I Danger for the Many-Worlds Interpretation?

[zonde]

If consciousness is not a matter of the physical state of your brain, then no theory of physics can account for your conscious experience. But this has nothing to do with whether the MWI is an interpretation of QM or a different theory; that depends only on whether the MWI makes different experimental predictions from standard QM. As MWI is currently formulated, it doesn't.

The point is that MWI does not make experimental predictions. MWI predicts experiences of experimenter. But experiences of any person are subjective until he compares them with experiences of other people and extracts objective content out of his subjective experience.
In science we have a model and physical reality. We compare the two. And we assume that that the model is independent from physical reality i.e. the part of physical reality that does the modeling and compares it with observations is sort of separated from the part of physical reality that we are modeling and observing. Otherwise it becomes hopelessly circular.
What is prediction in MWI? Observation? Does not seem so. Conscious experience of of experimenter? Who then is registering this conscious experience of experimenter? Is it experimenter #2? So is then observations of experimenter #2 what is predicted? Or again it is conscious experience of of experimenter #2?

 

[akvadrako]

Do you think classical probability must also be taken in a Many Worlds sense?

I wouldn't say it has to, but I think it provides an explanation. If you have two worlds, an up world and a down world, or measures of infinite worlds with equal proportions, then assuming the situation is completely symmetrical, one can say what it means to give 50% probability of experiencing each. The probability is real; no vanishing is required.

But saying that some situation is the same as a dice role doesn't provide an explanation in the same sense. It's only helpful if you find the randomness in a die role unproblematic.

 

[Minnesota Joe]

I actually do not like MW as its usually interpreted - that the different worlds are actually in some sense real - rather I like Gell-Mann's view - which is exactly what does real mean - he would prefer, as do I, treated on equal footing rather than real:

How do one of the branches treated on equal footing end up being actual on Gell-Mann's view?

 

[Minnesota Joe]

But saying that some situation is the same as a dice role doesn't provide an explanation in the same sense. It's only helpful if you find the randomness in a die role unproblematic.

I don't understand why anyone would ever think it unproblematic: given identical initial conditions there were 6 possible outcomes and nothing explains why you rolled a 1 instead of 2-6. It's just a brute fact. That's a very strong claim.

 

[DarMM]

How do one of the branches treated on equal footing end up being actual on Gell-Mann's view?

According to this view it isn't explained/ lies outside QM. Roland Omnès's book "The Interpretation of Quantum Mechanics" discusses it more particularly in the final chapter.

 

[PeterDonis]

The point is that MWI does not make experimental predictions. MWI predicts experiences of experimenter.

So does every other interpretation of QM, if you want to take this viewpoint. MWI is no different in this respect from other interpretations.

What is prediction in MWI? Observation? Does not seem so.

Why not?

 

[Quanundrum]

According to this view it isn't explained/ lies outside QM. Roland Omnès's book "The Interpretation of Quantum Mechanics" discusses it more particularly in the final chapter.

How is this then different from "Shut up and calculate" ?

 

[DarMM]

How is this then different from "Shut up and calculate" ?

You need to expand on exactly what you mean. "Shut up and calculate" is a short pithy phrase that David Mermin used and later retracted.

Saying that the quantum formalism doesn't make a deterministic statement about which outcome will occur is not the same as making no comment about nature or the formalism at all. So I wouldn't say they are the same.

 

[bhobba]

How do one of the branches treated on equal footing end up being actual on Gell-Mann's view?

What was it Newton said - Hypotheses non fingo (Latin for "I feign no hypotheses", "I frame no hypotheses", or "I contrive no hypotheses"). It was used by Isaac Newton in an essay, "General Scholium", which was appended to the second (1713) edition of the Principia.

This is the interpretation game. You pick the thing that worries you and that decides the interpretation you prefer. The question to ask is why does it worry you? Einstein initially worried about it, but eventually he put his finger on what really worried him - lack of objective reality as suggested by implications of entanglement. I have recently read a paper on the history of science that suggests complementarity was really entanglement, but obscurely expressed (Bohr was well known for being obscure and he also was a famous mumbler). So one could say that the real issue between Einstein and Bohr was how to view entanglement. Bohr thought it was a general feature of nature and was simply the way it is, Einstein was deeply worried to the point he thought QM incomplete.

Thanks
Bill

 

[bhobba]

How is this then different from "Shut up and calculate" ?

