are you going to change the philosophy? Are you going to stick with the straightforward way of reading a scientific theory as just telling us what the world is like — Interview with David Wallace

In the classic physics era, massive objects were supposed to attract one another at a distance, without any physical chanel of interaction between them, as by magic. Even Newton thought this was a problem, that the world could not possibly be that way, with actions at a distance. And yet the likes of Wallace were for two centuries quite happy to see Newtonian gravity as "the way the world was like"...

A scientific theory makes predictions about how the world behaves in quantitative terms. It doesn't tell you "what it's like" ontologically or qualitatively, never did, never will.

The only philosophy that needs to change here is that of strict determinism. Remove this assumption and you are left with a probabilistic interpretation of the wave function. You just need one world for that, but one open to surprises.

"This may be true if the probability is 0.5 vs 0.5. What if we wait shorter for the radioactive element to decay and the ratio is 0.58 (living cat) vs 0.42 (dead cat)? How many SA2 and SB2 are there then?"
— SolarWind
Still two, or many. It depends on how you resolve the fact that the BR appears to work in MWI, and that's something I'm not sure how to do... possibly that's a good reason not to buy into it, or maybe it's just something beyond my scope. — InPitzotl

It's even worse. The MWI supporters claim the world divides when the states are decohered. But decoherence is an exponentially decreasing process that is theoretically never complete. Therefore already the basic assumption is wrong and the MWI can be thrown into the garbage can.

You have a source? — Olivier5

https://www.youtube.com/watch?v=kTXTPe3wahc

Most of the video is so-so, nothing new if you've seen the material before although I did pick up a new insight or two. Sean Carroll comes on at 12:30, and his part is definitely worth watching. At 14:25 he says: "Does it happen infinitely often, we don't know," and then says the number is "gigumongous."

"gigumongous." — fishfry

I think the difference between infinite and gigumongous is rather technical.

Interestingly, the conservation of mass and energy would seem gigumongously violated by this constant burgeoning of a gigumongous number of new universes.

It does assume an infinity of worlds. — Olivier5

You're still counting the wrong thing. Believing that France exists is not making 67 million assumptions. Also, if those worlds are a problem with MWI, you should have a problem with them in QM.

Interestingly, the conservation of mass and energy would seem gigumongously violated by this constant burgeoning of a gigumongous number of new universes. — Olivier5

Something is majorly broken with this argument. If your wavefunction has A+B in it, and you have applied a force to a mass in A, does A then move twice as fast? Assuming it did, if we entertain collapse into A, did we lose mass to B going away?

The MWI supporters claim the world divides when the states are decohered. But decoherence is an exponentially decreasing process that is theoretically never complete. Therefore already the basic assumption is wrong and the MWI can be thrown into the garbage can. — SolarWind

That does not follow.

if those worlds are a problem with MWI, you should have a problem with them in QM. — InPitzotl

There's just one world in most QM interpretations.

There's just one world in most QM interpretations. — Olivier5

I don't think you grasp what a world is. If you have a wavefunction expressed as A+B, you have two worlds. If you have a superposition, you have multiple worlds.

"The MWI supporters claim the world divides when the states are decohered. But decoherence is an exponentially decreasing process that is theoretically never complete. Therefore already the basic assumption is wrong and the MWI can be thrown into the garbage can."
— SolarWind
That does not follow. — InPitzotl

What does not follow from what?

I don't think you grasp what a world is. If you have a wavefunction expressed as A+B, you have two worlds. If you have a superposition, you have multiple worlds — InPitzotl

I don't think superposition of states is a good way to think about QM.

What does not follow from what? — SolarWind

Everything you have after "Therefore" does not follow from what you have before "Therefore".

Can I guess? Are you perhaps suggesting MWI requires worlds be fully decohered? If that's the case, you just have a misconception of MWI. MWI isn't proposing that anything different happens with wavefunctions than that they evolve in accordance with Schrodinger's Equation. It's simply, more or less, the idea that measurement is entanglement; that when a device (or observer) measures something in superposition, the device (or observer) is entangling with those states.

MWI isn't proposing that anything different happens with wavefunctions than that they evolve in accordance with Schrodinger's Equation. — InPitzotl

Oh sorry. I actually thought "many worlds" had something to do with many worlds. And in every book about MWI it says that they divide at a measurement. But maybe you are talking about something completely different.

Oh sorry. I actually thought "many worlds" had something to do with many worlds. — SolarWind

It does, but you're attacking a straw man. A "world" is not an entire decohered universe.

And in every book about MWI it says that they divide at a measurement.

That's correct, but that division is an entanglement.

Radioactive-lump system evolves into a superposition (ignoring amplitudes):

|decayed> + |undecayed>

Cat-system measures this; that entangles the cat with these states:

|decayed>|dead cat> + |undecayed>|living cat>

Schrodinger measures this; that entangles Schrodinger with these states:

|Schrodinger-sees-dead-cat>|decayed>|dead cat>
+ |Schrodinger-sees-living-cat>|undecayed>|living cat>

There are now two worlds. We do not have to wait until Alice, or Andromeda, are decohered... the "Schrodinger-sees-dead-cat" no longer interacts with that living cat. Nothing is being added to the equation here; no partitions that aren't there already... we simply do this instead of collapsing when Schrodinger measures the cat-radioactive material system.

