ANALYSISApril 13, 2026
Should the presidential pardon power be limited?
The debate over limiting the U.S. presidential pardon power has intensified sharply in 2024–2025 following a series of high-profile and controversial clemency actions by both President Biden and President Trump. Biden issued a broad pardon to his son Hunter Biden in December 2024, pre-emptive pardons to public figures including Anthony Fauci and Mark Milley in January 2025, while Trump issued mass pardons to approximately 1,600 individuals involved in the January 6 Capitol attack on his first day back in office. Legislative proposals including a constitutional amendment (H.J. Res. 13) have been introduced in Congress to restrict the scope of the pardon power.
The Founders gave the president near-absolute power to forgive federal crimes — no exceptions written in, no congressional check required. After a string of pardons that critics call self-dealing and corruption shields, the real question is: can a democracy survive a mercy power with no limits, or does any limit on clemency open a door to political persecution?
Not familiar with this story? Get context →
Hamilton's design still fits today
C
Hamilton's argument in Federalist No. 74 was not naïve — it was a structural claim that clemency requires executive flexibility precisely when legislative passions run highest. The Founders deliberately placed this power in a single accountable executive because they understood that mobs and majorities are sometimes the threat, not the remedy. That architecture has not become obsolete.
L
Hamilton's design assumed informal constraints — shame, precedent, political cost — would do most of the work. What January 2025 demonstrated is that those constraints have collapsed for presidents who are departing or newly arriving: Biden faced zero electoral consequence for breaking his public pledge, and Trump faced none for immunizing political soldiers. When the informal guardrails fail, the constitutional text is all that remains — and right now that text permits everything we just watched.
C
Ford's pardon of Nixon was unpopular enough to cost him the 1976 election — that is Hamilton's accountability mechanism actually working. The argument that informal constraints have 'collapsed entirely' is doing a lot of work on the basis of two consecutive presidents who both faced unusual end-of-term circumstances, not a permanent structural failure.
L
Ford lost an election — Biden was already term-limited and had withdrawn from his race before the pardon. That is precisely the structural gap: the accountability mechanism you're citing simply did not exist for the president who most clearly violated his public pledge.
Hunter Biden pardon's structural failure
C
Biden's pardon of Hunter Biden was a serious breach of political integrity — he explicitly promised not to do it, then did it anyway, covering a decade of potential federal crimes with no stated rationale of innocence or systemic injustice. That is nepotism with constitutional cover, and it deserves to be named as such.
L
We agree on the facts, but notice what you just conceded: this is the strongest case for reform, and your own framing — 'nepotism with constitutional cover' — describes exactly the abuse channel H.J. Res. 13 targets. If the diagnosis is correct, the burden falls on critics to explain why the remedy is worse than leaving that channel open.
C
The remedy has to fit the diagnosis. The core failure with Biden was not kinship — it was the absence of political cost for a departing executive. A prohibition on pardons issued in a president's final months would target that specific gap without handing courts a veto over clemency throughout an entire administration.
L
A final-months prohibition is an interesting structural idea, but it would not have stopped Trump's day-one mass pardons — which means you need two separate reforms to close two demonstrated abuses, at which point you are building H.J. Res. 13 from scratch anyway.
January 6 pardons as political endorsement
C
Trump's January 6 mass clemency includes individuals convicted of assaulting police officers — that is not mercy toward the unjustly prosecuted, it is political reward for political soldiers. The conservative case for prosecutorial overreach applies to a handful of nonviolent trespassers. It does not survive Oath Keepers leaders convicted of seditious conspiracy by juries of their peers.
L
The question for any pardon is whether the executive believes the prosecution served justice — and reasonable people can dispute that across 1,600 highly varied cases. What conservatives are right to note is that embedding a 'corrupt purpose' standard in constitutional text would have invited courts to second-guess Ford on Nixon, Clinton on Marc Rich, and Obama on Chelsea Manning. The cure reaches well past the disease.
C
Seditious conspiracy convictions reviewed by bipartisan federal judges are not the same category as Nixon's pre-indictment pardon or Manning's commutation. You are using the hard cases to protect the easy ones, and the easy ones here include people convicted of beating Capitol Police officers.
L
Agreed those cases are indefensible — but the reform mechanism has to work for all future pardons, not just the ones we find indefensible today. That is exactly why the drafting of 'corrupt purposes' has to be precise enough to be judicially manageable, which it currently is not.
