Einstein's missed opportunity to rid us of 'spooky actions at a distance' | Hacker News Reader

Einstein's missed opportunity to rid us of 'spooky actions at a distance'

sciencex.com

Dr. W. M. Stuckey shows that the origin of quantum entanglement is none other than Einstein's own Principle of Relativity (No Preferred Frame of Reference).

Einstein's famous 1905 paper on Relativity applied this principle to Translational Frames, showing it requires the Universal constant c (speed of light) to be the same in all such frames. Were it not, the frame in which c was highest would be the only frame at rest.

But the same principle must require there to be no Preferred Orientation. This leads to the requirement that Planck's constant h be the same in all frames. If the Stern-Gerlach experiment could give results between +h and -h, then the orientation producing the maximum value would be a preferred frame.

And because of that, when Alice and Bob measure entangled quantum particles, their combined results must violate Bell's inequality.

But to my mind, the biggest take-away is that Einstein's Principle of Relativity absolutely requires that conservation can only be on average.

All right, that's a ridiculously condensed summary. Enough to make your head spin :-)

The title paper is for general audiences, and references the original paper at https://www.nature.com/articles/s41598-020-72817-7.pdf

I think I have a reasonable mental model of how certain QM processes work by visualising waves in transit coalescing as particles when measured.

But I have absolutely no idea how to visualise entanglement. Any tips? Or do we just have to shut up and calculate?

My understanding is,

Looking at a single object with a fixed angle camera produces similar observations to an entagled pair when the pair is in a similar configuration and where each object in the pair is observed by their own camera except one of the cameras sees a negated result.

Often you can visualize entanglement as a superposition of independently evolving states, when those states are solutions of Schrodinger equation.

The key to understanding entanglement for me was to understand that the wave function does not live in physical 3-D space, it lives in configuration space. A wave function that lives in physical 3-D space is a special case that applies only to a system that consists of a single unentangled particle. In that case, physical 3-D space and configuration space are the same. But in general, a wave function for N particles will live in a 3N-dimensional configuration space.

It seems to me that the configuration space of which you speak essentially explains the wave/particle duality. My B.S. Physics education from years ago never explained it as such as far as I can recall. I'm far from any expertise understanding, but this makes much sense. Any further pedagogical commentary would be most appreciated!

> configuration space ... explains the wave/particle duality

I don't think so. You can have a wave function in physical 3-D space as a (very common) special case and you still have the wave-particle dichotomy.

Why do you think configuration space explains WPD?

Really WPD more fundamentally arrives from the “conversion” of the probability wave (living in configuration space as you put it) into properties in 3D physical space. Measuring the QM system is analogous to a Fourier transformation and has the same mathematical limitations arising from a Fourier transform from wave to discrete space. Depending on the question you ask (tied to the convolution function) you get either discrete particle answers or wavelike answers. The probability wave does live in 3D space AFAICT, but the QM properties like spin, charge, momentum etc are tied into that 3D space and form a combined configuration space.

The really odd part to me is that at macro scales the probability waves collapse neatly into classic physics in 3D space, but still react in quantum fashion at small local atomic scales. As in the configuration spaces generally can only be determined for small subsystems but not a whole macro system without the “conversion” step.

The “conversion” of the probability wave--now that's much more specific with regards to the "configuration" space and well put. When I had first learned about WPD it really didn't strike me as odd or mysterious as historically described. In effect, I thought, both realities must be there simultaneously--that sounds like normal QM nature. The collapse of the duality simply depends on the "conversion" function. On this note, are there any fascinating presentations on WPD that take the "mystery" out of it?

Configuration space leads to the Everett interpretation, which leads to us being a superposition of observers, all of whom observed something that looked like a particle.

The way I see it is particles go all possible ways until you narrow things down with a measurement or something along those lines - an interaction that fixes where the thing is. (So in the two slit experiment with one particle it goes through both slits until its position is pinned down by hitting the screen.)

In the entanglement experiment described the particles have angular momentum every which way until the angular momentum of one is pinned down by measurement whereupon the other one is also pinned down to the opposite by conservation of momentum. There is still a sort of spooky action at a distance when that happens or perhaps a splitting of the multiverse 'at a distance' into many worlds where the spins point different ways.

I wish, for the love of god, that the early pop science about quantum physics hadn't used the phrase "when observed" when describing the wave collapse. If they had said "when interacted with by another force" (or anything along those lines), we wouldn't have loads of new age dummies talking about how the particle "knows" it is being observed by a conscious mind. No quantum woo, no Deepak Chopra.

I don't think you're giving the people who peddle bullshit enough credit. They'll just figure something else out.

> "when interacted with by another force"

But that is not the same, right? I mean, if it interacted with a force and no observation was made, the wave function doesn’t collapse, does it? Honest question (to avoid any defensiveness, I should disclose that I don’t subscribe to panpsychism).

> or anything along those lines

Any suggestions?

> if it interacted with a force and no observation was made, the wave function doesn’t collapse, does it?

Any two systems interacting will cause the collapse. It doesn't matter if the system is attached to a scientist or not.

> Any suggestions?

No, I'm a software developer, not a quantum physicist. :)

> Any two systems interacting will cause the collapse.

I suppose that means if a photon, say, is reflected by a mirror, that should collapse its wave function and any measurements after that should not have any effect on it?

Maybe it should be qualified what kind of interaction collapses wave function?

> I'm a software developer, not a quantum physicist.

Great, I’m not a quantum physicist either—yet here we are, talking about quantum physics!

> I suppose that means if a photon, say, is reflected by a mirror, that should collapse its waveform and any measurements after that should not have any effect on it?

I'm not sure if that example is the right one to use, but yes, that's roughly my understanding.

Okay. I actually thought the paradox was that wave collapses only if the photon is influenced by measurement activity but I’s been a while since I was reading up on this topic so I may need a refresher.

I suppose if we can calculate exactly how a given force would influence a photon, that would be essentially the same as “measuring” it.

Turns out you need to run your quantum computer in near-zero-kelvin conditions to avoid accidental 'observations' from collapsing the precious waves you're using to factorize the number 6.

That’s not what Schroedinger’s equation says though, it can combine the wave functions of two quantum systems, say a particle and a force, and show how they evolve with no problem, in fact that literally what it does. Quantum mechanics simply doesn’t have a complete account of collapse to a specific state.

Modern explanations tend to have interactions causing decoherence rather than observers and collapses.

Doesn’t decoherence just spread entanglement out to the rest of the world such that you end up in a situation analogous to Many Worlds?

It depends upon the QM interpretation you subscribe to. For example with many-worlds, there is no collapse (at all).

For "collapse" to occur, the interaction of the quantum system must be with a classical system. A classical system being something big, noisy that can be well approximated by classical mechanics.

Simple interaction between two systems doesn't cause "collapse" it makes the two systems become entangled. Classical systems are a bit contagious in this sense, anything that gets entangled with them becomes classical.

To be a bit more precise, this distinction between classical and quantum is a bit our fault. Everything is quantum at a fundamental level, classical system is one for which we do have not have a precise knowledge of the state of the system, instead we have a coarse representation. This should make more obvious in which way "classicalness" is contagious. Since the knowledge of a part was coarse, the knowledge of the newly entangled system is also necessarily coarse.

> if it interacted with a force and no observation was made

Consider the classic two-slit interference experiment. Whether the electron goes through the left or right slit can be treated a single qubit. Use a controlled-NOT gate to copy that qubit onto a second storage location, without observing either. Optionally drop the second qubit into a black hole to head off any claims about supposed future observations. Allow the electron to continue. Do you still observe interference pattens as in the non-copying version of the experiment? Why or why not?

Interesting throught experiment. I believe you would still see the interference since there are still two possible paths.

Using the word copy in conjunction with C-NOT is slightly misleading as the copies do not behave independently.

Tongue-in-cheek explanation: Maybe whoever wrote our simulation used shallow copy when they should have done a deep copy.

> Using the word copy in conjunction with C-NOT is slightly misleading as the copies do not behave independently.

