Could MWI create universes with fundamentally different physical laws?

[Suekdccia]

TL;DR Summary

Could MWI (Many Worlds Interpretation) create universes with fundamentally different physical laws?

Summary: Could MWI (Many Worlds Interpretation) create universes with fundamentally different physical laws?

Physicist John Wheeler proposed along with Hugh Everett and Bryce DeWitt the 'Many Worlds' Interpretation of Quantum Mechanics (although he expressed some doubts about its validity)

I have read, from an article in Nature, written by Martin Rees and Bernard Carr back in 1979 (https://www.nature.com/articles/278605a0) that Wheeler proposed some kind of multiverse containing universes with fundamentally different laws of physics:

"There may already be room for the anthropic principle in this picture. Wheeler envisages an infinite ensemble of universes, all with different coupling constants and so on. Most are 'stillborn', in that the prevailing physical laws do not allow anything interesting to happen in them; only those which start off with the right constants can ever become 'aware of themselves'. One would have achieved something if one could show that any cognisable universe had to possesses some features in common with our universe. Such an ensemble of universes could exist in the same sort of space as the Everett picture invokes."

I have also contacted with science historian Helge Kragh who in his book "*Higher Speculations: Grand Theories and Failed Revolutions in Physics and Cosmology*" he wrote about this as well:

"Nozick 1981, p. 129. The fecundity assumption is related to the principle of plenitude, but the latter refers only to the realization of possibilities in the actual world we live in. Whereas Nozick did not mention the anthropic principle, he did refer to Wheeler’s speculations of many universes with different laws of physics"

And he told me that Nozick referred to "Everett-Wheeler's Interpretation of Quantum Mechanics"

But how can this be true? Why did Nozick refer to Everett and Wheeler's many worlds interpretation if their universes would not have different fundamental laws?

I mean according to this interpretation all universes would have the same fundamental laws, so all of this does not make much sense to me.

However, I was told that Wheeler's cited comments (from Patton & Wheeler, Quantum Gravity, Ch. 9) hypothesize a situation where the various coupling constants arise in quantum mechanics during cosmogony, and thus different branches of the wavefunction get different values. Therefore, this would not be "some kind of variation of MWI" as much as just MWI applied to a particular form of quantum gravity.

But then my main question here is:

Does that mean that in MWI we could find universes (or branches) with fundamentally different laws of physics? Could we find universes with not just different fundamental constants but radically different fundamental laws of physics?

 

[PeterDonis]

Does that mean that in MWI we could find universes (or branches) with fundamentally different laws of physics?

Since the MWI is based on a specific set of laws of physics, it can't predict different ones.

The terminology of "parallel universes" and "multiverse" applied to the MWI is very confusing, given that those terms are also used in reference to very different (speculative) theories such as string theory, in which "different universes with different laws of physics" does make sense. But it doesn't in the MWI as used in standard QM.

 

[Michael Price]

You ask: Does that mean that in MWI we could find universes (or branches) with fundamentally different laws of physics? Could we find universes with not just different fundamental constants but radically different fundamental laws of physics?

Lets take an example: according to string theory the laws we see in our universe are low-energy manifestations of more fundamental laws. How strings wrap around higher dimensions determine things such as the masses of what appear, to us, to be lower energy fundamental particles - and even with a different number of large scale dimensions, different coupling constants, different standard models. According to MWI all these possibilities are actualised. And according to the anthropic principle most of these low energy universes would be uninhabitable by life. But the fundamental laws (string theory) would be the same for all these universes.

This is just an example since we don't know string theory (or M-theory) to be true. But it is an example of how things might pan out.

https://en.m.wikipedia.org/wiki/String_theory_landscape

 

[PeterDonis]

According to MWI all these possibilities are actualised

I don't know if using the term "MWI" to apply to this prediction of string theory is accurate. I'm not sure this prediction of string theory is due to quantum uncertainty or quantum fluctuations causing multiple decoherent outcomes. AFAIK this prediction of string theory is interpretation independent; that is, it is the same prediction regardless of which interpretation of ordinary QM you adopt.

