Entangled ≠ Communicating

Entanglement is one of the strangest features of quantum mechanics: two qubits can share a joint state such that measuring one instantly determines what you would observe in the other, no matter how far apart they are. This sounds — at first glance — like a recipe for faster-than-light communication. It is not, and understanding exactly why not is one of the deepest lessons in quantum information.

The Bell state and correlated outcomes

Start with the familiar $|\Phi^+\rangle$ Bell state:

|\Phi^+\rangle = \tfrac{1}{\sqrt{2}}\big(|00\rangle + |11\rangle\big).

Alice holds qubit 0, Bob holds qubit 1, and they travel to opposite ends of the galaxy. When Alice measures her qubit in the computational basis, she gets 0 or 1 with probability $\tfrac{1}{2}$ each. If she gets 0, the joint state collapses to $|00\rangle$ , so Bob's qubit is now also 0. If she gets 1, the state collapses to $|11\rangle$ , and Bob's qubit is 1.

The key word is random. Alice cannot choose which outcome she gets — it is decided at the moment of measurement. She therefore cannot encode any message into the random bit Bob observes.

What Bob actually sees

Before Alice measures, Bob's reduced density matrix is obtained by tracing over Alice's qubit. For $|\Phi^+\rangle$ the result is

\rho_B = \tfrac{1}{2}|0\rangle\!\langle 0| + \tfrac{1}{2}|1\rangle\!\langle 1| = \frac{I}{2},

the maximally mixed state. This is the same state Bob would have if the two qubits were uncorrelated and each independently prepared in a 50/50 mixture. It is also identical to what Bob sees after Alice has measured — he gets 0 or 1 with probability $\tfrac{1}{2}$ either way.

No local experiment on qubit 1 alone can distinguish "Alice measured" from "Alice did not measure." Correlations only become visible when Alice and Bob compare notes over a classical channel — which travels at light speed at best.

Why faster-than-light signaling stays impossible

Suppose Alice tries to signal by choosing which basis to measure in. If she measures in the computational basis, the same argument applies: Bob's marginals are $\tfrac{1}{2}$ , $\tfrac{1}{2}$ . If she instead measures in the Hadamard ( $X$ ) basis, the Bell state can be rewritten as

|\Phi^+\rangle = \tfrac{1}{\sqrt{2}}\big(|{+}{+}\rangle + |{-}{-}\rangle\big),

so Bob's outcomes are equally likely in the $X$ basis too. No matter what operation Alice performs on her qubit, Bob's reduced state remains $I/2$ . This is not a coincidence — it is a theorem that follows directly from the linearity of quantum mechanics and the partial-trace rule.

The distinction between entanglement and communication is fundamental: entanglement is a resource for correlating future measurements, not a conduit for information. When you use Bell states in quantum teleportation or superdense coding, you will see that those protocols always require an additional classical channel to complete the transmission. The entanglement provides a speed-up (or compression) but never replaces that classical leg.