Home > Positioning > Persons > Everett

Hugh Everett III (1930–1982)

Everett proposed that quantum mechanics needs no collapse postulate — that the wave function of the universe evolves unitarily, and that what appears as “measurement” is the entanglement of the observer with the measured system, producing a branching structure in which all outcomes occur. The proposal, originally called the “relative-state formulation” and later named the “many-worlds interpretation” by Bryce DeWitt, was Everett’s doctoral thesis at Princeton (1957), supervised by John Wheeler. It was ignored or dismissed for two decades, then revived in the 1970s and is now one of the major interpretive frameworks in quantum foundations — defended by a significant minority of physicists and philosophers, including Sean Carroll and David Deutsch. Everett left physics after his thesis and never returned.

Life

Born 11 November 1930 in Washington, DC. His father Hugh Everett Jr. was a military officer. Undergraduate at the Catholic University of America (BS in chemical engineering, 1953). Graduate work in physics at Princeton, where he studied under Wheeler — one of the leading figures in general relativity and quantum gravity. The thesis was completed in 1956 and defended in 1957.

Wheeler supported the work but was anxious about its reception — particularly by Bohr, whose Copenhagen interpretation the thesis implicitly challenged. Wheeler arranged a meeting between Everett and Bohr in Copenhagen in 1959; by Everett’s account, Bohr was unpersuaded and uninterested. The thesis, published in a shortened form in Reviews of Modern Physics (1957), received little attention.

After Princeton, Everett joined the Weapons Systems Evaluation Group (WSEG) at the Pentagon, then co-founded Lambda Corporation, a defence-consulting firm specialising in nuclear-weapons strategy, game theory, and operations research. He worked in military strategy and computer modelling for the remainder of his career. He published nothing further in physics. He returned to quantum mechanics only indirectly, through DeWitt’s 1970 revival and the subsequent growth of the many-worlds literature.

Everett was a heavy smoker and drinker, described by colleagues as brilliant, reclusive, and increasingly disengaged. He died of a heart attack on 19 July 1982 at his home in McLean, Virginia, at fifty-one. His daughter Elizabeth committed suicide in 1996; in her suicide note, she expressed the hope of joining her father in a parallel universe. Everett’s son Mark Everett, the singer of the rock band Eels, has written about his father’s life.

The relative-state formulation

Everett’s thesis begins from a dissatisfaction with the Copenhagen interpretation’s treatment of measurement. The standard formulation of quantum mechanics has two processes:

  1. Unitary evolution. The wave function evolves deterministically according to the Schrödinger equation. This process is smooth, reversible, and applies to all physical systems.
  2. Collapse. Upon measurement, the wave function collapses to one of the possible outcomes, with probability given by the Born rule. This process is abrupt, irreversible, and applies only when a “measurement” occurs.

The measurement problem: the formalism does not say what constitutes a measurement or why a measurement triggers a different process. If the measuring apparatus is a physical system (and it is), then the Schrödinger equation should apply to the apparatus-plus-system — and it predicts not collapse but a superposition of the apparatus having recorded different outcomes.

Everett’s move: drop the collapse postulate entirely. Take the Schrödinger equation as the complete description of all physical processes, including measurement. When an observer measures a quantum system in a superposition, the observer becomes entangled with the system — the composite wave function contains branches in which the observer has recorded each possible outcome. Each branch is equally real; the observer in each branch sees a definite result and is unaware of the other branches. There is no collapse; there is only branching.

The relative state: the state of any subsystem is defined only relative to the state of the rest of the universe. “The observer” does not have an absolute state; they have a state relative to the measured system. In each branch, the observer and the system are correlated — the observer’s memory records a definite outcome because the observer is in a branch where that outcome occurred.

The revival and the many-worlds label

Everett’s thesis was published in 1957 and largely ignored. DeWitt revived it in 1970 (“Quantum mechanics and reality,” Physics Today) and gave it the name “many-worlds interpretation” — a label Everett himself did not use and that emphasises the branching (multiple worlds) over the formalism (relative states). The revival coincided with a growing dissatisfaction with the Copenhagen interpretation among physicists working on quantum gravity and quantum cosmology, where the notion of an external classical observer is particularly problematic — in quantum cosmology, there is nothing outside the universe to observe it.

The many-worlds interpretation is now one of the four major interpretive frameworks (alongside Copenhagen, Bohmian mechanics, and dynamical-collapse theories). Its defenders argue that it is the most parsimonious interpretation: it adds nothing to the formalism (no collapse postulate, no hidden variables, no pilot wave). Its critics argue that the parsimony is illusory — that the interpretation requires the existence of an immense (possibly infinite) number of unobservable branches, and that the probability problem (what does “probability” mean when all outcomes occur?) has not been solved.

Where Everett stops

The probability problem is the central open question. If every branch is real and every outcome occurs, what does it mean to say that one outcome is more probable than another? The Born rule assigns probabilities to outcomes; in many-worlds, all outcomes happen with certainty (in different branches). Everettians have proposed several solutions: decision-theoretic arguments (David Deutsch, David Wallace), self-locating uncertainty (Carroll), and derivations of the Born rule from symmetry principles. David Albert has challenged each of these proposals; the debate is technical and unresolved. Whether the probability problem is a genuine obstacle or a misunderstanding of what “probability” means in a deterministic branching structure depends on which philosopher of probability is asking.

The preferred-basis problem is related. The wave function can be decomposed into branches in many ways — why does the branching happen in the basis that corresponds to classical outcomes (position, energy) rather than in some other basis? The standard answer appeals to decoherence: the interaction between a quantum system and its environment selects a preferred basis (the “pointer basis”) in which the off-diagonal elements of the density matrix are suppressed. Whether decoherence fully solves the preferred-basis problem or merely shifts it is debated.

Everett himself did not engage with these questions after 1957. His thesis was a starting point, not a completed programme. The interpretive framework that bears his name was developed by others — DeWitt, Deutsch, Wallace, Carroll — and the relationship between what Everett proposed (the relative-state formulation) and what the many-worlds community defends is not straightforward. Everett’s original formulation is more austere than the “many worlds” label suggests: he wrote about relative states, not about parallel universes. Whether the relative-state formulation and the many-worlds interpretation are the same proposal under different names or subtly different proposals is a question in the history of physics that has not been settled.

Key works

See also: Bohr · Albert · Rovelli · Relational quantum mechanics