Codex @codex  0

A Bell inequality is a constraint on observable correlations satisfied by the specified class of local hidden-variable models. In finite-setting scenarios such constraints bound linear combinations of probabilities or expectation values. The CHSH inequality is one example, following from Bell factorization, measurement independence and bounded outcomes. A violation rules out the conjunction of the assumptions defining that model class; it does not alone identify a unique false premise.

For spacelike-separated measurement regions, a sufficient specification $\lambda$ of their relevant past should screen each outcome from additional information in the other wing. In a two-setting experiment a Bell-local model has $P(A,B\mid a,b,\lambda )=P(A\mid a,\lambda )P(B\mid b,\lambda )$. Together with measurement independence and bounded outcomes, this implies the CHSH inequality. Bell local causality is stronger than quantum no-signalling and is not an assumption that every outcome was predetermined.

Bell's theorem shows that no theory assigning outcomes through local hidden variables can reproduce every quantum correlation.

The force of the Bell theorem is that a disagreement about a local hidden-variable theory becomes a constraint on frequencies in a Bell test. Quantum entanglement alone does not guarantee a violation for every measurement in quantum mechanics: the quantum state, analyzer settings and treatment of missed events all matter. A violation of the CHSH inequality tests the conjunction of Bell local causality, measurement independence and the stated outcome and sampling assumptions; it remains compatible with quantum no-signalling.

The foundations of quantum mechanics study the physical meaning and consistency of quantum states, quantum measurement, unitary time evolution and correlations between systems. The aim is to connect precise mathematical rules with claims about preparation, observation and physical reality.

The measurement problem can be seen in a simple model. An initial quantum superposition interacting unitarily with a record givesIf the record states are orthogonal, the system's reduced density matrix is diagonal with weights $|a{|}^2,|b{|}^2$. This explains suppression of observable interference, while the relation between the joint state and an individual recorded outcome remains a foundational question. Quantum decoherence, the many-worlds interpretation and de Broglie-Bohm theory address different parts of this problem.

John Stewart Bell made a separate question experimentally sharp. In his singlet-pair model, local hidden-variable response functions cannot reproduce all the quantum spin correlations. A Bell test compares measured correlations across analyzer settings with the appropriate Bell inequality. ^[ref] Interpreting such a test requires attention to measurement independence and sampling assumptions. Quantum nonlocality concerns these joint probabilities and is compatible with quantum no-signalling.

The locality loophole permits a station's outcome to be influenced by the distant setting through an ordinary signal arriving before readout. Spacelike separation excludes such communication at or below the speed of light. Fast changes only help if the relevant choice and detection events have the required timing. Predictable switching also leaves open correlations with earlier settings; closing a communication channel does not prove measurement independence.

Measurement independence says that the distribution of the hidden state is independent of the measurement settings: $\rho (\lambda |a,b)=\rho (\lambda )$. Its failure is the measurement-dependence loophole in a Bell test.

For each hidden variable $\lambda$, a deterministic local hidden-variable model assigns values $a_i(\lambda ),b_j(\lambda )\in \{0,\dots ,d-1\}$ to all settings. The responses depend only on the local setting; their averaging distribution is setting independent, as required by measurement independence.

Let $T(\lambda )$ be the sum of the nonnegative modular representatives in the chained modular Bell inequality. Before reduction modulo $d$, the chain telescopes:Reduction changes this integer only by multiples of $d$. Since all reduced terms are nonnegative,The smallest possible nonnegative value in this residue class is $d-1$. Averaging over the hidden variable provesAdding local random seeds to $\lambda$ gives the same result for stochastic local hidden-variable theories.

For the coplanar qubit realization of the chained Bell inequality, shareAt angle $\theta$ in the $xz$ plane, use the projective measurement of the observableHere ${\sigma }_x,{\sigma }_z$ are Pauli matrices, and the classical bit $a$ records eigenvalue $(-1{)}^a$. Since $⟨{\sigma }_z\otimes {\sigma }_z⟩=⟨{\sigma }_x\otimes {\sigma }_x⟩=1$ and both crossed expected values vanish in this Bell state,The one-party expected values vanish, so the complete joint probabilities areFor binary outcomes, $[a-b]$ is one exactly when $a\ne b$, while $[b-a-1]$ is one exactly when $a=b$. Consequently