 

[Michael Price]

I don't know if using the term "MWI" to apply to this prediction of string theory is accurate. I'm not sure this prediction of string theory is due to quantum uncertainty or quantum fluctuations causing multiple decoherent outcomes. AFAIK this prediction of string theory is interpretation independent; that is, it is the same prediction regardless of which interpretation of ordinary QM you adopt.

I guess we won't know for sure until M-theory is finalised. Perhaps ask on a new thread ...

 

[DarMM]

The String Theory landscape is nothing to do with MWI. It's to do with the range of Calabi-Yau backgrounds the compact dimensions can have and the induced range of vacua for the string theory.

 

[Michael Price]

Last time I looked, string theory was a quantum theory.

 

[DarMM]

Last time I looked, string theory was a quantum theory.

I'm not sure what you mean.

Indeed it is, but the landscape is due to the space of vacua which itself is due to the range of compactifications of the extra dimensions. This is a completely separate issue to those related to the interpretational issues in QM. These vacua would occur in a Copenhagen view as well.

 

[PeterDonis]

Last time I looked, string theory was a quantum theory.

Yes, but "quantum theory" is not the same as "MWI"; MWI is an interpretation of the theory, not the theory. As @DarMM and I have both said, the presence of zillions of vacua in string theory is interpretation independent.

 

[Michael Price]

I'm not sure what you mean.

Indeed it is, but the landscape is due to the space of vacua which itself is due to the range of compactifications of the extra dimensions. This is a completely separate issue to those related to the interpretational issues in QM. These vacua would occur in a Copenhagen view as well.

But the vacua might not be actualised, just remaining potential possibilities in Copenhagen. The universe is a bigger place in MWI.

 

[Michael Price]

Yes, but "quantum theory" is not the same as "MWI"; MWI is an interpretation of the theory, not the theory. As @DarMM and I have both said, the presence of zillions of vacua in string theory is interpretation independent.

See my reply to DarMM.

 

[DarMM]

But the vacua might not be actualised, just remaining potential possibilities in Copenhagen.

This isn't right at all. The choice of vacuum determines the Hilbert space and thus the set of available states for either MWI or Copenhagen. Unitary dynamics does not move you across vacua.

It's similar to alternate Hilbert spaces in QFT.

 

[Michael Price]

This isn't right at all. The choice of vacuum determines the Hilbert space and thus the set of available states for either MWI or Copenhagen. Unitary dynamics does not move you across vacua.

It's similar to alternate Hilbert spaces in QFT.

If you refer to Fock space, this a red herring.

 

[DarMM]

If you refer to Fock space, this a red herring.

No I'm not referring to Fock space. Why would that be a red herring anyway?

 

[akvadrako]

This isn't right at all. The choice of vacuum determines the Hilbert space and thus the set of available states for either MWI or Copenhagen. Unitary dynamics does not move you across vacua.

It's similar to alternate Hilbert spaces in QFT.

Unitary dynamics would not "move you" across vacua, but since M-theory is a quantum theory, all the vacua would be in the same Hilbert space. I don't see how it can be otherwise - that would imply string theory is prior to QM.

I suppose Copenhagen can also contain all those vacua, but then it would also be describing a multiverse within a single branch, which seems against the point of non-MW interpretations. I know there are some papers discussing the possibility of stringy worlds being equivalent to Everett worlds -- which would make them strictly incompatible with single-world interpretations. Though they haven't got much attention so I don't know how well received that view is.

 

[Michael Price]

Unitary dynamics would not "move you" across vacua, but since M-theory is a quantum theory, all the vacua would be in the same Hilbert space. I don't see how it can be otherwise - that would imply string theory is prior to QM.

I suppose Copenhagen can also contain all those vacua, but then it would also be describing a multiverse within a single branch, which seems against the point of non-MW interpretations. I know there are some papers discussing the possibility of stringy worlds being equivalent to Everett worlds -- which would make them strictly incompatible with single-world interpretations. Though they haven't got much attention so I don't know how well received that view is.

