This article is part of Cafe, TPM’s home for opinion and news analysis. It first appeared at Balls and Strikes.

Around 6 a.m. on Monday, October 12, roughly two dozen federal agents raided the Washington, D.C., residence of Representative Ilhan Omar, a Minnesota Democrat and the first Somali American member of Congress. The raid was part of the Trump administration’s sprawling investigation centered on Omar’s citizenship status. Over the past two weeks, federal officials have seized her phone, passport, notebooks, and computers. They’ve issued grand jury subpoenas to her family members. They’ve floated the possibility of criminal charges and prison time, and claimed that the probe may allow the Trump administration to strip Omar of her citizenship and facilitate her deportation.

Back in 2019, the last time the Trump administration directed federal authorities to comb through Omar’s immigration records, they found no evidence of wrongdoing. But actual legal wrongdoing is not the point. What’s “wrong” for Trump is the idea that a Black and Muslim woman he deems inferior could come to the United States and be granted rights as his equal.

Omar arrived in the United States in 1995 as a 12-year-old refugee of the Somali Civil War, and was naturalized as an American citizen in 2000. For years, Trump has disparaged Omar and other Somali Americans as “garbage,” and claimed that people from “the worst countries” should not come to the United States and start “telling us about the Constitution.” 

Under federal law, “having opinions about the Constitution” does not appear on the list of permissible reasons to revoke someone’s citizenship. Congress has authorized the Justice Department to pursue civil denaturalization proceedings only when the government alleges that someone should not have become a citizen in the first place — for example, by procuring that status “by concealment of a material fact or by willful misrepresentation.” Longstanding Supreme Court precedent requires the government to prove those allegations with “clear, unequivocal, and convincing evidence which does not leave the issue in doubt.”

The Supreme Court has also recognized that “citizenship obtained through naturalization is not a second-class citizenship,” and that a legal system in which “a group of citizens temporarily in office can deprive another group of citizens of their citizenship” is “completely incongruous” with “the very nature of our free government.” Accordingly, actual revocations of citizenship have been exceedingly rare in the modern era: Fewer than 150 Americans were denaturalized between 1968 and 2013, compared to more than 18 million Americans who were naturalized in that same period. And when the rare denaturalization does occur, the targets have tended to be people suspected of war crimes or terrorism.

The Trump administration is trying to change that. Between 1990 and 2017, the Justice Department filed an average of 11 denaturalization cases per year. During Trump’s first term, the average rose to 25 cases per year. Now, the Trump administration has directed the Department of Homeland Security to “supply” the Justice Department with “100-200 denaturalization cases per month,” and the DOJ has further instructed all employees in the Civil Division to “prioritize and maximally pursue denaturalization proceedings.”

Trump’s Justice Department is acting on these instructions. In April 2026, administration officials informed federal prosecutors that they had identified 384 Americans to target in a first wave of denaturalization cases. And earlier this month, the DOJ announced that it had filed 40 such denaturalization actions between August 21 and October 2. Additionally, researchers at the Transactional Records Access Clearinghouse have identified more than 100 denaturalization complaints filed by the Trump administration so far this year. Another database, maintained by Democracy Forward, shows nearly 200 civil denaturalization cases filed since January 2025.

The federal government’s ongoing attacks on Ilhan Omar are part of the Trump administration’s attempt to act as sole arbiter of who gets to be an American. It issues orders purporting to rescind birthright citizenship. It deploys federal agents to surveil, detain, and deport people of color en masse. And it is waging a denaturalization crusade that aims to make equal membership in society depend on race, ancestry, and Trump’s singular will.

This article first appeared at Balls and Strikes. If you enjoyed this post, please consider supporting the work of Balls & Strikes with a monthly or annual subscription. Click here to learn more and sign up.

This article is part of Cafe, TPM’s home for opinion and news analysis.

Donald Trump fancied himself an expert on television long before parlaying his titan-of-industry image, forged by “The Apprentice,” into politics. But his “reality-TV presidency” — complete with cliffhangers, confrontations, and eliminations — increasingly resembles a once-popular show in its later seasons, resorting to gimmicks as it scrambles to sustain an audience.

As someone who chronicled Trump’s primetime success and absorbed his claims (invariably false) about titanic ratings long after “The Apprentice” slipped off its lofty perch, there’s an odd familiarity to the desperate push toward the 2026 midterms. Trump’s shtick mirrors the tricks and traps of an aging reality show, from poorly vetted, swing-for-the-fences casting decisions to escalating levels of violence and risk, to, finally, lavish cash giveaways. (The main difference, actually, is that TV shows have to make good on their payout promises.)

In just the last few weeks, Defense Secretary Pete Hegseth announced plans for a live-streamed execution, only to see the president retreat from the idea, despite his initial endorsement of it; Trump chose Katie Zacharia as his press secretary with scant due diligence before firing her after a half-Scaramucci; and the president touted a variety of prize incentives to voters, including a $5,000 pledge to all adults — but only if the election results go his way — and to-be-determined cash awards to whistleblowers to advance his anti-fraud obsession.

Based on the polling, nothing really appears to be moving the needle. Among students of television, that really shouldn’t come as much of a surprise, since once a “show” enters this phase, there’s often not much that can be done to stave off the inevitable decline, as fickle viewers begin looking for something new, gravitating toward the next big thing.

Trump remains fond of pointing out when women are beautiful as an act of nose-thumbing rebellion against “woke” culture, but his recent gushing about the appearance of various men, and their bodies, has become a new source of fascination, as well as a gift to late-night comics. But it all fits the pattern of Trump as the producer of his own show, where he’s invariably the hero surrounded by enemies on all sides. Within that aura of showmanship, Trump also views himself as a casting genius uniquely able to identify star power, underscored by his repeated references going back years to “central casting.”

