Can Entanglement be Used for Faster-than-Light Communication?

[Allen_Wolf]

Let us say we have an entangled pair of photons with opposite spin, which we want to use to transmit information at a speed greater than the speed of light. One of the spins could be assigned as YES (1) and the other as NO (0). We keep one photon and send the other to the receiver. In case out photon is observed the combined function of both the photons collapses and the the message is automatically sent.

The issue with this system to send a message properly is that the result will be completely random and this prevents us from creating such a quantum communicator.

But can't we just observe our photon a few times until our photon's spin becomes that of the value encoded as NO (0) and cosequently make the receivers result as YES (1). If the receiver checks the result only after a fixed period of time and we adjust our results by that period, we will be able to send a message properly. Right?

 

[DrChinese]

But can't we just observe our photon a few times until our photon's spin becomes that of the value encoded as NO (0) and cosequently make the receivers result as YES (1). If the receiver checks the result only after a fixed period of time and we adjust our results by that period, we will be able to send a message properly. Right?

A couple of rules apply in the situation you describe.

a) There is no sense in referring to time ordering when measuring an entangled object. The quantum formula does not yield a statement such as "If I measure Alice first, then Bob will have X occur." I.e. there is no way to distinguish that from the statement "If I measure Bob first, then Alice will have X' occur."

b) Once a measurement is performed on Alice, Alice is no longer entangled with Bob on that basis. So that is (one reason) why your proposal won't work.

 

[hilbert2]

But can't we just observe our photon a few times until our photon's spin becomes that of the value encoded as NO (0) and cosequently make the receivers result as YES (1). If the receiver checks the result only after a fixed period of time and we adjust our results by that period, we will be able to send a message properly. Right?

I think that if you start doing repeat measurements on a particle belonging to an entangled pair, you will either get the same result from every measurement due to Zeno effect, or the interaction of your measuring apparatus with the particle will destroy the entanglement.

 

[entropy1]

A couple of rules apply in the situation you describe.

a) There is no sense in referring to time ordering when measuring an entangled object. The quantum formula does not yield a statement such as "If I measure Alice first, then Bob will have X occur." I.e. there is no way to distinguish that from the statement "If I measure Bob first, then Alice will have X' occur."

b) Once a measurement is performed on Alice, Alice is no longer entangled with Bob on that basis. So that is (one reason) why your proposal won't work.

Then you contradict yourself: if the first measurement would break the entanglement, then there is a time ordering. Or am I missing something?

 

[Boing3000]

Then you contradict yourself: if the first measurement would break the entanglement, then there is a time ordering. Or am I missing something?

The only time ordering here is that one photon is entangled before beeing measured, and not anymore after. There is no contradiction at all...

 

[Boing3000]

a) There is no sense in referring to time ordering when measuring an entangled object. The quantum formula does not yield a statement such as "If I measure Alice first, then Bob will have X occur." I.e. there is no way to distinguish that from the statement "If I measure Bob first, then Alice will have X' occur."

I think all these problems are fixed if:
1) the source is in the middle, Bod a little more farther from the source.
2) Bod & Alice decide to use a fixed angle, 90° of delta would do.
Bod knows he his measuring after Alice, and every observers agrees (although I am not sure about accelerating FoR)

b) Once a measurement is performed on Alice, Alice is no longer entangled with Bob on that basis. So that is (one reason) why your proposal won't work.

I always thought it was the only reason, what are the others ?

 

[DrChinese]

Then you contradict yourself: if the first measurement would break the entanglement, then there is a time ordering. Or am I missing something?

Once you measure Bob, Bob will no longer evidence any subsequent entanglement with Alice (on that basis). That is true regardless of whether Alice has already been measured or not. The point I am making is that time ordering only relates to Bob on Bob's side. There is no relevant ordering related to Alice vs. Bob.

Hopefully I am not contradicting myself myself.

 

[entropy1]

Once you measure Bob, Bob will no longer evidence any subsequent entanglement with Alice (on that basis). That is true regardless of whether Alice has already been measured or not. The point I am making is that time ordering only relates to Bob on Bob's side. There is no relevant ordering related to Alice vs. Bob.

Hopefully I am not contradicting myself myself.

I would say we only know the entanglement is broken from Alice's point of view, after the photon has been measured by Alice, and likewise for Bob, as @Boing3000 is putting forward. If we consider a time ordering of measurements, and Alice measures first, then we would still measure a correlation with Bob's later measurement.

 

[DrChinese]

I think all these problems are fixed if:
1) the source is in the middle, Bod a little more farther from the source.
2) Bod & Alice decide to use a fixed angle, 90° of delta would do.
Bod knows he his measuring after Alice, and every observers agrees (although I am not sure about accelerating FoR)

[Allen_Wolf: can't we just observe our photon a few times until our photon's spin becomes that of the value encoded as NO (0) ?]

I always thought it was the only reason, what are the others ?

Your 1 & 2 above don't fix anything (there is nothing to fix). You can certainly force - as you suggest - Alice's measurement to occur first. But the correlation function - i.e. the prediction of QM - is no different regardless of the time ordering.

