When David French, a conservative constitutional lawyer turned New York Times columnist, calls something the most monarchical act of an already monarchical presidency, it is worth understanding the mechanics. The act in question — Trump’s $1.776 billion “Anti-Weaponization Fund,” announced by the Department of Justice on May 19, 2026 — looks from the outside like a straightforward corruption story: Jan. 6 rioters might get paid, taxpayers foot the bill, and the dollar amount (1776, get it?) is a deliberate wink at the rioters’ rallying cry. All true. But the deeper problem, the one French and his co-hosts Jamelle Bouie and Michelle Cottle worked through on “The Opinions” two days later, is structural. Trump did not simply write a check to his allies. He sued his own IRS, engineered a collapse of the case before a judge could question whether it was real, and then extracted from a settlement between two arms of his own executive branch a fund with no judicial oversight, a commission appointed at his sole discretion, and a release of civil liability for himself and his family so broad that it functions as something the courts cannot actually issue: a civil pardon.
How a president sues himself
The constitutional structure of the fund is not an afterthought — it is the point. Understanding how Trump v. IRS was constructed, and why it was dropped, explains what came out of the wreckage.
Trump filed suit against his own IRS — an agency he controls — over the leak of his tax returns, an incident that allegedly occurred during his first term, when he also controlled the IRS. The defendant would ordinarily be the Department of Justice, also under his control. French’s observation: the moment a federal judge looked at the case and asked whether a real adversarial proceeding existed — one side suing itself — the architecture would collapse. Trump withdrew before that question was put formally.
What emerged from the withdrawal was not nothing. The administration, having dropped the case, entered into a unilateral agreement with itself: $1.776 billion drawn from the DOJ’s existing Judgment Fund, appropriated by Congress to settle legitimate claims against the government. A five-member commission appointed at Trump’s discretion, with no external oversight. Commission members may draw their expenses — hotel costs, per diems — from the fund they administer.
Attached was what French called extraordinary: a sweeping release of civil liability for Trump, his family members, the Trump Organization, and affiliated trusts, covering federal tax matters arising before the settlement date. The presidential pardon power is real and broad, but it reaches only criminal liability. A president who violated the law while in office remains civilly exposed after leaving — that is the gap that has now been papered over, by an agreement the administration made with itself. Former DOJ prosecutors, asked by journalists why the fund totals exactly $1.776 billion, said the figure is a deliberate signal: it echoes the Proud Boys’ planning documents titled “1776 Returns,” the chants rioters invoked as they breached the Capitol, and the general MAGA shorthand for revolutionary legitimacy. The White House did not deny it.
Who collects — and who cannot
Enrique Tarrio plans to apply for $2 to $5 million. The asymmetry between what the fund offers him and what civil law offers his victims is French’s sharpest point.
Tarrio — the former Proud Boys chairman convicted of seditious conspiracy and sentenced to 22 years before receiving a Trump pardon — told Reuters he expects to seek between $2 million and $5 million from the fund. He was not physically in Washington on January 6; a judge had banned him from the city after a prior conviction. He orchestrated the attack from a Baltimore hotel room, according to the jury that convicted him. Speaking separately to PBS News, he revised the estimate upward: “somewhere in the mid-tens of millions.” Michael Caputo, a Trump ally, filed the first known formal claim: $2.7 million.
Suppose there’s MAGA Mike. He took his flagpole and used it to beat a police officer, spent time in prison for assault, now he’s pardoned. He could file some sort of weaponization claim — maybe claiming he was poorly treated in prison, or something happened at trial. MAGA Mike, who beat a police officer with a flagpole, could get half a million dollars. Then you have Blue Bob. Blue Bob is protesting ICE in Minneapolis, and an ICE officer smacks him in the face or tasers him for no reason. Is Bob going to be able to apply to that fund? He can apply, but good luck. And if Bob tries to get compensation through a normal legal channel for a civil rights violation by a federal officer, there is just a massive web of immunities that really wall off federal officials from accountability.
French’s asymmetry has a precise legal shape. Qualified immunity, sovereign immunity, and the narrowing of Bivens claims (constitutional suits against federal officers) have been progressively constricted by the Supreme Court over two decades. A protester assaulted by an ICE agent faces multiple procedural barriers before reaching a merits hearing. A Jan. 6 defendant needs only to articulate some theory of “weaponization” and submit to a commission whose criteria Vice President Vance described, on the day the fund was announced, as open to “anybody.”
This is one side suing itself. And what comes out of it is a fund run entirely at his own discretion, with a release of liability that grants him, his family, and all parties to the lawsuit a version of a civil pardon — a thing the courts cannot issue.
— David French, The Opinions
Two Capitol Police officers moved faster than any other potential challengers. Harry Dunn, a former Capitol officer, and Daniel Hodges, an active Metropolitan Police officer — both injured defending the building on January 6 — filed suit in federal district court in Washington on May 20, arguing the fund lacks statutory authorization, the settlement is a corrupt sham, and the structure violates the Constitution and federal law.
The *standing* problem
Every constitutional violation needs someone with legal standing to challenge it in court. The Anti-Weaponization Fund has been structured to make that hard.
Standing doctrine requires plaintiffs to show a concrete, particularized injury from the challenged action. As a general rule, a taxpayer cannot sue the government simply because tax dollars are being spent illegally — that diffuse injury does not satisfy the requirement. Dunn and Hodges have a stronger claim: they are direct victims of the people who stand to collect, and the fund arguably injures them by recharacterizing their assailants as victims of the state. Whether a court agrees remains to be seen.
