The Universe is Deterministic

[godless dave]

All things in our experience have causes -- all of them.

As I'm sure you know, many events at the quantum level do not have causes as far as we can tell, so I'm puzzled why you would make that statement.

 

[Dorfl]

The double slit experiment can be explained by the transactional interpretation of QM, without needing hidden variables, or non-locality.

Yes. My point is that the double-slit experiment doesn't seem to address the issue of hidden variables at all, so it wasn't relevant to the OP.

 

[Cynic]

You're mistaken.

There is no dispute that according to standard QM the results we obtain from doing experiments are non-deterministic.

I'm disputing it, so I'm clearly not mistaken about that. I didn't say there was a "controvery" -- I was clarifying what I was disputing.

Of course not. But we're not talking about randomness, we're talking about determinism. The word "random" does not appear in the OP, and I'd rather not go there, because it's not a very well-defined word and we will inevitably fall into useless semantic arguments.

Yet the very first reply to the thread invoked randomness. It comes up and I don't see how it can be avoided -- it's practically the same issue. A genuinely random event implies non-determination by definition. A deterministic universe prevents it by definition. It's the hinge, whether you interpret the appearance of randomness to be reality or not. But if we can avoid it, sure. Not many people seem to understand it, you're right.

Google "bell's theorem" and look up the associated experiments.

While I do plenty of reading on my own, I don't accept a suggestion to do so as an argument. If you understand why one of those experiments necessitates a non-determinism, you should be able to explain it without refering me to it.

That said, what precludes non-local deterministic action? Without an understanding of what space actually is, discussing local versus non-local with certainty would seem problematic.

I mentioned on another thread that logic is an all-or-nothing proposition, and that we can't use it to support one part of our ideas and dismiss it when it gets inconvenient. At the root of things, I think that's what has happened. Non-determinism is illogical. While science is based on observations, it is also based on logic. Dismissing logic in science isn't any more valid than dismissing observation. The only resolution here is that what appears to be non-determinism doesn't describe reality. I think the general acceptance of an inherently illogical conclusion -- whatever history might be behind it -- is holding physics back.

Anyway, can you demonstrate something (yourself) that necessitates non-determinism? One would expect that the calling card of determinism is correlation and connectness.

 

[Dorfl]

I mentioned on another thread that logic is an all-or-nothing proposition, and that we can't use it to support one part of our ideas and dismiss it when it gets inconvenient. At the root of things, I think that's what has happened. Non-determinism is illogical. While science is based on observations, it is also based on logic. Dismissing logic in science isn't any more valid than dismissing observation. The only resolution here is that what appears to be non-determinism doesn't describe reality. I think the general acceptance of an inherently illogical conclusion -- whatever history might be behind it -- is holding us physics back.

Can you explain exactly where the logical contradiction lies in the belief that the development of a closed system is not completely dependent on its initial conditions?

 

[Yoink]

Non-determinism is illogical.

I'll leave the physics to others who understand it better, but this statement is just flat out wrong. It is, in fact, the inverse of the truth. There is no logical basis for our understanding of causality, as Hume demonstrated a good long time ago. If causality ceased to function tomorrow, I would certainly no longer be able to operate in this world (I would, presumably, cease to exist), but there is no logical bar to that eventuality.

Our understanding of causation is purely empirical; there's no logical basis to it whatsoever.

 

[sol invictus]

I'm disputing it, so I'm clearly not mistaken about that.

To be absolutely clear, you are disputing the statement "according to standard QM the results we obtain from doing experiments are non-deterministic"? (I should probably specify what I mean by "standard QM" - I meant QM in the Copenhagen interpretation, a topic you'll find in tens or hundreds of textbooks.)

If that's what you're disputing, you'd be better served reading a QM book than posting here.

If you want a prototypical example, prepare an electron in a state of spin up along the z-axis, then measure its spin along the x-axis. According to QM, the result is 50% likely to be up and 50% down, and there is nothing about the electron's state (or the detector's state) prior to the measurement that determines the result. Hence, the result is non-deterministic.

