Many Worlds Interpretation and act of measuring

[TrickyDicky]

But do you have a retrocausal theory that works?

I'm pretty sure there isn't or we'd know. But that is the case with MWI, that it just doesn't work, not even the way Copenhagen works being at least honest about it's limits in an upfront way(Heisenberg cut and nonunitary collapse).
Even disciples of Zurek like Jess Riedel admit MWI doesn't work. The reason we
are even discussing it is because is so popular, but that fact belongs to sociology rather than physics.

 

[bhobba]

Until they have been solved they are indeed killers and is cited as the main reason why the majority of realist quantum foundation experts reject MWI. It's not just my personal opinion.

Again you are confused between an issue that needs clarification and an actual problem. For example I gave a possible way out of the factorisation problem as explained by vacuum fluctuations. And its not generally accepted that Wallace doesn't derive the Born Rule - I think he does - but there is a lot of debate about it.

Thanks
Bill

 

[Quantumental]

Again you are confused between an issue that needs clarification and an actual problem. For example I cave a possible way out of the factorisation problem as explained by vacuum fluctuations.

Then you have to flesh it out to a real explanation, otherwise it's barely an idea.

And its not generally accepted that Wallace doesn't derive the Born Rule - I think he does - but there is a lot of debate about it.

Wrong, most people don't think he has succeeded. It's mostly Oxfordians who accept the derivation to some extent. There have been tons of papers arguing against it (Alastair Rae, Huw Price, Adrian Kent, David Albert etc.), even amongst MWI supporters it is highly controversial (Jacques Mallah, Eliezer Yudkowsky, Robin Hanson etc.)

 

[stevendaryl]

IMHO your premise 1. just generalizes your own difficulties to even consider that QM might be incomplete/wrong.

That statement couldn't be more incorrect. I don't have any trouble considering the possibility that QM might be completely wrong. It's just that there is absolutely no evidence for it being wrong.

 

[stevendaryl]

This is blatantly wrong.

No, it's not. I said that there is absolutely no evidence that QM fails when applied to macroscopic objects. That's true.

You can interpret the delay-erasure experiment, quantum pigeons etc. to imply that quantum mechanics is explained by retrocausality.

I didn't say otherwise.

 

[bhobba]

Then you have to flesh it out to a real explanation, otherwise it's barely an idea.

The point is that argument can be reversed. Its an issue that needs fleshing out.

Wrong, most people don't think he has succeeded. It's mostly Oxfordians who accept the derivation to some extent. There have been tons of papers arguing against it (Alastair Rae, Huw Price, Adrian Kent, David Albert etc.), even amongst MWI supporters it is highly controversial (Jacques Mallah, Eliezer Yudkowsky, Robin Hanson etc.)

And you know this precisely how? What survey (of physicists and mathematicians - not philosophers like Huw Price) have you done to know that?

Thanks
Bill

 

[stevendaryl]

I'm pretty sure there isn't or we'd know. But that is the case with MWI, that it just doesn't work, not even the way Copenhagen works .

The way I see it, there is no difference between MWI and Copenhagen, other than where you draw the boundary of what the "system" is.

 

[stevendaryl]

The factorization problem + born rule problem implies that QM does *not* imply MWI.

I'm speaking a little loosely. QM by itself, with no additional assumptions implies macroscopic superpositions. That seems inevitable to me. If you allow macroscopic superpositions, then you're 99% of the way to MWI.

 

[TrickyDicky]

The way I see it, there is no difference between MWI and Copenhagen, other than where you draw the boundary of what the "system" is.

That's like saying there is no difference between a living person and a corpse other than where you draw the boundary between life and death.
The distinction you mention is capital.

 

[Quantumental]

And you know this precisely how? What survey (of physicists and mathematicians - not philosophers like Huw Price) have you done to know that?

Well you're the one that stated that it's generally accepted that he did derive it, so I can ask the same to you. I am going by the papers I have come across. I am completely open to being wrong if you can supply more papers accepting his derivation than I just mentioned of people not accepting it. And I don't see why you exclude philoophers of physics who got the same mathematic credentials. This is afterall very philosophically loaded in the firt place.

 

[bhobba]

The way I see it, there is no difference between MWI and Copenhagen, other than where you draw the boundary of what the "system" is.

I tend to agree with that - except I would say it was what many call Copenhagen done right - Consistent Histories. The only difference is interpreting what a history is.

