Quantum Mechanics - why random??

[SkepticalScience]

Hey guys,

Thanks for helping me understand things a bit deeper.

One last question - i heard that at the quantum level, everything is random. These fluctuations don't operate under any of the classical physics laws.

Can someone point me to a paper or something that actually says this?

And what I can't understand is, are those fluctuations really completley random, or is it more that we just haven't figured out what the underlying cause is yet?

Like 10,000 years ago people thought lightning was completly random, but today we don't look at it that way.

Will the same thing be true of QM in the future??

SS

 

[SpaceFluffer]

Check out the latest issue of Scientific American - it's on Einstein but there's an article that discusses QM's probabilistic behaviour.

But don't use the word random - in science there's a huge difference between probabilistic and random.

Lightning is NOT random - the likelihood of it occurring has a probability distribution beneath the thundercloud. If it were 'random', lightning could randomly occur anywhere, thundercloud or not.

Quantum mechanics is similarly probabilistic. eg. there is a probability distribution describing the likelihood that a measurement will observe an electron (say) at each location in space. There are usually more favored locations, and usually locations where the probability is zero. This is far from random.

 

[rppa]

One last question - i heard that at the quantum level, everything is random. These fluctuations don't operate under any of the classical physics laws.

Can someone point me to a paper or something that actually says this?

This *$&#($* browser won't install shockwave, so I don't know if this is a satisfactory answer to your question, but it might be:
http://www.chemistry.ohio-state.edu/betha/qm/

This seems pretty good too, but I don't know if it's too technical for your taste. It starts with the classic double-slit experiment (which has a perfectly good classical model) then goes into the one-photon-at-a-time version (which doesn't).

And what I can't understand is, are those fluctuations really completley random, or is it more that we just haven't figured out what the underlying cause is yet?

QM is the best model we have right now. Under QM, they're truly random.

Like 10,000 years ago people thought lightning was completly random, but today we don't look at it that way.

We don't? Well, I guess you could describe it as chaotic, which means that it's beyond our ability to predict, so it's the next best thing to random.

Will the same thing be true of QM in the future??

Checking my magic 8-ball.... it says "Answer uncertain. Try again later."

Who knows? It might be, it might not. All we know is our current models. We don't have an underlying deterministic model that agrees with the experimental data.

One of the most bizarre predictions of the random QM version of the world is quantum entanglement, where a measurement of one particle seems to simultaneously affect the measurement of another distant particle.

Unfortunately, entanglement has been seen in the lab now, so QM still survives as an explanation.

 

[toddjh]

One last question - i heard that at the quantum level, everything is random. These fluctuations don't operate under any of the classical physics laws.

It's been a while, but the last I heard, some interpretations of QM say that it's true randomness, while others say it's simply unpredictability.

And what I can't understand is, are those fluctuations really completley random, or is it more that we just haven't figured out what the underlying cause is yet?

This is called the "hidden variable" hypothesis. There's a famous proposition (called Bell's Theorem) that more-or-less disproves it, at least in any useful form. Bell showed that the "random" outcomes of quantum-level measurements can be affected by the states of other particles an arbitrary distance away. The way it works is, two particles can become "entangled" -- some of their attributes (say, the polarizations of two photons) are connected. When you measure the attribute of one of those particles, the other particle is "forced" to adopt a related value, no matter where it is.

So, for example, the polarization of a photon you see here might be influenced by the polarization of another photon many light years away. That phenomenon is sometimes called "spooky action at a distance," because intuitively it looks like something must be travelling faster than light, and has been confirmed experimentally. The recent "teleporting photon" experiments are based on something similar.

So, in other words, even if there are "hidden variables," there is no way for us to exploit them in a useful fashion, even in principle. They might as well be completely random, for all the good they do us.

Jeremy

 

[drkitten]

Lightning is NOT random - the likelihood of it occurring has a probability distribution beneath the thundercloud. If it were 'random', lightning could randomly occur anywhere, thundercloud or not.

This statement is simply wrong.

It's actually very difficult to give a clear concise definition of the word "random" that is uniformly acceptable to statisticians and similar professional dealers in things random. But the notion of "random" certainly encompasses distributions that aren't uniform.

