The Assassination Conspiracy Reflex Is Now the Crisis

The most chilling part of the new assassination-attempt polling is not that Americans distrust the government. They do. The chilling part is that distrust now arrives preloaded, waiting at the scene before the blood is dry.

A NewsGuard/YouGov survey reported by NPR and WOSU¹ found that 30% of Americans believed at least one of three assassination attempts or alleged attempts against Donald Trump was staged. The survey, conducted April 28 through May 4 among 1,000 Americans, asked about the July 13, 2024 Butler, Pennsylvania shooting, the September 2024 Florida golf-course plot, and the April 25, 2026 White House Correspondents’ Association dinner shooting, according to WOSU/NPR¹. On the newest incident, only 45% called the attack real, 24% said it was staged, and 32% said they were unsure, according to the same WOSU/NPR report¹.

I think that is a civic emergency. Not because every “not sure” respondent is a conspiracy theorist. That would be lazy and unfair. The poll was fielded only three to nine days after the April 25 dinner shooting, and “I don’t know yet” can mean low information, caution, or lack of trust in early reporting rather than denial of reality, as the field dates in the WOSU/NPR account¹ make clear. The emergency is narrower and more serious: a large share of the country is ready, almost immediately, to classify political violence as theater.

That changes the function of violence in public life. An assassination attempt is supposed to trigger a shared sequence: protect the target, care for victims, investigate the facts, condemn the act. People can argue about motive, security failure, ideology, and punishment later. That shared first step matters because it tells would-be attackers that violence will isolate them rather than turn them into mascots for a faction’s information war.

The evidentiary record here is not just cable-news chatter. In the Butler shooting, the FBI said² it was investigating the July 13, 2024 rally shooting as an assassination attempt and potential domestic terrorism, and said the incident caused one victim’s death and injuries to Trump and other spectators. In the Florida golf-course case, the Justice Department said³ Ryan Wesley Routh was convicted by a federal jury on all five counts and sentenced to life in prison plus 84 months for attempting to assassinate Trump and assaulting a federal law-enforcement officer. In the dinner case, prosecutors have not proved their case at trial, and Cole Tomas Allen has pleaded not guilty, according to the Associated Press⁴. But the federal indictment⁵ alleges that Allen attempted to assassinate the president and assaulted a federal officer with a deadly weapon during the April 25 shooting at the Washington Hilton.

Those distinctions matter. Routh is a convicted defendant. Allen is accused. Butler involved a dead spectator, wounded people, and a killed gunman, according to the FBI². A serious citizen should treat those categories differently. But “staged” skips the hard work of distinction. It does not say, “I want the charging documents.” It says, “The event belongs to my narrative until reality proves otherwise.”

The counterargument deserves respect because it begins with something true. Americans have not lost trust for no reason. Gallup reported⁶ in 2025 that only 28% of U.S. adults had a great deal or fair amount of trust in newspapers, television, and radio to report news fully, accurately, and fairly, the lowest level in Gallup’s trend. Pew Research Center found⁷ in 2024 that only 22% of Americans trusted the federal government in Washington to do what is right just about always or most of the time. These are not fringe numbers. They are the atmosphere.

Nor is distrust just a mood. The Senate Intelligence Committee⁸ said in 2008 that several Bush administration public statements before the Iraq War were not supported by the available intelligence. The Justice Department inspector general⁹ found significant inaccuracies and omissions in the Carter Page Foreign Intelligence Surveillance Act applications, while also finding no documentary or testimonial evidence that political bias or improper motivation drove the opening of the FBI’s Crossfire Hurricane investigation. That combination is exactly why this is hard: institutions can be legitimate and still fail badly.

But institutional failure cannot become an all-purpose acid that dissolves every fact it touches. The right lesson of Iraq, FISA errors, Afghanistan, botched early reporting, and partisan media is not that nothing official can be believed. The lesson is that important claims need proof, documents, timestamps, video, adversarial review, and correction when the first account is wrong. Skepticism asks for a receipt. Conspiracy thinking declares the receipt forged before seeing it.

This is where the poll becomes more than a media-trust story. The United States is already living with an unusually high tolerance for political force. A PBS News/NPR/Marist poll¹⁰ reported in 2025 that 30% of Americans said people may have to resort to violence to get the country back on track, up from 19% about 18 months earlier. PRRI reported¹¹ in 2026 that 20% of Americans agreed that “true American patriots” may have to resort to violence to save the country. In that environment, mass doubt about whether attacks happened does not sit harmlessly in the corner. It weakens the basic social penalty for violence.

The mechanism is simple. If an attack on your political enemy is “staged,” you do not have to condemn it with the same force. If an attack on your political champion is “fake news,” you do not have to trust the investigation. If every event becomes an audition for the narrative most useful to your side, then political violence no longer shocks the system. It feeds the system.

I do not want official accounts treated as sacred texts. In fact, the only workable answer is almost the opposite: more evidence, faster, in forms that can be checked by people outside the government’s trust circle. Release timelines. Release surveillance video where doing so will not compromise security or witnesses. Publish charging documents, autopsy summaries, ballistics findings, after-action reviews, and corrections of early errors. Bring in inspectors general and bipartisan congressional reviewers when security failures are at issue. Make the factual baseline visible enough that citizens do not have to accept it on institutional prestige alone.

But leaders, media outlets, and platforms also need to stop treating “staged” narratives around political violence as just another spicy take. There is a difference between asking whether the Secret Service failed in Butler and claiming Butler was fake. There is a difference between questioning prosecutors’ theory in the Allen case and declaring the dinner shooting a performance. Platforms do not need to erase every false claim to make verified timelines easier to find than engagement bait. Politicians do not need to police private thought to condemn conspiracy claims from their own supporters. Newsrooms do not need to feign certainty to say what has been established and what has not.

My fear is that America has built an information culture in which the first instinct after a violent political event is not grief, alarm, or even anger. It is genre recognition. People see the images and instantly decide whether the plot feels like something their side would call real.

Here is my test for the next two months: if investigators release substantial video, timelines, and court evidence in the April 25 dinner case, and the share calling it staged falls sharply below 10%, then the main problem was delayed verification in a low-trust society. If, by midsummer, the staged number remains near 20% or higher despite inspectable evidence, then the darker diagnosis is right: conspiracy has become a standing reflex, not a response to missing facts. My prediction is the second outcome unless political leaders start paying a real social cost for laundering unreality when violence hits the other tribe.