| Bell inequality | |
|---|---|
| Name | Bell inequality |
| Caption | Schematic of an entanglement test using space-like separated measurements |
| Field | Quantum mechanics |
| Introduced | 1964 |
| Introduced by | John Stewart Bell |
| Related | Bell test experiments; CHSH inequality; Local realism |
Bell inequality
The Bell inequality refers to a family of mathematical inequalities derived by John Stewart Bell in 1964 that constrain statistical correlations predicted by any theory obeying local realism and certain classical assumptions. Violations of these inequalities by experimental data confirm that quantum mechanical predictions for entanglement cannot be reproduced by any local hidden-variable model, with profound consequences for foundations of quantum mechanics and emerging technologies such as quantum information and quantum cryptography.
Bell inequalities originated from attempts to formalize and test debates arising from the 1935 EPR paper by Albert Einstein, Boris Podolsky, and Nathan Rosen about the completeness of quantum mechanics. Building on work by David Bohm on hidden-variable models, Bell produced a theorem showing that certain statistical correlations predicted by quantum theory contradict the constraints satisfied by any local hidden-variable theory. Bell's 1964 paper "On the Einstein Podolsky Rosen paradox" led to a shift from philosophical discussion to experimentally testable statements. Subsequent theoretical contributions such as the CHSH inequality (Clauser, Horne, Shimony, and Holt) and the Clauser–Horne inequality refined testable forms. Empirical advances by experimentalists including John Clauser, Alain Aspect, Anton Zeilinger, and others culminated in progressively loophole-closing Bell test experiments spanning laboratories at institutions like University of California, Berkeley, Université de Paris-Sud, Austrian Academy of Sciences laboratories, and corporate research groups.
Bell inequalities express bounds on correlators or joint probabilities computed under assumptions of locality and realism. The simplest and most cited form is the CHSH inequality, which involves two observers (commonly Alice and Bob) each choosing between two measurement settings; classically the CHSH correlator S satisfies |S| ≤ 2, while quantum mechanics permits up to 2√2 (the Tsirelson bound). Other formulations include the original Bell (1964) inequality, the Clauser–Horne (CH) probability inequality, and multipartite inequalities such as the Mermin inequality and Svetlichny inequality for three or more parties. There are also entropic Bell inequalities based on Shannon entropy and inequalities tailored for continuous-variable systems, e.g., those using quadrature correlations relevant to quantum optics. In quantum information theory, Bell inequalities are linked to device-independent protocols where the violation quantifies certifiable randomness or key rates in device-independent quantum cryptography.
Derivations of Bell inequalities rest on specific assumptions: realism (measurement outcomes are determined by pre-existing properties or hidden variables), locality (no superluminal causal influences), and statistical independence or freedom of choice (measurement settings are uncorrelated with hidden variables). Alternative formulations may replace realism with determinism or counterfactual definiteness. Violations of Bell inequalities imply that at least one assumption must be abandoned; interpretations diverge on whether to reject locality (nonlocality), realism, or the freedom assumption. Hidden-variable models that reproduce quantum predictions while relaxing locality include de Broglie–Bohm theory; other approaches examine superdeterminism or retrocausal models. The theorem connects to broader conceptual frameworks including Bell's theorem, Kochen–Specker theorem, and debates in philosophy of physics about causality and ontology.
Starting with pioneering tests by Freedman and Clauser and notable experiments by Aspect, Grangier and Roger in the 1980s, Bell tests have progressively closed major experimental loopholes: the detection (fair-sampling) loophole, the locality (communication) loophole, and the freedom-of-choice loophole. Landmark "loophole-free" experiments around 2015 by groups at Delft University of Technology (Hensen et al.), NIST and Vienna (using trapped ions, superconducting systems, and photons) reported simultaneous closure of the major loopholes. Experimental platforms include entangled photons produced by spontaneous parametric down-conversion, trapped ions, nitrogen-vacancy centers in diamond, superconducting qubits, and atomic ensembles. Practical challenges remain: high-efficiency detection, fast random-setting generation, and space-like separation to enforce locality constraints.
Violation of Bell inequalities is a definitive operational signature of nonclassical correlations stronger than those explained by separable states. While entanglement is necessary for many Bell violations, not all entangled states violate a Bell inequality under every measurement scenario; concepts such as hidden nonlocality and activation show subtle structure. In quantum information theory, Bell-inequality violations underlie device-independent protocols for quantum key distribution (DI-QKD), randomness expansion, and self-testing of quantum devices. Quantitative measures such as nonlocality robustness relate to entanglement measures (e.g., entanglement entropy) and resource-theoretic treatments. Connections also exist to computational complexity in tasks like nonlocal games (e.g., the CHSH game) and separation results between classical and quantum communication.
Research extends Bell inequalities to multipartite systems, high-dimensional (qudit) settings, continuous variables, and network scenarios (e.g., bilocality inequalities). Generalized forms include device-independent entropic inequalities and inequalities tailored for asymmetric detection efficiencies. Measures of nonlocality quantify the degree of Bell violation and include nonlocal fraction, robustness to noise, and communication cost to simulate quantum correlations. The study of Bell inequalities also informs quantum foundations via concepts like information causality and macroscopic locality, and interfaces with fields such as quantum gravity when probing nonlocal features of proposed fundamental theories. Ongoing theoretical work explores the space of quantum correlations (the quantum set) compared with classical and no-signalling sets, using tools from convex optimization and operator algebras.