Just as an example, the roll of a fair die is random -- and specifically is uniform over the range 1..6. The roll of an unfair (biased) die is still random, but non-uniformly so. The height of a third grader is non-uniformly random and follows a very good approximation of a so-called "normal" distribution. The number of soldiers killed by rattlesnake bitesin any given year is random and follows a so-called Poisson distribution. The market share that a given company will have next year is random but is well-modelled by a Markov process. And so it goes.

 

[wipeout]

The mathematical definition of randomness is an incompressible bit-string, which in other words is information so messed up that it can't be expressed in a shorter form.

However, what gets called "random" in quantum physics is better to be called stochastic as there is a slight chance you could get some regularities in the "randomness" which would allow you to write it down in a shorter form. Most of the time it would be truly random in the mathematical sense, but some of the time it wouldn't be.

Anyway, that's just a technical point on the definition of randomness. It's okay to call quantum theory random in the nontechnical sense most people mean but if you want to be precise you call it "stochastic" or "a chance process" or something like that.

So, as to the stochastic nature of quantum mechanics...

It's that way as it's simply easier to believe everything is determined by chance.

There is no proof that it's stochastic. It could be determined by something we don't know about.

It's just easier to believe that it isn't.

 

[RussDill]

Personally, I like the concept of what can exist, does exist.

So if you have a list of possible outcomes from a quantum event, the fact that it occured means we are living in a reality where it occured. If there is a more likely outcome, we are more likely to be existing in that reality.

I think this is also known as the "many worlds theory"

 

[SkepticalScience]

Ok, let me ask you guys this:

Does science know what is causing these quantum fluctuations??

And toddjh, is the consequence of Bells Theorem that there IS no underlying cause of these fluctuations or that there is no "hidden" cause.

I guess I am having a tough time seeing what particle entanglements have to do with anything. If particles are behaving randomly because OTHER particles are being effected - then in theory, if we knew all the particles that are entangled, would we be able to predict their motions??

 

[RussDill]

Ok, let me ask you guys this:

Does science know what is causing these quantum fluctuations??

And toddjh, is the consequence of Bells Theorem that there IS no underlying cause of these fluctuations or that there is no "hidden" cause.

I guess I am having a tough time seeing what particle entanglements have to do with anything. If particles are behaving randomly because OTHER particles are being effected - then in theory, if we knew all the particles that are entangled, would we be able to predict their motions??

If you attempt to determine the polarization of a photon for instance, you are likely to change it's polarization. I don't really buy into the hidden variables.

 

[toddjh]

And toddjh, is the consequence of Bells Theorem that there IS no underlying cause of these fluctuations or that there is no "hidden" cause.

As far as I know, Bell's Theorem doesn't address the question of whether there's such a thing as true randomness in the QM world -- it only shows that it's impossible to rule out "spooky action at a distance" as a factor in what appears to us to be random. It may be that it's sometimes random and sometimes the result of entanglement. Or maybe the universe is one huge ball of string and it's always the result of entanglement -- we can't know, and that's the point.

I guess I am having a tough time seeing what particle entanglements have to do with anything. If particles are behaving randomly because OTHER particles are being effected - then in theory, if we knew all the particles that are entangled, would we be able to predict their motions??

No, because of the speed of light limitation. Think about how it works: Suppose you have two entangled particles -- you have one, and your friend has the other a few light years away. Measuring one will instantly fix the value of the other. Now, suppose you measure your particle and it has value X. Does it have value X because your friend already measured his particle, or does it have value X because you were the first to measure and that's just what it turned out to be? You can't know, even in principle, until you talk to your friend to find out which of you measured first, and that will take a few years.

In other words, you can't get any useful information out of the knowledge that a particle is entangled. Any information is still limited to the speed of light, and so it won't reach you until after the fact. That's important for what you're talking about because it means that it's still absolutely impossible to predict the result of a quantum measurement, even if you know that your particle is entangled, and where the other particle is.

Bell's Theorem is important because it disproves the idea you're talking about, that there might be some hidden information we're missing, and that if we knew it, the answer to quantum uncertainty would fall into our lap. You're not in bad company wondering about that -- Einstein himself was in that camp. But if the result of a quantum measurement could be affected by things happening halfway across the universe that we have no way of knowing about, how can you possibly hope for predictability? It's impossible, even in principle.