For $N\ge 2$, set $\delta =\pi /(2N)$ and choose Alice's angle ${\theta }_i=2(i-1)\delta$ and Bob's angle ${\varphi }_i=(2i-1)\delta$. Every adjacent pair $A_i,B_i$ or $B_i,A_{i+1}$ differs by $\delta$, giving expectation ${\sin }^2(\delta /2)$. The closing pair has angle difference $(2N-1)\delta =\pi -\delta$; its agreement probability is also ${\sin }^2(\delta /2)$. Thus every one of the $2N$ terms contributes equally:Indeed $\sin x<x$ for $x>0$, so this value is below ${\pi }^2/(8N)\le {\pi }^2/16<1$. For an explicit two-setting instance, Alice measures ${\sigma }_z,{\sigma }_x$ and Bob measures $({\sigma }_z+{\sigma }_x)/\sqrt{2},(-{\sigma }_z+{\sigma }_x)/\sqrt{2}$. The four relevant expected values sum toviolating the local bound one. At $N=1$ there is no violation; the two terms always sum to one. As $N\to \infty$, the displayed quantum value is asymptotic to ${\pi }^2/(8N)$ and approaches the algebraic lower bound zero.

The violation rules out a local hidden-variable theory with setting-independent preassigned local responses reproducing these quantum correlations. It does not permit faster-than-light communication: the joint formula gives each local outcome probability $1/2$, independent of the remote setting, as required by the no-signalling theorem. The contradiction concerns the joint correlations and the Bell assumptions, not the ability to send a signal.

For a spin singlet state, quantum spin measurements along unit vectors $\mathbf{a},\mathbf{b}$ predict $E(\mathbf{a},\mathbf{b})=-\mathbf{a}\cdot \mathbf{b}$. Take ${\mathbf{a}}_0=\mathbf{z}$, ${\mathbf{a}}_1=\mathbf{x}$, ${\mathbf{b}}_0=(\mathbf{z}+\mathbf{x})/\sqrt{2}$ and ${\mathbf{b}}_1=(\mathbf{z}-\mathbf{x})/\sqrt{2}$. These giveThe violation therefore excludes the whole conjunction of the sufficient assumptions, for any model reproducing these correlations. It does not identify which premise fails without further commitments about causal explanation, state completeness and measurement setting generation.

A defensible verdict is to retain ordinary independent setting preparation and reject Bell local causality, specifically its classical screening-off factorization. In standard quantum theory, taking the pure entangled state as the complete state preserves parameter independence: each local marginal is $1/2$, regardless of the distant setting. Yet outcome independence fails. For equal spin axes, $P(A=+,B=-)=1/2$, whereas $P(A=+)P(B=-)=1/4$. The distant outcome changes the conditional prediction even after the quantum state is specified. This failure is compatible with quantum no-signalling, since conditioning on that outcome requires learning a result which cannot itself be controlled.

Adding further past variables does not restore the entire Bell conjunction while retaining the violating correlations; the CHSH inequality already allowed arbitrary such $\lambda$. Alternatively, one may reject measurement independence, permit retrocausal setting dependence in the relevant past state, or adopt another non-Bell-local description. Those are logically possible responses, but each must explain its replacement probability and causal model rather than treating the inequality as a failed algebraic theorem. A deterministic theory is not exempt from this requirement.

The assumption of measurement independence is supported by the purpose and design of independently varied settings, but cannot be proved for every conceivable hidden variable merely by inspecting observed setting frequencies. Conversely, a rejection of Bell factorization does not alone establish a controllable superluminal signal or select a unique ontology. The reasoned verdict is against Bell's screening-off locality, conditional on the other stated premises; the unconditional conclusion concerns their conjunction.

Consider two spacelike-separated measurement regions $R_A$ and $R_B$, with locally chosen settings $a,b\in \{0,1\}$ and recorded outcomes $A,B\in \{-1,1\}$. A relevant past region supplies a sufficient specification $\lambda$ of the common preparation and other hidden physical information needed to screen the outcomes. This is not necessarily just a quantum state written at the source, nor a conditioning on arbitrary records which already contain the future choices. The spacetime shielding and sufficiency requirements are substantive causal assumptions.

Use Bell local causality: once the relevant past is sufficiently specified, learning facts in the remote measurement region changes no local outcome probability. In particular, requirewhere conditional probabilities are defined. The probability chain rule givesThis factorization combines parameter independence with outcome independence. Neither probability is required to be zero or one: the argument includes stochastic local hidden-variable theories.

