Bell's Theorem: Why Entanglement Isn't Spooky Messaging

Few ideas in physics are as widely quoted and as widely misunderstood as quantum entanglement. Einstein called it “spooky action at a distance.” Popular accounts suggest particles that instantly communicate across the universe. The reality is subtler, and in some ways stranger.

What entanglement is

In quantum mechanics a system is described by a state, and for two particles that state need not factor into “a state for particle one” times “a state for particle two.” Take two spin-1/2 particles prepared in the singlet state: if you measure the spin of each along the same axis, you always get opposite results, yet before measurement neither particle has a definite value on its own.

The pair has a well-defined joint property, “opposite spins,” without either member having an individual one. That is entanglement.

The Einstein–Podolsky–Rosen worry

In 1935 Einstein, Podolsky and Rosen argued that this made quantum mechanics incomplete. A natural alternative: the particles simply carry hidden instructions, like a pair of gloves sent in separate boxes. Open one box, find a left glove, and you instantly know the other is a right glove. Nothing travels between them; the information was there all along.

For three decades this looked like a matter of taste, since no experiment seemed able to tell the difference.

Bell changes the question

In 1964 John Bell showed it could be tested. Any theory in which outcomes are determined by pre-existing local hidden variables must obey a bound on how strongly measurements on the two particles can correlate. Quantum mechanics predicts that bound is broken.

The most common form is the CHSH inequality. Alice picks one of two measurement settings, and so does Bob. Compute a combination S of their four correlations:

Theory	Maximum value of S
Local hidden variables	2
Quantum mechanics	2√2, about 2.83

The 2√2 ceiling for quantum mechanics is known as Tsirelson’s bound.

What experiments found

Experiments beginning in the 1970s, with notable work by John Clauser and later Alain Aspect in the early 1980s, measured values of S well above 2, in agreement with quantum mechanics. Early tests had loopholes: detectors that missed too many particles, or settings that were not chosen quickly enough to rule out communication. Loophole-free tests, first reported in 2015, closed them. Clauser, Aspect and Anton Zeilinger shared the 2022 Nobel Prize in Physics for this line of work.

Why you still can’t send a message

If measuring one particle seems to affect the other, why can’t we signal with it? Because each party’s results, viewed alone, look completely random. Alice sees a fair coin flip no matter what Bob does. The correlation only appears when the two lists of results are brought together, and that comparison needs an ordinary channel, limited by the speed of light.

import random

def local_hidden_variable_pair():
    """Gloves in boxes: outcomes fixed at the source, so S can't exceed 2."""
    spin = random.choice([+1, -1])
    return spin, -spin

A model like this reproduces the “always opposite” part but cannot reproduce the full pattern of correlations across different measurement angles. That gap is exactly what Bell’s theorem exposes.

Why it matters beyond philosophy

Entanglement is a resource. It underpins quantum key distribution, whose security rests on the Bell test itself, and it is central to quantum computing and quantum networks. What began as a philosophical objection became an engineering tool.