The Everettian probability problem: deriving the Born rule and the Principal Principle from the non-probabilistic postulates of Everettian quantum mechanics

In the Everett (many-worlds) interpretation, every measurement outcome occurs, yet rational agents apparently should set their credences equal to Born-rule branch weights. Brown and Ben Porath (2020) argued that the Deutsch-Wallace (DW) decision-theoretic theorem neither derives the Born rule from non-probabilistic postulates alone nor justifies Lewis's Principal Principle (PP) linking chances to credences, and left five interlinked issues open: a non-question-begging physical derivation of strong (Gleason-style) non-contextuality; the status of branch weights as agent-independent chances from which the PP follows; the refutability problem posed by deviant branches (and the associated redundancy challenge to the DW theorem); the extension of the derivation to epistemically non-ideal agents; and the Gleason-route gaps for qubits and countable additivity. As of 2026, every strand remains open or actively contested: the redundancy dilemma was re-litigated by March (2024); Saunders' emergent-chance and branch-counting programs, Zhang's additivity-necessity results, and new self-location and future-record models all leave the normative link between physical facts and credences underived.

Background

In the Everett, or many-worlds, interpretation of quantum mechanics — also called Everettian quantum mechanics (EQM) — the universal quantum state evolves deterministically by the Schrödinger equation and never collapses. When a measurement-like interaction occurs, decoherence (the rapid dynamical suppression of interference between superposition components, caused by entanglement with the environment) splits the state into effectively independent branches, each containing a definite measurement record and a copy of the observer. All outcomes occur; there is no single factual result. The Born rule assigns to a branch with amplitude $c$ the weight $|c|^2$, which in collapse interpretations is the probability of that outcome. The interpretive puzzle, often called the incoherence problem or probability problem of EQM, is that if every outcome happens it is unclear what a numerical probability of an outcome is a probability of, and why any branch's inhabitants should expect Born statistics.

Two derivation strategies frame the problem. The first is Gleason-type: Gleason (1957) proved that any countably additive probability measure on the lattice of closed subspaces of a Hilbert space of dimension $D \ge 3$ that is non-contextual — assigning each projector (equivalently, each closed subspace) the same probability regardless of which mutually incompatible measurement context embeds it — is represented by a density operator, i.e. is the Born rule $Pr(E)=\mathrm{Tr}(\rho E)$. This route assumes strong non-contextuality and countable additivity (additivity over countably infinite orthogonal families, which de Finetti-style finite-additivity subjectivists reject), and it fails outright for qubits (two-level systems, $D<3$), since in two dimensions orthogonality imposes no nontrivial constraints on probabilities for projective measurements. Partial remedies widen the measurement class: Busch (2003) proves a Gleason-type theorem for $D \ge 2$ using positive-operator-valued measures (generalized measurements whose outcomes correspond to positive operators rather than projectors), and Wright and Weigert (2019) prove one for qubits using mixtures of projective measurements.

The second strategy is decision-theoretic. Deutsch (1999) and, in a rigorous form, Wallace (2012) proved the Deutsch-Wallace (DW) theorem: an agent in an Everettian universe whose preferences over quantum games (bets whose payoffs are delivered on different branches) satisfy certain rationality axioms — indifference to branching per se, indifference to microscopic inessentials, plus continuity and structural richness axioms — must act as if maximizing expected utility with credences equal to the Born weights. This is a Savage-style representation theorem (after Leonard Savage's decision theory: preferences satisfying the axioms are shown to be representable as expected-utility maximization) in which the credences, normally free parameters, are uniquely fixed by the unitary dynamics. The DW theorem covers Hilbert spaces of arbitrary dimension, an advantage over Gleason-type proofs. The rationality axioms were criticized early by Hemmo and Pitowsky (2007).

The chance-credence link. In the philosophy of probability, chances are objective, agent-independent probabilities; credences are rational degrees of belief; and Lewis's (1980) Principal Principle (PP) states that a rational agent's credence in an event, conditional on the event's chance being $x$ and on admissible background information, is $x$ itself. Wallace and Saunders claim that the DW theorem supplies, uniquely in physics, a derivation of the PP, with chances identified as branch weights, via cautious functionalism: chance is whatever plays the chance-slot in the PP. Earman (2018) argued, in a single-world algebraic setting, that the PP is a theorem of quantum probability theory given Gleason-type assumptions.