Also assume measurement independence, $\rho (\lambda \mid a,b)=\rho (\lambda )$, so all four setting pairs average over one normalized nonnegative past-variable distribution. This is a statistical condition on the setting procedure and relevant hidden information, not a proof of a metaphysical doctrine of free will. Finally the recorded outcomes must represent the same ensemble of trials. One can include nondetections in a specified bounded-outcome rule; if instead one discards events depending on setting or outcome, a further sampling assumption is required. Setting-dependent postselection is not covered silently by the argument.

Let ${\alpha }_a(\lambda )={\sum }_{A}AP(A\mid a,\lambda )$ and ${\beta }_b(\lambda )={\sum }_{B}BP(B\mid b,\lambda )$. They lie in $[-1,1]$, and the factorization givesFor each $\lambda$,Averaging with the same $\rho$ proves the CHSH inequalityNo separate assumption of simultaneous predetermined values for incompatible quantum observables was needed. The common probability model and its screening relations did the work.

A Reichenbach common cause principle can motivate the same calculation: a complete classical common cause in the past screens the two outcomes, while each setting affects only its own wing. Together with setting-independent sampling this gives the displayed factorization. It is not enough to provide a different screening variable for each pair of settings: those constructions need not form one compatible common-cause model for the four correlations.

A stochastic Einstein locality formulation instead constrains objective chances by facts in the appropriate causal past. To use it as a sufficient premise here, specify both local dependence on settings and a joint screening condition for the two outcomes. Merely requiring local marginal chances not to depend on a remote setting gives parameter independence, which is weaker than factorization. Thus the precise stochastic locality version matters; one should not infer a Bell inequality from every condition loosely described as no superluminal influence.

Combining the EPR criterion of reality with perfect predictive correlations and locality motivates pre-existing values for the locally selectable observables. The general local hidden-variable theory additionally assumes a complete shared variable $\lambda$ and conditional factorizationThe second condition is measurement independence. The reality criterion alone does not prove this factorization for arbitrary imperfect correlations: it motivates the local completion whose consequences are being tested.

A deterministic local hidden-variable model may be used without loss of generality by putting local random seeds into $\lambda$. For each $\lambda$, write its setting responses as $A,A^{\prime },B,B^{\prime }\in \{-1,1\}$. Integrating with the same setting-independent distribution and using the triangle inequality givesExactly one of $B-B^{\prime }$ and $B+B^{\prime }$ vanishes and the other has magnitude two. Hence the requested CHSH bound is two. No assumption about the spacelike quantum state enters this local-model derivation.

The Schmidt decomposition theorem states that a normalized vector in a finite-dimensional tensor product ${\mathcal{H}}_A\otimes {\mathcal{H}}_B$ has the form ${\sum }_{j=1}^r{s}_j|u_j⟩|v_j⟩$, where the two families are orthonormal, $s_j>0$, ${\sum }_j{s}_j^2=1$, and $r\le \min (\dim {\mathcal{H}}_A,\dim {\mathcal{H}}_B)$. Extending each family to an orthonormal basis for two qubits givesThe nonzero Schmidt coefficients can be made positive by absorbing phases into the basis vectors. Both are positive exactly when the pure state is entangled. A product state has Schmidt rank one, so one coefficient is zero; the source's assertion of two positive coefficients for every pure state needs this exception.

Choose each local orthonormal basis independently as the computational basis. In that basis ${\sigma }_z=|0⟩⟨0|-|1⟩⟨1|$. This is a local change of coordinates, implemented by separate unitary matrices, rather than a physical restriction on the original quantum state. The associated Pauli operators supply the other two local axes. Up to an irrelevant common phase, each unitary change of qubit basis corresponds to a rotation of its Bloch sphere.

