Deterministic quantum mechanics?

[Loren Booda]

Theoretically, if one had all of the classical information in the universe, could one predict quantum mechanical events? I believe this is a very basic and significant question that physicists need to confront.

 

[inha]

No. For example take all probabilistic phenomena. Question confronted, what's next?

 

[Careful]

No. For example take all probabilistic phenomena. Question confronted, what's next?

Perhaps Loren, you could phrase your question in more detail (for example what do you mean by quantum cosmology?). Nevertheless I can add at this moment that Inha's answer has a strong dependence on his interpretation of the measurement problem (it is not for sure at all that true random generators exist in nature - although they do exist in theory if we take QM as it stands now).

 

[caribou]

I've seen it called "quasiclassical" to reflect that everything is ultimately quantum in nature but appears classical to us above extremely small scales.

 

[Rade]

Did not the science of physics renunciate all hope of describing in classical detail the behavior of general systems at the microscopic spatial scale when they adopted the Schrodinger dynamics first published in 1926 in Annalen der Physik 79, 361 ? Is it not true that we can derive classical dynamics at macroscopic scale from QM, but never QM from classical--is this not a fundamental axiom of physics today ? Yet, --- what if the observed (e.g. experimentally) statistical distributions of the properties of things that exist could be understood mathematically as resulting from ontologically and a priori real "population distributions" rather than, as now, as QM calculations of assumed acausal reality (e.g., QM is real because it works--it is descriptive), then perhaps QM would not be necessary and the answer to the thread question--yes ?

 

[Careful]

**Did not the science of physics renunciate all hope of describing in classical detail the behavior of general systems at the microscopic spatial scale when they adopted the Schrodinger dynamics first published in 1926 in Annalen der Physik 79, 361 ? **

Not at all, although there are few who would bet their money on such approach in these times (but it is not ruled out yet). It is a bit naive (to say the least) to think that the publication of one crazy paper would suddenly wipe out the entire history of realistic physical thought.

** Is it not true that we can derive classical dynamics at macroscopic scale from QM, **

No, that is not known yet

** but never QM from classical--is this not a fundamental axiom of physics today ? **

That is true (at least when classical = local realism), but the question is whether those domains where CM cannot reach QM predictions do really exist in nature or not.

** Yet, --- what if the observed (e.g. experimentally) statistical distributions of the properties of things that exist could be understood mathematically as resulting from ontologically and a priori real "population distributions" rather than, as now, as QM calculations of assumed acausal reality **

Small note: you do not *observe* a probability distribution. Rather you postulate a probability distribution and try -amongst other things- to make a best fit with all current data available. If you want to understand this point better, you might want to check out any paper on Occam's raisor (which involves also the complexity of the theory).


**(e.g., QM is real because it works--it is descriptive), then perhaps QM would not be necessary and the answer to the thread question--yes ?[/QUOTE] **

Right, but such theory is not known yet; basically because IMO, it requires more detailed knowledge of the STRUCTURE of elementary particles. For example, we know virtually nothing about the electron

Cheers,

Careful

 

[masudr]

Theoretically, if one had all of the classical information in the universe, could one predict quantum mechanical events? I believe this is a very basic and significant question that physicists need to confront.

If you consult any introductory quantum mechanics text, you will find that a quantum state contains ALL the information one can know about a quantum object. Given the state, all we can do is predict probability distributions for outcomes of "ideal measurements".

Physicists confronted this basic and significant question a long time ago. As a pointer, you shouldn't use the internet forums as your primary reference of learning.

 

[DrChinese]

Theoretically, if one had all of the classical information in the universe, could one predict quantum mechanical events? I believe this is a very basic and significant question that physicists need to confront.

Under the theory of Quantum Mechanics, no amount of input information about the state of a system will pre-determine the results of all measurements that could be performed on that system. So in that sense, the answer is an unambiguous NO.

Of course, if you accepted Bell's Theorem (see how I cleverly worked this into a post which is not about Bell :tongue2: ) them you might reject the idea that particles have independent simultaneous values to their observables in the first place. Therefore your answer would still be NO.

