Questions About Quantum Theory: What's Wrong?

[reilly]

Daniel and Zz -- I'm pretty sure that my "Sakarov" is indeed Sakharov the great. And, funny, after I posted I realized Bardeen should be on the list, as well as Von Neumann. Bardeen rocks dude, as some might say today. But, hey, what do we know?
Regards,
Reilly

 

[dextercioby]

Well,Mr.Atkinson,if u mentioned a mathematician,it would be fair to mention all 3 of them:von Neumann,Wigner and Weyl...

Daniel.

P.S.If u mentioned Sakharov,it would be fair to include Kurchatov,as well,after all,in the Manhattan project,a dozen of brilliant physicists worked and people remember all their names (or at least should),but in the russian version,besides Kurchatov & Sakharov,i really doubt any westerner knows other names...

 

[Crosson]

You cannot explain the existence of certain particles much as you cannot explain the existence of this universe.

In some sense this quote is correct; it is foolish to seek some ultimate explanation that leaves no further questions i.e. to begin we must assume certain things.


Young's classical double slit experiment established without a doubt that light had wave like properties. A generation of physicists grew up accepting this wave nature of light as a fundamental assumption (experimental fact).

Because of physics we now have a reason why light has wave properties. Although this reason requires assumptions of its own, it is quite satisfying.

QM is a great theory, just like the work of Young and Fresnel. But anyone who says the experiments of QM can't be explained any other way, is shortsighted at best.

 

[dextercioby]

QM is a great theory, just like the work of Young and Fresnel. But anyone who says the experiments of QM can't be explained any other way, is shortsighted at best.

Okay,i understand the concept of evolution is science,but WHY would we seek an alternative theory (to QM,to GR,to SM) to account for experimental results which are in agreement with actual (partial,c'est vrai ) theories.

What i understand from your last post is that someone,in order not to be called "shortsighted" should desparately search for a new theory which would account,let's say for the first 11 sign.digits of g_{el}...

I cannot follow that logics.We must search for theories which would COMPLEMENT QM (& GR),not replace them...



Daniel.

P.S.Or if they do replace them,at least be able to reproduce the same results obtained by QM & GR at least with the same accuracy.

 

[Stingray]

Okay,i understand the concept of evolution is science,but WHY would we seek an alternative theory (to QM,to GR,to SM) to account for experimental results which are in agreement with actual (partial,c'est vrai ) theories.

Besides being consistent with experiments, any physical theory must also be logically self-consistent (of course, sufficiently elaborate experiments would reveal logical inconsistencies, but that doesn't mean that they are practical to perform). Ignoring everything else that has been discussed in this thread, it is clear that QM has serious difficulties when applied to spacetime geometry. This is a logical inconsistency in the theory that must be resolved despite the lack of any current experimental problems.

I also wonder how exactly the Copenhagen interpretation would be applied to quantum geometry... How would quantum cosmology work? As far as I know, these sorts of issues are what prompted Gell-Mann and Hartle to develop their ideas. Unfortunately, I haven't ever gotten around to reading their papers.

From another point of view, finding alternative formulations of the same theory has been useful historically. In classical mechanics, we had Newton's "interpretation," Lagrange's, Hamilton's, etc. Each of these is particularly suited to different types of problems, and lends itself to somewhat different types of intuition. Also, these ideas were fundamental for the development of QM. Even within QM, we have the Schrodinger and Heisenberg representations, as well as Feynman's path integrals, etc. While technically equivalent, it would be ridiculous to claim that it is only important to learn one of these formulations.

Reilly, your sarcastic comments towards caribou are a bit strange. I think we all agree that copenhagen is extremely useful for a wide variety of situations. That doesn't mean that other ideas are pointless (either pedagogically or as a matter of principle). They obviously must reproduce the usual results in all experimentally tested situations. It would be particularly elegant, for example, if Born's rule were to emerge as a limit of something else. This would have experimentally measurable consequences, although we might be quite far from being able to measure them.