Jeremy

 

[SkepticalScience]

OHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHH!


Holy Hell, todjh, that's the first explanation about this stuff that I actually "get".

Awesome!


THANKS SO MUCH!

Off to look up this cooky Bell's Theorem.

SS!!

 

[cbish]

Brian Greene's (Elegant Universe/String theory) new book, Fabric of the Cosmos does a great job in describing Quantum uncertainty, particle entanglement and Bell's studies. It also is rich in history as to how these ideas were developed and debated. (Einstein, Rosen et.al were of the idea that there wasn't quantum uncertainty, we just didn't have the instrumentation yet to see the real picture).

 

[Soapy Sam]

"Does science know what is causing these quantum fluctuations??- SkepticalScience

The logical difficulty with a chain of cause-and-effect is that it either must stop somewhere (so the lowest level phenomena are acausal- they "just happen") or the chain must be a closed loop, where a causes b...(add n steps here)...which causes a.

Odd that people would think fifteenth century sailing ships were "invisible" to Amerindians because they had no word for sailing ship, yet happily trust that a language such as English is fully able to meaningfully explain quantum fluctuations.

I'm sceptical about both notions, myself.

 

[Ziggurat]

Does it have value X because your friend already measured his particle, or does it have value X because you were the first to measure and that's just what it turned out to be? You can't know, even in principle, until you talk to your friend to find out which of you measured first, and that will take a few years.

Actually, it's worse than that: who measured first can actually depend on what reference frame you're in: an observer moving one way would say you measured first, but an oberver moving the other way would say your friend measured first, and they'd both be right (in their reference frame). Very spooky. Which makes this disclaimer especially important to maintaining our sanity when considering the problem:

In other words, you can't get any useful information out of the knowledge that a particle is entangled. Any information is still limited to the speed of light, and so it won't reach you until after the fact. That's important for what you're talking about because it means that it's still absolutely impossible to predict the result of a quantum measurement, even if you know that your particle is entangled, and where the other particle is.

Correction needed here, though:

Bell's Theorem is important because it disproves the idea you're talking about, that there might be some hidden information we're missing, and that if we knew it, the answer to quantum uncertainty would fall into our lap. You're not in bad company wondering about that -- Einstein himself was in that camp. But if the result of a quantum measurement could be affected by things happening halfway across the universe that we have no way of knowing about, how can you possibly hope for predictability? It's impossible, even in principle.

Not exactly. Bell's theorem doesn't disprove hidden variables. What it does do is disprove local hidden variable theories. Nonlocal hidden variables are still possible, but they're much less appealing: there may be some hidden variable that we can't measure that will predict the spin state of these particles, but whatever it is, it's non-local and applies to BOTH particles. So any hidden variable theory consistent with quantum mechanics is necessarily nonlocal and cannot save us from the spooky action-at-a-distance problem you describe above. And that non-locality bothered Einstein and others as much as (maybe more than) the randomness question.

 

[SpaceFluffer]

This statement is simply wrong.

It's actually very difficult to give a clear concise definition of the word "random" that is uniformly acceptable to statisticians and similar professional dealers in things random. But the notion of "random" certainly encompasses distributions that aren't uniform.

I say potato...

The dictionary definition of random is "Having no specific pattern, purpose, or objective". I was utilizing this definition of random, as opposed to the correct definition of a random variable - which is what you describe.

 

[SkepticalScience]

Ok, one other question.

Say, the universe only consisted of two sub-atomic particles. Would we know how modifications of one particle effects the other?

So say I have SubAtomic-Particle A and SubAtomic-Particle B. Say they are at rest, after all there are only 2 of them. I push down on A, and say, we notice that B goes up.

There wouldn't be any spooky-ness going on, cause we have reduced everything to 2 particles. . .right?

If that is right, then is it is a problem of complexity - that there are just too many sub-atomic particles to keep track of effecting too many other sub-atomic particles.

Also - that makes me wonder, if I change the position of one sub-atomic particle - does that mean, I could have possibly changed trillions and trillions of other sub-atomic particles?