Thanks
Bill

 

[stevendaryl]

I'm speaking a little loosely. QM by itself, with no additional assumptions implies macroscopic superpositions. That seems inevitable to me. If you allow macroscopic superpositions, then you're 99% of the way to MWI.

The counterargument that decoherence implies there ARE no macroscopic superpositions is, I think, an argument about semantics. Decoherence doesn't get rid of superpositions, it just makes it where, FAPP (For All Practical Purposes), we can ignore macrocroscopic superpositions, since interference between them is unobservable. Or maybe a different way of saying it is that if you start with a superposition of "Dead cat + Live cat", very rapidly, the situation changes to a superposition of "The Earth with a dead cat + The Earth with a live cat" and the "split" goes on to infect the entire universe.

That doesn't mean there are no macroscopic superpositions, it means such superpositions necessarily involve the entire accessible universe.

 

[Quantumental]

I'm speaking a little loosely. QM by itself, with no additional assumptions implies macroscopic superpositions. That seems inevitable to me. If you allow macroscopic superpositions, then you're 99% of the way to MWI.

I disagree because quantum mechanics itself is completely agnostic about the wavefunction itself. It doesn't say that the wavefunction exists. And all our experiments to date suggest there is in fact only 1 outcome and one world.

 

[TrickyDicky]

It's just that there is absolutely no evidence for it being wrong.

I said wrong or incomplete and there is plenty of evidence to consider the latter, think of all the peoplee working on Quantum gravity.
But the key point is that you cannot convert your own opinión about evidence (or lack of) about The status of QM in a logical premise to conclude that MWI must be Right. It is just not serious in a scientific Forum.

 

[bhobba]

And its not generally accepted that Wallace doesn't derive the Born Rule - I think he does - but there is a lot of debate about it.l

Well you're the one that stated that it's generally accepted that he did derive it, so I can ask the same to you.

Hmmmm. English was not my best subject, I was MUCH better at math - but I don't think that's what I said.

Thanks
Bill

 

[Quantumental]

Hmmmm. English was not my best subject, I was MUCH better at math - but I don't think that's what I said.

Indeed it is what you are saying by denying that it is generally accepted. but anyways uninteresting topic.

 

[stevendaryl]

I disagree because quantum mechanics itself is completely agnostic about the wavefunction itself. It doesn't say that the wavefunction exists. And all our experiments to date suggest there is in fact only 1 outcome and one world.

As for the first thing, it doesn't matter what the wave function "is". It matters what QM says are its implications.

So, as far as the predictions of QM, if you have a starting state $A$, and you have two alternative intermediate states $B_1$ and $B_2$, and one final state $C$, then in computing the probability of going from state $A$ to state $C$, you have to include interference terms from the two paths

$A \Rightarrow B_1 \Rightarrow C$

$A \Rightarrow B_2 \Rightarrow C$

In the case $B_1$ and $B_2$ are macroscopically distinguishable, then you're taking into account macroscopic superpositions. It doesn't matter what ontological status you give to the wave function. Now, it doesn't make any difference, in practice because if the intermediate states $B_1$ and $B_2$ are macroscopically distinguishable, then there is (usually) negligible amplitude for both of them to lead to the same final state $C$. So macroscopic superpositions are not observable. But as I said, they are consequences of QM. (Just unverifiable consequences)

But the flip side is true, as well. To say that "our experiments to date suggest there is in fact only 1 outcome and one world" to me means that you've measured the consequences of macroscopic superpositions and found those consequences false. That has not happened, for exactly the same reason that nobody has verified those consequences.

 

[stevendaryl]

I said wrong or incomplete and there is plenty of evidence to consider the latter. think of all the peoplee working on Quantum gravity.

Quantum gravity is about applying QM to spacetime itself, it seems to me. That doesn't imply that QM is incomplete.

But the key point is that you cannot convert your own opinión about evidence (or lack of) about The status of QM in a logical premise to conclude that MWI must be Right. It is just not serious in a scientific Forum.

The possibility of macroscopic superpositions is a consequence of QM. It seems to me that exploring the consequences of a scientific theory is appropriate in a scientific forum.

 

[atyy]

I don't. I know several people are working on assembling such theories (Huw Price, Ken Wharton, Matthew Leifer, Yakir Aharanov etc.). Some have even extended it to MWI (Lev Vaidman).
But it is what nature seems to imply in several experiments.

OK, so there is no consensus retrocausal theory at the moment either. I think TI is the most prominent one, but it still needs work.