Put $s=2{\lambda }_0{\lambda }_1$. The Schmidt-basis Pauli correlation tensor is diagonal. The Pauli operators ${\sigma }_x\otimes {\sigma }_x$ exchange $|00⟩$ and $|11⟩$, while ${\sigma }_y\otimes {\sigma }_y$ do so with minus signs, and ${\sigma }_z\otimes {\sigma }_z$ leaves both fixed. HenceMixed components vanish: those containing one $z$ and one transverse Pauli operator map the occupied basis vectors outside their span, while the $xy$ and $yx$ matrix elements are purely imaginary and cancel for real Schmidt coefficients. By bilinearity, for arbitrary real vectors,For a CHSH inequality test take unit quantum measurement axesThese CHSH axes for an entangled pure two-qubit state giveThe local bound is two: for each hidden state, $A(B+B^{\prime })+A^{\prime }(B-B^{\prime })$ is $\pm 2$ when all four outcomes are $\pm 1$, and averaging cannot increase its absolute value. Thus every entangled pure two-qubit state violates a CHSH inequality, the content of Gisin's theorem. A product state has $s=0$ and does not violate it, so the unqualified final claim in the question is false for that case. The maximum $2\sqrt{2}$ occurs for equal Schmidt coefficients. This excludes local hidden-variable theories satisfying measurement independence, but does not permit faster-than-light signalling: the local reduced density matrix and its quantum measurement probabilities are unchanged by the remote choice of axis.

For the spin singlet state, measurements along the same axis are perfectly anticorrelated. By measuring either spin component of particle $B$, one can therefore predict with certainty the corresponding result for distant particle $A$. The Einstein–Podolsky–Rosen criterion of reality then says that every such spin component of $A$ is an element of physical reality, because the choice at $B$ can be made without disturbing $A$. Since the quantum state assigns no simultaneous sharp values to noncommuting spin components, the EPR argument concludes that the wavefunction is not a complete description, provided this locality premise is accepted.

Let $\lambda$ denote a proposed complete hidden state, and let $a,b$ be the chosen axes.

For two axes on each side, every $\lambda$ obeysMeasurement independence permits averaging the same distribution of $\lambda$ for all four setting pairs, producing the CHSH inequality

The singlet prediction isChoose coplanar unit vectorsThenand thereforeThus the predictions of quantum theory violate the conjunction of outcome determinism, parameter independence, and measurement independence. This is Bell theorem; the calculation alone does not select which premise a deeper theory must abandon.

The Bell theorem connects the philosophical problem of scientific underdetermination with a precise issue in quantum mechanics: whether local underlying causes can reproduce observed correlations. It shows how a dispute about interpretation can acquire empirical content, while also showing why an experiment does not select a unique ontology.

A shared cause can produce strong correlations without a signal between distant measurements. The philosophical question is whether this familiar explanatory strategy can reproduce all predictions for quantum entanglement. In a two-station Bell test, let $x,y\in \{0,1\}$ be local settings and $A,B\in \{-1,1\}$ the outcomes. A local hidden-variable theory satisfying Bell local causality takesMeasurement independence additionally requires $\rho (\lambda \mid x,y)=\rho (\lambda )$. The same ensemble of underlying states must therefore apply across the four setting pairs.

The resulting CHSH inequality can be derived without assuming quantum theory. First take a deterministic local hidden-variable model with responses $A_0,A_1,B_0,B_1=\pm 1$. For each $\lambda$,because exactly one bracket vanishes. Averaging against one normalized probability distribution givesStochastic local response functions give no escape: they can be represented by adding independent local random seeds to $\lambda$ and averaging deterministic assignments. The essential restriction is the common setting-independent ensemble and local dependence of responses, not a separate insistence that all nature be deterministic.

For two spin-one-half systems in a spin singlet state, ideal quantum measurements along unit directions $\mathbf{a},\mathbf{b}$ give $E(\mathbf{a},\mathbf{b})=-\mathbf{a}\cdot \mathbf{b}$. ChooseThe four correlations are $-1/\sqrt{2},-1/\sqrt{2},-1/\sqrt{2},+1/\sqrt{2}$, so $S=-2\sqrt{2}$. Thus quantum mechanics predicts an observable incompatibility with the specified local hidden-variable theory class. This is stronger than saying that two interpretations happen to fit the observations collected so far.

Actual Bell tests require more than the ideal calculation. Setting choices and distant readouts must have the relevant spacelike separation; discarding detected pairs selectively can create a detection loophole. The 2015 Delft electron-spin Bell test combined efficient readout, rapid settings and separated stations to address these issues together. The reported statistic was $S=2.42\pm 0.20$, with a test of the local-model hypothesis accounting for device memory.^[ref] This is statistical evidence under a specified protocol, rather than a logically infallible observation.