Although “Survivor” producer Mark Burnett was the ostensible mastermind behind “The Apprentice,” Trump clearly felt like he internalized the tools of the trade. But these recent examples underscore the challenges of such sleight of hand, as well as how trying to bend an existing format can blow up in one’s face.

To those who might see the reality-TV analogy as strained, remember how Trump got here. The way “The Apprentice” carefully crafted Trump’s persona through creative editing, which was necessary to make sense of his behavior, is a story that’s been told and retold, primarily via confessionals and apologies from former staffers and the NBC marketing guru who helped launch the series.

Bill Pruitt, an early “Apprentice” producer, acknowledged in a 2024 Slate piece that he and his colleagues “played fast and loose with the facts,” creating “the myth” of Trump, beginning with the glittering boardroom built to stand in for the rundown space in his real-life offices.

Nor should it come as a shock that Trump harbors an inflated sense of his reality-TV acumen. Another producer, Jonathon Braun, recalled how Trump would make out-of-left-field choices about who to fire, forcing the producers to retrofit the entire episode to justify his chosen outcome. “Our job then was to reverse engineer the show and to make him not look like a complete moron,” Braun told New York Times reporters Russ Beuttner and Susanne Craig for their book “Lucky Loser: How Donald Trump Squandered His Father’s Fortune and Created the Illusion of Success.”

John Miller, the now-retired NBC executive, titled his 2024 essay/mea culpa “We Created a Monster: Trump Was a TV Fantasy Invented for ‘The Apprentice.’” As he told me in an interview in 2025, the qualities that Trump prized above all else were “loyalty and being telegenic,” adding, “Experience and competence are well down that list.” He also observed that Trump was both “manipulative,” and “extraordinarily easy to manipulate.”

Trump is gambling on a different, more serious playing field, but the inherent strategy is the same. For a president who sees himself as the ultimate showman, breathing life into a tiring act has notably led to some very rare backpedaling.

Zacharia offered a textbook illustration of those priorities, having effusively praised Trump on television — another attractive mouthpiece on the surface, appointed without evidence of the customary concerns about her qualifications or resume. 

Such corner-cutting can be perilous, and again, reality TV provides ample precedent. Take “The Bachelorette,” which picked Frankie Taylor Paul for the last edition despite questions about off-screen issues, then wound up having to shelve the entire season after TMZ posted video pertaining to a 2023 domestic-assault allegation.

Television producers regularly talk about “raising the stakes,” and stoking conflict or the perceived threat of violence is a go-to means of seeking to juice ratings. That might explain Hegseth’s tough-guy talk about a live-streamed execution, before the collective disgust over what the Daily Kos called a government-sponsored “snuff film” prompted the White House to back down.

Still, as critic Matthew Gilbert pointed out, televised executions seem less far-fetched than when “Network” satirized them 50 years ago. While daredevil televised stunts are as old as Evel Knievel, the possibility of seeing someone die — live! — has become another threshold for standing out amid a sea of content, such as “Free Solo” free climber Alex Honnold’s January special scaling a Taiwan skyscraper live on Netflix.

A less bloodthirsty gambit involves simply throwing extra prize money around, as “Survivor” did by doubling the potential payout for its 50th edition. Like physical altercations and tossed wine glasses, that usually feels like an exercise prone to yielding diminishing returns, to be pursued after exhausting most other options. 

Students of television, of course, will tell you that once a show enters this part of its lifecycle, there’s not much that can stave off the inevitable declines. Hit shows usually blaze hot, then fade, even if they endure for a long time like “Survivor,” like “The Apprentice” a  Burnett production, or “Dancing With the Stars.” That explains why something like “The Bachelor” franchise would roll the dice by selecting a bachelorette that ended up costing ABC and parent Disney millions in unused (so far, anyway) episodes.

Trump is gambling on a different, more serious playing field, but the inherent strategy is the same. For a president who sees himself as the ultimate showman, breathing life into a tiring act has notably led to some very rare backpedaling. And despite how effective the maestros at “The Apprentice” were in stitching together a better Trump, facing an angry electorate, he might be confronting his limits, finally, in terms of shaping reality.

This article is part of Cafe, TPM’s home for opinion and news analysis.

The Overton window on expanding — or packing or unpacking — the Supreme Court is now wide enough that it’s worth revisiting one inevitable part of the conversation: FDR’s Court-packing proposal in 1937. Misconceptions about the events abound in popular discussions.

Here’s the short version of the story you’ve probably heard: In 1935 and early 1936, the conservative-dominated Supreme Court handed down several decisions whose doctrines threatened important parts of FDR’s New Deal economic policies. After his overwhelming reelection in November 1936, he decided to pack the Supreme Court with six new justices, using the excuse that the sitting justices over the age of 75 weren’t able to handle the Court’s work effectively. Announced in January 1937, the Court-packing plan came as a surprise to political Washington, and immediately met strong resistance. FDR slogged on but the plan suffered several severe blows, including a letter from Chief Justice Charles Evans Hughes endorsed by liberal justice Louis Brandeis saying that the Court was completely up to date with its work. Other developments undercut the urgency for the president and his allies, including a decision favoring New Deal economic policies in March 1937 in which swing justice Owen Roberts cast the deciding vote and the retirement in May of conservative justice Willis Van Devanter. When the plan was formally interred in a mid-summer Senate vote, a New York journalist called Roberts’s vote the “switch in time that saved the nine,” and the phrase — and Roberts’s vote — have come to dominate today’s understanding of the episode.  

That version isn’t completely inaccurate and it isn’t a caricature, but it does contain some misconceptions, the most important of which is probably the focus on the captivating “switch in time” phrase. If we think that the story of the Court packing plan has some lessons for us today, it’s important to clear up these misconceptions. Historians have written books about the events surrounding the Court-packing plan and this essay won’t be able to explore all the ins-and-outs of a complicated set of events. Even more, the story has many moving parts, most of which are connected to others, so you can pick pretty much any place to start.

Was the Court-packing plan a surprise?

