Amnesty Transactional

“Jesus,” Saikrishna Bangalore Prakash notes in The Presidential Pardon, “is the most celebrated person not to receive a pardon.” 

Quite so. But second place must go, I think, to the Prophet Daniel, whose biblical story illustrates why a pardon power is needed. During the Babylonian Captivity, the story goes, Daniel became a valued adviser to the (possibly mythical) Persian King “Darius the Mede.” The Medes and Persians believed that an edict of their king, once written and proclaimed, could not be changed—not even by the king himself. Wicked advisers tricked Darius into decreeing a kingdom-wide 30-day ban on prayer—one that Daniel, an observant Israelite who had become Darius’s trusted adviser, had no choice but to violate. With a helpless shrug, Darius consigned the young man to the lions’ den, fully expecting him to be eaten. As it turned out, the lions miraculously refrained from eating him. As for the wicked counselors, Darius later fed them to the lions as consolation snacks. 

Prakash’s small book arrives at a fraught time for the Constitution in general and the pardon clause in particular. Those who find the issue perplexing would do well to start their research here.

It is safe to say that Trump’s abuse of the pardon power has no parallel in American history. Almost every president has granted a few that seem dodgy in retrospect; many have used them as an instrument of partisan politics; a few have used them as instruments of corruption. But in extent and scale, Trump’s pardons fall well below the subterranean ethical floor established even over the past 50 years. In pardoning 1,500 rioters convicted of involvement in the January 6 insurrection, Trump showed contempt for the law enforcement officers who protected the Capitol, and the system of government they preserved. His other pardons, from crypto fraudsters to foreign drug lords, reek with contempt for the very idea of law. Trump is also the first president to claim the power to undo a predecessor’s pardons, and the first to claim the power to pardon an offender convicted by a state, not the federal government. 

We have come a long way from the Medes and Persians: Article II, Section 2, Clause 1 of the U.S. Constitution grants the president “Power to grant Reprieves and Pardons for Offences against the United States, except in Cases of Impeachment.” The sweep of the power is breathtaking. Indeed, it is arguably wider than the royal pardon power that the Framers were imitating. British practice was familiar to the men who framed the clause. If they intended to include the British limitations, however, they stumbled: By the text of the Constitution, the power has been held to reside solely in the president (“he shall have Power”) and extend to all “Offences against the United States, except in Cases of Impeachment.” “The Supreme Court’s reading of the Constitution makes the president more powerful than the British Crown of the eighteenth century, for while Congress cannot regulate the presidential pardon power by statute, the British Parliament certainly could,” Prakash notes.

Shakespeare called mercy “an attribute to God himself.” The pardon power makes the president, in this area at least, something of a god. That is dangerous—but it is supposed to counteract the danger of a Constitution that does not take account of its own propensity to error, like the “law of the Medes and Persians.” But as with God’s mercy, the proper scope of the pardon clause cannot be entirely defined or fully explained. When does mercy become whim? When does it become a reward for or an instrument of crime? When is it just sleazy and disgusting? When is it godlike and when is it demonic?

The pardon clause has created puzzles and paradoxes literally since the dawn of the Republic, and presidents have been idiosyncratic in their approach. Barely a year into the first Washington administration, farmers in western Pennsylvania, incensed by a federal excise tax on whiskey, organized a rebellion and tarred and feathered two federal emissaries. President George Washington personally led the Army into the state, subdued the rebellion—and then pardoned its leaders even after a federal jury had convicted them of treason. No Darius the Mede he: It seemed wiser to the first president to conciliate the insurgent farmers than to insist on the letter of the law. Abraham Lincoln would pardon deserters, or even repentant traitors, with a few words (for example, “Let it be done”) scribbled on a letter begging for mercy.

These are grace notes. Prakash’s story truly begins with 1974, when Gerald Ford issued the infamous “full, free, and absolute pardon” to Richard Nixon for his crimes during Watergate. That act of clemency probably cost Ford an election in his own right, but has come to be viewed more charitably by some historians. “I am inclined to think the decision was brave,” Prakash writes. “What is clear is that, once again, many insisted that the clemency rested on ignoble motivations.” 