I think the difference between infinite and gigumongous is rather technical.

He's saying it's either infinite or a very large finite number.
— Olivier5

Interestingly, the conservation of mass and energy would seem gigumongously violated by this constant burgeoning of a gigumongous number of new universes. — Olivier5

You watched the vid from 14:25 and not from 12:30? Conservation of energy is the first point he handles. The universe has whatever energy it has. At each branch, the universe splits into slices, as you can think of them. The slices share the energy. No new energy is required.

He goes into detail on this point on his blog.

So even though there are more and more branches as time evolves, the contribution of each branch to the total energy is weighted by the factors |a_n|^2, and those numbers go down over time as branches split. The effects precisely cancel, so that the total energy of the universe (all branches included) is constant. It’s just that individual branches get “thinner” over time (their amplitudes get smaller), so they make smaller and smaller contributions to the total.

https://www.preposterousuniverse.com/blog/2021/01/28/energy-conservation-and-non-conservation-in-quantum-mechanics/

That article is well worth reading. It's a thorough layman-level discussion of conservation of energy in quantum physics. "Today I learned," actually last night when I read this piece, that energy isn't even conserved in our own slice of the universe. In quantum theory, energy is merely conserved "on average" and not necessarily all the time. How do you rescue true conservation of energy? With many-worlds! Sean Carroll is a hell of an expositor.

are you going to change the philosophy? Are you going to stick with the straightforward way of reading a scientific theory as just telling us what the world is like
— Interview with David Wallace

In the classic physics era, massive objects were supposed to attract one another at a distance, without any physical chanel of interaction between them, as by magic. Even Newton thought this was a problem, that the world could not possibly be that way, with actions at a distance. And yet the likes of Wallace were for two centuries quite happy to see Newtonian gravity as "the way the world was like"...

A scientific theory makes predictions about how the world behaves in quantitative terms. It doesn't tell you "what it's like" ontologically or qualitatively, never did, never will. — Olivier5

That would be an instrumentalist view of science, but Wallace takes a realist view, i.e., that a theory represents the structure of the world.

Newton thought that the "action at a distance" aspect of his gravitational law was a problem because he was also a realist. As are and were many scientists, including Einstein.

Wallace again:

Answering “yes” to the second question basically commits you to overturning a really pretty solid consensus in philosophy of science that scientific theories really do have to be understood as making claims about what the world is like, and aren’t just shorthands for claims about how experimental devices work. (And it’s a consensus that I think pretty much all scientists share when they’re not actively philosophising. Are there really astrophysicists who think that the reason for talking about stars is to model patterns of detections on photoplates, not vice versa?) — Interview with David Wallace

I don't think superposition of states is a good way to think about QM. — Olivier5

Superposition is a fundamental principle of quantum mechanics, regardless of interpretation. What alternative would you suggest?

Interestingly, the conservation of mass and energy would seem gigumongously violated by this constant burgeoning of a gigumongous number of new universes. — Olivier5

Just to reiterate @fishfry's point with a simple example. In MWI, the total energy of the universe is the weighted average of the energies of each branch. Suppose that there is only a single branch and the total energy is E. Now suppose that the universe splits into two branches such that:

$|\psi\rangle = \sqrt{\frac{1}{2}} |Branch 1\rangle + \sqrt{\frac{1}{2}}|Branch 2\rangle$

When an observer measures the energy, they will measure (approximately) E. So, from the observer's point-of-view, energy has been conserved on their branch. To calculate the total energy of the universe, it is necessary to square the amplitude of each branch and multiply by their respective energies. So:

$Total\space energy = \frac{1}{2}E + \frac{1}{2}E = E$

So the total energy of the universe has also been conserved.

. In MWI, the total energy of the universe is the weighted average of the energies of each branch. — Andrew M

Why the average and not the sum? What is the argument here?

That would be an instrumentalist view of science, but Wallace takes a realist view, i.e., that a theory represents the structure of the world. — Andrew M

That is not what a scientific theory ever does. Science is not religion.

Newton thought that the "action at a distance" aspect of his gravitational law was a problem because he was also a realist.

He could not figure out e.g. how come the star Syrius would exert a physical force on him (and him on Syrius) over such a vast distance, without any intermediary between them. My point was that, beyond Newton himself, very few people actually cared for this element of magic in the classic Newtonian theory of gravity. Most were content with this magical representation of the world.

So if you assess Newton's theory against your criteria of realism, he fails miserably. But if you assess his theory against the quality of the predictions it allowed, then Newton scored big time.

Superposition is a fundamental principle of quantum mechanics, regardless of interpretation. — Andrew M

No, it's not. You don't seem to know much about QM. Have you watched enough youtube videos yet?

Why the average and not the sum? What is the argument here? — Olivier5

Because that reflects the contribution of each branch to the wave function.