Yes and no. After one important decision in May 1935 going against the New Deal, FDR held a press conference and said that the Court wanted to take the country back to the horse-and-buggy days. Almost from the start of his first term FDR groused about the possibility that the Supreme Court would stand in the way of what he thought were programs needed to recover from the economic crisis the nation faced. Ideas about “doing something” about the Court floated around Washington starting in 1933, leaked from the administration to favored reporters and discussed at Georgetown dinner parties.

Still, FDR hadn’t directly targeted the Supreme Court during his 1936 reelection campaign. Instead, he regularly attacked the American Liberty League, an advocacy organization founded and funded by wealthy opponents of the New Deal. The League produced a series of pamphlets designed as templates for legal briefs laying out the constitutional arguments against a host of New Deal programs. When FDR went after the League, he was really taking on reactionary legal arguments he feared the Supreme Court would adopt. So, in the guise of the American Liberty League, the Supreme Court was part of the 1936 presidential campaign.

Supreme Court reform of some sort was in the air after FDR’s massive victory in November 1936. The development of the Court-packing plan was a closely held secret within the administration. The content of the announcement on February 5, 1937, was something of a surprise. But no one was surprised that FDR wanted to move some sort of Court reform through Congress.

Was there a switch in time?

No, but maybe yes. On June 1, 1936, the Supreme Court held a New York statute setting a minimum wage for women workers unconstitutional by a vote of 5-4, with Justice Roberts in the majority. Then, on March 29, 1937, the Court held an essentially identical Washington state statute constitutionally permissible by a vote of 5-4, with Roberts in the majority. Coming less than two months after the Court-packing plan’s announcement, Roberts’s “new position” is the switch in time.

It’s been known for decades, though, that Roberts actually cast his deciding vote in mid-December 1936, before the Court-packing plan was announced. Justice Harlan Fiske Stone was absent because he was recovering from a near-fatal illness. The justices divided four-to-four, with Roberts voting to uphold the Washington statute. Stone’s vote was necessary for the new majority but everyone knew that he’d vote to uphold the Washington statute (he’d joined the dissent in the New York case).  The case was essentially over when the Court took its initial vote. The formal vote was delayed until February 6, 1937, the day after the Court-packing plan was announced. Justice Felix Frankfurter later reported that Roberts came up with a lame explanation for his switch, saying that New York’s lawyers hadn’t explicitly asked the Supreme Court to overrule a controlling precedent but Washington’s lawyers did. (The excuse was lame because New York’s lawyers had indeed asked the Court to overrule the precedent, though in a short and somewhat cryptic paragraph.)  Whatever the reason, it didn’t come “in time” — that is, it didn’t come as a response to the February 5, 1937, unveiling of the Court-packing plan.

Remember, though, that Court reform was in the air. Maybe Roberts was picking up what we’d now call “vibes” about a threat to the Court, and maybe the lameness of his explanation was just a cover story for a change induced by the general atmosphere of threats to the Court as an institution.

Was the plan doomed from the start?

No. Vote counters knew that the House of Representatives would accept the plan. All the action was in the Senate, where there was fierce opposition from conservative Republicans and some Democrats, and unease among other Democrats. 

An adept politician, FDR held an ace in his hand, and everyone knew it. He promised one of the seats on an expanded Supreme Court to Senate majority leader Joseph Robinson of Arkansas. Robinson found the offer irresistible. He called in all the chits he had with his colleagues, putting together a potential majority for the plan one vote at a time. Political observers at the time weren’t sure what the outcome would be but they were inclined to think that Robinson could pull off a victory for FDR.

The final vote was scheduled for mid-July in an exceptionally hot (and un-airconditioned) Washington. Robinson held a meeting–social gathering where he continued to lobby wavering colleagues. The combination of heat and stress gave Robinson a fatal heart attack. The possibility of a majority for the Court-packing plan in the Senate always depended on personal ties between Robinson and other senators, and, with his passing on, the plan failed. 

Did FDR lose the battle but win the war? How?

Yes, yes, and maybe no. FDR lost the battle over expanding the Court but reached his main war goal — having a Court that wouldn’t completely obstruct his favored programs. As with almost all successful military campaigns, the path to the goal wasn’t straight. The war wasn’t over on March 29 when the nation learned that Roberts had voted to uphold Washington’s minimum wage law. FDR’s legal team worried that Roberts wasn’t going to be a reliable ally going forward — and indeed he wasn’t: a careful modern statistical study shows that Roberts reverted to his conservatism within a year.

The war wasn’t over on May 18, either, the day Justice Willis Van Devanter announced his intention to retire even though the retirement seemingly gave FDR the opportunity to replace a conservative with a reliable supporter of the New Deal. 

Van Devanter had wanted to leave the Court for years, mostly so that he could take care of his medically compromised wife. The obstacle was financial. Van Devanter was something of a penny-pincher, once complaining that he had been overcharged five cents for a package of seeds. With some reason, he worried that his financial condition would deteriorate if he left the Court. As the statutes stood at the beginning of 1937, lower court judges could either retire from the bench and continue to receive their judicial salaries or they could resign and take their pensions. The rules for Supreme Court justices were different. They had to resign, not retire — they could take their pensions, but not their salaries. And that mattered because judicial salaries for retired lower court judges couldn’t be reduced, but judicial pensions for them and for Supreme Court justices could be — and they were, as part of the austerity programs Herbert Hoover and FDR pursued at the Great Depression’s outset. That was what worried Van Devanter, because the statutes didn’t give him the retirement option. He would have to resign and be at the mercy of congressional decisions about judicial pensions. 