That's what the word "copy" means. If you flip a coin and copy that bit, you will observe that those copies do not behave independently either. If you want independent bits, flip two coins.

Similarly, "erasing" a (qu)bit technically consists of performing a exchange operation between it and a known-zero bit. In typical electronic computers, this would generally involve diffusion-like exchanges between the voltage level in a memory capacitor (such as a FET gate) and that on the GND rail, which has a much greater effective number of bits and therefore will stay mostly zero, but eventually requires a thermodynamic expenditure of known-valued bits (aka negentropy) from some external source to maintain its voltage level / bit zeroness. (This is rather simplified; there's lots of other sources of known-zero and known-one bits getting depleted and replenished, and the exact accounting depends on how you interpret various physical states information-theoretically.)

> In typical electronic computers,

That should have had a "for example" in front.

Calling it a copy seems to me to violate QM's no cloning theorem, but maybe it's just that we have slightly different definitions of 'copy'.

The no cloning theorem is just the rigorous quantum mechanical version of the fact that if you flip a coin and write down the result, then copy what you just wrote down, you don't get two independent coin flips.

I'm not sure you can really blame pop science - the concept seems to be there in the Copenhagen Interpretation eg. see point 4 https://en.wikipedia.org/wiki/Copenhagen_interpretation#Prin...

I think one of the appeals to actual physicists who do experiments is that is how things are usually set up - there's some equipment that makes a measurement and the Schrödinger equation stuff till collapse thing gives the correct result for what is observed. Obviously the universe got on ok for billions of years before physicists evolved so it's a simplification of reality.

Have you heard of Bohemian or pilot wave theory for QM? It's very rarely discussed except in darker corners of the internet but it's surprising to me it's not used as visualization. https://en.m.wikipedia.org/wiki/De_Broglie%E2%80%93Bohm_theo...

I didn’t know I was writing for the “darker corners of the internet”!

https://arstechnica.com/science/2017/07/a-brief-history-of-q...

Oh, yeah, Ars is pretty dark, but where it really comes up a lot is amongst child pornographers and terrorists. In fact, it's been shown that when a new pilot wave paper lands, productivity (such as it is) in those areas drops precipitously for a time. In fact it's a robust enough result that it forms a strong argument for increasing science funding.

More seriously, I interpreted "darker corners of the internet" in the parent to be a bit tongue in cheek, but to generally be indicating fora where there's a higher ratio of layman to expert, and crackpot to serious practitioner. There was no claim that it isn't discussed outside of that setting, just that it occurs more frequently there (as a proportion of QM discussions in general). This squares with my (poorly informed) general impression.

"have you not heard of this rarely talked-about theory that exists in the dark corners where only whspers live? Well here's a link to the wikipedia page"

>Or do we just have to shut up and calculate?

Edwin James had some interesting commentary on things like this:

"From his reply to EPR, we find that Bohr's position was like this: 'You may decide of you own free will, which experiment to do. If you do experiment E1 you will get Result R1. If you do E2 you will get R2. Since it is fundamentally impossible to do both on the same system, and the present theory correctly predicts the results of either, how can you say that the theory is incomplete? What more can one ask of a theory?'

While it is easy to understand and agree with this on the epistemological level, the answer that I and many others would give is that we expect a physical theory to do more than merely predict experimental results in the manner of an empirical equation; we want to come down to Einstein's ontological level and understand what is happening when an atom emits light, when a spin enters a Stern-Gerlach magnet, etc. The Copenhagen theory, having no answer to any question of the form: 'What is really happening when - - -?', forbids us to ask such questions and tries to persuade us that it is philosophically naive to want to know what is happening. But I do want to know, and I do not think this is naive; and so for me QM is not a physical theory at all, only and empty mathematical shell in which a future theory may, perhaps, be built."

https://bayes.wustl.edu/etj/articles/cmystery.pdf

...and which he goes on to makes some interesting observations about the Bell Inequalities.

"Just as Bell revealed hidden assumptions in vonNeumann's argument,so we need to reveal the hidden assumptions in Bell's argument. There are at least two of them, both of which require the Jeffreys view point about probability to recognize..."

Typo: Jaynes, not James

Ack. I can't believe I messed that up. He wrote an awesome book: "Probability Theory: The Logic of Science".

https://www.amazon.com/Probability-Theory-Science-T-Jaynes/d...

And there is a website with more information and a collection of his papers:

https://bayes.wustl.edu/

https://bayes.wustl.edu/etj/node1.html

You may find Caticha's Entropic Dynamics interesting:

https://arxiv.org/abs/1005.2357

http://dl.icdst.org/pdfs/files1/77964f05542451c01e8e420e975d...

That does look interesting. Thanks.

> But the same principle must require there to be no Preferred Orientation. This leads to the requirement that Planck's constant h be the same in all frames. If the Stern-Gerlach experiment could give results between +h and -h, then the orientation producing the maximum value would be a preferred frame.

OK, I think I understand that.

> And because of that, when Alice and Bob measure entangled quantum particles, their combined results must violate Bell's inequality.

Could you be slightly less ridiculously condensed here? Give a one-or-two-paragraph, accessible-to-the-semi-layman explanation of why this means the result must violate Bell's Inequality?

> But to my mind, the biggest take-away is that Einstein's Principle of Relativity absolutely requires that conservation can only be on average.

And the same request here. Why does the principle of relativity require that?

I can give it a try. The linked nature article had a lot of details.

> > But to my mind, the biggest take-away is that Einstein's Principle of Relativity absolutely requires that conservation can only be on average. > > And the same request here. Why does the principle of relativity require that?

The way I read this comment was that "the principle of relativity cannot conserve angular momentum on a per-trial basis".

In a Mermin Device a pair of entangled spin particles is set to two Stern-Gerlach experiments. The two particles has net (spin) angular momentum of 0 because that's was the net angular momentum of starting material. But if you measure the angular momentum of the two particles in two non-parallel directions, and if we also require that the only answers you are allowed to get are +hbar/2 or -hbar/2, then the sum of the angular momentum you get by adding +/-hbar/2 times one direction plus +/-hbar/2 times a different direction can never be 0.

If angular momentum cannot be preserved on a per-trial basis, then I suppose it must be preserved on average, because, I suppose if it isn't preserved on average, then I don't think you can say that angular momentum is preserved at all.

> And because of that, when Alice and Bob measure entangled quantum particles, their combined results must violate Bell's inequality.

> > And because of that, when Alice and Bob measure entangled quantum particles, their combined results must violate Bell's inequality. > > Could you be slightly less ridiculously condensed here? Give a one-or-two-paragraph, accessible-to-the-semi-layman explanation of why this means the result must violate Bell's Inequality?

The really short answer is that if we preserve the angular momentum on average then it entails that the correlations we observe from Mermin Device must match the correlations predicted by quantum mechanics, and therefore violate Bell's inequality for the same reason that predictions of quantum mechanics do.

In more detail, if we take the results of a measurement where Alice measures angular momentum in the vertical direction and Bob measures the angular momentum off vertical by theta degrees where Alice gets a result of +hbar/2, then in order for angular momentum to be preserved, Bob's measurement would have to be -cos(theta)hbar/2.

Of course Bob is only allowed to get hbar/2 or -hbar/2, so if we want angular momentum to be preserved on average then when we take an ensemble of trials, and filter out only those trials were Alice measures hbar/2, then the average of all of Bob's measurements for those trials should be -cos(theta)hbar/2. That requires that the probability Bob geting hbar/2 when Alice does is (1-cos(theta))/2 (= sin^2(theta/2)), which I believe is the value predicted by quantum mechanics. Once you have the predictions made by quantum mechanics, a violation of Bell's inequality follows by the usual arguments.

Thanks. That clarifies things a bit.

Invariance under velocity changes boosts the whole apparatus. And speeds less than C are observable, and do change w.r.t. an unboosted observer.

Rotational invariance would rotate both the source and the detector, and there would be no surprise that the possible results and statistics over them are unchanged.

The quantum surprise is that rotating the source relative to the detector leaves the possible results unchanged (though the statistics do change).