 

[dextercioby]

Besides being consistent with experiments, any physical theory must also be logically self-consistent (of course, sufficiently elaborate experiments would reveal logical inconsistencies, but that doesn't mean that they are practical to perform).

All partial theories developped and worldwide recognized so far are...

Ignoring everything else that has been discussed in this thread, it is clear that QM has serious difficulties when applied to spacetime geometry.

That is because the axioms of the nonrelativistic QM do not use the notion of spacetime.In fact,they're so abstract,that even the physical space (euclidean in Newtonian physics) is taken outta the picture...

This is a logical inconsistency in the theory that must be resolved despite the lack of any current experimental problems.

There is no logical inconsistence.QM doesn't explain gravity and spacetime,because it wasn't built to do it and,incidentally,every attempt to quantize gravity following the receipt given by the 6 axioms has failed so far.

I also wonder how exactly the Copenhagen interpretation would be applied to quantum geometry... How would quantum cosmology work? As far as I know, these sorts of issues are what prompted Gell-Mann and Hartle to develop their ideas. Unfortunately, I haven't ever gotten around to reading their papers.

From another point of view, finding alternative formulations of the same theory has been useful historically. In classical mechanics, we had Newton's "interpretation," Lagrange's, Hamilton's, etc. Each of these is particularly suited to different types of problems, and lends itself to somewhat different types of intuition. Also, these ideas were fundamental for the development of QM. Even within QM, we have the Schrodinger and Heisenberg representations, as well as Feynman's path integrals, etc. While technically equivalent, it would be ridiculous to claim that it is only important to learn one of these formulations.

Reilly, your sarcastic comments towards caribou are a bit strange. I think we all agree that copenhagen is extremely useful for a wide variety of situations. That doesn't mean that other ideas are pointless (either pedagogically or as a matter of principle). They obviously must reproduce the usual results in all experimentally tested situations. It would be particularly elegant, for example, if Born's rule were to emerge as a limit of something else. This would have experimentally measurable consequences, although we might be quite far from being able to measure them.

I know that I'm nitpicking,but if u decide to talk about QM,at least use its terminology properly.E.g.Schrödinger,Heisenberg & interaction (a.k.a.Dirac-Tomonaga-Schwinger) picture(s) .

Daniel.

 

[Stingray]

QM doesn't explain gravity and spacetime,because it wasn't built to do it and,incidentally,every attempt to quantize gravity following the receipt given by the 6 axioms has failed so far.

My point exactly. I was being a little loose with my wording. Fundamental physics as a whole is not logically self-consistent, and QFT/QM is obviously a part of this (I've been saying QM even when I mean QFT - sorry about the confusion). It is not very meaningful to say that QFT is just a stand-alone mathematical structure.

By the way, I don't understand your statement that the axioms of nonrelativistic QM are independent of spacetime. As I've learned them, they include Schrodinger's equation. What does that time derivative mean without a notion of spacetime (Newtonian or otherwise)? Can nonrelativistic QM be formulated on a classical curved background? I know QFT can, but you have to start from a completely different viewpoint than the one in textbooks. There might still be lingering issues as well. I don't know.

 

[ZapperZ]

Besides being consistent with experiments, any physical theory must also be logically self-consistent (of course, sufficiently elaborate experiments would reveal logical inconsistencies, but that doesn't mean that they are practical to perform). Ignoring everything else that has been discussed in this thread, it is clear that QM has serious difficulties when applied to spacetime geometry. This is a logical inconsistency in the theory that must be resolved despite the lack of any current experimental problems.

What does it mean to say something is "logically inconsistent"? What is logically inconsistent about QM? That it is built on a set of axiom that cannot be derived via First Principles? That it isn't built logically like mathematics?

Why does "difficulties when applied to spacetime geometry" implies "logical inconsistencies"? Does difficulty in applying BCS theory to High-Tc superconductors implies logical inconsistency of BCS theory, even when it is the MOST verified theory of a phenomenon in history? If we were to extend QM to include GR, does it then make it "logically consistent" when it wasn't before?

