By Emilia Wellesley · Published May 8, 2026 · Updated May 8, 2026
Last reviewed: May 8, 2026.
The screenshot is dated. The thread is archived. The original poster’s account is suspended. Working the chain backward on the 1938 War of the Worlds broadcast lands you in a strange place: the panic everyone remembers may have been smaller than the panic about the panic. The first viral hoax of the broadcast era is mostly a story about how a story propagates, and most of what people now write about it cites secondary write-ups of secondary write-ups.
Direct Answer: What Actually Happened on October 30, 1938
On Sunday, October 30, 1938, CBS aired Orson Welles and the Mercury Theatre on the Air’s adaptation of H.G. Wells’s 1898 novel, scripted by Howard Koch as fake news bulletins. Some listeners believed Martians had landed in New Jersey. Real-time panic was modest. The narrative of nationwide hysteria was largely amplified the next morning by newspapers competing with radio for advertising revenue. [1][2][3]
The Broadcast: A Halloween Episode That Sounded Like the News
The show went to air at 8 p.m. Eastern from CBS’s New York studios as the Halloween episode of The Mercury Theatre on the Air, a sustaining program with no commercial sponsor. Welles, twenty-three years old and already running a radical theater company, narrated and played the Princeton astronomer Professor Pierson. Howard Koch, hired weeks earlier at fifty dollars a week, restructured H.G. Wells’s Victorian novel as a contemporary news bulletin and relocated the Martian landing from Surrey to Grover’s Mill, New Jersey. [1][4]
The opening framing announced a dramatization. The broadcast then mimicked a routine evening of dance music interrupted by escalating news bulletins: a weather forecast, a hotel orchestra, then a reporter on the scene in New Jersey, then a heat-ray attack, then collapsing reporting from a transmitter going dark. Listeners who tuned in late, especially those switching over from the popular Edgar Bergen ventriloquist program on NBC during a musical break, missed the framing and dropped straight into what sounded like a national emergency. [1][4]
The Panic Narrative: Cantril, Newspapers, and the Stories That Stuck
The next morning’s headlines did the heavy lifting. The New York Times ran a front-page story on October 31, 1938, titled “Radio Listeners in Panic, Taking War Drama as Fact,” describing flooded police switchboards, families fleeing homes, calls to hospitals. Wire services pushed similar copy nationwide. Within a week the story had been laundered into received fact: a radio play had triggered mass hysteria in the United States. [3][5]
In 1940, Princeton psychologist Hadley Cantril cemented the panic narrative academically with The Invasion from Mars: A Study in the Psychology of Panic, the first book-length study of the broadcast. Working with Hazel Gaudet and Herta Herzog at the Princeton Radio Research Project, Cantril estimated about six million people heard the broadcast, that roughly 1.7 million understood it as a real news bulletin, and that around 1.2 million were “frightened or disturbed.” [6]
For decades, those numbers traveled. They got cited in journalism textbooks, in PBS documentaries, in undergraduate communication courses, in popular histories of media effects. The Cantril figures are why the 1938 broadcast became the canonical example whenever a journalist or media studies professor needed shorthand for “powerful media effects on a credulous public.”
Why the Panic Story Was Useful
A Halloween radio play that fooled millions reads as a parable. It served editors writing about Cold War propaganda, parents writing about television, scholars writing about the rise of mass communication. It still shows up in modern coverage of deepfakes and AI-generated misinformation. The story is portable. The story scales. That is why the chain of citations runs so deep.
The Revisionist Takedown: A Press Fabrication, Mostly
The chain breaks under closer reading. On October 28, 2013, Slate published Jefferson Pooley and Michael Socolow’s article “The Myth of the ‘War of the Worlds’ Panic,” arguing that the panic was largely a press fabrication. Their reading of the C.E. Hooper telephone ratings for the night put the Mercury Theatre’s audience at roughly two percent of polled households, an order of magnitude below Cantril’s six-million estimate. The Hooper interviewers asked respondents what they were listening to; almost none mentioned the Mercury broadcast. [2]
Pooley and Socolow argued that Cantril’s audience numbers leaned on a CBS-commissioned post-broadcast survey, with sampling and recall problems baked in by the very newspaper coverage Cantril was trying to study. Their broader claim: American newspapers, having lost classified and display advertising to radio across the Depression, used the Welles broadcast to argue that radio was an irresponsible news medium and should not be trusted with hard journalism. [2][7]
A. Brad Schwartz extended that case in his 2015 book Broadcast Hysteria: Orson Welles’s War of the Worlds and the Art of Fake News. Schwartz read roughly two thousand letters that listeners sent to Welles, the FCC, and CBS in the weeks after the broadcast. The letters describe a population that mostly enjoyed the show, recognized it as fiction, and was offended by newspaper coverage suggesting otherwise. A real minority were genuinely frightened. The leap from “a real minority” to “nationwide panic” happened in print, not on the street. [8]
The FCC Investigation
The Federal Communications Commission opened a formal review within hours, asked CBS for a transcript and recording, and closed the inquiry in December 1938 without sanctioning Welles, the Mercury, or the network. The Commission ruled that CBS would tighten its policies on simulated news bulletins and barred future complaints about the broadcast from being raised during station license renewals. The regulatory outcome was modest. The cultural memory was the opposite. [1][8]
Why This Story About a Story Matters Now
The 1938 broadcast is the cleanest available laboratory case for a pattern that recurs every time a new medium scares the established one. Newspapers in 1938 had reasons to discredit radio. Cable news in the 2000s had reasons to discredit blogs. Legacy outlets in the 2020s have reasons to characterize TikTok and Substack as misinformation engines. The pattern is not new. The Welles case is just the version with the best paper trail.
For working historians, the Welles case also exposes how durable a sourced-once claim can become. Cantril published one book in 1940. That book got cited, the citations got cited, and within thirty years the original methodological caveats had been sanded off. The Pooley-Socolow critique relied on archival ratings data that had always been available; nobody had bothered to compare it to Cantril’s numbers in print until 2013. The provenance was there. The chain of citation just had not been worked backward. [2]
The cultural memory survived because it was useful. It explained television panics in the 1950s, video-game panics in the 1990s, social-media panics in the 2010s, and AI panics now. Every generation gets to reach for the Welles broadcast the way previous generations reached for tulip mania. The reach is the point. Whether the underlying numbers support the reach is, apparently, secondary.
What the Modern Reader Should Take
Two things at once: the broadcast was a real cultural event and a real moment of confusion for some listeners, and the canonical “millions panicked” framing was largely manufactured by newspapers with a competitive interest in framing it that way. Both can be true. The strong revisionist position is not that nothing happened on October 30, 1938 — it is that the newspapers wrote the version that survived.
How a Story About a Story Propagates
The chain looks roughly like this: a Halloween radio drama airs, a real but limited number of listeners are confused, newspapers competing with radio amplify scattered reports into a national panic narrative, an academic study built partly on that newspaper coverage publishes hard numbers two years later, those numbers get cited for decades without re-checking the underlying data, a working historian and a journalism scholar finally compare the academic numbers against contemporary ratings data, and the corrected version reaches a smaller audience than the original myth.
That is recognizable. It is the same shape as a hundred internet hoaxes since: confident claim, secondhand citations, deferred forensics. The difference is the timeline. In 1938 the propagation took a generation. In 2026 it takes an afternoon. The mechanism is the same. The lesson is the same. Trace claims back to the primary source. Note when you cannot. Treat virality as evidence of nothing.
Frequently Asked Questions
Did War of the Worlds actually cause a nationwide panic in 1938?
No, not in the form usually described. Some listeners were genuinely frightened, especially those who tuned in late and missed the framing. Subsequent research by Pooley, Socolow, and Schwartz indicates the audience was much smaller than commonly cited and that the “millions panicked” framing was largely amplified by newspapers competing with radio. [2][8]
How many people actually heard the broadcast?
Hadley Cantril’s 1940 study estimated roughly six million listeners, with about 1.7 million understanding it as real news. C.E. Hooper telephone ratings from that evening suggested an audience of around two percent of polled households, dramatically lower than Cantril’s number. The discrepancy is the core of the revisionist case. [2][6]
Why did newspapers cover it as a panic?
American newspapers had been losing classified and display advertising revenue to radio throughout the Depression. The Welles broadcast offered a useful frame: radio was unreliable, sensational, and not to be trusted with serious news. The narrative served the print industry’s commercial interests in 1938. [2][7]
Did the FCC punish Welles or CBS?
No. The FCC opened a review almost immediately and closed it in December 1938 without sanctions. CBS agreed to tighten its policies on simulated news bulletins. The Commission also barred future complaints about the broadcast from being raised during station license renewals. The regulatory outcome was modest. [1]
Who actually wrote the script?
Howard Koch wrote the script under deadline pressure, with collaboration from John Houseman and Welles. Koch went on to co-write Casablanca and won an Academy Award for that screenplay. Koch’s Mercury contract gave him rights to the script, which became valuable as the broadcast’s reputation grew. [1][4]
What did Orson Welles say afterward?
Welles held a press conference the next morning where he appeared shaken and apologetic. He emphasized the show’s framing as fiction. In later interviews across his career, his stance shifted between humility and amusement, depending on the audience. The press conference photographs are the most-reproduced images of the entire episode. [4]
Was Cantril’s 1940 study peer-reviewed and respected?
It was a Princeton University Press monograph and became foundational in mass-communication research. Modern critics including Pooley and Socolow argue the study had methodological problems, particularly with audience estimates and reliance on a sample shaped by the very press coverage Cantril was studying. The book remains a landmark; its specific numbers are now contested. [2][6]
Are there recordings of the original broadcast?
Yes. CBS preserved a transcription disc, and the full one-hour broadcast is freely available through the Internet Archive and the Library of Congress National Recording Registry, which inducted the broadcast in 2003. The script itself has also been reprinted in multiple editions. [9][10]
What is the most current scholarly consensus?
Working media historians broadly accept the revisionist framing: the broadcast caused real but limited confusion, and the “nationwide panic” narrative was largely a press construction. The Cantril study remains historically important but its specific audience and panic numbers are no longer treated as authoritative. [2][8]
Why does the panic story keep getting repeated?
It is a useful parable. Editors reach for it whenever a new medium provokes anxiety: television, video games, social media, generative AI. The narrative is portable, it scales, and the original source got buried under decades of secondhand citation before the revisionist scholarship could catch up. [2]