You can't observe a photon as being YES at an angle (say 0 degrees) and then re-observe it a few times and suddenly it is NO at that angle. Hopefully you see the problems there, such as the photon not existing anymore. Even an electron that was spin up will continue to be spin up unless it interacts with something else. Again, I hope you see my point, which is that the OP example is not realistic on other levels.

 

[DrChinese]

If we consider a time ordering of measurements, and Alice measures first, then we would still measure a correlation with Bob's later measurement.

There is no demonstrable difference in the QM expectation regardless of order. If there is some difference, we have no way to tell.

On the other hand, you will notice the difference if you put your socks on after you put on your shoes.

 

[entropy1]

There is no demonstrable difference in the QM expectation regardless of order. If there is some difference, we have no way to tell.

I think I'm saying that you can't claim entanglement of the whole system is broken when one measurement is made, because if one measurement is made, the other will still correlate. So you have to consider the breaking of entanglement locally, or you have to throw out time ordering like you suggest. (which I like )

 

[Boing3000]

If we consider a time ordering of measurements, and Alice measures first, then we would still measure a correlation with Bob's later measurement.

Not if Alice choose another angle, and from Bob perspective either. If they just measure the same angle twice, they both get 100% matches (at that angle at that local place). That's what all non-entangle photon do anyway..

 

[Boing3000]

Your 1 & 2 above don't fix anything (there is nothing to fix)

The thing to fix is precisely that you don't know the YES/NO angle. But you can very well know(decide) the other site angle.

You can certainly force - as you suggest - Alice's measurement to occur first.

That's for sure, and that's important.

But the correlation function - i.e. the prediction of QM - is no different regardless of the time ordering.

I never said otherwise. But we are discussing the second measurement at each site.

You can't observe a photon as being YES at an angle (say 0 degrees) and then re-observe it a few times and suddenly it is NO at that angle.

And that's why the OP logically conclude that if the entanglement persist, it would be trivial to communicate FLT.

Hopefully you see the problems there, such as the photon not existing anymore.

Of course, but all we need to observe is the 50% photon remaining that should be 100% conserved by the second filter (which they are)

Even an electron that was spin up will continue to be spin up unless it interacts with something else. Again, I hope you see my point, which is that the OP example is not realistic on other levels.

That's those other levels I am interesting in. You know I have build my own "Beat a Bell's challenge" (with non-local hidden variable, and real logic (not imaginary/n'th dimensional)), and I would like to make it as close to physics as possible... (for example I can switch on or off the entanglement "persistance").
I am interested in adding other options...

 

[entropy1]

Not if Alice choose another angle, and from Bob perspective either. If they just measure the same angle twice, they both get 100% matches (at that angle at that local place). That's what all non-entangle photon do anyway..

Yes, the entanglement is broken after the measurement, locally. So for Alice to speak of a broken entanglement, she must have done her measurement, and for Bob to speak of broken entanglement, he has to have done his. By the way, I ment correlation in the ensemble of photons.

 

[stevendaryl]

I think I'm saying that you can't claim entanglement of the whole system is broken when one measurement is made, because if one measurement is made, the other will still correlate. So you have to consider the breaking of entanglement locally, or you have to throw out time ordering like you suggest. (which I like )

Entanglement is a correlation between distant measurements. I don't think it really makes sense to say that it's broken locally. It can't be broken for Alice and not for Bob.

Let me try to make it more precise. Suppose that we produce a twin pair of anti-correlated spin-1/2 particles at event $e_0$. One particle goes to Alice, and the other particle goes to Bob. Suppose that Alice measures her particle's spin twice, at event $e_1$ and later at event $e_2$. Bob measures his particle's spin twice, at events $e_3$ and $e_4$.

Then, regardless of the ordering in time of $e_1$ and $e_3$, their results will be strongly correlated. In particular, if Alice and Bob choose the same axis, $\vec{a}$, to measure spin relative to, then they are guaranteed to get opposite results. But the results at events $e_2$ and $e_4$ will not be strongly correlated. (They'll still be correlated, but not in the extreme sense that they will be guaranteed opposite results for measurements along any axis).

 

[entropy1]

Entanglement is a correlation between distant measurements. I don't think it really makes sense to say that it's broken locally. It can't be broken for Alice and not for Bob.

Let me try to make it more precise. Suppose that we produce a twin pair of anti-correlated spin-1/2 particles at event $e_0$. One particle goes to Alice, and the other particle goes to Bob. Suppose that Alice measures her particle's spin twice, at event $e_1$ and later at event $e_2$. Bob measures his particle's spin twice, at events $e_3$ and $e_4$.

Then, regardless of the ordering in time of $e_1$ and $e_3$, their results will be strongly correlated. In particular, if Alice and Bob choose the same axis, $\vec{a}$, to measure spin relative to, then they are guaranteed to get opposite results. But the results at events $e_2$ and $e_4$ will not be strongly correlated. (They'll still be correlated, but not in the extreme sense that they will be guaranteed opposite results for measurements along any axis).

