Results 1 to 11 of 11

Thread: Simple explanation of Bell's rejection of "Local Realism"

  1. #1
    Join Date
    Apr 2010
    Posts
    493

    Simple explanation of Bell's rejection of "Local Realism"

    HI all,

    I have watched a few good videos recently about Bell's theorem.

    https://www.worldsciencefestival.com...on-episode-21/
    https://www.youtube.com/watch?v=ZuvK-od647c
    https://www.youtube.com/watch?v=zcqZHYo7ONs

    Now I don't pretend to understand the physics but I think I understand the 'dumbed-down' explanations.
    I think i understand that 'spooky action at a distance' is confirmed, and there are no 'hidden variables' that the particles carry... so that somehow measuring one will affect/reveal the other even if on the other side of the universe.

    What i wanted to ask is:

    1. The experiment is using entagled particles.. does this therefore apply to all particles in the universe? Are all particles entangled?
    2. If "Local Realism" is rejected, what does this 'mean'? 'locality' (?) or 'Realism' (?) is rejected. What usage of the term 'realism' is being used here?
    3. How does it change / add to the previous understanding of Quantum Mechanics? i thought 'particles' were not localized anyway until they are observed, and can interact over macroscopic distances e.g. 2 slit experiment.
    4. Does this mean we are back at the concept 'fields'/ 'action at a distance'... wasn't the idea of an electric/magnetic field also a form of 'spooky action at a distance'? Newtonian gravity?
    5. Could it mean there is some FTL communication that we haven't yet understood or is this logically / causally impossible?

    I know this is very complex, but could everyone answer in relatively simple terms!? (if possible)
    Last edited by plant; 2020-May-07 at 03:31 AM.
    "It's only a model....?" :-)
    https://www.youtube.com/watch?v=m3dZl3yfGpc

  2. #2
    Join Date
    Oct 2005
    Posts
    27,229
    Quote Originally Posted by plant View Post
    What i wanted to ask is:
    You ask all the easy questions, don't you?
    1. The experiment is using entagled particles.. does this therefore apply to all particles in the universe? Are all particles entangled?
    All particles are entangled to some degree, but remember, in physics we don't have "a universe", we have "a system". The system is our way of describing the situation, an idealization-- not a claim on everything that is. The system cannot be the entire universe, for then it would include the scientist and their apparatus, and that's not how physics is set up. But if the system is complex enough, we may decide not to treat its entanglements, because they may have no more important effect than the effect of butterfly wings on the weather. It is somewhat similar to how every air molecule in the room right now has been influenced in some way by air molecules across the planet a few days ago, but we would never want to try to include or analyze or understand those effects with our physical description of the air around us, since they are so completely lost in the noise. So it is with entanglement, it is easily lost in the noise to the point that we say entanglement is "broken." This is a very important element of the "beables" of quantum mechanics (the quantities or properties we manipulate to describe a situation in language intended to sound like we are talking about what is), we must realize that the beables are holistic, meaning that you start with what you are regarding as the system, and the system then has a single wavefunction, regardless of how many components (like particles) it contains. But when we generate the concept of a system, we certainly ignore the entanglements between the system and its environment, and we may or may not be able to ignore the entanglements within the system itself.

    That's completely alien to classical mechanics, which is fundamentally reductionist (even though there is a single quantity called the "action" that is minimized in classical mechanics, that single holistic quantity reaches its minimum with respect to every variable independently, and those independent variables are viewed as the beables, not the holistic action, so the action is regarded as applying a constraint on the beables but is not itself the fundamental entity). Put differently, quantum mechanics treats observables as outcomes ruled by the beables, not as the beables themselves (that's the "Heisenberg gap" in a nutshell), whereas classical physics treats observables as being the same things as the beable properties (i.e., the velocity of a particle is a beable associated with the particle, not an outcome of an interaction the particle is subjected to by a velocity measurement). The holistic wavefunction is the full description of the system, it is the primary beable of any system, but making observations of that wavefunction allow us to reduce the system into a concept of a sum of parts.