While I do plenty of reading on my own, I don't accept a suggestion to do so as an argument. If you understand why one of those experiments necessitates a non-determinism, you should be able to explain it without refering me to it.

It's not a particularly simple argument, and this forum not intended as a classroom (nor would it work well as one).

Bell proved that any theory in which the results of the spin measurement I mentioned above are determined by the state of the electron and/or the state of the detectors makes a prediction which differs from QM's. The only way to obtain results consistent with QM is if the detectors can influence each other instantaneously no matter how far apart they are (which is what's meant by "non-local").

That said, what precludes non-local deterministic action?

Causality. Non-local hidden variables are impossible in relativistic theories. Moreover relativistic quantum field theory doesn't have them, yet such theories are the best tested in the history of science.

Non-determinism is illogical.

Nope. Probabilistic theories obey precisely the same set of logical rules as do deterministic ones.

What you really meant is "I don't like it".

 

[Cynic]

To be absolutely clear, you are disputing the statement "according to standard QM the results we obtain from doing experiments are non-deterministic"?

Yes.

I'd like to say upfront however, that I'm not a physicist -- I don't even play one on TV. No doubt you didn't need me to inform you of that. ;-)

My main goal here is to challenge myself, not demand to be taught, or shove my ideas down anyone's throat. I just take my skepticism seriously and guard against engaging in processes that might cause me to internalize someone else's ideas without giving me a change to fully vette my objections. So at all times please remember: I'm 100% sincere when I say I want to be proven wrong. If I'm right, so much the better, but I don't have any interest in harboring wrong ideas.

That said, not following up on my reservations wouldn't be particularly critical of me.

If you want a prototypical example, prepare an electron in a state of spin up along the z-axis, then measure its spin along the x-axis. According to QM, the result is 50% likely to be up and 50% down, and there is nothing about the electron's state (or the detector's state) prior to the measurement that determines the result. Hence, the result is non-deterministic.

But we don't know the electron's state prior to measurement so how can we say anything about it prior to measurement? For that matter, how do we know that the electron detected is the "same" electron? We don't really know what electrons are. Particle/wave duality doesn't mean that electrons are both; it means they are neither but can act like both. To be considered uncaused, we would need to know more about this nature. Since we don't, what these experiments say is that we don't know, but it looks like this.

Bell proved that any theory in which the results of the spin measurement I mentioned above are determined by the state of the electron and/or the state of the detectors makes a prediction which differs from QM's. The only way to obtain results consistent with QM is if the detectors can influence each other instantaneously no matter how far apart they are (which is what's meant by "non-local").

The trouble for me is, we're holding them to being a particle in this instance -- separate particles. What if entanglement means they effectively become parts of the same particle, where one isn't affecting the other, but affecting one is affecting the "other"? For all we know, there are dispersion force analogues at that level as well, and particles being interacted with that we're not taking into account at all. For all we know, there's one electron in the entire universe and we're just seeing bits of it here and there.

Moreover, our poor understanding of space and time means that we could be making huge assumtions regarding proximity. Gravity and universal expansion/contraction offer more than enough evidence that there's something we're missing in that regard.

My point isn't to toss about a bunch of half-baked crackpottery -- I respect the logic and science too much for that. I'm considerned that the book on this stuff has been shut prematurely, preventing whole generations from bothering to even look at it. There's too much we don't understand to declare with such certainty that something is truly non-deterministic. From the perspective of this world that we're trying to peer into, any influence the macroworld exerts there probably looks non-deterministic as well.

I'm reminded of "intelligent design" proponents and their notion of irreducibly complex structures, where upon the supposed discovery of one they invoke a designer because it can't occur naturally. That would be a pretty big mistake to make in biology, and it would be a bigger mistake to make in physics.

 

[Ziggurat]

But we don't know the electron's state prior to measurement so how can we say anything about it prior to measurement?