French is cautious about the legal path. The merits are real: no statute authorizes the fund, the Judgment Fund was appropriated for specific purposes, and the civil liability release has no precedent. But merits only matter if someone can get into court. The Supreme Court has been narrowing standing requirements; the fund has been structured to delay any ripe challenge until individual payments are made and a concrete adverse party emerges. By then, money will have moved.
What is clearer is the political pressure the announcement generated almost immediately. Senate Republicans who might otherwise have moved legislation paused briefly when the Dunn-Hodges suit dropped alongside news of the fund’s details. Whether that pause hardened into anything substantive was unresolved when this episode taped.
The box Republicans *built*
The slush fund arrived the same week Trump completed his revenge sweep of Republican primaries. The two stories are related: one shows what defiance costs, the other shows what compliance is purchasing.
The May 19 primaries were a demonstration of Trump’s grip on the Republican base. Thomas Massie of Kentucky — a libertarian-leaning seven-term congressman who had defied Trump on Iran, Israel aid, the Epstein files, and the debt ceiling — lost to Trump-endorsed Ed Gallrein, a former Navy SEAL, in what became the most expensive House primary in American history: more than $34 million in advertising, including over $9 million from AIPAC and pro-Israel groups with their own reasons to oppose Massie. Bill Cassidy, the Louisiana senator who voted to convict Trump in his second impeachment trial, came in third in his primary. Indiana Republicans who had blocked Trump’s mid-decade redistricting push lost to Trump-endorsed challengers across the board. Trump endorsed Ken Paxton — impeached by his own party’s state legislature, an admitted adulterer, and one of the 2020 election’s most prominent deniers — for the Texas Senate seat, over incumbent John Cornyn.
French describes the resulting position of remaining Republican members of Congress as a genuine trap: defy Trump and likely lose the primary, comply and lash yourself to what he and Bouie both call a sinking ship. The approval numbers support the metaphor. A mid-May ARG poll placed Trump’s disapproval at 65 percent against 31 percent approval. A New York Times/Siena generic ballot poll had Democrats ahead by 11 points in May — a margin that, if it held or widened through October, would produce a Republican rout regardless of how the maps were drawn.
Shane Massey, a South Carolina state senator who resisted the redistricting push, had made the structural argument explicitly: gerrymandering compresses Democrats into fewer districts and reduces the margin for error, so that a wave election produces larger Republican losses per seat than the original map would have. French endorsed the term Bouie coined for these maneuvers: dummy-manders. Massie’s concession speech, pointedly delivered long after the polls closed because, he said, “it took a while to find Ed Gallrein in Tel Aviv,” ended with a warning the crowd turned into something close to a demand: if the legislative branch always votes with the president, it has made itself a rubber stamp. The crowd chanted “Massie for president.” It was a joke. It was not entirely a joke.
The *NPC* problem
Bouie’s most interesting contribution is not about the slush fund. It is about why the administration does not seem to understand that any of this carries a political cost.
The MAGA political universe has developed an epistemic problem that goes beyond ordinary partisan motivated reasoning. The NPC frame — “nonplayable character,” borrowed from gaming, deployed online to describe political opponents as automated and unreal — reflects something Bouie takes seriously as a genuine belief, not just rhetoric. If opponents are not real agents making real choices, then the 2020 protests were Soros operations, lost elections are fraud, and a 65 percent disapproval rating is an artifact of a rigged polling apparatus.
This is not merely cynical. It is operationally disabling. An administration that cannot model its opponents as genuine actors making consequential choices cannot anticipate their responses. Bouie traces it in foreign policy — the administration, he argues, did not seriously reckon with the possibility that Iran would make choices of its own — and domestically in the slush fund itself. The assumption appears to be that public outrage will dissolve before November because public outrage always has. That assumption requires believing that the public is not a genuine actor.
The political environment that produced Trump’s 2024 victory was specific: low-propensity voters who believed he would restore economic prosperity, turning out on that promise. Two years into a second term defined by tariffs, ongoing inflation, and $1.776 billion going to people convicted of attacking the Capitol, that electorate does not exist in the same form. What exists instead is an energized Democratic primary base — Texas Democrats turned out in record numbers this cycle — a college-educated independent bloc that has moved sharply against the administration, and an economy that French describes as potentially on course for a “speed run” to the fall of 2008.
The planned Republican response — “woke, woke, woke,” in French’s parody of the playbook — is a campaign designed for a different electorate. Bouie does not think it works when the actual grievance is economic deterioration and the spectacle of conspicuous presidential corruption. What he expects instead is a collapse in Republican turnout alongside a surge in Democratic turnout: the mirror image of 2024, driven not by enthusiasm for Democrats but by revulsion at a specific record.
Three questions remain genuinely open as of this episode: whether Dunn and Hodges will establish standing sufficient to get a merits ruling before the fund distributes money; whether Senate Republicans’ brief hesitation hardened into anything legislative; and whether the generic ballot holds, widens, or narrows as the summer economy develops.
What is not open: the fund exists, has no judicial oversight, and is drawing applicants. The settlement that created it was between Trump and Trump. The civil liability release for his family is in the published text of the agreement. Commission members can reimburse themselves from the fund they administer.
What the piece leaves the reader with is a pair of vulnerabilities running in opposite directions. French’s constitutional analysis identifies a profound legal anomaly — this specific structure is so unusual that even a court inclined toward deference might find statutory hooks sufficient to strike it. Bouie’s political analysis identifies a different vulnerability: an administration so insulated from unwelcome information that it cannot pivot even when pivoting might matter. One of those vulnerabilities will resolve before November. The question is which one resolves first, and in whose favor.