'Corrupt purposes' standard is unworkable
C
Embedding 'corrupt purpose' in constitutional text would expose every contested pardon to federal court review — Ford on Nixon, Clinton on Marc Rich, Trump on Flynn, all litigated in real time. Ex parte Garland ruled the pardon power beyond legislative reach for a reason: courts are not equipped to adjudicate the political judgments clemency requires, and inviting them to do so creates a chilling effect on legitimate mercy.
L
The 'corrupt purposes' drafting problem is serious — we concede that. But the appropriate response is to demand precision from reformers, not to abandon the project. A bright-line prohibition on pardoning immediate family members or co-defendants in proceedings against the pardoning president is judicially manageable without requiring courts to evaluate executive motive in every clemency case.
C
A bright-line family prohibition would have stopped the Hunter Biden pardon — but it would not have touched January 6, which is the larger and more dangerous precedent. You are proposing to constitutionalize a rule that solves roughly half the problem you identified.
L
Then write two bright-line rules instead of one vague standard — family pardons and co-conspirator pardons are both judicially manageable without asking courts to read presidential intent. The drafting difficulty is real but it is an argument for better drafting, not for leaving both channels open.
Amendment threshold as reform safeguard
C
The 38-state ratification bar is a feature, not a bug. It ensures we do not carve up the Constitution because two consecutive presidents behaved badly in ways we found intolerable. The pattern of crisis-without-amendment after Nixon, after Marc Rich, after Flynn is not evidence of institutional failure — it is evidence that no genuine national consensus for structural change actually formed.
L
The Marc Rich pardon in 2001 and the Nixon pardon in 1974 each generated serious reform proposals that failed — and then we watched the same abuses recur. At some point, 'consensus hasn't formed' stops being evidence that the system is working and starts being evidence that the collective action problem is real. Two presidents of opposite parties abused the same power in the same month; if that doesn't approach the threshold for consensus, what would?
C
Both presidents acted in unusual end-of-term or start-of-term windows, not in the middle of normal governance — that suggests the problem is narrower and more specific than 'the pardon power is broken,' which is what a constitutional amendment implicitly declares.
L
If the problem is end-of-term and start-of-term abuses, that is still a structural problem with the power as written — and 'narrow and specific' abuses that both parties commit in the same month are exactly what amendments are for, not a reason to keep waiting for a crisis that somehow generates cleaner facts.
Conservative's hardest question
The most serious challenge to this argument is Biden's pardon of Hunter Biden — not because it is legally indefensible, but because it directly contradicted his own public pledge and involved his immediate family member, making the case for a familial-pardon prohibition feel less like partisan opportunism and more like a principled response to a concrete, named abuse. If even the reform's critics must concede the Hunter Biden pardon was a breach of integrity, the argument that electoral accountability is sufficient looks harder to sustain when the offending president was already term-limited and faced no electoral consequence at all.
Liberal's hardest question
The proposed prohibition on pardons issued for 'corrupt purposes' is genuinely difficult to operationalize in constitutional text — any standard vague enough to cover the cases in question may be specific enough to invite courts into every clemency decision, which Ex parte Garland correctly identified as constitutionally problematic. Reformers owe a precise answer to this drafting problem, and none has been offered yet.
Both sides agree: Both sides concede that the pardon power has been visibly abused by both Biden and Trump within the same calendar month, making purely partisan defenses of either president's actions intellectually untenable.
The real conflict: They disagree on a factual-structural question: whether electoral accountability has historically functioned as an adequate check on pardon abuse, with conservatives citing Ford's 1976 defeat as evidence it works and liberals citing Biden's term-limited, consequence-free pardon of his son as evidence it has now structurally failed.
What nobody has answered: If the informal political constraints on pardon abuse — shame, electoral cost, public norm — have genuinely collapsed for both parties simultaneously, what evidence exists that those same collapsed norms would produce the supermajority consensus required to ratify any constitutional amendment, and if that consensus is impossible, what does that imply about the republic's actual capacity for self-correction?
Sources
- U.S. Constitution, Article II, Section 2 — presidential pardon clause
- Federalist No. 74 (Alexander Hamilton, 1788)
- Ex parte Garland, 71 U.S. 333 (1866)
- DOJ Office of Legal Counsel opinion on self-pardons (1974)
- H.J. Res. 13, 119th Congress — proposed constitutional amendment limiting pardon power
- Statement of Rep. Steve Cohen (D-TN), January 9, 2025
- Washington Post investigation on Trump's use of pardon power
- Georgetown law professor Steve Vladeck on pre-emptive pardons
- University of Virginia law professor Saikrishna Prakash, book on evolution of the pardon power
- Public opinion polling data on pardon power limitations (cited in search results)