Realism vs Locality in Quantum Entanglement

[Markus Hanke]

As all observers will agree about the recorded macroscopic results of the interactions taking place at the two detectors, entanglement is not an observer-dependent phenomenon. What is observer-dependent is which measurement happened first.

Noted, with thanks. This makes sense to me.

If you find this situation paradoxical (as most people would)

I don't actually find it paradoxical, in fact, unless I am missing something crucial, I don't even see how it could possibly be considered a paradox at all. In my mind, I do not think of entanglement as involving any type of "action" over and above the initial interaction required to create the entangled state; to me it is just simply a statistical correlation between measurement outcomes thereafter. In many ways it is really just a trivial observation - like encountering someone whom you know is married on the street. If you meet the man, you know automatically that the other spouse ( who is potentially far away ) must be female, and vice versa, simply on account of your knowledge of them being in a heterosexual marriage. Your meeting that person does not in any sense of the word cause the gender of the absent spouse. At the same time, which of the two you meet is a random occurrence - it could be the man or the woman, with equal probability. The very same for the spin states in your example - whether, for a single measurement, you get spin-up or spin-down is essentially random, but you always know that, if the particles are entangled, the same measurement performed on the other particle must yield the opposite result, since they can't both be in the same spin state. There is no cause and effect here, just a correlation, and as we know correlation does not necessarily imply causation.

Correct me please if I am wrong on any of this.

 

[Nugatory]

I do not think of entanglement as involving any type of "action" over and above the initial interaction required to create the entangled state; to me it is just simply a statistical correlation between measurement outcomes thereafter.

That's right - there is no paradox as long as you're just thinking in terms of the statistical correlation. However, in your married couple example, you are going way beyond that; to be stay within the spirit of "just a correlation" you would have to say that the other spouse will be of the opposite sex if you look, but if you don't look you can make no statement.

To see the distinction you need more than one measurable attribute. Suppose every couple consists of a man and a woman, one blue-eyed and one brown-eyed, one tall and one short. If you and I each encounter one member of the couple, you meet a man and I meet a blue-eyed person... It does not mean that you met a brown-eyed man and I met a blue-eyed woman.

 

[zonde]

In many ways it is really just a trivial observation - like encountering someone whom you know is married on the street. If you meet the man, you know automatically that the other spouse ( who is potentially far away ) must be female, and vice versa, simply on account of your knowledge of them being in a heterosexual marriage. Your meeting that person does not in any sense of the word cause the gender of the absent spouse. At the same time, which of the two you meet is a random occurrence - it could be the man or the woman, with equal probability.

Exactly this type of model is proved incompatible with quantum mechanics predictions by Bell. If you are not familiar with Bell's theorem you can try DrChinese website
Even more it is ruled out by recent loophole free Bell inequality experiments.

 

[ddd123]

Correct me please if I am wrong on any of this.

The correct meaning of "paradox" is an impossible or contradictory result, so of course there's no paradox here, since nonlocal correlations are a logical possibility. But your marriage example isn't applicable because you cannot assign definite properties to each particle of the pair throughout their travel towards the measurement instruments, so that they can be compatible with the experimental results (i.e. local hidden variables).

 

[Markus Hanke]

Thank you Nugatory, zonde, and ddd123. So basically, what I missed in my classical analogy was that quantum objects do not actually have definite properties, until a measurement is performed, so the "entanglement information" isn't somehow carried along with them in the form of some hidden attribute or variable - it couldn't be, because no definite state has been "selected" yet. So for example, for two entangled electrons, they would exist in a superposition of all possible spin states for that particular setup, from the moment of entanglement until such time when a measurement of the spin is performed - only then is it meaningful to actually consider them as having a specific, definite spin state, and only then does entanglement become apparent when one compares the outcomes of measurements. In that case, it seems wrong to even consider the pair as separate, independent entities - the pair of electrons is really just one system described by one wave function, so if you perform a measurement on any part of the system, you collapse the wave function as a whole for both particles - which naturally reveals the correlation between the outcomes. Does that sound about right ?

Apologies if this stuff seems really basic to all of you, but as an amateur I am really trying to understand it correctly, as opposed to all the imprecise pop-sci babble that's out there. It's one thing to see it written down as a mathematical statement, but it's quite another to actually get an intuitive feel for how entanglement manifests. So thank you for your time in explaining it.

 

[ddd123]

You can have nonlocal hidden variables in Bohmian mechanics. So "existing in a superposition" is one way to interpret the result, the most common, but it's far from being the only one, there are lots.

 

[Markus Hanke]

You can have nonlocal hidden variables in Bohmian mechanics. So "existing in a superposition" is one way to interpret the result, the most common, but it's far from being the only one, there are lots.

Ok, noted

 

[ebos]

Just because something can't be perceived by our senses and our machines (yet) doesn't mean it is not real.

 

[DrChinese]

Just because something can't be perceived by our senses and our machines (yet) doesn't mean it is not real.

The issues about quantum realism have little or nothing to do with senses or machines. It has to do with classical ideas about the physical world, which we already know are inaccurate and a poor description of it. The quantum view is more accurate and more useful.

 

[Neandethal00]

@Markus Hanke

I have read most of the Link you posted. I have heard/read the following sentence so many times

"a particle doesn't have a certain property (the particle is not IN one state or another) until that property is measured."

Many say this is the conclusion/interpretation of QM.

Will it be totally wrong if we rephrase the sentence as

"Nothing exists until it interacts with other(s)".

Another way of stating this words is

"To a particle nothing exists until it interacts with another particle(s)".

It changes its properties due to interaction.

Interaction(s) of an object is the proof of its existence. I think double slit experiment
can also be explained following this logic.

 

[Neandethal00]

@Markus Hanke

I have read most of the Link you posted. I have heard/read the following sentence so many times

"a particle doesn't have a certain property (the particle is not IN one state or another) until that property is measured."

Many say this is the conclusion/interpretation of QM.

Will it be totally wrong if we rephrase the sentence as

"Nothing exists until it interacts with other(s)".

Another way of stating this words is

"To a particle nothing exists until it interacts with another particle(s)".

It changes its properties due to interaction.

Interaction(s) of an object is the proof of its existence. An interaction always change something about the object.
I think double slit experiment can also be explained following this logic.

 

[Markus Hanke]

"Nothing exists until it interacts with other(s)".

In my mind, that wouldn't be consistent with things like the Aharonov-Bohm effect; a particle traverses a region of vanishing E and B fields, before hitting a detector. It turns out the particle's trajectory is indeed affected by the underlying vector potential, even though no interactions in the usual sense have taken place. So clearly, the particle did exist prior to interacting with the detector.

It guess this ultimately depends on what exactly you mean by "interaction".