But your own (as well as others) belief in MWI seems motivated by the simplicity of the idea that we just extend the idea of the superposition and wavefunction to entail the univere as a whole. But how would that work? There is nothing external to the universe, hence nothing to decohere it. There is also no explanation of fundamental ontology in MWI. This is where David Wallace is still struggling to come up with some deeper ontology than the wavefunction which goes far beyond the quantum formalism as is explained by Jeffrey Barrett (author of the standard entry of Everett) in this paper page 36: http://www.socsci.uci.edu/~jabarret/bio/publications/everett4.pdf
This again enters the territory of the problem of factorization too. You yourself accept that there is no solution to this at this present time, yet you don't seem troubled by it. I have to ask: Why not?

I'm glad you think I believe in MWI, because I don't! I just try to keep an open mind. I agree that there is no version of MWI on which there is consensus about its viability, even among the distinguished proponents of the interpretation.

Why not just accept for instance de-Broglie Bohm then? Sure it has problems with non-locality, but it's no worse than factorization, and it does seem to be what nature tell us. Particles moving in the pattern of waves. It has no probability problems like MWI does...Or as mentioned retrocausality as several experiments seem to suggest.. Why blindly favour MWI?

As I understand it, dBB cannot be accepted without experimental evidence, because dBB is an interpretation with a common sense reality. If one accepts dBB without experimental evidence, then one is saying that the Bohmian trajectories are not real, which would be against the point of dBB. dBB is not discussed much because there is consensus about it. It works for non-relativistic QM, and QED. There are proposals for extending it to the full standard model, and maybe they are correct, but whether they are needs to be carefully checked, and the difficulty is at the point where a professional has to do it.

 

[dx]

One of the central points in Copenhagen is that you cannot have a "universal wavefunction." In any application of quantum theory, there must be some heavy objects which are placed "outside the description" so to speak, which define the situation and the possible phenomena that can appear. The classically described situation is specified by the classically defined parameters like X or θ that enter into the Schrodinger differential equation. The fact that these measuring bodies can also be subjected to measurement only implies that there must be additional heavy bodies which must be introduced relative to which these measurements take place. There is no objective "classical" vs "quantum" cut, but in any application of qm, there must be some bodies which are taken to be outside the description, without which the parameters entering into the schrodinger equation would not make any sense.

 

[atyy]

Quantum gravity is about applying QM to spacetime itself, it seems to me. That doesn't imply that QM is incomplete.

If Copenhagen is taken as QM, then there is an argument that QM is incomplete when applied to spacetime itself. The reason is that Copenhagen has a macroscopic observer, who presumably has a lab on a piece of classical spacetime. So far the only plausible proposal for non-perturbative quantum gravity is AdS/CFT. And so far only the observer on the "classical" boundary has full QM. For example, in the proposed firewall resolution of Papadodimas and Raju, I think the infalling observer sees no violation of QM, but does not have "full QM" either. This may change of course. People are trying to extend the lessons from AdS/CFT to dS, and also trying to figure out how much an observer in the bulk has QM.

 

[stevendaryl]

One of the central points in Copenhagen is that you cannot have a "universal wavefunction."

I suppose so, but you don't need a universal wavefunction in order to have the possibility of macroscopic superpositions, which is all that you need for the possibility of "alternate worlds". The frustrating thing about Copenhagen is that there is this boundary between the "observer" and the "system". On one side of the boundary, things are treated classically, and on the other side things are treated quantum-mechanically. But there is nothing in QM that suggests a maximum size of the system on the QM side.

 

[stevendaryl]

If Copenhagen is taken as QM, then there is an argument that QM is incomplete when applied to spacetime itself. The reason is that Copenhagen has a macroscopic observer, who presumably has a lab on a piece of classical spacetime. So far the only plausible proposal for non-perturbative quantum gravity is AdS/CFT. And so far only the observer on the "classical" boundary has full QM. For example, in the proposed firewall resolution of Papadodimas and Raju, I think the infalling observer sees no violation of QM, but does not have "full QM" either. This may change of course. People are trying to extend the lessons from AdS/CFT to dS, and also trying to figure out how much an observer in the bulk has QM.

Thank you. That's the closest I've seen to evidence that QM is incomplete. But what's the conclusion? In quantum gravity, you have no observers, so what is done with the probability amplitudes (which I assume are still there)? What do they mean in the absence of an observer?

 

[TrickyDicky]

Quantum gravity is about applying QM to spacetime itself, it seems to me. That doesn't imply that QM is incomplete.