I don't think you really understand the problem. You prepare an electron with spin along the +z axis. You can call this a measurement if you like, but it doesn't matter: everything that quantum mechanics has to say about the electron's spin is known. Now you do another measurement of its spin, but this time along the x axis. What do you find? A 50/50 chance of getting either +x or -x.

There are a number of ways to approach this problem. Under the Copenhagen interpretation Sol mentioned, you've got a probabilistic collapse event occurring because of your second measurement, and quantum mechanics is NOT strictly deterministic. There's no getting around that if you accept the Copenhagen interpretation.

It's true that this is not the only interpretation that is available to us. The issue centers around what "collapse" means, and what's really going on during a measurement. There are alternatives to the Copenhagen interpretation, though we can't really experimentally differentiate between them at this point. My personal inclination is to be agnostic about the whole thing: collapse is what happens when we want to stop using quantum mechanics, so I don't see too much point in obsessing about what quantum mechanics says about it.

Is it possible that there's some strict determinism lurking underneath the whole thing? Sure. But I don't think you can prove it any more than you can disprove it. By stating that quantum mechanics is deterministic, you're basically saying that a whole set of interpretations of quantum mechanics is wrong. Which is an acceptable position, but you're getting yourself in trouble because you don't seem to really understand the differences between these various interpretations.

For that matter, how do we know that the electron detected is the "same" electron? We don't really know what electrons are.

We know enough that this is REALLY not part of the problem.

My point isn't to toss about a bunch of half-baked crackpottery -- I respect the logic and science too much for that. I'm considerned that the book on this stuff has been shut prematurely, preventing whole generations from bothering to even look at it.

Your fears are misplaced. People think about this stuff all the time.

 

[Cynic]

There are a number of ways to approach this problem. Under the Copenhagen interpretation Sol mentioned, you've got a probabilistic collapse event occurring because of your second measurement, and quantum mechanics is NOT strictly deterministic. There's no getting around that if you accept the Copenhagen interpretation.

It's true that this is not the only interpretation that is available to us. The issue centers around what "collapse" means, and what's really going on during a measurement. There are alternatives to the Copenhagen interpretation, though we can't really experimentally differentiate between them at this point. My personal inclination is to be agnostic about the whole thing: collapse is what happens when we want to stop using quantum mechanics, so I don't see too much point in obsessing about what quantum mechanics says about it.

Is it possible that there's some strict determinism lurking underneath the whole thing? Sure. But I don't think you can prove it any more than you can disprove it. By stating that quantum mechanics is deterministic, you're basically saying that a whole set of interpretations of quantum mechanics is wrong. Which is an acceptable position, but you're getting yourself in trouble because you don't seem to really understand the differences between these various interpretations.

To be clear, I've been trying to say what you just did -- that it's an interpretation issue. I don't think QM is "wrong", just not complete. I'm also not saying that QM is deterministic, but that QM doesn't necessarily demand that the universe be deterministic.

It's true that I don't understand the whole of it. I'm just trying to maintain that we don't know enough to say one way or another and that in that position, it's silly to side with that which is completely unlike everything else we've ever experienced.

 

[ben m]

My main goal here is to challenge myself, not demand to be taught

Your questions are such that the answers are unintelligible without a pretty good understanding of the experiments and the math. Remember, it took 70 years for physicists to get from the first hints of quantum mechanics (Planck 1901) to the conclusive experimental disproof of local realism (Freedman and Clauser 1972); we teach this stuff to physics majors, if at all, only when they have three years of background.

I think that "challenging yourself" along these lines is necessarily equivalent to "learning modern quantum mechanics". There is no sense in which you can get a quickie conceptual overview, and still be able to distinguish good results from bad results.

 

[ben m]

It's true that I don't understand the whole of it. I'm just trying to maintain that we don't know enough to say one way or another and that in that position, it's silly to side with that which is completely unlike everything else we've ever experienced.

What makes you think that we don't know enough?