Shut up and calculate does not worry about the issue of how to define observation, measurement or whatever you want to call it, in a purely quantum way.

Thanks
Bill

 

[Minnesota Joe]

According to this view it isn't explained/ lies outside QM. Roland Omnès's book "The Interpretation of Quantum Mechanics" discusses it more particularly in the final chapter.

What was it Newton said - Hypotheses non fingo (Latin for "I feign no hypotheses", "I frame no hypotheses", or "I contrive no hypotheses"). It was used by Isaac Newton in an essay, "General Scholium", which was appended to the second (1713) edition of the Principia.

In the video Gell-Mann dismisses as meaningless the idea that two branches are "equally real". Unfortunately he doesn't justify this with anything except incredulity there. So he seems to have at least that much of a hypothesis: before branching things are on the same footing and after branching, there is one actuality.

Do you know the missing argument? Why is it meaningless to talk about two real branches?

This is the interpretation game. You pick the thing that worries you and that decides the interpretation you prefer. The question to ask is why does it worry you? Einstein initially worried about it, but eventually he put his finger on what really worried him - lack of objective reality as suggested by implications of entanglement. I have recently read a paper on the history of science that suggests complementarity was really entanglement, but obscurely expressed (Bohr was well known for being obscure and he also was a famous mumbler). So one could say that the real issue between Einstein and Bohr was how to view entanglement. Bohr thought it was a general feature of nature and was simply the way it is, Einstein was deeply worried to the point he thought QM incomplete.

I have at least two authors telling me how charismatic Bohr was, so it cracked me up to learn he was a mumbler. And Einstein didn't work well with others. Geniuses, what are you going to do?

Anyway, I know the EPR paper stresses those things, but apparently Einstein thought that the non-local aspect of entanglement was the most important issue. He thought Podolsky, who was responsible for the writing of the paper, made a mess of it and they reportedly had a falling out after.

 

[bhobba]

In the video Gell-Mann dismisses as meaningless the idea that two branches are "equally real".

I just had a look at the video again. What he says is what does equally real mean because real is a rather vague term. He prefers on equal footing except for probabilities. Real is one of those terms that is very difficult to pin down exactly - philosophers have been arguing what it is for centuries with no progress as far as I can see. Gell-Mann was just trying to remove that vagueness. While I can't think off hand of another way to do it, a lot of interpretation stuff is just personal preference - Gell-Mann does not like real so replaces it with on equal footing which he prefers - as do I.

Yes the mannerisms of some famous scientists is an interesting topic. I hadn't heard Einstein didn't work well with others though. But gleaning things from recollection, especially others, is fraught with danger as I was reminded again only today. I answered a question on another forum on a topic I have no direct experience with. It was a story someone told me years ago at work - of course I made that clear in the post. One reply was - and today on things that never happened. From his profile he knew a lot more about the subject than me so I would say he is likely right. Human recollection can be quite flawed, especially if its second or third hand, so its best to try and avoid doing what I did, but sometimes its hard to resist. I am sure most have done the exercise on you tell someone in a class something, and they pass it on to the next person and so on. Then the last person repeats what they were told - and it often bears no relation to what the first person was told. For Einstein my reference is Pais - Subtle is the Lord - but even that is fraught with the recollection issue.

Thanks
Bill

 

[DarMM]

In the video Gell-Mann dismisses as meaningless the idea that two branches are "equally real". Unfortunately he doesn't justify this with anything except incredulity there. So he seems to have at least that much of a hypothesis: before branching things are on the same footing and after branching, there is one actuality.

Do you know the missing argument? Why is it meaningless to talk about two real branches?

If you look at his papers and those of others who've worked on similar problems like Omnès or people who work in quantum information it's just basically that QM is probabilistic. It treats them all equally in the sense that each history is assigned a probability of occurrence, but only one history actually occurs. This is no different in a sense to how brownian motion or any other stochastic process treats all paths on equal footing, but only one actually occurs.

He thought Podolsky, who was responsible for the writing of the paper, made a mess of it and they reportedly had a falling out after.

This is because Podolsky leaked details of the paper to the popular press.

 

[DarMM]

But saying that some situation is the same as a dice role doesn't provide an explanation in the same sense.

Again Copenhagen people wouldn't claim it does, they say QM doesn't provide a mechanism for how the history is selected. It's not that they're saying this is an explanation or satisfying in some way, they're saying the quantum formalism doesn't give such an explanation.

