What These Experiments Do and Do Not Prove

Foundations experiments are powerful because they turn philosophical pressure into precise constraints. They do not merely say that quantum mechanics is strange. They identify classes of classical-looking explanations that fail, and they show where quantum theory survives demanding laboratory tests.

The same experiments are often overinterpreted. Bell tests do not enable faster-than-light signaling. Contextuality tests do not ban every hidden-variable theory. Delayed-choice and quantum eraser experiments do not prove that consciousness creates reality. A trustworthy reading keeps the theorem, the experiment, and the interpretation distinct.

For the main experimental pages, see Bell Inequality Experiments, Loophole-Free Bell Tests, Kochen–Specker Contextuality Experiments, Delayed-Choice Experiments, and Quantum Eraser Experiments.

Experiments Constrain Classes of Theories

The key phrase is “classes of theories.” A no-go theorem starts from assumptions and proves that any model satisfying those assumptions must obey a constraint. An experiment then tests whether nature obeys the constraint in a real implementation.

For Bell experiments, the tested local hidden-variable structure has the form

$$P(A,B\mid x,y) = \int d\lambda\, \rho(\lambda)\, P(A\mid x,\lambda)\, P(B\mid y,\lambda),$$

with suitable independence assumptions. Bell inequalities follow from structures of this kind. Quantum mechanics predicts violations for suitable entangled states and settings, and experiments observe such violations.

The conclusion is not “all hidden variables are impossible.” The conclusion is narrower and stronger:

Local hidden-variable explanations satisfying the tested Bell assumptions fail.

For Kochen–Specker contextuality, the excluded class is different. The target is a noncontextual value assignment to projective measurements, not a spacelike-local model of separated systems. For delayed-choice and quantum eraser experiments, the target is again different: naive stories in which systems carry fixed classical wave-or-particle labels independent of the final measurement context.

The experiments therefore form a map of constraints, not one single slogan.

Against Naive Realism

The experiments make trouble for a naive picture in which every measurable quantity has a pre-existing value that is merely revealed by measurement, independent of the measurement arrangement.

Double-slit, delayed-choice, and quantum eraser experiments show that interference depends on which alternatives remain coherent and which measurement context is realized. A path record, even an unread physical record, can suppress interference. An erasing measurement can recover phase-sensitive correlations only in conditioned subensembles. This is not how ordinary ignorance about a hidden classical path behaves.

Kochen–Specker-type results sharpen the point. In Hilbert-space dimension at least three, one cannot assign values to all projectors while preserving the functional relations among compatible projective measurements and keeping those values independent of context.

The careful lesson is not that reality is unreal. The lesson is that classical value-attribution rules fail when applied globally to quantum observables. Any realist interpretation that remains viable must use a subtler structure than “all observables had ordinary context-independent values all along.”

Against Local Hidden Variables

Bell experiments give the clearest empirical constraint in the foundations story. Under the Bell assumptions, correlations between separated outcomes must obey Bell inequalities such as CHSH:

$$\lvert S\rvert\le2.$$

Quantum mechanics predicts, and experiments observe, violations of such inequalities in suitable scenarios. Modern loophole-free tests close the major detection and locality loopholes under explicit assumptions about setting independence, event definitions, and statistical analysis.

The responsible statement is:

Observed Bell violations rule out local hidden-variable models satisfying
the assumptions used to derive the tested inequality.

This is already a profound result. It means the EPR hope for a local classical completion of quantum mechanics does not survive the experimentally tested Bell framework.

But it is not the same as saying:

• no hidden-variable theory of any kind can exist;
• determinism alone is excluded;
• nonlocal interpretations are excluded;
• every possible freedom-of-choice loophole has been logically eliminated;
• relativity has been experimentally violated in the sense of usable superluminal signals.

Bohmian mechanics, for example, is explicitly nonlocal. Superdeterministic models reject setting independence. Objective-collapse models modify dynamics. Different interpretations pay different prices, but Bell experiments do not select one winner by themselves.

Against Noncontextual Hidden Variables

Kochen–Specker contextuality targets a different classical intuition. It asks whether measurement outcomes can be understood as revealing pre-existing values that are independent of which compatible observables are measured alongside them.

