974, it is a “fundamental rule that no one may be judge in his own case,” and hence, “the President cannot pardon himself.”

The modern view, backed by a 1973 Justice Department legal opinion, is that a sitting president cannot be prosecuted and punished until he leaves office—or is removed from his office. To support this view, adherents point to Justice Joseph Story’s 1833 Commentaries on the Constitution, in which he says:

There are other incidental powers, belonging to the executive department, which are necessarily implied from the nature of the functions, which are confided to it. Among these, must necessarily be included the power to perform them, without any obstruction or impediment whatsoever. The President cannot, therefore, be liable to arrest, imprisonment, or detention, while he is in the discharge of the duties of his office; and for this purpose his person must be deemed, in civil cases at least, to possess an official inviolability.

Yet note the wording “in civil cases at least.” Story seems hesitant to extend that protection beyond a certain point. He is right of course to worry that a president’s official time and attention should not be drained away by sundry civil suits. However, he is far less certain it appears, about whether this prerogative of avoiding being dragged into court applies to a criminal indictment. 

More significantly, note what Story writes immediately following the above quote: “But [the president] has no authority to control other officers of the government, in relation to the duties imposed upon them by law, in cases not touching his political powers.” “Other officers” would certainly include prosecutors and judges, and echoing Wilson, include the possibility of a president being prosecuted for an act not tied to his constitutional functions.

Story’s interpretation is consistent with what Alexander Hamilton had to say on this topic in Federalist No.69: “The President of the United States would be liable to be impeached, tried, and, upon conviction of treason, bribery, or other high crimes or misdemeanors, removed from office; and would afterwards be liable to prosecution and punishment in the ordinary course of law,” suggesting a president would first be removed from office through impeachment proceedings and then prosecuted. Yet Hamilton neither here, nor anywhere else, precludes a president from being indicted while in office—even if, as Hamilton implies, the actual prosecution might follow an impeachment removal. Nor does Hamilton have anything to say about a president who has been indicted before being elected. He might have doubted that such a person could ever be elected in the first place, but it would be equally doubtful that he would think that the prosecution after a president leaves office could be avoided by a newly sworn-in president pardoning himself. 

There is also a deeper argument against self-pardons deducible from the structure of the Constitution as a whole and the president’s place within it.  

The pardon power is commonly thought to be an act of mercy necessary as a backstop against injustice. Granting pardons as a matter of justice has a long tradition in the United States along with developed institutional mechanisms and processes to advance this constitutional purpose. But mercy and justice are not the fundamental reasons for the pardon power. The fundamental reason for the pardon power is to equip the president and the government with the ability to secure the public good in times of an emergency or crisis. This is why it is vested in a president rather than a court. And it is why the most interesting, controversial, and consequential pardons have little to do with justice—and may even be unjust, such as pardoning leaders of rebellions or insurrection. While mercy may be a by-product of such a pardon, the purpose is not mercy for the guilty party but rather peace or tranquility for the polity as a whole. President Ford’s pardon of former president Nixon was justified this way.

The classic account of the pardon as an instrument of emergency power in service of the public good is John Locke’s chapter “Of Prerogative” in his Second Treatise on Government. There Locke shows that it is sometimes necessary to violate the law in service of the public good. The example he uses to illustrate this proposition is a person living in a neighborhood threatened by a wildfire who burns down his neighbor’s house in order to save his own and the rest of the neighborhood. Locke argues that an executive might rightly pardon the arsonist because the public good was at stake. Locke does not craft an example in which the executive breaks the law and pardons himself. He pardons someone else to illustrate the need for an executive designed to serve the public good—not to serve himself. It is not only that one should not be a judge in one’s own legal case, it is also that the exercise of this prerogative should be about the public good and not about the executive himself.Seeing the pardon as an instrument of emergency power also explains why legalists are so insistent that the pardon power is plenary. Emergency prerogative is vested in a president to solve the problem of providing power to address unforeseen contingencies. Such power is broad and seemingly unlimited because the unforeseen threats to the constitutional order are unlimited. However, it makes no sense for a president to deploy such power if the threat to the constitutional order comes from the president himself. In such a situation, “self-pardons” would extend the crisis rather than address it. In such a situation, issuing a self-pardon would contradict the purpose of the president’s emergency power (another reason no one in 1787 thought to explicitly state the obvious that self-pardons are impermissible).

Although we both think that a pardon of Trump would not be in the nation’s long-term public interest because of the negative precedent it would set with respect to the rule of law and the president’s duty in upholding it, it is also true that a case can be made that the public good could be served by pardoning Trump given the depth and intensity of political polarization in the United States.  

But that argument only makes sense if some president other than Trump pardoned him—whether his victorious opponent in 2024 or his own vice president if Trump were to step down from office. Someone else would be in a constitutional position to assess the public good—indeed, with a potential cost to his or her own political fortunes. A self-pardon is inconsistent with the gravity of the decision of “high politics” required in a case like this.

On the best reading of the Constitution, a self-pardon is never legitimate.

Gary J. Schmitt is Senior Fellow in the Social, Cultural & Constitutional Studies Program at the American Enterprise Institute.

Jeffrey K. Tulis is Professor of Government and Professor of Law at the University of Texas at Austin.