Hatton Sumners, the chair of the House Judiciary Committee, had long thought that treating Supreme Court justices worse than lower court judges was silly. Some lawyers worried that “retirement” required that the judge be available to serve as needed — something now known as “senior status” — and that the Constitution didn’t allow that for Supreme Court justices. Sumners disagreed. In 1936 he introduced a bill to eliminate the difference. It failed. The next year, with Supreme Court reform on the table, he tried again. With some misgivings, and aware of the important role Sumners would play in shepherding the Court-packing plan through the House, FDR’s lawyers decided that Sumners’s bill wouldn’t undermine support for Court-packing.  The House passed the bill on February 10, the Senate on February 26, and FDR signed it on March 1, whereupon Van Devanter decided to leave the Court.

Why wasn’t the ability to appoint a justice enough to get FDR to abandon the politically costly Court-packing campaign? The answer: Senator Joe Robinson, to whom FDR has promised a court seat. Despite that promise, Robinson wasn’t a full-throated New Dealer and FDR’s lawyers worried that he might end up as unreliable as they feared Roberts would be. FDR took his time filling the vacancy Van Devanter created, with the excuse that Robinson was still needed to push the Court-packing plan through the Senate. A month after Robinson died, FDR nominated Senator Hugo Black to the Court, giving him a reasonably solid five-to-four majority. The next year conservative Justice George Sutherland, who like Van Devanter had wanted to leave the Court for several years, retired, and FDR nominated Solicitor General Stanley Reed to succeed him. 

By then the war really was over. The Supreme Court didn’t invalidate any major New Deal programs after 1937, although it did sometimes interpret major statutes — especially the National Labor Relations Act — in ways less friendly to labor than FDR’s labor allies would have liked. Democrats suffered severe losses in the 1938 midterm elections, mostly because the nation was in a severe economic recession induced by FDR’s too-rapid return to the orthodoxy of balanced budgets — but maybe the Court-packing episode made it easier for some voters to abandon the Democratic Party.

In the even longer sweep, the conservative war against legal liberalism continued. Liberals liked to describe a “New Deal settlement” that coupled judicial deference to legislatures on matters of economic policy with judicial activism on civil rights and civil liberties. Recent scholarship has made the case that conservatives never really accepted the so-called settlement, making some tactical concessions while biding their time — which came toward the end of the 20th century.

Did 1937 set a “precedent” for Court expansion?

Maybe yes, but a lot depends on how you describe the precedent. Focus on the politics of Court size and the answer is that the 1937 episode might have created what the British call a constitutional “convention” against changing the size of the Supreme Court simply to affect predicted outcomes — but that kind of convention can itself be changed when politicians think that it stands in the way of getting something important done. And focus on the politics of the federal courts more generally and the answer is that changing their powers to advance partisan goals is standard practice.

The Court’s size has been fixed at nine since 1869. Before that the Court grew to accommodate national growth — a “good government” reason for changing its size. Sometimes the timing of expansion was affected by partisan concerns. And in the decade of the Civil War, Congress expanded the Court, shrank it, and expanded it again mostly for partisan reasons but with some real “good government” explanation as well. 

FDR’s Court-packing plan was packaged in “good government” terms — too many of the Court’s justices were old and unable to keep up a work pace adequate to needs. That was a fig-leaf covering the partisan motivation and FDR and his allies quickly abandoned it. The fig-leaf that’s offered in today’s discussions is “thirteen circuits, thirteen justices,” but no one really thinks that there’s a strong good-government reason for having one justice for each circuit. The Court-packing plan’s failure might be understood as establishing a convention against changing the Court’s size unless there’s a real “good government” reason for doing so. 

Unwritten constitutional conventions like that are always up for grabs. One problem is distinguishing between practices that persist because no one cares enough to challenge them and practices that persist because people think they ought to follow them. The complicated politics that account for the failure of FDR’s Court-packing plan — Robinson’s death, Roberts’s change in vote, Sumners’s bill and Van Devanter’s retirement — weaken the case for saying that 1937 shows that there’s a convention against changing the Court’s size to affect predicted outcomes. Since then, the issue of the Court’s size didn’t matter enough for anyone to try to do something about it — until now. So maybe having a Court with nine members results from a practice rather than a convention.

Even more, conventions can be fragile — certainly they’re not as robust as provisions written into the constitutional text. A scholar writing in 1925 offered an example. From Washington’s presidency to the early 20th century American presidents stayed on American soil (broadly defined, it turns out). U.S. presidents stayed on the U.S. side of the Mexican border when they met their Mexican counterparts on the bridge connecting El Paso and Juarez. In 1914 Woodrow Wilson reluctantly declined an invitation to England to celebrate the anniversary of the treaty ending the war of 1812. Within five years he went to Versailles, and the convention disappeared. 

The picture is much more clear when we look at the politics of the federal courts generally. Congress has adjusted their powers for good government reasons, of course, but it also has routinely tinkered with the federal courts to change predicted outcomes. When Congress feared that the national courts would rule against workers seeking compensation for injuries, it shunted those cases to state courts. When Congress feared that state courts would treat Blacks badly, they allowed the cases to be shifted into the federal courts. The list goes on and on. Seen in that light, changing the Court’s size is just another example of treating the federal courts as one of the many tools politicians can use to achieve their goals.

Lessons from the past?

One of the first things graduate students in history are told is that they should be wary about drawing lessons from the foreign country that is the past. If lessons are to be drawn, though, they should be drawn on the basis of as full an understanding as possible of what really happened — an understanding informed by the story I’ve told, and of course by the even more complete story that can be found in the historical record.

In Rough Edges, Mike Rothschild writes about fringe groups, conspiracy theories and how the Internet broke our brains. This column is part of TPM Cafe, TPM’s home for opinion and news analysis.

People love a grand conspiracy. The bigger the scheme is, the more people get drawn into it. And the more details and players that are involved, the more believable it seemingly becomes. After all, who would go to the trouble of making up something so complicated? It must be true.

