Abolish unilateral presidential pardons

A Proposed Constitutional Amendment to Reform and Constrain the Presidential Pardon Power

The pardon power granted to the President of the United States under Article II, Section 2 of the Constitution has, since its inception, been understood as an act of mercy — a release valve for a legal system prone to the excesses of prosecutorial zeal, jury error, and changing moral consensus. Alexander Hamilton defended it in Federalist No. 74 as a necessary counterweight to the rigidity of law, arguing that inflexible justice is often injustice. For much of American history, this rationale held. Pardons were granted sparingly, reviewed through a formal Justice Department process, and understood as acts of genuine clemency rather than instruments of political self-preservation.

That understanding has eroded. The pardon power as currently construed is among the most unchecked authorities in the American constitutional system. The President may pardon any person for any federal offense, at any time, for any reason or no reason, with no requirement of review, no requirement of explanation, and no mechanism for override. Courts have largely declined to impose limits. Congress has no formal role. The pardoned individual need not have been charged, tried, or even identified by name at the time the pardon is granted. No other democracy with a comparable executive grants its head of government equivalent unilateral authority over criminal accountability.

The rot this creates is not hypothetical. Presidents have used the pardon power to shield close associates from accountability for crimes committed in service of the President’s own political agenda. They have dangled pardons as implicit rewards for silence, warping the incentives of potential cooperating witnesses in federal investigations. In the most extreme interpretation — one that has never been definitively rejected by the Supreme Court — a President might pardon himself, rendering the chief executive effectively immune from federal criminal law while in office and beyond. The Framers designed a system of checks and balances precisely to prevent this kind of concentrated immunity from consequence. The pardon power, as currently constructed, is a hole in that design large enough to swallow the whole.

This amendment does not propose abolishing clemency. Clemency is a legitimate and humane function of executive power. The purpose of this reform is narrower and more surgical: to ensure that the pardon power cannot be weaponized as a shield for the President, the President’s inner circle, or anyone whose prosecution might threaten the President’s political interests. Mercy toward the genuinely deserving is a virtue. Immunity for the politically connected is corruption wearing mercy’s clothing. This amendment separates the two.

The first operative provision of this amendment strips the President of the power to pardon any individual who has been investigated by, charged by, or convicted of crimes related to conduct undertaken in connection with a federal election, the obstruction of federal law enforcement, or the obstruction of a congressional investigation. These categories represent the precise circumstances under which the pardon power is most likely to function as a tool of self-protection rather than genuine clemency. A President whose associates engaged in electoral fraud, witness tampering, or contempt of Congress has an obvious personal interest in their silence and an obvious mechanism — the pardon — with which to purchase it. That conflict of interest must be structurally prohibited, not merely condemned.

The second operative provision prohibits the President from issuing any pardon — preemptive or otherwise — to themselves, to any member of their immediate family, or to any individual who served in a senior role in their administration at the time the underlying conduct occurred. Self-pardons are an affront to the most basic principle of the rule of law: that no person stands above accountability. The argument that a President can pardon themselves would have struck the Framers as a definition of tyranny. That this question remains legally unresolved in the twenty-first century is itself an indictment of the system’s failure to anticipate how aggressively the office could be exploited.

The third operative provision establishes a Pardon Review Panel — a bipartisan body composed of retired federal judges appointed jointly by the majority and minority leaders of both chambers of Congress — with authority to review and, by supermajority vote, nullify any presidential pardon issued within sixty days of the President leaving office or any pardon issued to an individual under active federal investigation at the time of issuance. This is not a veto over ordinary clemency decisions. It is a circuit breaker against the most transparently self-serving exercises of the power, concentrated at the moments of greatest risk: the final days of an administration and the point of active criminal exposure.

Some will object that any constraint on the pardon power undermines executive authority and creates dangerous uncertainty about the finality of presidential acts. This objection mistakes the nature of what is being proposed. Executive authority derives its legitimacy from the Constitution, and the Constitution is not an instrument of unlimited personal power — it is a framework for accountable governance. A pardon issued to purchase silence is not a legitimate exercise of executive authority; it is an abuse of it. Naming that abuse and prohibiting it does not weaken the executive — it restores the executive to its proper function.

Others will raise the concern that defining the prohibited categories creates new ambiguities and invites litigation. This concern has merit and deserves serious attention in the drafting process. The language of the prohibited categories must be precise enough to provide clear guidance while broad enough to capture the conduct it targets. This amendment does not purport to offer final legislative text; it offers a framework of principles that a drafting process conducted by constitutional scholars, former prosecutors, and defense attorneys should refine. The existence of ambiguity at the margins is not a reason to leave the core problem unaddressed.

A third objection holds that this amendment is a solution looking for a problem — that abusive pardons are rare and the existing political accountability mechanisms are sufficient. History refutes this. The Iran-Contra pardons issued by President George H.W. Bush effectively terminated an independent counsel investigation into potential criminal conduct implicating the President himself. The pardons issued by President Trump to associates convicted of crimes arising from investigations that touched directly on his own conduct — Roger Stone, Paul Manafort, Michael Flynn — followed a pattern that any neutral observer would recognize as the use of clemency to protect political allies and deter cooperation with prosecutors. These are not aberrations. They are consequences of a structural vulnerability that predictably produces the same abuses when the conditions are right.

