Is the Many Worlds Interpretation Widely Accepted Among Physicists?

[A. Neumaier]

Pop-sci has the full intellectual authority of institutions like Caltech, MIT, University of Tel Aviv, and Cambridge behind it. Don't you agree this is not good?

I'd even go further and say, it's very bad and harmful for science! First of all it gives the wrong impression about what science is about in the public and what to expect from science as an accurate quantitative description (not explanation) of the world and possible applications in engineering it might imply later.

The problem is that both scientists and the general public at large enjoy this game too much to let it die.
See also https://www.physicsforums.com/posts/5445587 and the post after it.

 

[vanhees71]

Well, but I'd think that science itself is so exciting that you do not need to make it more exciting by rather telling science-fiction stories rather than the science itself. Science fiction is nice, and from time to time I like to read an SF story too, but then I read it as an SF story and not as a popularized explanation of science. The latter is a very difficult business, and of course you must make it somehow interesting to the general audience, simplify it as much as possible to make it understandable etc. but I think it's harmful for pure science and the legitimation of the large costs covered by tax-payer's money to a large extent if you sell science as science fiction or some sensational story of not yet explored results (like the "faster-than-light neutrinos" (->technical trouble with glas fibers and time-keeping), BICEP2's B-modes (->dust), the claim of having found some new particle way below the discovery "threshold" of $5 \sigma$ confidence level...). In the long term this destroys the credibility of science and scientists in the public, and imho the public has a right to be correctly informed about what science has to really say and not some science fiction. If you want to have science fiction you pay for watching a movie or an SF book but not billions for pure science!

 

[atyy]

Same here - and its rather common:
http://www.mat.unimi.it/users/galgani/arch/heisenberg25amer_j_phys.pdf

Evidently Dirac did as well and led him to his much better formulation:
https://en.wikipedia.org/wiki/Paul_Dirac
'Dirac's attention was drawn to a mysterious mathematical relationship, at first sight unintelligible, that Heisenberg had reached. Several weeks later, back in Cambridge, Dirac suddenly recognised that this mathematical form had the same structure as the Poisson Brackets that occur in the classical dynamics of particle motion. From this thought he quickly developed a quantum theory that was based on non-commuting dynamical variables. This led him to a more profound and significant general formulation of quantum mechanics than was achieved by any other worker in this field.'

Maybe it was because:
'Dirac was famously not bothered by issues of interpretation in quantum theory. In fact, in a paper published in a book in his honor, he wrote: "The interpretation of quantum mechanics has been dealt with by many authors, and I do not want to discuss it here. I want to deal with more fundamental things."'

Thanks
Bill

In fact Dirac was very bothered by issues of interpretation in quantum theory. He believed they would be solved by the incompleteness of quantum theory.

 

[atyy]

Dirac was a very wise guy ;-).

Yes, he was. Like most great physicists, he was concerned about the interpretational problems of quantum theory.

 

[atyy]

Well, I've my trouble to understand Heisenberg's papers, starting with his famous Helgoland paper. Maybe, I'm wrong, but as far as I understood, Heisenberg was a strong advocate of the necessity of a classical/quantum cut, but he never could come up with an objective way to define it.

Of course, because the cut is subjective.

 

[vanhees71]

That's interesting. I thought he was only pretty unsatisfied with QED and renormalization and all that. Do you have some reference, where Dirac explains his quibbles with interpretation? I always thought he belongs to the "no-nonsense physicist" type, who don't bother about metaphysics but are after the utmost clear exhibition of the mathematical structure behind the theories and models they think about, at least this is what I have read by Dirac so far (including his famous QT textbook, his papers on QT, the introduction of annihilation and creation operators and QFT, the canonical quantization of Hamiltonian systems with constraints (Dirac brackets), radiation-reaction problem in classical electrodynamics, GR (he wrote one of the shortest and best introductions to GR)).

 

[vanhees71]

Of course, because the cut is subjective.

Good, than it's not physics, and I can forget about it ;-).

 

[atyy]

Good, than it's not physics, and I can forget about it ;-).