Zz.

 

[caribou]

Caribou -- I'm happy for you, and sad for the rest of us. I guess we've been mistaken all along. I used practical Copenhagen in my Ph.D thesis, perhaps I should redo my work, and my various published papers. Given your expertise, what do you think? I suspect I'm not the only person in the Forum facing such difficulties. Do we need to come up with, as you state it, a theory other than QM in order to save our status as physicists, active or retired? I ask because of your imperative, "no doubt about it" directive to develop this other theory. And all this time I thought I had a reasonable clue about what I was doing. And worse yet, your directive very much invalidates most of the physics from the 1920s until recently. Looks to me like an impending crisis.

I have no idea how you arrived at these conclusions, as my post suggests nothing even remotely like that.

The Copenhagan interpretation works just fine in most respects, as long as you don't ask certain questions that many never ask. These are questions, for example, like how a theory with "external observers" is supposed to apply to the universe as a whole or how wave functions can "collapse" with no interaction.

These issues don't matter in most situations but Copenhagen is simply inadequate for something like quantum cosmology, hence Hartle's interest in the clarification of quantum mechanics for the field he and Hawking helped found in a modern sense.

Decoherent histories is simply about clarifying following quantum mechanics by following quantum mechanics and seeing where it leads.

 

[caribou]

Reilly, you can lead a horse to water, but you can't lead water to a horse.

Instead of talking of horses and water, how about you talk about instantaneous wave function collapse over galactic distances with no interaction causing the collapse?

It's not a time-consuming challenge. Just sum up in a few words what you think of it.

 

[Chronos]

Instead of talking of horses and water, how about you talk about instantaneous wave function collapse over galactic distances with no interaction causing the collapse?

It's not a time-consuming challenge. Just sum up in a few words what you think of it.

That is no more mysterious than any other wave function collapse. Observation is the interaction causing the collapse.

 

[Stingray]

Why does "difficulties when applied to spacetime geometry" implies "logical inconsistencies"?

To put it most succintly, it is logically inconsistent to claim that GR and QFT (+standard model) together form a fundamental description of our universe. On a purely mathematical level, either one of these theories works just fine without the other. But experiment has shown that we can't throw out either of them (at the appropriate scales). We also can't combine them in any consistent way (so far). It is therefore obvious that fundamental physics needs to be extended.

I am calling this a "logical inconsistency" because we are arriving at the problem without any direct experimental evidence. The only sense in which experiment is involved is that both GR and QM are essentially our favorite (minimal) extrapolations of all known experiments (at least we think that they are consistent with everything).

Does difficulty in applying BCS theory to High-Tc superconductors implies logical inconsistency of BCS theory, even when it is the MOST verified theory of a phenomenon in history?

BCS theory has never been claimed to be fundamental physics. So no, there is no logical inconsistency implied by it being difficult to apply to high-Tc superconductors. It is just a limit to the model (apparently - I'm far from an expert).

Now, you might object that both GR and QM are also just models, and that labelling them "fundamental physics" is just arrogant semantics. I can argue against this, but there isn't much point. The statement that there are known problems with {GR + QFT} stands, and you can call them whatever you like.

If we were to extend QM to include GR, does it then make it "logically consistent" when it wasn't before?

Yes.

 

[caribou]

It's even worse for me. As a lowly experimentalist, all I've been doing throughout my carreer have been to "shut up and calculate". I had very little time, nor inclination to make any philosophical interpretation - mainly because I care more about what I can show physically, and the fact that philosophical interpretation on this has not produced any significant contribution to the advancement of physics.

Then it seems your lack of time and inclination to study interpretation means there might be a slight chance you were aware that lecturing me on what consistent histories does and does not say could be taking a risk on the off-chance you hadn't fully understood the theory.