The source paper. Brown and Ben Porath (2020), published in the Pitowsky memorial volume, critically examines these claims. They distinguish Wallace's decision-theoretic weak non-contextuality (measurement neutrality: rational indifference between different physical realizations of the same abstract measurement) from strong Gleason-style non-contextuality across incompatible contexts, and argue the DW theorem establishes at most the former; the most plausible physical defense of strong non-contextuality, Timpson's (2011) dynamical argument from the unitary measurement process, presupposes that the standard account of that process is independent of the Born rule. They argue that the Wallace-Saunders derivation of the PP presupposes the principle, merely filling its chance-slot with branch weights, so that branch weights are at best functionally or emergently chancy rather than agent-independent chances. They press the methodological problem of deviant branches: branching guarantees observers whose statistics violate any pre-ordained statistical test of the Born rule, so the Popper-Gillies-Deutsch falsifiability rule (a theory whose assigned probabilities clash with observed frequencies thereby has a scientific problem) appears to force Everettian observers in such branches to question their theory — a tension sharpened by the redundancy challenge of Dawid and Thébault (2014), endorsed with a caveat by Read (2018). Finally, they quote Wallace's admission that the DW theorem is silent on what to do when the quantum state is not known, or when the agent is uncertain about whether quantum mechanics is true; Wallace's unified approach and the Greaves-Myrvold (2010) evidential theory of branching worlds are the standing, debated attempts to cover such epistemically non-ideal agents.

Problem Statement

Overarching objective: determine whether the Born rule and the normative link between branch weights and rational credences — the Principal Principle (PP), or its Everettian analogue — can be derived from the explicitly non-probabilistic postulates of Everettian quantum mechanics alone; if not, characterize the minimal probabilistic or normative input that any such derivation requires — minimal in the sense of logical strength (the weakest such input, i.e. one entailed by every other probabilistic or normative input sufficient for the derivation, or, if no unique weakest input exists, a characterization of the minimal alternatives). As left open by Brown and Ben Porath (2020), the problem has five linked components, all serving this single objective:

(i) Non-contextuality: provide a non-question-begging, purely physical (unitary/dynamical) derivation of the strong, Gleason-style non-contextuality condition — that the probability assigned to a projector is independent of which incompatible measurement contexts embed it — without implicitly invoking probabilities; or show that no such derivation is possible within EQM. (Timpson's dynamical argument is the standing candidate, but it presupposes that the standard account of the measurement process is independent of the Born rule.)

(ii) Chances: either give a fully theory-internal, non-decision-theoretic account on which branch weights are bona fide agent-independent chances from which the PP follows, or establish that only a functional or emergent notion of chance is available to the Everettian — and then say precisely whether the PP is a theorem, a definition, or a postulate of the theory.

(iii) Deviant branches and refutability: formulate a non-ad-hoc account of statistical refutation for Everettian agents — one that preserves standard refutability in non-deviant branches and gives a principled verdict for observers in branches whose statistics deviate from Born weights — thereby settling whether the Deutsch-Wallace theorem is redundant (as the Dawid-Thébault/Read line urges and the source's considerations reinforce) or essential (as March 2024 argues).

(iv) Epistemically non-ideal agents: extend the decision-theoretic derivation of the Born rule to agents uncertain about the state preparation or about whether quantum mechanics is true, and determine whether the PP or its analogue holds for them.

(v) Gleason-route gaps: close or delimit the technical gaps of Gleason-based derivations — the failure for qubits ($D<3$) under projective measurements and the reliance on countable rather than finite additivity — or prove that no derivation in the stated framework avoids them.

Partial progress exists on (iv) and (v) (Wallace's unified approach; Greaves-Myrvold's evidential theory; Busch's and Wright-Weigert's qubit Gleason-type theorems for wider measurement classes) and candidate theories exist on (ii) (Saunders' branch-counting; the 2025-2026 self-location, physical-probability and future-record models), but no strand is closed, and (i)-(iii) remain actively contested as of 2026. An answer may take the form of a derivation, an impossibility or non-uniqueness result, a methodological account, or a combination; any such answer must address all five components (i)-(v), which are the constituent parts of the single objective rather than five independent tasks — an answer that leaves any of them unaddressed does not resolve the task. What resolves the task is a definite verdict on that objective — whether the Born rule and the PP (or its Everettian analogue) are derivable from the non-probabilistic postulates alone and, if not, which minimal probabilistic or normative input is required — argued across those components; a component counts as addressed by a partial result, by a delimitation, or by an argument that it does not bear on the verdict, and need not be closed outright for the verdict to stand.

