The Many Worlds Interpretation and DARPA

Hugh Everett III wrote one of the most consequential papers in the history of physics and then spent the rest of his career doing classified work for the Pentagon. These two facts are connected in ways that have not been examined in any public document.

Everett developed the many-worlds interpretation of quantum mechanics as his doctoral dissertation at Princeton in 1957, working under physicist John Wheeler. The theory proposed that the wave function of quantum mechanics never collapses. Instead, every quantum event causes the universe to branch into separate realities, one for each possible outcome. Every measurement, every interaction, every quantum event that could go multiple ways does go multiple ways, in separate branches of an ever-proliferating multiverse. There is no collapse. There is no randomness. There is only branching.

Niels Bohr rejected the theory. The physics community largely ignored it. Everett left academic physics and went to work for the Weapons Systems Evaluation Group, a Pentagon defense analysis agency, where he spent the next twenty-five years doing classified work that has never been publicly described in any detail. He published almost nothing in physics after his dissertation. He died of a heart attack in 1982 at 51.

The Weapons Systems Evaluation Group, known as WSEG, was established in 1948 as an independent analytical body serving the Joint Chiefs of Staff. Its function was to apply rigorous mathematical and scientific analysis to questions of military strategy, weapons system design, and nuclear war planning. WSEG employed some of the most technically capable people in the country during the Cold War. Its work was classified at the highest levels and its products went directly to the senior military and civilian leadership making nuclear strategy decisions.

Everett joined WSEG in 1956, a year before his dissertation was formally submitted. He worked there through most of his career, eventually becoming a senior analyst. The specific content of his WSEG work is not in the public record. What is known from biographical sources is that his work involved game theory applications to nuclear strategy, optimal targeting analysis, and the mathematics of nuclear exchange scenarios. He was working on the formal analysis of situations involving multiple simultaneous outcomes with branching probability structures.

The connection between that work and the many-worlds interpretation he had just developed is not incidental. Many-worlds is precisely a framework for thinking about systems where all outcomes occur rather than one outcome being selected. The mathematics of a universe that branches at every quantum decision point is structurally related to the mathematics of decision trees in game theory and to the analysis of scenarios where multiple simultaneous futures must be evaluated. Whether Everett's classified defense work applied many-worlds thinking explicitly, or whether the two bodies of work informed each other implicitly, is not established in any available document.

Nuclear war planning in the 1950s and 1960s confronted a specific analytical problem. The outcome of a nuclear exchange depended on a vast number of interacting variables: the number of warheads that successfully launched, the number that survived interception, the accuracy of delivery systems, the hardness of targets, the sequence of strikes, the decision-making of the adversary under attack. Each of these variables had probability distributions. The interaction of all of them produced an enormous space of possible outcomes.

Standard probabilistic analysis treats this as a problem of expected values: you calculate the probability-weighted average outcome across all scenarios. Many-worlds thinking offers a different framing: all outcomes occur, in different branches. The question is not which outcome will happen but which branch you want to be in. The strategic implications of that reframing are not trivial. If every exchange scenario that is physically possible occurs in some branch, the question of how to plan for nuclear war changes character. You are not trying to maximize your probability of a good outcome. You are trying to ensure that the branches in which you survive are more numerous or more probable than the branches in which you do not.

Whether this framing was applied in Everett's classified work is unknown. Whether it influenced the development of nuclear strategy in ways that are visible in declassified strategic documents from the period is a question that has not been asked in the academic literature on either Everett or Cold War nuclear strategy. The overlap between the man who invented the framework and the classified work he spent his life doing has been noted by Everett's biographers and then not pursued.

The academic reception of Everett's dissertation was almost uniformly negative at publication. John Wheeler, his advisor, was sympathetic but encouraged Everett to soften the theory's claims before publication, which Everett did in the published version of the dissertation. The shortened published version removed some of the more philosophically explicit material from the longer original. Bohr and the Copenhagen school found the theory unnecessary and philosophically objectionable. The consensus of the physics community was that the Copenhagen interpretation was adequate and that Everett's branching universes were an extravagant solution to a problem that did not exist.

Everett attended one physics conference after his dissertation and attempted to discuss the theory with Bohr's group. He found no receptive audience and did not return to academic physics. The theory was largely forgotten until 1970, when physicist Bryce DeWitt published a popularization of Everett's work in Physics Today that used the phrase "many worlds" for the first time. That article brought the theory to a wider audience and began the gradual rehabilitation of many-worlds as a serious interpretation of quantum mechanics.

By the time many-worlds began receiving serious academic attention, Everett had been doing classified defense work for over a decade and had no interest in re-entering academic debates about his dissertation. He was aware of the growing interest in his theory and found it somewhat amusing that physics was catching up to work he had done and abandoned fifteen years earlier. He gave a small number of interviews in the late 1970s and attended a conference on the theory in 1977. He died five years later without having returned to active physics research.

Everett's work at WSEG and at the defense consulting firm he subsequently founded, Lambda Corporation, remains classified. The specific analytical products he produced, the strategic questions he was assigned to answer, and the methods he applied to those questions are not in the public record. Biographers who have written about Everett have been unable to access his classified work files through FOIA requests, as the relevant materials remain protected under national security exemptions.

Lambda Corporation, which Everett founded after leaving WSEG, continued defense consulting work including work for DARPA. The specific DARPA programs Lambda worked on have not been identified in any publicly available document. Everett ran Lambda until his death and the company continued under other management afterward. Its classified work product from the Everett era has not been released.

The question of whether the many-worlds interpretation of quantum mechanics was applied to defense problems during the Cold War, and whether those applications produced results that remain classified, is not answerable from the available evidence. The man who developed the theory spent his entire career doing classified defense work that used the same mathematical structures his theory describes. His files have not been released. The intersection of his two bodies of work has not been examined publicly. It may never be.

Hugh Everett invented many-worlds quantum mechanics in 1957 and immediately went to work for a classified Pentagon agency doing nuclear strategy analysis. He published nothing further in physics and spent twenty-five years applying mathematical frameworks structurally related to his theory to classified defense problems. His WSEG and DARPA work files have not been released. The connection between the most radical interpretation of quantum mechanics ever proposed and the classified nuclear strategy work of the man who proposed it has not been examined in any public document. The files that might answer the question remain classified.