Well, I've studied the subject part-time for a couple of years and, as just one example, I would never think "incompatible" meant contradictory. However, as I have said before, specialists in interpretation are making errors and, in fact, that's one of those that they make, so it would be extremely unfair to hold anyone else to account. So I didn't and I don't.

That's why it's not a big deal to me that you were trying to "correct" me on something I know doesn't need correction. If it was obvious then it wouldn't have taken decades to be found in the quantum mechanics. Only Bohr with his genius for insight anticipated it.

But I don't want to talk about this anymore. It's not important for most physics anyway. We can agree to disagree and leave it at that.

 

[Locrian]

It's not important for most physics anyway.

Actually, can you give any instance in physics in which it is important? It's total lack of importance is part of the point other posters are trying to make.

 

[caribou]

That is no more mysterious than any other wave function collapse. Observation is the interaction causing the collapse.

I see.

What about this problem of "interaction-free measurement" in which if a particle has a wave function which means it could be detected at A at time 1 or at B at time 2, we know by simple logic that if there was no interaction at A at time 1 then the particle will later be detected at B at time 2.

The wave function collapsed at time 1 because nothing happened. Or it collapsed at time 1 because we observed that nothing happened.

Either way, the wave function collapsed without any physical interaction.

Now that seems a bit strange to me.

The physicists whose work I was describing have found that wave function collapse is simply a mathematical shortcut and not a physical effect.

Now that makes more sense to me.

What do you think?

 

[caribou]

Actually, can you give any instance in physics in which it is important? It's total lack of importance is part of the point other posters are trying to make.

Well, it's unlikely to make new predictions. It's mainly about resolving the long-standing issues.

Just understanding how macroscopic quasiclassical physics emerges from microscopic quantum physics is one such issue. Removing the idea of external observeration being needed for the universe in quantum cosmology would be another. And then there is taming paradoxes like delayed-choice, EPR, GHZ and so on.

This thread originally asked what's wrong with quantum mechanics. I'm relating that it appears that nothing is really wrong with it and it seems to have less problems than a lot of people think.

How important anyone thinks that is would be up to them.

 

[elas]

We ditch copenhagen only if we find another interepretation which can predict physical phenomena as well, if not better than copenhagen and (hopefully) consistent with other theories (GR, for example).

Glad to here there is a debate going on somewhere, pity it is not on PF; but I disagree with your last paragraph, theories predict, interpretations explain. My case is that Quantum theory does predict with great accuracy, but the Standard Model interpretation does not explain at all well; therefore we should concentrate on interpretation before trying to improve further our ability to predict. Put another way, it could be that a better explanation would show that Quantum theory is far more complete than is realized at present and therefore the search for alternative theories (string etc) is a waste of time and effort.

 

[Eye_in_the_Sky]

reply to post #76

Of course, many measurements, particularlry on large systems induce small changes in the system. But that's not the issue. The issue is "before and after", whether a coin toss, winning in poker, ascertaining the temperature of bath water -- who wants to injure their child with water that is too hot. While you may have some notions about the temp of the bath water -- a Baysean situation -- you don't know until you measure -- with your hand or foot, or with a thermometer. The water temp does not change, but your head does -- you go from "I don't know" to "I know". That's collapse.

That is only 'half' of the story of "collapse". It is the 'half' which concerns the knower – i.e. the subject.

The other 'half' of the story concerns the system which has been measured – i.e. the object. What did it go from? ... and what did it get to?

And this is my point. If one asserts that the "collapse" phenomenon is the same in a quantum scenario as it is in a classical scenario – (not merely with respect to the subject, but) also with respect to the object – then one has to (at least implicitly) assume the physical existence of "hidden variables" in the quantum case.
_______________

QM state vectors,based on the famous complete set of measurements, apart from a phase factor, by definition, give a complete description of the system at hand. That's as basic as it gets.

It is unclear to me why you have brought up the subject of a "CSCO". Is it because you are interpreting a statement such as the following in terms of CSCO's?

