This morning, a new arXiv paper 2105.13519 appeared providing detail of an EPR steering experiment showing that if it is assumed that the EPR steering requires quantum collapse, then more than one bit had to be communicated faster than light. It is a non-trivial experiment done by an international collaboration with much attention to detail. Special attention is given to address the various loopholes. Of course, these details would have to be checked. But let's assume it is all correct.

Since faster than light travel is not considered to be possible, is it reasonable to conclude that what this experiment really shows is that quantum collapse does not happen? In other words, can this experiment be regarded as an indication that all interpretations of quantum mechanics based on quantum collapse are wrong?


This answer assumes that the word "collapse" in the question refers to the specific idea that the cited experiment is designed to address, namely that something physically happens simultaneously at both locations when one member of an entangled pair is measured.

We've always known that the collapse idea (as defined above) is not compatible with special relativity, because the time-order of two spacelike-separated measurement events $A$ and $B$ is observer-dependent. If $A$ and $B$ measure objects that are entangled with each other, then which of those two measurements should trigger the alleged collapse simultaneously at both locations? The measurement that happens first? But which one happens first? It's observer-dependent, because they're spacelike-separated.

That doesn't prove that the collapse idea is wrong, though, because an addict could argue that special relativity is wrong instead. The experiment doesn't change this situation. Experiments don't cure addictions.

  • 2
    $\begingroup$ As an addict, I might add that the collapse is compatible with special relativity as long as no information is sent between the two measuring events. The measuring events don't cause one another. They just happen, as two space-separated events can happen simultaneously. It's exactly the assumption that no information has been sent that is contested in this experiment. $\endgroup$ May 31 '21 at 15:23
  • $\begingroup$ @DescheleSchilder What exactly do you mean by "collapse"? Are you using that name for the idea that the thing being measured influences (becomes entangled with) the thing that's measuring it? That's perfectly consistent with relativity, of course, but that's not what the experiment cited in the question is trying to address. The experiment is trying to address the idea that something happens simultaneously at both locations when one member of an entangled pair is measured (EPR "steering"). My answer is responding to that specific idea, which is the usual connotation of the "collapse" language. $\endgroup$ May 31 '21 at 21:57
  • $\begingroup$ I mean that the thing being measured instantaneously (simultaneously) changes over a spacelike distance. This will have an "entangling" effect on the measurement devices. The entity being measured causes this entanglement. But if the collapse is not caused by anything, there is compatibility with SR. If information is transferred (to make the faraway place know about the measuring event) then collapse will not be compatible with SR as information cannot be sent FTL. Maybe in that case the collapse is just non-instantaneous. $\endgroup$ May 31 '21 at 22:19
  • $\begingroup$ @DescheleSchilder. In the paper it is explicitly stated that more than one bit of information must have been set faster than light, if quantum collapse took place. So, as I understand it, this means that there is indeed information being sent. $\endgroup$ Jun 1 '21 at 3:22
  • $\begingroup$ @flippiefanus Huh? I didn't read the paper. Information was observed to travel FTL? $\endgroup$ Jun 1 '21 at 3:30

Collapse is generally considered wrong now, but only if the collapse is absolute. If the collapse is relative, i.e. observer dependent, then it is still compatible with causality etc.


Here is an argument for the process of collapse of our knowledge of entangled quantum processes.

Certain quantum processes are entangled and yet are capable of showing the determinacy of entanglement. For example, one can use an electric circuit to separate ions and electrons [QM objects] on the plates of a capacitor. Then you separate the plates and put them in different places. As long as you do not know the direction of the potential difference of the electric circuit, your knowledge about the charge on the capacitor plates is zero.
After you measure the direction of the current of your experimental setup the quantum mechanical wave function of the un-knowledge collapses and you know, which plate is charged with more electrons and which with more ions.
Applying a voltmeter to the plates you discharge the plates and proof the prediction from the measurement of the current from the experimental setup. If the capacity of the plates is large and your measurement is performed quickly, you can repeat the measurement several times to verify your results.

However, there are processes where the generation of the entanglement and the subsequent measurement of the results is ambiguous. The point of this uncertainty is the claim that the wave functions of the entangled particles remain entangled until measurement.

Take the process of generation of entangled photon pairs. It is possible to design quantum dots, producing pair of photons, entangled in their polarization. The photons of each pair have the directions of polarization with an random angle but entangled by an angle difference. Say the angle for the first photon may be 33°, than the angel of the second photon is 33°+180°. The point is, that the production of photons with always the same angle of polarization was not possible. And the proof, that an entanglement exist, was done by statistical methods. Not knowing the angles of polarizations of the photons, you should place the polarizers for the measurement in an arbitrary direction and not every measurement will be successful.

Instead of claiming the interference of wave functions until measurement one may claim that the entanglement was finished from the moment of pair production and exists until the disturbance from measurement processes. Both claims are not provable and meaningless for cryptographic and other technical uses.

  • $\begingroup$ Not sure I understand your argument. I don't doubt the existence of entanglement. That does not proof that there is something like quantum collapse. In the Many Worlds interpretation you can still have entanglement. $\endgroup$ Jun 1 '21 at 3:19
  • $\begingroup$ @flippiefanus Entanglement is very real. The uncertainty - for example for the spin of entangled particles - is very real because of our inability in the measurement process, which is accompanied by the destruction of the entanglement. But the argumentation, that the values of the spins directions arises only during the measurement should be equally valid as the argument, that these values arises at the moment of the entanglement. Both is not verifiable but the second point of view makes it much easier :-) $\endgroup$ Jun 1 '21 at 3:49
  • $\begingroup$ What about the argumentation in the answer from the link below your question? $\endgroup$ Jun 1 '21 at 3:51

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.