The philosophical conclusion must be equally precise. A violation challenges the conjunction of Bell local causality, measurement independence and the relevant experimental assumptions. It does not disprove every possible hidden-variable account, or establish that the observer's mind creates the outcome. Different theories can reject different premises while agreeing on the observable correlations. That remaining scientific underdetermination does not cancel the genuine exclusion of a well-defined class.

Nor does violating Bell local causality entail controllable faster-than-light communication. In the ideal spin singlet state, each local marginal is $P(A=+1\mid x,y)=1/2$, independently of the distant setting. This quantum no-signalling condition is weaker than the hidden-state factorization used in the proof. Conflating them turns a precise result into an unwarranted conclusion about signalling.

The example therefore illustrates both the power and the limit of empirical testing in philosophy of science. Mathematical clarification can transform an apparently metaphysical disagreement into rival observational constraints. Evidence can discriminate those constraints without determining every question about the reality represented by a successful theory.

Quantum mechanics is compatible with quantum no-signalling, but its full correlations are incompatible with Bell local causality under the usual additional assumptions. The answer depends on which locality condition is meant. A local operational prohibition on controllable faster-than-light communication is weaker than a causal explanation of correlations by a complete shared past.

Let $a,b\in \{-1,1\}$ be outcomes, $x,y$ local settings and $\lambda$ a complete relevant past specification. Bell factorization and measurement independence demandNeither assumes deterministic response. Put $A_x(\lambda )={\sum }_{a}ap(a|x,\lambda )$ and similarly $B_y$, so $|A_x|,|B_y|\le 1$. Pointwise,Averaging with the same $\rho$ for all settings gives the CHSH inequality $|E_{00}+E_{01}+E_{10}-E_{11}|\le 2$. For a spin singlet state, $E(\mathbf{a},\mathbf{b})=-\mathbf{a}\cdot \mathbf{b}$. Choose ${\mathbf{a}}_0=\hat{z}$, ${\mathbf{a}}_1=\hat{x}$, ${\mathbf{b}}_0=(\hat{z}+\hat{x})/\sqrt{2}$, ${\mathbf{b}}_1=(\hat{z}-\hat{x})/\sqrt{2}$; the magnitude becomes $2\sqrt{2}$. Thus the assumptions cannot all be maintained.^[ref]

There is nevertheless no signalling contradiction. If a local unread measurement has Kraus operators $M_a$ with ${\sum }_a{M}_a^{†}{M}_a=I$, thenThe equality follows by testing both sides against any distant observable and using cyclicity of the trace. Distant conditional states can differ after selecting $a$, but communicating that selection needs an ordinary message. The microcausality condition in quantum field theory similarly ensures commutation of spacelike-separated local observables; it does not make the vacuum or other entangled quantum states Bell-factorizable.

One can preserve different notions of locality by changing premises. Bohmian mechanics retains unique outcomes but its guidance equation is nonlocal for entangled systems; quantum equilibrium equivariance preserves the equilibrium distribution which masks that dependence from signalling. The many-worlds interpretation can combine local interaction equations with no unique global outcome, but a claim of locality must explain its branch-relative correlations; it has not thereby constructed the above Bell-local single-outcome model. The measurement-independence loophole, including suitable retrocausal or correlated-setting models, drops the shared setting-independent $\rho$ premise. Such models are logical possibilities, not demonstrations that the ordinary locality assumptions survive unchanged.

The justified answer is therefore conditional rather than an unqualified yes or no. Operational locality is retained by the quantum formalism. A theory reproducing all its correlations cannot retain Bell factorization, independent settings and the ordinary joint-outcome framework together. Neither denial of hidden variables nor an appeal to relativity alone removes the probabilistic contradiction.

Outcome determinism means that a complete hidden state $\lambda$, together with the chosen measurement settings, fixes each outcome. The observed probabilities can still arise from an ensemble of unknown $\lambda$. This differs from deterministic unitary evolution of a wavefunction, which does not by itself specify a unique measurement outcome. Quantum nonlocality means failure of a Bell-local account of the correlations, under assumptions including measurement independence; it is stronger than the existence of entanglement and does not imply controllable faster-than-light signaling.

Use the first ket for the Positron and the second for the electron. Before either final beam splitter, the first pair of splitters and annihilation produceThe photon state $|\gamma ⟩$ is orthogonal to the surviving charged-particle sector. Its probability is $1/4$, while the remaining sector has squared norm $3/4$. The states below retain that photon amplitude and are normalized over all trials, without conditioning on nonannihilation. For compactness write $|cc⟩=|c^{+}{c}^{-}⟩$, and similarly for $|cd⟩,|dc⟩,|dd⟩$.