The quantum-mechanical state-vector description is "complete".

This is a statement about "physical reality". It purports that the "real factual situation" pertaining to the system at hand is completely characterized by the state vector; i.e. "hidden variables" have no physical existence.

So, perhaps then this is the point which you have been trying to make all along:

Do not attempt to interpret the state vector in terms of the "object" (i.e. the system at hand), but do so only in terms of the "subject" (i.e. the knower). Then, a quantum "collapse" scenario is no different from a classical one.

If that is your point, then I reply:

Of course, "collapse" will then be the same! You have chosen to disregard the one respect in which it can be different. But when that respect is taken under consideration, it turns out that "collapse" can then be said to be the same only if the state-vector description is not "complete"; i.e. "hidden variables" exist.
_______________

I still don't get the need for hidden variables -- in my dissertation I used QED to compute radiative corrections for various electron-nucleon scattering experiments, which helped map out the electromagnetic structure of nucleons. Should I be worried that I didn't use hidden variables?

In order to perform calculations, it suffices to use a minimal interpretation of the "shut-up-and-calculate" genre. Therefore, there would be no need whatsoever to invoke the notion of "hidden variables". Invoking such notions may, however, become relevant in the context of statements made regarding the nature of "reality".

 

[Chronos]

Reality is subjective, not objective. We are incapable of accessing 'objective' reality. The only reality we can observe is necessarily subjective. The mere act of making an observation perturbs the nature of a system in a very fundamental way. And we can only observe systemic effects because we cannot observationally isolate fundamental elements of systems. We are limited to observing their interactions - i.e., their relationships one to another. Generally relativity and QT are fundamentally connected in that respect - all interactions are relative.

 

[Mazuz]

Reality is subjective, not objective.

What about the viewpoint that our subjective reality is an illusion created by an objective universe.

Perhaps you only meant that by our very nature escaping a subjective viewpoint is impossible.

 

[dextercioby]

_______________It is unclear to me why you have brought up the subject of a "CSCO". Is it because you are interpreting a statement such as the following in terms of CSCO's?

The quantum-mechanical state-vector description is "complete".

This is a statement about "physical reality". It purports that the "real factual situation" pertaining to the system at hand is completely characterized by the state vector; i.e. "hidden variables" have no physical existence.

That is a logical interpretation of an axiom.The first axiom.If u reject it,by claiming that the "hidden variables",which are obviously excluded by 1-st principle and by the claim that all one needs to know is an CSCO and solve SE,have "physical existence",then u don't have QM anymore.You'd have to reject the whole theory.Can u do that...?Iff you have a viable alternative;for 80 yrs one doesn't have that and my opinion is he won't...

You can't fight against a postulate (e.g.von Neumann's about collapsing state vector when measuring),without refuting all theory.As i said,as in other physical theories,postulates form not only a logical structure,but also a unitary structure...It's like having something against Einstein's second postulate of SR,just because this postulate asserts "c" constance withoout taking into account "hidden variables",or who knows what else...

Daniel.

 

[vanesch]

The problem we are discussing is written in Aristotles meditations.

I guess you mean Descartes

Patrick.

 

[vanesch]

What is the essential postulate from which one can derive all of QM?

The superposition principle: if A and B are two possible physical states, then
a |A> + b |B> is also an existing physical state.

And then you need some embellishment (Hilbert spaces, operators etc...) to make this meaningful but this is the essential idea of quantum theory.

cheers,
Patrick.

 

[ZapperZ]

To put it most succintly, it is logically inconsistent to claim that GR and QFT (+standard model) together form a fundamental description of our universe. On a purely mathematical level, either one of these theories works just fine without the other. But experiment has shown that we can't throw out either of them (at the appropriate scales). We also can't combine them in any consistent way (so far). It is therefore obvious that fundamental physics needs to be extended.