On the other hand, there are those that argue that the theory of Bohmian Mechanics has the potential to answer such questions in the affirmative IF you had all of the state information about all of the particles in the universe. As I understand it, in BM there are contributions to the "deterministic" evolution of particle trajectories from particles that are space-like separated.

I think it is acknowledged by most that it is NOT POSSIBLE, in principle, to know all of the state information about all of the particles in the universe. This would be true regardless of what theory you think is applicable; because it would take an even bigger universe to store that much knowledge. And so on... and your answer is again NO.

 

[Careful]

**
Of course, if you accepted Bell's Theorem (see how I cleverly worked this into a post which is not about Bell :tongue2: ) **

Oh, but I do accept this theorem; it is not of such mathematical complexity that there are details which cannot be understood :tongue2: (it remains a wonder for me why they call this ``a very deep result´´ or ``one of the biggest discoveries ever´´ ...).

**
them you might reject the idea that particles have independent simultaneous values to their observables in the first place. **

You might indeed but you might also not. You might want to read Sorkin's paper on Occam's raisor - see it as a christmas gift .

** Therefore your answer would still be NO. **

Wrong: you should learn that physics is not about popularity polls and neither about miss math elections (although in practice one could claim this to be true).


**
On the other hand, there are those that argue that the theory of Bohmian Mechanics has the potential to answer such questions in the affirmative IF you had all of the state information about all of the particles in the universe. As I understand it, in BM there are contributions to the "deterministic" evolution of particle trajectories from particles that are space-like separated. **

Correct, but I doubt that BM has anything substantial to add (apart from some interpretational liberties one could take, the theory is entirely equivalent to SQM where the latter can make predictions at all)

**
I think it is acknowledged by most that it is NOT POSSIBLE, in principle, to know all of the state information about all of the particles in the universe. **

Nah, it is merely stated so without any profound comments

**
This would be true regardless of what theory you think is applicable; because it would take an even bigger universe to store that much knowledge. **

That is outrageously false: QM contains many more degrees of freedom than CM does (as an easy excercise proves it to be). If QM would contain LESS degrees of freedom, then I would be all ears

I invite Masudr to consult an elementary GR textbook

Anyway, good holidays to all of you.

Cheers,

Careful

 

[masudr]

I invite Masudr to consult an elementary GR textbook

I'm sorry, every question is framed within it's domain of applicability. I assumed this question was framed in the domain of non-relativistic QM.

I invite Careful to consider that perhaps Masudr has consulted more elementary books than Careful cares to know.

Seasons greetings!

 

[Careful]

**I'm sorry, every question is framed within it's domain of applicability. I assumed this question was framed in the domain of non-relativistic QM. **

?? Nobody mentioned this particular framework and the word UNIVERSE usually rings the GR bell in my head. Moreover, it is a bold (and probably mistaken) assumption that you can exclude GR in the description of elementary particle physics (the question was very general and should be treated as such).

** I invite Careful to consider that perhaps Masudr has consulted more elementary books than Careful cares to know. **

Dito to masudr. However, from such an educated person I would expect a broader point of view than the mere quotation of the QM fairy tale ESPECIALLY when the question is about the application of QM on the scale of the universe (I am sure masudr is familiar with the problems at hand).
It is not because this is a QM forum that information should only flow in one direction.

Cheers,

Careful

 

[DrChinese]

DrChinese said: I think it is acknowledged by most that it is NOT POSSIBLE, in principle, to know all of the state information about all of the particles in the universe. This would be true regardless of what theory you think is applicable; because it would take an even bigger universe to store that much knowledge.

Nah, it is merely stated so without any profound comments

That is outrageously false: QM contains many more degrees of freedom than CM does (as an easy excercise proves it to be). If QM would contain LESS degrees of freedom, then I would be all ears

I think it is true that "most" do NOT think it is possible, in principle, to know all of the state information of the universe. I can easily acknowledge that you are not most.

