Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Randomness being an intrinsic property? Quantum entanglement?

  1. Sep 30, 2010 #1
    I've searched these forums hardcore about these questions and the wide range of answers is so confusing to me, so I hope that maybe if I provide some examples and specific questions, I may better understand.

    I always hear that quantum particles exhibit "intrinsic" randomness in the states they take on, but how is this any different from me tossing a die repeatedly and saying "Look, this die is intrinsically random -- notice how the rolling average gets closer and closer to 3.5." But let us pretend that a skeptic looks at this and wonders if there are hidden variables that account for what I think is "random."

    Of course, we may later find that "Well, if we know something about gravity, the type of die, the weight, the air resistance, the distance between the die and the surface it strikes, any deformation that occurs between the die and surface, the physical makeup of the die and surface, the initial position of the die, the forces acting on it when the die is released, etc -- we can know something about what value we'll get."

    Would these variables be considered "local hidden variables?" How is any of this different from the case of quantum particles? I always hear that "local" quantum variables have been ruled out and that "non-local" variables may still be in the running, but what does it mean for something then to be local or non-local? How do we know all variables have been accounted for when we have not yet penetrated down to the fundamentally small and fast (namely I refer to Planck fundamentals, here, of which we're supposedly 20 orders of magnitude away from or something like that)?

    How can we tell that quantum randomness is "intrinsically/fundamentally random" if there are possible variables to explain the actions?

    Regarding the non-local variables, I've heard of Bell's Theorem even though I don't know *why* it rules out local variables for QM (if we applied Bell's Theorem to a die system, would we be able to show that there are other variables involved?). I hear quantum entanglement is a common misconception explaining that information still doesn't travel FTL, but the wavefunction collapse still occurs despite the object distance, implying that collapse is purely an observer phenomenon. This makes me wonder if wave collapse really occurs at all -- is it possible that these "potential states" don't really exist, but that one state just IS and we won't know about it until it acts on something else in a discernible way that allows us to derive its state? As in, we don't know what something IS until we look but we model its potential states via probability? What evidence is there for non-local variables in QM or that particles can all communicate with each other?

    I've become increasingly frustrated with these questions because despite the hours I spend pouring over this stuff, I can't seem to find answers.
  2. jcsd
  3. Sep 30, 2010 #2


    User Avatar
    Science Advisor
    Gold Member

    Give yourself a little time. It takes a bit of effort to follow the reasoning and there is often initial resistance. :smile:

    Regarding Bell: Imagine you make a simple assumption that there is in fact a cause for what we see with entangled particles. And that cause exists in the past light cone of the entangled particles. Bell's Theorem then shows us that there MUST be non-local effects at work.

    On the other hand: If you reject the idea that particles have observables even when not being observed, then Bell's Theorem allows us to reject non-local effects as a required explanation.

    Take your pick. :smile:

    If you have not seen the derivation of Bell's Theorem - I have a page which explains that - then you should probably follow it through before going much further.

  4. Sep 30, 2010 #3
    Thanks for the response, but what do you mean by "past light cone" of entangled particles?
  5. Sep 30, 2010 #4


    User Avatar
    Science Advisor
    Gold Member

    If something causes something else, it must have occurred in the past - by our usual definition of cause and effect. And if there was a contribution from a cause, presumably it was nearby - although this is actually an open question. So the light cone is defined by the speed of light c multiplied by the time in the past in which the contribution may have happened. Clearly, the further back in time you go, the bigger the light cone.
  6. Oct 1, 2010 #5
    What DrC said, but let me elaborate, because I've shared your confusion on this.

    Utltimately, it isn't any different. You don't need to know about Bell or light cones. It's well known that fully classical treatments of quantum phenomena are quantitatively inadequate. It's an open question wrt whether fully quantum treatments of quantum phenomena are qualitatively inadequate. In either case we're dealing with our ignorance. The reason that certain quantum phenomena are called intrinsically or fundamentally random is simply because the quantum formalism is fundamental to the classical formalism (ie., it quantitatively accounts for things that a so-called qualitative classical formalism can't quantitatively account for), and that certain quantum phenomena cannot be precisely, even quantitatively, predicted by the qm formalism -- hence, they're deemed 'intrinsically' or 'truly' or 'fundamentally' random. But, this isn't a statement about reality. It's just a statement about what the most fundamental theory in physics has to say about reality. In other words, the statement that 'qm phenomena are truly random' has a particular meaning, and when you understand what that particular meaning is then, well, you'll understand it and not disagree with it. But nobody knows what this has to do with an underlying reality, because nobody knows what qm, or classical physics for that matter, has to do with an underlying reality.
    Last edited: Oct 1, 2010
  7. Oct 2, 2010 #6
    I know basic physics only so be gentle.