    So quantum mechanics starts with a holistic existence of any system, glued together with entanglements, but when the system is coupled to stochastic influences from its environment, those entanglements are "broken" in the sense of getting averaged over so badly they are like the motions of the air molecules around us-- we will have no choice but to treat the details as random, and ignore entanglement, which allows us to compartmentalize the beables into individual particles and so forth. But without the coupling to the environment, a system is always just a single thing in quantum mechanics, it cannot be reduced into its parts. Indeed I think it is more fruitful to think of the "Heisenberg gap" as the gap between a holistic system itself, and the way it effectively breaks up into individual components when coupled to a complex environment (often a measuring device), moreso than to think of it as quantum vs. macroscopic.
    2. If "Local Realism" is rejected, what does this 'mean'? 'locality' (?) or 'Realism' (?) is rejected. What usage of the term 'realism' is being used here?
    Realism means that the outcome of measurements are controlled by the properties of the objects being measured, like a set of instructions that say "if someone measures quantity X I will produce the outcome x," whether or not the measurement is actually done. It means we treat nature as a kind of "answer man" that is ambivalent to what we choose to actually determine as true, because the truth is independent of what we choose to know. Local means these properties that determine the outcomes are carried with the particles themselves, so when they leave each other's light cones, the various instruction books become independent entities. This is the aspect of nature that Einstein was so sure must be true that he felt quantum mechanics had to be incomplete because it's not the holistic way that quantum mechanics is set up. It is possibly the most astonishing aspect of nature that it prefers quantum mechanics to Einstein's intuition, as though quantum mechanics took a complete shot in the dark and happened to hit the bullseye with remarkable prescience.
    3. How does it change / add to the previous understanding of Quantum Mechanics? i thought 'particles' were not localized anyway until they are observed, and can interact over macroscopic distances e.g. 2 slit experiment.
    Entanglement is deeper than indeterminism, because you can hold (as some did prior to Bell's theorem) that the Heisenberg uncertainty principle stems from incomplete knowledge about the system. You simply say that the wavefunction is not the only information available to nature, because it isn't the full instruction book. Bell proved that with entanglement, it's simply impossible to describe what happens with instruction books carried by the particles, so either there are no instruction books (and we are forced to think of something fundamentally random going on, god is rolling dice), or the instruction books are holistic and cross outside the light cones. It is an interesting question as to which one Einstein would have chosen had he lived to see Bell's theorem and the experimental refutation of local realism. I think he would have found it deeply unsettling.
    4. Does this mean we are back at the concept 'fields'/ 'action at a distance'... wasn't the idea of an electric/magnetic field also a form of 'spooky action at a distance'? Newtonian gravity?
    Fields always obey the rule that correlations cannot propagate faster than the speed of light. So if I move an electron here, the electromagnetic field over there cannot change until a signal reaches it. So it's not action at a distance at all, it's propagation of a signal. Entanglement actually doesn't change that-- it is still not possible to propagate a signal faster than light. But here we get into the difference between a "signal" and an "effect." Personally, I don't think we should use the concept of "effect" at all, like measuring something here "affects" the measurement over there, expressly because it never propagates any kind of signal and could never be used for communication faster than light. So the concept of a signal is a well defined notion in physics, but the concept of an effect is not well defined and probably doesn't exist at all in physics. If we have two entangled particles separated by a great distance, and I measure one of them, you will not detect anything new in your particle, and you will not know anything new about your particle until I tell you the outcome, via normal communication channels. For me, I would say this means we are seeing for the first time that the observers are also part of a holistic system, and can no longer regard themselves as something separate from the system they observe.
    5. Could it mean there is some FTL communication that we haven't yet understood or is this logically / causally impossible?
    If relativistic quantum mechanics is right, there is not FTL communication allowed by entanglement. In essence, we only see the need to treat a system as entangled if we have knowledge of observations on all its parts, and we can't get that knowledge except via non-FTL channels. So we are also seeing, for the first time, the role of our knowledge about a system appearing in how we regard the system as behaving. This is also a ramification of the loss of local realism.
    Last edited by Ken G; 2020-May-07 at 06:25 AM.