However, I think there have been authors who have previously demonstrated that the amount of information it would take to describe a single particle's classical observables - such as position, momentum, etc. - easily equals or exceeds that which can be stored in a single particle. You would need that to be able to calculate, using BM, the trajectories of particles. So to describe one universe, you'd need at least an additional universe to store that description. But then you'd need to describe the universe where you are storing that information too. Etc.

So I conclude that it is not possible, in principle, to accomplish the result postulated by the OP - even under the concepts of BM. I acknowledge that some BMers will not be satisfied by this reasoning. (For anyone who accepts QM, the answer was already a big NO.)

Of course, if you are a local realist, then you are already in denial and therefore logic (Bell's Theorem) and experiment (Aspect etc.) aren't really a factor. :rofl:

 

[Loren Booda]

Agreed that, when only considering the microscopic and conventionally classical, quantum mechanics is not determinable. Is there however a macroscopic version of Q. M. so that the correspondence principle in regards to scale is reversed - like the quantum cosmology of the big bang's first three minutes being spread across the sky relative to us?

 

[Careful]

**I think it is true that "most" do NOT think it is possible, in principle, to know all of the state information of the universe. I can easily acknowledge that you are not most. **

You don't seem to have understood that I was merely making fun of your use of the verb ``to THINK´´.

**
However, I think there have been authors who have previously demonstrated that the amount of information it would take to describe a single particle's classical observables - such as position, momentum, etc. - easily equals or exceeds that which can be stored in a single particle. You would need that to be able to calculate, using BM, the trajectories of particles. So to describe one universe, you'd need at least an additional universe to store that description. But then you'd need to describe the universe where you are storing that information too. Etc. **

That is obviously true when you take SQM as the *correct* BASIS theory; this was clearly not what I was aiming at as you should know...

**
Of course, if you are a local realist, then you are already in denial and therefore logic (Bell's Theorem) and experiment (Aspect etc.) aren't really a factor. :rofl: **

Not at all, local realists usually are much more aware of the fineprint of Bell's theorem than QM worshippers are Again, you should take a look at this paper concerning Occam's raisor (just a friendly remark - I promised to behave until the end of the year )

 

[masudr]

...it is a bold (and probably mistaken) assumption that you can exclude GR in the description of elementary particle physics...

It is, on the contrary, very bold to assume you can include GR on any issue regarding QM, until we have a satisfactory mathematical formalism to do so.

 

[Careful]

It is, on the contrary, very bold to assume you can include GR on any issue regarding QM, until we have a satisfactory mathematical formalism to do so.

Just because you might have missed something: I am one of these crazy persons who says you should not quantize gravity at all. Until further notice, I remain convinced that QM (when applied to the correct distance scales) *is* a statistical description of a classical gravito/electromagnetic process (that is why I insist so much that Bell test results are treated with the correct egards.). In such a programme, there is NO need for a new mathematical machinery, but there IS a huge demand for realistic matter models of elementary particles. Such particle models ARE already partially available (for the electron say) and guess what?? The nonlinear corrections due to Einstein Maxwell theory on the Compton scale (vis a vis flat Minkowski electromagnetism) are immense (and gravitation is extremely important in this regard).

Cheers,

Careful

 

[DrChinese]

You don't seem to have understood that I was merely making fun of your use of the verb ``to THINK´´.

Again, you should take a look at this paper concerning Occam's raisor (just a friendly remark - I promised to behave until the end of the year )

I missed that the first time I read it, so a point for you.

I assume you are referring to Sorkin's 1982 paper, http://physics.syr.edu/~sorkin/some.papers/ . This is kinda interesting, I have a definite interest in this paper. However, I am not sure exactly how you relate it to this thread. He does provide an estimate of the amount of storage required to express certain values (his 3.4). Is this what you had in mind? Or perhaps the idea that the theory with fewest parameters is the best?

 

[DrChinese]

I am one of these crazy persons who says you should not quantize gravity at all.

That doesn't seem weird to me at all. I wake up at least 4 days a week with that view.