References

1. Harvey R. Brown and Gal Ben Porath, 'Everettian probabilities, the Deutsch-Wallace theorem and the Principal Principle', in M. Hemmo and O. Shenker (eds.), Quantum, Probability, Logic: The Work and Influence of Itamar Pitowsky, Jerusalem Studies in Philosophy and History of Science (Springer, Cham, 2020), pp. 165-198. DOI: 10.1007/978-3-030-34316-3_7, https://doi.org/10.1007/978-3-030-34316-3_7; preprint arXiv:2010.11591, https://arxiv.org/abs/2010.11591
2. David Deutsch, 'Quantum theory of probability and decisions', Proceedings of the Royal Society of London A 455, 3129-3137 (1999). DOI: 10.1098/rspa.1999.0443, https://doi.org/10.1098/rspa.1999.0443
3. David Wallace, The Emergent Multiverse: Quantum Theory according to the Everett Interpretation (Oxford University Press, 2012). DOI: 10.1093/acprof:oso/9780199546961.001.0001, https://doi.org/10.1093/acprof:oso/9780199546961.001.0001
4. Andrew M. Gleason, 'Measures on the closed subspaces of a Hilbert space', Journal of Mathematics and Mechanics 6, 885-894 (1957). DOI: 10.1512/iumj.1957.6.56050, https://doi.org/10.1512/iumj.1957.6.56050
5. David Lewis, 'A Subjectivist's Guide to Objective Chance', in R. C. Jeffrey (ed.), Studies in Inductive Logic and Probability, Volume 2 (University of California Press, 1980), pp. 263-294. DOI: 10.1525/9780520318328-009, https://doi.org/10.1525/9780520318328-009
6. Meir Hemmo and Itamar Pitowsky, 'Quantum probability and many worlds', Studies in History and Philosophy of Modern Physics 38, 333-350 (2007). DOI: 10.1016/j.shpsb.2006.04.005, https://doi.org/10.1016/j.shpsb.2006.04.005
7. Hilary Greaves and Wayne Myrvold, 'Everett and Evidence', in S. Saunders, J. Barrett, A. Kent and D. Wallace (eds.), Many Worlds? Everett, Quantum Theory, and Reality (Oxford University Press, 2010), pp. 264-304. DOI: 10.1093/acprof:oso/9780199560561.003.0011, https://doi.org/10.1093/acprof:oso/9780199560561.003.0011
8. Christopher G. Timpson, 'Probabilities in Realist Views of Quantum Mechanics', in C. Beisbart and S. Hartmann (eds.), Probabilities in Physics (Oxford University Press, 2011), pp. 201-230. DOI: 10.1093/acprof:oso/9780199577439.003.0008, https://doi.org/10.1093/acprof:oso/9780199577439.003.0008
9. P. Busch, 'Quantum States and Generalized Observables: A Simple Proof of Gleason's Theorem', Physical Review Letters 91, 120403 (2003). DOI: 10.1103/PhysRevLett.91.120403, https://doi.org/10.1103/PhysRevLett.91.120403
10. Victoria J. Wright and Stefan Weigert, 'A Gleason-type theorem for qubits based on mixtures of projective measurements', Journal of Physics A: Mathematical and Theoretical 52, 055301 (2019). DOI: 10.1088/1751-8121/aaf93d, https://doi.org/10.1088/1751-8121/aaf93d
11. Richard Dawid and Karim P. Y. Thébault, 'Against the empirical viability of the Deutsch-Wallace-Everett approach to quantum mechanics', Studies in History and Philosophy of Modern Physics 47, 55-61 (2014). DOI: 10.1016/j.shpsb.2014.05.005, https://doi.org/10.1016/j.shpsb.2014.05.005
12. James Read, 'In defence of Everettian decision theory', Studies in History and Philosophy of Modern Physics 63, 136-140 (2018). DOI: 10.1016/j.shpsb.2018.01.005, https://doi.org/10.1016/j.shpsb.2018.01.005