No, you can't do physics without it. Unless you have MWI or Bohmian Mechanics or consistent histories etc. This is the great failure of Ballentine and Peres.

 

[A. Neumaier]

No, you can't do physics without it. Unless you have MWI or Bohmian Mechanics or consistent histories etc. This is the great failure of Ballentine and Peres.

This is off-topic here. I therefore created another thread about it.

 

[atyy]

That's interesting. I thought he was only pretty unsatisfied with QED and renormalization and all that. Do you have some reference, where Dirac explains his quibbles with interpretation? I always thought he belongs to the "no-nonsense physicist" type, who don't bother about metaphysics but are after the utmost clear exhibition of the mathematical structure behind the theories and models they think about, at least this is what I have read by Dirac so far (including his famous QT textbook, his papers on QT, the introduction of annihilation and creation operators and QFT, the canonical quantization of Hamiltonian systems with constraints (Dirac brackets), radiation-reaction problem in classical electrodynamics, GR (he wrote one of the shortest and best introductions to GR)).

http://blogs.scientificamerican.com/guest-blog/the-evolution-of-the-physicists-picture-of-nature/

"And when this new development occurs, people will find it all rather futile to have had so much of a discussion on the role of observation in the theory, because they will have then a much better point of view from which to look at things."

"The difficulties in quantum theory are of two kinds. I might call them Class One difficulties and Class Two difficulties. Class One difficulties are the difficulties I have already mentioned: How can one form a consistent picture behind the rules for the present quantum theory? These Class One difficulties do not really worry the physicist. If the physicist knows how to calculate results and compare them with experiment, he is quite happy if the results agree with his experiments, and that is all he needs. It is only the philosopher, wanting to have a satisfying description of nature, who is bothered by Class One difficulties."

"I should like to say a little more about the Class One difficulties. I feel that one should not be bothered with them too much, because they are difficulties that refer to the present stage in the development of our physical picture and are almost certain to change with future development."

"So I shall say that if we can find a way to describe the uncertainty relations and the indeterminacy of present quantum mechanics that is satisfying to our philosophical ideas, we can count ourselves lucky. But if we cannot find such a way, it is nothing to be really disturbed about. We simply have to take into account that we are at a transitional stage and that perhaps it is quite impossible to get a satisfactory picture for this stage."

 

[secur]

Pop-sci has the full intellectual authority of institutions like Caltech, MIT, University of Tel Aviv, and Cambridge behind it. Don't you agree this is not good?

I'd even go further and say, it's very bad and harmful for science!

The problem is that both scientists and the general public at large enjoy this game too much to let it die.

There's an insight article "Why I won't read your new theory" (or similar) by PeterDonis noting how many crackpots think they can invent new science off the top of their heads. Physicists themselves bear considerable responsibility (i.e. blame) for their misguided ambitions. This pop-sci pseudo-science, from respected authors, is terrible. People figure if that's the best real scientists can do, they can do better with no knowledge at all - and they're right.

We can't stop the general public from enjoying this stupid game. And they can't be blamed for believing stuff presented under the imprimatur of major institutions. It's up to physicists themselves to knock it off, and communicate only sensible science to the public.

I always thought [Dirac] belongs to the "no-nonsense physicist" type, who don't bother about metaphysics ...

Yes, and he still was in 1963 (when he wrote the Scientific American article quoted in previous post). But in the 20 years he had left he finally changed his mind and tentatively decided the "collapse" was caused by ... well, I'd rather not say because you won't believe it. Read the biography "The Strangest Man".

 

[A. Neumaier]

People figure if that's the best real scientists can do, they can do better with no knowledge at all - and they're right.

They are not right - it is only that the (perhaps) best real scientists do to turn science into vey simplified but animated popular science that dispenses with formulas. They are usually told by the scientists that there is a BIG difference between real science (whose language is math) and popular science (whose language is plain English) - but they ignore this important extra piece of information.