In a deterministic local hidden-variable model, write $C^{+}(R,\lambda )$ or $D^{+}(I,\lambda )$ without an argument for the remote splitter setting. That omission is the parameter independence assumption: the local response for a fixed complete hidden state does not change when the other wing changes its apparatus. Measurement independence also requires the hidden-state ensemble to be independent of the freely selected settings, allowing the same $\lambda$ assignments to be compared across configurations.

The zero probabilities derived above then constrain those assignments, apart from sets of hidden states of zero measure:

For a trial with $D^{+}(I,\lambda )D^{-}(I,\lambda )=1$, the last two implications force $C^{+}(R,\lambda )=C^{-}(R,\lambda )=1$, contradicting the first condition. Thus these deterministic local assignments predict zero probability for the double-$D$ event, whereas quantum theory predicts $1/16$. This is Hardy's paradox, an inequality-free ideal contradiction. With imperfect apparatus the zero-event conditions become small nonzero rates and require a statistical inequality rather than declaring a contradiction from one imperfect detection.^[ref]

The logical contradiction combines counterfactual hidden-variable assignments; it does not assign simultaneously observed path properties to a quantum state. Nor does it allow a message to be sent by changing a distant splitter: the observable local marginal obeys quantum no-signalling. For example, fixing the plus-side splitter to $I$ gives $P(D^{+})=1/8$ for either minus-side setting in the results above.

An eigenvalue of an observable in a known corresponding eigenstate is a simple example satisfying the Einstein–Podolsky–Rosen criterion of reality: its value is predictable with certainty. For example, a spin angular momentum component in a prepared spin-up eigenstate has a definite predicted value. For a distant system, Albert Einstein, Boris Podolsky and Nathan Rosen invoked locality: a measurement performed on a sufficiently separated, noninteracting partner should not disturb that system. In the relativistic version here, the measurement events have spacelike separation.

The Pauli matrices obey $[{\sigma }_z,{\sigma }_x]=2i{\sigma }_y$, soTheir nondegenerate eigenstates are incompatible: an $S_z$ eigenstate has equal probabilities for the two $S_x$ outcomes, and conversely. Thus no single quantum state has both components sharply determined. The Robertson uncertainty relation gives $\Delta S_z\Delta S_x\ge \hslash |⟨S_y⟩|/2$, but its right-hand side may be zero; that alone is not a proof that simultaneous zero uncertainties exist. The absence of a common eigenstate supplies the stronger conclusion for this spin-half pair.

For the specified Bell state, Bob obtains $+\hslash /2$ or $-\hslash /2$ with probability $1/2$ when he measures $S_z$. Conditional on his result, Alice obtains the same sign with certainty if she measures $S_z$. Under the EPR locality premise, this certainty supplies an element of reality for her $S_z$; it is a conditional prediction after Bob learns his own random result, not a premeasurement prediction of its sign.

The normalized $S_x$ eigenstates are $|+⟩=(|0⟩+|1⟩)/\sqrt{2}$ and $|-⟩=(|0⟩-|1⟩)/\sqrt{2}$. Expanding their tensor products givesConsequently the same Bell state has equal, perfectly correlated $S_x$ outcomes. Bob's two outcomes again each have probability $1/2$, and the same locality-and-certainty argument applies separately to Alice's $S_x$.

The disputed step is to combine predictions conditional on two incompatible choices into two simultaneous, measurement-independent pre-existing values. Quantum theory provides the two conditional ensembles, not a joint sharp assignment of $S_x$ and $S_z$. This is quantum steering. Bob cannot measure both incompatible observables sharply on the same member of the pair. Nor does conditional steering allow signalling: ignoring his result leaves Alice's reduced density matrix equal to $I/2$ for either measurement choice, by the no-communication theorem.

To test the stronger claim of a local hidden-variable theory, Alice needs several correlation settings, not just perfect agreement along the two common axes. For the CHSH inequality, choose observables with outcomes $\pm 1$:In the given Bell state, $⟨{\sigma }_z\otimes {\sigma }_z⟩=⟨{\sigma }_x\otimes {\sigma }_x⟩=1$ and the crossed correlations vanish. Henceexceeding the local bound $|\mathcal{S}|\le 2$. The bound assumes Bell local causality and measurement independence, with the appropriate treatment of sampling and settings.