The pardon power was designed for cases of genuine injustice: the wrongly convicted, the excessively sentenced, those whose crimes reflected desperation rather than predation, those whose cases illuminate the failures of a legal system that does not dispense justice equally across lines of race, class, and geography. This amendment does not touch any of those cases. The President retains full pardon authority for the vast majority of federal offenses and the vast majority of circumstances. What is removed is the subset of the power that has been shown, repeatedly, to function not as mercy but as obstruction.

Reform of the pardon power enjoys rare potential for bipartisan support precisely because the abuse of that power does not belong to a single party. Both parties have witnessed presidents of their own affiliation issue pardons that tested or exceeded the boundaries of legitimate clemency. Both parties’ members of Congress have, at various moments, introduced legislation seeking to constrain the power, without constitutional authority to do so effectively. A constitutional amendment is the only instrument with sufficient force to resolve what legislation and judicial interpretation have failed to settle.

The process of ratification will be appropriately demanding. An amendment requires two-thirds approval in both chambers of Congress and ratification by three-fourths of the states. This is a high bar, designed to ensure that only changes reflecting genuine national consensus alter the foundational law. The difficulty of the path is not a reason to avoid it — it is a reason to make the argument clearly and persuasively, to build the coalition of reformers across partisan lines, and to trust that a democratic people, given sufficient time and sufficient clarity about what is at stake, will choose accountability over impunity.

The argument for this amendment ultimately rests on a single, foundational claim: that no person who holds executive power should be able to use that power to insulate themselves or their associates from the legal consequences of abusing it. This is not a partisan claim. It is not a claim about any particular president or any particular party. It is a claim about what a constitutional republic is for. A republic in which the chief executive can commit crimes, surround themselves with people who commit crimes on their behalf, and then erase the legal consequences of those crimes with a stroke of a pen is not fully a republic. It is a republic with a monarchy concealed inside it, waiting for the right occupant of the office to activate it.

The design of the American constitutional system has always relied on what Madison called auxiliary precautions — structural mechanisms that check power without depending on the virtue of the individuals who hold it. Madison understood that parchment barriers alone were insufficient; that the system had to be designed so that ambition countered ambition, that power resisted power. The unilateral pardon power, in its current form, is a gap in that design. It is a place where ambition need not be countered because no structural counter exists. This amendment is an auxiliary precaution. It closes a gap that two and a half centuries of constitutional experience have revealed to be dangerous.

Opponents will argue that the Framers intended the pardon power to be broad and would have included limits if they wanted them. This argument proves too much. The Framers also did not anticipate political parties, standing armies, the administrative state, nuclear weapons, or a communications environment in which a president can address the entire nation instantaneously. Constitutional interpretation has never been a matter of frozen original intent applied mechanically to unanticipated circumstances. It has always been a living negotiation between foundational principles and the realities that those principles must govern. The foundational principle at stake here — that no one is above the law — is as old as the Constitution itself. This amendment applies it to circumstances the Framers did not foresee.

There is also a democratic legitimacy argument for this amendment that deserves emphasis. When a president pardons an associate to prevent cooperation with investigators, the victims of that corruption are not only the prosecutors whose cases are disrupted. They are the voters who cast their ballots without full knowledge of what was done in their name. Democratic self-governance depends on accountability, and accountability depends on the possibility of legal consequence. A pardon power that can be used to erase legal consequences for conduct that distorted the democratic process strikes at the very foundation of the system the President swears an oath to protect and defend.

The amendment proposed here is not a radical restructuring of American government. It does not transfer the pardon power to Congress, to the courts, or to a commission. It does not impose bureaucratic process on the ordinary exercise of clemency for ordinary cases. It removes a specific, demonstrably abused subset of the power from the unilateral control of the one person with the greatest personal interest in its misuse. It adds a limited review mechanism for the most obviously compromised exercises of the power. It draws a bright line — no self-pardons, no pardons for election crimes, no pardons for those who obstructed the investigations that threatened the President — and trusts the courts to enforce it.

The path to ratification will require sustained political will and a public that understands what is at stake. It will require members of Congress willing to advance a reform that reduces the power of the office they may one day occupy or hope to see their party occupy. It will require a willingness to look past the immediate partisan calculus — which president was most egregious, which party benefits from reform now — to the longer-term institutional interest in a presidency that cannot corrupt its own accountability. Those are difficult asks. But the American constitutional tradition has, at its best, shown itself capable of exactly this kind of institutional self-correction: the recognition, often belated but ultimately real, that concentrated and unchecked power is incompatible with the promise of self-governance.

The case for this amendment is, in the end, the case for the rule of law itself. Not the rule of law as a rhetorical flourish deployed in partisan argument, but the rule of law as a structural reality — a system in which consequences follow from conduct, in which power does not exempt its holders from accountability, and in which the instruments of justice cannot be hijacked by those they are designed to constrain. The presidential pardon power, reformed as this amendment proposes, would remain what Hamilton intended: an act of mercy available to a compassionate executive in the service of justice. What it would no longer be is what it has too often become: a get-out-of-jail-free card for the politically powerful, funded by the democratic legitimacy of an office the American people did not intend to place beyond the law.