 

[zonde]

The point is that MWI does not make experimental predictions. MWI predicts experiences of experimenter.

So does every other interpretation of QM, if you want to take this viewpoint. MWI is no different in this respect from other interpretations.

In other interpretations outcome of experiment is objective physical fact independent from experiences of experimenter. This is no so in MWI, in MWI outcome is subjective. That's the difference.

What is prediction in MWI? Observation? Does not seem so.

Why not?

MWI predicts that every possible outcome happens. In order to mach what is observed it predicts that experimenter observes only one of the possible outcomes. So outcome of experiment does not stand on it's own, you have to include experimenter into model and certain relations between outcomes and experimenter. Only then you get what's observed.

 

[Minnesota Joe]

If you look at his papers and those of others who've worked on similar problems like Omnès or people who work in quantum information it's just basically that QM is probabilistic. It treats them all equally in the sense that each history is assigned a probability of occurrence, but only one history actually occurs. This is no different in a sense to how brownian motion or any other stochastic process treats all paths on equal footing, but only one actually occurs.

That just sounds like a Copenhagen Interpretation. Now with more decoherence.

 

[Minnesota Joe]

In other interpretations outcome of experiment is objective physical fact independent from experiences of experimenter. This is no so in MWI, in MWI outcome is subjective. That's the difference.

MWI predicts that every possible outcome happens. In order to mach what is observed it predicts that experimenter observes only one of the possible outcomes. So outcome of experiment does not stand on it's own, you have to include experimenter into model and certain relations between outcomes and experimenter. Only then you get what's observed.

This seems to be what Sabine Hossenfelder is complaining about in the OP. She is saying that the measurement problem is reintroduced by the prescriptive, non-mathematical part of the theory--the part that attempts to account for our experiences.

 

[DarMM]

That just sounds like a Copenhagen Interpretation. Now with more decoherence.

It is basically the Copenhagen interpretation, but expanded to include sequences of events, i.e. histories, rather than singular experiments alone, and a more precise notion of "classicality". I'm not sure what you mean by "now with more decoherence", decoherence is involved but it's only an aspect of the expansion of Copenhagen. One also has the phase cell analysis for the demonstration that classical logic is a subset of QM. This demonstration is independent of decoherence and there is much more.

I didn't mention decoherence above, so I could be misunderstanding your remark as I'm not sure of the context.

 

[Minnesota Joe]

It is basically the Copenhagen interpretation, but expanded to include sequences of events, i.e. histories, rather than singular experiments alone, and a more precise notion of "classicality". I'm not sure what you mean by "now with more decoherence", decoherence is involved but it's only an aspect of the expansion of Copenhagen. One also has the phase cell analysis for the demonstration that classical logic is a subset of QM. This demonstration is independent of decoherence and there is much more.

I didn't mention decoherence above, so I could be misunderstanding your remark as I'm not sure of the context.

Not your fault, but mine and multiple conversations. Decoherent histories was being compared favorably to MWI earlier in the thread, so I was just trying to determine what it explains.

 

[zonde]

This seems to be what Sabine Hossenfelder is complaining about in the OP. She is saying that the measurement problem is reintroduced by the prescriptive, non-mathematical part of the theory--the part that attempts to account for our experiences.

In her video Sabine replaced experimenter with detector and that way made MWI philosophically more acceptable but less elegant and therefore unacceptable for MWI proponents.
But anyways there have to be collapse at some point where we say: from this point onward we will consider only one outcome. If this point happens to be within MWI model then MWI looses it's main selling point - no collapse (Sabine's argument). If it's outside MWI model then MWI does not produce objective predictions (my argument).

 

[Elias1960]

AFAIU about an interpretation which is name "consistent histories":

It is basically the Copenhagen interpretation, but expanded to include sequences of events, i.e. histories, rather than singular experiments alone, and a more precise notion of "classicality". ... One also has the phase cell analysis for the demonstration that classical logic is a subset of QM.

If it contains statements that are in conflict with classical logic, it should be renamed "inconsistent histories" and thrown away.

If there is something in the mathematical apparatus worth not to be thrown away, one should reformulate it in a logically consistent way. The key would be to follow Bell's paper "Against measurement" and stop to name something a measurement which depends not only on the system itself but also on the context.

 

[PeterDonis]

In other interpretations outcome of experiment is objective physical fact independent from experiences of experimenter. This is no so in MWI, in MWI outcome is subjective. That's the difference.