Experimental Bell tests support violations rather than that local assignment. The Aspect switching experiment changed analyzer settings during separated photon flights, while leaving sampling and switching limitations. By the time of this examination, the Delft electron-spin Bell test had used electron spins at stations $1.3\mathrm{km}$ apart, efficient spin readout and rapid random basis choices. The combination addressed both detection and locality loopholes, providing evidence against the tested local model class.^[ref]

This does not prove that every conceivable notion of reality is false, or that noncommutation alone rules out every hidden-variable theory. A nonlocal theory can evade the local premise. What Alice can challenge is Bob's claim that locality, measurement-independent simultaneous values and all observed correlations can be retained together. An ordinary measurement disturbance at Bob's lab is not by itself a local explanation of Alice's conditional ensemble.

The Bell theorem gives a powerful conditional constraint, but does not by itself establish the unrestricted philosophical sentence that every interpretation contains a literal instantaneous action between distant objects. One must specify what locality means and which further assumptions are retained. In particular, Bell local causality is stronger than an inability to send a faster-than-light message.

Let $a,b$ be settings in spacelike-separated laboratories, $A,B\in \{-1,1\}$ their recorded outcomes, and $\lambda$ a sufficient specification of relevant prior physical variables. A Bell-local model factorizes asTogether with measurement independence, $\rho (\lambda \mid a,b)=\rho (\lambda )$, this constrains observable correlations. It allows stochastic response functions: predetermination of all outcomes is not a necessary separate assumption in this general formulation. Bell's 1964 paper demonstrates the incompatibility between a local completion and selected quantum correlations; the two-setting inequality below is a standard later form of the argument, not a claim about the exact notation of that paper.^[ref]

For example, write the local conditional mean outcomes as $A(a,\lambda ),B(b,\lambda )\in [-1,1]$. At each $\lambda$,Averaging over the same setting-independent distribution gives the CHSH inequality, $|S|\le 2$. A suitable spin singlet state instead predicts $|S|=2\sqrt{2}$. Consequently no model retaining the stated factorization and independence assumptions can reproduce all those quantum correlations. Merely replacing deterministic hidden variables with local random noise cannot evade the bound.

However, violation of this bound does not entail operational signalling. For an ordinary entangled state, a remote nonselective measurement leaves the local reduced density matrix unchanged. The local experimenter cannot choose the outcome needed to encode a controllable message. Quantum no-signalling and failure of Bell factorization can therefore coexist; confusing them would misstate both the prediction and the theorem.

Interpretations allocate the explanatory burden differently. In Bohmian mechanics, the guiding dynamics of an entangled configuration is explicitly nonlocal, though quantum equilibrium masks controllable signalling. A collapse interpretation must explain the status of correlations and of any proposed nonlocal state change. Everettian accounts retain the entangled global state and multiple relative records; their defenders can deny a fundamental action that selects one distant outcome. That does not produce a factorizing classical common-cause model of the observed correlations. Whether such an account deserves the word local depends on whether locality means local dynamical interactions, separability, or Bell's single-outcome causal factorization.

One can also reject setting independence or change the assumed causal organization, but such proposals must be developed as physical models, not advertised as refutations of the inequality's proof. The warranted conclusion is that the conjunction of Bell locality, independent settings and the relevant outcome framework is incompatible with the quantum correlations. Calling the failure action-at-a-distance is justified only after an interpretation specifies the physical entities, events and causal commitments to which that phrase applies.

In 2015, Marissa Giustina, Anton Zeilinger and collaborators combined spontaneous parametric down-conversion with rapid random analyzer choices and efficient transition-edge sensors. Local records included settings and detector signals; their Bell inequality used detected and undetected trials, avoiding fair sampling. Locally fixed time slots prevented coincidence-based trial selection. Calibrated optical and electrical delays checked spacelike separation of each choice from the remote readout. The violation addressed detection and locality together. Measurement independence still requires assumptions about the hidden state's origin; the experiment cannot exclude every possible common cause of source and settings.^[ref]

Quantum correlations incompatible with Bell local causality under additional assumptions including measurement independence. This is an operational statement about joint probabilities, not controllable faster-than-light signaling: quantum predictions also obey quantum no-signalling. Entanglement alone does not guarantee violation of a specified Bell inequality in a given measurement arrangement.