And nobody loves a conspiracy more than President Trump. He and his inner circle especially seem to have fallen in love with what has literally been dubbed “The Grand Conspiracy.” It’s a hopelessly complex, entirely unevidenced theory claiming that a high-ranking group of political figures have been plotting in the shadows for years to stop Donald Trump from making America great again through investigations, indictments and impeachments. It has all the hallmarks of the type of grandiose schemes that Trump supporters especially fall for: countless players, sprawling plots, intricate secrets, and a total lack of evidence that it actually took place.

Despite lacking cogent answers to foundational questions such as “what it is” and “who is doing it,” revenge for the Grand Conspiracy has been a driving force in the administration, prompting an investigation that has trundled along over the last year and a half, involving dozens of officials looking for some reason to indict the president’s enemies.

And there are many, many enemies. They include a greatest hits of Trump’s obsessions and grievances, from former special counsel Jack Smith to minor players like former FBI agent Peter Strzok, and even former President Barack Obama himself. According to Grand Conspiracy believers, they are all powerful Deep State figures who supposedly worked together in a single, years-long scheme to destroy Trump and undo the will of the people. 

Precisely what these conspirators have actually done remains unclear — even, seemingly, to the prosecutors themselves, who, per the Wall Street Journal, complained to the Justice Department that they weren’t getting much guidance from the local U.S. attorney. The probe is now being run out of Washington, where it continues to baffle those assigned to move it forward. 

In the last few weeks, there have been a host of indications that DOJ efforts to prosecute these figures — and tie them all together into one grand scheme — are falling apart. And, as the Grand Conspiracy prosecution flails at DOJ, the claims also aren’t doing much in the court of public opinion, where Republicans are looking for anything they can find to shore up their crumbling midterm prospects.

This article is part of Cafe, TPM’s home for opinion and news analysis.

Significant polling errors most often result when late-deciding voters break differently from historical patterns. That’s likely to happen this year, and not in a way that favors Republicans. 

Start with the observation that the current midterm environment is utterly unlike any in recent history. The CNN average of recent polls pegs the proportion of voters who disapprove of Trump at 65 percent, and a recent You/Gov Economist poll found that of those who disapprove, 84 percent of them disapprove of him “strongly.” We’ve had no recent midterm cycle in which most American voters strongly disapprove of the president. 

If we look at the distribution of Trump disapprovers, the implications for Republicans become even more ominous. Let’s say that Trump’s most loyal MAGA voters make up 30 percent of the electorate and assume, plausibly, that most of them associate largely with Americans of similar views. Since this group is largely devoid of Trump disapprovers, the upshot is that an overwhelming majority of voters not in this group — some 70 percent of the electorate — are Trump disapprovers. And since there are also almost no undecided voters among Trump loyalists, the portion of the electorate that describes itself as “undecided” also must consist almost entirely of Trump disapprovers.

To see how this may matter, consider the observation that most of us are more heavily influenced than we realize by the behavior and beliefs of peers. (This was the core premise of my 2019 book, Under the Influence: Putting Peer Pressure to Work.) As decades of careful studies have shown, peer behavior strongly influences almost everything we do and think — what we eat, how much we exercise, whether and how much we smoke and drink, how much we weigh, and which candidates we support.

Now imagine the decision facing an undecided voter. She’s almost surely not a Trump loyalist, and virtually all of her close friends and relatives are likely to disapprove of Trump, a majority of them strongly. They’re unhappy about gas prices, the cost of groceries, corruption, the war in Iran, the behavior of ICE, and more. Herself a likely Trump disapprover, she’s almost surely unhappy about many of these things as well. Late deciders in past elections often tended to break toward candidates who share their political identity. If this year’s late decider is a conservative, that would predict she’d vote Republican. But given everything we know about human behavior, casting that vote would be difficult in the current environment. It would mean bucking the views of almost everyone she knows and cares about.

How much might all this matter? A recent generic ballot poll estimated that 8 percent of likely voters remain undecided. If these voters would have split 4D-4R in a normal midterm environment, a 6D-2R split in today’s climate would hardly seem unlikely. And that four percentage point swing would almost surely be more than enough to erase any pro-R polling bias from other sources.

But there’s more. Peer influences shape not only the decision about whom to vote for but also the decision about whether to vote at all. This influence will of course cause current polling averages to further understate Democratic strength.

In a recent interview with MS NOW host Jen Psaki, former Obama adviser David Plouffe was understandably reluctant even to discuss the possibility that current polls might be understating Democratic strength. Acknowledging this possibility could of course encourage some people to slack off, thinking the prize already in hand. 

But understating Democratic prospects also entails risk. As we’ve seen, Republican seats are vulnerable in many areas that earlier Democrats would never have thought to contest. There’s reason to believe that the playing field may yet warrant further expansion. 

This article is part of Cafe, TPM’s home for opinion and news analysis.

President Trump has in recent days doubled down on a September executive order, rebranding Artificial Intelligence (AI) to Super Intelligence (SI). Those unwilling to embrace his preferred terminology — “the highly accepted new and more accurate term,” as he described it on Truth Social — would be deemed “THE ENEMY!” The whole thing may seem on first glance to be silly or innocuous. In reality, however, the rebranding of AI to SI isn’t just about nomenclature. 

Trump has waded into a debate on what the ultimate outcome of AI may be: one in which the world, because of AI, becomes a utopia, or one in which the world becomes a dystopia due to AI.

Beneath the debate over artificial intelligence sits a bigger question about where the technology is going. “Superintelligence” marks the point where AI stops merely matching or beating humans at specific tasks. It describes machines that outthink the best human minds in nearly every field, from reasoning and planning to scientific discovery and problem-solving.

Silicon Valley has broadly divided into two schools over what that transition would mean. On one side are the techno-optimists and accelerationists, represented by figures such as Marc Andreessen and, in a more qualified way, Peter Thiel, who see increasingly powerful AI as a means of breaking technological stagnation and producing extraordinary advances in science, productivity and human prosperity. Andreessen has gone so far as to describe AI as a “universal problem solver,” while Thiel has argued that AI could finally reaccelerate technological and economic progress after decades of relative stagnation. In this view, AI, as it becomes superintelligence, means we are, in simple terms, moving towards a utopian society.