Say you put a red marble and a blue marble in two envelopes. You randomly post one envelope to Australia. One year later, you open the other envelope. You now know the color of the marble in Australia.

What's the difference between this and quantum entanglement?

actually the marbles change colour every 1 second and you can take them really far apart, meaning one can go at a really high speed and distance trough universe, so the time would have dilated for it. when you open them both have same color

That sounds logical. Relativity theory allows you to age one object faster than another by changing their relative speed. Nobody would claim there was information travelling faster than light in your example.

What makes people say information travels faster than light with quantum entanglement?

If the marbles were changing colour at random but still alwasy different colours, that would suggest info is doing so.

If they change color in sync then that is a known property of both marbles. Say you have a marble that is blue if the number of seconds is even and red otherwise. Knowing the color of one marble is enough to know the color of the other marble without information travelling between the marbles?

Not my area. My understanding: colours sync exactly on measurement (when you look at them).

> Knowing the color of one marble is enough to know the color of the other marble

I guess so.

> without information travelling between the marbles?

The marble colours are in sync on measurement. Somehow that info has travelled instantaneously. You just can't use it to send information, at all..

above is just my understanding. I have no background in this. Just a programmer.

They don't.

This would be a 'local hidden variable' theory. According to wikipedia these have largely been ruled out:

Most advocates of the hidden-variables idea believe that experiments have ruled out local hidden variables

Source: https://en.wikipedia.org/wiki/Bell%27s_theorem#Bell_inequali...

It seems that hidden variables have always been made to be simple hidden states. But we now know about RNGs and seeds and such.

A shared RNG seed is essentially entanglement.

This delves more into complex hidden variables, that normal analyses ignore: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC137470/

> A shared RNG seed is essentially entanglement.

Classical entanglement, which is not good enough to explain quantum entanglement.

It can explain quantum entanglement, seems to be a version of superdeterminism.

In QM, experiments show us that entangled particle spin probabilities vary non-linearly with the angle between detectors (even if those detectors are far apart).

This means that either: 1) locality is broken.. state is somehow transmitted faster than the speed of light between particles. 2) realism is broken.. god plays dice with the universe

But there's also a 3rd, which is: the choice of detector angle is not an independent variable (a necessary assumption for Bell's inequalities to hold).. instead the state of the universe is pre-determined and the experimenter's choice of detector angle is known beforehand so there is no need for spooky action at a distance. This isn't a very popular explanation since it provides no reason as to why we don't see this weird lack of independence elsewhere.

Well, if superdeterminism is an explanation there is nothing to explain ;-)

Apparently this common explanation is wrong. If you actually measure one, the other actually changes. Although you can't use that to send information, so einstein's faster-than-c restriction isn't violated.

Maybe this will help https://html.duckduckgo.com/html?q=bell%27s%20inequality%20s... I imagine the youtube links might be more comprehensible.

If the other changed, you could ue those changes to send information by varying time between changes (pulse width modulation). Instead when you measure one you go from "not knowing which one is red or blue" to "knowing color of both".

You can't.

Exactly. Because they don't change. If you detect one color, it stays the same.

I think you can only tell if it’s changed by measuring the thing and comparing the results with the other person.

AFAIK you don't need to compare. It's like random number generator, but you have two complementary generators. When one generates 1, the other generates 0. You don't know what you will get next, but you know what was last and you know that other person got opposite number.

Only if the measurements align. If they do you get perfectly correlated numbers. If the angles are 90 degrees apart you get completely unrelated numbers.

The problem comes in when the angle between your two measurements is anything else. The chance that the measurements match is based on the cosine of that angle. There's no way for this to happen if the measurements are independent.

If you try to write two equations, where the first equation takes the secret particle state and first angle and gives you 1 or 0, and the second equation takes the secret particle state and second angle and gives you 1 or 0, you won't be able to reproduce the odds you get in the real world. Only equations that know both angles will work.

No, just because something changes doesn't mean it can be used to send information. You need changes you can control.

The thing about the change of "colour" in this analogy is you don't know in which direction it changes. So let's say you observe you "marble" through a "purple filter", which gives has:

- a 50% chance of being transparent to your marble (corresponding to a red-blue superposition marble collapsing to a purple marble)

- a 50% chance of being opaque to your marble (corresponding to red-blue superposition marble collapsing to a green marble).

The issue is that when you learn your marble is purple, while you know with 100% certainty the marble in australia is green, there is no way you can send information to Australia using that. This is because the other 50% of the time, your marble will be green, and the marble in Australia is purple.

So if I'm sitting in Australia, when I measure the marbles in my envelopes with purple filters, all I see is purple marbles 50% of the time and green marbles 50% of the time no matter what measurements you are performing at your end. So you can't send me messages by performing measurements at your end because you can't change the statistics of those measurements.

But you'll know the answer to every measurement I performed, if you've measured the other marble with a purple filter too.

So, how "it changes from state we don't know to a different state after measurement" differs from "we don't know what state it is, but after measurement we know"? How do you know state changes after measurement when you don't know which state it is before measurement? Does it really change, or do our knowledge of that state changes? That's why I say that state doesn't change, we only know what state it is after measurement.

Only if you assume a single universe. If, on the other hand, the universe split into both options, then nothing "changed" when you looked at one of them. You just no longer have a way of interacting with the sub-universe where the ball was the other color.

This is fairly far from my field, but as I understand it, that would be a hidden variable interpretation of QM [1], and specifically in that analogy a local hidden variable theory. That's what Einstein himself wanted.

There is a famous test, Bell's inequality [2], that specifically rules out local hidden variable interpretations of QM.

Nonlocal hidden variable interpretations, such as De Broglie - Bohm theory [3], are potentially still on the table, however.

It is somewhat ironic that Bell's theorem is sometimes presented in popular media as a general disproof of all hidden variable theories, in a context where locality is taken for granted -- because Bell himself seems to have been partial to nonlocal hidden variable theories. An article by the same Mermin mentioned in the OP is worth a read, on this subject [4].

[1] https://en.wikipedia.org/wiki/Hidden_variable_theory

[2] https://en.wikipedia.org/wiki/Bell%27s_theorem

[3] https://en.wikipedia.org/wiki/De_Broglie%E2%80%93Bohm_theory

[4] https://cqi.inf.usi.ch/qic/Mermin1993.pdf

Anyone know if anyone has followed up on Caroline Thompson's work after she passed away?

"The Chaotic Ball: An Intuitive Analogy for EPR Experiments"

https://arxiv.org/abs/quant-ph/9611037

Haven't found anything yet myself, but would love to know -- that looks quite interesting!

Wikipedia's Simple English page on Bell's inequality actually has a nice overview of a simplified way of thinking about quantum entanglement: https://simple.wikipedia.org/wiki/Bell%27s_theorem

The way I understand it, this is exactly the simplified (and incorrect) explanation.

Say SOMEONE ELSE puts the marbles in two envelopes and sends them to you and your friend in Australia. (it's someone else because we don't actually create the entangled particles, we just "get" them)

The marbles being red and blue (or both red or both blue, depending on what you're measuring) from the beginning would be a LOCAL hidden variable. It's local because it's been predetermined at the moment of creation and the marbles carry the property on themselves and it's hidden because you don't know how/why the person putting the marbles in those envelopes decided those colors and you can't see them until you open the envelope (measure the particle).

This way if you don't open your envelope, your friend's envelope contains a marble that's 50/50 red or blue and the color will be the predetermined one no matter what you do with your marble at home. So whatever decides the marble's color has nothing to do with your marble, it's local to the friend's one.

The actual measurements work differently. It's been experimentally proven many times that at the moment you look at your marble, the other marble's 50/50 probability of being red and blue shifts substantially to, for example 75/25. And that's without it having any way of knowing that you've seen your marble. So there are hidden variables that we don't understand, but they're not local. They somehow affect both marbles.

In real life there aren't only two colors and the probabilities aren't those nice numbers, but you get the principle.

Can you explain how is this shift in probability measured?

The experiment is repeated many times, statistics is computed over results, and gives probabilities and correlation.

How much work has been put into ensuring that the observed samples aren't biased?