  3. #3
    Join Date
    Apr 2010
    Posts
    493
    Realism means that the outcome of measurements are controlled by the properties of the objects being measured, like a set of instructions that say "if someone measures quantity X I will produce the outcome x," whether or not the measurement is actually done. It means we treat nature as a kind of "answer man" that is ambivalent to what we choose to actually determine as true, because the truth is independent of what we choose to know. Local means these properties that determine the outcomes are carried with the particles themselves, so when they leave each other's light cones, the various instruction books become independent entities. This is the aspect of nature that Einstein was so sure must be true that he felt quantum mechanics had to be incomplete because it's not the holistic way that quantum mechanics is set up. It is possibly the most astonishing aspect of nature that it prefers quantum mechanics to Einstein's intuition, as though quantum mechanics took a complete shot in the dark and happened to hit the bullseye with remarkable prescience.

    So.... does Bell mean that EITHER localism OR realism must be rejected or could it be both?

    But how could you have ‘realism’ (particles have properties even if we don’t observe them) without those properties being ‘local’ (contained within the particles)?
    How could you have ‘local’ properties without those properties being also ‘real’?
    _—————-
    on a tangent to this...i have read proponents of Nik Bostrom suggest this might be that we are in a simulation trying to minimise processor use so that only things that are observed need to be ‘rendered’? The atoms comprising a little cafe in Paris only has to be rendered when i visit......
    "It's only a model....?" :-)
    https://www.youtube.com/watch?v=m3dZl3yfGpc

  4. #4
    Join Date
    Oct 2005
    Posts
    27,229
    Quote Originally Posted by plant View Post

    So.... does Bell mean that EITHER localism OR realism must be rejected or could it be both?
    It could be both. This is like the Godel proof that first-order arithmetic cannot be both complete and consistent, but it still might be neither for any given set of axioms (such as the Peano axioms, for example, which seem pretty innocuous so everyone hopes it is incomplete but not inconsistent, but we can't know which because there is another Godel proof that says you can't prove the consistency of any system that is rich enough to formulate the Godel statement).
    But how could you have ‘realism’ (particles have properties even if we don’t observe them) without those properties being ‘local’ (contained within the particles)?
    Good question. But this is what is held by people who think the wavefunction is a "real thing"-- the wavefunction is not carried with the particles, it is holistic. It's success is why quantum mechanics points to a holist quality of reality.
    How could you have ‘local’ properties without those properties being also ‘real’?
    The way to not have realism is to let god roll dice. So a local theory says the probabilities are carried by the particles, but when an observation is done, the outcome is not completely determined by those properties.
    _—————-
    on a tangent to this...i have read proponents of Nik Bostrom suggest this might be that we are in a simulation trying to minimise processor use so that only things that are observed need to be ‘rendered’? The atoms comprising a little cafe in Paris only has to be rendered when i visit......
    In my view, Bostrom is trying to do something with physics that it is not designed to do, and is not scientific. The scientific approach is not to ask "is nature really a simulation," it is to ask, "what power over, or understanding of, our situation can we gain by imagining that nature is a simulation." Anyone who holds that nature follows algorithmic rules, like Newton's laws, is essentially asserting that nature works just like a simulation. Whether it actually is, or just works like it is, is clearly not a question that physics is meant to address. Ironically, Bostrom is trying to find evidence that nature is a simulation in the fact that it is not algorithmic, which is reminiscent of Penrose's point about consciousness being non-algorithmic. But look at what we have-- if nature is algorithmic, there's your simulation right there, and if it is not algorithmic, apparently we can also argue it is a simulation trying to conserve memory. So what point do we have in a hypothesis that works regardless of what we observe in nature?
    Last edited by Ken G; 2020-May-09 at 05:58 PM.