But I am not sure I follow the other part of what you said about electro/gravito interactions. Surely you are not saying that electromagnetism and gravity should be unified... are you?

 

[Rade]

**** Is it not true that we can derive classical dynamics at macroscopic scale from QM, **No, that is not known yet Cheers,Careful

But (1) does not the Ehrenfest theorem show that a form of QM wave equation leads to Newton's equations of motion on the average (see Bohm, 1951, Quantum Theory), and (2) does not the Wentzel-Kramers-Brillouin (WKB) approximation allow for wave functions which, as stated by Bohm (1951) on page 268, is "the same as the classical probability distribution function". So, from the above it does appear that at least some aspects of classical dynamics are well known to be derived from QM--at least according to Bohm. But I no expert at all in this field, so I look forward to your correction of my errors.

 

[Careful]

**
I assume you are referring to Sorkin's 1982 paper, http://physics.syr.edu/~sorkin/some.papers/ . This is kinda interesting, I have a definite interest in this paper. However, I am not sure exactly how you relate it to this thread. **

It relates to your Bell comments on this thread and my nagging that IF you would have a classical explanation which would fit the RAW data in Bell experiments very well, then this theory would be a gazillion times preferred over quantum mechanics (remember that you also have to store the spatial wave functions in your computer).

**
He does provide an estimate of the amount of storage required to express certain values (his 3.4). Is this what you had in mind? Or perhaps the idea that the theory with fewest parameters is the best? **

Not quite, his Occam's razor principle is a balancing act between theory complexity and accuracy of prediction as it should be. It kills off oversimplified theories which make bad predictions as well as too complex theories which do make accurate fits. The ideal theory is the one which makes a particular experimental outcome very likely and is not more complex than it should be (of course it is a subject of debate how to define theory complexity ). For this Occam's razor principle to apply, it is of ultimate importance that we have detailed information about Bell test setups as well as the RAW (unprocessed) data at hand.

 

[Careful]

That doesn't seem weird to me at all. I wake up at least 4 days a week with that view.
But I am not sure I follow the other part of what you said about electro/gravito interactions. Surely you are not saying that electromagnetism and gravity should be unified... are you?

No, I am not saying that at all. In quantum gravity, one of the main (unadressed so far) puzzles (I refuse to accept the MWI interpretation as a credible theory - what is the probability that you are in one universe (P = 0)? and what do you need these other universes for (Occam's razor)? and what makes that you are just in one universe and not in another (consciousness and zombie copies)?) is how to avoid (more or less equal weighted) superpositions of cats (to use Schroedingers example). Another problem is how to deal with the measurement: surely one would like to have ONE dynamics for the microworld and on long distance scales R is not compatible with the local physics we are used to in the macroworld.

Therefore, one may legitimatly wonder to what degree the simplest theory (= CM) extends to the microscale. It is not an accident that spacetime locality is the most simplifying principle in nature; therefore it seems much more logical to study in detail Einstein Maxwell theory BEFORE you start thinking about new exotic mathematics and physics (it turns out that EM theory is more than complex enough with a wealth of unexpected results). That is why I believe LR deserves much more attention (in the QG community) and why experimentators should drop their bias in presenting results about Bell tests.

Cheers,

Careful

 

[Careful]

But (1) does not the Ehrenfest theorem show that a form of QM wave equation leads to Newton's equations of motion on the average (see Bohm, 1951, Quantum Theory), and (2) does not the Wentzel-Kramers-Brillouin (WKB) approximation allow for wave functions which, as stated by Bohm (1951) on page 268, is "the same as the classical probability distribution function". So, from the above it does appear that at least some aspects of classical dynamics are well known to be derived from QM--at least according to Bohm. But I no expert at all in this field, so I look forward to your correction of my errors.

Sure, this is is correct but by far not sufficient (see my previous post).