SS

 

[RussDill]

Ok, one other question.

Say, the universe only consisted of two sub-atomic particles. Would we know how modifications of one particle effects the other?

So say I have SubAtomic-Particle A and SubAtomic-Particle B. Say they are at rest, after all there are only 2 of them. I push down on A, and say, we notice that B goes up.

There wouldn't be any spooky-ness going on, cause we have reduced everything to 2 particles. . .right?

If that is right, then is it is a problem of complexity - that there are just too many sub-atomic particles to keep track of effecting too many other sub-atomic particles.

Also - that makes me wonder, if I change the position of one sub-atomic particle - does that mean, I could have possibly changed trillions and trillions of other sub-atomic particles?

SS

Any measurements or changes to a particle is done with another particle. In the case of two particles, it doesn't take long to figure out which particle it will be. Even as you add more particles, whatever particle you are using to measure is invaraibly intangeled with the other particles, does that effect your measurement?

 

[Tez]

Re: Re: Quantum Mechanics - why random??

It's been a while, but the last I heard, some interpretations of QM say that it's true randomness, while others say it's simply unpredictability.



This is called the "hidden variable" hypothesis. There's a famous proposition (called Bell's Theorem) that more-or-less disproves it, at least in any useful form. Bell showed that the "random" outcomes of quantum-level measurements can be affected by the states of other particles an arbitrary distance away. The way it works is, two particles can become "entangled" -- some of their attributes (say, the polarizations of two photons) are connected. When you measure the attribute of one of those particles, the other particle is "forced" to adopt a related value, no matter where it is.

So, for example, the polarization of a photon you see here might be influenced by the polarization of another photon many light years away. That phenomenon is sometimes called "spooky action at a distance," because intuitively it looks like something must be travelling faster than light, and has been confirmed experimentally. The recent "teleporting photon" experiments are based on something similar.

So, in other words, even if there are "hidden variables," there is no way for us to exploit them in a useful fashion, even in principle. They might as well be completely random, for all the good they do us.

Jeremy

I disagree with most of this.

Bell himself "believed in" (consistently advocated) some sort of physical realism (hidden variables). The point of his theorem is simply that such a theory must be nonlocal. Note, the alternative is to deny realism at some level, and thats not particularly nice either.

There is nothing that says hidden variables are necessarily hidden - a deeper theory may well "reveal" them (much as thermodynamics had "hidden atoms"). Nor does the randomness of such a theory have to be more than the indeterminism of observers constrained to not know the precise conditions of every object in the universe. (An example is Bohm-de Brogle theory, where some formulations are deterministic - the randomness arises from the sort of randomness we get in classical thermodynamics, where the observer isn't able to know the specific microstate.)

Its an interesting question (for classical theories as well as quantum ones) whether there is such a thing as "objective randomness": Randomness that can somehow "exist" independent of the inherent indeterminism of observers trying to process information about the world around them. Describing something as random seems like such an observer specific thing to me...

 

[wipeout]

From what I've read of recent work on the fundamentals of quantum theory, the EPR correlations are best described as non-classical and not as non-local, and no effect is passing from one particle to the other and so there's no spooky-action-at-a-distance.

The problem lies in applying our classical physics way of thinking to quantum physics and then effects that go faster-than-light and go backwards in time or involve negative probabilities seem to turn up.

However, I find the idea of these strange effects a lot less horrifying than the recent work on the fundamentals of quantum theory that is supposed to clarify things like EPR correlations.

This work has an infinity of different but equally legitimate descriptions of events emerging out of an infinity of possible but interfering events and Schrodinger's Cat popping up all over the damn place.

"Mind-warping" is the most polite term I can think of for it.

 

[toddjh]

Re: Re: Re: Quantum Mechanics - why random??

Bell himself "believed in" (consistently advocated) some sort of physical realism (hidden variables). The point of his theorem is simply that such a theory must be nonlocal. Note, the alternative is to deny realism at some level, and thats not particularly nice either.

Sorry, guys. I thought the "local" part was implied by the context, and I didn't feel the need to add more terms to what is already a potentially confusing subject.

Jeremy