The other school sees doomsday. Its clearest voices are Eliezer Yudkowsky and Nate Soares, whose recent book carries an intentionally unambiguous title: “If Anyone Builds It, Everyone Dies.” They argue that creating an intelligence vastly greater than our own before solving the problem of controlling it could result in human extinction. That concern extends well beyond them. Geoffrey Hinton and Yoshua Bengio, two pioneers of modern artificial intelligence, are among hundreds of researchers and technology leaders who have publicly warned, as I recently did, that mitigating the risk of extinction from AI should be treated alongside threats such as nuclear war.

The underlying fear is remarkably simple: humanity has never created an intelligence greater than its own. If a superintelligent system develops goals that diverge from ours — or simply pursues objectives in ways its creators never anticipated — we may discover too late that being its creator does not make us its master. For this school, “superintelligence” does not evoke abundance or technological liberation. It evokes the possibility of doomsday.

At September’s United Nations General Assembly meeting, President Trump shunned the idea of global standards and regulations. In doing so, he has firmly planted himself in the camp of the utopians who see regulation, especially by a multilateral organization, as choking off the possibility of a utopian techno-state. There are several implications of Trump’s push towards superintelligence but three are especially important to consider.

First, preemptively rebranding AI as “superintelligence” normalizes the danger. When the White House, federal agencies, and the largest tech companies talk about superintelligence as something America should build and lead, the debate shifts. The question is no longer whether to build it. The question becomes how to win the race.

Second, it speeds up that race. Trump says America cannot “stifle” AI because it is ahead of China and must maintain that advantage. For anyone who believes uncontrolled superintelligence could end civilization, this is the nightmare: two great powers convinced they cannot afford to slow down.

Third, it takes some of the force out of the warning. AI safety advocates spent years using “superintelligence” to describe a future threshold carrying potentially existential risks. Trump now uses it as an optimistic label for technologies that already exist. Indeed, his executive order temporarily defines “Super Intelligence” as the very technologies already covered by the federal definition of artificial intelligence. If that usage sticks, saying “superintelligence is coming” no longer sounds quite like an alarm. The president has already declared that its era has arrived.

And then there is Elon Musk. Few people illustrate the tension better. Musk has spent years warning that advanced AI could pose an existential threat to humanity. Yet when Trump inaugurated his “Era of Super Intelligence,” Musk was standing right next to the president. In October, Musk went further, pledging to rename SpaceXAI to SpaceXSI. 

That does not necessarily mean Musk has abandoned his warnings. But it does illustrate how dramatically the politics surrounding those warnings have changed. The question for the doomer camp may no longer simply be whether humanity should create superintelligence. It may be what influence their warnings can retain now that the United States government has made achieving utopia through superintelligence a matter of national ambition.

This article is part of Cafe, TPM’s home for opinion and news analysis.

Donald Trump’s ability to jail millions of people without even a bond hearing depends on what the Supreme Court thinks about one man’s case. 

This book excerpt is part of Cafe, TPM’s home for opinion and news analysis.

On June 25, 2013, the Roberts Court handed down their decision in Shelby County v. Holder. The 5-4 decision gutted the Voting Rights Act of 1965’s “preclearance” condition, which required states and localities with a history of racial discrimination in voting practices to get permission from the federal government before they could change any policy related to their elections. The country had changed dramatically, Chief Justice Roberts wrote for the majority; the “strong medicine” of the Voting Rights Act, he argued, was no longer needed, let alone justified. But Michael G. Miller and I show throughout our new book, An American Problem: How the Country Built – and the Supreme Court Broke — The Voting Rights Act just how misguided that reasoning was. In the parts of the country freed from preclearance by Shelby County, our electoral institutions have grown demonstrably less open to Americans of color.

When Mike and I started working on what would become An American Problem, a promise made by Justice Roberts in Shelby County still held. “Section 2” of the 1965 Voting Rights Act, he wrote — the section containing the tools allowing litigators to attack election laws and policies that disproportionately harmed voters of color ± “is permanent and applies nationwide.” Although the Shelby County majority was happy to invalidate the preclearance regime that June morning, they told the country that tools remained still for removing discriminatory laws from the statute books.

But this is no longer the case.

This article is part of TPM Cafe, TPM’s home for opinion and news analysis. It was originally published at Balls and Strikes.

On Tuesday, October 6, Justice Samuel Alito released So Ordered, his addition to the Supreme Court’s library of vanity books written by justices who decided that their $306,600 salaries are not enough. Alito’s book is particularly unnecessary because most of it is not even original content, but a collection of adapted speeches he’s given since his confirmation in 2006. As it turns out, a lesser-known perk of being a Supreme Court justice is that you can get paid to publish old talking points thrown together by clerks whose names you forgot long ago.

Much of the coverage of Alito’s book is due to ongoing speculation that he could retire soon, ensuring that a Republican president and a Republican-controlled Senate can nominate and confirm his successor. While promoting So Ordered, though, Alito has said that he plans to “serve for a certain number of additional years,” and “intends” to “stick around through this term.” On the one hand, I understand why this is a story; on the other, given that Alito is a lifelong Republican with an alarming history of flying January 6-adjacent flags over his homes, if Democrats take the Senate in the 2026 midterms, I will not be shocked if he changes his mind.

This article is part of Cafe, TPM’s home for opinion and news analysis.

As insurgent candidates offering critiques of the Democratic Party won primary after primary in 2025 and 2026, pundits were left wondering: were the Democrats having their own “Tea Party moment”? Back in 2010, the Tea Party was key to securing massive GOP legislative gains, even as its reactionary edge cost the GOP winnable senate seats in places like Delaware and Nevada.