Jimmy Carter, in this as in so many other areas an outlier, promised an amnesty for Vietnam War–era draft evaders and kept that promise. But he refused to pardon individuals except on recommendation of the Department of Justice’s Office of the Pardon Attorney (a post created by Congress in 1894). George H. W. Bush pardoned the Iran-Contra conspirators; Bill Clinton pardoned his half-brother Roger, and a fugitive financier, Marc Rich, whose wife Denise “had made substantial donations to the expected William Clinton Presidential Library and Hillary Clinton’s campaign.” George W. Bush in 2007 used his clemency power to spare I. Lewis (“Scooter”) Libby, his vice president’s chief of staff, who was convicted of perjury and obstruction of justice for his role in thwarting an investigation into a leak of a CIA officer whose husband had been critical of the run-up to war in Iraq. 

Except in Libby’s case, Prakash notes, Bush was somewhat stingy with his mercy—as was Barack Obama:

Bush and Obama knew their reputations would not suffer from rejecting most pardon applicants. Yet the aversion declined over time, and supposed lessons were forgotten, leading President Obama’s successors to be more adventurous and aggressive. Exploiting the pardon pen for political or personal purposes was perhaps too tempting, bordering on irresistible.

The last two presidential transitions saw chief executives wield the pardon pen in various unseemly ways—pardoning family members involved in questionable financial doings, on the one hand; pardoning convicted insurgents who tried to destroy the republic on the other. 

Further, as his second term progresses, Donald Trump’s pardons are growing more flagrant and bizarre. On December 1, he pardoned former Honduran President Juan Orlando Hernández, a convicted drug kingpin who sent 400 tons of cocaine into the United States and once boasted that he was going to “stuff the drugs right up the noses of the gringos.” The next day, Trump posted on Truth Social that he was terminating all of Joe Biden’s pardons signed by an autopen. (One president can’t terminate a previous president’s pardon. And the reviled autopen was invented and first used by none other than President Thomas Jefferson, considered by some to have had some knowledge of the Constitution.) Never tiring of his hunt for new norms to violate, Trump on December 11 pardoned Tina Peters, a former Arizona official whose offense was allowing access to state voting equipment to “Stop the Steal” activists seeking to challenge the 2020 presidential election. 

The joker in this latest pardon: Peters was convicted in a state court of a state crime. For some time now—since, oh, I don’t know, roughly 1788—everyone who read the Constitution understood the phrase “against the United States” to give the president power only in federal cases. Peters’s lawyers are now demanding her release, on the utterly meretricious claim that “the President of the United States has the power to grant a pardon in any of the states of the United States.” Not content with that, Trump has taken his bloated distortion of the pardon power global: He threatened retribution against Brazil when it refused to drop all charges against its former President Jair Bolsonaro, and has also demanded that the government of Israel pardon his crony Benjamin Netanyahu. 

Thus, in America in 2025, the quality of mercy is, well, pretty doggone strained. Prakash is open-minded to a degree that will frustrate some readers who want definite answers. But he is clear in his distaste for the sordid hugger-mugger of contemporary pardon politics. In 2025, we are, he suggests, living in a “pardon dystopia,” a slovenly collapse into the lower depths of presidential ethics. If his answer to our dilemma is less concrete than one might wish (“my view is that we need fewer misguided or unjust pardons and more praiseworthy ones”), that may just flow from the difficulty of limiting a godlike attribute.

Prakash believes that the current morass is not the work of Trump alone. He questions Biden’s amnesty for all those convicted of marijuana possession and use and his commutation of sentences of most prisoners facing the federal death penalty:

Biden’s marijuana pardons and death penalty commutations open a brave (and frightening) new world of policy pardons, where the president is pardoning not because of a person’s particular circumstances or because of a rare need to pacify rebels, but merely because the president disapproves of the underlying policy that the law reflects. 

Well, perhaps. (Surely it is relevant that nearly half of the American states have ended the farce of marijuana prohibition?) Besides, is this really a “brave new world”? During his time in office, Thomas Jefferson was widely criticized for pardoning the four public figures convicted under the Adams administration’s grotesquely one-sided Sedition Act. These were defendants convicted of supporting Jefferson in the run-up to his 1800 election. They were, in fact, guilty of this “offense.” Jefferson argued that the act (which had expired at the end of Adams’s term) was unconstitutional as well as a bad idea. Surely this smacks of “policy pardoning”? Similarly, Andrew Johnson liberally pardoned rich southern slaveowners and Confederate public officials, in part because he hoped for their support in a presidential run—but also to forestall measures that would give political power to Black Americans. Pure politics, policy, or both?