It does, I'm afraid. QM as it is currently cannot be made to account for gravity, this implies quantum theory must be changed, corrected or completed as a theory in order to accommodate it in a quantum gravity theory. I suspect you are aware of this.

The possibility of macroscopic superpositions is a consequence of QM. It seems to me that exploring the consequences of a scientific theory is appropriate in a scientific forum.

First of all you should describe mathematically what you mean by macroscopic superposition. You seem to take for granted a concept of superposition(the strong superposition principle) that is not the one implemented in quantum theory and which is prevented precisely by the issues of factorization, preferred decomposition system/environment... that have been discussed here as obstacles to MWI.(It is also prevented by things like superselection rules if you extend the concept to relativistic quantum theory).
If you start by assuming precisely what is seen by the experts as serious issues of MW, you cannot logically draw valid conclusions.

 

[dx]

There is indeed no maximum size. It depends on the situation. In the double slit experiment, we can do two things. We can either use the diaphragm with the slits as a measuring body, and therefore outside the description, in which case we get interference. Or we can study the momentum exchange between the particle and the diaphragm, in which case we would be including the diaphragm on the "quantum side". In the latter case, there will be additional bodies introduced relative to which the momentum transferred to the diaphragm is controlled, and in that case we lose interference. So the diaphragm can either be a "classical" or a "quantum" object, depending on exactly what you are doing. The point however is that there must always be some bodies which are outiside the description. Without those bodies, the parameters in the algebraic or differential equations of which the matrices or wavefunctions are solutions are not defined.

 

[atyy]

Thank you. That's the closest I've seen to evidence that QM is incomplete. But what's the conclusion? In quantum gravity, you have no observers, so what is done with the probability amplitudes (which I assume are still there)? What do they mean in the absence of an observer?

Quantum gravity, and especially quantum cosmology, is the main reason for trying something like MWI or the Gell-Mann and Hartle version of consistent histories.

 

[stevendaryl]

It does, I'm afraid. QM as it is currently cannot be made to account for gravity, this implies quantum theory must be changed, corrected or completed as a theory in order to accommodate it in a quantum gravity theory. I suspect you are aware of this.

I would say that quantum field theory, with current methods, is incapable of accommodating gravity. I think of QM as the more general theory, of which NRQM and QFT are instances.

First of all you should describe mathematically what you mean by macroscopic superposition.

I just did.

 

[stevendaryl]

Quantum gravity, and especially quantum cosmology, is the main reason for trying something like MWI or the Gell-Mann and Hartle version of consistent histories.

That's what I thought. So if someone (TrickyDicky, for example) is using quantum gravity as an argument against MWI (or some other observer-free variant of QM), then he's got things exactly backwards.

 

[Jimster41]

I am struggling to come to terms with the different interpretations, and get a basic understanding of QM. But at this point the "ensemble" interpretation seems most appealing. Partly because it seems consistent with the notion that time is "Universal".

Since the word "time" sort of implies a metric of "time state" maybe better put as - it implies that there is not really any such thing as a "state". The "pure state" and the superposition of histories it implies only has meaning as an observer held memory/prediction of how a set of future measurement events will probably "go". But that in reality... there is only the "going".

-[/PLAIN] from http://en.wikipedia.org/wiki/Ensemble_interpretation[/URL]
"Consider a classical die. If this is expressed in [URL='http://en.wikipedia.org/wiki/Bra%E2%80%93ket_notation']Dirac notation, the "state" of the die can be represented by a "wave" function describing the probability of an outcome given by:

Where it should be noted that the "+" sign of a probabilistic equation is not an addition operator, it is a standard probabilistic or Boolean logical OR operator. The state vector is inherently defined as a probabilistic mathematical object such that the result of a measurement is one outcome OR another outcome."At least to me, then the question of why our predictions of what will happen are probabalistic, seems a bit less... bizzare. A single non-infinite set of truly hidden variables, like a "bulk" would suffice, as a possible cause? And this seems sort of synonymous with the proposed existence of "the future", which is pretty realistic.

Hopefully I'm not off topic, or too mistaken in my attempt to follow.

 

[TrickyDicky]

I just did.

If you refer to your #332, that is clearly not enough, you just said macroscopic superpositions are not observable, and used the concept of strong superposition that is at odds with FAPP QM and as I said is prevented by things like superselection sectors and the issues about preferred decomposition discussed in this thread and put forward by the majority of experts in the community as clearly not solved, that you ignored in your answer (also nicely sumarized in dx's #340).