Superposition is a fundamental principle of quantum mechanics, regardless of interpretation.
— Andrew M

No, it's not. — Olivier5

You're mistaken. From Paul Dirac's classic textbook:

The discussion in the preceding section about the limit to the gentleness with which observations can be made and the consequent indeterminacy in the results of those observations does not provide any quantitative basis for the building up of quantum mechanics. For this purpose a new set of accurate laws of nature is required. One of the most fundamental and most drastic of these is the Principle of Superposition of States. We shall lead up to a general formulation of this principle through a consideration of some special cases, taking first the example provided by the polarization of light. — The Principles of Quantum Mechanics, Ch.1: The Principle Of Superposition, p4 - Paul Dirac

Also you haven't said what your alternative way of thinking about QM is. For example, how would you describe the double-slit experiment without assuming superposition?

What Dirac is talking about is the superpositions of wavefunctions. Waves can add to one another, as in sound1 + sound2 = sound3.

how would you describe the double-slit experiment without assuming superposition? — Andrew M

I would assume that the photon behaves as a wave until it interacts with something, at which point somehow it behaves as a particle.

I would assume that the photon behaves as a wave until it interacts with something, at which point somehow it behaves as a particle. — Olivier5

That is childish. An interaction always exists: The Van der Waals forces from the laboratory, the natural radioactivity and the gravity of the earth.

That is childish. An interaction always exists: The Van der Waals forces from the laboratory, the natural radioactivity and the gravity of the earth. — SolarWind

Van der Waals forces are inter molecular. Gravity does not affect light.

Van der Waals forces are inter molecular. Gravity does not affect light. — Olivier5

Matter is an antenna for electromagnetic waves via displacement of charge distribution (Van der Waals) and partially absorbs the EM waves.

With gravitation the deflection of light is known.

What Dirac is talking about is the superpositions of wavefunctions. Waves can add to one another, as in sound1 + sound2 = sound3. — Olivier5

OK, we're on the same page then. You seemed to be saying you could think about QM without that.

how would you describe the double-slit experiment without assuming superposition?
— Andrew M

I would assume that the photon behaves as a wave until it interacts with something, at which point somehow it behaves as a particle. — Olivier5

Without detectors at the slits, the photon still interacts with the apparatus at the slits. Yet an interference pattern is still observed.

For an even simpler example, when a photon interacts with a beam splitter, the interaction is described by a wave function (i.e., there is amplitude for both the photon transmitting and reflecting). This is exploited by Mach-Zehnder interferometers, for example.

So your assumption can't be correct.

the photon still interacts with the apparatus at the slits — Andrew M

Not the photons passing the slits.

"the photon still interacts with the apparatus at the slits"
— Andrew M

Not the photons passing the slits. — Olivier5

And what about the many neutrinos that are flowing through the earth, and what about the virtual particles?

the photon still interacts with the apparatus at the slits
— Andrew M

Not the photons passing the slits. — Olivier5

Those photons as well. Amplitude for every possible path through the apparatus contribute to the observed interference pattern - including when the photons are emitted one by one. The characteristics of the apparatus (such as the positions and widths of the slits) determine the paths the photon can take and thus the specific interference pattern.

Even with photon detectors at the slits, the resulting photon position information can be erased, restoring the interference pattern.

The issue can be simplified by considering a Mach-Zehnder interferometer where there are only two possible paths. Photons sent one at a time through an MZI will exhibit wave interference even though they interact with components of the MZI itself (such as beam splitters, mirrors and sample liquid on the paths).

This damn thread is :fire:

So to emphasize, MWI doesn't add the assumption of multiple worlds; the assumption leading to multiple worlds is already there. The traditional interpretation has a world with a dead cat and a living cat in it; those are worlds. MWI instead removes the assumption that some collapses are ontic whereas others are just apparent. — InPitzotl

A "world" is not an entire decohered universe.

Nothing is being added to the equation here; no partitions that aren't there already... — InPitzotl

such as beam splitters, mirrors and sample liquid on the paths). — Andrew M

Ok, so a photon still behaves as a wave when it is reflected by a mirror. Fair enough. That doesn't make for an infinite number of universes, though.

Total energy=1/2E+1/2E=E — Andrew M

So at every branching, the total energy of the universe is divided by 2? And likewise with its mass, I suppose. Since there is a gigamongous number of branching per nanosecond, it follows that if the MWI was true, our universe would become empty of all matter and energy quite rapidly, like in a few seconds.

So at every branching, the total energy of the universe is divided by 2? And likewise with its mass, I suppose. Since there is a gigamongous number of branching per nanosecond, it follows that if the MWI was true, our universe would become empty of all matter and energy quite rapidly, like in a few seconds. — Olivier5

You have a cake with a weight of 1 pound. You divide the cake in two pieces. Now you have two half cakes, each with a weight of 1/2 pound. Global weight is conserved. The universe has some amount of energy, and it's divided among the worlds. The sum of the energy of all the worlds is always the same.

I do share your concern that if the total energy of the universe (universe being all the worlds taken together) is finite, then this places a limit on how many worlds there can be; because as the number of worlds increases, the amount of energy in each world gets smaller. I don't know how Sean Carroll handles this problem.