Anyway, in the realm of small things---atoms, photons, nuclei, etc.--- indeterminacy is the only thing we've ever experienced. There are exactly zero data, and indeed zero experiences of any sort, suggesting anything else. From that point of view, it's silly to side against "everything we've ever experienced".

 

[sol invictus]

Yes.

Then please find a reference anywhere which agrees with you. Remember, you challenged my assertion about what standard QM says. So the onus is on you to find a source that agrees with you.

You won't be able to, of course.

I'm reminded of "intelligent design" proponents and their notion of irreducibly complex structures, where upon the supposed discovery of one they invoke a designer because it can't occur naturally. That would be a pretty big mistake to make in biology, and it would be a bigger mistake to make in physics.

Biology doesn't have theorems.

There are a number of ways to approach this problem. Under the Copenhagen interpretation Sol mentioned, you've got a probabilistic collapse event occurring because of your second measurement, and quantum mechanics is NOT strictly deterministic. There's no getting around that if you accept the Copenhagen interpretation.

Apparently Cynic doesn't agree.

It's true that this is not the only interpretation that is available to us. The issue centers around what "collapse" means, and what's really going on during a measurement. There are alternatives to the Copenhagen interpretation, though we can't really experimentally differentiate between them at this point.

None of them are both local and deterministic, because no such theory is possible at all (let alone a mere re-interpretation of QM).

My personal inclination is to be agnostic about the whole thing: collapse is what happens when we want to stop using quantum mechanics, so I don't see too much point in obsessing about what quantum mechanics says about it.

You don't think the standard model of particle physics describes measuring devices?

Is it possible that there's some strict determinism lurking underneath the whole thing? Sure. But I don't think you can prove it any more than you can disprove it.

I beg to differ. Not only can you disprove it, Bell did - albeit with some assumptions, but the discussion should therefore be phrased in terms of which of those assumptions you're prepared to abandon.

By stating that quantum mechanics is deterministic, you're basically saying that a whole set of interpretations of quantum mechanics is wrong. Which is an acceptable position, but you're getting yourself in trouble because you don't seem to really understand the differences between these various interpretations.

Worse, he's stating that a certain explicit, specific set of assumptions about the world are wrong.

I'm also not saying that QM is deterministic, but that QM doesn't necessarily demand that the universe be deterministic.

QM plus two assumptions (locality and what you might call reality) does imply that the universe cannot be deterministic. Until you understand that, there's no point in debating it, because that's the starting point.

By the way, the interpretation I find most reasonable is in a sense fully deterministic, and it's certainly local and causal... but it's deterministic in a strange way that evades the theorem (by not satisfying the "reality" postulate), and which also implies that the results we observe for measurements are not determined by the state before the measurement.

 

[Ziggurat]

You don't think the standard model of particle physics describes measuring devices?

I don't think we apply the standard model of particle physics to measuring devices. That may be a purely practical limitation (I don't know how you can apply that model to 10²⁰ particles), but it's still a limitation for basically every measurement I can think of involving wave function collapse.

I beg to differ. Not only can you disprove it, Bell did - albeit with some assumptions

Exactly. Those assumptions are reasonable, but they're not proven.

but the discussion should therefore be phrased in terms of which of those assumptions you're prepared to abandon.

I don't disagree with that.

 

[orange31]

Finding an electron's position and vector momentum simultaneously
cannot be done. The electron's wave function is pretty large. I guess you'd call that 'non-deterministic' (I'm not a physicist either).

That doesn't mean we live in a 'non-deterministic universe'. For example, even though a bowling ball has a wave function, in reality it's so incredibly tiny that the Uncertainty principle has no meaning.

Don't forget, QM is the realm of the 'very small'.
What is kind of interesting is how the double slit experiment has been found valid for buckyball molecules of 60 carbon atoms, meaning molecular wt 60 x 12 = 720. That's kind of 'big' in the QM world, isn't it?

 

[Yoink]

Finding an electron's position and vector momentum simultaneously
cannot be done. The electron's wave function is pretty large. I guess you'd call that 'non-deterministic' (I'm not a physicist either).