What do you think? Is mainstream physics about to go "oh shit, we didn't take into account that we might be biased. Thank you random person on the internet for pointing this out".

So me and my friend are continuously getting those envelopes. I am opening all the envelops one after another and found that they are distributed 50/50 to red/blue.

Same if my friend is opening the envelops.

Now for all the opened envelops if I have got 10 red balls. Now if my friend open the paired envelops, he will probably get 7 blue and 3 red.

My observation of the balls had an effect on his side and shifted probabilities on his side.

If that's what you mean, what does observation or measuring even mean? How do the balls know the envelop has been opened.

This explanation seems like it can't be correct and must be a simplification as well. If this was measurable in a way that's described here you'd be able to transmit information.

I always imagined the two "marbles" as possibly being two similar but differing clocks instead. The clocks will align more or less often depending on how similarly they're set and how fast each run. With this analogy you can come up with any distribution that fits your fancy.

Its probably a silly analogy but it lets me cling my notions of no spooky action.

> The actual measurements work differently. It's been experimentally proven many times that at the moment you look at your marble, the other marble's 50/50 probability of being red and blue shifts substantially to, for example 75/25. And that's without it having any way of knowing that you've seen your marble. So there are hidden variables that we don't understand, but they're not local. They somehow affect both marbles.

The rest of your explanation was super easy to grok (thank you!) but this part I can't wrap my head around. If the balls can be red or blue, and it's 50/50 before, how would the probability go to 75/25? I would expect it to either stay at 50/50 (no change) or to 100% (because the other ball is known).

Can you elaborate on this part? This is really fascinating.

This is completely incorrect, to the point where what you were trying to correct was actually more accurate, though incomplete.

The usual setup is that for any given axis, each person always measures 50:50. Measuring your own doesn't change the odds of the other.

Knowing the _results_ of your own does. For the same axis, the correlation is exact. For axes with an angle theta between them, we get a correlation of R ~ cos(theta/2).

The upshot is that there is no underlying (classical) probability distribution that can give rise to this that can explain things for all measurement axes. This is sometimes glossed as "correlation without correlata".

I was trying to make the original marbles in envelopes comparison work a little better (the "holy f*ck this doesn't make sense" aspect), not to actually explain the phenomenon. I hope no one got confused. There's plenty of better sources and smarter people to get the accurate explanation from.

It's not my field, but I remember reading that your example doesn't represent entanglement because when put into envelopes, one marble is already red and another blue.

In quantum entanglement they are both truly and really random until you measure one. And it's not random in a sense that you closed your eyes when putting them into envelope. They actually both don't have a "selected" color. They "snap into one of two colors" when you measure (look at) one. And the "unbelievable" thing is that when you measure one, the other one immediately snaps into opposite color, no matter how far it is.

I don’t think this is correct. The two particle system is prepared in a perfectly known state (e.g. both spins up). There’s nothing random about it. Randomness only occurs at the measuring device, if it not aligned with the direction of the spin of the incoming particle.

Nope. They don't "have" their own state until one of them is measured. But they do have a correlated state which exists before measurement, which says they have opposite/same states. The individual states arise only after measurement. I'm not a physicist, but wrote a Quantum Simulator.

Roughly speaking: if you can see your marble through a purple filter, the one in Australia will turn out to be perfectly green.

You could have something similar to quantum entanglement if these were some strange kind of marbles which cannot have a well defined color and size at the same time and are magically linked.

If you look at the marble you got and it's red (or blue) the size becomes indeterminate. Focusing now on the size you will find it's large or small, but the color becomes indeterminate. It could be red the next time you look at it.

When you take your entangled marble, look at the color and see it's red you know the other marble is in the "blue" state (and the entanglement is broken). If someone looks at the color of that marble you know they will find it's blue. But if they look at the size before looking at the color it could be large or small (and looking now at the size of your marble will tell you nothing about it) and if they look at the color later it could be red or blue.

In the classical case, if there is a large red marble in one envelope and a small blue marble in the other it doesn't matter in what order you look at the color and the size. You will always know what the other person found.

In the quantum case, if both look at color first they will find complementary colors. If they both look at size first they will find complementary sizes. But the second measurement will be uncorrelated. And if they make the measurements in a different order, everything will be uncorrelated.

Sabine Hossenfelder gives the best explanation to address this that I've found

https://www.youtube.com/watch?v=j6Mw3_tOcNI&ab_channel=Sabin...

Welcome to Bell's Casino. You and your partner will be playing our famous two-coin game today. We have hidden two coins under these two opaque cups. You and your partner are to guess the orientation, heads or tails, of both of the coins. Guess correctly and win $1. Guess incorrectly and lose your $1 bet.

In order to help you out, after you two have made your guess we are going to give you two a chance to back out and lose nothing. After your prediction we are going to reveal one coin to you and another coin to your partner. Together you and your partner will have an opportunity to back out, but the catch is that you two are not allowed to communicate!

Instead of communicating, you can raise either a red flag or a green flag after seeing your coin. Similarly, your partner can raise either their red flag or their green flag after seeing their coin. If you both raise the same colour flag, the game keeps going and we see if you win or lose. If you both raise different colour flags, the game stops and you lose nothing.

To ensure you don't cheat, we've separated you and your partner by 200 million kilometers and you have one minute to raise one of your flags after seeing your coin, otherwise you lose the game. (Alternatively you are your partner are separated by 400 meters and you have 100 nanoseconds to raise one of your flags.)

Good luck.

---

The above casino game cannot be beaten using envelopes of marbles, but it can be beaten (i.e. positive expected value) using envelopes of entangled particles. See quantum pseudo-telepathy.

This argument is called "Bertlmann's Socks"

https://en.wikipedia.org/wiki/Reinhold_Bertlmann#Bertlmann%E...

The other replies explain why it's wrong, but here's a link to Bell's refutation for good measure

http://cds.cern.ch/record/142461/files/198009299.pdf

What seems to be missing from the replies posted so far is the notion of coherence, and the choice of measurement basis.

What separates a coherent "quantum" superposition, say, |0> + |1>, from a probabilistic "non-quantum" 50:50 mixture is that I can choose a measurement basis in which the coherent state always yields a definite result, say "1", whereas measuring the mixed state always yields a 50:50 mixture of "0"s and "1"s.

A continuous sweep of the angle of the measurement basis generally results in an interference pattern, the amplitude of which can be used to assess the fidelity of the quantum state.

(I get paid to work on quantum communication and related experiments.)

If you do the experient with an entangled source and two Stern Gerlach detectors oriented the same way it's not so interesting a bit like the marbles in envelopes - either they both red of green. The interesting bit is if you rotate them a bit the correlation varies like cos(the angle) between them. So correlation 1 at 0 degrees, 0 at 90 degrees, -1 at 180 degrees and about 0.98 at 10 degrees. But how does nature or whatever know the angle between them when they are far apart? In most 'hidden variables' scenarios the correlation at 10 degrees is more like 0.89 or a linear change and that is basically the essence of Bell's theorem and experiments - you can't get the correlations without the particle at one end kind of knowing the set up at the other, or 'non locality' as Bell called it.

So many confusing answers in this thread, but yours is clear and enlightening (and factually correct, afaik).

Your answer seems to focus on the key un-addressed subtlety.

Why is the difference in orientation of the detector necessarily linear? What is the control aspect of this experiment where classical-system shows this linear pattern? Or can the argument be made more fundamentally?

You can think of an example where things are classical, the particles start with some definite orientation randomly determined at the start and if they are within say 45 degrees of the angle of the detector they go one way, over 45 the other and it's not so hard to figure in that case it will vary linearly. As to how to prove the general case I don't know. Try Bell's paper?

As an aside I don't think the classical 'hidden variables' situation has a prefered orientation which contradicts the premise of the featured article that you get the odd entanglement effects so as to not have a prefered orientation.

Thanks for the help. I just want to point out this [detector orientation] is likely a big area where non-physics people might get tripped up:

If you told me causally this detector which measures electrons/photons/whatever and varies by the cosine of the orientation, I don't think any (non-physics person) would bat an eye; it seems like a pretty normal thing a sensor might do.

