Why is it difficult to accept the idea of nonlocality in quantum mechanics?

[elbeasto]

I read that measurement of one property affects the outcome of the other entangled item. My question, why must the measurement of one system "affect" the other? More specifically, why can't their properties through time and space be determined by some initial action.

For example, let's say we have 2 hypothetical batting machines that can hit a racket ball in _exactly_ the same way 100% of the time inside a racket ball room. We set each machine to hit identically. The initial position of impact( from bat to ball ) would determine its spin, position, and momentum. Because the hits are _identical_, all of their properties will be identical. ( The rooms are identical in dimensions as well )

Why can't we apply this to photon properties? I know photons will be opposite of each other but the idea I am trying to get across is the initial force that sends the photons on their way determines all of their properties which is why they are always in sync with each other. Not that their is some hidden communication between them. Can someone please explain why I am wrong?

 

[elbeasto]

I have an add-on to the question. As I began to think about the dimensions of the room I realized that if everything was not completely symmetrical, the spin of the hypothetical ball would be different once it hit a wall unlike its counterpart. Is this a known issue?

 

[DrChinese]

Why can't we apply this to photon properties? I know photons will be opposite of each other but the idea I am trying to get across is the initial force that sends the photons on their way determines all of their properties which is why they are always in sync with each other. Not that their is some hidden communication between them. Can someone please explain why I am wrong?

What you are describing is something which is classical. Photons are quantum particles and they obey a different set of rules. One of those is the Heisenberg Uncertainty Principle (HUP). No one really knows the mechanics of the HUP (if there are any). But the rule itself applies in this situation and it leads to limits.

A variation on your idea was introduced in 1935, known as EPR. A response 30 years later, by Bell, demonstrated clearly that common classical ideas are incompatible with this rule. Specifically, the idea of locally separate properties with well defined but unknown values is incompatible with the HUP. This is often phrased as saying that local hidden variables are incompatible with quantum mechanics.

 

[elbeasto]

Thanks for your response DrChinese!

Could you elaborate a little bit? For example, I argue that some initial force send 2 photons on their way as mirror images of each other and there is some hidden variable ( something within the photon for example ) that could predict any subsequent motion of the photon. Because I argue that there are some hidden variables that will correctly and accurately determine any property of the photon at any given time, your argument is HUP. But if I believe that I can define any property ( more importantly all properties simultaneously ) of a photon at any given time, why should I believe in HUP? I guess what I am asking is can you clarify why HUP disproves hidden variables? From what I read it appears as though limits are created that say as the accuracy of one measurement goes up the accuracy of the other goes down. So I can understand how my scenario violates that. But why did HUP disprove this scenario to begin with? It appears as though my scenario is only wrong if I accept that HUP is valid.

Sorry if my questions are poorly phrased. I seem to be having a problem with just phrasing my question exactly as I see it in my head.

 

[jtbell]

I argue that some initial force send 2 photons on their way as mirror images of each other and there is some hidden variable ( something within the photon for example ) that could predict any subsequent motion of the photon.

A "local hidden variable" model like this can easily account for the results of experiments with pairs of photons in which they pass through polarizers that are oriented at 90 degrees with respect to each other. However, nobody has (yet) come up with such a model that can also account for the results of experiments in which the polarizers are oriented at other angles.

J. S. Bell proved a theorem (now known as Bell's Theorem) that says that all local hidden variable theories must make predictions which satisfy a certain inequality (the Bell inequality) whose exact form depends on the experimental setup. Quantum mechanics generally makes predictions which violate the Bell inequality. A number of experiments have tested this. They have all (so far as I know) given results which agree with the predictions of QM and violate the Bell inequality.

Some people argue that the experiments have "loopholes" which make the Bell inequality violations illusory; that is, if we could perform an experiment that closes all the loopholes, the results would satisfy the Bell inequalities (and falsify QM in the process). Some argue that Bell's Theorem uses incorrect assumptions and therefore its conclusion is also invalid. These are both small-minority positions, as far as I can tell. People like to argue about it in this forum. Look at the two threads on the first page of this forum, which have the largest number of posts.

You can find links to key papers on Dr. Chinese's site, including proofs of Bell's Theorem for specific experimental situations. My own first exposure to this stuff was via an article by N. David Mermin, with a title something like "Is the Moon There When Nobody Looks?" (you can find a link to it on Dr. Chinese's site).

 

[DrChinese]

Thanks for your response DrChinese!