    Though,it seems whatever happens inside the universe is deterministic.
    Above atoms(outside of quantum realm) cause and effect never seemed to fail us,everything is deterministic even humans.
    But from what i've read the quantum laws of gravity and "some other stuff" is different from the world of the "above the atoms".
    So like DrChinese said that cause and effect need the traditional time because x interacts with y and results in z,this is our intuitive perception of time ,since time is largely based on gravity this becomes a problem mainly because quantum mechanic's gravity (and the "some other things") are different from the "traditional" gravity and the "some other things" :P

    We just cannot apply the same linear understanding of time we have in quantum mechanics,its so far has been good for survival throughout the millenniums but its not very helpful understanding subatomic world.

    So i can only guess that dark energy/mater etc that make up of the 90+% of the universe have to do something with this seemingly "true randomness",cause and effect might take form in a different way or have a very different logic...but still some kind of logic(patern,sequence,structure,even if its too complex to humans to understand) ,true randomness at least inside of universe seems like scifiction to me,it will take time to understand it ,just as all unexplainable ,"random" ,weird things humanity faced since its dawn.

    I mean if this true randomness was...well,true,couldn't we see it in real life? nothing so far is random(quantum randomness) in life ,i think if it was, nothing would work correctly.We would pick a glass of water and it would turn into a volcano or disappear.But those things never happen shows that the different logic or True randomness doesnt change the "normal" world and it stays inside the atoms,is this correct?

    My knowledge is poor in physics,i only know few theoretical things and perhaps i have terrible mistakes in my understanding ,but i have a tendency to try and understand my surroundings hence me babbling things i know very little in a forum (:
  8. Oct 3, 2010 #7
    can anyone explain?
  9. Oct 3, 2010 #8
    I was reading your post, and then ... anyway you might want to put your question in the philosophy forum.

    Ok, to be more straightforward and honest about it, yes, there's an answer to your question (and, to be honest I haven't read it yet), but you should post it in the philosophy forum.

    Don't be discouraged some actual physicists lurk there from time to time.
    Last edited: Oct 3, 2010
  10. Oct 3, 2010 #9
    Ok just asnwer this:
    Do you think quantum randomness is True random or there are other variables that we dont know and cause this seemingly random activity?
  11. Oct 3, 2010 #10


    User Avatar
    Science Advisor
    Gold Member

    In a "stochastic" theory, the input variables are unknown and the results can appear random - much as you suggest. There is no evidence at all that Quantum Mechanics is a stochastic theory, although it is certainly possible. Bohmian Mechanics does supply a stochastic element to QM, although at the cost of non-local action.

    The problem with most stochastic theories is that they make predictions which are experimentally falsifiable. Ergo, they get ruled out pretty quickly. Other than Bohmian and a few other non-local theories, there really are no candidate theories which are capable of jumping through the myriad of requirements which exist. The requirements are draconian, in fact, and I do not expect anything to be discovered anytime soon which might change the landscape. After 80+ years, there has not been much of anything to appear which shakes the foundations. But who knows?
  12. Oct 3, 2010 #11
    So we're waiting for the next Einstein to throw a theory and makes us happy.
    Another question:
    Is it true that "what happens in QM stays in QM" ,as far as it goes for randomness?
  13. Oct 4, 2010 #12
    randomness at microscopic levels gives practically deterministic laws at macroscopic scales due to statistics of large numbers. I say practically, since it is conceivable that unusual macroscopic events might occur, but the expectation times are generally longer than the age of the universe never mind the few thousand years we've been around to witness such events. If you throw 100 coins on a table, eventually they'll all land heads up, but even if you throw once a second, this will usually happen only once every ~30 million years.

    So "deterministically" you get ~50/50 split between heads/tails.

    Also remember that the quantum probability densities evolve exactly deterministically according to the Schrodinger equation, so the statistical predictions have some stability.
  14. Oct 4, 2010 #13
    I see,but if the statistical predictions at macro have stability doesnt that make it NON-(true)random at micro but rather with too many variables and a very complex system but still deterministic?
  15. Oct 4, 2010 #14
    No, a (chaotic) deterministic dynamical system with an invariant measure (probability density) is no different to a fundamentally random one with the same probability distribution from the point of view of macro observations.

    Of course, it's different at the micro level, but since we don't know what the micro level is wrt quantum particles we just assume it's fundamentally random, unless someone comes up with an underlying deterministic theory (eg pilot wave theories are usually based on deterministic dynamics of particles)

    Actually, in deterministic systems you have the poincare recurrence theorem, which predicts all points in some subspace of the total phase space will recur with predictable periodicity, whereas with truly random dynamics the recurrence would be with average times. So I suppose you could watch a quantum system for several cycles of its recurrence time and see if the dynamics cycle predictably. Problem is recurrence times for even simple systems are impractically huge, which is why the poincare recurrence theorem is regarded as irrelevant in classical statistical physics.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook