Many-Worlds Interpretation of Quantum Mechanics — SEP (Vaidman)

Reference

Vaidman, L. (2026). Many-Worlds Interpretation of Quantum Mechanics. In E.N. Zalta & U. Nodelman (Eds.), The Stanford Encyclopedia of Philosophy (Summer 2026 Edition). First published 2002; substantive revision June 17, 2026.

Core Idea

The MWI holds that there are numerous parallel worlds existing in the same space and time as our own. Recognizing these additional worlds makes it possible to remove fundamental randomness and action at a distance from quantum theory. The MWI provides a solution to the measurement problem.

Two Parts of the Theory

1. A theory yielding the time evolution of the quantum state of the single Universe (Schrodinger equation / relativistic generalization). Rigorous, mathematically unproblematic.
2. A prescription setting up correspondence between the quantum state of the Universe and our experiences (approximate, FAPP - for all practical purposes).

What Is a World?

A world is the totality of macroscopic objects (stars, cities, people) in a definite, classically described state. In the MWI there is only ONE Universe containing a vast multiplicity of worlds. Temporal asymmetry: a world traces back to a unique past but branches into a multitude of futures.

Personal Identity (“Who am I?”)

“I” is defined at a time by a complete classical description of my body and brain. I correspond to a unique past self but multiple future selves. Meaningless to ask “Which future self will I be?” - I will correspond to them all. Related to Parfit’s critique of personal identity (1986).

Measure of Existence

The measure of existence of world i is mu_i = |alpha_i|^2 (squared amplitude in the decomposition of the universal wave function). Parallels probability measure in Everett (1957).

Probability in the MWI

• No genuine probability: all outcomes are realized. But the illusion of probability can be explained.
• Probability Postulate (Born-Vaidman rule): an observer should set subjective probability of an outcome proportional to the total measure of existence of all worlds with that outcome.
• Self-location ignorance probability: post-measurement uncertainty (sleeping pill experiment).
• Deutsch (1999) / Wallace: derivation from decision theory (rational betting). Criticized (Kent, Albert, Price).
• Zurek (2005): envariance symmetry derivation of Born rule.
• Sebens & Carroll (2018): self-location uncertainty proof; criticized by Kent (2015), McQueen & Vaidman (2019).

MWI and (Non)Locality

• Bell’s argument cannot get off the ground in MWI because Bell’s theorem requires a single outcome. The MWI is dynamically local - no action at a distance.
• Nonseparability (entanglement) remains: a “world” is a nonlocal concept.
• Deutsch & Hayden (2000): local descriptions of quantum information using Heisenberg picture.

Tests of the MWI

• In principle distinguishable from collapse theory via interference between worlds (undoing a quantum experiment with macroscopic devices) - but strictly gedanken, far beyond current technology.
• If a collapse process exists, MWI fails; collapse models predict minute violations of energy conservation, not yet observed.
• Statistical evidence supports the Probability Postulate as strongly as Born rule in other frameworks.

Objections

• Ockham’s razor: criticized for multiplying entities; defended as most economical in LAWS (no collapse postulate).
• Preferred basis problem: solved by decoherence - locality of interactions defines the preferred basis.
• Wave function is not enough (Bell, Maudlin): wrong type of object (configuration space vs 3+1 space-time). MWI counters with density of particles as primitive ontology.
• Derivation of probability postulate: naive counting-of-worlds derivations give wrong predictions; measure of existence resolves.
• Social behavior: quantum Russian roulette - Behavior Principle makes MWI believers behave conventionally.

Why the MWI?

• Avoids collapse postulate (genuinely random + action at a distance).
• No experimental evidence favoring collapse.
• Resolves paradoxes (Schrodinger’s cat, Wigner’s friend, Elitzur-Vaidman interaction-free measurement).
• Favored by quantum information researchers: quantum computation as parallel processing across parallel worlds (Deutsch & Jozsa 1992) - though questioned (Steane 2003, Duwell 2007).
• Berenstain (2020): MWI continues the pattern of scientific revolutions that decenter humans (heliocentrism, Darwin, relativity).
• Still lacks consensus (Allori 2023; Gibney 2025 survey).