A new paper in Foundations of Physics argues that quantum theory does not need an external guarantee that different observers will always agree on measurement outcomes. In “Intersubjective Agreement about Measurement Outcomes is Unnecessary in QBism,” philosophers Gino Elia, Jennifer Carter, and Robert Crease examine one of quantum mechanics’ most unsettling thought experiments—Wigner’s Friend—and use it to defend a distinctive interpretation of quantum theory known as QBism. Their central claim is provocative: objectivity in quantum mechanics may not depend on every observer possessing the same facts, quantum states, or measurement records. Instead, objectivity can emerge from the way agents use the quantum formalism while interacting with one another and with the physical world.
Wigner’s Friend was introduced by physicist Eugene Wigner in 1961. In the thought experiment, a friend inside a sealed laboratory measures a quantum system, such as a particle whose state can be described as a superposition of two alternatives. From the friend’s perspective, the measurement produces a definite result. Wigner, standing outside the laboratory and treating the entire lab—including the friend—as a quantum system, may instead assign a state in which the friend, the apparatus, and the measured particle remain mathematically entangled. The same physical episode is therefore represented differently by the two observers. Modern variations of this scenario have intensified the debate by asking whether quantum mechanics can consistently describe agents who use the theory on other agents.
The tension becomes especially sharp when the question is whether Wigner and the friend must eventually agree about what happened. Some interpretations attempt to preserve a single, observer-independent measurement outcome. Others accept that facts may be relative to an observer or a physical perspective. QBism, short for Quantum Bayesianism, takes a different route. It treats a quantum state not as a literal description of an observer-independent physical condition, but as an individual agent’s personal expectations about the consequences of possible actions on the world. Probabilities in this framework are Bayesian degrees of belief. When an agent performs a measurement, the resulting outcome is not merely the discovery of a pre-existing value; it is an experience generated through an encounter between the agent and a quantum system.
This does not mean that QBism reduces reality to private opinion or permits agents to believe anything they wish. The quantum formalism imposes strict mathematical constraints on the probabilities an agent may consistently assign. A quantum state is represented by a density operator, and the Born rule converts that state into probabilities for possible measurement outcomes. In QBist language, the Born rule is not simply a rule for calculating an objective distribution hidden inside nature. It is a normative rule that guides an agent in organizing expectations about future experiences. Two agents may begin with different information and assign different states, yet both remain accountable to the same quantum principles when making predictions.
The authors focus on what QBism calls reciprocity: Wigner and his friend are not detached spectators positioned outside one another’s descriptions. Each is a physical system capable of acting on the other. The friend acts on the measured system and receives an outcome; Wigner can later interact with the laboratory, the friend, or a record of the experiment. These interactions are not merely channels for transferring already completed facts. They are physical events that can alter what each agent is entitled to expect next. The formalism is therefore applied from a standpoint. Wigner’s state assignment concerns his anticipated experiences arising from actions on the laboratory, while the friend’s assignment concerns the friend’s own anticipated experiences. Neither description automatically replaces the other.
This approach also challenges the idea that quantum states should function as universally shared snapshots of reality. In classical physics, it is often natural to imagine that a system possesses a complete state that different observers can, in principle, discover and record. Quantum mechanics resists this picture. Even an agent with maximal information about a quantum system cannot generally predict every possible measurement outcome with certainty. Quantum theory contains an irreducible statistical residue: some future outcomes remain unpredictable, not merely because of practical ignorance but because of the structure of the theory itself. The paper presents this limitation as central to QBism’s account of measurement. A measurement outcome is personal in the sense that it is an experience for a particular agent, but it is not arbitrary, because it results from a constrained physical interaction.
Critics of QBism have argued that this position threatens the intersubjective basis of science. If measurement outcomes belong to individual agents, what prevents two observers from disagreeing indefinitely? Philosopher of physics Steven French has raised concerns that QBism may not guarantee agreement between Wigner and the friend, either about quantum states or about outcomes. The new paper responds that demanding such a guarantee may mistake the role of scientific objectivity. Agreement is practically valuable, the authors acknowledge, but an ultimate metaphysical mechanism ensuring convergence is not required for physics to function as an objective practice. Scientists communicate, compare records, calibrate instruments, and establish shared procedures. These activities create stable patterns of coordination without requiring the world to supply an observer-free certificate for every fact.
The argument draws heavily on phenomenology, a philosophical tradition concerned with how objects and experiences are constituted from particular perspectives. In this context, assigning a quantum state is described as a form of objectification. To assign a state is to treat something as a quantum system—to identify it as the kind of entity to which the formalism can meaningfully be applied. This process is neither a complete capture of the thing itself nor a purely subjective invention. The object exceeds the description. A quantum system can be modeled through a Hilbert-space state, measurement operators, and probabilities, but it is never exhausted by those mathematical representations. The formalism succeeds precisely because the world is richer than any one of its descriptions.
This phenomenological interpretation helps explain why quantum theory can be used consistently by agents occupying radically different positions. The theory does not need to change its fundamental rules whenever the system under discussion changes from an electron to a measuring device or to a human observer. Every system can, in principle, be treated as a physical participant in an interaction. At the same time, no state assignment becomes a complete, view-from-nowhere account of that system. Wigner may represent the friend as part of a larger quantum process, while the friend experiences a definite result within the laboratory. These descriptions can be related through later interactions, but they need not be identical in advance or collapse into a single universal narrative.
The implications extend beyond the philosophical puzzle of Wigner’s Friend. Contemporary no-go theorems and laboratory experiments inspired by observer-independence debates have shown that several intuitive assumptions cannot all be maintained simultaneously. Researchers must choose which commitments to preserve: universal facts, locality, freedom of measurement choice, or the unrestricted applicability of quantum theory to observers and laboratories. The authors of the new study do not present an experiment that proves QBism, nor do they claim to eliminate every interpretation of quantum mechanics. Instead, they offer a conceptual defense of the idea that objectivity can be grounded in shared practices, reciprocal physical interactions, and disciplined use of the quantum formalism rather than in identical private experiences.
That conclusion may sound less dramatic than the phrase “agreement is unnecessary,” but its consequences are substantial. The paper is not saying that communication, reproducibility, or scientific consensus are unimportant. It is arguing that these achievements should not be confused with a deeper requirement that all agents must possess the same quantum state or experience the same outcome from an external point of view. In QBism, science remains possible because agents are embodied participants in a common world, capable of acting on one another, learning from consequences, and revising their expectations. Objectivity is therefore not a frozen inventory of facts existing independently of every perspective. It is an ongoing achievement made possible by a formalism that respects both the individuality of experience and the resistance of the physical world.
Subject of Research: QBism, Wigner’s Friend, quantum measurement, phenomenology, and the philosophical foundations of intersubjective objectivity
Article Title: Intersubjective Agreement about Measurement Outcomes is Unnecessary in QBism
Article References: Elia, G., Carter, J. & Crease, R. “Intersubjective Agreement about Measurement Outcomes is Unnecessary in QBism.” Foundations of Physics 56, Article 45 (2026).
Image Credits: AI Generated
DOI: 10.1007/s10701-026-00947-5
Keywords: QBism; Phenomenology; Intersubjectivity; Wigner’s Friend; Quantum mechanics