The phenomenon is statistical. In your example (which is a bit too simple but still), the analogous surprise would be something like each party guesses what's in the envelope before they open it, and the accuracy of their guess is too high to explain by chance.

I'm a little sceptical of these authoritative-sounding "overviews" that are essentially plugs for the author's thesis - the paper providing the justification for the title quietly inserted in this case right at the end of the article.

Personally, I'd prefer third party summaries of the thesis when it has been established as an interesting contribution, and the original article to stick to what is actually accepted by the mainstream; or at the very least to be more up-front that this is actually based on a new paper by the author.

That's not to say this paper is wrong - I'm not remotely qualified to judge (and it happens regularly in articles plugged on HN); I just find the way these things are presented as a bit iffy.

I agree, motivation is important (because it reflects on the quality of the content).

Does someone have a good explanation/intuition for why you cannot exploit quantum entanglement to send information faster than light?

If me observing the particle in Australia alters the probability distribution of your particle in USA, can't I only observe the particle when I want to communicate 1 and never observe it when I want to communicate 0?

Edit: thanks a lot for the answers! I guess it boils down to the fact that the Australian guy cannot condition his decision on the (unknown) spin of his particle -- if he could (eg: had access to the local hidden information) then he would be able to update the USA's probability distribution instantaneously and use it to communicate

When the person at the other end looks at their particle, and sees it either 1 or 0 and has no other information, how do they know if you've looked at your particle or not?

And even if they would know, due to the distance and general relativity, there is no concept of one of the two persons looking first. It could be that from Alice's point of view Bob looked later, while from Bob's point of view Alice looked later. So, in which direction did the information go?

Initially, your particle in Australia and its entangled twin in the USA exist in a superposition of 0 and 1. When you "measure" the state of your particle, you force it to assume a definite state, and entanglement forces the other particle to assume e.g. the opposite definite state ("spooky action at a distance"). This allows you to synchronize information across large distances, but you cannot send anything, because you cannot chose the outcome of the quantum measurement.

If you can guarantee a random distribution of the color of the balls, it seems like a decent way to handle secure key echange.

Yes, that is in fact one of the practical ways to do quantum key distribution.

What if we manage to get our whole world into a superposition and only collapse to the world where the particles did contain the message we wanted? Sorry... I've been reading a lot of Greg Egan lately ;)

I have a bag with two balls, one red, one blue. I randomly send one of the balls to Australia, one of the balls to the USA. The moment you look at your ball in Australia, you also know the color of the ball in the USA. How would you use that fact to send information?

The crux with quantum mechanics is that you can show (very easily, understandable by the layman, look up the Bell inequality) that in the quantum case the ball only takes on a color the moment you look at it, and so instantaneously is setting the color in the USA as well.

However, the basic setup still applies. You cannot send information by merely observing something.

How does a ball know that it has been observed and in effect changing the color of the other paired ball?

That's the "spooky action at a distance" (in the case of the entangled particles, obviously; with the two balls it's not so spooky). Spooky.

Maybe this is a basic question - but what I don’t understand is why this is called “spooky” action.

My intuition is you have two particles, and you don’t know what concrete states they are in, but you know all possible states (that may be represented as some sort of system of equations).

By observing a single particle you unlock a variable in that system of equations and can therefore solve the whole thing. To me it would be more straightforward to say the concrete state of the particle is simply unknown until it is observed. The concept of superposition seems like an overly complex description for this phenomenon.

I understand my view is wrong, but I don’t understand how I’m wrong

What you are describing is a hidden variable theory - i.e. there is some concrete state of the particles, but it is hidden.

John Bell demonstrated that in order for a hidden variable theory to make predictions in agreement with quantum mechanics, it must have nonlocal interactions, which means any workable hidden variable theory must also be pretty spooky.

https://en.wikipedia.org/wiki/Bell's_theorem

Here's a layman's explanation of why "hidden local variables" theory doesn't match experiments:

https://www.wired.com/2014/01/bells-theorem/

In other words, modeling particle pairs as having matching static hidden "meta data" in them doesn't work. They do act as if there is instantaneous communication between the particles, but in a limited way that prevents us from using them for instant communication. Quantum mechanics is a weird tease, having magical properties that always serve up loopholes when we try to leverage the magic for real-world benefits. The quantum universe seems built by insurance lawyers who are masters at screwing consumers with fine-print when they go to make a claim.

I observe my particle here, and in doing so its state is decided.

The state of the entangled particle over there, a light year away (for example) is also decided. Instantly. Faster than the speed of light. Nothing travelled from here to there. No particle, no photon, nothing. How does over there "know" that I did something over here?

Sure feels kind of spooky.

To me it would be more straightforward to say the concrete state of the particle is simply unknown until it is observed.

It's not just unknown. It's undecided. It has no concrete state. It's not that it IS a one or a zero and you just don't know it. It's not yet been decided whether it's a one or a zero, but as soon as the decision is made for one of the entangled particles, the decision is also made for the other one, a light year away. Instantly. Spooky.

When you observe it, you collapse the wave-function of which color the ball has into a particular value (red or blue). Before you observed it, the ball was in a superposition of the two colors. And this collapse instantaneously also collapses the wave-function to the American ball.

Now, that is obviously not true for macroscopic objects like balls. Those are not in a superposition of colors until they are observed, but it is true for quantum objects like electrons.

> Now, that is obviously not true for macroscopic objects like balls. Those are not in a superposition of colors until they are observed, but it is true for quantum objects like electrons.

But then what is it that can I do with two entangled electrons that I can't do with two literal billiard balls known to be different colors than one another?

You cannot paint a billiard ball with a superposition of colors.

But what is the implication of that? What do those words cause to be different in practice?

The randomness in the observed color of the ball. (The observable represented by the "color operator" could have two or more distinct eigenvalues.)

You can prepare a pair of photons in a state such that when you measure the polarization of both of them along the same axis, for whatever direction you want to choose, you get the same result. But they are entangled, each photon considered separately is not in a well-defined state.

You can also prepare two photons in the same state, so the have the same polarization for some direction chosen at that time. But the measurements along other axis won't be perfectly correlated (if they are correlated at all).

The red/blue color example is too simple to be interesting.

Obligatory "not a quantum physicist," but the only way to observe something is to throw something at it (e.g. a photon) and see what bounces back. The problem is that when you throw something at it, you're interacting with and affecting the ball.

This is not quite correct. There are ways to gain information about a quantum state without interacting with it: https://en.m.wikipedia.org/wiki/Interaction-free_measurement

Do (non)measurements taken this way 'collapse the wave function' anyway? Or can you only get information that is still open to change during the actual measurement?

"Collapse" is just one way to explain how measurements happen. These are "actual" measurements, because they give you actual information.

Yeah, this is another popular non-explanation. (It does not explain why the ensuing randomness is subject to strict and perfectly deterministic laws).

Parallel realities. In one red goes to the US in another blue. In another vice versa. They decohere to one with someone who's seen red one who's seen blue.

The ball example is insufficient and misleading. It is unfortunately too simple as you need 3 inputs and 2 output to demonstrate the effect (aka Mermin device).

The way to think about it is a box with 3 buttons. There is no such thing as 'observation', the only way you can interact is to push one of the 3 buttons and as a result the box will output either a red or green light.

You must push a button to get the light, but the button may mutate the internal state of the box. Using this model, there's nothing special about human or conscious observation. Every interaction via a particle or otherwise is simply pushing a button.

The crazy thing is.. no matter how clever an algorithm you write to drive the lights from the buttons, you cannot match the observed probabilities. (100% if the same button is pushed, 25% if different buttons are pushed).

> Using this model, there's nothing special about human or conscious observation. Every interaction via a particle or otherwise is simply pushing a button.

But there is something kind-of-special about the box with the buttons and the lights.

Not every interaction is simply pushing a button that lights one lamp or another. Keeping the analogy, the result of an interaction between two particles may be a combination of the "red on", "green on" states. You need to keep adding particles to have a box with buttons and lamps that works as expected.