  5. #5
    Join Date
    Apr 2011
    Location
    Norfolk UK and some of me is in Northern France
    Posts
    9,153
    If Bell can be satisfied by a simulation, within which prediction is possible provided the simulation is complex enough, then that simulation need not be complete and certainly not consistent. So Godel cannot say if we live in such a simulation, nor can we unless like Neo in The Matrix, we are able to go beyond the simulation. A good enough simulation can be untestable by us without being complete or consistent
    sicut vis videre esto
    When we realize that patterns don't exist in the universe, they are a template that we hold to the universe to make sense of it, it all makes a lot more sense.
    Originally Posted by Ken G

  6. #6
    Join Date
    Oct 2005
    Posts
    27,229
    Quote Originally Posted by profloater View Post
    If Bell can be satisfied by a simulation, within which prediction is possible provided the simulation is complex enough, then that simulation need not be complete and certainly not consistent.
    Do you mean that what we normally regard as a simulation must satisfy the Bell inequality, whereas reality does not? We do normally think of simulations as being imbued with local realistic qualities, but Bostrum must be aware of that, so he must be generalizing the concept of a simulation to something more akin to a simulation carried out by a quantum computer-- thereby able to escape the Bell inequality. But that's my point, if Bostrum can still regard the universe as a simulation even after discovering it doesn't obey the Bell inequality, then clearly we are not saying much of anything when we call it a simulation.
    So Godel cannot say if we live in such a simulation, nor can we unless like Neo in The Matrix, we are able to go beyond the simulation.
    What I think would have been a better end to The Matrix is, ***spoiler alert*** if Neo discovered that the computer-world he discovered, and the aliens, were also just a simulation by something else. That would have made the important point that we have to ask, what is different about a concept of reality that is a simulation, and one that is not? For example, prior to Newton's laws, people often imagined that what happened was due to the whims of powerful gods that could do anything they wanted, and afterward, people tended to think everything happens for algorithmic reasons. But which ones of those is a simulation, and which one isn't? I don't think the entire concept of "simulation" is well posed. As soon as you discover people are brains in vats, and you see the vats and see the vat overlord, isn't your next question going to be, why isn't the overlord world also going to turn out to be a simulation? How is any reason that anything happens different from a cog in a machine?
    A good enough simulation can be untestable by us without being complete or consistent
    When we discovered Newton's laws, why didn't people suddenly start to think the universe was a simulation ruled by those laws? So it's not really about how good is the simulation (Newton's laws were pretty good, they looked infallible until we discovered their issues), it's about the fact that anything we do to understand reality ends up turning it into a kind of simulation. We'll always be the ones doing the simulating, that is all we will ever have.

  7. #7
    Join Date
    May 2003
    Posts
    6,146
    Quote Originally Posted by plant View Post
    So.... does Bell mean that EITHER localism OR realism must be rejected or could it be both?
    You can definitely choose to reject both locality and realism, but I think on this question that even if you reject realism, as it is usually viewed, that doesn't help keep locality. That is, I think that even if you assume the things we're measuring do not have real values (perhaps the values we measure are fundamentally randomly determined at the moment of measurement), I think that those measurement results have to allow nonlocal influences to make everything work out.

    Quote Originally Posted by plant View Post
    But how could you have ‘realism’ (particles have properties even if we don’t observe them) without those properties being ‘local’ (contained within the particles)?
    You can go the other way: it's definitely possible to construct a hypothetical mechanism whereby quantum particles can have real properties, but those properties have to be able to influence nonlocal measurements. Bohm's pilot wave model is just such a model.

    Quote Originally Posted by plant View Post
    How could you have ‘local’ properties without those properties being also ‘real’?
    You could have the results of measurements involve a certain amount of fundamental randomness, rather than being determined wholly by the properties of the particles. When you measure the polarization of a photon, it could be not just determined by some hidden property of the photon, but randomly determined by the arrangement of photon and measuring device when measured (but not influence by anything distant, so that it stays local). There's nothing wrong with that possibility in principal, except that I think it still doesn't allow you to match the results we actually see when we do experiments. But it's not logically impossible.

