Possibility that all current interpretations of QM are wrong

[Demystifier]

MWI is the interpretation you necessarily get if you take the equations of quantum mechanics and don't add any magic. It is the most well-defined interpretation.

It's the most well defined intrinsically, as an interpretation of the Schrodinger equation itself. But when it comes to extrinsic interpretation, namely to interpretation of the experimental observations in terms of those equations, MWI has serious problems.

 

[DarMM]

MWI is the interpretation you necessarily get if you take the equations of quantum mechanics and don't add any magic. It is the most well-defined interpretation.

You also have to take the wavefunction as representing a real wave and you have a lack of clarity with the Born rule. This can be sorted out, but MWI after sorting out the Born rule has very similar ambiguities to other interpretations.

 

[mfb]

But when it comes to extrinsic interpretation, namely to interpretation of the experimental observations in terms of those equations, MWI has serious problems.

I see that differently, but this has been discussed a lot already and I don't want to start yet another iteration of that discussion.

 

[DarMM]

I see that differently, but this has been discussed a lot already and I don't want to start yet another iteration of that discussion.

If at some point you want to discuss it, it would be interesting to hear as even most MWI authors don't see it that way. They see getting experimental observations out as requiring a lot of additional structure with the details unresolved.

 

[Elias1960]

MWI is the interpretation you necessarily get if you take the equations of quantum mechanics and don't add any magic. It is the most well-defined interpretation.

I disagree. You have to add a lot of magic. And the equations of quantum mechanics are simply ignored.

From the Schroedinger equation follows, last but not least, a continuity equation for $\rho(q)=|\psi(q)|^2$. According to MWI, this is a continuity equation for something which does not exist.

 

[mfb]

If at some point you want to discuss it, it would be interesting to hear as even most MWI authors don't see it that way. They see getting experimental observations out as requiring a lot of additional structure with the details unresolved.

No, my point was that we discussed it way too often already. One of many threads.

According to MWI, this is a continuity equation for something which does not exist.

Of course it exists! It's the reason to care about the squared amplitude and therefore a critical element of hypothesis testing, see above.

But as I said: I'm not interested in yet another iteration of these discussions, this will be my last post on MWI here.

 

[kith]

No, my point was that we discussed it way too often already. One of many threads.

I just reread part of this thread and FWIW I'm still interested in understanding your position. Now that @DarMM has noticed it, I'm actually hoping that he uses his magic powers and translates it like he did with the thermal interpretation. ;-)

 

[entropy1]

Do you think that an interpretation as follows could be possible in the future?
The wave function is objectively real (no hidden variables). There is wave function collapse but it doesn't happen instantaneously, it's a physical process that occurs at sublight speed. And it is a local theory. Is this a possible future interpretation?

If I am correct, the worlds in MWI are orthogonal, so if a certain measurement yields A, B and C for worlds, and for sake of argument you end up in world B, then that may subjectively be seen as a collapse (into world B).

 

[physika]

Is there a possibility that none of the current interpretations of QM are right?

Yes.

or perhaps one is correct.

Collapse Models
has different predictions, so it can be tested.

 

[Elias1960]

If I am correct, the worlds in MWI are orthogonal, so ...

If MWI worlds are orthogonal, this would require an additional structure which nobody has proposed yet.
So, let's simply denote the set of all those orthogonal worlds with Q, and a particular world with $q \in Q$. Just a denotation. Then, the most general wave function would be a $\mathbb{C}$-valued function on Q. So, there would be a preferred base, namely the configuration space base Q.

But a main selling point of MWI is their claim that they don't need any additional structure, they simply take QT as it is. Different from evil dBB, which prefers Q.

 

[PeterDonis]

If MWI worlds are orthogonal, this would require an additional structure which nobody has proposed yet.

No, it wouldn't. It just requires that the different terms in the entangled wave function that describes the system + observer + environment after measurement are orthogonal. In practice this won't be exactly true, but it will be true to a very good approximation once decoherence has happened.

 

[Elias1960]

No, it wouldn't. It just requires that the different terms in the entangled wave function that describes the system + observer + environment after measurement are orthogonal.

And how does one split the wave function if one has nothing but the wave function, without any additional structure?

Instead of defining such a structure in precise terms, MWI uses handwaving verbal descriptions and refer to particular situations where things are obvious, like Schroedinger's cat. Of course, for Schroedinger's cat it does not matter if you use configuration space or phase space, the dead and the living cat will be (approximately) orthogonal.

But this is not what a reasonably well-defined interpretation should provide. Once one claims to have no additional structure, one should not use verbal handwaving about Schroedinger cats to define those worlds but use precise well-defined fundamental structures.

 

[vanhees71]

The physical predictions of QM are interpretation independent.

If you have alternative models, which make different predictions than standard QM you have physically different theory, which can be tested against QM, and then an appropriate experiment must decide about which theory is the better one.

 

[PeterDonis]

how does one split the wave function

MWI doesn't "split" the wave function. Everything in the MWI is unitary and preserves information; there is no "splitting".

 

[Elias1960]

MWI doesn't "split" the wave function. Everything in the MWI is unitary and preserves information; there is no "splitting".

What are, in this case, those "different terms in the entangled wave function that describes the system + observer + environment after measurement"? IMHO there are no such "different terms" without some structure which makes them different, and without some well-defined operation which splits the wave function into such "different terms".

(My main objection against MWI is that it is not a well-defined interpretation at all. It is diffuse handwaving using imprecise notions which seem plausible only in some particular situations. If one asks for clarification of the meaning of these notions, one never gets precise answers.)

 

[PeterDonis]

What are, in this case, those "different terms in the entangled wave function that describes the system + observer + environment after measurement"?

The terms that come from the interaction Hamiltonian that entangles the system + observer + environment. In the simplest case, where we don't include a separate environment, say the system is a qubit in the z-spin up state and the measurement is a Stern-Gerlach measurement in the x direction. Then the initial state is a product state of the qubit z-spin up (written in the spin-x basis since that's what we're going to measure) and the observer's "waiting to observe result" state. So we have (ignoring normalization):

$$
\Psi_\text{before} = \left( | + \rangle + | - \rangle \right) | \text{waiting to observe result} \rangle
$$

After the measurement, the qubit's state is now entangled with the observer's state, so we have:

$$
\Psi_\text{after} = | + \rangle | \text{observed x spin up} \rangle + | - \rangle | \text{observed x spin down} \rangle
$$

Showing that the unitary evolution induced by the interaction Hamiltonian takes us from $\Psi_\text{before}$ to $\Psi_\text{after}$ is straightforward given the definition of a Stern-Gerlach measurement in the x direction.

 

[kith]

What are, in this case, those "different terms in the entangled wave function that describes the system + observer + environment after measurement"? IMHO there are no such "different terms" without some structure which makes them different, and without some well-defined operation which splits the wave function into such "different terms".

I would put it like this: In the MWI, the fundamental ontological object is the universal wavefunction / state. Whether the "different terms" are there or not, isn't a property of the wavefunction or of the Hilbert space but depends on how (or whether) the Hilbert space is divided into subspaces. This is something which the observer does in order to describe his experiments.

So in a sense, it is the observer who puts the many worlds into the MWI. They aren't inherently present or well-defined in the universal wavefunction and the Hilbert space.