No, in MWI the outcome is not subjective. In each branch, all observers agree on what the outcome of a particular experiment was.

The difference between MWI and other interpretations is that in the MWI, there is not a single result of a measurement; instead, all possible results happen, each in a different branch.

MWI predicts that every possible outcome happens.

In order to mach what is observed it predicts that experimenter observes only one of the possible outcomes.

No, it doesn't. It predicts that the copy of the experimenter in each branch observes only one outcome. But none of those copies is "the same" as the experimenter before the measurement, so you can't say that any of the outcomes is "the" outcome that "the" experimenter observed. That's the difference between the MWI and other interpretations; in other interpretations, you can say that "the" experimenter observed just one outcome. In MWI, you can't.

outcome of experiment does not stand on it's own, you have to include experimenter into model and certain relations between outcomes and experimenter. Only then you get what's observed.

Including the relations between the experimenter and the system being experimented on has to be done in any interpretation of QM; it's part of the minimal "shut up and calculate" math of QM.

I can see a way of making "prediction is not observation" true in the MWI, but it's not quite the way you are describing. In the MWI, if we look at the experimenter subsystem as a quantum system, not just one particular branch but the degrees of freedom in that subsystem as a whole, then after the experiment, that subsystem does not have a well-defined state at all, because it is entangled with the system that was experimented on. (The system that was experimented on doesn't have a well-defined state either.) That's what MWI predicts. But that means that, considering the system as a whole, not just one particular branch in isolation, there is no "observation"--the subsystem that is supposed to "observe" is not in a well-defined state at all, so it can't have "observed" anything since that requires it to be in a well-defined state.

 

[Minnesota Joe]

I just had a look at the video again. What he says is what does equally real mean because real is a rather vague term. He prefers on equal footing except for probabilities. Real is one of those terms that is very difficult to pin down exactly - philosophers have been arguing what it is for centuries with no progress as far as I can see. Gell-Mann was just trying to remove that vagueness. While I can't think off hand of another way to do it, a lot of interpretation stuff is just personal preference - Gell-Mann does not like real so replaces it with on equal footing which he prefers - as do I.

Except Gell-Mann didn't say vague. He said "equally real" has no useful meaning. A vague term can still have useful meaning. What is meant by a "pile of clothes" is vague, because who knows what a 'pile' exactly is, but I understand it well enough to step over it.

I was just curious what he was on about there, because it sounded rather dismissive.

 

[A. Neumaier]

the copy of the experimenter in each branch observes only one outcome. But none of those copies is "the same" as the experimenter before the measurement, so you can't say that any of the outcomes is "the" outcome that "the" experimenter observed.

This would be strange since then it would be impossible to have identity across measurements for anything, even in the branch that we actually observe...

 

[PeterDonis]

This would be strange since then it would be impossible to have idenitty across measurements for anything, even in the branch that we actually observe...

Your "the branch that we actually observe" is a misuse of language when describing the MWI. There is no single "we" after the measurement, because "we" branch along with everything else. After the measurement there are multiple copies of "we", and each copy observes a different measurement result.

As I have said in previous discussions of the MWI, it's hard even to talk about the MWI using ordinary language, because ordinary language has assumptions built into it that the MWI violates.

 

[bahamagreen]

How short is the period of time for a world between MWI branching?

Is it perhaps universally short enough (sub Planck time) that the individual worlds would never have a history, just dt, like an initial condition?

 

[zonde]

I can see a way of making "prediction is not observation" true in the MWI, but it's not quite the way you are describing. In the MWI, if we look at the experimenter subsystem as a quantum system, not just one particular branch but the degrees of freedom in that subsystem as a whole, then after the experiment, that subsystem does not have a well-defined state at all, because it is entangled with the system that was experimented on. (The system that was experimented on doesn't have a well-defined state either.) That's what MWI predicts. But that means that, considering the system as a whole, not just one particular branch in isolation, there is no "observation"--the subsystem that is supposed to "observe" is not in a well-defined state at all, so it can't have "observed" anything since that requires it to be in a well-defined state.

Ok, you have doubts that MWI has well defined "observer" at all. But let's assume we can speak about "observers" in MWI.

No, it doesn't. It predicts that the copy of the experimenter in each branch observes only one outcome. But none of those copies is "the same" as the experimenter before the measurement, so you can't say that any of the outcomes is "the" outcome that "the" experimenter observed. That's the difference between the MWI and other interpretations; in other interpretations, you can say that "the" experimenter observed just one outcome. In MWI, you can't.