I've always wondered - what constitutes an "observation"? How does the particle know it has been observed?

Particles don't know anything, but you have to interact with it in order to observe it. You have to bounce a photon off of it or something like that in order to get any information out of it.

https://en.wikipedia.org/wiki/Quantum_decoherence

You can't observe something without sending information. In order to make an observation, you must interact with whatever is being observed, so that information about the interaction can come back to you.

In the bag example above, we can observe the Australian ball and know the color of the American ball, and we cannot use this interaction in Australia to send information to America. But we cannot avoid sending information to the Australian ball when we observe it.

>> You cannot send information by merely observing something.

This is, at the least, very poorly phrased. As explained above, not only can you send information by observing something, it's impossible not to do so. The question here is where the information goes.

My understanding is that any kind of measurement will do, it doesn't matter how you get it. You could use photons to do the measurement but there are other ways which all have the same result.

It's a mystery how the particle "knows" (In other words, nobody knows when the wave function collapses) but one popular interpretation is that the particle exists in all states, i.e. in a pure description of reality. When any quantum system interacts with it, then it becomes entangled with the result of that measurement, branching it into a new universe (edit for clarification: a new world where it was as if it was never a wave, and it was always a particle). That's my understanding of the many-worlds theory.

That entanglement propagates across nearby particles, so it doesn't have anything to do with eyes or consciousness. If the air molecules around your body interact with the particle then that entanglement propagates through your body and places you in the new world.

Re: When any quantum system interacts with it, then it becomes entangled with the result of that measurement, branching it into a new universe. That's my understanding of the many-worlds theory.

This is a case of a simple theory that indeed models the mystery well. However, it seems "wasteful" in that it would branch into gazillion trees of reality. In Occam's Razor, does "simplicity" include quantity of "stuff" needed? Because sometimes the brute force algorithm/model is the "simplest" if we ignore quantity of stuff and time, such as bubble-sort. Bubble-sort is one of the simplest sorting algorithms known, but is inefficient from a time and resource standpoint.

If there are "free" dimensions to spare out there, then the "wasteful" multi-verse model may not really be wasteful. We humans are used to thinking in terms of economic trade-offs, and a model that uses up large quantities of space/time rubs our instincts wrong.

If true, the theory means that in some universe somewhere I'm a billionaire who married a supermodel.

If the universe is a mathematical object there being an infinity of universes isn't any more wasteful than there being an infinity of integers for example. From Occam's point of view it's simpler if all integers exist rather than there being a cap if that were even logically possible. So yeah go supermodel!

Re: If the universe is a mathematical object...

Math is a modeling technique, not a "thing". To me it doesn't make sense to say the universe "is" math. Maybe it's a machine "running" math notation (programming code), but that's not the same as it "being" math.

(Is "God" the server admin?)

It's been hypothesised: "the physical universe is not merely described by mathematics, but is mathematics" https://en.wikipedia.org/wiki/Mathematical_universe_hypothes...

not proven of course.

Suppose God is running multiple identical instances of the universe on several machines. Are we in a particular one?

We wouldn't know or care and it would make no difference. It only becomes an "issue" if there is a mutation or glitch that breaks symmetry.

The universe isn't mathematical, it is explained by math, a country is not Chinese because I wrote a tour guide in Chinese about it. Infinite universes isn't really applicable here, you're thinking of a growing block universe. A simpler point is a block universe. https://en.m.wikipedia.org/wiki/Eternalism_(philosophy_of_ti...

But if Chinese was the only language (which is a more correct analogy), the country might in fact be Chinese.

>>"However, it seems "wasteful" in that it would branch into gazillion trees of reality."

In a way, it could be interpreted as very efficient. Only the branches where some "measurement" is done are "calculated". I suppose the others are garbage collected at the end of time, or something like that.

And maybe it's not a tree, but a graph of universes. In the same way that a universe split in two, two universe could also fuse into one when they share the previous state. Somehow it feels like this have to be connected to reversible vs. non-reversible computation.

Ah.. it's a good feeling being a fearless dilettante.

Re: Only the branches where some "measurement" is done are "calculated". I suppose the others are garbage collected at the end of time, or something like that.

But that's adding complexity back into it. You are increasing complexity of the theory/model by adding a complex cleaner/trimmer in order to reduce the quantity of resources consumed.

I think you're mixing up metaphysics and human intuition with what the math describes. The current math says there may be essentially infinite worlds created in infinite time, where yes there is least one in which you are a billionaire married to a supermodel. The only constraint is in the properties of nature (e.g. a world will never be created in which an electron has 0 spin).

However, I agree with you that it seems implausible because it implies absurd situations like, there is a world in which someone lives a life of celebrity because every time they roll some dice it always lands on 6, and every time they flip a coin it lands on heads, etc.

Even worse, many-worlds doesn't really solve the problem anyway - it still doesn't explain WHY you only observe one result, when the Schrodinger equation predicts several. That is, why can't you see the other worlds?

That doesn't seem like a question that can be answered mathematically, does it? That's like asking, why do electrons have a spin of 1/2? Why is the speed of light 299,792,458 m/s? These are just properties of the universe.

Not really. It's the same question as the measurement problem: Schrodinger's equation predicts that a particle can exist in many places at the same time, with different amplitudes, and interact with particles in all those places. However, if we want to predict the particle's movement after it encounters a detector, we need to update the wave function to set its probability to 1 at the position of the detector and 0 everywhere else - otherwise, our predictions are measurably wrong.

Now, the question is: what causes this discontinuity in the equations of motion? Why is interaction with a detector different than interaction with another particle? Many Worlds simply reframes this problem, but doesn't get rid of it. In MWI, you would say 'the particle moves in all universes according to the wave function, until it interacts with a detector, possibly interfering with versions of itself in other universes. Then, when it encounters the detector, the world line of the detector splits - in some universes it passes the detector, in others it doesn't. However, it no longer interacts with other versions of itself,so we must update the wave function inside the universe where it passed the detector'.

> Now, the question is: what causes this discontinuity in the equations of motion? Why is interaction with a detector different than interaction with another particle?

> Many Worlds simply reframes this problem, but doesn't get rid of it.

Maybe I'm misunderstanding. It's like asking "why is there a difference between me jumping in a swimming pool and someone else jumping in it? I don't get wet when someone else is swimming." The difference is... one of you is in the pool. It's not going to spontaneously make the other person wet.

In MWI the difference is that if it interacts with a particle, you're not entangled, the particle is. If it interacts with a detector then you're entangled. So, there is no difference except for what gets entangled.

What that means is the wave function can only appear to collapse when you entangle. If some particle entangles, it will collapse for that particle and branch into a new world, but you're not in that world; for you it's still a waveform.

Edited for clarity.

> In MWI the difference is that if it interacts with a particle, you're not entangled, the particle is. If it interacts with a detector then you're entangled. So, there is no difference except for what gets entangled.

I don't think that is the whole story. If you want to predict the motion of a particle correctly, you still need to update the Schrodinger equation after interaction with the detector, but not after interaction with another particle. And this is independent of whether you personally look at the detector or not, even if the detection occurs outside your light-cone. This is evident from the fact that MWI still needs both the Schrodinger equation and the Born rule to accurately predict experimental results.

> What that means is the wave function can only appear to collapse when you entangle. If some particle entangles, it will collapse for that particle and branch into a new world, but you're not in that world; for you it's still a waveform.

But this is not true for macroscopic objects. The motion of a detector, and indeed even the motion of a particle after it interacts with a detector, does not behave like a wave, regardless of whether I have ever interacted it. Even if the interactions are space-like separated from myself, I can still predict them with classical mechanics, and confirm when the data finally reaches me. For example, I can predict the location of a particle in a double slit experiment if I know that there is a detector at one of the slits, regardless of where in the universe that experiment happens. How can I be entangled to a detector that exists outside my past light-cone? But then, I can't predict the outcome of a double slit experiment without a detector near the slits, regardless of how close I am to the experiment.

This still shows to me that there is an observer-independent collapse happening when a particle interacts with a detector, where we don't have a physical description of what a detector actually is.

