Bell’s Theorem as Historical Turning Point

Bell’s theorem changed the status of the EPR debate. Before Bell, the argument over completeness, locality, and hidden variables could seem interpretive: different views of the same successful quantum predictions. Bell showed that broad classes of local hidden-variable explanations obey inequalities that quantum mechanics can violate.

That shift made the foundations of quantum mechanics experimentally testable. The theorem is not merely a philosophical slogan, and Bell tests are not merely demonstrations of entanglement. Together they constrain a precise class of local explanations for quantum correlations.

This page is historical. The formal theorem statement and compact assumptions live in Bell Theorem and CHSH Inequality.

From EPR to Bell

The EPR argument claimed that quantum mechanics is incomplete if one accepts locality and the EPR reality criterion. Bohm’s spin version then reformulated the issue with a normalizable spin-singlet state. Bell’s 1964 paper asked whether local hidden variables could reproduce the spin correlations for every choice of analyzer directions. His inequalities showed a conflict with quantum predictions under locality and independence assumptions. The 1969 CHSH proposal supplied the four-setting form used below.

Local Hidden Variables

A hidden-variable model supplements the quantum state with additional variables, usually denoted $\lambda$, intended to complete the description of each run. In a Bell scenario, Alice and Bob choose measurement settings $x$ and $y$ and obtain outcomes $a$ and $b$.

A standard local hidden-variable model has the structure

$$P(a,b\mid x,y) = \int d\lambda\, \rho(\lambda)\, P(a\mid x,\lambda)\, P(b\mid y,\lambda).$$

Here $\rho(\lambda)$ is a normalized nonnegative distribution. Setting independence means $\rho(\lambda\mid x,y)=\rho(\lambda)$. Local factorization means that, once the common variables are specified, the joint outcome distribution is the product of local response distributions. This includes both independence from the distant setting and the absence of residual outcome correlations conditional on $\lambda$. Merely requiring each marginal to ignore the distant setting is weaker and does not imply a Bell inequality.

This is not the same as saying every realist view is ruled out by one line of algebra. Different interpretations reject or modify different assumptions. Bell’s theorem is powerful precisely because it states what assumptions lead to inequalities and what experimental violations exclude.

Bell Inequalities

The 1969 CHSH form provides a particularly simple experimentally usable version. Alice chooses between settings $a$ and $a'$, Bob chooses between $b$ and $b'$, and each outcome is encoded as $\pm1$. Define the correlation

$$E(x,y) = \langle A_xB_y\rangle.$$

The CHSH combination is

$$S = E(a,b) + E(a,b') + E(a',b) - E(a',b').$$

Local hidden-variable models satisfying the CHSH assumptions obey

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

Quantum mechanics predicts larger values for suitable entangled states and measurement choices. For the spin singlet,

$$E(\mathbf a,\mathbf b) = -\mathbf a\cdot\mathbf b,$$

and suitable directions give

$$\lvert S\rvert=2\sqrt2.$$

This violation is not a failure of ordinary probability arithmetic inside quantum mechanics. It is a failure of the local hidden-variable factorization to reproduce quantum correlations.

Experimental Testability

Bell’s theorem made it possible to ask nature a sharper question. An experiment can prepare correlated systems, choose measurement settings on separated sides, estimate correlations, and compare them with a Bell inequality.

The historical experimental arc includes:

• the CHSH proposal, which put Bell’s idea into a form suited to real correlation measurements;
• early optical and atomic-cascade tests;
• Aspect’s time-varying analyzer experiments in the early 1980s;
• modern loophole-free tests using photons, separated atoms, superconducting circuits, and solid-state spins;
• device-independent quantum-information protocols that use Bell violation as an operational resource.

Each step required careful attention to assumptions. Detection efficiency, locality of setting choices, source timing, statistical analysis, and independence assumptions all matter. The experimental pages in this chapter separate those issues from the theorem itself.

For the experimental survey, see Bell Inequality Experiments.

A violation excludes the local models satisfying the stated independence assumptions. It does not select a unique interpretation or imply controllable faster-than-light communication. The distinction between entanglement, Bell violation, and local signaling is developed in the exercises below.

Exercises

1. For the singlet choose $\mathbf a=\hat{\mathbf z}$, $\mathbf a'=\hat{\mathbf x}$, $\mathbf b=(\hat{\mathbf z}+\hat{\mathbf x})/\sqrt2$, and $\mathbf b'=(\hat{\mathbf z}-\hat{\mathbf x})/\sqrt2$. Calculate $S$.

Solution

The first three correlations are $-1/\sqrt2$, whereas $E(a',b')=+1/\sqrt2$. Thus $S=-4/\sqrt2=-2\sqrt2$. Reversing an output convention can change the sign, but $\lvert S\rvert$ exceeds the local bound $2$.

2. What assumption is represented by the factorization $P(a,b\mid x,y,\lambda)=P(a\mid x,\lambda)P(b\mid y,\lambda)$?

Solution

It represents local factorization: conditional on $\lambda$, the outcomes are independent and each response depends only on its local setting. Independence of $\rho(\lambda)$ from the settings is an additional assumption needed for the displayed Bell model.

3. Why is Bell violation stronger than simply observing entanglement?

Solution

Entanglement means a state is not separable across a subsystem split. Bell violation means that observed correlations cannot be reproduced by local hidden-variable models satisfying the Bell assumptions. Some entangled states do not violate a particular Bell inequality, so Bell violation is a stronger operational property.

4. Why does Bell violation not imply faster-than-light signaling?

Solution

Bell violation concerns joint correlations collected after comparing data from both sides. Quantum mechanics preserves local marginal statistics: one party cannot choose a remote outcome or setting to control the other party’s unconditioned local distribution. Classical communication is still required to compare the correlations.

References

• 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.
• J. S. Bell, “On the Einstein Podolsky Rosen Paradox,” Physics 1, 195-200, 1964, DOI: 10.1103/PhysicsPhysiqueFizika.1.195.
• J. S. Bell, Speakable and Unspeakable in Quantum Mechanics, 2nd ed., Cambridge University Press, 2004.
• D. Bohm and Y. Aharonov, “Discussion of Experimental Proof for the Paradox of Einstein, Rosen, and Podolsky,” Physical Review 108, 1070-1076, 1957, DOI: 10.1103/PhysRev.108.1070.
• 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.
• J. F. Clauser, M. A. Horne, A. Shimony, and R. A. Holt, “Proposed Experiment to Test Local Hidden-Variable Theories,” Physical Review Letters 23, 880-884, 1969, DOI: 10.1103/PhysRevLett.23.880.
• A. Einstein, B. Podolsky, and N. Rosen, “Can Quantum-Mechanical Description of Physical Reality Be Considered Complete?,” Physical Review 47, 777-780, 1935, DOI: 10.1103/PhysRev.47.777.