13. Charles T. Sebens and Sean M. Carroll, 'Self-locating Uncertainty and the Origin of Probability in Everettian Quantum Mechanics', The British Journal for the Philosophy of Science 69(1), 25-74 (2018). DOI: 10.1093/bjps/axw004, https://doi.org/10.1093/bjps/axw004; preprint arXiv:1405.7577, https://arxiv.org/abs/1405.7577
14. Adrian Kent, 'Does it Make Sense to Speak of Self-Locating Uncertainty in the Universal Wave Function? Remarks on Sebens and Carroll', Foundations of Physics 45(2), 211-217 (2015). DOI: 10.1007/s10701-014-9862-5, https://doi.org/10.1007/s10701-014-9862-5
15. John Earman, 'The Relation between Credence and Chance: Lewis's Principal Principle Is a Theorem of Quantum Probability Theory', preprint, PhilSci-Archive 14822 (2018), http://philsci-archive.pitt.edu/14822/
16. Paul Tappenden, 'Everettian theory as pure wave mechanics plus a no-collapse probability postulate', Synthese 198, 6375-6402 (published online 2019; print issue 2021). DOI: 10.1007/s11229-019-02467-4, https://doi.org/10.1007/s11229-019-02467-4
17. Simon Saunders, 'The Everett Interpretation: Probability', in E. Knox and A. Wilson (eds.), The Routledge Companion to Philosophy of Physics (Routledge, 2021), pp. 230-246. DOI: 10.4324/9781315623818-21, https://doi.org/10.4324/9781315623818-21; preprint arXiv:2103.03966, https://arxiv.org/abs/2103.03966
18. Simon Saunders, 'Branch-counting in the Everett interpretation of quantum mechanics', Proceedings of the Royal Society A 477, 20210600 (2021). DOI: 10.1098/rspa.2021.0600, https://doi.org/10.1098/rspa.2021.0600; preprint arXiv:2201.06087, https://arxiv.org/abs/2201.06087
19. Eleanor March, 'Is the Deutsch-Wallace Theorem Redundant?', Philosophy of Physics 2(1) (LSE Press, 2024). DOI: 10.31389/pop.65, https://doi.org/10.31389/pop.65
20. Richard Dawid and Karim P. Y. Thébault, 'Decoherence and Probability', Philosophy of Science 93(1), 143-169 (published online 2025; print issue 2026). DOI: 10.1017/psa.2025.10110, https://doi.org/10.1017/psa.2025.10110; preprint arXiv:2410.01317, https://arxiv.org/abs/2410.01317
21. Simon Saunders, 'Physical Probability in the Everett Interpretation, and Bell Inequalities', in Local Quantum Mechanics (Oxford University Press, 2026), pp. 352-382. DOI: 10.1093/9780197792766.003.0014, https://doi.org/10.1093/9780197792766.003.0014; preprint arXiv:2512.00575, https://arxiv.org/abs/2512.00575
22. Lev Vaidman, 'Probability of Self-Location in the Framework of the Many-Worlds Interpretation', Entropy 27(4), 416 (2025). DOI: 10.3390/e27040416, https://doi.org/10.3390/e27040416
23. Jiaxuan Zhang, 'Summing to Uncertainty: On the Necessity of Additivity in Deriving the Born Rule' (preprint 2026), arXiv:2603.06211, https://arxiv.org/abs/2603.06211
24. Jonathan Baxter, 'The Future-Record Model of Everettian Probability' (preprint 2026), arXiv:2609.06130, https://arxiv.org/abs/2609.06130
25. Avi Levy and Meir Hemmo, 'Locality and probability in relativistic quantum theories and hidden variables quantum theories', Foundations of Physics 55, article 73 (2025). DOI: 10.1007/s10701-025-00885-8, https://doi.org/10.1007/s10701-025-00885-8; preprint arXiv:2503.12967, https://arxiv.org/abs/2503.12967
26. Olga Nánásiová, Karla Čipková and Michal Zákopčan, 'Problem of Existence of Joint Distribution on Quantum Logic', Entropy 26(12), 1212 (2024). DOI: 10.3390/e26121121, https://doi.org/10.3390/e26121121