Don't you have the same problem in classical mechanics? Let's say you're standing at the edge of a pond, and you see waves rippling across the surface. The deviation in height of the surface of the water is described by h = cos(r + t) where r is the distance from the centre of the pond and t is the current time.

Why can you see the solution of the equation for the entire surface of the pond at once, but only for a single instant of time at any given moment?

It's not the same thing, because classical mechanics explicitly models the time - it can predict that at time T the system is in one state, and indeed when I look at a the system at time T, I see it in a single state.

Conversely, the Schrodinger equation gives an amplitude to the same particle/wave at many locations at time T. However, when you look for it at time T at all of those locations at once, you only find it in one of them. If you perform the experiment many times, you will find it at all of those locations some amount of the time. But then, if you try to use the Schrodinger equation to model movement before AND after interaction with the detector, you will not be able to find the particle at any position that doesn't match what the detector initially saw.

That is, say the Schrodinger equation predicts the particle has the same amplitude at locations X and Y. Then, after interacting with something at locations X and Y at time T1, it will have some amplitude at locations X1, X2, Y1, Y2 at time T2.

Now, if we try an experiment where the interaction at time T1 happens with a particle, and you have detectors at positions X1, X2, Y1, Y2, you will find it with equal probabilities at any of the 4 locations. However, if at X and Y there is a detector, and you detect the particle at X, it will never be found at positions Y1 or Y2. You have to update the Schrodinger equation after you find out that the particle is found at X, which is never how classical mechanics work.

Isn't the problem that you're only looking at the system in a single world W, when viewing all solutions requires viewing it in multiple worlds?

I mean, I get that time is a little different in that you will eventually experience and remember all possible solutions as you stand there watching the system, because classical time is a linear chain of events. In the multi-world case, it's a branching chain, and your experience and memories of the different solutions are stuck in their own branches.

That does make worlds weird and different from the other dimensions, but we accepted time as being weird and different from space for a very long time.

> Now, if we try an experiment where the interaction at time T1 happens with a particle, and you have detectors at positions X1, X2, Y1, Y2, you will find it with equal probabilities at any of the 4 locations. However, if at X and Y there is a detector, and you detect the particle at X, it will never be found at positions Y1 or Y2. You have to update the Schrodinger equation after you find out that the particle is found at X, which is never how classical mechanics work.

This makes total sense if it's actually a wave and the particle is merely a solution for a particular world W. The detector didn't change anything about the wave. It just coupled you to the wave system earlier, so now your branch of the many-world tree can only see the subset of solutions that correspond with whatever you detected. The only thing that has changed, though, is your ability to see the other solutions. You branched earlier, so now each branch you exist in only sees a subset of the full solution.

That said, I am not a physicist. The many worlds explanation was just the first thing that actually made sense to me about quantum mechanics. It's so conceptually simple.

> This makes total sense if it's actually a wave and the particle is merely a solution for a particular world W. The detector didn't change anything about the wave. It just coupled you to the wave system earlier, so now your branch of the many-world tree can only see the subset of solutions that correspond with whatever you detected. The only thing that has changed, though, is your ability to see the other solutions. You branched earlier, so now each branch you exist in only sees a subset of the full solution.

This explanation only works if either the detector is not itself made of particles, or if there is a detector wave that you could become entangled with by observing.

But the first one can essentially be discarded, and the second one is not experimentally confirmed. The equations happen the way I described whether you observe the detector or not. The detector could be hundreds of light years away from you, but you would still be able to predict what happened after the particle hit it with classical mechanics. So one particle's interaction with a detector instantly branches at least its entire future light-cone, but two particles interacting doesn't have the same effect. So at what scale does this happen? Or in what conditions?

Re: doesn't explain WHY you only observe one result

If the alternative universes are in different dimensional planes, it's pretty obvious why we couldn't observe them.

But why then can particles in different universes interact with each other (or even themselves)?

I'm not sure cross-universe communicating is necessary in the multiverse model. The splitting just resembles communication from our perspective in that it makes the probabilities look "rigged".

That still doesn't explain interference patterns in double-slit experiments, especially in double-slit experiments with a single photon/electron at a time. Those can only be explained by the particle/wave traveling through both slits and then the two versions interacting with each other.

Ever heard of pilot wave theory? https://en.m.wikipedia.org/wiki/De_Broglie%E2%80%93Bohm_theo... It's an easy visualization that can be shown with speakers and liquid

But to fit experimental observations, the "wave" would have to be faster than light.

Yeah. On the other hand it does give decent intuition about how certain interactions would result, especially if talking about massive particles traveling much slower than light.

I remember reading an article here a while back that involved a macroscopic re-creation of the double slit experiment results, but where mere observation remained possible, because light did not sufficiently influence the substrate. In that experiment the particles were droplets traveling on top of a set of waves, working in the pilot wave fashion.

Any attempt to use anything of similar scale to the particles to observe which slit the drop went through would break the interference pattern, but mere light did not, allowing one to visually see how a pilot wave style interpretation could work, if it were not for that whole (photons travel at the speed of light, so these would need to be faster than light propagating pilot waves) thing.

Indeed it looks like flubert linked a video from an earlier study of the same basic mechanics, prior to the more recent one that included the double slit experiment replication.

I was under the impression that the double slit experiment with the oil drops does not replicate the quantum mechanics expected interference pattern.

http://math.mit.edu/~bush/wordpress/wp-content/uploads/2017/...

Here's one interesting silicone oil pilot-wave video:

https://www.youtube.com/watch?v=nmC0ygr08tE

> one popular interpretation is that the particle exists in all states, i.e. in a pure description of reality. When any quantum system interacts with it, then it becomes entangled with the result of that measurement, branching it into a new universe (edit for clarification: a new world where it was as if it was never a wave, and it was always a particle).

The two-slit experiment contradicts this. You get different results depending on when you perform the observation(s).

So the new world is a world where the particle was originally a wave, and became a particle when it was observed. Not a world where the particle was always a particle.

"A new interpretation of quantum mechanics suggests that reality does not depend on the person measuring it"

https://phys.org/news/2020-10-quantum-mechanics-reality-pers...

https://www.mdpi.com/2073-8994/12/9/1533

Phew!

Re: "How does the particle know it has been observed?"

It's as if God's code looks something like:

    if (event.thisParticle.isBeingObserved()) {
       thisParticle.assignAttributes();
    }

"extreme late binding of all things?"

Lack of compile-time type-checking gives us dodgy politicians and lawyers.

You might find this helpful: https://en.wikipedia.org/wiki/Quantum_decoherence

In many formulations, e.g. multiverse, the apparatus doing the measuring (doesn't have to be a human or anything) becomes entangled with the thing being measured. This is still not super well understood.

I assume, for the sake of my own sanity, that "observation" means the particle becomes a cause for some kind of effect, e.g. colliding with something in a way that changes the something's state. Quantum mechanics experts, please don't tell me it's weirder than that.

In Quantum Mechanics things travel as waves and interact as particles. Between interactions, everything is described by the Schrodinger Wave Function, which evolves to include every possible path. But at the point of an interaction the wave function must collapse to satisfy conservation rules.

Consider an electron fired at a dual slit with a phosphor screen. While traveling from the electron gun, thru the slits, to the screen, the electron is described by a Wave Function. It has no fixed position or momentum. The Schrodinger wave passes through both slits and interferes with itself on the other side. The wave function evolves into a series of lines.

But when the electron interacts with the screen it always appears as a single point. It must do so by the laws of conservation. At the interaction it must have a specific location and momentum in order for there to be conservation of charge, momentum, energy, etc.

This interaction is enough to 'collapse the wave function'. No 'observation' is required.

How does this happen? There is no localized mechanism that can possibly make this work. The conservation laws are not local restrictions. They are universal.

Please note that this is my own explanation of now QM works, and does not necessarily reflect the official position of any school of thought. It does, however, reflect the actual use of Quantum Mechanics, in that systems evolve via the Schrodinger Equation and interactions must obey conservation laws. And No, it cannot explain how entanglement works.