That doesn't mean we live in a 'non-deterministic universe'. For example, even though a bowling ball has a wave function, in reality it's so incredibly tiny that the Uncertainty principle has no meaning.

Don't forget, QM is the realm of the 'very small'.
What is kind of interesting is how the double slit experiment has been found valid for buckyball molecules of 60 carbon atoms, meaning molecular wt 60 x 12 = 720. That's kind of 'big' in the QM world, isn't it?

Very small things aren't part of the universe?

 

[sol invictus]

I don't think we apply the standard model of particle physics to measuring devices. That may be a purely practical limitation (I don't know how you can apply that model to 10²⁰ particles), but it's still a limitation for basically every measurement I can think of involving wave function collapse.

I didn't ask whether "we apply" it, but rather whether "it describes" measuring devices. This thread isn't about a practical question, it's about something in principle.

I think most physicists believe that a measuring device is simply a large collection of particles which interact with each other according to the standard model (plus gravity). As a practical matter it's difficult to use that knowledge to make predictions about the behavior of the device, but one can use it to address the in-principle question of whether the world is deterministic.

Do you agree?

 

[Ziggurat]

I didn't ask whether "we apply" it, but rather whether "it describes" measuring devices. This thread isn't about a practical question, it's about something in principle.

I understand that distinction. But I think you're not taking the practical problem as seriously as you should. This isn't just a matter of us not applying our model in some situations, it's a matter of us never applying quantum mechanics when a collapse of the wave function occurs. I do not think it is a coincidence that the collapse process, which we have all these various competing interpretations for, is confined to situations we never model with any accuracy.

As a practical matter it's difficult to use that knowledge to make predictions about the behavior of the device, but one can use it to address the in-principle question of whether the world is deterministic.

Do you agree?

I agree that one can use that knowledge to address that question, but I do not think we can answer the question precisely because the collapse process, which is at the heart of this whole thing, always and only happens when we're not looking (in a sense), when we're discarding quantum mechanics because we can't (for practical reasons) keep using it.

 

[sol invictus]

This isn't just a matter of us not applying our model in some situations, it's a matter of us never applying quantum mechanics when a collapse of the wave function occurs. I do not think it is a coincidence that the collapse process, which we have all these various competing interpretations for, is confined to situations we never model with any accuracy.

I don't think it's a coincidence either. The evidence is growing stronger and stronger that collapse is nothing more or less than the fact that with lots of particles, the wavefunction is very susceptible to almost instant decoherence. It's nothing more than the fact that (1-e)^{10^23} is a very small number even when e is very close to zero.

And I don't think it's true that we never model collapse accurately. Mesoscopic systems that maintain coherence are becoming more and more common. That's what a quantum computer is, after all. So I think experiment has already pushed back the edges of what constitutes a measurement, and I suspect that will accelerate.

I agree that one can use that knowledge to address that question, but I do not think we can answer the question precisely because the collapse process, which is at the heart of this whole thing, always and only happens when we're not looking (in a sense), when we're discarding quantum mechanics because we can't (for practical reasons) keep using it.

Well, either you think measurement devices are made of particles that obey the known laws of physics, or you don't and you think there's some new and mysterious law of physics that applies to voltmeters but not electrons.

Which is it?

 

[Dancing David]

Please demonstrate an experiment that you feel necessitates a non-deterministic universe.

Fusion in the sun, according to Coloumb's law two protons will always repulse each other more strongly the closer they get, Always , so you can't have fusion at the core of the sun, the greater the pressure on the two protons, the harder they push away from each other.

So... how do you get fusion of two protons in the sun?

There is a probability that one or the other will just happen to exist next to the other because of QM. They do not behave like hard little balls they behave like waveforms. So deteministically they can never approach each other. Probablistically they can.

 

[Dancing David]

Check out the practical experiments:
http://en.wikipedia.org/wiki/Bell's...re_violated_by_quantum_mechanical_predictions