The theorem says that in dimension at least three, sharp quantum measurements cannot be assigned noncontextual definite values in a way that respects the relevant functional relations. Experiments test operational versions of this idea using compatible measurements, contextuality inequalities, and carefully specified assumptions about measurement compatibility.

The responsible statement is:

Contextuality experiments rule out tested noncontextual models under
the implemented compatibility and measurement assumptions.

They do not rule out contextual hidden-variable theories. They also do not mean merely that the apparatus mechanically disturbs the system. The issue is deeper: the value assigned to a measurement outcome cannot be treated as independent of the measurement context in the way the noncontextual model requires.

No Faster-Than-Light Signaling

Bell violations are sometimes called “nonlocal,” but nonlocal correlations are not the same as controllable faster-than-light communication.

In an operational no-signaling theory, Alice’s local outcome statistics do not depend on Bob’s remote setting:

$$\sum_B P(A,B\mid x,y) = P(A\mid x),$$

independent of $y$. Quantum mechanics satisfies this condition for ordinary local measurements on separated systems.

Bell experiments reveal correlations after Alice and Bob compare their records through an ordinary classical channel. Before comparison, each side sees local statistics that cannot be used to infer the distant setting. The same point appears in quantum eraser experiments: the unsorted signal data do not reveal which marker measurement basis was chosen.

This distinction is essential. The experiments force us away from Bell-local hidden-variable explanations, but they do not provide a communication channel outside the light cone.

No Consciousness-Caused Collapse Proof

Delayed-choice and quantum eraser experiments are often described in language that makes the observer sound magical. That is not what the experiments require.

Which-path information can be carried by a physical marker: polarization, recoil, an idler photon, an internal state, or an environmental degree of freedom. Interference can disappear because alternatives become entangled with distinguishable marker states, even if no person reads the marker. Conversely, conditional interference can be recovered when marker outcomes are sorted in an erasing basis, again without invoking consciousness as a dynamical ingredient.

The measurement problem remains a real conceptual problem, but these experiments do not solve it by proving that minds collapse wavefunctions. They show that physical measurement context, entanglement, distinguishability, and conditioning matter.

For a careful statement of what measurement formalism leaves open, see What Measurement Formalism Does Not Settle and Measurement Problem as Historical Problem.

Theorem, Experiment, Interpretation

Many confusions disappear when three layers are kept separate.

First, a theorem proves a conditional mathematical result. Bell theorem, CHSH, and Kochen–Specker theorem each specify assumptions and derive a constraint or contradiction.

Second, an experiment implements a physical scenario and tests an observable consequence. Real experiments must address detector efficiency, timing, setting choices, compatibility, finite statistics, source quality, and data-analysis rules.

Third, an interpretation says what the formalism means. Interpretations differ over ontology, state update, probability, nonlocality, and the status of measurement outcomes. They must respect the experimental constraints, but the constraints do not normally determine a unique interpretation.

This separation is not pedantry. It prevents overclaiming in both directions. Skeptics should not dismiss successful experiments by attacking a different theorem. Enthusiasts should not turn a precise violation into a sweeping metaphysical slogan.

Interpretations Still Open

The experimental record strongly constrains acceptable interpretations, but it does not close the interpretive landscape.

Any viable interpretation must account for:

• Bell inequality violations without usable faster-than-light signaling;
• contextuality constraints on value assignments;
• interference, loss of interference, and conditional recovery of interference;
• the Born rule and state-update practice;
• the emergence of stable records in macroscopic apparatus;
• agreement with ordinary laboratory quantum mechanics.

Different interpretations meet these constraints differently. Copenhagen-like approaches emphasize measurement context and the limits of classical description. Everettian approaches keep unitary evolution and interpret outcomes through branching structure. Bohmian mechanics uses definite configurations with nonlocal dynamics. Objective-collapse theories modify the dynamics and are constrained by precision experiments. Informational and operational approaches treat the quantum state less as a microscopic material wave and more as a tool for organizing probabilities.

This page does not adjudicate among them. Its point is more basic: foundations experiments rule out important simple pictures and force any remaining picture to be explicit about its assumptions.