I am calling this a "logical inconsistency" because we are arriving at the problem without any direct experimental evidence. The only sense in which experiment is involved is that both GR and QM are essentially our favorite (minimal) extrapolations of all known experiments (at least we think that they are consistent with everything).

But isn't this putting the cart before the horse? The FACT that this is still a highly active research area means that you are already set in your ways that GR and QM/QFT cannot be merged. And if they are successful in merging those two, now it becomes logically consistent? I find that highly logically inconsistent!

Again, even when GR and QM/QFT cannot be made into a unified theory, I do not see why they are both logically inconsistent UNTIL there are experimental evidence to point to such a notion. Until we get to a scale where they both should work equally well and we can see where they both deviate, then we can't say anything. This is playing by your rule of requiring "direct experimental evidence". You too cannot claim of logical inconsistencies without directly experimental evidence. Last time I checked, we have no such evidence where QM/QFT and GR can be tested on equal grounds.

BCS theory has never been claimed to be fundamental physics. So no, there is no logical inconsistency implied by it being difficult to apply to high-Tc superconductors. It is just a limit to the model (apparently - I'm far from an expert).

What does being "fundamental" (another area we can debate on) has anything to do with what you are applying? You are using an example of an ongoing evolution of an idea and pointing out that just because it STILL cannot be merged into another, it is logically inconsistent. If we apply that principle, almost everything that we have in physics are logically inconsistent. There's nothing "fundamental" about this.

I find it highly puzzling that when there are still issues regarding the merging of QM with GR, it is QM that is pointed out to be "logically inconsistent". If you look at the degree of certainty in terms of experimental observations, QM outstrip GR by orders and orders of magnitude. The validity of QM can be found in all of your modern electronics. We can manipulate, engineer, and change various parameters to test many parts of QM EASILY. The body of evidence for QM is astounding. Now look at GR. I'm not claiming that it is wrong, but c'mon people. Look at the nature of the evidence and how many there are! Not a single evidence from GR can come even close to the degree of certainty of, let's say, the evidence for an energy gap in the superconducting state of a superconductor!

Yet, what do we get? QM cannot agree with GR, so QM must be logically inconsistent. I find that conclusion to be highly illogical based on the wealth of experimental evidence alone.

Zz.

 

[ZapperZ]

I see.

What about this problem of "interaction-free measurement" in which if a particle has a wave function which means it could be detected at A at time 1 or at B at time 2, we know by simple logic that if there was no interaction at A at time 1 then the particle will later be detected at B at time 2.

The wave function collapsed at time 1 because nothing happened. Or it collapsed at time 1 because we observed that nothing happened.

Either way, the wave function collapsed without any physical interaction.

Now that seems a bit strange to me.

The physicists whose work I was describing have found that wave function collapse is simply a mathematical shortcut and not a physical effect.

Now that makes more sense to me.

What do you think?

Say what?!

Isn't "interaction-free measurement" an oxymoron? Can you please construct a QM state that fits into your description above?

Zz.

 

[dextercioby]

The superposition principle: if A and B are two possible physical states, then
a |A> + b |B> is also an existing physical state.

And then you need some embellishment (Hilbert spaces, operators etc...) to make this meaningful but this is the essential idea of quantum theory.

cheers,
Patrick.

I think your post can be resumed in one word only

LINEARITY...

Daniel.

 

[vanesch]

I think your post can be resumed in one word only

LINEARITY...

