Case Study 1: EPR and the Completeness of Quantum Mechanics

Background

On May 15, 1935, Albert Einstein, Boris Podolsky, and Nathan Rosen published a paper in Physical Review titled "Can Quantum-Mechanical Description of Physical Reality Be Considered Complete?" This four-page paper, universally known as the EPR paper, launched a debate about the foundations of quantum mechanics that would not be resolved for three decades — and whose implications continue to shape physics, philosophy, and technology today.

The paper did not claim that quantum mechanics was wrong. Its predictions, EPR acknowledged, were correct. Instead, they argued that quantum mechanics was incomplete — that there must be additional variables ("hidden variables") needed to fully describe physical reality.

The EPR Setup in Modern Language

Let us restate the EPR argument using the formalism of this chapter.

The State

Alice and Bob prepare a pair of spin-1/2 particles in the singlet state:

$$|\Psi^-\rangle = \frac{1}{\sqrt{2}}(|0\rangle_A|1\rangle_B - |1\rangle_A|0\rangle_B)$$

where $|0\rangle = |\!\uparrow_z\rangle$ and $|1\rangle = |\!\downarrow_z\rangle$. The particles are then separated and sent to distant laboratories.

EPR's Criterion of Reality

EPR introduced a seemingly innocuous definition:

Criterion of Reality: "If, without in any way disturbing a system, we can predict with certainty (i.e., with probability equal to unity) the value of a physical quantity, then there exists an element of physical reality corresponding to this physical quantity."

The Argument

Step 1: Alice measures $\hat{S}_z$ on her particle.

If she obtains $+\hbar/2$ (spin up), the state collapses to $|0\rangle_A|1\rangle_B$, and Bob's particle is certainly in $|1\rangle_B$ (spin down in $z$). By the criterion of reality, $S_z^{(B)} = -\hbar/2$ is an element of reality.

If she obtains $-\hbar/2$, Bob's particle is in $|0\rangle_B$ (spin up in $z$). Either way, Bob's $S_z$ is determined.

Step 2: Alternatively, Alice measures $\hat{S}_x$ on her particle.

The singlet state can be rewritten in the $x$-basis:

$$|\Psi^-\rangle = \frac{1}{\sqrt{2}}(|+\rangle_A|-\rangle_B - |-\rangle_A|+\rangle_B)$$

where $|\pm\rangle = \frac{1}{\sqrt{2}}(|0\rangle \pm |1\rangle)$ are eigenstates of $\hat{S}_x$. By the same reasoning, Alice can determine Bob's $S_x$ with certainty.

Step 3: The locality assumption.

Alice is far from Bob. Her choice of measurement ($\hat{S}_z$ or $\hat{S}_x$) cannot disturb Bob's particle. Therefore, both $S_z$ and $S_x$ are simultaneously elements of reality for Bob's particle.

Step 4: The contradiction.

But $\hat{S}_z$ and $\hat{S}_x$ do not commute: $[\hat{S}_z, \hat{S}_x] = i\hbar\hat{S}_y \neq 0$. According to quantum mechanics, non-commuting observables cannot simultaneously have definite values. This contradicts step 3.

EPR's conclusion: Quantum mechanics is incomplete. There must be "hidden variables" that predetermine the outcomes.

Bohr's Response

Niels Bohr responded almost immediately, also in Physical Review. His response was notoriously difficult to follow, but the essential argument was:

The EPR criterion of reality assumes that we can reason about what Bob's particle "would have been" if Alice had made a different measurement. In quantum mechanics, the measurement context matters. The physical situation — including the entire experimental arrangement — defines what constitutes a phenomenon. You cannot mix results from incompatible experimental setups.

In modern language: Bohr was essentially arguing against counterfactual definiteness — the idea that unperformed measurements have definite outcomes.

The Partial Trace Perspective

Our chapter's formalism illuminates the debate. The reduced density matrix of Bob's particle is:

$$\hat{\rho}_B = \text{Tr}_A(|\Psi^-\rangle\langle\Psi^-|) = \frac{1}{2}\hat{I}$$

This is the maximally mixed state — Bob's particle, considered alone, contains zero information about any spin direction. The correlations exist only in the joint state.

When Alice measures $\hat{S}_z$: - If she gets $+\hbar/2$ (probability 1/2): $\hat{\rho}_B \to |1\rangle\langle 1|$ - If she gets $-\hbar/2$ (probability 1/2): $\hat{\rho}_B \to |0\rangle\langle 0|$

The ensemble average of Bob's state is still $\hat{\rho}_B = \frac{1}{2}(|0\rangle\langle 0| + |1\rangle\langle 1|) = \frac{1}{2}\hat{I}$ — unchanged. Alice's measurement does not change any local observable of Bob's particle. This is why no faster-than-light signaling is possible.

Bell's Resolution (1964)

John Stewart Bell provided the crucial insight that transformed the EPR debate from philosophy to physics. He showed that any theory satisfying:

1. Locality: No faster-than-light influences
2. Realism: Measurements reveal pre-existing values

must obey certain statistical inequalities (Bell inequalities). He then showed that quantum mechanics predicts violations of these inequalities.

The CHSH Inequality

The most experimentally useful form, due to Clauser, Horne, Shimony, and Holt (1969):

$$|S| = |E(a, b) - E(a, b') + E(a', b) + E(a', b')| \leq 2$$

where $E(a, b) = \langle \hat{\sigma}_a \otimes \hat{\sigma}_b \rangle$ is the correlation when Alice measures along direction $a$ and Bob along $b$.

For $|\Phi^+\rangle$ with optimal measurement angles ($a = 0, a' = \pi/2, b = \pi/4, b' = -\pi/4$):

$$|S|_{\text{QM}} = 2\sqrt{2} \approx 2.83$$

This exceeds the classical bound of 2.

Experimental Confirmation

Every experiment has confirmed quantum mechanics and violated the Bell inequality. Nature is not locally realistic in the EPR sense.

Discussion Questions

1. EPR assumed that "locality" and "realism" are both true. Bell showed that at least one must be abandoned. Which do you think is more natural to give up, and why?

2. The partial trace shows that Alice's measurement does not change any local observable of Bob's particle. How does this reconcile with the "spooky action at a distance" that troubled Einstein?

3. Some interpretations of quantum mechanics (e.g., many-worlds) maintain locality by giving up a single definite outcome. Others (e.g., Bohmian mechanics) maintain realism by accepting non-locality. What are the trade-offs?

4. The 2022 Nobel Prize citation specifically mentioned "experiments with entangled photons, establishing the violation of Bell inequalities." Why did it take 60+ years from EPR (1935) to loophole-free Bell tests (2015)?

5. Consider the coefficient matrix formalism: the singlet state has $C = \frac{1}{\sqrt{2}}\begin{pmatrix}0 & 1 \\ -1 & 0\end{pmatrix}$. The fact that $\det(C) = \frac{1}{2} \neq 0$ (rank 2) is what makes the EPR argument possible. If the rank were 1 (a product state), there would be no correlations to argue about. In what sense is the rank of $C$ the "mathematical root" of the EPR puzzle?