A Responsible Reading Checklist

When encountering a strong claim about a foundations experiment, ask:

• Which theorem or inequality is being invoked?
• Which assumptions are used in the theorem?
• Which physical loopholes does the experiment close?
• Which assumptions remain?
• Are local marginal statistics being confused with joint correlations?
• Is postselection or coincidence sorting essential?
• Is the claim about data, theorem, or interpretation?
• Does the statement imply more than the cited experiment actually tested?

This checklist makes the experiments more impressive, not less. Their strength comes from precision.

Cross-Links

• Foundations Experiments and Quantum Reality
• EPR Argument
• Bell Inequality Experiments
• Loophole-Free Bell Tests
• Kochen–Specker Contextuality Experiments
• Delayed-Choice Experiments
• Quantum Eraser Experiments
• Bell Theorem
• CHSH Inequality
• Kochen–Specker Theorem
• Entanglement in Foundations
• Local Measurement Statistics
• What Measurement Formalism Does Not Settle
• Measurement Problem as Historical Problem
• Common Historical Misconceptions

References

• J. S. Bell, “On the Einstein Podolsky Rosen Paradox,” Physics 1, 195-200, 1964, DOI: 10.1103/PhysicsPhysiqueFizika.1.195.
• S. Kochen and E. P. Specker, “The problem of hidden variables in quantum mechanics,” Journal of Mathematics and Mechanics 17, 59-87, 1967.
• N. D. Mermin, “Hidden variables and the two theorems of John Bell,” Reviews of Modern Physics 65, 803-815, 1993, DOI: 10.1103/RevModPhys.65.803.
• A. Aspect, J. Dalibard, and G. Roger, “Experimental Test of Bell’s Inequalities Using Time-Varying Analyzers,” Physical Review Letters 49, 1804-1807, 1982, DOI: 10.1103/PhysRevLett.49.1804.
• B. Hensen et al., “Loophole-free Bell inequality violation using electron spins separated by 1.3 kilometres,” Nature 526, 682-686, 2015, DOI: 10.1038/nature15759.
• N. Brunner, D. Cavalcanti, S. Pironio, V. Scarani, and S. Wehner, “Bell nonlocality,” Reviews of Modern Physics 86, 419-478, 2014, DOI: 10.1103/RevModPhys.86.419.
• X.-S. Ma, J. Kofler, and A. Zeilinger, “Delayed-choice gedanken experiments and their realizations,” Reviews of Modern Physics 88, 015005, 2016, DOI: 10.1103/RevModPhys.88.015005.
• M. Schlosshauer, “Decoherence, the Measurement Problem, and Interpretations of Quantum Mechanics,” Reviews of Modern Physics 76, 1267-1305, 2005, DOI: 10.1103/RevModPhys.76.1267.
• A. Peres, Quantum Theory: Concepts and Methods, Kluwer, 1995.
• J. S. Bell, Speakable and Unspeakable in Quantum Mechanics, 2nd ed., Cambridge University Press, 2004.

Exercises

1. A popular article says, “Bell tests prove that faster-than-light communication is real.” What is wrong with that statement?

Solution

Bell tests show violations of inequalities obeyed by local hidden-variable models under stated assumptions. The violations appear in joint correlations after the two sides compare their data. Each side’s local marginal statistics remain independent of the remote setting, so the experiment does not provide a controllable faster-than-light communication channel.

2. Explain why “Bell theorem disproves determinism” is too broad.

Solution

Bell theorem rules out local hidden-variable models satisfying the Bell assumptions, including locality and setting independence. Determinism by itself is not the target. A deterministic theory that is explicitly nonlocal, such as Bohmian mechanics, is not excluded by Bell theorem in the same way a Bell-local deterministic theory is.

3. In one sentence, distinguish a theorem from an experiment in this chapter.

Solution

A theorem derives a constraint from specified assumptions, while an experiment tests whether real measurement statistics obey or violate that constraint under a concrete implementation with stated loophole controls.

4. Why does a quantum eraser not prove that consciousness creates collapse?

Solution

The relevant ingredient is physical distinguishability of marker states and conditional sorting of joint data. A marker can carry which-path information without being read by a person, and eraser interference appears only in conditioned subensembles. No consciousness-dependent dynamical rule is needed to explain the observed statistics.