Well, from a mathematical point of view, of course, the "superposition principle" and "linearity" are about the same. But there is something physical to the "superposition principle" which is maybe not captured by the term "linearity".
"Linearity" seems to be a requirement on the kinds of equations or so of a theory. For instance, one is tempted to say that Maxwell's equations are "linear". But people who say "linear" usually think of "first order approximation". You can potentially think of small non-linearities "correcting" the linear theory.
But the superposition principle in QM is not so much about the equations. It is about the possible states a system can be in. And here, the strange, bold and weird properties of QM all come together: if you postulate that configuration "A" is a possible state of your system (be it a particle, a field, a solar system, an atom, whatever) and if "B" is also a possible, different, state of your system, then there exists a DIFFERENT state for each complex couple (a,b) modulo a common factor, which is described by a |A> + b |B>.
This contains the essence, and all the weirdness, of QM. If "sitting on your chair" is one of your states, and "lying on your bed" is another one, then there are, by fundamental postulate, a miriad of different states you can be in, namely a x "lying on your bed" + b x "sitting on your chair", for each couple of complex numbers (a,b) modulo a complex factor.
This at first totally absurd idea is the very foundation of quantum theory. If you tweak it, you don't have a quantum theory anymore.

The other aspect of the superposition principle is less fundamental, but nevertheless important, namely the hypothesis that the time evolution operator U(t,t') is a linear operator over the state space. This one comes closer to your "linearity" requirement. It is conceivable that one could modify this (small non-linear corrections) and still talk about a kind of quantum theory. But you cannot do away with the first "state space" superposition.

cheers,
Patrick.

 

[vanesch]

Again, even when GR and QM/QFT cannot be made into a unified theory, I do not see why they are both logically inconsistent UNTIL there are experimental evidence to point to such a notion. Until we get to a scale where they both should work equally well and we can see where they both deviate, then we can't say anything. This is playing by your rule of requiring "direct experimental evidence". You too cannot claim of logical inconsistencies without directly experimental evidence. Last time I checked, we have no such evidence where QM/QFT and GR can be tested on equal grounds.

You are right of course that the clash between GR and QM does not mean that QM has to be logically inconsistent. In fact, as far as I understand (which is not much) the most successful candidates to resolve the issue (superstrings and loop quantum gravity) stick in fact to QM and modify the gravity part. However, they are still far from achieving their goals, even just on paper, not even talking about experiments, so I'd say that the judge is still out (and will be - for a long time !).

Nevertheless, quantum theory (in the Copenhagen version) does have a serious inconsistency, or at least, an issue that should be resolved one day, and that is the projection postulate. For all its practical value (and no, you don't have to rewrite your PhD because you used it), the issue remains: what sets apart an interaction we label "measurement" from an interaction we consider "part of the system, in the hamiltonian" so that it does TWO TOTALLY INCOMPATIBLE THINGS to the wavefunction ?
Now, as I said, this is, for the foreseeable future, NO PRACTICAL ISSUE, because we are still doing quantum experiments which are so remote from our macroscopic, "classical" world that we can put a "Heisenberg cut" anywhere between the system and us, with identical results (thanks to decoherence theory). But it is an issue of principle, no ? And it might be touched upon by the eventual modifications needed to deal with gravity. Also, maybe one day, when our quantum experiments WILL have reached a level of sophisitication that is unheard of today, the issue will have practical consequences.

The second point is, that even if we have no direct experimental access, we know that the very early universe, as well as in the vincinity of black holes, quantum theory AND GR must play a role. So the clash between GR and QM is real, because real situations exist where both should be important. It is not that they deal with non-overlapping domains, even if we have no direct experimental access to their domain of overlap yet.

So although as of today, and the near (and even not-so-near) future, quantum theory as we know it, using Born's rule, gives satisfying results, and leads to many interesting applications and fine science, its limits are "in view": collapse or no collapse should be resolved one day, and QM/GR should be resolved one day. And I wouldn't make any bet that QM will come out of it without any modification. It might be. It might not.

cheers,
Patrick.

 

[ZapperZ]

Nevertheless, quantum theory (in the Copenhagen version) does have a serious inconsistency, or at least, an issue that should be resolved one day, and that is the projection postulate. For all its practical value (and no, you don't have to rewrite your PhD because you used it), the issue remains: what sets apart an interaction we label "measurement" from an interaction we consider "part of the system, in the hamiltonian" so that it does TWO TOTALLY INCOMPATIBLE THINGS to the wavefunction ?