Agreed on all points. Yes, there are papers claiming the equivalence of worlds or universes from different multiverse schemes (inflation, MWI, string landscape) but they seem confused to me.

 

[DarMM]

Unitary dynamics would not "move you" across vacua, but since M-theory is a quantum theory, all the vacua would be in the same Hilbert space. I don't see how it can be otherwise - that would imply string theory is prior to QM.

I don't understand this. Even simple quantum field theories like $\phi^4$ have multiple Hilbert spaces as different states induce disconnected GNS representations.

I suppose Copenhagen can also contain all those vacua, but then it would also be describing a multiverse within a single branch, which seems against the point of non-MW interpretations

I don't understand this either I have to say. What do you mean "multiverse in a branch"?

 

[DarMM]

What I'm saying here is that since the vacua live in different folia of the observable algebra even in MWI there wouldn't be a branch with one vacuum and a branch with another.

 

[PeterDonis]

since M-theory is a quantum theory, all the vacua would be in the same Hilbert space

This is not correct. As @DarMM pointed out, even in ordinary QFT different vacua are in different Hilbert spaces. This was an important motivation for the algebraic approach to QFT.

 

[akvadrako]

Well, we agree that M-theory is a theory defined on top of the quantum framework? So there must be a Hilbert space where M-theory lives. As I understand it (I don't know much about string theory), each vacua also generates a universe described by another theory in the quantum framework with a different Hilbert space.

I don't understand this either I have to say. What do you mean "multiverse in a branch"?

How do you understand Copenhagen to also contain the string landscape? Basically by "a branch" I mean Copenhagen describes a single branch - so if you say all stringy universes are actualized in Copenhagen, the world there is really a multiverse.

 

[akvadrako]

This is not correct. As @DarMM pointed out, even in ordinary QFT different vacua are in different Hilbert spaces. This was an important motivation for the algebraic approach to QFT.

If this is true then it must be that string theory is not a theory defined on top of QM; that's what I've always understood. I agree it also creates other Hilbert spaces for specific vacua.

 

[PeterDonis]

we agree that M-theory is a theory defined on top of the quantum framework? So there must be a Hilbert space where M-theory lives

This is an invalid argument. @DarMM and I have both explained why.

 

[PeterDonis]

If this is true then it must be that string theory is not a theory defined on top of QM

"A theory defined on top of QM" is vague. String theorists consider string theory to be a quantum theory. But it's certainly not an ordinary non-relativistic quantum theory defined on a single Hilbert space. You apparently think that anything that can be described as a "quantum theory" must be defined on a single Hilbert space. That is simply not true, at least not as the physicists actually working on quantum field theory and string theory are using the term.

 

[charters]

In string landscape eternal inflation, every MWI branch/decoherent history contains arbitrarily many bubble universes, and different bubbles instantiate different string vacua. So as Peter and Dar say, this multiplicity of vacua is interpretation neutral and distinct from MWI. Tegmark calls this Level 2 vs Level 3 multiverses.

In MWI, you would have a "world" where a bubble of vacua type A nucleates closest (in the inflationary spacetime) to a bubble of type B, and another "world" where A nucleates closest to a type C bubble. But all bubble types still exist in each MWI world, whereas other interpretations just cut off all but one MWI branch, but keep all the bubbles of the ontological banch.

With a different cosmology theory, maybe you only have one vacua per branch because you've made inflation non-eternal. But this is not where most of the thinking on this idea is at.

 

[akvadrako]

In string landscape eternal inflation, every MWI branch/decoherent history contains arbitrarily many bubble universes, and different bubbles instantiate different string vacua. So as Peter and Dar say, this multiplicity of vacua is interpretation neutral and distinct from MWI. Tegmark calls this Level 2 vs Level 3 multiverses.

In MWI, you would have a "world" where a bubble of vacua type A nucleates closest (in the inflationary spacetime) to a bubble of type B, and another "world" where A nucleates closest to a type C bubble. But all bubble types still exist in each MWI world, whereas other interpretations just cut off all but one MWI branch, but keep all the bubbles of the ontological banch.