Could you elaborate a little bit? For example, I argue that some initial force send 2 photons on their way as mirror images of each other and there is some hidden variable ( something within the photon for example ) that could predict any subsequent motion of the photon. Because I argue that there are some hidden variables that will correctly and accurately determine any property of the photon at any given time, your argument is HUP. But if I believe that I can define any property ( more importantly all properties simultaneously ) of a photon at any given time, why should I believe in HUP? I guess what I am asking is can you clarify why HUP disproves hidden variables? From what I read it appears as though limits are created that say as the accuracy of one measurement goes up the accuracy of the other goes down. So I can understand how my scenario violates that. But why did HUP disprove this scenario to begin with? It appears as though my scenario is only wrong if I accept that HUP is valid.

Sorry if my questions are poorly phrased. I seem to be having a problem with just phrasing my question exactly as I see it in my head.

Your question and phrasing are fine. In EPR, this argument was made much as you make it. Their idea was that you could take 2 particles (I always use photons because they are easy to visualize) that have conserved properties, and then predict their outcomes in such a way as to violate the HUP.

So suppose we have entangled Alice and Bob. You measure p on Alice and q on Bob to desired precision. From the conservation rules, you could then know both p and q for Alice (and Bob too). That would violate the HUP, if true.

But actual experiments show that when you measure p on Alice, you make Bob's q completely indeterminate. And vice versa. Now, keep in mind that this is NOT obvious at first glance. Initially, you come to the conclusion that you have violated the HUP. Analysis of the data will seem as if it shows this. However, it took Bell (1965) to show us that this is not so. If you are not familiar with Bell, I recommend you read about it and then return here. We can then continue the conversation. If you have already read about it, then let me know that.

Or see my web page which explains a bit more:

Theorem:An Overview with Lotsa Links
[/url]

 

[zonde]

I read that measurement of one property affects the outcome of the other entangled item. My question, why must the measurement of one system "affect" the other? More specifically, why can't their properties through time and space be determined by some initial action.

I you ask why some people think that photon properties can't be determined by some initial action then I will suggest reading this https://www.physicsforums.com/showpost.php?p=2817138&postcount=18". I like the explanation in this post as I used one very similar for myself.

 

[elbeasto]

Thanks for the post Zonde, it makes sense to me and I hate to sound stubborn ( and I know I will ) but it just hard for me to accept that way of thinking. I appreciate your help though!

I read your page as well DrChinese, it was very helpful. I had previously read about Bell, but your page was a bit easier to follow. One thing I notice as I am reading through all of these articles and papers is that I find myself clicking a lot of links. It is like I am reading an array of pointers and each pointer is simply pointing to another location in the array. As I read, there are a lot of terms that are either unfamiliar to me or vague to the point that I cannot make a connection between the term and its relevancy to the topic at hand. Which leads me to look up ( or in most cases re-look up ) the term.

Rather than take up your time, because I just started this independent digging not too long ago so I will have a _lot_ of questions, could you point me in a good starting direction? I have already started watching the Feynman lectures. Someone somewhere said that was a good place to start and it appears as though it is but the more advice the better.

One last question though ( haha ) at what point( or size ) do we say an object stops following the laws of what I think is referred to as classical and begin following laws of quantum mechanics? Please correct me if I am wrong but it appears that anything smaller than an atom follows quantum mechanics but anything bigger follows classical. If I am wrong, what object larger than an atom follows quantum mechanics. If I am not, what is the limit of the size any quantum particle can be before it starts to follow classical laws?

Thanks again!

 

[zonde]

Thanks for the post Zonde, it makes sense to me and I hate to sound stubborn ( and I know I will ) but it just hard for me to accept that way of thinking. I appreciate your help though!

If by "way of thinking" you mean nonlocality of QM then I do not suggest that you should accept that. I was just trying to provide most simple illustration that the problem is quite real (as that is sometimes quite hard to accept that from more sophisticated explanations).

In reality photon experiments that attempt to prove this theoretical finding rely on so called fair sampling assumption. That is photon pair detections are not perfect in experiments, there is quite large part of unpaired single photon detections at both sites. Now fair sampling assumption simply states that if you would have detected all photons so that all photon detections would be paired between sites you would have obtained the same statistical result as in the case of inefficient detection.
If you do not accept this assumption then photon experiments doesn't prove nonlocality.

One last question though ( haha ) at what point( or size ) do we say an object stops following the laws of what I think is referred to as classical and begin following laws of quantum mechanics? Please correct me if I am wrong but it appears that anything smaller than an atom follows quantum mechanics but anything bigger follows classical. If I am wrong, what object larger than an atom follows quantum mechanics. If I am not, what is the limit of the size any quantum particle can be before it starts to follow classical laws?

There are performed double slit experiments with fullerene molecules (C60) so I don't think that size is very good guide for separating classical and QM domains.