In addition to his marijuana amnesty and death row commutations, Biden, on the night before he left office, also issued pardons to figures like General Mark Milley, who had incurred Trump’s wrath by helping thwart Trump’s suggestion that the military “just shoot [George Floyd protesters] … just shoot them in the legs or something.” Trump had suggested that Milley deserved punishment, perhaps “DEATH!” 

“Though contentious enough,” Prakash writes, “these were the less divisive of Biden’s midnight pardons.” Biden also pardoned his son Hunter and other members of his family. Prakash sees corruption in these. “For many years, Joseph Biden had been involved in a sordid business, where he was the product,” and the pardons protected family members who had profited from Biden’s office. 

This “both sidesism” is liable to enrage hard-core Biden loyalists, but I am not one and I never thought that Biden’s approach to the family business was defensible, nor his hobnobbing with millionaires. There is a counterargument, though. Trump had already promised to use criminal justice as a political tool to jail his critics, and had been no less explicit about his plans to jail Hunter Biden, and “the entire Biden crime family.” This was a lawless threat to misuse the law. Blocking a malicious prosecution of the Bidens, sleazy though their conduct may have been, is not the moral equivalent of interfering with the ordinary course of justice. It was, instead, impeding corrupt misuse of state power.

(And as for Hunter, that’s a special case. A father who would not pardon a child whom Trump had promised to destroy would, in my judgment, not deserve the title of “father” at all. Criticize Biden as you will, I think the Hunter pardon was among his best work.)

The heart of The Presidential Pardon is not so much a history of pardon as a meditation on its appropriate extent. Does the word pardon include other forms of clemency? Prakash supplies a useful taxonomy—full pardons (that wipe away convictions and any sentence imposed for them), commutations (orders that reduce or wipe away sentences but not the convictions), remissions (exemptions for fines owed to the government), amnesties (pardons offered to an entire group), preemptive pardons (which means the recipient cannot even be tried), blanket pardons (recipients cannot be convicted of any crime committed during a time period or activity), and conditional pardons (orders that exempt the recipient from punishment as long as he or she does certain things). The word offenses does not indicate that the president can only pardon crimes—civil fines and forfeitures may be forgiven as well, though not penalties awarded not to government itself but to individuals. 

Can a president pardon himself? The self-pardon question is a version of what philosophers call “the omnipotence paradox,” summarized in a George Carlin routine: “If God is all powerful, can He make a stone so big that He Himself can’t lift it?” Scholars, and the Justice Department’s Office of Legal Counsel, say presidential self-pardon is a paradox too far; Prakash is not so sure. 

The book might usefully have included the one limit on the pardon imposed after 1789, which also marks the only successful effort to limit presidential power in any respect. Beginning in 1865, Andrew Johnson issued a stream of sweeping pardons, apparently in hopes of restoring the slave-owning elite to power and packing Congress with former Confederates who would support him for election as a Democrat in 1868. The Framers of the Fourteenth Amendment, in Section 3, barred former U.S. officials who had turned coat and served the Confederacy from returning to federal or state office. They pointedly provided that no presidential pardon could lift that disability; only Congress, by a vote of two-thirds, could restore a traitor’s right to hold office.

(That notation, of course, leads us into the question of whether the Supreme Court usurped part of Section 3 in 2024 when it proclaimed that Colorado could not bar Trump from its primary ballot, even granting that he was what Colorado Secretary of State Jena Griswold called an “oath-breaking insurrectionist.” I would have enjoyed reading Prakash’s view on that.)

In the last part of the book, Prakash imagines ways to make the pardon power better. He suggests that the Supreme Court might at some point insert itself into adjudging the applicability and scope of the pardon pen. True, in this area as in others, the Court’s recent decisions seem favorable to presidential absolutism. “But the more the abuses pile up, the more tempting it will be for the Court to reconsider its precedents and pronouncements. Stay tuned.” Under present doctrine, Congress cannot legislate itself a role except by a constitutional amendment. Such amendments might also make some offenses unpardonable; might give Congress a legislative veto over pardons; might repose some or all of the power in other actors, including judges; or might bar any pardon of a family member. 

Prakash briefly mentions my own favorite nostrum. As long as we are imagining constitutional amendments, I think this one could actually pass: The pardon power should remain with the president—but he or she should be barred from issuing any pardon between Election Day and Inauguration Day. The really scandalous pardons tend to occur during the last days of a president’s term, when the voters can’t punish the president or the president’s party. Let incumbent presidents use the power as they see fit; but then let the president or party face the voters, as Gerald Ford had to do.