Well, here's where we differ. I can't tell if, even if we buy into CI, that we have a "logical inconsistency" or simply it offends our "tastes"! We find it uncomfortable to say that an electron occupies BOTH H atoms simultaneously in an H2 molecule, or that the superconducting current flows in BOTH directions at the same time in the Delft and Stony Brook's SQUID experiment. But nature owes us nothing to make us comfortable. To have something flowing in BOTH direction at the same time can be argued to be "logically inconsistent", but this assumes a priori that our common sense on how things should behave is valid. And we all know that our "common sense" are built on classical underpinnings. Our concept of "time", "position", "momentum", "energy", etc. are all classical ideas. When we force those into where it doesnt' fit, OF COURSE we will get strange answers (square objects through round holes).

My point is that what is there to distinguish between CI having inconsistent interpretation with us forcing something to confine to our tastes? We all agree that ALL experiments so far have agreed with QM's predictions. I find it less "offensive" to have QM make predictions that offends and contradicts my common sense. Usually, when that happens, it signifies new physics. An electron can fractionalize separately into its spin and charge components? Bring it on!

The second point is, that even if we have no direct experimental access, we know that the very early universe, as well as in the vincinity of black holes, quantum theory AND GR must play a role. So the clash between GR and QM is real, because real situations exist where both should be important. It is not that they deal with non-overlapping domains, even if we have no direct experimental access to their domain of overlap yet.

But that's what I said earlier. Till we get to THAT scale, we have no direct experimental evidence. Since the existence of black holes are still indirect, performing QM vs. GR experiments there are still a long way off. We have no experiments as of yet, and in the near future, to test such things. Thus, using such a scenario to imply "logical inconsistency" of QM is premature and certainly, at least in my book, illogical.

Take note that, at the very simplest level, QM HAS incorporated a "quantization" of gravitational potential. This is seen in the recent experiment of neutrons falling in gravitational fields.[1] While this isn't the GR effects we are looking for, it is at least another indication that QM has more gravitational consideration in it than GR has for QM.

Zz.

[1] V.V. Nesvizhevsky et al. Nature v.415, p.297 (2002).

 

[Stingray]

ZapperZ, you almost seem to be intentionally misinterpreting my posts. My last one was stated more precisely than the others. I said, for example, that

On a purely mathematical level, either one of these theories [QFT or GR] works just fine without the other.

But we are not mathematicians. Experiment demands that the two theories cannot be considered as independent axiomatic systems (except for the very useful purpose of approximation ). There must be a single underlying theory which at least reproduces both of these ideas in the regimes that they have been tested. My statement was that naive forms of this combined theory are logically inconsistent. So I implied that BOTH GR and QFT are inconsistent in this sense. You don't seem to like that terminology, so feel free to come up with a different word.

The FACT that this is still a highly active research area means that you are already set in your ways that GR and QM/QFT cannot be merged.

I am completely confident that GR and QM/QFT CAN be merged in the sense I mentioned above (reproducing known experiments). I am almost as confident that this will require significant modification of one or probably both of these theories. As they stand today in textbook form, they are incompatible even at their most basic level. I don't think this can be resolved in any trivial way. Something has to give. Unfortunately, I think that we are arguing semantics again...

Again, even when GR and QM/QFT cannot be made into a unified theory, I do not see why they are both logically inconsistent UNTIL there are experimental evidence to point to such a notion. Until we get to a scale where they both should work equally well and we can see where they both deviate, then we can't say anything.

Huh? We can perform imaginary experiments using the theory that we know, which gives us nonsense. We don't need a (real) experiment to tell us that something that has to change.

Now, it may very well be true that we will never find the "correct" theory of quantum gravity without experimental help, but that's a separate issue.

Also, what was your point in quoting Nesvizhevsky's paper? It is a nice experiment, but I don't think anyone really expected QM to fail at that level. It's even common to assign something like that as an undergraduate homework problem (the theoretical portion, obviously).