Progressive triumphs in Democratic primaries during president Trump’s second administration might suggest something similar is on the way. Zohran Mamdani beat Andrew Cuomo in the race for New York’s mayoralty, Abdul El-Sayed’s triumphed against establishment pick Haley Stevens in the Michigan Senate primary, and Graham Platner easily dispatched — despite scandals of such magnitude he would ultimately have to withdraw as the party’s general election candidate — Gov. Janet Mills in the Maine Senate primary. Each candidate offered fresher rhetoric, more ambitious politics, and a critique of the Democratic Party, allowing them to edge out — or in some cases destroy — their opponents. Cue the invocations of the Tea Party and questions about whether or not “the Democratic Party might very well be about to explode.”  

But comparisons between the Democratic Party’s fresh crop of candidates and the Tea Party do not really stand up to scrutiny. Tea Partiers were motivated by an anti-systemic energy that had deeper roots than the current ire of Democratic voters, and the demographics of the GOP meant the party ultimately embraced a “burn it down” approach to governance. The insurgent left within the Democratic coalition remains focused on conventional liberal issues like human rights, affordability, and taxing the wealthy. New, animating issues for voters — such as ending military funding for Israel and harsher regulation of data centers — still do not see voters anywhere close to the Tea Party’s reactionary fervor to remake, or remove, much of the U.S. government. Democratic voters — and the candidates they elect — may find that current leadership is not fit for their objectives, but replacing a few key stakeholders is a different dynamic from the Tea Party’s anathema towards any Republican interested in “normal” governance.

This op-ed is part of Cafe, TPM’s home for opinion and news analysis.

Last week, New York City announced a groundbreaking $131.5 million settlement with DoorDash. The largest such resolution in city history, the case involved over 260,000 workers who were paid late and in some cases not paid at all. 

It also raises the question: If these underpayments happened in New York City, where DoorDash knew it was under scrutiny, what’s happening to gig workers everywhere else when no one is watching?

The census makes politics. Representatives are allocated to states based on the count of the population, and states redistrict their maps accordingly. Thus, it’s perhaps not surprising that different partisan forces work behind the scenes to make census data in their own image. 

In advance of the 2030 Census, the Trump Administration has recently announced that it plans to drastically upend an esoteric set of rules known as the “residence rules.” These are the protocols that dictate who is counted where for the upcoming census. Although Congress is ultimately responsible for overseeing the census and determining the rules, it absconded with that responsibility decades ago. This has created a vulnerability that the Trump administration seeks to exploit. In changing the rules, the Trump administration is hoping to alter the political map. As TPM has reported, civil servants have been ordered to stay silent. 

Data Are Made, Not Found: A Story of Politics, Power, and the Civil Servants Who Saved the US Census tells the story of what it took for dedicated bureaucrats to pull off the 2020 Census, despite a pandemic, natural disasters, financial constraints, and political interference. This book shows what it takes to make the census so that each and every one of us can fight to ensure that democracy’s data remain intact and legitimate. The ins and outs of who gets counted where might seem boring, but this is what makes the process so vulnerable to interference. When the public and our elected officials do not pay attention to details, technocratic moves become a form of policymaking by other means.

This article is part of TPM Cafe, TPM’s home for opinion and news analysis. It was originally published at The Conversation.

President Donald Trump loves a good pardon. His generosity is so well known, in fact, that an industry of lobbyists, lawyers and other insiders has reportedly sprung up to facilitate clemency for wealthy criminals. Clemency, meaning mercy or leniency, lets the president reduce the punishment for, or even forgive, federal crimes.

For defendants hoping to avoid jail and walk away from millions of dollars in financial sanctions, paying, say, $600,000 to a lobbying firm to influence the president may seem like a good investment.

House Democrats recently reported that Trump’s second-term clemency – which includes both pardons, which wipe away every part of a conviction, and commutations, which typically leave convictions intact but shorten a person’s sentence – has saved convicted defendants nearly $1.7 billion in forgiven financial obligations.

Some of that amount – about $700 million – had not yet been officially imposed by courts because the clemency came through so quickly. Much of the remaining $1 billion had been ordered as restitution, meant to repay victims’ losses.

As a legal scholar researching the evolution of modern presidential clemency, I have reviewed every individual pardon and commutation granted by U.S. presidents since 1989, spanning seven administrations, from George H.W. Bush to Trump’s second term. I have found that while other presidents have canceled tens of millions of dollars’ worth of restitution and fines, none of them have come anywhere near the billion-dollar mark – until now.

What is restitution?

The main financial penalties faced by federal defendants are fines, forfeiture – where a defendant has to give up, say, the Ferrari they bought with the profits from their drug business – and restitution.

Restitution is an order to pay back the criminal’s gains to the victims of the crime. For example, if the CEO of a company tricks people into investing $600 million into a hydrogen-powered truck that was never operational, the court may order that, if convicted, he pay back the people he defrauded.

Restitution is rarely able to return all of someone’s lost money, but it is supposed to prevent the fraudster from enjoying the fruits of their crime.

Getting away with it

Since January 2025, Trump has granted clemency to a lengthy list of big-ticket fraudsters, including a cryptocurrency billionaire whose exchange pleaded guilty to anti–money laundering violations and CEOs convicted of payroll tax fraud.

Many of these pardons and commutations have coincided with the use of lobbying firms or generous donations to causes close to Trump’s heart: his inaugural committee, his reelection campaign or MAGA Inc. One successful pardon-seeker’s mother attended a $1 million fundraising dinner.

Clemency seekers use a variety of ways to catch the president’s attention. The end result is that victims of crime – and taxpayers, who have to pay for these huge white-collar investigations and trials – are often left with nothing.

Unfettered power

The U.S. Constitution grants the president nearly unbounded power to pardon whomever he likes, whenever he likes and for whatever reason he likes. This is why, since 1894, the Office of the Pardon Attorney has tried to impose some guidelines on the process and decide which cases should be eligible for clemency at all.