 

[Demystifier]

I am convinced that pop science makes much more good than harm. Yes, it creates a few crackpots, but they are an exception rather than a rule. Most people who read popular science do not become crackpots. Instead, they become people with some rough understanding of science aware of the fact that their understanding is only rough. Besides, some of them eventually become true scientists themselves, precisely because they initially got interested in science by pop-science books.

 

[stevendaryl]

To be clear, MWI is more than just superpositioning, because superpositioning resolves back to a single "world", whereas MWI allows these worlds to ever increase in number.

Personally, I doubt the "I" in MWI. If it's an interpretation, there should be no way to differentiate it from other "interpretations". But if there are many worlds - and an ever increasing number of them - then there is an ever-increasing amount of information in anyone of those worlds.

I would say that it's definitely an interpretation of no-collapse quantum mechanics. If you introduce a physical collapse, then that makes it a different theory than pure, unitary evolution quantum mechanics.

The "many" in Many Worlds is just about the fact that an arbitrary wave function can be written as an infinite superposition:

|\psi\rangle = \sum_j C_j |\phi_j\rangle

The only thing different about Many Worlds is that (1) the wave function describes the entire world, and (2) the different |\phi_j\rangle can be macroscopically different. In a sense, you get MWI by dropping constraints on the application of QM, rather than by adding new assumptions.

 

[stevendaryl]

I am convinced that that pop science makes much more good than harm. Yes, it creates a few crackpots, but they are an exception rather than a rule. Most people who read popular science do not become crackpots. Instead, they become people with some rough understanding of science aware of the fact that their understanding is only rough. Besides, some of them eventually become true scientists themselves, precisely because they initially got interested in science by pop-science books.

Thank you. I agree.

 

[stevendaryl]

One other argument in favor of MWI for cosmology is that for the first 14 billion years or so after the Big Bang, there were no observers, so the operational interpretation of QM as a way to predict probabilities for the outcomes of observations seems inappropriate.

 

[Demystifier]

One other argument in favor of MWI for cosmology is that for the first 14 billion years or so after the Big Bang, there were no observers, so the operational interpretation of QM as a way to predict probabilities for the outcomes of observations seems inappropriate.

Observers were absent even 4 billion years ago, but nobody uses MWI in planetology to understand how planets were formed 4 billion years ago. The reason for using MWI in cosmology is different. It is the idea that cosmology is supposed to study Universe as a whole, in which case there is no external observer which could observe the whole Universe from the outside. But in my opinion this reason for using MWI is misleading too, because no cosmological measurement really measures the Universe as a whole. You always watch this or that galaxy, and only a very coarse-grained picture of the galaxy.

 

[stevendaryl]

Observers were absent even 4 billion years ago, but nobody uses MWI in planetology to understand how planets were formed 4 billion years ago.

Planetology is usually applied classical (non-quantum) physics, though.

The reason for using MWI in cosmology is different. It is the idea that cosmology is supposed to study Universe as a whole, in which case there is no external observer which could observe the whole Universe from the outside. But in my opinion this reason for using MWI is misleading too, because no cosmological measurement really measures the Universe as a whole. You always watch this or that galaxy, and only a very coarse-grained picture of the galaxy.

Well, I suppose the evolution of a galaxy would fit into the Copenhagen Interpretation if you viewed the first 14 billion years as a very time-consuming experiment, and assume that there is no definite state of the galaxy until astronomers evolve to study it.

 

[.Scott]

I would say that [MWI is] definitely an interpretation of no-collapse quantum mechanics. If you introduce a physical collapse, then that makes it a different theory than pure, unitary evolution quantum mechanics.

The "many" in Many Worlds is just about the fact that an arbitrary wave function can be written as an infinite superposition:

|\psi\rangle = \sum_j C_j |\phi_j\rangle

The only thing different about Many Worlds is that (1) the wave function describes the entire world, and (2) the different |\phi_j\rangle can be macroscopically different. In a sense, you get MWI by dropping constraints on the application of QM, rather than by adding new assumptions.

MWI, as I have heard the term used, suggests many independent worlds that continuously separate from each other and never recombine. In contrast, non-MWI would say that there is never any "separation" and that all "worlds" interact continuously - never reaching any independence.