With a different cosmology theory, maybe you only have one vacua per branch because you've made inflation non-eternal. But this is not where most of the thinking on this idea is at.

I totally agree with this, except for the possibility that sometimes vacua and branches might be two words to describe the same thing: for some time separately evolving regions of the wavefunction.

 

[DarMM]

If this is true then it must be that string theory is not a theory defined on top of QM

It is, it's just that QM allows a theory to have separate Hilbert spaces.

As @charters said above a Copenhagen view would have bubbles with each vacua, but only a single "vacua neighbourhood layout". MWI would have all layouts. The multiverse of vacua is separate to MWI.

 

[akvadrako]

"A theory defined on top of QM" is vague. String theorists consider string theory to be a quantum theory. But it's certainly not an ordinary non-relativistic quantum theory defined on a single Hilbert space. You apparently think that anything that can be described as a "quantum theory" must be defined on a single Hilbert space. That is simply not true, at least not as the physicists actually working on quantum field theory and string theory are using the term.

Maybe M-theory can be defined on multiple Hilbert spaces, but each instance would then be defined on one of those spaces.

 

[DarMM]

Maybe M-theory can be defined on multiple Hilbert spaces, but each instance would then be defined on one of those spaces.

It's really no different from how it is in QFT. The theory has multiple disconnected state spaces. Each one would be an "instance" which is slightly different. See $\phi^4$ with a double well for a simpler example.

 

[PeterDonis]

vacua and branches might be two words to describe the same thing

Not if "branches" means what it does in the MWI.

Maybe M-theory can be defined on multiple Hilbert spaces, but each instance would then be defined on one of those spaces.

An "instance" here is not the same as a "branch" in the MWI.

 

[akvadrako]

Not if "branches" means what it does in the MWI.

Why not? That's what those papers are trying to show, that the creation of bubble universes creates separate branches. I wouldn't say I buy it, but I don't see why it's ruled out either.

An "instance" here is not the same as a "branch" in the MWI.

I mean an instance of a string theory landscape. As in you can describe a landscape as existing on some Hilbert space. If this isn't true and a landscape requires multiple Hilbert spaces to be defined, then this is my misunderstanding.

 

[PeterDonis]

If this isn't true and a landscape requires multiple Hilbert spaces to be defined

As I understand it, it does; the landscape is equivalent to multiple different vacua in QFT (each individual "universe" in the landscape is equivalent to a different vacuum state), which require multiple Hilbert spaces to be defined.

 

[akvadrako]

As I understand it, it does; the landscape is equivalent to multiple different vacua in QFT (each individual "universe" in the landscape is equivalent to a different vacuum state), which require multiple Hilbert spaces to be defined.

I understand that too, but that those are the outputs of the theory. To be defined at all requires some space to live on.

 

[PeterDonis]

To be defined at all requires some space to live on.

That would be the "space" of all possible vacua, i.e., the landscape. Not a single Hilbert space.

 

[Suekdccia]

Since the MWI is based on a specific set of laws of physics, it can't predict different ones.

The terminology of "parallel universes" and "multiverse" applied to the MWI is very confusing, given that those terms are also used in reference to very different (speculative) theories such as string theory, in which "different universes with different laws of physics" does make sense. But it doesn't in the MWI as used in standard QM.

In that case how could Wheeler refer to MWI when he proposed that there were parallel universes with different fundamental laws? String Theory was hardly developed at that time, so what did Wheeler want to say with this?

That is my point: If "standard" MWI could not produce other universes with fundamentally different laws then what is happening here?

 

[charters]

I think you are just reading into the word "law" too much. In a vacua with 4 spatial dimensions you have an inverse cube "law" instead of an inverse square "law" for massless, unconfined gauge interactions. So that's something we call a law that changes across vacua. But the uncertainty principle or speed of light are laws that must hold in all vacua.

Whether MWI can have very early branching into different vacua does not require string theory per se - its conceivable to make this anthropic argument in any form of quantum cosmology/gravity where the universe starts off in a high energy false vacua. String theory just does a somewhat better job at defining the stable and metastable points in the vacuum decay chain.