These guidelines generally require that a clemency seeker have been released from prison for five years before they are eligible to request a pardon. If they never went to prison, the five years is calculated from the date of sentencing. Many presidents adhere to this norm: My study shows that between 94% and 100% of people pardoned by George H.W. Bush, George W. Bush and Barack Obama met this criteria. Because he gave so many preemptive pardons, Joe Biden’s percentage was lower, at around 68%, but still well over half.

In contrast, out of the 216 individual pardons granted by Trump in his second term so far, only 34 – about 15% – have gone to people who waited that long.

The money ladder

A look back at recent past administrations shows an erratic but increasing amount of financial obligations canceled. While some presidents have been frugal with their financial forgiveness – George H.W. Bush and George W. Bush remained under the million-dollar mark, while Joe Biden was just over it – others have been less so.

Bill Clinton erased nearly $10 million in fines and restitution through pardons, most notoriously pardoning Marc Rich, a fugitive financier whose ex-wife contributed more than a million dollars to various Democratic causes and the Clinton Presidential Library.

Obama more than doubled Clinton’s record, forgiving $23 million in financial sanctions.

But Trump surpassed them both in his first term, canceling nearly $195 million in financial obligations. The amount of money saved by convicted defendants during his second term is exponentially higher. One could even say that, as Trump likes to put it, “We’ve never seen anything like it.”

A new role for commutations

Since the main function of commutations is to let people out of prison sooner, commutations usually leave financial obligations in place. During his first term, Trump primarily granted commutations that left “intact and in effect” all conditions of a person’s sentence except for imprisonment. This let numerous defendants out of jail but kept them on the hook for nearly $3 billion in fines and restitution.

That has changed during Trump’s second term. Today, his commutations routinely cancel a recipient’s prison sentence “with no further fines, restitution, probation or other conditions.”

This is how, through commutations alone, Trump has canceled more than $300 million in penalties and restitution since he returned to office.

A word of caution

One figure that is unavailable in public records is whether any given defendant has started to pay, or has fully paid off, their financial obligations. So these figures for every administration represent a ceiling: the full amount of fines, restitution or forfeiture that a judge imposed on a convicted defendant. How much was actually paid down can usually only be estimated, though there are exceptions.

For example, Jason Galanis, who pled guilty to defrauding the Oglala Sioux Nation, had paid just over $2 million of his $80 million restitution before his sentence was commuted and the restitution erased in 2025. But we only know that because he unsuccessfully asked the court to return the money.

One useful indicator of how much someone may have already paid is the speed at which they got clemency. The faster they receive clemency, the less time they have to pay off their penalties. The median delay between sentencing and receiving a pardon during Biden, Obama and George H.W. Bush’s administrations was over 20 years.

During Trump’s second term, in contrast, things are moving faster. Trump has pardoned or given commutations to at least 44 recipients within a year of their sentences who together account for over $270 million in financial obligations.

This includes a nursing home executive who pled guilty to withholding his employees’ payroll taxes – which prosecutors said he spent on a yacht – and was pardoned 12 days after being sentenced, ensuring that he could avoid some $4 million in restitution. A crypto company was pardoned hours before it was due to pay a $100 million fine for violating anti–money laundering laws. A digital media founder was ordered to repay defrauded investors $36 million. His sentence was commuted, and his restitution canceled within four months of sentencing.

In these cases, clemency was issued before most of the money was able to reach the government or the victims it was owed to. The faster clemency is granted, the more likely it is that fraudsters get to walk away without paying their victims anything.

This article is republished from The Conversation under a Creative Commons license. Read the original article.

This article is part of TPM Cafe, TPM’s home for opinion and news analysis. It was originally published at Balls and Strikes.

On the morning of Sunday, September 20, police officers in the Chicago suburb of Evanston responded to a 911 call about a confrontation in the street. When the officers arrived, they found a man on the ground who reported injuries to his head, neck, and teeth. Soon, the officers learned that the injured man was a U.S. citizen, and his assailants were federal immigration agents.

The Department of Homeland Security put out a statement on Tuesday claiming that the beaten man “resembled the target” of an immigration enforcement operation, and that he was “not cooperative” with agents’ investigation. “Once the individual finally identified himself,” said DHS, immigration officers “left the scene without further incident.” 

In Church, Merch, and State, Sarah Posner writes about the intersection of religion and politics in the United States. This column is part of Cafe, TPM’s home for opinion and news analysis.

President Donald Trump’s “border czar” Tom Homan delivered a no-holds-barred defense of Trump’s mass deportation regime at the Georgia Faith and Freedom Coalition’s Victory Dinner in Duluth, Georgia, last week in an event that appeared designed to bolster the campaigns of two major Republican candidates running for U.S. Senate and governor in the state. 

Homan’s speech signaled that an unabashed defense of the administration’s immigration brutality is a key part of its midterm strategy to mobilize right-wing evangelical and Catholic voters to back Republicans. In a surprisingly profanity-laced speech that castigated critics of Immigration and Customs Enforcement (ICE), including Democrats and federal judges, Homan became another administration official to inject himself into midterms campaigning this cycle, this time to lend a Christian imprimatur to Trump’s racist and violent assault on immigrants.

This article is part of TPM Cafe, TPM’s home for opinion and news analysis. It was originally published at Balls and Strikes.

For months, President Donald Trump has been coping with his myriad failures and plummeting approval ratings by trying to make it harder for people to vote his party out of power in the 2026 midterms. To that end, in March, he issued an executive order that would require the U.S. Postal Service to draft a rule that would upend vote-by-mail procedures in dozens of states, and effectively block the agency from processing ballots that do not comply with the rule’s exacting requirements. Trump did not attempt to hide his motives for doing so: Legal restrictions on mail-in voting, he told Republican lawmakers, would “guarantee the midterms” for a party that is otherwise in danger of losing them badly.