I'm saying that after one of the two measurements, say $e_1$, Alice's, the entanglement is (A) broken or (B) not broken. In case of (A), what does that mean for Bob? If he is yet going to measure, and the entanglement is broken before that, there will be found no correlation (in an ensemble) when he measures. If he has already measured, then there will be a correlation. However, the correlation should not depend on if Bob already has measured or not, the more if there is no time-ordening. You can't say the entanglement is broken 'at a certain moment'. You can say it is broken 'after both measurements'.

In case of (B), the entanglement is not broken after measurement of Alice, since the principle of symmetry, Bob's measurement won't break it either, which is impossible.

Interpreting this one could say that (1) there is no time ordening, or (2) breaking entanglement cannot depend on the time of measurement of the other and is hence local.

The effect of breaking the entanglement is a result of projecting the state on the eigenvector of the operator, which is a local effect.

 

[stevendaryl]

I'm saying that after one of the two measurements, say $e_1$, Alice's, the correlation is (A) broken or (B) not broken.

I was trying to substitute a more precise claim, and you're going back to a vaguer statement. What correlation are you saying is broken or not broken?

As I said, the very first measurement by Alice is correlated with the very first measurement by Bob, regardless of what times those measurements are made.

In case of (A), what does that mean for Bob? If he is yet going to measure, and the correlation is broken before that, there will be found no correlation (in an ensemble) when he measures.

Correlation with what? As I said, the first measurement made by Bob will be strongly correlated with the first measurement made by Alice. The second measurement made by Bob will not be strongly correlated with the second measurement made by Alice. Are you in agreement with these two claims? What further question is there about it?

 

[entropy1]

What correlation are you saying is broken or not broken?
Correlation with what?

Sorry, should be: "entanglement".

As I said, the very first measurement by Alice is correlated with the very first measurement by Bob, regardless of what times those measurements are made.

I agree. My reply concerns if you claim the "entanglement is broken by measurement".

 

[PeroK]

I'm saying that after one of the two measurements, say $e_1$, Alice's, the entanglement is (A) broken or (B) not broken. In case of (A), what does that mean for Bob? If he is yet going to measure, and the entanglement is broken before that, there will be found no correlation (in an ensemble) when he measures. If he has already measured, then there will be a correlation. However, the correlation should not depend on if Bob already has measured or not, the more if there is no time-ordening. You can't say the entanglement is broken 'at a certain moment'. You can say it is broken 'after both measurements'.

In case of (B), the entanglement is not broken after measurement of Alice, since the principle of symmetry, Bob's measurement won't break it either, which is impossible.

Interpreting this one could say that (1) there is no time ordening, or (2) breaking entanglement cannot depend on the time of measurement of the other and is hence local.

The effect of breaking the entanglement is a result of projecting the state on the eigenvector of the operator, which is a local effect.

This is not correct. With entanglement, you have a two-particle system. There are not two separate quantum systems here: there is one system of two particles. A measurement is a measurement of the system and the system is either a pair of entangled particles or it is not.

After the measurement, you have two one-particle quantum systems.

 

[stevendaryl]

Sorry, should be: "entanglement".

I know. But it's not clear what "entanglement is broken" means, although it is perfectly clear what it means to say that this measurement is or is not correlated with that measurement.

 

[entropy1]

This is not correct. With entanglement, you have a two-particle system. There are not two separate quantum systems here: there is one system of two particles. A measurement is a measurement of the system and the system is either a pair of entangled particles or it is not.

After the measurement, you have two one-particle quantum systems.

That is correct. I guess the state of the system is the solution.

 

[DrChinese]

It is worthwhile to recall: no one understands the precise underlying mechanism by which entanglement effects occur. When a measurement separates a particle from an entangled state, no one knows exactly how that mechanism operates either. Instead, each person has a mental model - that is what we are really discussing.

 

[Derek P]

Sorry, should be: "entanglement".
I agree. My reply concerns if you claim the "entanglement is broken by measurement".

Entanglement is indeed broken by the first observation. Consider an entanglement where observation of the two particles has two possible outcomes [1,1] or [0,0]. If the first measurement returns a 1 then the possibility of observing [0,0] has gone for good. So although the correlation has yet to be observed pending the second measurement, the entanglement no longer exists because one of the original possibilities no longer exists.

 

[Derek P]

After the measurement, you have two one-particle quantum systems.

Or an extended entanglement where the unmeasured particle is entangled with the first particle, its detector and the obligatory cat.

 

[entropy1]

Entanglement is indeed broken by the first observation. Consider an entanglement where observation of the two particles has two possible outcomes [1,1] or [0,0]. If the first measurement returns a 1 then the possibility of observing [0,0] has gone for good. So although the correlation has yet to be observed pending the second measurement, the entanglement no longer exists because one of the original possibilities no longer exists.

And if you look at the last measurement first?

 

[Derek P]

And if you look at the last measurement first?

Looking doesn't come into it. An epistemological interpretation of entanglement avoids all the problems because by the time the observer gets all the results, the physical objects could have colluded somehow. Emphasis on the somehow. And also on the fact that such a model studiously avoids saying what is going on in the real world up until that point. <shrug> Some people are happy with that approach.