This seems like terminology issue. I call a "copy of the experimenter" in single branch an "experimenter". And it stands to reason. Experimenter observes single outcome - that's the claim. So the copy of experimenter observing particular outcome is distinct from another copy of experimenter that observes different outcome - these copies should be called distinct "observers".

No, in MWI the outcome is not subjective. In each branch, all observers agree on what the outcome of a particular experiment was.

The difference between MWI and other interpretations is that in the MWI, there is not a single result of a measurement; instead, all possible results happen, each in a different branch.

So we have different groups of observers that observe the same outcome withing the group. Observers from one group can't communicate with observers from different group. And we call these groups of observers "branches". That way for all observers it appears that their observations are objective.

But then what is prediction? It's not that particular outcome happens objectively. Prediction is that group of observers observe the same outcome. So we have prediction and we can go on to the next step - test this prediction.
Who is going to be the experimenter and what measurements he is going to perform to observe this "group of observers observing the same outcome"? And there is the problem - we don't have any superobservers that are not already included into prediction. We can't compare this prediction with reality in a theory neutral way. MWI is modeling it's own confirmation. So it's not scientific theory, it is philosophical theory (we don't test philosophical models against anything - we just embrace them if we consider them useful or ignore them if we consider them useless).

Including the relations between the experimenter and the system being experimented on has to be done in any interpretation of QM; it's part of the minimal "shut up and calculate" math of QM.

I don't follow you there. Where in the math is that relation?

 

[PeterDonis]

you have doubts that MWI has well defined "observer" at all

No, I'm just pointing out that you can't use the word "observer" the way it is usually used when talking about the MWI. The same goes for many other ordinary language words, as I have already pointed out.

let's assume we can speak about "observers" in MWI.

No, you can't just "assume" this if you mean by "observer" what is usually meant, because the usual meaning of "observer" makes assumptions that the MWI violates. So "assuming" that you can just speak normally about observers in the MWI means you are being inconsistent.

I call a "copy of the experimenter" in single branch an "experimenter".

Yes, but you can't call the copy "the" experimenter because there isn't just one of them after the measurement. Nor can you say anything about "experimenters" that assumes that there is only one of them after the measurement, or that one particular copy is "the same" as the experimenter before the measurement and the other ones aren't. Doing any of those things means you're being inconsistent.

the copy of experimenter observing particular outcome is distinct from another copy of experimenter that observes different outcome - these copies should be called distinct "observers".

Ok, this is fine. But then you actually have to do it and accept the implications. See below.

what is prediction? It's not that particular outcome happens objectively. Prediction is that group of observers observe the same outcome.

With the above meaning of "observers", so that "group of observers" means "the copies of a group of observers that are all in the same branch", yes.

Who is going to be the experimenter and what measurements he is going to perform to observe this "group of observers observing the same outcome"?

It doesn't matter; all of the observers in the group (in a particular branch, since that's what we're using "observers" and "group" to mean) can communicate with each other (since they're all in the same branch) and verify that they all observed the same outcome. This communication is a classical process and doesn't require any quantum branch points anywhere, so it doesn't raise any quantum measurement issue.

we don't have any superobservers that are not already included into prediction.

We don't need any. See above.

We can't compare this prediction with reality in a theory neutral way.

Yes, we can, for this particular prediction. But that's because you restricted this particular prediction to one branch. The kinds of predictions of the MWI we can't compare with reality in a theory neutral way are predictions about the relative weights of multiple branches (or, for that matter, the existence of multiple branches).

Where in the math is that relation?

In your choice of the measurement operator.

 

[Morbert]

Decoherent histories was being compared favorably to MWI earlier in the thread, so I was just trying to determine what it explains.

You might find this additional video by Gell-Mann useful for seeing what DH potentially brings to the table

https://www.webofstories.com/people/murray.gell-mann/163?o=SH
DH is very closely related to Copenhagen, but it ventures where Copenhagen dares not.

 

[Minnesota Joe]

You might find this additional video by Gell-Mann useful for seeing what DH potentially brings to the table

https://www.webofstories.com/people/murray.gell-mann/163?o=SH
DH is very closely related to Copenhagen, but it ventures where Copenhagen dares not.

Definitely interesting. Thanks! Ha! Copenhagen "sort of strains credulity...it's not a convincing fundamental presentation". That's putting it mildly.