That's what I thought. So if someone (TrickyDicky, for example) is using quantum gravity as an argument against MWI (or some other observer-free variant of QM), then he's got things exactly backwards.

I didn't use it as an argument against MWI, but as an example of in which sense QM is incomplete.

 

[julcab12]

I haven't worked through the paper, but doesn't having an objective wave function collapse imply nonlocal interactions (according to Bell's theorem)?

.. Been busy lately. Oh, It's a bit of contradiction but i don't think that's the case. The general formalism is that state vector evolves through quantum jump including the basic assumption that the state of the system is entirely described by a vector in Hilbert space. I think the jump creates the illusion of superposition. Observer normalized that state into a objective state and so on. But i don't know. I really have in issue with non locality as being absolute. Non-local correlations is not the same as non-locality in the equations of motion -- which is real to me. It can be normalized as unitary evolution or treat each states as real and probabilistic.

I suppose so, but you don't need a universal wavefunction in order to have the possibility of macroscopic superpositions, which is all that you need for the possibility of "alternate worlds". The frustrating thing about Copenhagen is that there is this boundary between the "observer" and the "system"..

..

...If you're taking it literal then yes! I supposed it's a natural consequence BUT probability distribution and statistics only requires multiplicity as a framework (talking about position) not as a real objective value simply because probability deals specific to averaging. Superposition 'like' behaviors (almost but not as complex as quantum does) are also seen in macro - Einstein ring /cross ( / ) and they're not doing statistical distribution on galaxies, star etc to make sense of it. I think this is not just confined to classical-specific approach but should be taken as 'general' consideration. When we see multiplcity, weirdness or superposition etc etc. We don't stop at the paradigm as 'literal' but investigate what makes that apparent phenomenon regardless of the interpretation.

 

[bhobba]

The general formalism is that state vector evolves through quantum jump including the basic assumption that the state of the system is entirely described by a vector in Hilbert space.

I don't think that's how collapse theories work. They often introduce a small non-linearity to explain collapse but by and large things still evolve by Schroedinger's equation in a deterministic way even though it contains an indeterministic element.

Superposition 'like' behaviors (almost but not as complex as quantum does) are also seen in macro - Einstein ring /cross

Its nothing like quantum superposition eg it doesn't require complex numbers that imply interference effects - at least as far as I can see.

Thanks
Bill

 

[Rajkovic]

bhobba, do you think that there is no collapse?

 

[bhobba]

bhobba, do you think that there is no collapse?

There is nothing in the formalism of QM about collapse - its interpretation dependant - some interpretations have it (eg GRW) and some don't (eg MW).

However even for those that have it, it has morphed into a different question with our better understanding of decoherence. The question now is - why do we get any outcomes at all. Most interpretations stand powerless before it - simply saying - somehow (my ignorance ensemble is like that) but for a few its trivial (eg BM, GRW and MW).

Thanks
Bill

 

[TrickyDicky]

But the objection applies equally to any probabilistic theory. If you a flip a coin some number of times, it's possible to get arbitrarily long sequences of heads-up. Strictly speaking, no finite amount of information can confirm or refute a probabilistic theory. That's true whether or not we consider many-worlds. In practice, we use a cut-off and declare that a probabilistic theory has been refuted if the chance that it is correct is below the cut off. But this leaves a possibility that we come to the wrong conclusion--accept a false theory, or reject a true theory--just because we by chance had an "atypical" run.

When we consider many-worlds, there will obviously be some worlds where the results of experiments will differ significantly from the predictions of QM, and in those worlds, the researchers will come to the wrong conclusion that QM is mistaken.

I would like to come back here to explain why the objection about "wrong" probabilities made on the MWI doesn't apply to any probabilistic theory, in particular why it doesn't apply to QM as a probabilistic theory.
There is a reason why there is consensus that the Born rule cannot be properly derived in MWI unless one postulates(as opposed to derive) decoherence, which implies the Born rule as explained in this thread.
Segregating all possible distributions of outcomes in different isolated worlds doesn't make sense from any probabilistic sound theory.
For the worlds with probability one and zero obviously no notion of probability can arise, but also in those worlds where only a determined probability is fixed there is no possible concept of probability, it would also be purely deterministic. And if there is different outcomes they would obtain the same probability we do. In other words either you have determinism or access to enough variety of outcomes that the predictions would lead to the same probability theory we have, getting rid of the need of worlds and branching.