Cheers,

Careful

 

[DrChinese]

Not quite, his Occam's razor principle is a balancing act between theory complexity and accuracy of prediction as it should be. It kills off oversimplified theories which make bad predictions as well as too complex theories which do make accurate fits. The ideal theory is the one which makes a particular experimental outcome very likely and is not more complex than it should be (of course it is a subject of debate how to define theory complexity ). For this Occam's razor principle to apply, it is of ultimate importance that we have detailed information about Bell test setups as well as the RAW (unprocessed) data at hand.

Ideal=his ideal. Theories with different inputs and different output precisions can all be good. Newtonian gravity - as a theory - is absolutely useful, even in the presence of Einstein's GR. I would not agree that one is objectively better than the other in all cases.

 

[Careful]

Ideal=his ideal. Theories with different inputs and different output precisions can all be good. Newtonian gravity - as a theory - is absolutely useful, even in the presence of Einstein's GR. I would not agree that one is objectively better than the other in all cases.

Well, objectively, it seems to me that the theory which is known to have the broader range and the higher accuracy is always preferred. However, I do admit that the relation between Newtonian Mechanics and GR is much more subtle that the one with QM since NM is actually a simpler theory and still does a wonderful job at the scales of everyday life. Therefore, in the right circumstances, it is more profitable to use NM in practice.

 

[inha]

Perhaps Loren, you could phrase your question in more detail (for example what do you mean by quantum cosmology?). Nevertheless I can add at this moment that Inha's answer has a strong dependence on his interpretation of the measurement problem (it is not for sure at all that true random generators exist in nature - although they do exist in theory if we take QM as it stands now).

The only measurement problem I have first hand experience with is a scintillator detector dying on me. Wanna guess my philosophical stance on QM?

 

[Careful]

The only measurement problem I have first hand experience with is a scintillator detector dying on me. Wanna guess my philosophical stance on QM?

Nah, I bet you are pleased with copenhagen ... but if you want to explain your point of view, go ahead.

 

[Locrian]

But (1) does not the Ehrenfest theorem show that a form of QM wave equation leads to Newton's equations of motion on the average (see Bohm, 1951, Quantum Theory),

A lot of good that does you! How exactly do you plan to use those approximations on wave equations? How do you get from wave equations that are operated upon in QM to the objects that are operated upon in classical mechanics? Where is the thorough derivation of this?

I would love to see someone try to generate equations of motion of a rocket under thrust using principles of quantum mechanics. Of course it can't be done, and because of more than just problems with calculation.

and (2) does not the Wentzel-Kramers-Brillouin (WKB) approximation allow for wave functions which, as stated by Bohm (1951) on page 268, is "the same as the classical probability distribution function".

And what would that show, exactly?

No, the idea that you can get classical mechanics from quantum mechanics is a myth perpetuated by poor textbooks. In textbooks they create imaginary situations where QM and CM end up with similar results over many measurements, but in reality these situations rarely occur; more often, trying to use QM to produce classical equations for a situation is either impossible or a failure. Nevermind the fact that "proof" by example is no proof at all! There is no mathematical proof that shows that classical mechanics can really be produced from quantum mechanics, and in practice it is a miserable waste of time to try. This idea fails in both philosophy and practice.

 

[Careful]

No, the idea that you can get classical mechanics from quantum mechanics is a myth perpetuated by poor textbooks. In textbooks they create imaginary situations where QM and CM end up with similar results over many measurements, but in reality these situations rarely occur; more often, trying to use QM to produce classical equations for a situation is either impossible or a failure. Nevermind the fact that "proof" by example is no proof at all! There is no mathematical proof that shows that classical mechanics can really be produced from quantum mechanics, and in practice it is a miserable waste of time to try. This idea fails in both philosophy and practice.

Sensible reply Just two small questions : (a) did you waste too much time in trying this ? (b) what is the conclusion you draw ?

Cheers,

Careful

 

[Locrian]

(a) did you waste too much time in trying this ?

I believed it much longer than I should have and didn't waste nearly enough time trying it.

(b) what is the conclusion you draw ?

That mathematical paradigms are only useful (valid?) within the experimental regime that produced them.