Physics Van 3-site Navigational Menu

Physics Van Navigational Menu

Q & A: entanglement, quantum collapse, and relativity

Learn more physics!

Most recent answer: 11/23/2018
When an entangled particle "collapses", physics says that the other particle collapses simultaneously. But that's non compatible with relativity, which says that simultaneity depends on the place and velocity of the observer. So the question is: - How can quantum simultaneity be compatible with relativity? Do the entanglement state is dependent on the observer, in the mean that a particle can be entangled for one observer and collapsed to other? - What experiments fundament the answer? Is the answer only theoretical?
- Marraco

That's a very important fundamental question. Relativity survives its encounter with quantum "measurement" processes intact, barely.

People sometimes say that relativity requires that "nothing can travel faster than c". That can lead to confusion about what is "something". Relativity does run into big problems if either

1. a conserved quantity


2. information

travels faster than c. At first glance it seems that this quantum collapse involves faster-than-c information transfer, but it turns out that it doesn't.

Let's say that spins A and B are remote and entangled. A measurement of A then gives a result that determines what must happen if B is measured on the same axis, and vice-versa. Doesn't that send information faster than-c from whichever was measured "first" (according to some particular observer's frame) to the other? Not really, because the result of the measurement is purely random. What do I mean by "random"? I mean that no chain of events near A can help predict the result at A. There are no local variables that lead to the result. This is very unlike, say, an ordinary telegraph message, in which there are some things going on near one end of the line that lead causally to the message sent, e.g. "Help. Stagecoach attack."  There would be big problems if that message started off at the "receiver" and then went to the "sender" and on to the fingers of the telegraph operator, etc. 

The quantum "signal" seems, according to some observers, to travel from A to B. Other observers give the direction as B to A. As usual, time order isn't invariant for space-like separated events. Because there is absolutely no causal chain at either end leading up to the content of the quantum signal, there's no reason to logically insist on who is the sender and who is the receiver. So long as quantum random signals are truly random all the way down, the direction of transmission does not have to be invariant. Relativity survives without any paradoxes.

Various experiments showing violations of the Bell Inequalities all show this absence of local causal variables. I think some even involve moving detectors so that each detector sees its own event first, in its own frame. 

Mike W.


(published on 07/04/2013)

Follow-Up #1: The Speed of Entanglement

In quantnum entanglement, would a change in one member of a pair be followed instantaneously by that change in the other regardless of separation? Or does the information have to travel at the speed of light?
- Sam (age 17)

Hi Sam,

Operations on an entangled particle do in fact affect the partner particle instantaneously, if you choose to describe things as if events at one particle affect the other one. (As we describe above, that doesn't really capture the strange connection between the events at the entangled particles.) Also as explained above, this does not violate the postulates of relativity, since randomness prevents any information being sent.

Actually, it was proven recently that if entangled partners interacted at any finite speed, then you could send faster-than-light signals. Since this wouldn't be consistent with special relativity, it is a proof that the interaction must be instantaneous ("infinite speed"). 

Hope that helps,

David Schmid

(published on 01/08/2014)

Follow-Up #2: entanglement and simultaneity

You don't answer to the first fundamental question: In SR we can't provide a bunch of clocks that are simultaneous (In SR we haven't absolute simultaneity). But in the QM entanglement we postulate that the wave function collapse simultaneously occurs in any frame that watch the first measurement at A, and this is the main paradox.
- Ali Ayat (age 26)
Tehran, Iran

You raise an important point. If you try to picture "collapse" as something that actually happens, then it seems to violate the speed limit. As we explain above, it does not actually do anything that makes problems for relativistic physics, e.g. it doesn't transmit information faster than c.

When people try to make modifications of quantum mechanics to make the collapse a real process, they do run into problems with relativistic paradoxes. One way of dealing with those issues is to simply drop the whole idea of collapse. That means accepting that all the different possible outcomes occur, an idea called the "Many Worlds" interpretation of quantum mechanics. When signals from the remote events reach each other, the amplitude for getting inconsistent signals from the different events is zero. So each observer, using only signals traveling at c or slower, only sees consistent "entangled" outcomes.

I know that sounds weird. 

Mike W.

(published on 11/23/2018)

Follow-up on this answer.