    If you really, really like locality, there are a few possibilities that Bell's Theorem leaves open. One principle that the construction of Bell's Theorem depends on is called "contrafactual definiteness" (often abbreviated CFD). This is the notion that it's meaningful to talk about what would have happened if we had made different choices. We do this all the time in everyday life: "If only I had remembered to bring my umbrella, I wouldn't be soaking wet right now" or "I'm happier working here than if I had taken the job offer from the other company". Bell's argument depends on being able to meaningfully ask questions like "if the scientists running observations at lab B had set their polarizing filter at a different angle, they would still have to have gotten results consistent with the results at lab A, as well as the results they saw at the angle they really did measure". If you decide that such questions are meaningless (you didn't bring an umbrella, and you can't meaningfully discuss what might have happened if you did, because that's not what happened; the polarizing filter was set at the angle it was set at, and you can't talk about what you might have measured if it were set differently), you can't frame the argument that Bell used. There could be a certain sense that this is a little like realism, but moreso: not only do particles have real properties, but it's considered out of bounds to talk about what the universe might be like if those properties were anything different from what they were, or if any measurement decisions were made differently, but it goes pretty far beyond what physicists usually mean when they talk about realism in this context.
    Conserve energy. Commute with the Hamiltonian.

  8. #8
    Join Date
    Oct 2005
    Posts
    27,229
    What's more, it can be argued that the whole point of physics is to answer questions like "what will happen if I do such-and-such," that is what makes scientific thinking powerful. Imagine going to a doctor who gives you a diagnosis of cancer, and you say, what will happen if I decide to undergo chemotherapy? You certainly don't want the doctor to say, "well, I'm not only a doctor, I'm a philosopher, and I've decided that everything we do is determined by our nature. So whether or not you undergo chemotherapy depends on factors outside your control, so why are you asking me anyway?" You'd respond, "my nature is to wonder what will happen if I do various things, so please just answer my question." In other words, it makes no difference if some kind of superdeterminism is the "actual truth", physics cannot be useful that way, we must use principles like contrafactual definiteness because they are built into scientific thinking.

  9. #9
    Join Date
    May 2003
    Posts
    6,146
    Quote Originally Posted by Ken G View Post
    What's more, it can be argued that the whole point of physics is to answer questions like "what will happen if I do such-and-such," that is what makes scientific thinking powerful. Imagine going to a doctor who gives you a diagnosis of cancer, and you say, what will happen if I decide to undergo chemotherapy? You certainly don't want the doctor to say, "well, I'm not only a doctor, I'm a philosopher, and I've decided that everything we do is determined by our nature. So whether or not you undergo chemotherapy depends on factors outside your control, so why are you asking me anyway?" You'd respond, "my nature is to wonder what will happen if I do various things, so please just answer my question." In other words, it makes no difference if some kind of superdeterminism is the "actual truth", physics cannot be useful that way, we must use principles like contrafactual definiteness because they are built into scientific thinking.
    Yes, I agree. It's really weird to imagine the (counterfactual!) world where we never consider counterfactuals.
    Conserve energy. Commute with the Hamiltonian.

  10. #10
    Join Date
    Oct 2005
    Posts
    27,229
    Yes, rejecting counterfactuality is philosophically allowed because anyone can leave the domain of how the physicist thinks in order to build a world view that is not useful for doing physics, but that's of no real consequence to physics, as the latter is not about building world views-- it's about informing actions via successful predictions. People do use physics to create world views, but I've never thought it's such a good idea. In physics, the right way to use a world view is as a mindset for creating a sense of comfort and understanding when applying some theory, it was never meant to be taken seriously and has a pretty poor track record when this is forgotten.

  11. #11
    Join Date
    Apr 2011
    Location
    Norfolk UK and some of me is in Northern France
    Posts
    9,153
    I was suggesting that Bell’s test does not allow us to test whether we are in a simulation. It remains untestable like the god concept which in fact a simulation is just another, more acceptable for some, god concept. Like Rosencrantz tossing coins in Stoppard’s play cannot assume he has passed into the afterlife. I guess Bell shows that individual trials are not necessarily independent of history, like the tossing coin experiment, maybe they are sometimes! That would be counterfactual like being regularly lucky or unlucky.
    sicut vis videre esto
    When we realize that patterns don't exist in the universe, they are a template that we hold to the universe to make sense of it, it all makes a lot more sense.
    Originally Posted by Ken G

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •