What Was the Manhattan Project, and How Secret Was It Really?

Background

In August 1939, physicists Albert Einstein and Leo Szilard wrote to President Franklin Roosevelt warning that Nazi Germany might be developing an atomic weapon and urging the United States to begin its own research. That warning, combined with British scientific findings on the feasibility of an atomic bomb, led to the formal establishment of the Manhattan Engineer District in 1942, placed under the command of Brigadier General Leslie Groves, an Army engineer who had just overseen construction of the Pentagon.

Under Groves and scientific director J. Robert Oppenheimer, the project grew into one of the largest industrial and scientific undertakings in history: roughly 130,000 people worked across dozens of sites, most significantly Los Alamos, New Mexico (bomb design), Oak Ridge, Tennessee (uranium enrichment), and Hanford, Washington (plutonium production), at a total cost of nearly $2 billion in 1940s dollars. The project successfully tested the first atomic device at the Trinity site in New Mexico on 16 July 1945, and the resulting weapons were used against Hiroshima on 6 August and Nagasaki on 9 August 1945, ending the Second World War within days.

How the Secrecy Worked

Groves's central security principle, in his own words, was that "each man should know everything he needed to know to do his job and nothing else." The project was divided into rigid compartments: construction workers and lower-level technicians typically had no idea what the overall effort was building, while only a small circle of senior scientists and officials understood its complete scope. A uranium-enrichment technician at Oak Ridge might work for years without learning that a separate facility in New Mexico was designing the weapon their material would arm.

The compartmentalisation reached remarkably high in government. President Roosevelt did not inform his own Vice President, Harry Truman, of the project's existence. Truman learned the full details only on 25 April 1945, thirteen days after becoming president following Roosevelt's death, in a briefing from Secretary of War Henry Stimson and General Groves. He witnessed the weapon's power directly for the first time at the Trinity test less than three months later, and authorised its combat use within weeks of that.

The Espionage That Got Through

Against the American public and its own government, the compartmentalised system worked essentially as designed until Truman's public announcement on the day of the Hiroshima bombing. Against Soviet intelligence, it did not. Klaus Fuchs, a German-born theoretical physicist who had fled to Britain before the war, joined the Manhattan Project's British delegation in 1943 and transferred to the Theoretical Division at Los Alamos in 1944, working under Hans Bethe on the implosion design used in the Nagasaki weapon. Throughout this period, Fuchs was passing detailed technical information, including bomb-design specifications, directly to Soviet intelligence contacts.

Fuchs was not the only source. Machinist David Greenglass, working at Los Alamos, passed additional design information through a separate network that eventually implicated Julius and Ethel Rosenberg. None of this espionage was detected by wartime security, which had been built to prevent leaks through ordinary channels rather than to catch a cleared scientist deliberately passing information. Discovery came only in 1949, when the US Army's Venona programme, decrypting wartime Soviet diplomatic cables, identified Fuchs's role. Venona's own extreme secrecy meant its intercepts could not be used as courtroom evidence, and Fuchs was ultimately caught only after voluntarily confessing to British intelligence in January 1950, a confession that then led directly to Greenglass, Harry Gold, and the Rosenbergs.

Common Misconceptions

The project's secrecy is sometimes described as a simple, unqualified success, when it succeeded against one threat, public and congressional disclosure, while failing comprehensively against another, Soviet espionage, for the entire duration of the war. Both things are documented and both are true simultaneously; treating "the secret held" as a single verdict misses the specific, narrower thing the security system actually protected against.

It is also sometimes assumed that Fuchs was caught through conventional wartime counterintelligence work. He was not. His discovery depended entirely on a codebreaking programme, Venona, that did not bear results on his specific case until 1949, four years after the war Fuchs's information had helped shape had already ended.

Current Consensus

Historians agree on the documented record in full: the project's compartmentalised security worked as intended against the American public and Congress, at least one Soviet espionage network operated inside it undetected for the war's duration, and Venona's 1949 decryptions, followed by Fuchs's own 1950 confession, are what ultimately exposed that failure. Declassified since the 1946 Atomic Energy Act transferred control to civilian authority, the Manhattan Project is one of this site's rare fully documented "secrecy" cases, with no live dispute about what happened, only about how differently its two kinds of secrecy, public and adversarial, should be judged.

Why This Story Endures

The Manhattan Project endures as a subject because it demonstrates two opposite lessons about secrecy at once, using the same programme as evidence for both: security built around institutional discipline and compartmentalisation can hold for years against the audience it is explicitly designed to hold against, while remaining almost entirely defenceless against a single motivated individual with legitimate access and a different loyalty. Every subsequent US government secrecy failure this site covers inherits some version of that same asymmetry, including the Cambridge Five, the British spy ring that Venona's same decryption effort began unravelling in 1951, two years after it identified Fuchs.

MKUltra and COINTELPRO are the taxonomy's other named examples of "once-classified state programmes now in the public record," and the comparison is instructive: both of those were exposed by external investigation and forced disclosure decades after the fact, while the Manhattan Project's core existence was always going to become public the moment its weapon was used, and its real secrecy failure lay entirely in a different direction that had nothing to do with when the bomb itself would be revealed. Why the Cold War produced so many conspiracy theories traces how episodes exactly like this one, genuine, high-stakes government secrecy proven real, primed a public that would spend the following decades finding hidden government projects newly plausible, whether or not the specific project in front of them actually existed. The Manhattan Project is part of this site's government projects cluster, within the broader secret societies and covert operations coverage.