> In Quantum Mechanics things travel as waves and interact as particles. Between interactions, everything is described by the Schrodinger Wave Function, which evolves to include every possible path. But at the point of an interaction the wave function must collapse to satisfy conservation rules.

That would make some sense if by interaction you mean "interaction with the macroscopic environment". When small-enough quantum systems (like two particles) interact there is no collapse and the evolution is unitary.

> This interaction is enough to 'collapse the wave function'. No 'observation' is required.

How do you distinguish the interactions that 'collapse the wave function' from those who do not?

The idea that Measurement = "interaction with the macroscopic environment" is part of the Copenhagen interpretation, not a requirement of QM itself.

Aside: (Personally, I see this more as Bohr's way of dodging questions he had no answer to, and not a viable way to think about Quantum Mechanics. A better answer would have been "I don't know. Let's figure it out." But that was impossible for political reasons. Bohr was being attacked by Einstein for 's sake. He can be forgiven for adopting Ali's "rope-a-dope" tactics if he felt that Einstein was trying to destroy his entire field in its infancy. But I find "there is no quantum world" simply unacceptable.)

Now to answer your question as best I can, an interaction must collapse the wave function when it is required to fulfill a conservation rule. For example, if an electron is captured by a nucleus it becomes bound and emits a photon. This is an interaction that must conserve momentum, angular momentum, energy, and charge. Because of that, the electron can no longer be represented by a non-localized wave function. The universe must concentrate those properties down to a point in order to "do the accounting" necessary for the conservation rules.

No, I don't know how it does that. But then, NONE of the available interpretations answer that question. This indicates to me we are thinking about it wrong.

What I like about Stuckey's paper is that it adds another factor: besides conservation rules the universe seems to require that "measurements" obey the Relativity Principle (No Preferred Frame of Reference). I have yet to figure out how to incorporate that.

Is the idea of “collapsing the wave function” a requirement of QM itself? In that context, a “measurement” would be to be what you call “an interaction that must collapse the wave function”.

And your answer is simply wrong. An excited atom can emit a photon, for example, and the system will still be described by a “non-localized wave function”. It won’t even be well defined if the spontaneous emission has happened or not yet.

The evolution of a quantum system according to Schrödinger’s equation doesn’t violate conservation rules. And, in case it’s not clear, the quantum system described by the wave function in the example above is the atom-photon(-or-not) pair.

You’re definitely thinking about it wrong.

> the quantum case the ball only takes on a color the moment you look at it, and so instantaneously is setting the color in the USA as well

We don't actually know that this is an accurate description of what is happening, although it is consistent with what is happening.

Very likely, the underlying physical process still operates below the speed of light. "Instantaneous" isn't something that makes physical sense in this context.

There have been experiments confirming entanglement. It does not "tell" the particle at the speed of light. The mutual wave function collapses as a whole unit.

it is very easy to measure something entangled at the same time (or at least within a margin that is faster than light travel) and confirm you always get the correct results. If the wave function didn't collapse together, you would get results that break quantum laws, such as measuring two entangled particles with both up-spin.

You misunderstand the objection. The fact that you didn't expand on your definition of simultaneity suggests that you're missing a lot of background here. There are many ways that physics could appear to violate causality with collapse without actually sending information faster than light (e.g. how most of the multiverse theories work).

If the ball only takes a color when it's observed (loosely), couldn't that timing be exploited to convey information?

By viewing the color of the ball you’re not notifying the other ball to change its state(in a way that can be measured). You will only know that once the ball is observed it will have a certain color.

The only way I can foresee this being used to transmit data faster than light is that if you both agree to perform an action depending on what color ball they see. If you both view the ball at some agreed point in the distant future, you will instantly know what action the other person will do.

By looking, you're forcing it to take a color if it didn't have one. You can't tell if it was randomly chosen by your action, or if it was chosen by the other side's action before yours, without communicating by some other means.

No, in fact it doesn't matter who opened the box first.

Imagine we both open our boxes at around the same time, and communicate each other results at the speed of light.

- From my point of view, I opened the box first, got red and caused you to get blue. Your message confirms my hypothesis.

- From your point of view, you opened the box first, got blue and caused me to get red. My message confirms your hypothesis.

We can repeat this experiment thousands of times, every time, we both get a 50/50 chance of red vs blue, and every time, the hypothesis is confirmed.

In order to know who really was first, the only way is to wait for the other person result. For example: when you open the box, start a timer, stop it when you receive my message. It the time is less than the times it takes for light to travel between us, I was first. But in order to have this information, you have to wait, so it is not faster-than-light anymore.

>> Does someone have a good explanation/intuition for why you cannot exploit quantum entanglement to send information faster than light?

Because physicists can not tell the difference between a particle whose "wave function collapsed" and one that didn't.

The faster than light communication would be equivalent to modulating the collapsedness of a stream of particles by measuring-or-not their entangled counterparts. Since the is no discernable difference between particles pre and post collapse, no information can be transmitted.

Entanglement works like synchronized clocks. You can't send information, but you can coordinate actions to a similar effect. Much like you can coordinate actions of separate groups using synchronized clocks with no communication except for passing time.

That's the "hidden local variables" theory. Experiments seem to suggest it's wrong. How one pair is observed appears to affect (change) the other instantaneously. But, not in a way we can use to our advantage to get instant communication.

Exact mechanism is different from synchronized clocks, it's an illustration how you can have correlation without communication and how you can use it in practice, an this correlation works with both entanglement and synchronized clocks.

It seems some aspects can be explained by (modelled as) synchronization, but not all. "Spooky action at a distance" resembling faster-than-light communication still exists.

The two top explanations are either some info can travel faster than light (from our perspective), or the universe forks into copies when needed.

This is my favourite explaination. I have had this bookmarked for over 10 years and keep coming back to it:

http://www.felderbooks.com/papers/bell.html

This is a Google Tech Talk I gave about this about 8 years ago:

https://www.youtube.com/watch?v=dEaecUuEqfc

It is based on this (unpublished) paper:

http://www.flownet.com/ron/QM.pdf

The key insight is that measurement and entanglement are actually the same physical phenomenon. A measurement is nothing more than a very large network of mutual entanglements.

you have to send classical information to get useful stuff out of the entanglement like quantum teleportation.

The simplest one is that you need to be able to tell the partner when you've made your changes and that you're ready for them to measure

By this analogy, (non-local) hidden variable theories are the ether of quantum mechanics: The ether determines the true value of simultaneity, and hidden variables determine the true outcomes of measurements.

For both theories the physics surrounding them just happens to make their presence undetectable. In the case of the ether, the ether wind just happens to shrink the arms of the Michelson-Morley interferometer by exactly the amount needed to prevent the interference pattern from detecting the ether wind. In the case of hidden variable theories, the predicted joint probability distributions just happen to make the hidden variable values themselves uninferable.

Please correct errors in the following: polarization of photons is another form of entanglement, and I seem to recall there are others. Does this result solve the "action at a distance" problem for all of them?

When I wrote my original post I seriously considered adding the restriction to 'entangled spin states', but then decided not to.

First, if confused more than clarified things. And second, I suspect the principle does apply to all forms of entanglement.

Now if I could only prove that, I'd be on my way to Stockholm. :-)

so this means the quantum computing is not possible ?

For quantum computing to be impossible, basically all of quantum mechanics would have to be wrong.

There are enormous engineering challenges with quantum computing, but no fundamental challenges.

There are some who argue those engineering challenges make theoretical compliance moot:

https://spectrum.ieee.org/computing/hardware/the-case-agains...

(I've also just submitted that link to HN separately fwiw)

So the argument is that the chance of having an error goes up exponentially as you add more qubits, 1:1 with the size of the problem state?

Well that's easy enough to understand, even though I have absolutely no idea if it's true or not.

If the error necessarily goes up exponentially, and error correction or dampening can not possibly work, I would count that as new knowledge about the basics of quantum mechanics.

Basically, either quantum computing works or we'll learn a lot more about quantum mechanics we didn't already know.

Reply on news.ycombinator.com