I think an easy contrast between MWI vs. non-MWI is the Schrödinger Cat. An MWI proponent would say that there are independent instances of both a dead cat and live one. I would say that the experiment is impossible. You can't separate the cat or the radioactive decay particle from the rest of the universe. A decay decision is imposed on the particle based on its participation in the universe and the result is a cat that is definitely dead or alive - but not both. The interaction between the cat and the particle itself affects the timing of the particle decay. The reason I say this is that the alternative is to say that the result seen when the box is open is the result of a decision made with information that did not exist before the experiment started - that a parameter was added to the universe. In MWI, this parameter is what would differentiate between one world and another. In non-MWI, superpositioning never creates a state with information other than that it started with.

I don't know what creates apparent "collapse", but my thought is that some inconsistencies occur as the wave function involves more and more mass. The problem with this "inconsistency" notion, is that it presumes that ultimately only one path will prove out to be free of inconsistencies - but I can't imagine anything that would drive that number to 1 rather than 0 or some large number. Still, something drives events towards apparent collapse. What we observe and remember are fully collapsed events - not a blurred unresolved super-positioned history seen with a blurred unresolved super-positioned mind.

 

[vanhees71]

Well without anybody doing physics you don't need to bother with "interpretation" of theories concerning physics at all ;-)).

 

[stevendaryl]

MWI, as I have heard the term used, suggests many independent worlds that continuously separate from each other and never recombine.

No. In the original development of MWI by Everett, he never assumed anything other than unitary evolution according to Schrödinger's equation. The thing about worlds "splitting off from one another and never recombining" is not an additional assumption; it follows from decoherence. If you have a superposition of two states, \alpha |\phi_1\rangle + \beta |\phi_2 \rangle, the only way to "observe" such a superposition is through interference. To have interference effects, you need to have a "final state" |\psi_{final}\rangle such that both

|\phi_1\rangle \Rightarrow |\psi_{final}\rangle

and

|\phi_2\rangle \Rightarrow |\psi_{final}\rangle

have non-negligible probabilities. But if |\psi_1\rangle and |\psi_2\rangle are macroscopically distinguishable, then there is no such final state |\psi_{final}\rangle reachable from both. If you make a macroscopic change, such as a cat dying, then from then on, the world will be different because that cat died. So effectively, macroscopically distinguishable states never recombine. So you might as well treat them as separate possible worlds. But there is no need for any special, irreversible "splitting" process.

I think an easy contrast between MWI vs. non-MWI is the Schrödinger Cat. An MWI proponent would say that there are independent instances of both a dead cat and live one. I would say that the experiment is impossible. You can't separate the cat or the radioactive decay particle from the rest of the universe.

That's exactly what MWI says. Rather than a superposition of a live cat and a dead cat (which is so unstable as to be practically impossible), the situation evolves into an either/or of (1) a world in which the cat is alive, and (2) a world in which the cat is dead.

A decay decision is imposed on the particle based on its participation in the universe and the result is a cat that is definitely dead or alive - but not both. The interaction between the cat and the particle itself affects the timing of the particle decay. The reason I say this is that the alternative is to say that the result seen when the box is open is the result of a decision made with information that did not exist before the experiment started - that a parameter was added to the universe. In MWI, this parameter is what would differentiate between one world and another. In non-MWI, superpositioning never creates a state with information other than that it started with.

I'm not sure I understand the distinction you are making. The idea behind MWI is just ordinary quantum mechanics, but where you move the boundary of what you consider "the system" to include more and more of the universe. You can either view

• the atom of uranium as "the system", and the cyanide canister is measurement device, or
• the atom + cyanide canister is the system, and the cat is an observer, or
• the atom + cyanide canister + cat is the system, and the person opening the box is the observer, or
• the atom + cynanide canister + cat + person is the system, and a second person is the observer, or
• ...
• the entire universe is the system

However, if you want to describe the system using a pure state (wave function) as opposed to a mixed state (density matrix), then that is only possible if the system is (at least temporarily) isolated. Only at the very small (a single atom) and the very large (the whole universe) can the system be considered isolated, so the use of pure states only applies at those two levels.

I don't know what creates apparent "collapse", but my thought is that some inconsistencies occur as the wave function involves more and more mass. The problem with this "inconsistency" notion, is that it presumes that ultimately only one path will prove out to be free of inconsistencies - but I can't imagine anything that would drive that number to 1 rather than 0 or some large number. Still, something drives events towards apparent collapse. What we observe and remember are fully collapsed events - not a blurred unresolved super-positioned history seen with a blurred unresolved super-positioned mind.

Well, MWI definitely doesn't predict that anyone sees blurred superpositions.

 

[Demystifier]

Planetology is usually applied classical (non-quantum) physics, though.

So is cosmology (except at very early times).

 

[Demystifier]

Well, I suppose the evolution of a galaxy would fit into the Copenhagen Interpretation if you viewed the first 14 billion years as a very time-consuming experiment, and assume that there is no definite state of the galaxy until astronomers evolve to study it.

Exactly!

 

[secur]

People figure if that's the best real scientists can do, they can do better with no knowledge at all - and they're right.

They are not right - it is only that the (perhaps) best real scientists do to turn science into vey simplified but animated popular science that dispenses with formulas. They are usually told by the scientists that there is a BIG difference between real science (whose language is math) and popular science (whose language is plain English) - but they ignore this important extra piece of information.

I am convinced that pop science makes much more good than harm. Yes, it creates a few crackpots, but they are an exception rather than a rule. Most people who read popular science do not become crackpots. Instead, they become people with some rough understanding of science aware of the fact that their understanding is only rough. Besides, some of them eventually become true scientists themselves, precisely because they initially got interested in science by pop-science books.

Certainly a lot of popular science books are good. For instance "The Milky Way" by Bart Bok; "Road to Reality" by Roger Penrose. Even the bad ones are, perhaps, 2/3 good. But the errors (mostly in philosophy not physics) are terrible. For instance David Deutsch in "Beginning of Infinity" actually tries to "prove" that infinity exists in the real world! Clearly he's never even heard of Aristotle, Kant, and others who long ago dissected his elementary errors. Richard Feynman in "Lectures" said (paraphrasing, don't have the book handy) that QED "explains why the cream in your coffee swirls, it even explains why you like cream in your coffee." Then there's "Quantum Suicide" - 'nuff said.

You both feel that modern physics is basically in great shape and little peccadilloes like these can be shrugged off as harmless. Whereas I see a big problem: hubristic rejection of real-world experimental data in favor of sheer fantasy.

Such books (and attitudes) do, indeed, help attract the next generation of scientists. But they repel the cream of the crop.

 

[atyy]

Does MWI assume classical spacetime?

 

[Demystifier]

Certainly a lot of popular science books are good. For instance "The Milky Way" by Bart Bok; "Road to Reality" by Roger Penrose.

The "Road to Reality" is really good, but I wouldn't classify it as popular. It requires certain mathematical literacy. I would call it semi-popular.

 

[Demystifier]

Does MWI assume classical spacetime?

It depends. You can apply MWI to any quantum theory, from non-relativistic QM to quantum gravity.

 

[stevendaryl]

You both feel that modern physics is basically in great shape and little peccadilloes like these can be shrugged off as harmless. Whereas I see a big problem: hubristic rejection of real-world experimental data in favor of sheer fantasy.

I'm probably going to be sorry that I asked, but what's an example of real-world experimental data that is being rejected by physicists out of hubris?

 

[stevendaryl]

Does MWI assume classical spacetime?

MWI really amounts to an attempt to apply quantum mechanics to the entire universe, instead of splitting the universe into observer + system. I don't think that it depends on any particular assumptions about what the universe is like, other than that it can be described by quantum mechanics.

However, there is a huge problem with MWI, which is reconciling its picture of amplitudes evolving unitarily with the way the world appears to us. It seems possible to me that we have to assume something special about the universe in order for MWI to reproduce a world like ours (with macroscopic objects having approximately definite positions and momenta, and with microscopic objects obeying the Born rule for probabilities).